UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The effectiveness of proprioceptive training in the ACL reconstructed knee Liu, Teresa Yeong Lih 1998

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_1998-0520.pdf [ 7.77MB ]
Metadata
JSON: 831-1.0077173.json
JSON-LD: 831-1.0077173-ld.json
RDF/XML (Pretty): 831-1.0077173-rdf.xml
RDF/JSON: 831-1.0077173-rdf.json
Turtle: 831-1.0077173-turtle.txt
N-Triples: 831-1.0077173-rdf-ntriples.txt
Original Record: 831-1.0077173-source.json
Full Text
831-1.0077173-fulltext.txt
Citation
831-1.0077173.ris

Full Text

THE EFFECTIVENESS OF PROPRIOCEPTIVE TRAINING IN THE A C L RECONSTRUCTED K N E E by TERESA Y E O N G LIH LIU B.Sc.(PT), The University of British Columbia, 1994 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in THE F A C U L T Y OF G R A D U A T E STUDIES School of Human Kinetics We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA August 1998 © Teresa Yeong Lih Liu, 1998 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of iCl/jL/JCS The University of British Columbia Vancouver, Canada DE-6 (2/88) 11 ABSTRACT Purpose The main purpose of this study was to determine the effectiveness of proprioceptive training in the A C L reconstructed limb. The second purpose of this study was to determine the relative contribution of isokinetic strength and peak hamstring torque time to functional ability. Methods Ten subjects with unilateral A C L reconstructed limbs were randomly assigned to two experimental groups. Group One (Strength Training Group) consisted of five subjects who were placed on a 12 week general lower body strength training program. Group Two (Proprioceptive Group) consisted of five subjects who were placed on a 12 week proprioceptive training program for the lower extremities. Peak hamstring torque time (PTT) was measured using the protocol described by Small et a l . l . Average concentric and eccentric torques of the quadriceps and hamstring muscles were measured using the KTN-COM isokinetic dynamometer. Functional ability was determined by the one-legged single hop for distance (SLHD) and the one-legged timed hop. Subjective scores were obtained from the Lysholm and Gillquist Knee Scoring Scale and the Tegner and Lysholm Activity Scale. Results Both training protocols were found to influence peak hamstring torque time of the A C L reconstructed limb. No significant differences were found between or within the two experimental groups. However, a group by test occasion interaction effect on peak hamstring torque time was found. A curvilinear relationship between PTT and test occasion was evident for both experimental groups. The strength training group demonstrated a slowing of PTT at 6 weeks, while the proprioceptive training group demonstrated an improvement in PTT at 6 weeks. At the end of the 12 weeks, both experimental groups regressed toward their baseline PTT values, such as the strength training group demonstrated an improvement in PTT while the proprioceptive training group demonstrated a slowing of PTT in the latter six weeks. Ill There was a significant group by test occasion interaction effect on isokinetic strength measures. The proprioceptive group demonstrated greater isokinetic strength gains than the strength training group after 12 weeks of training. Both experimental groups demonstrated similar significant gains in functional ability. Both groups also demonstrated similar significant gains in the subjective assessment (Lysholm and Gillquist Knee Scoring Scale) and the subjective analysis of physical function (Tegner and Lysholm Activity Scale) after 12 weeks of training. Regressional analyses indicated isokinetic strength of the quadriceps and the hamstring muscles to have significant effects on functional ability (SLHD). Average concentric hamstring torque was found to have the most significant effect on functional ability. Conclusion Both strength training and proprioceptive training have an influence on peak torque time. It is proposed that the two types of training influence PTT through different neuromuscular mechanisms. Strength training is proposed to positively influence PTT by increasing fast twitch/slow twitch (FT/ST) muscle area ratio and to negatively influence PTT by decreasing muscle spindle sensitivity. Proprioceptive training is proposed to positively influence PTT by improving coordination and neural activation of the appropriate muscles. Coordination relies on proprioception and kinesthesia. However, the effectiveness of proprioceptive training is dependent on appropriate progression and repetition. Early integration of speed, force, and complexity of movement into proprioceptive training may decrease the effectiveness of training. Thus, the integration of these components should only occur i f precision of performing motor tasks is not compromised. Both strength training and proprioceptive training may have beneficial effects on subjective scores and functional ability. Greater isokinetic strength gains can be observed with proprioceptive training than with strength training as training occurs over time. These greater isokinetic strength gains are proposed to be secondary to improved muscle coordination and neural activation rather than actual muscular hypertrophy. iv The strength of the lower extremities contribute significantly to functional ability. The strength of the hamstring muscle appear to play a greater role in functional ability than the quadriceps. However, due to the small sample size used in this study, the ability to generalize these results may be limited. Table of Contents Abstract ii Table of Contents v List of Tables vii List of Figures x Acknowledgements xiii Chapter One INTRODUCTION 1 Chapter Two LITERATURE REVIEW 3 Embryology of the A C L 3 Gross Anatomy of the A C L 4 Microanatomy of the A C L 6 Vascular Supply of the A C L 7 Neuroanatomy of the A C L 8 Function of the A C L 12 The Neurophysiology of Proprioception and Kinesthesia 17 The Role of Proprioception in the Function of the Knee Joint 25 The Role of the A C L in Neuromuscular Control: Possible Mechanisms 32 Surgical Considerations Relating to Proprioception 34 Current Literature on the Proprioceptive Function of the A C L 38 Rehabilitation 44 Chapter Three METHODOLOGY LITERATURE REVIEW 49 Kinetic Communicator Dynamometer 49 Reflex Hamstring Contraction Latency 55 Functional Testing 59 Subjective Scoring Scales 63 Chapter Four DEFINITIONS OF TERMS 64 Chapter Five PILOT STUDY 65 Purpose 65 Subjects 66 Methods 67 Statistics ' 70 Results 71 Conclusion 79 Chapter Six METHODOLOGY 80 Purpose 80 Subjects 82 Methods 85 Statistics 91 Chapter Seven RESULTS 92 Descriptive Data of Subjects 92 Subjective Results 95 Objective Results 101 Results of Regressional Analyses 130 Chapter Eight DISCUSSION 139 Limitations 150 Chapter Nine S U M M A R Y A N D CONCLUSION 152 Summary 152 Conclusion 155 Chapter Ten CLINICAL R E L E V A N C E A N D RECOMMENDATIONS 156 Clinical Relevance 156 Recommendations 157 Chapter Eleven REFERENCES 158 Appendix A POWER ANALYSIS 171 Appendix B SUBJECTIVE RATING SCALES & SUBJECTIVE ASSESSMENT SHEET 173 Subjective Assessment Sheet 174 Lysholm and Gillquist Knee Scoring Scale 175 Tegner and Lysholm Activity Scale 176 Appendix C TRAINING PROTOCOLS 179 Group One (Strength Training) 180 Group Two (Proprioceptive Training) 181 Appendix D RELIABILITY OF S Y M M E T R Y INDEXES 183 V l l List of Tables Table 3.1 Cronbach Coefficient Alpha for Two Methods of Measuring Reaction Time in the Dominant and Nondominant Hamstring Musculature 57 Table 3.2 Mean Hamstring Reaction Time as Measured Using Peak Torque Time 57 Table 3.3 Mean Hamstring Reaction Time as Measured Using Electromyograph Time 57 Table 5.1 Descriptive Data for Pilot Study 66 Table 5.2 PTT Data for Test Occasion One 71 Table 5.3 Descriptive Data of Different Preset Velocities 72 Table 5.4 PTT at Three Different Angular Velocities (RM A N O V A ) 72 Table 5.5 Pearson Product-Moment Correlation Coefficients for Test Occasion One 73 Table 5.6 Regressional Analyses for Test Occasion One 74 Table 5.7 Comparison of PTT and Peak Torque (@ 190°/sec) Between Test Occasions 78 Table 5.8 Intraclass Correlation Coefficients 78 Table 6.1 A Study Design 86 Table 6.IB Study Design 86 Table 7.1 Descriptive Data of Subjects 92 Table 7.2 Descriptive Data of Subjects (One-Way A N O V A ) 93 Table 7.3 Measures at Test Occasion One (One-Way A N O V A ) 93 Table 7.4 Lysholm and Gillquist Knee Scoring Scale Results 95 Table 7.5 Descriptive Data for Lysholm and Gillquist Knee Scoring Scale Scores 95 V l l l Table 7.6 Lysholm and Gillquist Knee Scoring Scale (RM A N O V A ) 95 Table 7.7 Tegner and Lysholm Activity Scale Results 98 Table 7.8 Descriptive Data for Tegner and Lysholm Activity Scale Scores 98 Table 7.9 Tegner and Lysholm Activity Scale (RM A N O V A ) 98 Table 7.10 Descriptive Data for Average Torques (Nm) at Test Occasion One 101 Table 7.11 Descriptive Data for Average Torques (Nm) at Test Occasion Two 102 Table 7.12 Average Concentric Quadriceps Torque of the A C L Reconstructed Limb (RM A N O V A ) 102 Table 7.13 Average Concentric Quadriceps Torque of the A C L Intact Limb (RM A N O V A ) 104 Table 7.14 Average Eccentric Quadriceps Torque of the A C L Reconstructed Limb (RM A N O V A ) 107 Table 7.15 Average Eccentric Quadriceps Torque of the A C L Intact Limb (RM A N O V A ) 109 Table 7.16 Average Concentric Hamstring Torque of the A C L Reconstructed Limb (RM A N O V A ) 109 Table 7.17 Average Concentric Hamstring Torque of the A C L Intact Limb (RM A N O V A ) 111 Table 7.18 Average Eccentric Hamstring Torque of the A C L Reconstructed Limb (RM A N O V A ) 111 Table 7.19 Average Eccentric Hamstring Torque of the A C L Intact Limb (RM A N O V A ) 113 Table 7.20 Descriptive Data for Functional Hop Tests at Test Occasion One 116 Table 7.21 Descriptive Data for Functional Hop Tests at Test Occasion Two 116 Table 7.22 One-Legged Time Hop Performance of the A C L Reconstructed Limb (RM A N O V A ) 116 IX Table 7.23 One-Legged Timed Hop Performance of the A C L Intact Limb (RM A N O V A ) 118 Table 7.24 SLHD Performance of the A C L Reconstructed Limb (RM A N O V A ) 121 Table 7.25 SLHD Performance of the A C L Intact Limb (RM A N O V A ) 123 Table 7.26 Descriptive Data for PTT 125 Table 7.27 PTT of the A C L Reconstructed Limb (RM A N O V A ) 125 Table 7.28 PTT of the A C L Intact Limb (RM A N O V A ) 128 Table 7.29 Pearson Product-Moment Correlation Coefficients for A C L Intact Limb 130 Table 7.30 Pearson Product-Moment Correlation Coefficients for A C L Reconstructed Limb 130 Table 7.31 Average Concentric Quadriceps Torque Effect (ACL Reconstructed) 131 Table 7.32 Average Eccentric Quadriceps Torque Effect (ACL Reconstructed) 132 Table 7.33 Average Concentric Hamstring Torque Effect (ACL Reconstructed) 133 Table 7.34 Average Eccentric Hamstring Torque Effect (ACL Reconstructed) 134 Table 7.35 Average Concentric Quadriceps Torque Effect (ACL Intact) 135 Table 7.36 Average Eccentric Quadriceps Torque Effect (ACL Intact) 136 Table 7.37 Average Concentric Hamstring Torque Effect (ACL Intact) 137 Table 7.38 Average Eccentric Hamstring Torque Effect (ACL Intact) 138 List of Figures Figure 2.1 The Tripartite Area of Origin 5 Figure 2.2 Mechanoreceptors in the Anterior Cruciate Ligament 11 Figure 2.3 Hypothesized Neurogenic Contribution to Progressive Knee Instability 25 Figure 5.1 Regressional Analysis for Right 190°/s and 210°/s 75 Figure 5.2 Regressional Analysis for Right 190°/s and 230°/s 75 Figure 5.3 Regressional Analysis for Right 2107s and 2307s 76 Figure 5.4 Regressional Analysis for Left 1907s and 2107s 76 Figure 5.5 Regressional Analysis for Left 1907s and 2307s 77 Figure 5.6 Regressional Analysis for Left 2107s and 2307s 77 Figure 7.1 Lysholm and Gillquist Knee Scoring Scale Scores Pre and Post Intervention vs. Group 96 Figure 7.2 Scatter Plot of Lysholm and Gillquist Scores 97 Figure 7.3 Tegner and Lysholm Activity Scale Scores Pre and Post Intervention vs. Group 99 Figure 7.4 Scatter Plot of Tegner and Lysholm Scores 100 Figure 7.5 Average Concentric Quadriceps Torque of the A C L Reconstructed Limb Pre and Post Intervention vs. Group 103 Figure 7.6 Scatter Plot of Average Concentric Quadriceps Torque of the A C L Reconstructed Limb 104 Figure 7.7 Average Concentric Quadriceps Torque of the A C L Intact Limb Pre and Post Intervention vs. Group 105 Figure 7.8 Scatter Plot of Average Concentric Quadriceps Torque of the A C L Intact Limb 106 Figure 7.9 Average Eccentric Quadriceps Torque of the A C L Reconstructed Limb Pre and Post Intervention vs. Group 107 xi Figure 7.10 Scatter Plot of Average Eccentric Quadriceps Torque of the A C L Reconstructed Limb 108 Figure 7.11 Average Concentric Hamstring Torque of the A C L Reconstructed Limb Pre and Post Intervention vs. Group 110 Figure 7.12 Average Eccentric Hamstring Torque of the A C L Reconstructed Limb Pre and Post Intervention vs. Group 112 Figure 7.13 Scatter Plot of Average Eccentric Hamstring Torque of the A C L Reconstructed Limb 113 Figure 7.14 Average Eccentric Hamstring Torque of the A C L Intact Limb Pre and Post Intervention vs. Group 114 Figure 7.15 Scatter Plot of Average Eccentric Hamstring Torque of the A C L Intact Limb 115 Figure 7.16 One-Legged Timed Hop Performance of the A C L Reconstructed Limb Pre and Post Intervention vs. Group 117 Figure 7.17 Scatter Plot of One-Legged Timed Hop Performance of the A C L Reconstructed Limb 118 Figure 7.18 One-Legged Timed Hop Performance of the A C L Intact Limb Pre and Post Intervention vs. Group 119 Figure 7.19 Scatter Plot of One-Legged Timed Hop Performance of the A C L Intact Limb 120 Figure 7.20 SLHD Performance of the A C L Reconstructed Limb Pre and Post Intervention vs. Group 121 Figure 7.21 Scatter Plot of SLHD Performance of the A C L Reconstructed Limb 122 Figure 7.22 SLHD Performance of the A C L Intact Limb Pre and Post Intervention vs. Group 123 Figure 7.23 Scatter Plot of SLHD Performance of the A C L Intact Limb 124 Figure 7.24 Peak Torque Time of the A C L Reconstructed Limb Pre and Post Intervention vs. Group 126 Figure 7.25 Scatter Plot of PTT of the A C L Reconstructed Limb (Linear Regressional Lines) Figure 7.26 Scatter Plot of PTT of the A C L Reconstructed Limb (Quadratic Regressional Lines) Figure 7.27 Average Concentric Quadriceps Torque Effect of the A C L Reconstructed Limb Figure 7.28 Average Eccentric Quadriceps Torque Effect of the A C L Reconstructed Limb Figure 7.29 Average Concentric Hamstring Torque Effect of the A C L Reconstructed Limb Figure 7.30 Average Eccentric Hamstring Torque Effect of the A C L Reconstructed Limb Figure 7.31 Average Concentric Quadriceps Torque Effect of the A C L Intact Limb Figure 7.32 Average Eccentric Quadriceps Torque Effect of the A C L Intact Limb Figure 7.33 Average Concentric Hamstring Torque Effect of the A C L Intact Limb Figure 7.34 Average Eccentric Hamstring Torque Effect of the A C L Intact Limb X l l l Acknowledgements I would like to thank my thesis committee members, Dr. Jack Taunton, Dr. Pat McConkey, and Dr. Donna Mclntyre, for their unending support and encouragement throughout this post graduate program. I would also like to thank Dr. Tim Inglis for offering his support and knowledge. Completion of this degree was a component of the Sports Physiotherapy Fellowship Program at the Allan McGavin Sports Medicine Clinic. Funding for this study was provided by the B.C.M.S.F. Research Grant. 1 Chapter One INTRODUCTION The anterior cruciate ligament (ACL) is the most commonly ruptured ligament in the human knee joint.2 It is estimated that half a million people in the United States sustain a clinically significant A C L injury annually.^ The A C L plays a major role in maintaining the normal function of the knee. Like all ligaments, the A C L guides joint motion by preventing unphysiologic and excessive motions.3 In addition, it works in partnership with the muscles and articular surfaces to transmit the loads of activity across the joint. Once the integrity of the A C L is disrupted, a reduction in functional ability and stability of the affected knee is often seen. Despite all the information to date on the A C L , the understanding of its structure, function, and biology is not to the point where the ligament can be restored to its pre-injury state. Numerous studies have proposed that one of the possible factors limiting the A C L reconstructed or A C L deficient limb from achieving pre-injury status is altered proprioception. It is believed that injury to the A C L results in altered somatosensory information that adversely affects motor control. This may subsequently lead to an increase risk for recurrent injury, decreased performance, or both. Histologically, it has been demonstrated that the human A C L contains mechanoreceptors that can detect changes in tension, speed, acceleration, direction of movement, and the position of the knee jointA 5 Currently, proprioceptive training following an A C L injury is emphasized in the attempt to maximize the sensory information mediated by the joint and musculotendinous afferents to dynamically stabilize the joint.6 Although diminished or altered proprioception and kinesthesia have been documented after A C L injuries, very few prospective studies have examined the effectiveness of proprioceptive training after A C L injuries. Of those prospective studies found in the literature, none involved the A C L reconstructed population. The main purpose of this study was to determine the effectiveness of a proprioceptive training protocol versus a strength training protocol in restoring the neuromuscular function of the A C L reconstructed limb. 2 The second purpose of this study was to determine the relative contribution of isokinetic strength and peak hamstring torque time to functional ability. 3 LITERATURE REVIEW Embryology of the ACL The appendicular skeleton consisting of the pectoral and pelvic girdles plus the bones of the upper and lower limbs appears quite early in the embryologic process.7 The limbs begin to appear as small elevations from a slight lateral ridge extending along each side of the trunk. 8 The forelimb buds show first, at the level of the more caudal cervical segments and the hind-limb buds are level with the lumbar and upper sacral segments. Although the upper limbs appear earlier in utero and are relatively more mature, the lower extremities develop more rapidly than the upper extremities at birth and throughout the early neonatal growth. ? The development of the knee joint can first be detected in the fourth week of development, or the 5 mm stage of human embryo, from a concentration of mesenchyme. ^  The development of the knee joint is quite rapid, and by six weeks, a very obvious knee joint is discernible. The cruciate ligaments of the human knee joint first appear as condensations of vascular synovial mesenchyme (in the blastoma) at about 6.5 weeks of development, well before joint cavitation occurs, and remains extrasynovial at all times.^ > 7 The A C L begins as a ventral ligament and gradually invaginates with the formation of the intercondylar space. From its earliest appearance, the A C L changes very little to achieve its final form J By ten weeks, the A C L and the posterior cruciate ligament (PCL) are separate from each other and are easily distinguished from one another by the direction of their parallel fibers.3 By the twentieth week of development, the cruciate ligaments resemble those of the adult.3 4 Gross Anatomy of the ACL The A C L is intra-articular but is extrasynovial within its synovial envelope.9 Proximally, it attaches to the posterior aspect of the intercondylar surface of the lateral femoral condyle.7> 9, 10 T n e femoral attachment is shaped like a segment of a circle, with its anterior border straight and its posterior border convex.9 The femoral attachment covers an area of approximately 2 cm2.7 Distally, the A C L attaches anteriorly and laterally to the tibial spine. Portions of the anterior fascicles of the ligament occasionally will blend with the anterior horn of the lateral meniscus. On rare instances, the posterior fascicles will blend with the posterior horn of the lateral meniscus.9 The tibial attachment has a surface area of approximately 3 cm2.? The tibial insertion of the A C L is both larger and stronger than its femoral origin. It has been observed that the insertion sites of the A C L pass through a transition zone before inserting into the bone.3, 11 The abrupt change from flexible ligamentous tissue to rigid bone is mediated through this zone of fibrocartilage and mineralized fibrocartilage, allowing a graduated change in stiffness and a reduction in stress concentration at the attachment sites. The transition zone is divided into four morphologically distinct zones: 1. Zone one is the ligament substance. 2. Zone two represents a region of fibrocartilage. 3. Zone three represents mineralized fibrocartilage. 4. Zone four is bone. Due to the orientation of its bony attachments, the A C L is a three-dimensional fan-shaped structure that runs anteriorly, medially, and distally across the joint from the femur to the tibia. 10 As the A C L courses through the knee joint, it turns on itself in a slight outward (lateral) spiral.9 On average, the A C L is 40 mm in length, 10 mm in width, and 5 mm in thickness.?' 10 The A C L is divided into the anteromedial, the intermediate, and the posterolateral bands (Figure 2.1). 12 5 Figure 2.112 The Tripartite Area of Origin. ( A M = Anteromedian I = intermediate, PL = posterolateral) The anteromedial band is composed of the fascicles arising from the most anterior portion of the tibial attachment and inserting on the most medial and proximal portions of the femoral attachment.^ The posterolateral band is composed of the fascicles arising from the posterior portion of the tibial attachment and inserting on the most lateral and distal parts of the femoral attachment. A l l fascicles of the A C L are taut during the final 30° of knee extension. 12 The fascicles of the anteromedial band are taut principally in flexion while the posterolateral band becomes progressively taut as knee extension occurs. While this designation provides a general scheme as to the dynamics of the A C L throughout the arc of motion, one should keep in mind it tends to oversimplify somewhat. For even though the fascicles of the ligament are summarily divided into three bands, the A C L is actually a continuum of fascicles that provides stability in any position of the knee joint. 12 6 Microanatomy of the ACL The A C L is composed of multiple fascicles of collagen tissue.9> 10 The smallest structural unit is the fibril, 150 nm to 250 nm in diameter, which is grouped together to form fibers of 1 um to 20 um in diameter. Groups of fibers then form the subfascicular units, 100 um to 250 um in diameter, surrounded by an endotendineum. Three to twenty subfascicular units group together to make up a fasciculus, which may range from 250 wm to several millimeters in diameter. The fasciculus is surrounded by an epitendineum and the entire group of fascicles form the A C L that is surrounded by the paratenon. In a study using dogs, humans, and rabbits, Clark and Sidles 13 challenges the details of the organizational pattern mentioned above. Clark and Sidles^ observed only two distinct orders segmentation. The A C L was found to be composed of multiple 20 um width collagen fiber bundles separated by a network of fine fibrils and cells. The bundles then group into fascicles that vary in size from 20 to 400 um in diameter. 7 Vascular Supply of the ACL The ligamentous branches of the middle genicular artery and the branches of the inferior genicular arteries provide the blood supply to the A C L . 14 These blood vessels richly endow the synovial fold that covers the A C L and form a "web-like network of periligamentous vessels that ensheath the entire ligament." 14 The synovial fold covering originates from the posterior inlet of the intercondylar notch and extends to the tibial insertion of the ligament, where it joins the synovial tissue of the joint capsule distal to the infrapatellar fat pad. The middle genicular artery enters the upper one third of the A C L and its vascular branches course both proximally and distally along the ligament. The inferior medial and lateral genicular arteries enter the A C L through the medial and lateral fat pads respectively. Smaller connecting branches, derived from the periligamentous vessels, penetrate the A C L transversely and anastomose with a network of endoligamentous vessels.3» 14 These vessels, along with their supporting connective tissues, are oriented in a longitudinal direction and run parallel to the collagen bundles within the ligament. Minimal blood supply is derived from the ligamentous osseous junctions of the A C L A 14 8 Neuroanatomy of the ACL Nerve Supply of the Knee Joint According to Kennedy^, there are two distinct groups of articular nerves for the knee joint. The posterior group consists of the prominent posterior articular nerve, which is a branch of the tibial nerve, and the terminal branch of the obturator nerve. The anterior group consists of articular branches of the femoral, common peroneal, and saphenous nerves. Kennedy^ found the largest and most prominent nerve to be the posterior articular nerve. The posterior articular nerve branches from the posterior tibial nerve above the knee in the popliteal fossa. It then wraps around the popliteal vein and artery before it becomes the popliteal plexus, which penetrates the posterior capsule of the knee joint. Branches of this plexus supplies the posterior capsule, the posterior fat pads, the posterior oblique ligament, the medial and lateral ligaments (MCL, LCL) , the PCL, the A C L , and the posterior parts of the annular ligaments surrounding the lateral and medial menisci. The medial articular nerve is a branch of the saphenous and/or the obturator nerves, and it appears to be distributed to the medial and anteromedial aspect of the fibrous capsule, the M C L , the medial meniscus, the patella ligament, the infrapatellar fat pad, and the medial part of the patellar periosteum. The lateral articular nerve takes a recurrent course after branching from the common peroneal nerve and innervates the capsule of the superior tibiofibular joint, the inferolateral tissues of the knee joint, and the peroneal muscles. Innervation of the A C L The A C L is innervated by the posterior articular nerve.4 Branches of the posterior articular nerve that innervate the A C L course along the periligamentous and endoligamentous vessels, where some of the nerves ramify and send nerve fibers into the collagenous substance of the A C L . Recent morphologic and physiologic investigations have convincingly demonstrated that the A C L contains complex mechanoreceptors of different types that probably make significant contributions to motor control and proprioception. The 9 presence of mechanoreceptors within the A C L has been identified by a number of histological studiesA 5, 15, 16 Kennedy^ found neural elements similar to Golgi tendon organs within multiple clefts in the tibial origin of the A C L and in its richly vascularized synovial covering. Similarly, Schultz et al.16 found fusiform corpuscles resembling Golgi tendon organs at the surface of each cruciate ligament beneath the synovial membrane. More recently, Schutte et al.5 identified three morphologically distinct mechanoreceptors in the A C L : 1. Two types of Ruffini end-organs. 2. Pacinian corpuscles. 3. Free nerve endings. Schutte et al.5 concluded that the specialized receptors and free nerve endings within the A C L constitute 1% of the area of the A C L while Zimny et al.15 suggested that they compromise 2.5% of the area. Mechanoreceptors located in joint capsules and ligaments are thought to serve at least two major functions. 1 ?> 18 One, they provide important sensory feedback necessary for proprioception and kinesthesia. Two, they play a role in the initiation and coordination of appropriate muscle firing patterns, or reflexes, to stabilize the knee joint during functional activities. The exact mechanism by which muscular reflexes are initiated by mechanoreceptor feedback is a subject of debate. Mechanoreceptors are found in the A C L as well as the PCL, medial and lateral collateral ligaments, menisci, patellar tendon, and infrapatellar fat padA 19, 20 Joint afferents are traditionally classified into four different groups according to their fiber diameter and conduction velocities.^! The four groups are: 1. Group I afferents have diameters from 10-18 um and conduction velocities greater than 60m/s. 2. Group II afferents have diameters from 5-12 um and conduction velocities of 20-70 m/s. 3. Group III afferents have diameters from 1-7 wn and conduction velocities of 2.5-30 m/s. 10 4. Group IV afferents have conduction velocities less than 2.5 m/s. Group I fibers emanate from Golgi tendon organlike endings. The Group II fibers originate from the Pacinian corpuscles and Ruffini endings. Group III and IV myelinated and unmyelinated afferents originate from free nerve endings.21 Mechanoreceptors can be divided into four types:^ 1. Type I mechanoreceptors are Ruffini-like receptors. 2. Type II mechanoreceptors are Pacinian-like receptors. 3. Type III mechanoreceptors are similar to Golgi tendon organs. 4. Type IV mechanoreceptors are free nerve endings that convey pain. Fast adapting mechanoreceptors are the most sensitive indicator of change in the tension of the ligament. 5 They identify acceleration at the initiation and termination of movement. Slow adapting mechanoreceptors are capable of continuous activity that can be quantitatively altered in response to variation in the tension of the ligament. These receptors identify motion, position, and angle of motion. 5 Ruffini end-organs (Figure 2.2) are mechanoreceptors that have a very low threshold and respond to slight changes in tension within the ligament. 5 They are capable of a prolonged discharge due to their slow adaptation, and therefore, play a role in conveying the proximity of the joint to its limit of motion in flexion and extension.^ The second morphologically distinct Ruffini end-organ identified in the A C L by Schutte et al^ resembled a Golgi tendon organ (Figure 2.2). Golgi tendon organs are also characterized by slow adaptation but have a high threshold for excitation. They respond at the extremes of motion and may be responsible for mediating protective reflex arcs.6 Ruffini end-organs and Golgi tendon organs have been described as the dominant receptor types in joint capsules.22 The Pacinian corpuscle (Figure 2.2) is.a.fast adapting mechanoreceptor that is activated by movements of the knee joint, regardless of position. 5 These receptors have a low threshold for excitation and are inactive in the immobile joint or when the joint is moving at constant speed, but are responsible for signaling acceleration and deceleration 11 of the joint. Thus, Pacinian corpuscles are considered solely as dynamic mechanoreceptors.23 Due to the information provided by these mechanoreceptors in the A C L , the central nervous system can appreciate speed, acceleration, direction of movement, and the position of the knee jointA 5 Free nerve-endings (Figure 2.2) constitute a pain-receptor system for the tissues of the knee joint. An individual's apparent insensitivity to immediate pain after an A C L injury may be explained by the small concentration of free nerve-endings found in the ligament. ^  A . Ruffini B. Golgi C. Pacinian Tendon Organ Corpuscle D. Free Nerve Ending End-Organ Figure 2.2 ^ Mechanoreceptors in the Anterior Cruciate Ligament 12 Function of the ACL Mechanical Role According to Ellison and Berg?, to understand the function of the A C L , an understanding of knee joint kinematics is essential. Arthrokinematics refers to the study of movement at a joint that accompanies physiological range of motion.24 Osteokinematics refers to the study of the movement of a bone in space without regards to the effects at the joint.24 Since the knee is a modified ovoid joint, there are two degrees of freedom of movement. The ranges of motion that occur at the knee joint are flexion, extension, internal rotation, and external rotation. Flexion and extension occur about a transverse axis in the sagittal plane while internal and external rotation occurs about a vertical axis in the transverse plane.? In the sagittal plane, flexion and extension of the knee occur by a combination of rolling and sliding motions. The arthrokinematics in the sagittal plane is due to the shape of the femoral condyles.? Anteriorly, the femoral condyles are flatter and oval-like in geometry. Posteriorly, the condyles are more curved or spherical. Due to the geometry of the femoral condyles, rolling predominates in the early degrees of flexion while sliding prevails in the latter degrees of flexion.? In the horizontal plane, internal rotation and external rotation occur as a result of spinning about a vertical axis. Of special interest is the screw home mechanism, which entails external rotation of the tibia relative to the femur with terminal knee extension.? The screw home mechanism gives the knee maximum joint stability in preparation for body weight loading in the single-leg stance phase of the gait cycle.? This mechanism is a result of the inequality of bony geometry observed in the sagittal plane. Due to the smaller surface area of both the lateral tibial plateau and lateral femoral condyle, the lateral side of the knee joint achieves congruency at 30° of flexion.8 A continuation of simple extension is therefore impossible, however internal rotation of the femur about a vertical axis is possible. As this occurs, full congruence of the lateral condyle is delayed and continuing extension is possible. Therefore, internal rotation of the femur (on the tibia) must occur at 30° of flexion to achieve full knee extension. 13 With an understanding of knee joint kinematics, the function of the A C L as a passive restraint can be better appreciated. The major function of the A C L is to prevent excessive anterior tibial translation in various degrees of flexion. 12, 25 According to Butler et al.25, the A C L provides 85% of the restraining force to anterior tibial displacement at 30° and 90° of knee flexion. In flexion, the anteromedial band of the ligament is the most important in preventing anterior tibial translation, whereas the posterolateral band contributes mostly to stability in extension.?' 12 The A C L also prevents hyperextension of the knee, limits excessive tibial rotation (internal rotation more so than external rotation), and is a secondary restraint to both valgus and varus stresses.?' 12' 26 Finally, the A C L enhances the screw home mechanism as the knee approaches terminal extension.? Stability of the knee in terminal extension is of maximal importance, especially to athletes who participate in activities involving vigorous cutting, jumping and rapid deceleration. If the A C L is ruptured or plastically elongated, the synchronous rolling and gliding of the knee joint becomes abnormal and the knee becomes vulnerable to degenerative changes secondary to unguided femoral motion.?' 27 A significant injury to the A C L often results in anterolateral instability of the knee, especially if the lateral collateral ligament and the posterolateral complex of the capsule are also damaged.3' 28 Thus, decreased stability of the knee associated with functional disability is a common feature of the A C L deficient knee.29 According to Torzill et al.30, there are four factors that affect joint stability: 1. Passive soft tissue structures. 2. Active muscle force. 3. Externally applied forces and moments, as occur during normal physiologic functions such as gait and stair climbing. 4. Geometry of the articulating surfaces. Traditionally, the mechanical role of the A C L as a passive restraint against excessive joint displacements has been emphasized. The integrity of the A C L plays an important role in determining the passive joint stability of the knee. Passive joint stability 14 is determined by the joint geometry and the mechanical properties of the tissues within and around the joint. According to Noyes et al.31, a complete failure of the human A C L occurs at stress levels of about 1725N, while bone avulsions and ligamentous microfailures occur at lower stress levels. However, it has been demonstrated in vitro that during strenuous activities such as downhill skiing, the load on the knee joint and its ligaments may substantially exceed potential injury levels.32 Consequently, the knee joint must rely on mechanisms other than the mechanical properties of its ligaments to maintain functional joint stability during strenuous physical activities and to prevent injuries to the ligaments. Functional joint stability can be defined as the condition in which the joint is stable and does not give symptoms during physical activities. The lack of a strict relationship between the passive stability and the functional stability of the knee joint has been established in several studies.33, 34 Functional joint stability is determined by the interaction of a number of factors, including the passive restraints of the joint, the joint geometry, the friction between the cartilage surfaces, and the load on the joint caused by compression forces resulting from body weight and the muscles acting on the joint.21 According to Johansson et al.21, the load imposed on the joint is one of the most important determining factors of functional joint stability. Using cadaver knees, Markolf et al.35 demonstrated that by increasing the load on the knee joint greatly enhances the anterior-posterior, medial-lateral, varus-valgus, and torsional stiffness. This implies that when the load on the knee joint is increased, the application of a given external force will result in less tibial displacement, and thus, functional joint stability is increased. Markolf et al.35 concluded that joint load, whether generated by gravitation, dynamic, or muscular forces, is an important protective mechanism that avoids ligament strain. Similar results were produced in a study conducted by Torzill et al.30 T j s m g n m e intact cadaveric knees and four A C L deficient knees, anteroposterior translation was measured at 0°, 15°, 30°, 45°, and 90° of knee flexion after the application of an 15 anteroposterior force of 100 N , a joint compressive load of 0 N , 111 N , 222 N , 333 N , or 444N, and a quadriceps force of 0 or 133 N . The investigators found a significant decrease in total anteroposterior translation after the application of a joint-compressive load, a quadriceps force, or a combination of both. This occurred for all combinations of joint-compressive load and quadriceps force, at all flexion angles tested, and in both the intact and the A C L deficient knees. Thus, the activity of the muscles surrounding the knee joint can have a direct impact on functional joint stability. Thus, although a precise diagram of the neurocircuitry of dynamic knee joint control is not agreed on by investigators, there is clear evidence that the neuromuscular system is capable of altering the strains imposed on the passive restraints of the knee and contributes to functional joint stability during physical activities.36-40 Numerous studies have demonstrated that while a hamstring contraction reduces A C L strain, a quadriceps contraction increases A C L strain.38, 41-45 Renstrom et al.38 observed that hamstring activity decreased A C L strain relative to the passive normal strain at angles greater than 60° whereas isometric quadriceps activity significantly increased the strain within the A C L at angles of 0° to 45°. A in vivo study conducted by Beynnon et al.44 measured the strain in the intact A C L during active range of motion (ROM) of the knee, with or without an ankle weight, and during isolated isometric quadriceps and hamstring contractions. Strain values were also measured during simultaneous contractions of the quadriceps and the hamstrings. Active R O M was found to produce an A C L strain response that depended on the angle of knee flexion and the level of quadriceps activity. In the study, the A C L strain values were at a minimum with 90° of knee flexion and were at a maximum with terminal knee extension during active R O M . With the addition of an ankle weight, the increase in the activity of the quadriceps resulted in a greater A C L strain value. Isometric quadriceps contractions at 15° and 30° of knee flexion were shown to increase the A C L strain whereas there were no changes in the A C L strain value relative to the relaxed state at 60° and 90° of flexion. Isometric hamstrings contractions did not produce a change in ligament strain at any flexion angle. Simultaneous contraction of the quadriceps and the hamstrings produced an increase in strain value only at 15° of flexion. Therefore, resisted knee extension 16 exercises between 0° and less than 60° of knee flexion with the tibiofemoral joint unloaded (open kinetic chain) should be avoided in the early stages of rehabilitation after A C L reconstructions.44 According to Borsa et al.46, the functional instability that occurs after an injury to the A C L is due to the combined effects of excessive tibial translation and a lack of reflex stabilization of the muscles surrounding the knee joint. The lack of reflex stabilization of the knee is thought to be secondary to a diminished sensory feedback mechanism. Thus, the A C L is thought to have a sensory role as well as its well-known mechanical role in the knee joint. Sensory Role Since the discovery of mechanoreceptors in the A C L , it has been suggested that sensory information from the ligament assists in providing dynamic stability to the knee joint by contributing to proprioception and kinesthesia, and thus, to muscular coordination and reflexesA 5, 18, 29 j^e role proprioception plays in the prevention and progression of injury is currently receiving a great deal of attention in the literature.^7* 48 The results of numerous studies support the existence of a sensory role for the ACL.40, 49-51 Somatosensory evoked potentials (SEPs) were recorded by Pitman et al.^l at the cerebral cortex upon electrical stimulation of the A C L . SEPs measure the electric potentials evoked in the cerebral cortex upon stimulation of a peripheral neuroreceptor. The results of the study provide direct evidence for the presence of a proprioceptive function of the A C L . Krauspe et al.50 demonstrated the afferent fibers of the A C L were activated by local mechanical stimulation of the A C L and by movements of the knee joint. In this study, passive motion alone was suffice to elicit impulse generation, but at full extension and combined with rotation, a higher degree of stimulation was generated. Finally, by using a dynamic histological technique, Madey et al.^2 established a pathway between the spinal dorsal ganglion and the posterior capsule of the knee and the A C L . 17 The Neurophysiology of Proprioception and Kinesthesia Both proprioception and kinesthesia, first introduced by Sherrington^ and Bastian545 respectively, are specialized types of the sense of touch. Both are involved in the control of movement and posture. Proprioception and kinesthesia arise from discharge of sensory receptors in the skin, musculotendinous units, ligaments, and joint capsules, from centrally generated motor commands, and from interactions between these afferent and efferent signals.6> 55 These sensory receptors transduce mechanical deformation to a neural signal that modulates conscious and unconscious responses. Visual and vestibular centers also contribute afferent information to the central nervous system (CNS) regarding body position and balance.47 Proprioception and kinesthesia are hypothesized to contribute to the motor programming for neuromuscular control required for precision movements and muscle coordination.0^ 47 Thus, proprioception and kinesthesia play a role in functional joint stability. According to Gandevia55j the two terms, proprioception and kinesthesia, have become almost synonymous and are used to cover almost anything concerned with the control of movement. However, proprioception and kinesthesia are two distinct terms. Proprioception refers to the unconscious sensation contributing to movement control, or reflex control of movement without the need of conscious awareness.53 Kinesthesia involves conscious awareness; these include sensations of limb movement and position, sensations of force and heaviness, and sensations of the timing of muscular contractions.54 Collectively, kinesthesia and proprioception encompass a group of sensations.55 First is the traditional sensation of position and movement of the limbs and trunk. Second, there are sensations related to muscle force, including effort, tension, heaviness and stiffness. Third, sensations exist for timing of muscle contractions. Fourth, there is a sensation of body posture and size as part of a "schema" encompassing more than one joint. Other definitions of proprioception and kinesthesia, often more simplistic, can be found in the literature. For example, Barrack et al.56 refers to proprioception as the conscious awareness of the limb position in space and kinesthesia as the awareness of 18 joint motion. Beard et al.29 divided proprioception into three components: (1) static awareness of joint position, (2) kinaesthetic awareness, and (3) closed-loop efferent activity that is required for the reflex response and the regulation of muscle stiffness. For the purpose of this thesis, unless otherwise specified, the term proprioception will be used to encompass both the conscious and unconscious sensation of joint movement and joint position. The sense of proprioception is made possible by mechanoreceptors that transduce modulated neural signals which are transmitted via cortical and reflex pathways through different nerve fiber types that are specific for each modality.22 An increased stimulus of deformation is coded by an increased afferent discharge rate or an increased population of activated receptors.57 The sense of proprioception is proposed to be carried along by large-diameter myelinated nerve fibers that have high conduction velocities, while the sensations of pain are conducted by smaller diameter nerve fibers that have slower conduction velocities.22, 57 Sensory information from receptors in the joint capsules, ligaments, skin and musculotendinous units are transmitted via articular nerves through the posterior roots to the spinal cord and ascend through the posterior column (dorsal-column-lemniscal) where they are then conducted to the post-central gyrus of the cerebral cortex.22 it has been noted there is a disproportionately high cortical representation for the lower limb, which has been taken as evidence of the importance of proprioceptive input from the lower extremity.22 Conventionally, the sensory units in the limbs that are responsible for proprioception and kinesthesia are divided into three categories: skin, joint, and muscle. ^ 7 Each of these categories of afferents is considered to be independent and act in parallel with the others. However, it should be noted that within the area of a joint, tendons and periarticular connective tissue and capsule may not be separable as to function when loads are transmitted across them. Thus, proprioception and kinesthesia appear to rely on simultaneous activity in a number of types of afferent endings. 19 However, in the literature, there is the continual debate as to which group of receptors are the main source of proprioception, and specifically, of kinesthesia. Skin Mechanoreceptors It is generally agreed upon that muscle receptors are the main contributors to kinesthesia and proprioception in the literature. There is much debate among investigators on whether joint and skin receptors have only a facilitatory function or actually contribute to proprioception and kinesthesia. Edin and Abbs^S has demonstrated that a majority of the low threshold cutaneous mechanoreceptors located on the back of the human hand respond to movements at one or more nearby joints. Cohen et al.^9 demonstrated that active arm movement evoked complex patterns of tactile activity in the anterior somatosensory cortex of awake, behaving, primates. These results has added impetus to the argument that skin receptors may play an important role in kinesthesia and motor control. Clark et al.60 tested position sense in the human knee joint, with anesthesia of skin, joint, or skin and joint together. The investigators blocked cutaneous afferents by anesthetizing a 15 centimeter band of skin around the knee and blocked joint afferents by injecting a local anesthetic into the joint space. Clark et al.60 concluded that joint awareness does not depend on sensory input from either receptors in the joint or the skin around the joint. The insignificant results may be secondary to the lack of total sensory loss produced. The problem in eliminating just cutaneous afferents is that in most joint, it is difficult to anesthetize the skin alone. Edin and Johansson0^ conducted a study examining the relative importance of cutaneous inputs in kinesthesia. The study produced an unexpected finding that skin afferent inputs may have precedence over muscle spindle inputs with regard to kinesthetic and proprioceptive processing as expressed by perception and motor responses. Edin and Johansson0^ proposed that the relative importance of cutaneous input to kinesthesia and proprioception have been underestimated for the following reasons: 1. "The exquisite sensitivity of skin afferents to skin deformation has only been recently quantified and has not previously been taken into account."(p.247) 20 2. "The possibility that skin afferents may provide information about joint configuration by means of encoding patterns of skin deformation rather than as simple "joint angle transducers" has not been considered."(p.247) 3. "Distant mechanical effects of stretching muscles have not been taken into proper account when interpreting experimental data related in particular to the involvement of muscle spindle inputs in signaling imposed joint movements."(p.247) Thus, Edin and Johansson^! concluded that although skin mechanoreceptors are not the only source of information about joint position, the importance of these receptors has been underestimated in the past and that, under some circumstances, the information provided by these afferents has precedence over all other sources of proprioceptive information. Muscle Mechanoreceptors Muscle spindles are distributed throughout the belly of the muscle and send information to the nervous system about either the muscle length or rate of change of its length. Muscle spindles have a sensory and motor innervation. 55 The sensory component consists of one primary and several secondary spindle receptors. The primary spindle endings innervate both the nuclear bag fibers and the nuclear chain fibers. The secondary spindle endings usually innervate only the nuclear chain fibers. Primary spindle endings give rise to group la afferents and secondary endings to slower conduction group II afferents. Primary spindle afferents are much more sensitive to dynamic stimuli such as stretch or tendon taps than secondary afferents. The high sensitivity of primary spindle endings means that their exact intramuscular location modifies their response, as all regions of a muscle do not lengthen uniformly. 5 5 The motor innervation consists of fusimotor axons, or gamma-motorneurons, which innervate several spindles; and the less frequent skeletofusimotor axons, or beta-motorneurons, which innervate intrafusal and extrafusal muscle fibers. 5 5 Fusimotor axons are distinguished by their innervation of particular types of intrafusal muscle fibers and their action on primary and secondary afferents. There are two types of fusimotor axons, dynamic and static. Dynamic fusimotor axons innervate bag] (dynamic) fibers while static fusimotor axons innervate bag2 (static) fibers and nuclear chain fibers 21 Golgi tendon organs are uniquely placed to record muscle unit forces.-5-5 Over 90% of a tendon organ is located at the muscle junctions, with the remainder in the tendon. This encapsulated receptor is "in series" with a few muscle fibers and "in parallel" with others inserting around it. The number of muscle fibers inserting into a tendon organ ranges from 5 to 50, with 10 to 20 being common in humans.55 The adequate stimulus for tendon organs seems to be contractile forces acting at the receptor rather than passive stretch. They have a lower threshold for active force than for that produced by passive stretch. Tendon organs are innervated by lb afferents. These lb afferents have conduction velocities that overlap those of primary muscle spindle afferents but with a fractionally lower mean value.55 Since the response of muscle spindle afferents is known to be a function of muscle length, muscle afferents are able to provide an unconfounded, unidirectional signal of joint movement.57 Around 1960, due the results of various experiments, spindle discharges were thought to lack any access to conscious sensation, or kinesthetic awareness.55 Instead, information from muscle spindles were seen as reserved for the unconscious control of movement, or proprioception. However, after a study conducted by Goodwin et al.62 in 1972, the pendulum swung, and currently, muscle afferents are generally believed to be the principal source of kinesthetic and proprioceptive signals. In the study conducted by Goodwin et al.62 i n 1972, it was demonstrated that when vibration of 100 hertz was applied to the tendon of the biceps or the triceps muscle, an illusion of elbow motion was created. Subjects in the study consistently thought the elbow was the in position that it would have assumed if the vibrated muscle had been stretched. Thus, excitation of muscle afferents, by vibration of the muscle, caused sensations of joint movement and position. In another study conducted by Goodwin et al.63 ; where the joint and cutaneous afferents of the index finger or the whole hand were inactivated by anaesthesia, kinesthetic sensations were still evoked by passive joint movements. It was also noted in the study that when subjects tensed their muscles acting around the index finger, their kinesthetic awareness was enhanced. 22 Grigg et a l . 0 4 studied the individual's ability to detect passive joint movement and duplicate positions of the hip shortly after total hip replacement. The results indicated the ability to detect joint position was retained after total hip replacement. This lead the authors to conclude that joint position sense is not totally dependent on the joint capsule or the surfaces of the hip joint, and that extracapsular components (i.e. muscle receptors) exist. In summary, there is strong evidence that muscle afferents play an important role in proprioception and kinesthesia. However, this role is probably not a simple one. For example, in the study conducted by Goodwin et al.63 where kinesthesia persists after paralyzing joint afferents but preserving muscle afferents, the resulting kinesthesia was deteriorated from the normal. Moberg65 also demonstrated that when the hand was anesthetized and finger muscle afferents were unanesthetized, kinesthetic acuity in the fingers was much lower than normal. Thus, it appears that in order to have normal proprioceptive sensations and normal acuity, it is necessary to have sensory systems other than those in muscles functioning simultaneously. In addition, in studies utilizing longitudinal vibration of a muscle tendon is probably no a selective activator of muscle spindle afferents. Activation of muscle spindle afferents by vibration may also increase muscle tone, which may ultimately affect joint afferents owing to the mechanical coupling between some muscles and joint capsules. Joint Mechanoreceptors Joint motion will stretch the joint capsule on one side of the joint and may also compress it against the underlying bone.57 Structures on the other side of the joint will be unloaded by the rotation. Ligaments may also be loaded in rotational movements of the joint. Therefore, it is possible that mechanoreceptors originating in these tissues may have a role in proprioception and kinesthesia. It is still a popular belief that the contribution of joint afferents to proprioception is important only at limits of joint motion. 18, 21, 23 jn[s belief originated from most of the early investigations that found a majority of joint afferents to signal near full flexion and extension, while few units have their active range in the mid-position (i.e. midrange 23 units).18, 21, 23, 55 Thus, it was concluded that true joint receptors do not contribute to proprioception and kinesthesia since they could not inform the CNS about joint angle throughout the full range of movement. This popular hypothesis would explain the results of the study conducted by Grigg et al.64 where the hip joint capsule was surgically removed and deficits in kinesthesia were not observed when the joint was moved through intermediate positions. However, in study conducted by Ferrell66 in 1980 using feline knees, a higher proportion of midrange was observed than previously in the posterior articular nerve of the knee, located in the capsule as well as in the knee joint ligaments. Ferrell66 also observed a higher frequency of ligamentous afferents among the midrange units than among the end range units in the knee joint capsule. Midrange units or afferents probably arise from Ruffini endings.21 Thus, it appears that joint receptors may contribute to proprioception and kinesthetic awareness throughout the full range of joint motion. In a study to assess the role of joint receptors in human kinesthesia when intramuscular receptors can not contribute, Ferrell et a l .^ 7 concluded that the discharge of joint receptors can produce perceived signals of joint motion and that under normal conditions these receptors may duplicate the kinesthetic input from muscle spindle afferents. Kinesthetic acuity was tested at the distal interphalangeal joint of the middle finger when the hand was positioned so that the joint was effectively disengaged from its muscular attachments. Despite the new found evidence for joint mechanoreceptors contributing to proprioception and kinesthesia, the popular belief is still that muscle afferents play the predominate role in proprioception and kinesthesia. According to Grigg et a l .^ 7 , the following reasons support the lack of major proprioceptive roles for joint afferents, especially those originating from ligamentous structures such as the A C L : 1. "There is a scarcity of nerve endings. While there are relatively large numbers of afferent neurons having endings in the joint capsule, there appear to be not more than a few ligament afferents that originate in the ACL."(p.l2) 24 2. "Ligament afferents would function as nonspecific limit sensors. A C L afferents are not active when the knee is in a neutral position. They have been shown to respond to stimuli that would cause tensile loading of the A C L : rotations into the limit of flexion, extension, and internal or external rotation. Since ligament tension can be caused by a number of different types of rotations, the mechanoreceptors in them are poor candidates to subserve proprioceptive sensations of movements in specific directions. While neurons innervating the collateral ligaments are less well documented, they also appear to be tension sensitive, and thus to be subject to the same reservations as those stated for A C L afferents."(p!2) However, with the discovery of midrange afferents within the knee joint capsule and ligaments, the debate regarding the significance of joint afferents in proprioception and kinesthesia continues. In addition, the small number of midrange units found in joints does not necessarily mean that the midrange units have less influence on proprioception and kinesthesia. The strength of such an influence can not be assumed to be directly proportional to the number of units engaged.21 Many studies have demonstrated that both proprioception and kinesthesia awareness of the knee are altered after an injury to the ACL.29, 40, 68-72 According to Lephart et al.47, kinesthesia awareness is essential for proper joint function in sports, activities of daily living, and occupation tasks, whereas proprioception is essential in modulating muscle function and initiating reflex stabilization of joints. 25 The Role of Proprioception in the Function of the Knee Joint Diminished proprioception in joints are secondary to injury. Decreases in proprioception may lead to further instability secondary to poor coordination of the dynamic stabilizers.4 This ultimately may initiate a vicious cycle of reinjury secondary to proprioceptive deficiency. 21, 47 since proprioception is thought to be involved in mediating the control of muscular coordination and reflexes, it is the postulation that these mechanisms of the knee depend, to a certain degree, on the proprioceptive contribution of the ACL. 4> 29, 68 Kennedy4 proposed that a decrease in the proprioception of the knee, secondary to an injury of the A C L , may contribute to progressive knee instability (Figure 2.3). After studying professional ballet dancers for knee joint laxity, Barrack et al.56 concluded that decreased joint position sense (reproduction of passive positioning) may indicate below normal protective reflexes, which may increase the chances of acute or chronic injuries and degenerative changes to occur. INITIAL KNEE INSTABILITY REPETITIVE MAJOR FAILURE OF AND MINOR INJURIES MECHANORECEPTOR FEEDBACK LOSS OF REFLEX MUSCULAR SPLINTING Figure 2.3 4 Hypothesized Neurogenic Contribution to Progressive Knee Instability Protective Mechanisms Traditionally, joint afferents have been regarded as "limit detectors", responsible for triggering protective reflexes when the joint is threatened by hyperextension, hyperflexion, or excessive rotation.21 However, this traditional view contains considerable conceptual problems, since ligamento-muscular protective reflexes do not 26 seem to be able to act in time to protect the ligaments from injury unless the threatening event is very slow.73, 74 j n a s t u dy conducted by Yasuda et al.74} ^ w a s demonstrated that following a lateral impact, elongation of the M C L and A C L reached peak values by 70 ms. This finding lead to the conclusion that contraction of the leg musculature would not protect the M C L and A C L from injury when a lateral impact load is applied to the knee. The results of the study conducted by Yasuda et al.74 a r e m agreement with the results obtained by Pope et.al.73 m m e n - study, Pope et al.73 reported that 34 ms would elapse between the loading of the M C L and ligament rupture, while 89 ms would pass between ligament loading and the initiation of a ligamento-muscular protective reflex. The calculation of the latter latency seems to be accurate, as it agrees well with the findings of Petersen and Stener753 who showed that a tap at the site of a partial rupture of the M C L activated the sartorius muscle with a latency to the E M G of 76 ms. According to Johansson et al.21, this "time argument" has been widely used to place the mechanical properties of the knee joint ligaments among the most important factors subserving joint stability. However, this view disregards the crucial point in the discussion about the importance of muscles in the control of knee joint stability: the state of the changeable muscle stiffness at the time of the displacement, which will to a considerable degree determine the load on the joint and thereby the actual stability of the joint at the time of the trauma.21 Stiffness is defined as the ratio of force change to length change.23 Stiffness of a muscle can be described as two components: 1. Intrinsic muscle stiffness. 2. Reflex-mediated muscle stiffness. According to Johansson et al.21, the intrinsic muscle stiffness is determined by the viscoelastic properties of the muscle and the existing acto-myosin bonds. Intrinsic muscle stiffness depends greatly on the existing acto-myosin bonds or the degree of muscle contraction in a given moment. Therefore, intrinsic muscle stiffness is partly the result of the preceding reflex-mediated muscle stiffness. Reflex-mediated muscle 27 stiffness is determined by the excitability in the gamma-motorneuron pool, descending commands, autogenic and heterogenic reflexes. The variability in the overall muscle stiffness during dynamic condition is largely due to modulation of the reflex-mediated component. Thus, when the knee is subjected to force threatening to displace it, there are two possible ways in which the muscles can protect the knee from displacement-^, 49, 76-78-1. By a preparatory (preset) increase in the muscle stiffness, which will increase the load on the knee and thereby enhance the functional stability of the joint. 2. By a direct reflex counteraction of the applied force or the resulting displacement. Sensory information from ligamentous afferents are hypothesized to contribute to both mechanisms. However, since simple protective reflexes have been shown to be too slow to protect the joints and associated ligaments during fast perturbations, the main contribution of ligamentous afferents to functional joint stability may be a continuous preparatory adjustment of intrinsic muscle stiffness though the reflex-mediated component.21 Maintenance of Joint Integrity In a study conducted by Barrett et al.79s the investigators examined the question, "To what extent, therefore, does lack of good proprioception predispose to the development of degenerated change?"(p.53) Three groups were involved in this study: Group I - Humans with normal knees. Group II - Humans with osteoarthritic knees. Group III - Humans with replaced knees. Results demonstrated that static awareness of joint position deteriorated with age and was further impaired at all ages in osteoarthritic knees. Individuals with total knee replacements exhibited greater accuracy than did those with osteoarthritic knees. Barrett et a l . 7 ^ concluded that though loss of static joint position sense may be a consequence of the process of osteoarthritis, it may equally be a primary factor in the initiation of joint deterioration. In a study using dogs, O'Connor et al.^O examined the effects of denervation of the of the knee joint. Four experimental groups were used: 28 Group A - Articular neurectomy was carried out. Group B - The A C L was cut, but the knee was left normally innervated. Group C - The A C L was cut, and the knee was also denervated. Group D - Control group. Several criteria were used to judge the effect of joint denervation. O'Connor et al.80 observed the locomotion of the dogs and they sacrificed the animals and looked for histological evidence of osetoarthritis. The results of the study were: Group A - The dogs demonstrated no signs of locomotor problems and they did not develop arthritis. , Group B - The dogs limped until they were sacrificed. They developed arthritis. Group C - The dogs limped until they were sacrificed. They developed the most severe arthritis. Three of the 15 dogs in this group developed subchondral fractures. According to O'Connor et al.80, the results show that an animal uses proprioceptive information from joint afferents in maintaining the integrity of the joint. Through proprioceptive feedback, the animal with the unstable joint can learn to adopt movement strategies that minimize tissue stress and potentially damaging stimuli. The results of the study suggest that when proprioceptive feedback is absent, damage to the joint occurs. Muscular Co-Contraction Numerous studies have indicated the hamstrings as essential antagonists in maintaining joint stability in the extension motion.^?, 81-83 Draganich et al.82 demonstrated that there is co-activation of the hamstrings with the quadriceps during monoarticular quasi-static knee extension. The E M G signals of all of the knee flexors and extensors increased with increasing loads on the ankle, and also increased as the knee extended. This observation supports the hypothesis that the activity of the hamstrings, combined with the restraint of the A C L , acts to prevent excessive anterior displacement of the tibia during functional activities. Baratta et al.81 concluded that co-activation of the hamstrings during maximal effort, slow isokinetic contractions of the quadriceps is necessary to aid the A C L in 29 maintaining joint stability, equalizing the articular surface pressure distribution, and regulating the joint's mechanical impedance. Furthermore, a reduced co-activation pattern of the antagonist may increase the risk of A C L damage.81 The study included three groups of subjects: Group 1 - Nonathletic and healthy subjects, none of whom were active in any sports or exercise program. Group 2 - High performance athletes. None of the subjects in this group had performed any hamstring exercises prior to the test. Group 3 - High performance athletes. A l l subjects in this category routinely exercised their hamstring muscles (i.e. hamstring curls). Baratta et al.81 observed that high performance athletes in Group 2 with hypertrophied quadriceps demonstrated strong inhibitory effects on the hamstring coactivation patterns when compared to Group 1, while those in Group 3 had a coactivation pattern similar to Group 1. Hagood et al.83 observed that an increase in limb velocity into knee extension resulted in an increase in the contribution of the hamstrings, especially during terminal extension. The authors explained this increase in hamstring activity as a way to provide the limb with "dynamic braking such that it does not exceed its physiological range of motion limits."83 Since A C L strain significantly increases in the terminal range of knee extension, it appears the hamstrings may have an ACL-emulating role. 3? Finally, the literature has shown that specific training can improve the contributions of the hamstrings to knee joint stiffness as well as reducing laxity, thereby reducing the risk of ligamentous damage. 81 The literature suggests that the mechanoreceptor system of the A C L may be one of the mechanisms responsible for reflex co-activation of the hamstrings with the quadriceps.4' 5, 15, 18 i f m e A C L is solely responsible for providing the sensory information required for the co-activation of the hamstrings, this muscular activity would be absent in A C L deficient knees. In a study conducted by Solomonow et al.37, hamstring activity pattern was monitored during loaded knee extension (quasi-isometric 30 conditions) in two groups: A C L deficient and A C L intact. Activity pattern of the hamstrings was also monitored during mechanical loading of a cat's A C L . It was hypothesized that the A C L might be involved in providing "reflex sensory information regarding ligament loading or deformation that may request the muscles to assist in maintaining joint stability as conditions arise when the ligaments are overloaded."^7 Solomonow et al.-*7 demonstrated that mechanical loading of a cat's A C L resulted in an increase in the electromyograph (EMG) activity of the hamstrings. This increase in the E M G activity of the hamstrings was evident only with the application of a high mechanical load (130N). Similar responses were elicited in both the A C L intact and A C L deficient knees during loaded knee extension, indicating the existence of an alternative reflex arc, unrelated to the A C L receptors, that was available to maintain joint integrity. Solomonow et a l . ^ 7 proposed that a direct reflex arc exists from the A C L to the hamstrings with the purpose to maintain functional joint stability. In addition, a secondary reflex arc exists from the receptors in the muscles or joint capsule, providing activation of the hamstrings upon knee instability. This secondary reflex arc has a longer response time than the direct reflex arc, as well as having a simultaneous inhibitory input to the quadriceps, and it can provide sufficient joint stability i f the hamstrings are exercised and are well conditioned.-*7 Solomonow et al.-*7 concluded that the hamstrings are a load regulator of the knee during extension, increasing their activity when the A C L is overloaded or when the "geometric configuration disallows the ligament to perform satisfactorily in maintaining stability."(p.211)^7 It is doubtful, however, that either the direct ACL-hamstring reflex arc or the secondary reflex arc can protect the intact or reconstructed A C L if high loads are applied rapidly A 7& Under rapid loading conditions, the ligament may be loaded and ruptured before sufficient muscle tension to protect the ligament can be generated. Contrary to the findings of Solomonow et al.-*7, in a study using chlorase-anesthesized cats, Pope et al.-*6 failed to produce any reflex hamstring activity with mechanical loading of the A C L (up to 125 N). According to Pitman et a l .^ l , it is difficult to develop a model in the cat, and especially in the human, for accurate and significant 31 mechanical stimulation. Mechanical loads are complex and may be affected by anesthesia or by the duration or type of load. Therefore, as summarized by Solomonow et a l . 7 ^, the sources of antagonist activity are probably the motor cortex, the velocity-sensitive component of the muscle spindle afferents, as well as other receptors in the joint capsule and ligaments. Grabiner et al.84 conducted a study to assess whether an automatic hamstrings excitation could be elicited during maximum effort isometric knee extension performed in minimum elapsed time. Results failed to support the contention of an automatic A C L -hamstrings synergy as proposed by Solomonow et al.-*7. However, contrary to other studies found in the literature, Grabiner et al.84 conducted the study using static knee extension conditions while others have used isokinetic and quasi-dynamic conditions. 32 The Role of the ACL in Neuromuscular Control: Possible Mechanisms In 1944, Ivar Palmar^ speculated that the ligaments supply the CNS input that makes neuromuscular control of the knee joint possible. The exact mechanism by which muscular reflexes and the continuous regulation of muscle stiffness are initiated is a subject of some debate. It is uncertain whether some impulses generated from the mechanoreceptors result directly in stimulation of alpha motor neurons or whether the effect is primarily on the gamma efferent muscle spindle system. It should be noted that the current interest in the neuroreceptor function of the A C L is really a resurgence of interest in a function that has been suggested in the past. The number of studies on skeletomotor effects caused by natural stimulation of sensory endings in the knee joint is limited. Few studies using E M G have indicated that the activity in the hamstring muscle might be altered by increased tension in the A C L . 37, 81 Yet several studies have indicated that ligament stretches with moderate loads (i.e. less than 130 Newtons) do not exert much influence directly on the skeletomotor system.36, 84 Freeman and Wyke86 i n 1967 theorized that Type I and II joint mechanoreceptors project directly onto the muscle spindle system and influence the gamma-motor neurons, which ultimately has an effect on the primary muscle spindle afferents. These investigators concluded that reflexes from articular mechanoreceptors take part in the normal reflex coordination of the muscle tone in posture and movement. Thus, the mechanoreceptors of the A C L , through the gamma efferent muscle spindle system, may participate in the regulation and pre-programming of the muscular stiffness around the knee joint and thereby of the knee joint stiffness. 18, 49 Currently, Johansson and coworkers 18, 21, 23 ^ strong proponents of this theory. After conducting several thorough investigations, Johansson and coworkers^, 21, 23 concluded that modest stretching of the cruciate and collateral ligaments (5 to 40 Newtons) elicits activity in the low-threshold mechanoreceptors.49, 76, 77 Activity in these ligamentous afferents was found to primarily influence the gamma efferent muscle spindle system, rather than causing direct reflex effects on the alpha motorneurons. Johansson and coworkers^, 21, 33 ^ found the effects of the ligamentous receptors on the gamma efferent muscle spindle system to be potent and frequent and may induce major changes in responses of the primary muscle spindle afferents. As the activity in the primary muscle spindle afferents modifies stiffness in the muscle through the reflex-mediated component, the A C L mechanoreceptors may, through the gamma efferent muscle spindle system, participate in the coordination and continuous preparatory adjustment of muscular stiffness around the knee joint during dynamic conditions. Specifically, Sojka et al .? 7 demonstrated that stretching of the posterior cruciate ligament of the ipsilateral knee causes changes in dynamic and/or static sensitivity of the primary muscle spindle afferents to sinusoidal stretching. The changes in the sensitivity of the primary muscle spindles were concluded to be due to reflex actions of stretch/tension-sensitive receptors in the cruciate ligaments onto fusimotor neurons. Similar results were found with stretching of the A C L and the collateral ligaments.49, 76 As proposed by Solomonow et al.-*7, Johansson et al.*8 also concluded there was the possibility of the A C L directly influencing the alpha motor neuron neural pathway, but only at high mechanical loading. 34 Surgical Considerations Relating to Proprioception The significance of proprioception is apparent with expanded research into its role in normal joint function, specifically in dynamic stabilization^ 21, 48 initial work in proprioception focused on the documentation of diminished proprioception in injured joints. Since then, the role of proprioception in the initiation and propagation of injury and in the rehabilitation that follows has been further elucidated. From this, its role in surgery is gradually being defined.^? Preservation of Afferents According to Safran et al.87, the role of surgery in proprioception, and the current role of proprioception in surgery, is unclear due to the lack of scientific research and clinical studies. It is undetermined to what degree are the biomechanics of the knee joint and the vulnerability of its ligaments depend on the innervation of the repaired or reconstructed A C L . Data provided by several investigators, especially Johansson et al.49, suggest that the innervation of the A C L is important for the stability of the joint and for the integrity of the ligament. Thus, the loss of innervation of the A C L may be responsible for the altered biomechanics of the knee and the vulnerability of its ligaments after injury or operative treatment. According to Safran et al.87, operations around the knee joint should be done in such a way that the integrity of its mechanoreceptors and afferent nerves of its surrounding structures, such as the capsule, the collateral ligaments, the fat pad, the synovium, and the perimeniscal tissue, are preserved. The primary goal during surgery would be to save as much sensory function as possible. The recent trend towards arthroscopic surgery of the knee, ankle, and shoulder may have benefits in the reduced compromise of proprioception postoperatively as compared to standard open procedures. 87 with the preservation or restoration of the "sensory" function of disrupted ligament, symptoms such as functional instability and muscle weakness may be avoided.50, 87 Regeneration of mechanoreceptors Another important issue to consider when discussing the significance of proprioception in A C L surgeries is regrbwth, or regeneration, of sensory afferents. 35 Denti et al . -" studied the fate of mechanoreceptors in torn and reconstructed A C L in animals and humans. The study has involved three specific groups: Group 1 - Humans with untreated complete A C L tears. (n=20) Group 2 - Sheep with three types of A C L reconstruction: Bone-patellar-bone graft; Bone-patellar-bone augmented with ligament augmentation device (LAD); Leeds-Keio artificial ligament. (n=T2) Group 3 - Humans with failed A C L (semitendinosus) reconstruction. (n=2) The results indicated that while morphologically normal mechanoreceptors remained in the torn ligament for 3 months post-injury in Group 1, their number gradually decreased after this time, and that there was a total absence of corpuscles and free nerve endings in lesions that were 1 year old. In Group 2, there was the presence of mechanoreceptors in A C L reconstructions with the autologous patellar tendon grafts. Mechanoreceptors were identified at 3 months and continued to be evident up to 9 months after the procedure. No mechanoreceptors, however, were found when the A C L reconstruction was done with an Leeds-Keio artificial ligament. Group 3, which consisted of two subjects with A C L reconstructions using the semitendinous tendon, demonstrated a significant number of Ruffini and Pacinian corpuscles, especially near the tibial insertion. It should be noted that while the A C L reconstructed knees in Group 3 demonstrated significant laxity during clinical and instrumental testing, neither subject reported episodes of giving way or instability. In addition, during arthroscopy, the semitendinous graft was found to be slack and degenerated. Denti et al.33 reasoned that the absence of functional instability in Group 3 may be due to the presence of mechanoreceptors found, which may be supplying useful proprioceptive information even in a mechanically nonfunctional graft. Although Denti et al.33 demonstrated repopulation of mechanoreceptors after A C L autograft reconstruction, whether these sensory receptors recovered their physiological roles was not determined. The absence of mechanoreceptors in those animals with A C L reconstructions using the artificial Leeds-Keio ligament suggests that the use of an artificial ligament is not an appropriate choice for A C L reconstruction. 36 Other histological studies using experimental animal models have supported the notion that repopulation of mechanoreceptors occurs after A C L reconstruction. Tsujimoto et al.88 compared A C L repair with patellar tendon augmentation to excision of the A C L stump with patellar tendon autograft reconstruction in goats. Although more mechanoreceptors were found in the repair and augmentation ligaments, mechanoreceptors were found after both types of surgeries. In the repair and augmentation group, the knee was also more stable, the reconstructed A C L stiffer, and there was improved gait and recovery. Tsujimoto et al.^8 noted that few if any mechanoreceptors were evident prior to 6 months postoperatively and that the quantity of receptors found after surgery was less than that of the normal A C L . Based on the findings of their study, the investigators concluded there may be value in preserving the remnants of the A C L during surgical reconstruction. The functional significance of this repopulation of mechanoreceptors demonstrated is still currently unknown. Preservation of the Natural A C L According to Safran et a l .^ 7 , "the next alternative for promotion of mechanoreceptor function is acute reconstruction of the A C L or other damaged ligaments, with the natural ligament insertion sites attached to the graft." Many studies have found the majority of A C L mechanoreceptors to be located at the insertion sites of the ligament.5> 15, 16, 19 Therefore, it is very likely that by preserving the natural A C L insertion stumps, there is great potential for the original A C L mechanoreceptors to transfer their function to the reconstructed ligament i f they heal under appropriate tension.**7 In addition, these natural A C L stumps may provide the ideal environment for the repopulation of mechanoreceptors within the graft. However, a concern with this approach of sewing stump ends into the graft is the risk of developing excessive tissue at the insertions of the graft.**7 Graft Tension One final consideration in A C L reconstruction relating to knee proprioception is the tension of the A C L graft.**7 Since all mechanoreceptors need to be mechanically deformed or loaded to transmit impulses to the central nervous system, surgical 37 procedures may enhance or restore proprioception indirectly through restoration of appropriate tension of capsulo-ligamentous structures. Proprioception has been demonstrated to be diminished in individuals with knee laxity.89 Despite intensive research in this area, the source and the importance of the new population of mechanoreceptors within the A C L grafts are undetermined. It is possible the receptors supply the A C L graft by regrowth, regeneration, growth from the surrounding tissues, dedifferentiation of other cells, or some other mechanism.87 Also, it has not yet been demonstrated that these mechanoreceptors actually function. Thus, the enhanced proprioception after A C L reconstruction may be due to training or enhanced functioning of other proprioceptors from sources other than the new A C L . Another unanswered question is how much repopulation is needed in the graft to cause a detectable difference in the functional stability of the operated knee.87 38 Current Literature On the Proprioceptive Function of the A C L Although the recent studies on the proprioceptive function of the A C L have enhanced our understanding of why surgical reconstructions do not create normal knee function, the question of what are the effects of reconstruction on the proprioceptive function of the involved knee remains. Current research findings on the effects of A C L reconstruction on knee proprioception provide no consensus.34, 90, 91 j n e discrepancy between studies may be due to the different measures of proprioception and/or kinesthesia used, time after surgery, age of the subjects, and different surgical techniques. In addition, the ultimate significance of statistical differences of proprioception as determined in research studies is unknown. Some studies have demonstrated that A C L reconstruction restores proprioception to a sensitivity equivalent to that of uninjured controls.90? 91 Others have found that proprioception after A C L reconstruction is better than proprioception of A C L deficient knees, but is still diminished when compared to uninjured controls. 34, 71 studies have also found that a proprioceptive deficit exists in the postsurgery knee as compared with the contralateral, uninjured control knee.29 No published studies have compared proprioception in the same subject before and after A C L reconstruction to assess i f a loss or gain of proprioception occurs with surgery. Proprioceptive Studies Proprioception of the knee has been measured by various methods in the literature. Examples of the proprioceptive measures used in the literature are: 1. Detection of passive motion. 2. Reproduction of passive positioning. 3. Relative reproduction of passive positioning. 4. Static awareness of joint position. 5. Standing balance. It should be noted that although many of the investigators state that proprioception is being measured when the above methods are used, kinesthesia rather than proprioception is really what is being measured in most cases. The main purpose of a number of recent studies has been to evaluate the effect(s) of A C L injury and A C L reconstruction on clinical proprioceptive measures. 39 Corrigan et a l . o y measured both the ability to reproduce passive positioning and to detect passive motion of the knee joint in individuals with torn ACLs and age-matched controls. When compared to the controls, both the ability reproduce passive positioning and to detect passive motion in the A C L deficient knees were found to be significantly diminished. Barrack et al.40 studied the ability to perceive passive motion of the knee joint in those with arthroscopically proven complete A C L tears and found the A C L deficient limb to have a significantly higher threshold for detecting a change in joint position compared to the normal contralateral knee. In contrast, the control group showed no significant difference in the perception of passive motion between limbs. Based on the popular postulation that muscle afferents play a strong role in kinesthesia, this decline in awareness after A C L injuries may be attributed to a loss of muscle receptor function, secondary to muscle atrophy. 18, 39, 63 However, multivariate analysis confirmed that the changes observed in kinesthetic awareness of the A C L deficient knee were due to the loss of the A C L rather than other variables such as isokinetic strength, time since injury, and the subject's age. Other studies have also found no correlation between muscle strength and measurements of proprioception.^, 71 Results of these studies seem to confirm that the changes observed in knee proprioception after A C L injuries are solely due to the loss of the proprioceptive function of the A C L . However, the relationship between muscle strength and muscle spindle sensitivity is unclear at this point.92 MacDonald et al.90 studied the perception of passive motion in three groups of subjects: 1. A C L deficient. 2. Hamstring tendon-ligament augmentation device A C L reconstructions. 3. Bone-patellar tendon-bone A C L reconstructions. No statistical significant differences were found between the groups. In agreement with Barrack et al.40, MacDonald et al.90 also found no significant correlation between the threshold of perception of passive movement (TPPM) differences between involved and uninvolved knees and either injury to surgery time interval (ACL reconstructed subjects) 40 or injury to follow-up (ACL deficient subjects). If the loss of the A C L simply results in a loss of passive restraint that eventually leads to stretching of capsular structures, possibly causing a decrease in the response of capsular receptors and thereby contributing to the loss of proprioception, one would expect a significant loss of proprioceptive ability in chronically but not acutely A C L injured knees.40, 90 According to MacDonald et al.90, the results of their study lend support to the theory that proprioceptive loss can be a cause of joint laxity and not just the result of it. Barrett34 studied static awareness of joint position in three groups: 1. Normal knees. 2. A C L deficient knees. 3. A C L reconstructed knees. In the study, the A C L deficient knees had significantly poorer position sense than the normal and A C L reconstructed knees of age-matched subjects. The normal intact knee had the most accurate static joint position sense. More importantly, clinical assessments of A C L laxity and subjective knee scoring systems were observed to correlate poorly with an individual's satisfaction and the functional capability of the knee. Static awareness of joint position, however, was found to correlate closely with the functional outcome of the knee and an individual's satisfaction with the knee. This may explain why those with insufficient ACLs feel their knees are unstable despite minimal laxity observed in clinical assessments. It may be that after an A C L rupture, proprioceptive sensation must arise from other sources in the knee, such as the collateral ligaments and the joint capsule.34, 3? However, because an A C L deficient knee moves in a non-physiological axis, the remaining proprioceptive output is non-physiological and disorganized. Therefore, the individual feels the knee to be unstable due to "cortical interpretation and analysis of knee position that is disturbed."34(p.836) Also, the unphysiological gait of the individual with an A C L deficient knee may cause abnormal loading in the joint and lead to degenerative changes in the long term.34 Using three different outcome measures as indicators of proprioception (reproduction of passive positioning, relative reproduction, and detection of passive of 41 passive motion), Co et al.93 concluded that a well done A C L reconstruction with appropriate rehabilitation can result in proprioception that is essentially equal to that of the contralateral limb. Govett^l observed that static joint position awareness was significantly worse in A C L deficient knees than in A C L reconstructed knees. When compared to the control group, the surgical group was not significantly different in proprioceptive function. EMG/Reflex Testing A number of studies have used E M G techniques to evaluate the changes in muscle firing patterns and reflex muscle contraction(s) following A C L injury.29, 94-96 Altered muscle coordination has been observed in those with A C L deficiency.^?, 98 Higher hamstring moments secondary to altered muscle coordination have been observed in those with A C L insufficiency during functional testing.97-99 j t j s hypothesized that the observed altered muscle coordination (increased activity of muscles that work synergistically with the ACL) is an essential adaptation to secure knee stability in A C L deficient knees.97, 100 Ciccotti et al.94 compared the electromyographic activity of normal, rehabilitated A C L deficient, and A C L reconstructed knees during seven functional activities. The seven functional activities were: walking, ramp ascending, ramp descending, stair ascending, stair descending, running, and cross-cutting. Results indicated the following: 1. Rehabilitation alone for the A C L deficient knee does not restore the E M G patterns of the normal knee during the performance of the seven functional activities, while bone-patella-bone reconstruction does restore the normal E M G synchrony. 2. A greater disparity in performance may be noted between A C L deficient and A C L reconstructed or A C L intact individuals during more strenuous lower extremity activities. Branch et al.99 performed dynamic EMGs in A C L deficient knees and found abnormalities in muscle firing patterns including decreased activity in the gastrocnemius and quadriceps muscle groups and increased activity in the hamstring muscle during stance phase. Branch et al.99 also monitored activities with and without bracing and 42 found that the addition of a derotational brace did not alter the muscle firing pattern in the A C L deficient knee. Wojtys et al.96 assessed the neuromuscular performance in normal and A C L deficient knees. An apparatus delivered an anteriorly directed step force to the posterior aspect of the leg while anterior tibial translation was monitored and E M G signals were recorded at the medial and lateral quadriceps, medial and lateral hamstrings, and gastrocnemius muscles. Testing was done at 30° of knee flexion with the foot fixed to a scale to monitor weightbearing, while the tibia remained unconstrained. The results of the study indicated that timing and recruitment order of the observed muscles were altered by A C L injury. Specifically, those individuals with A C L deficient knees who had higher subjective functional rating scores favored the same voluntary recruitment order as most normals, which was hamstrings-quadriceps-gastrocnemius muscles, while those with lower subjective scores favored their A C L antagonist quadriceps muscle first in their voluntary response. Finally, it was observed that chronic A C L deficient individuals remained slower in their muscle reaction times in all five muscle groups when compared to the normals. Other investigators have studied the changes in the neuromuscular function of the hamstring muscle after A C L injuries.29, 68, 70, 96 Some investigators have suggested the reflex hamstring contraction latency (RHCL) to be a measure of dynamic knee proprioception.29, 68 Clinically, RHCL can be defined as the reaction time of the hamstrings to a posterior-anterior force applied to the tibia.29 The measurement of reaction time is one way to quantify the neuromuscular response from the onset of a stimulus to reaction at the muscular level. 1 It is the assumption that as the time lag of the R H C L shortens, the stress to ligaments and other knee joint structures becomes lessJ In two studies, Beard et al.29, 68 measured the RHCL in A C L deficient knees and found the mean latencies in the affected legs were nearly doubled of that in the unaffected legs. No correlation was found between passive anterior-posterior laxity and the frequency of reported "giving-away" of the knee.29, 68 There was, however, a significant positive correlation between the RHCL differential and the level of reported 43 instability. Beard et a l / y ' " ° concluded that the loss of protective reflex between the A C L and the hamstrings in those with A C L deficiency is likely to be a contributing factor in the decreased joint stability experienced by these individuals.68 44 Rehabilitation Rehabilitation is an important component of both the surgical and non-surgical interventions of A C L injuries. Although there is no universally accepted rehabilitation regime, most agree that the general goals of rehabilitation are to: 1. Develop a strong, mobile, and functionally stable knee capable of withstanding the forces created during activities of daily living (ADL) and sport activities. 2. Minimize the complications of surgery. 3. Minimize the risk of future joint deterioration in the involved knee joint. Open Kinetic versus Closed Kinetic Chain A primary concern in the rehabilitation of a newly reconstructed A C L is the amount of strain placed on the graft during therapeutic exercises.3> 6, 27, 101 At present, the relationship between graft strain behaviour imposed by exercise and the biological response of the healing graft is essentially unknown. 102 However, if the rehabilitation process is too conservative, many complications such as prolonged knee stiffness, loss of range of motion, and atrophy of the thigh musculature may occur.27, 101 As a result of this dilemma, numerous studies have been done to investigate the strain within the A C L during exercises commonly used in rehabilitation.^, 41-44, 101, 103-106 Some investigators have specifically examined the effects of open kinetic chain exercises and closed kinetic chain exercises on the A C L . 103, 105, 106 Yack et al.106 found significantly less anterior tibial translation during closed kinetic chain exercises in those with A C L deficient knees than during open kinetic chain exercises. In addition, it was observed that while anterior tibial translation increased with an increase in loading during open kinetic chain exercises, it was not significantly different during closed kinetic chain exercises. Ohkoshi et al.105 examined the mean shear force exerted on the tibia at 15°, 30°, 60°, and 90° of knee flexion while the subject stood on both legs. The mean shear force exerted on the tibia demonstrated a posterior drawer force at each of the knee flexion angles tested. In addition, shear forces were estimated to be negative in more than 90% of the population in positions with the knee flexed at 30° or more and the trunk anteriorly flexed at 30° or more. Ohkoshi et al.105 concluded that it would be beneficial 45 to include closed kinetic chain exercises such as the "half-squats" in the early stages of A C L rehabilitation for the following reasons: (1) to enhance proprioceptive function of the affected lower limb(s), (2) to prevent bone atrophy, (3) it is functional, and (4) to cause simultaneous contractions of the quadriceps femoris muscles and the hamstrings. Therefore, closed kinetic chain exercises are applicable in the early stages after A C L reconstruction when excessive strain to the graft is undesirable. Accelerated Rehabilitation Recently, the concept of accelerated rehabilitation after surgery has been proposed and implemented by some investigators to overcome the many complications of A C L reconstructions. 101, 104, 107 Components of an accelerated rehabilitation program are: 1. No initial range of motion restrictions 2. Weightbearing of the limb as tolerated by the patient 3. Closed kinetic chain exercises 4. Proprioceptive training Shelbourne et al.101 observed that participants (post A C L reconstruction) of an accelerated rehabilitation program had less anterior knee pain, more knee joint mobility, and better strength in the lower limbs than those on a more conservative program. The study also indicated that the accelerated rehabilitation program did not compromise the integrity of the A C L grafts. Histologic analysis of graft tissue from individuals from both groups failed to indicate any adverse reaction to the physical stresses inherent in the accelerated rehabilitation protocol. MacDonald et alJ04 examined the effects of an accelerated rehabilitation program after A C L reconstructions with combined semitendinosus-gracilis autograft and a ligament augmentation device. Again, no detrimental effects on the A C L reconstructed knees were found. However, opponents of this type of rehabilitation argue that because the relationship between graft strain and the healing response of the graft is currently unknown, it is impossible to safely recommend such exercise programs without running the risk of permanently elongating the graft or disrupting its fixation sites. 102 Regardless, the success of accelerated rehabilitation has been observed in a number of studies. 101, _104, 107 p o r example, in the study by 46 MacDonald et al.104, 32 of the 34 subjects returned to their pre-injury activity level at approximately 20 to 34 months post-reconstruction. Proprioceptive Training There is indirect evidence suggesting that the existing motor programs after an A C L injury can be modified according to the altered sensory situation to regain nearly normal neuromuscular function.7^, 89, 94, 108, 109 \ \ 7 e m U s t remember that although the afferent feedback from the A C L may be important for proprioception and functional joint stability, afferent fibers with sensory endings in the A C L are only a minor part of the total innervation of the knee joint. Thus, it is of interest to determine whether denervation of the A C L secondary to injury or surgery could not be compensated for by the numerous afferent fibers that supply the capsule and other ligaments of the knee joint. These sensory endings can also be activated by forces or movements that cause tension to the A C L . 7 6 , 77 j t h a s D e e n demonstrated that increasing the tension of the collateral ligaments or the posterior cruciate ligament elicits similar effects on the gamma efferent muscle spindle system.76, 77 Thus, these afferents may signal sensory information to the same neuronal circuit(s) the A C L does. However, although the effects of the afferent fibers from the posterior articular nerve of the knee on spinal neurons have been demonstrated in general, little information is available about the relevance of different afferent inputs, from the capsule and the ligaments, to these spinal circuits.22 Wojtys et al. 110 conducted a study to determine the effects of various effects of various exercise regimes (isotonic, isokinetic, and agility) on the muscle reaction time the muscles crossing the knee joint (gastrocnemius muscle, medial hamstring muscle, lateral hamstring muscle, lateral quadriceps, and medial quadriceps). These muscles are capable of knee joint compression and that may ultimately aid the A C L in preventing excessive anterior tibial translation. To measure muscle reaction times, an apparatus delivered an anteriorly directed step force to the posterior aspect of the leg while anterior tibial translation was monitored and E M G signals were recorded. According to the investigators, the importance of muscle reaction times to normal extremities for joint stabilization and injury prevention is unclear. In addition, although weight training is an 47 important part of most athletic conditioning programs, the effect of these programs on neuromuscular function remains unclear. The study included four groups: Group I - Healthy volunteers on a isokinetic program. Group II - Healthy volunteers on a isotonic program. Group III - Healthy volunteers on an agility program. Group IV - Controls. A l l three training groups exercised for 30 minutes three times per week for 6 weeks. The results indicated that while the agility-trained group generally improved in muscle reaction time in all categories (spinal reflex, intermediate response, and voluntary response), the voluntary response times in the isokinetic group were slower in all five muscle groups, with the medial hamstring and medial quadriceps muscles showing significant delays. The isotonic group's responses showed insignificant changes after 6 weeks of training. Thus, while agility exercises can potentially improve muscle reaction time to anterior tibial translation at the knee joint, isotonic and isokinetic strength training of the lower extremities do not appear to improve this parameter. Zatterstrom et al.72 demonstrated the standing balance of subjects with chronic A C L insufficiency improved with physiotherapy (3 to 6 months), and follow up examination after 36 months proved persistent normalization of the single-limbed balance on both sides (ACL intact and A C L insufficient). After studying the ability to detect passive joint motion in 12 ballet dancers and 11 age-matched controls, Barrack et al.^9 concluded that kinaesthetic awareness can be improved through extensive athletic training. Success after ligament reconstruction, therefore, may not depend directly on the tightness or strength of the reconstruction, but rather on the recovery of proprioception. High levels of athletic performance may elude the individual with a reconstructed A C L due to the loss of proprioceptive nerve-endings, causing a loss of the normal feedback mechanism and reflex postural tone of the muscles.34 As mentioned, the A C L may have a neurophysiological proprioceptive role that is just as important as its mechanical role in maintaining joint stability. Since a decrease in proprioception of the knee can contribute to progressive joint instability, leading to rapid 48 joint deterioration, the inclusion of proprioceptive training in the rehabilitation of A C L lesions is considered to be beneficial.4, 47 The ultimate goal of proprioceptive training is to regain proper neuromuscular control of the affected joint(s).47 As part of regaining proper neuromuscular control of the knee joint, dynamic (or reflex) control of the hamstring muscles must be restored.29, 68, 70 Walla et a l .m observed that high functional rating scores were correlated to the presence of "reflex control" of the hamstrings in individuals with A C L deficient knees. He then concluded that dynamic hamstring control is important in this population. Giove et al.l 12 found higher levels of sports participation in those whose hamstring strength was equal to or more than their quadriceps strength. Studies also indicate that the RHCL of an A C L deficient knee can be improved through techniques that facilitate rapid hamstring contractions and emphasize the proprioceptive feedback from around the knee. In addition, a positive correlation between improvements in RHCL and functional gain has been observed. 70, 109 Thus, it appears that a proprioceptive training program for A C L deficient or A C L reconstructed knees should include techniques that will facilitate rapid recruitment of the hamstring muscles. Despite all the research indicating the importance of proprioception in the function of the knee joint, proprioceptive training protocols in clinical setting still appear to be lacking. Based on the database collected, Beard and Fergussonll^ reported that 72% of the centers surveyed had the main objective of increasing muscular strength for the A C L deficient knee while only 12% had the objective of obtaining dynamic stability via proprioceptive training. 49 Chapter Three METHODOLOGY LITERATURE REVIEW Kinetic Communicator Dynamometer Isokinetic dynamometers are frequently used by clinicians and researchers as assessment tools. In order for a measurement tool to be clinically useful, the tool must be considered to be reliable, valid, and sensitive to clinically important changes(s). Reliability is the first prerequisite for validity and sensitivity to change. The Kinetic Communicator (KJN-COM®) dynamometer (Chattex Corp., Chattangooga, TN) system is a hydraulically driven, microcomputer-controlled device for the test, measurement, and rehabilitation of human joint function. The KIN-COM user performs a movement against a resistance which the machine provides via a rotating lever arm system. The machine-controlled movement modes include isokinetic, semi-isotonic, and passive joint motion. The KIN-COM can induce concentric, eccentric, or isometric contraction of the involved muscle(s). Farrell et al. 114 conducted a study to assess the reliability and validity of the KIN-C O M device operating system, with the unit operating in a manner designed to imitate human use of the machine. The study focused on the following functions of the unit: 1. The lever arm position. 2. The lever arm velocity. 3. The force measurement systems. The three functions of the KIN-COM were tested under both static (no lever arm movement) and dynamic (operating) conditions. Measurements made by the KIN-COM system were compared to measurements made simultaneously with external and independent devices designed to copy the KIN-COM's system. The results of the study indicated the following: 1. Average lever arm velocity is within 1.5% of the target velocity. 2. Acceleration time of the lever time to target velocity is within 0.16 seconds (worst case). Deceleration is affected by lever arm direction and not target velocity, and it is equal to approximately 50% of the acceleration time. 3. The force measuring system is within 3% of the actual applied load(s). 50 Thus, Farrell et a l . 1 1 4 concluded that the KIN-COM is acceptable for most clinical or research applications. Although the KIN-COM is generally agreed upon as a reliable measuring tool in the literature, there are still some methodological issues that are commonly debated on in the literature. Some of these issues are: 1. What is the most reliable and valid value derived from the K I N - C O M to quantify strength? 2. What is the significance of preload? 3. What is the significance of the dynamometer application arm length? 4. What is the significance of having an intercontraction pause? 5. Should data derived from isokinetic tasks be "trimmed" of the nonisokinetic values produced at the extremes of the range of motion? 6. What is the most reliable or clinically relevant test protocol for strength testing? It is beyond the scope of this thesis paper to discuss all the significant findings in the literature relating to these methodological issues. A brief synopsis, however, will be provided. Strength Quantification In a study conducted by Tis and Perrinl 15? the relationship between average force, average torque, peak force, and peak torque were examined when measuring isokinetic strength of the knee extensor and flexor muscles. Each subject's concentric and eccentric quadriceps femoris and hamstring muscle strength was evaluated at 90°/second, which was performed through the range of 5 to 90 degrees. A preload of 75N was used and data were gravity corrected. The results indicated the following: 1. For knee extensor musculature, r values ranged from 0.82 to 0.94 and 0.91 and 0.99 for concentric and eccentric contractions, respectively. 2. For knee flexor musculature, r values ranged from 0.72 to 0.95 and 0.86 to 0.96 for concentric and eccentric contractions, respectively. The investigators concluded that the use of any of these isokinetic variables would be appropriate when reporting isokinetic data. However, when using peak and average 51 measures interchangeably, preload and range of motion must remain consistent between and among subjects. Preloading Preloading is the setting of minimal resistance of the lever arm that the muscle must overcome in order to trigger the movement of the arm in the predetermined direction. Preloading assists muscle activation in the initial portion of the isokinetic task. 116 Jensen et al.l 17 conducted a study examining the effect of two preload levels on isokinetic performance of the knee extensors on the KIN-COM. The low preload level was 50N and the high level was 75% of a maximal voluntary isometric contraction (MVIC), performed at the start angles of 100° and 30° of knee flexion. The high preload condition had a significant increase in torque at 1, 5, 10, and 15 degrees concentrically and 1,5, 10, 15, and 20 degrees eccentrically into the range of motion. In comparing the two preload conditions over the whole torque curve, there was a significant difference in average torque values both concentrically and eccentrically, no significant difference in peak torque in either contraction, and a significant shift in peak torque angle with concentric contractions only. The investigators of this study concluded that a high preload level significantly increases torque production in the beginning of the curve and also the average torque for the entire curve. 11 ? According to Hoens and Strauss 116, preloading permits recruitment of motor units and stretch of the series elastic component. This assists in obtaining maximal activation of the muscle(s) earlier in the range of motion of the isokinetic task. Without preloading, submaximal activation of the muscle group is likely during the acceleration phase of the movement. During this phase, motor units would be recruited and the torque produced against the resistance pad of the dynamometer would be less than that produced under maximal isokinetic conditions. Setting the preload at a level that eliminates either a positive or negative inflection at the initiation of an isokinetic task can aid in minimizing the impact artifact and the oscillations resulting from the inertia of the segment, and in producing a true isokinetic strength curve shape. 116 52 Dynamometer Application Arm Length A key recommendation when using isokinetic dynamometers is that the knee and the dynamometer axes are oriented co-axially. When the knee and the dynamometer axes are oriented co-axially, it is thought that the torque scores recorded by the dynamometer are not affected by the length of the dynamometer application arm.H8 Co-axial alignment represents the ideal mechanical situation, facilitates analysis of exercise and test situations, and serves as a basis for comparisons of torques within and between subjects or patients. Kramer et al.118 conducted a study to determine if the length of the dynamometer application arm affected torques during reciprocal concentric-eccentric contraction cycles at 60°/second angular velocity. Healthy subjects were tested using dynamometer application arm lengths corresponding to 33%, 67%, and 95% of the distance from the knee axis to a resistance pad placement that just contacted the dorsum of the foot. Results indicated that the torques produced at the 33% length were approximately 39% lower than those produced at the 67% length, and the torques produced a the 67% length were approximately 10% lower than those produced at the 95% length. The investigators of the study believe that the discrepancy between torque readings at different dynamometer application arm lengths is due to the displacement of the knee axis away from the rotational axis of the dynamometer, such that the two axes were no longer co-axial. According to Kramer et al.H8, displacement of the knee axis from the dynamometer axis, such that the subject application arm is longer than the dynamometer application arm, produces a dynamometer torque less than that recorded when the two axes are co-axial. If the knee axis is displaced such that the subject application arm is shorter than the dynamometer application arm, the dynamometer torque will be greater than the recorded under the condition of co-axial alignment. Thus, torque measurement recorded using different application arm lengths are not directly comparable. 118 Intercontraction Pause In the study conducted by Wessel et al.l 19, the effect of an intercontraction pause on the measurement of isokinetic torque of the knee extensors was examined. A pause of 0.25 or 1.0 second was used. The results indicated that peak and average concentric 53 torques were greater for the 0.25 second pause compared to the 1.0 second pause. Intercontraction pause had no effect on the eccentric phase of the knee extensors. Truncating Data In a study conducted by Hoens and Strauss 116^  m e effect of deleting nonisokinetic phases of movement from isokinetic strength evaluations was examined. While the term "isokinetics" denotes a type of muscular contraction which accompanies a constant rate of limb movement, periods of acceleration and deceleration exist in the context of isokinetic exercise and evaluation. The acceleration and deceleration periods of "isokinetic" exercise/evaluation limit the duration of the period of constant angular velocity in exercises/evaluations with isokinetic dynamometry. According to Hoens and Strauss 1 the reliability and validity of the measurement of torque produced by the muscle groups involved in the task is dependent on several factors, including the achievement and maintenance of the preset velocity by both the dynamometer and subject. A technique to correct for measurement error in the extremes of range for isokinetic tasks is to eliminate data generated in these regions from the analysis.!^ However, there appears to be no consistency in the portion of the R O M excluded in the literature. In the study conducted by Hoens and Strauss the isokinetic strength and endurance of the trunk extensors and flexors in normal male and female subjects were evaluated concentrically from 25° of extension to 30° of flexion at angular velocities of 20, 40, and 60 degrees per second, on two occasions. Full range of motion average torque (FRAT) and truncated range of motion average torque (TRAT) were calculated. The results indicated that direct FRAT and TRAT measures, although highly correlated, were significantly different at each of the three angular velocities. Derived agonist/antagonist ratios had slightly lower correlations and were significantly different from each other at 40 and 60 degrees per second. The investigators of the study concluded that the nonisokinetic phases of an isokinetic task significantly influence trunk average torque recorded by the dynamometer, particularly at faster velocities and with direct measures. 11°" 54 Standard Protocol Despite all the research that has been done on isokinetic dynamometry, there is a lack of consensus as to what specific protocol or methodology should be used for strength testing of the particular regions of the body. 120 Differences in testing protocols, participant characteristics and type of dynamometer do not allow valid comparisons between studies. Some investigators have determined test-retest reliability using average torque, while others have used peak torque values. 120, 121 Furthermore, the number of trials used to establish average or peak torques differ between studies. In some studies, agonistic and antagonistic movements were assessed separately, while others used reciprocal tests. H 4 ' 122 Similarly, eccentric and concentric movements were performed separately or in one continuous movement. 120 Finally, pretest sessions were performed in some studies, while there were no familiarization tests in others. 120 55 Reflex Hamstring Contraction Latency Reflex hamstring contraction latency was first suggested by Beard et al.29 as a dynamic measurement of proprioception. Neuromuscular performance can be assessed at three levels 11^: 1. Spinal stretch reflex response (Ml) . 2. Intermediate response (M2). 3. Voluntary or volitional response (M3). Reflex hamstring contraction latency has been assessed at all three levels in the literature.29, 68, 96, 110 j n a study conducted by Wojtys et al.HO, the neuromuscular performance of normal lower extremities was assessed using E M G . These investigators found the following: 1. The spinal stretch reflex responses of the lower limb occur between 26 and 130 msec after the onset of anterior tibial translation force. 2. The intermediate responses is very reproducible, occurring between spinal stretch reflex and volitional activity. It routinely occurs just before volitional activity at 110 to 216 msec after the onset of anterior tibial translation force. 3. Volitional activity is of the largest amplitude and of longest duration. It occurs between 156 msec to 431 msec after the onset of anterior tibial translation force. Currently, there are two popular methods of measuring the neuromuscular function of the hamstring muscle, which are;l> 68 1. The Vicon (Oxford Metrics, Oxford, England) Interfaced Knee Displacement Equipment (VIKDE). 2. The Kinetic Communicator (KIN-COM®) dynamometer (Chattecx Corp., Chattangooga, TN). When using the VIKDE to measure reflex hamstring contraction latency, the subject is measured in a position of single-limb stance (full weight-bearing) with the knee flexed 30°.29, 68 An anteriorly directed shear force is applied to the tibia and surface EMGs record the activity of the hamstrings in response to the applied force. The latency of contraction is defined as the time from initial tibial movement, identified by accelerometry, to the onset of increased hamstring E M G activity. When using this 56 method of assessing the neuromuscular function of the hamstring muscle, all three types of muscular responses (spinal reflex response, intermediate response, volitional response) may be assessed. There are two methods of measuring RHCL using the K I N - C O M . 1 The first method involves measuring peak torque time (PTT). Peak torque time was first defined by Ihara and Nakayama^O as the "time before the hamstring exerts maximum force."(p.314) Peak torque time is the time lapse between the initial mechanical movement of the dynamometer lever arm and the generation of maximum torque by the hamstring musculature in response to the sudden forward movement of the dynamometer lever arm. In addition to analyzing PTT, Ihara and Nakayama^O also looked at peak torque value (PTV), rising torque time (RTT), and rising torque value (RTV) in their study. Of the four parameters, PTT seems to be the most important parameter representing hamstring reaction time. The second most important parameter of reaction time seems to be the RTT because it represents the time the hamstrings are about to react. A l l these parameters are considered to be volitional responses of neuromuscular function. The second method of measuring hamstring reaction time using the KIN-COM isokinetic dynamometer is electromyograph time (EMGT).l Electromyograph time measures the time lapse between the initial mechanical movement of the dynamometer lever arm and the first myoelectrical hamstring activity that is recorded. 1 In both methods of measuring hamstring reaction time using the KIN-COM, the subject sits on the table of the KIN-COM with 90° of hip flexion and 90° of knee flexion and is stabilized at the waist and distal thigh. 1 The dynamometer lever arm is secured to the subject's testing leg, 2 cm above the medial malleolus, and it moves unexpectedly forward (into full knee extension) at a angular velocity of 230°/second.l a The subject is instructed to draw the leg backward against the forward force of the lever arm as quickly as possible once the movement of the lever arm is felt. a Ihara and Nakayama'0 used a speed of 210°/second, which is just below the maximal angular velocity of knee flexion during normal gait. 57 Small et a l . 1 did a comparative study on the two methods of measuring RHCL using the KIN-COM isokinetic dynamometer in the normal population. Based on the results of the study, both the PTT and EMGT are reliable methods for measuring hamstring reaction time (Table 3.1). Table 3.1: Cronbach Coefficient Alpha for Two Methods of Measuring Reaction Time in the Dominant and Nondominant Hamstring Musculature * • Method Dominant Nondominant Peak Torque Time 0.82 0.90 Electromyograph Time 0.90 0.86 However, due to the sensitivity of the EMGT, there are some difficulties associated with anticipation of the lever arm. Anticipation of the lever arm can be identified by an increase in myoelectrical activity prior to actual lever arm movement. 1 According to Small et a l . l , this heightened pre-stimulus E M G activity can not be accurately differentiated from E M G activity representing actual response to lever arm. Consequently, full subject compliance is essential for obtaining accurate measurements using the E M G T protocol. In their study, Small et al.l a i s o attempted to determine if there is a significant difference in RHCL between dominant and nondominant limbs. The study found no significant difference (p<0.05) in reaction time as measured by PTT and E M G T between the dominant and nondominant limbs (Tables 3.2 and 3.3). The authors concluded that the uninvolved lower extremity may be used as a control. Table 3.2: Mean Hamstring Reaction Time as Measured Using PTT (N=40) Leg Mean (msec) SD Range Dominant 380.5 35.9 130-500 Nondominant 394.1 45.5 260-500 Table 3.3: Mean Hamstring Reaction Time as Measured Using Electromyograph Time (N=40)] Leg Mean (msec) SD Range Dominant 30.5 1.41 10-120 Nondominant 33.6 1.94 10-150 Wojtys et al.l 10 reported a lack of correlation between muscle reaction time assessed by E M G and the time to peak muscle torque. They concluded both parameters 58 depend on neural pathways and occur sequentially. Also, the investigators felt that both parameters appear to be needed for effective limb protection but seem to be affected by different training factors. It should be noted, however, the method in which PTT was measured by Wojtys et al.96, 110 j s different from that described by Small et a l . l . 59 Functional Testing Functional tests are often included in the evaluation of anterior cruciate ligament deficiency or are used as a method to monitor rehabilitation.^7* 123-127 Barber et al.126 studied the effectiveness of five various hopping, jumping, and cutting type tests in determining lower extremity limitations in A C L deficient knees. The study involved 93 subjects with A C L intact knees and 35 subjects with A C L deficient knees. The five functional tests were: (1) one-legged single hop for distance, (2) one-legged vertical jump, (3) one-legged timed hop, (4) shuttle run without pivot, and (5) shuttle run with pivot. For each functional test, the limb symmetry index was calculated. To calculate the symmetry index for the one-legged single hop for distance and the one-legged vertical jump, the mean of the involved limb is divided by the mean of the noninvolved limb and the result is multiplied by 100. To calculate the symmetry index for the one-legged timed hop and the two shuttle runs, the mean of the noninvolved limb is divided by the mean of the involved limb and the result is multiplied by 100. In the normal population, no statistical significance was found between right and left lower limb scores (limb symmetry index) as related to sports activity level, gender, or dominant side. 126 This allowed an overall symmetry index score to be established for the normal population as a whole. The majority (92% to 93%) of the normal population scored in the 85% limb symmetry category for the one-legged single hop for distance and the one-legged timed hop. The study found that neither the two shuttle run tests nor the vertical jump test are useful in detecting lower limb functional limitations. More than 90% of the A C L deficient population scored in the normal limb symmetry range on the two shuttle run tests; subjects tended to compensate by running at one-half velocity and guarding both legs during turning and cutting maneuvers. For the vertical jump test, 27% of normal subjects scored outside the normal limb symmetry range. It is the recommendation of the authors that at least two hop tests are used when assessing functional limitations in the A C L deficient population. 126 This i s because while 60% of the A C L deficient knees performed abnormally on at least one of the two hop tests in the combined test analysis, the range of abnormality decreased to 42% or 50% (depending on the test used) i f only one hop test was used. 60 Noyes et a l . 1 ^ ' studied the sensitivity of four different types of one-legged hop tests. The four one-legged hop tests were: (1) single hop for distance, (2) timed hop, (3) triple hop for distance, and (4) cross over hop for distance. The analysis showed that 62% of the A C L deficient population performed abnormally on at least one of the two hop tests conducted. When only one test was conducted, the percentage of abnormal results decreased to between 49% and 52%. The authors concluded that these hop tests have a low sensitivity rate. However, the high specificity and low false-positive rates allow these hop tests to be used as confirmation tests for lower limb dysfunction. 127 In a study conducted by Kramer et al.l28, t n e test-retest reliability of the single hop for distance test following A C L reconstruction was examined. According to the investigators of this study, in documenting the single hop for distance, the criterion measurement has been calculated using the mean distance of three hops and that of the longest hop. Whether the three hops are equivalent, and both data analysis strategies lead to similar scores, or whether using the longest hop leads to significantly higher scores than using the mean of three hops, is unclear. The primary purpose of the study was to determine test-retest reliability of the single hop for distance, when using the distance hopped on each leg and the hop index as criterion measurements, and using the two-data analysis strategies: 1. The mean of three hops. 2. The longest of three hops. Each test occasion (two in total, within a 5-day interval) consisted of three trials for each leg. The mean distance of the three hops and that of the longest hop were determined (in cm) for each leg. The mean hop index score was determined by dividing the mean distance hopped on the involved leg by that hopped on the noninvolved leg, and the longest hop index was determined by dividing the longest hop on the involved leg by the longest hop on the noninvolved leg. Results of the study demonstrated acceptable reliability for the distance hopped on each leg and the hop index, regardless of whether the mean of three hops or the longest hop was used in calculations (all ICO0.75). Although hop distances calculated using the longest hop were significantly greater than those determined using the mean of three hops, hop indexes were not affected by data 61 analysis strategy. The percentage difference between the scores determined on occasion one and occasion two averaged approximately 3%. This indicates that any learning component associated with performing the single hop for distance is minimal and is not manifested by a systematic tendency for improvement simply as a result of repeating the test. As a result, researchers can have confidence in the data generated on one occasion. Booher et al.129 conducted a study examined the reliability of the following single-leg hop tests in healthy subjects: 1. Hop for distance. 2. Six meter timed hop. 3. Thirty meter agility hop. Healthy subjects chose a random sequence of testing that was performed on both lower extremities. Subjects performed two testing trials for each single-leg hop test on each occasion. An independent t-test revealed no significant difference between dominant and nondominant extremities for all three dependent variables. An A N O V A repeated-measures analysis revealed significant differences between the test trials within and between sessions for all three dependent variables. The fact that the test scores were consistently improving during the second trial of each day indicated that motor learning was occurring. However, when the mean of two test trials was analyzed, no significant difference existed between two (hop for distance and timed hop) of the three dependent variables. By taking the mean of the two trials for each day and comparing days, the investigators obtained greater stability of measures as demonstrated by the intraclass correlation coefficient, standard error of measurement, p value, and confidence interval. Although the agility hop was still significantly different between test days, Booher et a l . 1 2 9 felt it was still a stable measure for clinical usage because: 1. The 95% confidence interval was 15.05 to 18.07 seconds. This is considered to be a small confidence interval. 2. When the difficulty of a 30 meter single-leg agility hop test is considered, a standard error of measurement of 0.77 seconds and a mean difference of 0.47 seconds between occasions are very small. 62 Thus, the investigators of this study concluded that the three single-leg hop tests examined were reliable and contained small measurement error as performed in the study. 63 Subjective Scoring Scales The most commonly used subjective scoring scale for A C L lesions is the modified Lysholm and Gillquist Knee Scoring Scale (1985). 130 The original version was first introduced in 1982.131 The maximum score in which a subject can achieve on this scoring scale is 100 (maximum function). The intrapersonal coefficient of variation of this scale is 3% and the interpersonal is 4%. 130 The test-retest correlation coefficients are 0.97 and 0.90 respectively. 130 The Tegner and Lysholm Activity Score (1985) is graded from 0 to 10 and covers activities in daily life and recreational and competitive sports. 130 in a study by Tegner and Lysholm 13 0, significant differences in scores (modified Lysholm and Gillquist Knee Scoring Scale) at different activity levels were obtained in those with A C L deficient knees. For example, the mean score for subjects at activity levels 5-10 was 83 + 10, and for subjects at activity level 0, it was 53 + 16 (p < 0.001).130 The authors argued that the activity scale is a "valuable complement to the functional score because limitations in knee function may be masked by an involuntarily low activity level."!30 64 1. A C L deficient: 2. A C L reconstruction: 3. Isolated A C L rupture: 4. Functional j oint stability: 5. Proprioception: 6. Kinesthesia: 7. Isokinetic: Chapter Four DEFINITION OF TERMS A condition which the A C L has been ruptured and it has not been surgically repaired or reconstructed. A surgical procedure which attempts to restore the integrity and function of the ruptured A C L . A complete tear of the A C L that is not associated with other ligamentous, capsular, meniscal or bony injury. The condition in which the joint is stable and does not give symptoms during physical activities. The unconscious sensation of joint movement and joint position. Proprioception is involved in the control of movement and posture. The conscious sensation of joint movement and joint position. Kinesthesia is involved in the control of movement and posture. The dynamic muscular contraction when the velocity of movement is controlled and maintained constant by a special device. 8. Concentric muscle contraction: A muscle contraction that involves the approximation of the origin and insertion of the muscle. 9. Eccentric muscle contraction: A muscle contraction that involves the origin and insertion of the muscle to move away from each other. 10. Coordination: 11. Balance: Highly skilled multimuscular automaticity. It is the bodily equilibrium or the ability to maintain the center of body mass over the base of support without falling. 12. Engram Automatic motor patterns. 65 Chapter Five PILOT STUDY Purpose A pilot study was conducted to determine the test-retest reliability of measuring peak hamstring torque time. The protocol described by Small et al.l was used, m the study conducted by Small et a l . l , m e main purpose was to determine i f there was a significant difference between two methods of recording hamstring PTT using the KIN-C O M isokinetic dynamometer (peak torque time vs. electromyograph time). Although Small et al.l provided evidence that measuring PTT using the KIN-COM is reliable within an occasion, the test-retest reliability between occasions was not addressed. Unless the test-retest reliability between occasions of an equipment is determined, clinical and research decisions based on the data collected (from the equipment) may be questionable. For instance, an improvement in a subject's scores may be the result of learning to perform better with the equipment, decreased apprehension, and/or simple variation attributable to repeated testing, rather than treatment effect. In addition, in the two studies that used the KIN-COM to assess hamstring neuromuscular function, two different velocities were used for the forward movement of the lever arm, 2307second and 2107second. At present, it is unclear whether PTT measured at different velocities may be compared to each other. Therefore, before valid comparisons may be made between results of different studies, it must be shown that there are no significant differences between PTT measured at different velocities. It is also beneficial to determine whether or not we can accurately predict peak torque times at high velocities from those times assessed at lower velocities. At present, the maximum angular velocity that the KIN-COM used for this study can be programmed for is 2507second. Although this angular velocity appears to be high, it is just slightly above the velocity of the knee extending during the stance phase of the gait cycle. The quadriceps fires during the last 12% of the gait cycle preparing the limb for weight acceptance, and during this time the knee normally extends at a rate of 233° per second.132 66 Subjects Six subject, 4 females and 2 males, were recruited from the university population for this pilot study. Informed consent was obtained for all subjects involved in this pilot study. Only subjects without a history of bilateral lower extremity pathology were eligible to participate in the study. A brief interview revealed that all subjects engaged in recreational or competitive sports activities at least three times per week. A l l individuals involved in the pilot study were right dominant. Table 5.1 summarizes the descriptive characteristics of subjects involved in the pilot study. Table 5 ! : Descriptive Data for Pilot Study N Minimum Maximum Mean Standard Deviation Age (year) 6 23.98 29.88 25.88 2.18 Height (cm) 6 156.10 193.40 170.00 13.73 Weight (lbs) 6 108.00 171.60 135.20 21.50 67 Methods Forward lever arm movement of the KIN-COM isokinetic dynamometer provided the unanticipated stimulus that evoked an eccentric response of the hamstring muscles. There were two test occasions for each subject. In the first test occasion, three different velocities were used for the forward (extension) movement of the lever arm: 230°/second, 210°/second, and 190°/second. In the second test occasion, only 190°/second was used. A backward (flexion) angular velocity of 10°/second was used for all testing occasions. A l l testing occasions occurred at the Allan McGavin Sports Medicine Clinic. Prior to actual testing, all subjects were required to complete 10 minutes of stationary bicycling. This was followed with five repetitions of static hamstring stretch, each held for 20 seconds. Efforts were taken to ensure the subjects had a thorough understanding of the testing procedure. A l l orders of testing were randomized by side and forward angular velocity of the lever arm for the first test occasion. The order of testing from the first test occasion was duplicated for the second test occasion. Each subject was seated on the KIN-COM with their pelvis and tested thigh secured to the bench. The subject's trunk was also comfortably, but firmly, strapped to the back of the KIN-COM bench seat. The axis of motion of the K I N - C O M was aligned with the lateral femoral condyle of the tested knee. The dynamometer lever arm was secured to the subject's testing leg 2 cm above the medial malleolus. A l l tests were performed between 0 and 90 degrees of knee flexion. Gravity compensation was performed for all test occasions. A low acceleration rate was used for all test occasions. Subjects were asked to close their eyes and were fitted with earphones. The music used ranged from classical music to classic rock. The purpose of this procedure was to eliminate any visual or audio cues. Once a subject was secured to the KIN-COM bench and all visual or audio cues were removed, the subject was allowed four practice trials. After a one minute rest interval, actual testing began. 68 Test Occasion One For the first test occasion, six trials were performed at each angular velocity. There was a five minute rest interval between each velocity test session. Thus, there were three test sessions in the first test occasion for each of the lower extremity. Subjects were instructed to maximally contract their hamstring muscle in response to forward lever arm movement. This was achieved by pulling back into the lever arm pad which was moving forward into full knee extension. Subjects were asked to completely relax their musculature after the lever arm achieved full extension and to refrain from anticipating future lever arm movement. Time of lever arm movement was varied according to the following intervals: 5, 3, 8, 6, 5, and 8 seconds between repetitions. This was done to minimize the subjects' tendency to anticipate the forward movement of the lever arm. Test Occasion Two For the second test occasion, six trials were performed at 1907second. Thus, there was only one test session in the second test occasion for each of the lower extremity. After much contemplation and discussion, 1907second was chosen as the angular velocity to be used in this thesis project to assess time to hamstring peak torque. Therefore, the test-retest reliability of 1907second needed to be known. The angular velocity of 1907second was chosen for the following reasons: 1. A l l potential subjects will have had their A C L reconstructed with their semitendinosus (part of the hamstring muscles). It is quite likely that the affected hamstring group will be under neurological inhibition during the window of time subjects will be involved in study. Therefore, when using velocities such as 2307second, we may not allow enough time for the affected hamstring group to reach potential peak torque. 2. If we assume that the A C L does contribute to knee proprioception and kinesthesia, then we also must assume that the potential subjects will have impaired knee proprioception and kinesthesia. Thus, we need to ensure that the angular velocity we choose to assess time to hamstring peak torque will provide a large enough window of time for peak torque to be reached and still be challenging for the subjects. 3. To decrease the likelihood of sustaining a strain to the hamstring muscles. 69 Otherwise, the procedures used in the first test occasion were reproduced exactly for the second test occasion. Time to peak torque was obtained from ASCII files saved from the KIN-COM isokinetic dynamometer. The slowest and the fastest times to hamstring peak torque were eliminated from each test session for each subject. The mean of the remaining four trials was used as the time to hamstring peak torque for each subject. 70 Statistics The criterion score (peak torque time of the hamstring muscle) used for statistical analysis was the mean of the four trials retained from the six trials completed during one test session. Peak torque values used in statistical analysis were also obtained in the same manner as the peak torque time values. For test occasion one, a repeated-measures analysis of variance was conducted to determine if there were any differences between peak torque times assessed at the three isokinetic velocities. Pearson product-moment correlation coefficients were calculated to determine the relationship of peak torque times measured at the three isokinetic velocities. Regression was then performed to determine the accuracy of peak torque times obtained at lower isokinetic velocities in predicting those obtained at higher velocities. For test occasion two, intraclass correlation coefficients were calculated to determine the reliability of measuring peak hamstring torque times at the angular velocity of 190° per second between occasions. The level of significance for all statistical tests were set at p<0.05. 71 Results Test Occasion One The main purpose of the first test occasion was to establish the validity of comparing peak torque times collected at different velocities as well as their correlational and regressional relationship. The following table summarizes the descriptive statistics for test occasion one. Table 5.2: PTT Data for Test Occasion One (msec) Velocity N Minimum Maximum Mean Std. Deviation Right-1907sec 6 360.00 480.00 445.83 43.32 Right-2107sec 6 380.00 467.50 423.75 31.01 Right-2307sec 6 365.00 442.50 403.75 29.74 Left-1907sec 6 427.50 522.50 471.58 35.43 Left-2107sec 6 387.50 512.50 424.17 45.65 Left-2307sec 6 370.00 475.00 408.75 39.65 The results obtained in test occasion one of the pilot study for hamstring peak torque times at 2307second coincide with the values Small et al.l f o u n d in their study. Small et al.l f o u n ( i p e a k torque time values of X = 380.5 + 35.9 milliseconds (dominant leg) and X = 394.1 + 45.5 milliseconds (nondominant leg). It appears from the descriptive data that peak torque of the hamstring muscle was achieved faster by the subjects as the forward angular velocity of the lever arm was increased. Dietz et al.l33 showed that the latency of muscle contraction after perturbation depends on the rate and amplitude of acceleration, and therefore on the force. In their study, Dietz et al.l33 examined the electromyogram responses of the leg musculature and the corresponding joint movements following a perturbation of the limb during walking on a treadmill, produced by a randomly timed treadmill acceleration impulse, either predictable, or unpredictable in its amplitude and rate of acceleration. The investigators found that the onset latency (of the gastrocnemius) muscle was shorter (65ms) for high accelerations and longer (85 ms) for lower accelerations. Although the KIN-COM enables the user to select a low, medium, or high acceleration rate, little information is available regarding what these settings exactly mean. After analyzing the ASCII files, it appears that although a low acceleration rate was selected for all test 72 sessions and occasions, each of the three velocities had a different acceleration rate. Data in Table 5.3 demonstrate that as the forward angular velocity of the lever arm was increased, the acceleration rate also increased. Specifically, data in Table 5.3 indicates that in a given period of time (i.e. 0.08 seconds), the lever arm achieved a higher angular velocity when the preset velocity was 210°/second (or 2307second) than when the preset velocity was 190°/second. Similarly, the lever arm traveled a greater angular displacement in a given amount of time when the preset velocity was 210°/second (or 230°/second) than when the preset velocity was 190°/second. Table 5.3: Descriptive Data of Different Preset Velocities Velocity of Lever Velocity at Angle at 0.08s Velocity at Angle at 0.10s Arm 0.08s (From 90°) 0.10s (From 90°) 190°/second 116.297sec 87.04° 154.467sec 84.12° 210°/second 128.547sec 86.62° 171.047sec 83.42° 2307second 140.587sec 86.12° 187.257sec 82.75° Thus, the tendency for the hamstring muscle to achieve peak torque faster as the forward angular velocity of the lever arm was increased may be explained by the findings of Dietz e t a l . 1 3 3 . The data in Table 5.2 suggest that there was a tendency for the dominant limb to respond slightly faster than the nondominant limb. However, a repeated-measures A N O V A found insignificant differences (p<0.05) in PTT between the lower extremities at all three isokinetic velocities. Table 5.4 summarizes the repeated-measures A N O V A conducted to determine i f there were significant differences between peak torque times measured at the three different isokinetic velocities. Table 5.4: PTT at Three Different Angular Velocities Tests of Within-Subjects Effects (RM A N O V A ) Source Type III Sum of Squares df Mean Square F Sig. Sessions® 190°, 17092.10 2 8546.05 30.582 .000 210°,&2307sec Session * Side 1093.35 2 546.67 1.956 .168 Error (Session) 5588.89 20 279.44 73 Tests of Between-Subjects Effects Source Type III Sum of Squares df Mean Square F Sig. Side 971.361 1 971.361 0.258 .622 Error 37581.111 10 3758.111 As indicated by the results, the peak torque times measured at the three different velocities were significantly different. As mentioned previously, there was no significant difference in PTT measured between the left and right lower limbs. Also, there was no significant interaction between isokinetic angular velocity and the side (dominant or nondominant) of limb being tested. Table 5.5 summarizes the correlations between the three different velocities used to assess time to hamstring peak torque at test occasion one. Table 5.5: Pearson Product-Moment Correlation Coefficients for Test Occasion One Variable Right-1907sec Right-2107sec Righ-2307sec Left-1907sec Left-2107sec Left-2307sec Right-1907sec r 1.000 .761* .626 .763* .577 .579 Sig. (one-tailed) .039 .092 .039 .115 .114 N 6 6 6 6 6 6 Right-2107sec r .761* 1.000 .880* .706 .803* .782* Sig. (one-tailed) .039 .010 .059 .027 .033 N 6 6 6 6 6 6 Right-2307sec r .626 .880* 1.000 .394 .459 .541 Sig. (one-tailed) .092 .010 , .220 .180 .134 N 6 6 6 6 6 6 Left-1907sec r .763* .706 .394 1.000 .877* .861* Sig. (one-tailed) .039 .059 .220 .011 .014 N 6 6 6 6 6 6 Left-2107sec r .577 .803* .459 .877* 1.000 .933** Sig. (one-tailed) .115 .027 .180 .011 .003 N 6 6 6 6 6 6 Left-2307sec r .579 .782* .541 .861* .933* 1.000 Sig. (one-tailed) .114 .033 .134 .014 .003 N 6 6 6 6 6 6 * Correlation is significant at the 0.05 level (1-tailed). ** Correlation is significant at the 0.01 level (1-tailed). As demonstrated by the data in Table 5.5, the correlation between 1907degrees and 2107degrees, and between 2107degrees and 2307degrees were statistically significant for both lower extremities. Although it is evident that all three velocities for 74 the left leg were strongly correlated, the degree of association was still stronger between 1907second and 2107second, and between 2107second and 2307second, than between 1907second and 2307second. The lack of significant correlation found between 1907second and 2307second for the right lower extremity may be explained by the findings of Small et a l . 1 . In their study, Small et a l . 1 found the reliability of the PTT protocol was 0.82 and 0.90 for the dominant and nondominant limb, respectively. Thus, the lack of correlation found between 190°/second and 2307second for the right lower extremity may be due to the decreased reliability of the dominant limb (compared to the nondominant limb) and the small sample size used in this pilot study. The following table summarizes the regression analysis of test occasion one. Table 5.6: Regressional Analyses for Test Occasion One Side Independent Variable Dependent Variable Significance of Regression A N O V A Right 1907second 2107second 0.079 Right 1907second 2307second 0.184 Right 2107second 2307second 0.021 Left 1907second 2107second 0.022 Left 1907second 2307second 0.028 Left 2107second 2307second 0.007 From the data collected, it is evident that peak torque times measured at lower isokinetic velocities were predictive of those measured at higher velocities for the left lower extremity (the nondominant limb). However, only those peak torque times measured at 2107degrees were predictive of those measured at 2307second for the right lower extremity (the dominant limb). 75 480 460 440 420 Right @ 2107s 400 380 360 Rsq = 0.5792 340 360 380 400 420 440 460 480 500 R i g h t ® 1907s Figure 5 ! Regressional Analysis for Right 1907s and 2107s 460 440 420 R i g h t ® 2307s 400 380 360 Rsq = 0.3917 340 360 380 400 420 440 460 480 500 R i g h t ® 1907s Figure 5.2 Regressional Analysis for Right 1907s and 2307s 76 460 440 420 Right @ 2307s 400 380 360 Rsq = 0.7749 360 380 400 420 440 460 480 Right® 2107s Figure 5.3 Regressional Analysis for Right 2107s and 2307s 520 480 460 L e f t ® 2107s 440 Rsq = 0.7693 460 480 500 L e f t ® 1907s 540 Figure 5.4 Regressional Analysis for Left 1907s and 2107s 480 420 440 460 480 500 520 540 Left® 1907s Figure 5.5 Regressional Analysis For Left 1907s and 2307s Left® 2107s Figure 5.6 Regressional Analysis For Left 2107s and 2307s 78 Test Occasion Two The main purpose of the second test occasion was to establish the test-retest reliability (between occasions) of measuring hamstring peak torque time at the isokinetic angular velocity of 1907second using the protocol described by Small et a l . 1 . For the second test occasion, only five of the six original subjects returned. Therefore, the sample for the second test occasion consisted of three females and two males. Table 5.7: Comparison of PTT and Peak Torque (@ 1907sec) Between Test Occasions Variable N Minimum Maximum Mean SD PTT @ Occasion One-Right (ms) 5 360.00 480.00 442.50 47.57 PTT @ Occasion Two-Right 5 395.00 480.00 455.00 34.46 Peak Torque @ Occasion One-Right (Nm) 5 38.75 130.50 69.20 39.10 Peak Torque @ Occasion Two-Right 5 39.50 131.25 77.50 35.25 PTT @ Occasion One-Left 5 427.50 522.50 469.40 39.15 PTT @ Occasion Two-Left 5 420.00 485.00 460.00 30.26 Peak Torque @ Occasion One-Left 5 36.75 131.00 60.40 39.87 Peak Torque @ Occasion Two-Left 5 28.75 142.75 67.05 44.29 Time Interval From Test Occasion One to Test Occasion Two (week) 5 12.00 25.00 18.60 5.22 Reliability coefficients were calculated using the intraclass correlation coefficient. The reliability coefficient describes the degree to which scores on a measure represent something other than measurement error.134 Table 5.8: Intraclass Correla tion Coefficients Variable Intraclass Correlation Coefficient (3,k) Right - PTT (ms) 0.922 Left - PTT 0.901 Right - Peak Torque (Nm) 0.986 Left - Peak Torque 0.984 The reliability of the PTT protocol used in the pilot study was 0.92 and 0.90 for the right (dominant) and the left (nondominant) limb, respectively. The reliability of peak torque was 0.99 and 0.98 for the right and the left limb, respectively. 79 Conclusion Based on the findings of test occasion one, comparisons between PTT measured at different velocities should not be made. Based on the Pearson product-moment correlation coefficients calculated in this pilot study, there is a strong relationship between peak torque times measured at 1907second and those measured at 2107second, and between peak torque times measured at 210°/second and those measured at 2307second. Also, it appears that the degree of association between PTT measured at different velocities decreases as the isokinetic velocities become farther apart. Based on the regression analyses, the ability to predict the PTT of a high isokinetic angular velocity from a lower isokinetic angular velocity appears to be dependent on dominance. Similar to the trend observed in the correlational analyses, the ability to predict peak torque times of higher velocities diminishes as the isokinetic velocities become farther apart. Therefore, volitional peak torque times measured at more widely separated isokinetic velocities are less predictive of each other. Based on the results of test occasion two, the reliability of measuring peak hamstring torque time at the isokinetic angular velocity of 1907second is 0.92 and 0.90 for the dominant and the nondominant limb, respectively. 80 Chapter Six METHODOLOGY Purpose Both strength training and proprioceptive training have been emphasized in the literature as important components of the rehabilitation process after A C L injuries. Although the effectiveness of strength training is well-established in the literature, the effectiveness of proprioceptive training is not. In addition, the effects of strength training on neuromuscular function has been examined in the literature while little has been done to examine the effects of proprioception training on strength. Since an A C L injury often results in decreased function of the involved knee joint, despite restoration of normal joint arthrokinematics via surgical intervention, it is important to optimize factors during the rehabilitation phase, such as strength and proprioception, that may positively influence the present and future function of the individual. Thus, the effectiveness of rehabilitation In restoring these factors must be determined. Numerous investigations have examined the effects of A C L injuries on knee proprioception, but very few prospective studies have been done to examine the effectiveness of proprioceptive training after A C L injuries. Thus, the main objective of this study was to determine the effectiveness of proprioceptive training versus strength training in restoring the normal neuromuscular function of the A C L reconstructed knee, or specifically, the peak torque time of the hamstring muscle. The second objective of this research was to determine the relative contribution of isokinetic strength and peak hamstring torque time to the performance of two functional hop tests. The following hypotheses were tested in this study: HI 1: The strength training group will demonstrate significantly greater strength gains in the A C L reconstructed knee than the proprioceptive training group at the end of the 12 weeks. HI2: The proprioceptive training group will demonstrate greater improvements in peak torque time of the A C L reconstructed knee compared to the strength training group at the end of the 12 weeks. 81 HI3: The proprioceptive training group will demonstrate greater improvements in the performance of the two functional hop tests on the A C L reconstructed knee compared to the strength training group at the end of the 12 weeks. HI4: The proprioceptive training group will demonstrate greater improvements in the subjective scores compared to the strength training group at the end of the 12 weeks. HI5: Peak torque time will have a significant effect on the performance of the two functional hop tests. HI6: Isokinetic strength will have a significant effect on the performance of the two functional hop tests. 82 Subjects Ten subjects who all had their disrupted A C L reconstructed by two orthopaedic surgeons of the Allan McGavin Sports Medicine Clinic were recruited for this study. The total number of subjects needed for the study were determined by power analysis (Appendix A). Due to ethical considerations, a control group was not established for this study. A l l potential subjects were initially contacted by mail. A list of potential subjects was obtained from the two orthopaedic surgeons who have agreed to work with this study. Subjects were selected based on the following criteria: 1. Individuals who were 18 to 38 years old. 2. Only one knee was involved. 3. An isolated A C L injury that had been surgically reconstructed. (Isolated A C L injury = the collateral ligament(s) did not require surgical intervention and the meniscus (or menisci) was not completely removed) 4. A minimum of 6 months since the date of operation. 5. No previous history of significant injury to or pathology in the A C L reconstructed knee. 6. No significant secondary damage to the involved knee joint as a result of chronic A C L deficiency. (Significant secondary damage = the collateral ligament(s) required surgical intervention and the meniscus (or menisci) was completely removed) 7. Normal hip and ankle joint function. 8. No neurologic diseases. 9. No vestibular or visual disturbances. 10. No arthritic condition(s) involving the lower extremities. 11. Was not attending physiotherapy sessions (within the last 6 weeks). 12. Possessed full range of motion (both active and passive) in the operated knee. 13. Was able to participate in light recreational activities without inducing significant pain or obvious swelling in the operated knee. (Light recreational activities = cycling, swimming, walking) 14. The absence of pain and swelling in the operated knee during normal A D L . 83 Individuals were excluded if they had received intense proprioceptive training post A C L reconstruction. Individuals were also excluded if they had any cardiovascular, respiratory, systemic, metabolic condition limiting exercise tolerance. Prior to participating in the study, the subjects were required to read and fill out the informed consent form. Once informed consent was obtained, subjects were asked to fill out a subject information sheet designed for this study (Appendix B). Ethical approval was acquired from the Committee on Human Experimentation of the University of British Columbia. Subjects were withdrawn from the study if negative effects (such as pain and swelling in the A C L reconstructed knee) induced by the experimental protocol persisted after 48 hours or were made worse with subsequent exercise sessions. However, no subjects were withdrawn from the study secondary to persisting pain or swelling in the A C L reconstructed knee. The 10 subjects were randomly assigned to one of two experimental groups. Group One (Strength Training Group = ST Group) consisted of five subjects who were placed on a general lower limb strength training program (Appendix C). Group Two (Proprioceptive Training Group = PT Group) consisted of five subjects who were placed on a proprioceptive training program6 that had a special emphasis on facilitating rapid hamstring contractions (Appendix C). Group One (ST Group) consisted of three males and two females with an average age of 24.66 years, who had undergone surgical reconstruction for an isolated unilateral A C L rupture (three right, two left). The subjects in Group One were involved in the following sports prior to their A C L injury: soccer, varsity volleyball, ice hockey, rugby, and hiking. Group Two (PT Group) consisted of one male and four females with an average age of 25.04 years, who also had undergone surgical reconstruction for an isolated unilateral A C L rupture (three right, two left). The subjects in Group Two were involved in the following sports prior to their A C L injury: judo (national level), karate, soccer, varsity ultimate, and squash. b If we consider the definitions of proprioception and coordination, it is probably more correct to state that subjects in Group Two were placed on a coordination training program. However, for the purpose of this thesis, the term "proprioceptive training" is used. 84 A l l subjects were asked to perform the assigned exercises three times a week. A l l subjects were asked to come to the University of British Columbia campus three times a week, each time for approximately one hour, to perform their exercises under the supervision of the investigator(s). In some instances, the exercise sessions took place off campus, at locations more convenient for the subjects. Details of each exercise session for all subjects were recorded on a data sheet designed for this study. In addition, all subjects were asked to keep a record of all other physical activities they are involved in for the duration of the study. The main purpose for maintaining these physical activity records was to assist the investigators in determining the accuracy of the Tegner and Lysholm Activity Scale scores. Each subject was required to complete 12 weeks of the assigned exercise protocol. 85 Methods Exercise Protocols Appendix C contains the details and the progression of the exercises prescribed to each experimental group. Compliance was monitored directly through supervision and indirectly through data sheets. The exercises assigned to Group Two (PT Group) can be divided into two types: 1. Balance 2. Coordination Balance is the bodily equilibrium or the ability to maintain the center of body mass over the base of support without falling.48 Coordination can be defined as highly skilled multimuscular automaticity.136 Balance and coordination are closely interrelated as balance can be defined in the following three ways 48 : 1. The ability to maintain a position. 2. The ability to voluntarily move. 3. The ability to react to a perturbation. Throughout the 12 weeks of proprioceptive training, the maintenance of proper alignment of the lower extremities was emphasized during all exercises. In the first 6 weeks symmetry and precision of movements were emphasized. In the latter 6 weeks, speed was gradually introduced and integrated into the exercise program. Measures The following descriptive information was assessed when subjects entered the study: a) Age in years. b) Weight in kilograms. c) Height in centimeters. d) Activity level prior to A C L injury (using the Tegner and Lysholm Activity Scale). e) Time interval from surgery to entering into the study in months. f) Time interval from initial injury to surgery in months. 86 The outcome measures of this study were: 1. Subjective scores: a) Lysholm and Gillquist Knee Scoring Scale b) Tegner and Lysholm Activity Scale 2. Performance on the one-legged single hop for distance (SLHD) and the one-legged timed hop. 3. Peak torque times for each hamstring muscle. 4. Isokinetic average eccentric and concentric torques of the hamstring and quadriceps muscles. Study Design Table 6 ! A and 6.IB demonstrate the study design used for this study. Group Test Occasions For Measures Occasion One: Within 7 days prior to the start of the 12 weeks. Occasion Two: Within 7 days after the completion of the 12 weeks. One (Strength Training) Measures: Subjective Scores Isokinetic Strength Functional Hop Tests Descriptive Information Measures: Subjective Score Isokinetic Strength Functional Hop Tests Two (Proprioceptive Training) Measures: Subjective Scores Isokinetic Strength Functional Hop Tests Descriptive Information Measures: Subjective Scores Isokinetic Strength Functional Hop Tests Table 6.IB: Study Design Group Test Occasions for Measure Occasion One: Within 7 days prior to the start of the 12 weeks. Occasion Two: In the 6 t h week of the 12 weeks. Occasion Three: Within 7 days after the completion of the 12 weeks. One (Strength Training) Measure: Peak Torque Time Measure: Peak Torque Time Measure: Peak Torque Time Two (Proprioceptive Training) Measure: Peak Torque Time Measure: Peak Torque Time Measure: Peak Torque Time 87 Subjective Scores The modified Lysholm and Gillquist Knee Scoring Scale and the Tegner and Lysholm Activity Scale were the two subjective scoring scales used in this study. The purpose of administering the modified Lysholm and Gillquist Knee Scoring Scale and the Tegner and Lysholm Activity Scale is to detect any changes in the subjects' level of activity and function before and after 12 weeks of intervention. Examples of the above subjective scoring scales are included in Appendix B. Function Hop Tests The one-legged single hop for distance (SLHD) and the one-legged timed hop were conducted in the manner described by Booher et al.l29 p o r the one-legged single hop for distance test, a 4 meter measuring tape that was marked in centimeters (and inches) was secured to the floor. Two practice trials, each separated by 30 seconds, were performed by each subject prior to actual testing. A period of 60 seconds was given after subjects completed their practice trials. Subjects were instructed to stand on one leg and to hop as far as possible, landing on the same limb. Subjects were positioned with their toes touching the zero mark. The distance from the zero mark to their heels was measured and recorded. Each limb was tested twice. A 30 second period was given after subjects completed each test trial. The A C L intact limb was always tested first. The mean of the two trials for each limb was calculated and used as the criterion score. For the one-legged timed hop test, a distance of six meters was measured. Two practice trials, each separated by 30 seconds, were performed prior to actual testing. A period of 60 seconds was given after subjects completed their practice trials. Subjects were encouraged to use large one-legged hopping motions to perform a series of hops over the total distance. Subjects were positioned with their toes touching the zero mark. Timing of the test began when the subjects sprang forward and ended when they crossed over the finish mark. Time was recorded by the investigator to the nearest 1/100 second. Each limb was tested twice. A 30 second period was given after subjects completed each test trial. The A C L intact limb was always tested first. The mean of the two trials was calculated and used as the criterion score. 88 Peak Hamstring Torque Time Peak torque times of the hamstring muscles were measured according to the procedure described in the pilot study of the second test occasion. The only change made to that described in the pilot study was the length of the force arm, which was positioned such that it was no longer than 70% of the length of the lower leg (or of the tibia). The A C L intact limb was always tested first. Test results were saved as ASCII files and transferred onto 3.5" computer discs. The ASCII files recorded the factor of "time" in 0.01 second or 10 millisecond intervals. The slowest and the fastest times to hamstring peak torque were eliminated from the six trials performed by each limb of the subjects. The mean of the remaining four trials was used as the criterion score. Unlike the other measures used in this study, PTT was measured on 3 separate test occasions. PTT was measured prior to the start of the prescribed exercise program, at 6 weeks into the program, and upon the completion of the program. Peak hamstring torque times were assessed after muscle torque testing was completed for the tested leg at test occasions one and three. There was a 5 minute rest interval between the completion of muscle torque testing and the assessment of PTT. Muscle Torque (Strength) Muscle torque was measured using the KIN-COM. The KIN-COM has been shown to be a reliable and valid machine for testing and t r a in ing .Quadr i ceps and hamstring (concentric and eccentric) muscle torques were assessed bilaterally at 45 degrees per second. The following test protocols were performed (in order): 1. A C L intact Quadriceps at 457second. 2. A C L intact Hamstring at 457second. 3. A C L reconstructed Quadriceps at 457second. 4. A C L reconstructed Hamstring at 457second. A l l tests for measuring quadriceps muscle torque were performed between 10 and 90 degrees of knee flexion. To protect the A C L graft, the last 10 degrees of knee extension was eliminated during quadriceps muscle torque testing. A l l tests for measuring hamstring muscle torque were performed between 0 and 90 degrees of knee flexion. 89 Gravity compensation was performed for all isokinetic strength testing. A low acceleration rate was used for all tests. Prior to testing, all subjects completed 10 minutes of stationary cycling followed by three repetitions of static quadriceps and hamstring stretching, holding each repetition for 20 seconds. This was to ensure the subjects did not develop muscle cramping while being tested on the KIN-COM. A l l subjects were also given verbal instructions regarding the test protocol and the use of the KIN-COM dynamometer. Once the subject was familiar with the procedure, he/she was seated on the KIN-C O M with their pelvis and tested thigh secured to the bench. The subject's trunk was also comfortable, but firmly, strapped to the back of the KIN-COM bench seat. The axis of motion of the KIN-COM was aligned with the lateral femoral condyle of the tested knee. The dynamometer lever arm was approximately 70% of the length measured from the lateral knee joint line to the distal portion of the medial malleolus. Once positioned, subjects were given a practice session for each muscle tested. Practice sessions consisted of five repetitions, three of which were performed submaximally followed by two which were performed with maximum effort. After a 2 minute rest interval, subjects were asked to perform repetitions with maximal effort until three torque curves that overlapped each other were produced on the K I N - C O M screen. There was a 10 second intercontraction pause between each concentric and eccentric phase. Test results were saved onto 3.5" computer discs. Surgical Procedure The surgical technique used to reconstruct the injured anterior cruciate ligaments is similar to that described by Gomez et al.135 In this technique, the ipsilateral semitendinosus tendon is dissected proximally, but left undisturbed distally. A 6 mm drill hole is placed through the proximal tibia, exiting at the attachment of the A C L . This drill hole is positioned so that the remnants of the original A C L will surround it. A lateral incision 4 cm long is made at the flare of the lateral femoral epicondyle. The iliotibial band is incised and dissection is made into the popliteal space to expose the site of exit of the semitendinosus tendon from the joint through the posterior capsule. The semitendinosus graft is threaded through the drill hole and the remaining stump of the 90 A C L . The resulting ligament/tendon complex is then passed through the posterior capsule of the knee and attached to the lateral intramuscular septum in the over the top fashion. The tendon is fixed to roughened bone just inferior to the attachment of the lateral intermuscular septum, and the stump of the A C L is sutured to the adjacent periosteum. This technique attempts to preserve mechanoreceptors in this area, which would otherwise be removed with other surgical techniques. 91 Statistics Statistical analysis began with an one-way analysis of variance to determine i f there were any significant differences between the two groups on any of the descriptive and outcome measures assessed at test occasion one. Repeated-measures analysis of variance was the chosen statistical tool to determine i f there were any significant differences at the end of the 12 weeks between Group One and Group Two for each of the four outcome measures used in this research study. Correlation Analyses Pearson product-moment correlation coefficients were calculated to determine which of the following variables significantly correlated with the two functional hop tests (performances measured at test occasion two): 1. Average concentric quadriceps torque (measured at the test occasion two). 2. Average eccentric quadriceps torque (measured at test occasion two). 3. Average concentric hamstring torque (measured at test occasion two). 4. Average eccentric hamstring torque (measured at test occasion two). 5. PTT of the hamstring muscle (measured at test occasion three) Those variables which significantly correlated with the two functional hop tests were used for subsequent regressional analyses The level of significance for all statistical tests were set at p<0.05. t 92 Chapter Seven RESULTS Descriptive Data of Subjects Table 7.1 provides a description of the 10 subjects involved in this study. Table 7.1: Descriptive Data of Subjects Variable Group Mean Std. 95% Confidence Minimum Maximum Deviation Interva for Mean Lower Bound Upper Bound Height One 175.10 10.64 161.88 188.32 165.70 190.80 (cm) Two 168.14 10.34 155.30 180.98 158.10 179.70 Weight One 93.30 •35.18 49.62 136.98 61.50 150.00 (kg) Two 67.10 13.96 49.77 84.43 52.00 90.00 Age One 24.66 2.66 21.35 27.97 22.50 29.30 (year) Two 25.04 3.68 20.47 29.61 20.40 30.50 Activity Level One 8.60 1.14 7.18 10.02 7.00 10.00 Prior To Injury (Tegner) Two 8.80 1.10 7.44 10.16 7.00 10.00 Time From Surgery To One 12.20 8.34 L84 22.56 7.00 27.00 Study (month) Two 6.70 0.84 5.66 7.74 6.00 8.00 Time From Injury To One 23.75 26.84 -9.57 57.08 2.00 55.00 Surgery (month) Two 29.00 57.62 -42.55 100.55 1.00 132.00 A l l subjects were right dominant. In Group One, 3 subjects had their right A C L reconstructed and 2 subjects had their left A C L reconstructed. In Group Two, 3 subjects had their right A C L reconstructed and 2 subjects had their left A C L reconstructed. Group One consisted of 3 males and 2 females. Group Two consisted of 1 male and 4 females. An one-way analysis of variance demonstrated no significant differences between the two groups in any of the descriptive data collected at test occasion one. 93 Table 7.2: Descriptive Data of Subjects (One-Way A N O V A ) Variable Sum of Squares df Mean Square F Sig. Height (cm) Between Groups 121.10 1 121.10 1.10 .325 Within Groups 881.27 8 110.16 Weight (kg) Between Groups 1716.10 1 1716.10 2.40 .160 Within Groups 5730.00 8 716.25 Age (year) Between Groups 0.361 1 0.361 .04 .856 Within Groups 82.66 8 10.33 Activity Level Prior To Between Groups 0.10 1 0.100 .08 .784 Injury (Tegner) Within Groups 10.00 8 1.25 Time From Surgery To Between Groups 75.62 1 75.62 2.15 .181 Study (month) Within Groups 281.10 8 35.14 Time From Injury To Between Groups 68.91 1 68.91 .03 .858 Surgery (month) Within Groups 16163.88 8 2020.48 In the one-way analysis of variance conducted to determine if the two groups were different in any of the four outcome measures assessed at test occasion one, a significant difference in average concentric and eccentric torque of the quadriceps muscle of the A C L reconstructed limb was found between the two groups. There are two possibilities for these differences: 1. There are more males in Group One (ST Group) than in Group Two (PT Group). Thus, the significantly greater quadriceps strength (of the A C L reconstructed limb) demonstrated by the strength training group at test occasion one may be secondary to the different female to male ratios between the two groups. 2. Although the time interval from surgery to entering the study was found to be not significantly different between the two groups, the mean time interval for Group One was greater than Group Two. Thus, it is possible that subjects in the strength training group recovered a greater percentage of their pre-injury quadriceps strength than those in the proprioceptive training group. 94 Table 7.3: Measures at r "est Occasion One (One-Way A N O V A ) Torque (Nm) Sum of Squares df Mean Square F Sig. Average Concentric Between Groups 10304.00 1 10304.10 10.0 .013 Torque of Quadriceps 0 Within Groups 8242.00 8 1030.25 Average Eccentric Between Groups 16564.90 1 16564.90 14.4 .005 Torque of Quadriceps 0 Within Groups 9205.20 8 1150.65 Subjects in the strength training group completed a mean of 31.6 + 3.44 sessions while subjects in the proprioceptive training group completed a mean of 31.4 + 1.82 sessions of the total 36 sessions. 95 Subjective Results Lysholm and Gillquist Knee Scoring Scale The following tables provide a description of Lysholm and Gillquist scores obtained from the subjects in Group One and Group Two before and after the prescribed training protocols. Table 7.4: Lysholm and Gillquist Knee Scoring Scale Results Grou Stren 3 One (n=5) gth Training Group Two (n=5) 'roprioceptive Training Subject Pre Post Change Subject Pre Post Change 1 95 100 5 1 91 95 4 2 99 100 1 2 75 86 11 3 95 100 5 3 93 95 2 4 91 96 5 4 88 85 -3 5 83 88 5 5 91 100 9 Table 7.5: Descriptive Data for Lysholm and Gillquist Knee Scoring Scale Scores Mean Std. Deviation 95% Confidence Interval Group Occasion Lower Bound Upper Bound One 1 92.60 6.07 85.70 99.50 (ST) 2 96.80 5.22 90.74 102.85 Two 1 87.60 7.27 80.70 94.50 (PT) 2 92.20 6.46 86.15 98.25 There was a significant difference in the scores between test occasions. There was no significant difference between groups nor was there a group by test occasion interaction effect on the Lysholm and Gillquist scores. An interaction effect is the combined effect of two or more independent variables (i.e. group and test occasion) on a dependent variable (i.e. Lysholm and Gillquist score). 134 Table 7.6: Lysholm and Gillquist Knee Scoring Scale Tests of Within-Su bjects Effects (RM A N O V A ) Source Type III Sum of Squares df Mean Square F Sig. Test Occasion 96.80 1 96.80 11.22 .010 Occasion * Group 0.20 1 0.20 0.02 .883 Error (Occasion) 69.00 8 8.62 96 Tests of Between-Subjects Effects Source Type III Sum of Squares df Mean Square F Sig. Group 115.20 1 115.20 1.63 .237 Error 565.00 8 70.62 110 100. Score 90 J 80 70 O 5 Q 5 One I •*2 Group Two • Pre I I Post Figure 7.1 Lysholm and Gillquist Knee Scoring Scale Scores Pre and Post Intervention vs. Group Figure 7.1 is a boxplot. Boxplots give an overview of within group distributions. The horizontal line in the middle of the boxes marks the median (score) of the sample. The edges (or the hinges) of each box mark the 25 t h (bottom hinge) and the 75 t h (top hinge) percentiles. Thus, the central 50 percent of the data values fall within the range of the box. The length of the box (the difference between the values of the hinges) is called the hspread. The vertical lines extending up and down from each box (whiskers) show the range of values that fall within 1.5 hspreads of the hinges. Boxplots are also very useful in demonstrating any outliers or extremes within the distributions. Outliers are numeric values that does not fall within the range of most scores in a distribution. Extremes are similar to outliers, but more deviant (than outliers) from the distribution. Outliers and extremes can distort statistical associations or correlations. 134 97 Due to the number of subjects used in this study, the presence of outliers and extremes were common. However, unless these outliers or extremes demonstrated a trend that was opposite of the general distribution, they will not be discussed in detail. Figure 7.1 indicates subject 5 of Group One was an outlier while subject 2 of Group Two was an extreme at test occasion one. Both subjects possessed Lysholm and Gillquist scores that were lower than the general distribution. However, at the end of the 12 weeks, subject 2 of Group Two was no longer an extreme while subject 5 of Group One remained an outlier. 110, , 100 . Score 90 , 80 , 70 Group • T w o ° One 1.0 2.0 Test Occasion Figure 7.2 Scatter Plot of Lysholm and Gillquist Scores (Linear Regressional Lines) Figure 7.2 is a scatter plot with linear regressional lines. Scatter plots are useful in graphic representations of the relationship between two variables (i.e. group and test occasion). Figure 7.2 demonstrates the lack of a group by test occasion interaction effect on Lysholm and Gillquist scores. It is clear from Figure 7.2 that both experimental groups made similar gains in Lysholm and Gillquist scores after 12 weeks of training. 98 Tegner and Lysholm Activity Scale The following tables provides a description of the subjective scores obtained from using the Tegner and Lysholm Activity Scale from the subjects in Group One and Group Two before and after the prescribed training protocols. Group One (n=5) Group Two (n=5) Strength Training Proprioceptive Training Subject Pre Post Change Subject Pre Post Change 1 7 8 1 : 1 6 6 0 2 8 8 0 2 5 7 2 3 4 7 3 3 3 6 3 4 4 6 2 4 4 6 2 5 6 7 1 5 5 9 4 Table 7.8: Descriptive Data For Tegner and Lysholm Activity Scale Scores Mean Std. Deviation 95% Confidence Interval Group Occasion Lower Bound Upper Bound One 1 5.80 1.79 4.25 7.35 2 7.20 0.84 6.07 8.33 Two 1 4.60 1.14 3.05 6.15 2 6.80 1.30 5.67 7.93 A significant difference in Tegner and Lysholm scores was found between test occasions. There was no significant difference between groups nor was there a group by test occasion interaction effect on Tegner and Lysholm scores. Table 7.9: Tegner and Lysholm Activity Scale Source Type III Sum of Squares df Mean Square F Sig. Test Occasion 16.20 1 16.20 18.51 .003 Occasion * Group 0.80 1 0.80 0.91 .367 Error (Occasion) 7.00 8 0.88 Tests of Between-Subjects Effects Source Type III Sum of Squares df Mean Square F Sig. Group 3.20 1 3.20 1.24 .297 Error 20.60 8 2.58 99 One Group Two Figure 7.3 Tegner and Lysholm Activity Scale Scores Pre and Post Intervention vs. Group Figure 7.3 indicates that subject 5 of Group Two was an outlier after 12 weeks of proprioceptive training. This subject's Tegner and Lysholm Activity score was higher than the general distribution of Group Two at the end of the 12 weeks. There were no outliers or extremes in Group One. 100 Score Test Occasion Figure 7.4 Scatter Plot of Tegner and Lysholm Scores (Linear Regression Lines) Figure 7.4 indicates the lack of a group by test occasion interaction effect on the Tegner and Lysholm scores. Although subjects in Group Two appeared to demonstrate greater improvements than those in Group One after 12 weeks of training, it was not statistically significant. Thus, both experimental groups made similar gains in Tegner and Lysholm scores after 12 weeks of training. 101 Objective Results Isokinetic Torque (Strength) Isokinetic values used for statistical analyses were average torques generated at 457second. Absolute torque values were used for statistical analyses rather than symmetry indexes. Symmetry indexes (i.e. dividing the A C L reconstructed knee average torque by the A C L intact knee average torque and multiplying the quotient by 100) were not used for statistical analyses because the reliability coefficients (ICC) observed for absolute torques were higher than those observed for ratios or symmetry indexes (Appendix D). Ratio data are particularly appealing clinically because scores may be standardized and the subject may act as his or her own control. 121 Although ratios are frequently used to describe the extent of abnormality, the reliability of ratio data may be questionable. Because the ratio is the quotient of two absolute numbers, each of which may vary in two directions upon repeat testing, the reliability of ratio data may be less than that associated with the absolute scores used to calculate the ratio. 121 Table 7.10: Descriptive Data for Average Torques (Nm) at Test Occasion One Average Torque (Nm) Measured at 457second Group One ( x ± S D ) Group Two ( x ± S D ) Total (x ± SD) Concentric Quadriceps (Nm) n = 5 n = 5 N= 10 A C L Reconstructed 144.40 + 31.42 80.20 ±32.76 112.30 + 45.39 A C L Intact 149.40 ±40.30 116.80 ±30.23 133.10 + 37.72 Eccentric Quadriceps (Nm) n = 5 n = 5 N = 10 A C L Reconstructed 161.00 + 38.25 79.60 ±28.95 120.30 ±53.51 A C L Intact 172.60 + 59.23 142.00 ±43.14 157.30 ±51.44 Concentric Hamstring (Nm) n = 5 n = 5 N = 10 A C L Reconstructed 71.00 ±24.63 53.20+ 15.53 62.10 + 21.56 A C L Intact 80.80 ±26.58 61.00 ±23.95 70.90 + 26.04 Eccentric Hamstring (Nm) n = 5 n = 5 N = 1 0 A C L Reconstructed 85.40 ±28.35 58.60+ 17.16 72.00 ± 26.22 A C L Intact 93.80 + 37.62 78.00 + 26.62 85.90 + 31.83 102 Table 7 ! 1: Descriptive Data for Average Torques (Nm) at Test Occasion Two Average Torque (Nm) Measured at 457second Group One (x ± SD) Group Two fx±SD) Total (x ± SD) Concentric Quadriceps (Nm) n = 5 n = 5 N = 10 A C L Reconstructed 142.80 + 40.91 104.20 ±34.66 123.50±41.13 A C L Intact 154.80 ±36.81 132.60 ±31.20 143.70 ±34.23 Eccentric Quadriceps (Nm) n = 5 n = 5 N = 10 A C L Reconstructed 164.00 ±46.67 116.80 ±37.92 140.40 ±47.18 A C L Intact 175.40 ±37.18 150.40 ±34.36 162.90 ±36.23 Concentric Hamstring (Nm) n = 5 n = 5 N = 10 A C L Reconstructed 73.00 ± 16.93 61.20 +19.72 67.10+18.41 A C L Intact 79.20 ±24.61 65.40 ±23.42 72.30 ±23.79 Eccentric Hamstring (Nm) n = 5 n = 5 N = 10 A C L Reconstructed 85.40 + 23.74 71.60 ± 17.17 78.50 ±20.84 A C L Intact 99.40 + 32.89 85.00 ±22.44 92.20 ±27.61 Average Concentric Quadriceps Torque of the A C L Reconstructed Limb There was a significant difference between Group One and Group Two in the average concentric quadriceps torque of the A C L reconstructed limb. Also, there was a significant difference between the two test occasions and a significant group by test occasion interaction effect on the average concentric quadriceps torque. Table 7.12: Average Concentric Quadriceps Torque of the A C L Reconstructed Limb Tests of Within-Su bjects Effects (RM A N O V A ) Source Type III Sum of Squares df Mean Square F Sig. Test Occasion 627.20 1 627.20 7.71 .024 Occasion * Group 819.20 1 819.20 10.073 .013 Error (Occasion) 650.60 8 81.32 Tests of Between-Subjects Effects Source Type III Sum of Squares df Mean Square F Sig. Group 13209.80 1 13209.80 5.54 .046 Error 19091.00 8 2386.38 103 300 200. Nm 03 100. • Pre • Post One Group Two Figure 7.5 Average Concentric Quadriceps Torque of the A C L Reconstructed Limb Pre and Post Intervention vs. Group Figure 7.5 indicates that subject 3 of Group Two was an outlier after 12 weeks of proprioceptive training. The subject obtained an average concentric quadriceps torque of the A C L reconstructed limb that was higher than the general distribution at the end of the 12 weeks. There was no outliers or extremes in Group One. 104 220 200 180 160 Concentric Quad 140 \ (Nm) 120 100 80 60 40 Test Occasion Figure 7.6 Scatter Plot of Average Concentric Quadriceps Torque of the A C L Reconstructed Limb (Linear Regressional Lines) Figure 7.6 demonstrates an group by test occasion interaction effect on the average concentric quadriceps torque of the A C L reconstructed limb. It is clear from Figure 7.6 that Group Two made greater gains in average concentric quadriceps torque of the A C L reconstructed knee than Group One after 12 weeks of training. Average Concentric Quadriceps Torque of the A C L Limb There was no significant difference between Group One and Two in the average concentric quadriceps torque of the A C L intact limb, nor was there a significant group by test occasion interaction effect. There was, however, a significant difference in the average concentric quadriceps torque between test occasions. Table 7.13: Average Concentric Quadriceps Torque of the A C L Intact Limb Tests of Within-Su bjects Effects (RM A N O V A ) Source Type III Sum of Squares df Mean Square F Sig. Test Occasion 561.80 1 561.80 9.15 .016 Occasion * Group 135.20 1 135.20 2.20 .176 Error (Occasion) 491.00 8 61.38 105 Tests of Between-Subjects Effects Source Type III Sum of Squares df Mean Square F Sig. Group 3753.800 1 3753.800 1.583 .244 Error 18971.000 8 2371.375 220 One Group Two Figure 7.7 Average Concentric Quadriceps Torque of the A C L Intact Limb Pre and Post Intervention vs. Group Figure 7.7 indicates that subject 3 of Group Two was an outlier at test occasion two. The subject obtained an average concentric quadriceps torque of the A C L intact limb that was higher than the general distribution of Group Two after 12 weeks of training. There was no outliers or extremes in Group One. 106 220 Test Occasion Figure 7.8 Scatter Plot of Average Concentric Quadriceps Torque of the A C L Intact Limb (Linear Regressional Lines) Figure 7.8 indicates the lack of a group by test occasion interaction effect on the average concentric quadriceps torque of the A C L intact limb. It is clear from Figure 7.8 that both experimental groups made similar gains in the average concentric quadriceps torque of the A C L intact limb after 12 weeks of training. Average Eccentric Quadriceps Torque of the A C L Reconstructed Limb There was a significant difference between Group One and Group Two in the average eccentric quadriceps torque of the A C L reconstructed limb. Also, there was a significant difference between the two test occasions and a significant group by test occasion interaction effect on the average eccentric quadriceps torque of the A C L reconstructed limb. 107 Table 7.14: Average Eccentric Quadriceps Torque of the A C L Reconstructed Limb Tests of Within-Su bjects Effects (RM A N O V A ) Source Type III Sum of Squares df Mean Square F Sig. Test Occasion 2020.05 1 2020.05 8.03 .022 Occasion * Group 1462.05 1 1462.05 5.81 .042 Error (Occasion) 2012.40 8 251.55 Tests of B etween-Subjects Effects Source Type III Sum of Squares df Mean Square F Sig. Group 20672.45 1 20672.45 7.64 .025 Error 21659.60 8 2707.45 300 200 Nm 100 One Group Two Figure 7.9 Average Eccentric Quadriceps Torque of the A C L Reconstructed Limb Pre and Post Intervention vs. Group Figure 7.9 indicates that subject 2 of Group Two was an outlier at test occasion one. This subject had an average eccentric quadriceps torque of the A C L reconstructed limb that was lower than the general distribution of Group Two at test occasion one. However, the subjects was no longer an outlier by test occasion two. There were no outliers or extremes in Group One. 108 300 200 Eccentric Quad (Nm) 100 1 Figure 7.10 Scatter Plot of Average Eccentric Quadriceps Torque of the A C L Reconstructed Limb (Linear Regressional Lines) Figure 7.10 demonstrates a group by test occasion interaction effect on the average eccentric quadriceps torque of the A C L reconstructed limb. It is clear from Figure 7.10 that Group Two made greater gains in average eccentric quadriceps torque of the A C L reconstructed limb than Group One after 12 weeks of training. Average Eccentric Quadriceps Torque of the A C L Intact Limb There was no significant difference between Group One and Group Two in the average eccentric quadriceps torque of the A C L intact limb. Also, there was no significant difference between the test occasions and no significant group by test occasion interaction effect on the average eccentric quadriceps torque of the A C L intact limb. Group • T w o ° One 2.0 Test Occasion 109 Table 7.15: Average Eccentric Quadriceps Torque of the A C L Intact Limb Source Type III Sum of Squares df Mean Square F Sig. Test Occasion 156.80 1 156.80 .43 .529 Occasion * Group 39.20 1 39.20 .11 .751 Error (Occasion) 2903.00 8 362.88 Tests of Between-Subjects Effects Source Type III Sum of Squares df Mean Square F Sig. Group 3864.20 1 3864.20 1.07 .331 Error 28824.60 8 3603.08 Average Concentric Hamstring Torque of the A C L Reconstructed Limb There was no significant difference between Group One and Group Two in the average concentric hamstring torque of the A C L reconstructed knee. Also, there was no significant difference between the two test occasions and no significant group by test occasion interaction effect on the average eccentric hamstring torque of the A C L reconstructed limb. Table 7.16: Average Concentric Hamstring Torque of the A C L Reconstructed Limb Tests ofWithin-Su bjects Effects (RM A N O V A ) Source Type III Sum of Squares df Mean Square F Sig. Test Occasion 125.00 1 125.00 2.46 .156 Occasion * Group 45.00 1 45.00 .88 .374 Error (Occasion) 407.00 8 50.88 Tests of Between-Subjects Effects Source Type III Sum of Squares df Mean Square F Sig. Group 1095.20 1 1095.20 1.54 .250 Error 5684.60 8 710.58 110 Nm One Two Group Figure 7.11 Average Concentric Hamstring Torque of the A C L Reconstructed Limb Pre and Post Intervention vs. Group Figure 7.11 indicates that subject 3 of Group Two was an extreme score at test occasion two. This subject obtained an average concentric hamstring torque of the A C L reconstructed knee that was higher than the general distribution of Group Two after 12 weeks of training. There were no outliers or extremes in Group One. Average Concentric Hamstring Torque of the A C L Intact Limb There was no significant difference between Group One and Group Two in the average concentric hamstring torque of the A C L intact limb. Also, there was no significant difference between the test occasions and no significant group by test occasion interaction effect on average concentric hamstring torque of the A C L intact limb I l l Table 7.17: Average Concentric Hamstring Torque of the A C L Intact Limb Tests of Within-Su bjects Effects (RM A N O V A ) Source Type III Sum of Squares df Mean Square F Sig. Test Occasion 9.80 1 9.800 .39 .550 Occasion * Group 45.00 1 45.000 1.79 .218 Error (Occasion) 201.20 8 25.150 Tests of Between-Subjects Effects Source Type III Sum of Squares df Mean Square F Sig. Group 1411.20 1 1411.20 1.18 .308 Error 9535.60 8 1191.95 Average Eccentric Hamstring Torque of the A C L Reconstructed Limb There was no significant difference between Group One and Group Two in the average eccentric hamstring torque of the A C L reconstructed limb. However, there was a significant difference between the test occasions and a significant group by test occasion interaction effect on the average eccentric hamstring torque of the A C L reconstructed limb. Table 7.18: Average Eccentric Hamstring Torque of the A C L Reconstructed Limb Source Type III Sum of Squares df Mean Square F Sig. Test Occasion 211.25 1 211.25 5.50 .047 Occasion * Group 211.25 1 211.25 5.50 .047 Error (Occasion) 307.00 8 38.38 Tests of Between-Subjects Effects Source Type III Sum of Squares df Mean Square F Sig. Group 2060.45 1 2060.45 2.19 .177 Error 7519.80 8 939.98 112 140 One Group T w o Figure 7.12 Average Eccentric Hamstring Torque of the A C L Reconstructed Limb Pre and Post Intervention vs. Group There were no outliers or extremes for either experimental group. 113 120 100 80 Eccentric Hamstring (Nm) v } 60 40 20 Test Occasion Figure 7.13 Scatter Plot of Average Eccentric Hamstring Torque of the A C L Reconstructed Limb (Linear Regressional Lines) Figure 7.13 demonstrates a group by test occasion interaction effect on average eccentric hamstring torque of the A C L reconstructed limb. It is clear from Figure 7.13 that Group Two made greater gains in average eccentric hamstring torque of the A C L reconstructed limb than Group One after 12 weeks of training. Average Eccentric Hamstring Torque of the A C L Intact Limb There was no significant difference between Group One and Group Two in the average eccentric hamstring torque of the A C L intact limb, nor was there a group by test occasion interaction effect. There was, however, a significant difference in average eccentric hamstring torque of the A C L intact limb between test occasions. Table 7.19: Average Eccentric Hamstring Torque of the A C L Intact Limb Tests ofWithin-Su bjects Effects (RM A N O V A ) Source Type III Sum of Squares df Mean Square F Sig. Test Occasion 198.45 1 198.45 7.30 .027 Occasion * Group 2.45 1 2.45 .09 .772 Error (Occasion) 217.60 8 27.20 114 Tests of Between-Subjects Effects Source Type III Sum of Squares df Mean Square F Sig. Group 1140.05 1 1140.05 .62 .452 Error 14620.40 8 1827.55 160 140 120 Nm 100 80 60 40 Q 3 * 3 * 4 • Pre Post One Group Two Figure 7.14 Average Eccentric Hamstring Torque of the A C L Intact Limb Pre and Post Intervention vs. Group Figure 7.14 indicates subject 3 of Group Two as an outlier at test occasion one and an extreme score at test occasion two. This indicates that subject 3 had a average eccentric hamstring torque of the A C L intact knee that was higher than the general distribution of Group Two at both test occasions. Subject 4 of Group Two was an extreme (lower) score at test occasion two. 115 160 140 120 Eccentric Hamstring (Nm) 100 Test Occasion Figure 7.15 Scatter Plot of Average Eccentric Hamstring Torque of the A C L Intact Limb (Linear Regressional Lines) Figure 7.15 demonstrates the lack of a group by test occasion interaction effect on the average eccentric hamstring torque of the A C L intact knee. It is clear from Figure 7.15 that both experimental groups made similar gains in average eccentric hamstring torque of the A C L intact knee after 12 weeks of training. Functional Hop Tests Symmetry indexes were not used for statistical analyses of functional hop tests because the reliability coefficients (ICC) observed for absolute values were higher than those for ratios or symmetry indexes (Appendix D). Therefore, absolute values were used for statistical analyses. 116 Table 7.20: Descriptive Data for Functional Hop Tests at Test Occasion One Functional Hop Test Group One ( x ± S D ) Group Two (x ± SD) Total (x ± SD) Timed Hop (sec) n = 5 n = 5 N = 10 A C L Reconstructed 2.34 ± 0.22 2.75 + 0.58 2.54 ±0.47 A C L Intact 2.27 + 0.25 2.21 ±0.20 2.24 ± 0.22 Single Hop For Distance (cm) n = 5 n = 5 N = 10 A C L Reconstructed 160.34 ±33.76 133.30 ±38.32 146.82 + 38.32 A C L Intact 168.20 ±29.94 165.61 +24.02 166.90 ±25.63 Table 7.21: Descriptive Data for Functional -lop Test at Test Occasion Two Functional Hop Test Group One (x ± SD) Group Two (x ± SD) Total (x ± SD) Timed Hop (sec) n = 5 n = 5 • N = 10 A C L Reconstructed 1.82 ±0.25 2.03 + 0.23 1.92 ±0.25 Normal 1.74 ±0.26 1.84 ± 0.15 1.79 ±0.21 Single Hop for Distance (cm) n = 5 n = 5 N = 10 A C L Reconstructed 181.04 ±32.57 150.95 + 35.90 166.00 ±36.00 A C L Intact 179.25 ±23.76 171.34 ± 29.10 175.24 + 25.37 One-Legged Timed Hop Performance of the A C L Reconstructed Limb There was no significant difference between Group One and Group Two in the performance of the one-legged timed hop, nor was there a significant group by test occasion interaction effect. There was, however, a significant difference in the performance of the one-legged timed hop between test occasions. Table 7.22: One-Legged Time Hop Performance of the A C L Reconstructed Limb Tests ofWithin-Su bjects Effects (RM A N O V A ) Source Type III Sum of Squares df Mean Square F Sig. Test Occasion 1.925 1 1.925 30.694 .001 Occasion * Group 4.851E-02 1 4.851E-02 .773 .405 Error (Occasion) 0.502 8 6.272E-02 Tests of Between-Subjects Effects Source Type III Sum of Squares df Mean Square F Sig. Group .476 1 0.476 2.524 .151 Error 1.508 8 0.189 117 One Group Two Figure 7.16 One-Legged Timed Hop Performance of the A C L Reconstructed Limb Pre and Post Intervention vs. Group Figure 7.16 indicates subject 2 of Group Two was an outlier at both test occasions. This subject's performance on the one-legged timed hop was worse than the general distribution of performances for Group Two. Subject 5 of Group Two was also an outlier, but only at test occasion two. This subject's performance on the one-legged timed hop was better than the general distribution. Subject 5 of Group One was an outlier at test occasion one. The subject's performance on the one-legged timed hop was better than the general distribution of Group One at test occasion one. 118 4.0 3.5 . 3.0 T H (sec) 2.5 2.0 15 4 1.0 B ° e e" D - • e Group • Two D One 1.0 2.0 Test Occasion Figure 7.17 Scatter Plot of One-Legged Timed Hop Performance of the A C L Reconstructed Limb (Linear Regressional Lines) Figure 7.17 demonstrates the lack of a group by test occasion interaction effect on the one-legged timed hop performance of the A C L reconstructed limb. It is clear from Figure 7.17 that both groups made very similar improvements in the performance of the one-legged timed hop. One-Legged Timed Hop Performance of the A C L Intact Limb There was no significant difference between Group One and Group Two in the one-legged timed hop performance of the A C L intact limb, nor was there a significant group by test occasion interaction effect. There was, however, a significant difference in the one-legged timed hop performance of the A C L intact limb between test occasions Table 7.23: One-Legged Timed Hop Performance of the A C L Intact Limb Source Type III Sum of Squares df Mean Square F Sig. Test Occasion 1.01 1 1.01 44.76 .000 Occasion * Group 3.121E-02 1 3.121E-02 1.39 .273 Error (Occasion) 0.18 8 2.252E-02 119 Tests of Between-Subjects Effects Source Type III Sum of Squares df Mean Square F Sig. Group 2.205E-03 1 2.205E-03 .029 .868 Error .600 8 7.504E-02 One Group Two Figure 7.18 One-Legged Timed Hop Performance of the A C L Intact Limb Pre and Post Intervention vs. Group Figure 7.18 indicates subject 3 and 5 of Group Two as extremes scores at test occasion two. The performance of subject 3 on the one-legged time hop was worse than the general distribution of Group Two at test occasion two. The performance of subject 5, however, was better than the general distribution at test occasion two. Subject 3 of Group One was an outlier at both test occasions. The performance of subject 3 on the one-legged timed hop was slower than the general distribution of Group One. 120 2.2 TH (sec) Group • Two D One Test Occasion Figure 7.19 Scatter Plot of One-Legged Timed Hop Performance of the A C L Intact Limb (Linear Regressional Lines) Figure 7.19 demonstrates the lack of a group by test occasion interaction effect on the one-legged timed hop performance of the A C L intact limb. Although subjects in Group One demonstrated a greater improvements in their one-legged timed hop performance on the A C L intact knee after the 12 weeks, it was not significant. Thus, both experimental groups made similar improvements in the one-legged timed hop performance of the A C L intact limb after 12 weeks of training. One-Legged Single Hop for Distance Performance of the A C L Reconstructed Limb There was no significant difference between Group One and Group Two in the one-legged single hop for distance performance of the A C L reconstructed limb, nor was there a significant group by test occasion effect. There was, however, a significant difference in the performance of the one-legged single hop for distance between test occasions. 121 Table 7.24: SLHD Performance of the A C L Reconstructed Limb Test Source Type III Sum of Squares df Mean Square F Sig. Occasion 1838.40 1 1838.40 26.31 .001 Occasion * Group 11.63 1 11.63 0.17 .694 Error (Occasion) 558.94 8 69.87 Tests of Between-Subjects Effects Source Type III Sum of Squares df Mean Square F Sig. Group 4079.80 1 4079.80 1.61 .240 Error 20232.11 8 2529.01 300. 200 Cm 100 Q 2 One 03 •*2 • Pre Post Group Two Figure 7.20 SLHD Performance of the A C L Reconstructed Limb Pre and Post Intervention vs. Group Figure 7.20 indicates subject 3 of Group Two as an outlier (high score) and subject 2 as an extreme (low score) at test occasion one. However, both subjects were within the general distribution of Group Two at test occasion two. Subject 2 of Group One was an outlier at test occasion two. This subject's performance on the one-legged single hop for distance at test occasion two was better than the general distribution of Group One. 122 300 200. SLHD (cm) 100 . 0 1.0 Test Occasion Figure 7.21 Scatter Plot of SLHD Performance of the A C L Reconstructed Limb (Linear Regressional Lines) Figure 7.21 demonstrates the lack of a group by test occasion interaction effect on the one-legged single hop for distance performance of the A C L reconstructed knee. It is clear from Figure 7.21 that both experimental groups made very similar improvement in the one-legged single hop for distance performance of the A C L reconstructed limb. One-Legged Single Hop for Distance Performance for the A C L Intact Limb There was no significant difference between Group One and Group Two in the one-legged single hop for distance performance of the A C L intact limb, nor was there a significant group by test occasion interaction effect. However, there was a significant difference in the performance of the one-legged single hop for distance between test occasions. • • • • • • i • Y Group • T w o D One 123 Table 7.25: SLHD Performance of the A C L Intact Limb Tests of Within-Subjects Effects (RM [ A N O V A ) Source Type III Sum of Squares df Mean Square F Sig. Test Occasion 351.961 1 351.961 9.820 .014 Occasion * Group 35.378 1 35.378 .987 .350 Error (Occasion) 286.732 8 35.841 Tests of Between-Subjects Effects Source Type III Sum of Squares df Mean Square F Sig. Group 137.812 1 137.812 .098 .762 Error 11254.158 8 1406.770 240 220 i 200 Cm 180 160 140 120 02 IMHiiill One Two • Pre • Post Group Figure 7.22 SLHD Performance of the A C L Intact Limb Pre and Post Intervention vs. Group Figure 7.22 indicates subject 2 of Group One as an outlier at test occasion two. This subject performed one-legged single hop for distance performance better than the general distribution of Group One at test occasion two. 124 220 Test Occasion Figure 7.23 Scatter Plot of SLHD Performance of the A C L Intact Limb (Linear Regressional Lines) Figure 7.23 demonstrates the lack of a group by test occasion interaction effect on the one-legged single hop for distance performance of the A C L intact knee. It is clear from Figure 7.23 that both experimental groups made similar improvements in the one-legged single hop for distance performance of the A C L intact knee. Peak Torque Time Table 7.26 contains the descriptive data for peak torque times obtained from the three test occasions. 125 Table 7.26: Descriptive Data for PTT PTT (msec) Group One ( x ± S D ) Group Two (x ± SD) Total (x ± SD) Occasion One n = 5 n = 5 N = 10 A C L Reconstructed 482.50 + 24.87 478.00 ±20.57 480.25 ±21.65 A C L Intact 499.00 ± 32.09 498.00 ± 34.07 498.50 ±31.21 Occasion Two n=5 n = 5 N = 10 A C L Reconstructed 498.50 + 23.22 456.00 ±6.52 477.25 + 27.57 A C L Intact 487.00 ±35.86 493.50 ±33.01 490.25 ± 32.67 Occasion Three n = 5 n = 5 N = 10 A C L Reconstructed 480.00 ± 19.12 472.00+ 11.10 476.00 ± 15.33 A C L Intact 506.00 ±28.59 490.00 + 29.84 498.00 ±28.81 Peak Torque Time for the A C L Reconstructed Limb There was no significant difference between Group One and Group Two in the peak hamstring torque times, nor was there a difference in PTT between test occasions. There was, however, a significant group by test occasion interaction effect on PTT. Table 7.27: PTT of the A C L Reconstructed Limb Tests ofWithin-Su bjects Effects (RM A N O V A ) Source Type III Sum of Squares df Mean Square F Sig. Test Occasion 95.42 2 47.71 0.23 .798 Occasion * Group 2205.42 2 1102.71 5.29 .017 Error (Occasion) 3332.50 16 208.28 Tests of Between-Subjects Effects Source Type III Sum of Squares df Mean Square F Sig. Group 2520.83 1 2520.83 3.94 .082 Error 5117.50 8 639.69 126 540 One Group Two Figure 7.24 Peak Torque Time of the A C L Reconstructed Limb Pre and Post Intervention vs. Group Figure 7.24 indicates subject 4 of Group One as an outlier at test occasion three. This subject had a slower PTT than the general distribution of Group One. Subjects 3 and 5 of Group Two are indicated by Figure 7.24 as extreme scores at test occasion two. These two subjects were no longer extremes at test occasion three. 127 540 Test Occasion Figure 7.25 Scatter Plot of PTT of A C L Reconstructed Limb (Linear Regressional Lines) It is evident in Figure 7.25 that the linear regressional lines does not provide an adequate description of the group by test occasion interaction effect on PTT. The scatter plot indicates there may be a curvilinear relationship rather than a linear relationship between the PTT and test occasion. 128 540 520 , 500 . 480 , PTT (ms) V 460 , 440 Group • Two D One 1.0 2.0 3.0 Test Occasion Figure 7.26 Scatter Plot of PTT of A C L Reconstructed Limb (Quadratic Regressional Lines) Figure 7.26 provides a better description of the group by test occasion interaction effect on PTT of the A C L reconstructed limb. It is evident the two experimental groups demonstrated very different trends in PTT within the 12 weeks of training. Peak Torque Times for the A C L Intact Limb There was no significant difference between Group One and Group Two for PTT of the A C L intact limb. Also, there was no significant difference in PTT between test occasions and no significant group by test occasion interaction effect. Table 7.28: PTT of the A C L Intact Limb Tests of Within-Su bjects Effects (RM A N O V A ) Source Type III Sum of Squares df Mean Square F Sig. Test Occasion 427.92 2 213.96 0.30 .748 Occasion * Group 656.25 2 328.12 0.45 .644 Error (Occasion) 11603.33 16 725.21 Tests of Between-Subjects Effects Source Type III Sum of Squares df Mean Square F Sig. Group 91.88 1 91.88 0.05 .821 Error 13491.67 8 1686.46 130 Results of Regressional Analyses Correlational Analyses Tables 7.29 and 7.30 summarize the statistically significant results of correlational analyses. There were no significant correlations found between any of the six variables and one-legged timed hop. Four of the six variables were found to have significant correlations with the one-legged single hop for distance. Table 7.29: Pearson Product-Moment Correlation Coefficients for A C L Intact Limb SLHD Coefficient Significance (two-tailed) Concentric Quadriceps 0.637 0.033 Eccentric Quadriceps 0.653 0.041 Concentric Hamstring 0.719 0.019 Eccentric Hamstring 0.658 0.038 Table 7.30: Pearson Proc Limb uct-Moment Correlation Coefficients for , SLHD Coefficient Significance (two-tailed) Concentric Quadriceps 0.856 0.002 Eccentric Quadriceps 0.777 0.008 Concentric Hamstring 0.849 0.002 Eccentric Hamstring 0.667 0.035 Regressional Analyses Regressional analyses were performed to determine the relative contribution of isokinetic torque (strength) to the performance of one-legged single hop for distance test. Two regressional analyses were performed to determine the effects of peak torque time and isokinetic torque on the performance of the two functional hop tests. They were: 1. A C L reconstructed knee (N = 10). 2. A C L intact knee (N = 10). Regressional Analyses for the A C L Reconstructed Limb The effects of isokinetic torque on functional performance (SLHD) of the A C L reconstructed knee were tested by regressional analyses. The following variables had significant effects on functional performance: 1. Average concentric torque of the quadriceps. Table 7.31: Average Concentric Quadriceps Torque Effect (ACL Reconstructed) R R Square Adjusted R Square Std. Error of the Estimate 0.856 0.733 0.699 19.7353 240 220 200 SLHD (cm) 1 6 0 Rsq = 0.7329 80 100 120 140 160 180 200 220 Average Concentric Quadriceps Torque (Nm) Figure 7.27 Average Concentric Quadriceps Torque Effect of the A C L Reconstructed Limb 132 2. Average eccentric torque of the quadriceps. Table 7.32: Average Eccentric Quadriceps Torque Effect (ACL Reconstructed) R R Square Adjusted R Square Std. Error of the Estimate 0.777 0.604 0.554 24.0281 240 220 200 180 SLHD (cm) 1 6 0 140 Rsq = 0.6040 100 120 140 160 180 200 220 Averge Eccentric Quadriceps Torque (Nm) Figure 7.28 Average Eccentric Quadriceps Torque Effect of the A C L Reconstructed Limb 133 3. Average concentric torque of the hamstring muscle. Table 7.33: Average Concentric Hamstring Torque Effect (ACL Reconstructed) R R Square Adjusted R Square Std. Error of the Estimate 0.849 0.720 0.685 20.1942 240 220 200 SLHD (cm) Rsq = 0.7203 50 60 70 80 90 100 Average Concentric Hamstring Torque(Nm) Figure 7.29 Average Concentric Hamstring Torque Effect of the A C L Reconstructed Limb 134 4. Average eccentric torque of the hamstring muscle. Table 7.34: Average Eccentric Hamstring Torque Effect (ACL Reconstructed) R R Square Adjusted R Square Std. Error of the Estimate 0.667 0.445 0.375 28.4559 240 220 200 180 SLHD (cm) 1 6 0 i 140 120 100 Rsq = 0.4446 50 60 70 80 90 100 110 120 Average Eccentric Hamstring Torque (Nm) Figure 7.30 Average Eccentric Hamstring Torque Effect of the A C L Reconstructed Limb Regressional Analyses for the A C L Intact Limb The effects of isokinetic torque on functional performance (SLHD) of the A C L intact knee were tested by regressional analyses. Analyses found the following variables had significant effects on functional performance: 1. Average concentric torque of the quadriceps muscle R R Square Adjusted R Square Std. Error of the Estimate 0.673 0.452 0.384 19.9104 240 220 200 180 SLHD ( c m ) l 6 0 140 120 100 Rsq = 0.5017 100 120 140 160 180 200 220 Average Concentric Quadriceps Torque (Nm) Figure 7.31 Average Concentric Quadriceps Torque Effect of the A C L Intact Limb 2. Average eccentric torque of the quadriceps muscle. R R Square Adjusted R Square Std. Error of the Estimate 0.653 0.427 0.355 20.3743 240 220 200 180 SLHD (cm) 1 6 0 140 120 100 Rsq = 0.3584 120 140 160 180 200 220 240 Average Eccentric Quadriceps Torque (Nm) Figure 7.32 Average Eccentric Quadriceps Torque Effect of the A C L Intact Limb 3. Average concentric torque of the hamstring muscle. R R Square Adjusted R Square Std. Error of the Estimate 0.797 0.635 0.590 16.2508 240 220 200 SLHD (cm) 180 Rsq = 0.6112 50 60 70 80 90 100 110 Average Concentric Hamstring Torque (Nm) Figure 7.33 Average Concentric Hamstring Torque Effect of A C L Intact Limb 4. Average eccentric torque of the hamstring muscle. R R Square Adjusted R Square Std. Error of the Estimate 0.614 0.377 0.299 21.2433 240 220 200 SLHD (cm) Rsq = 0.5158 80 100 120 140 160 Average Eccentric Hamstring Torque (Nm) Figure 7.34 Average Eccentric Hamstring Torque Effect of the A C L Intact Limb 139 Chapter Eight DISCUSSION In the last decade, proprioception training has been emphasized in the literature as an important component of rehabilitation programs for individuals with A C L injuries. Numerous studies have documented altered knee proprioception after A C L injuries. However, the effectiveness of proprioceptive training in A C L deficient or A C L reconstructed limbs is not well established in the literature. To the best of the investigators' knowledge, this is the first prospective and randomized study examining the effectiveness of proprioceptive training in individuals with A C L reconstructed knees. Thus, it is difficult to directly compare the results of this study with current findings in the literature. In addition, the method in which PTT was measured in this study has not been used previously as an outcome measure. The method used to measure PTT was duplicated from that described by Small et a l . l , w j m ^ e exception of the forward angular velocity of the dynamometer lever arm. Thus, the validity of direct comparisons between this study's PTT results and those of the literature may be questionable. The results of this study indicate that both strength training and proprioception training (of the lower extremities) have an influence on peak hamstring torque time of the A C L reconstructed limb. It is proposed that they influence PTT through different neuromuscular mechanisms. Statistically, this influence was found to be dependent on the interaction of two factors: 1. The type of training. 2. The duration of training. Peak torque time is a measure of volitional activity of muscle(s) and we propose it relies on the following factors: 1. The sensitivity of the muscle spindles. 2. Kinesthetic awareness of the knee joint. 3. The ability to recruit as many motor units as possible in the shortest amount of time. 4. The coordination of the limb to the specified task. It is the postulation of this study that there is sensory information originating from the A C L graft and joint afferents that affects the primary spindle afferents via the gamma efferent muscle spindle system. Thus, the A C L graft and joint afferents may contribute to 140 the continuous preparatory adjustment of muscular stiffness through the gamma efferent muscle spindle system. PTT can be regarded as the ability of the hamstring muscle to increase its stiffness at the time of a displacement (i.e. the forward movement of the dynamometer lever arm). As indicated in the review of the literature, functional joint stability is greatly determined by joint load caused by the contraction (or stiffness) of its surrounding muscles.35 Hence, PTT may be regarded as a dynamic measure of the neuromuscular function of the knee joint. A curvilinear relationship between PTT and test occasion was observed for both experimental groups. The strength training group demonstrated a slowing of PTT while the proprioceptive training group demonstrated an improvement of PTT at 6 weeks. Both experimental groups regressed toward their baseline PTT values at the end of the 12 weeks, such as the strength training group demonstrated an improvement in PTT while the proprioceptive training group demonstrated a slowing of PTT in the latter 6 weeks. The trends observed in the strength training group and the proprioceptive training group at 6 weeks are supported by findings of Wojtys et al.l 10. j n their 6 week study, Wojtys et al.HO observed the time to PTT for knee flexion improved by the largest margin (38 ms) in the agility trained group compared to the isotonic or isokinetic trained group. Wojtys et al.l 10 a i s o observed that the time to PTT for knee flexion slowed (31 ms) in the isotonic trained group. However, these observations were not statistically significant between the three groups (agility, isotonic, and isokinetic). The improvement in PTT observed by Wojtys et al. 110 and by investigators of this study is proposed to be due to improved coordination and neural activation of the A C L reconstructed limb secondary to the prescribed proprioceptive training. Coordination is defined as highly skilled multimuscular automaticity.l3^ The ability to coordinate depends on: 1. Appropriate firing and timing of specific musculature to produce the desired motor pattern. 2. Appropriate inhibition of specific musculature not involved in the desired motor pattern. 3. Practice. 141 ' The appropriate neural activation and inhibition of specific musculature relies on proprioception and kinesthesia. According to Kottkel365 the most important aspect of practice to develop coordination engram is the building of inhibition so that muscles that should not participate in the pattern are inhibited. Wojtys et al.l 10 had difficulty explaining the slowing of PTT demonstrated by their isotonic group at the end of the 6 weeks. The findings of Wojtys et al.l 10 G f this study at 6 weeks are in contrast to those of Hakkinen and Komi92. Hakkinen and Komi92 conducted a study examining the effects of heavy resistance strength training, 3 times per week for 16 weeks, on neuromuscular performance both in voluntary and reflex contractions of the quadriceps muscle. These investigators reported an improvement in the time to peak muscle torque of the quadriceps muscle at 8 weeks and at 16 weeks. This improvement was attributed to an increase in the fast twitch/slow twitch (FT/ST) muscle area ratio. A fast motor unit produces a higher tension and reaches a peak twitch tension in a shorter time than a slow unit.92 A possible rationale for the slowing of PTT demonstrated by the strength training group at 6 weeks may be the lack of similarity between the training mode and the testing mode. The emphasis of the strength training program (used in this study) was to recruit the maximum number of motor units possible so that a maximum force or muscular contraction was produced. The emphasis was not on the ability to recruit maximally in a short period of time. Rather, slow concentric and eccentric contractions were emphasized. However, the testing of PTT relies on one's ability to maximally recruit their hamstring muscle in the shortest amount of time possible. Another possible rationale for the slowing of PTT demonstrated by the strength training group at 6 weeks is a decreased sensitivity of the muscle spindles secondary to the prescribed training.92 fri their 16 week study examining the effects of strength training, a significant decrease in reflex E M G amplitude/force amplitude ratio of the quadriceps was demonstrated. Hakkinen and Komi92 suggested this result may imply a change in the sensitivity of the muscle spindles. These investigators proposed that a decrease in sensitivity of the muscle spindles may result from the morphological changes 142 of the intrafusal fibers caused by strength training. A decrease in muscle spindle sensitivity of the hamstring muscle may ultimately affect kinesthesia (i.e. to detect stretching of the hamstring muscle) and the stiffness of the muscle, and thus, peak hamstring torque time. The trends observed at the end of the 12 weeks are difficult to explain. It is possible that the observed improvement in PTT (from the sixth week) demonstrated by the strength training group is secondary to an increase in FT/ST area ratio, as proposed by Hakkinen and K o m i 9 2 . As stated previously, Hakkinen and K o m i 9 2 observed improved PTT of the quadriceps muscle at 8 weeks while Wojtys et al.l 10 observed slowing of PTT at 6 weeks of isotonic training. It may be that an increase in FT/ST area ratio is time dependent and thus, an improvement of PTT for the strength training group was not observed until the end of the 12 weeks. An increase in FT/ST area ratio may also be dependent on the degree of overloading specified for isotonic training. Loads of 80% to 120% of one maximum repetition were used for the study conducted by Hakkinen and K o m i . 9 2 . A maximum weight resistance tolerated for 3 sets of 10 repetitions was used by Wojtys et al.l 10 a n ( j m j s s m d y . It is possible that subjects in the strength training group were unable to advance to loads similar to those used in the study by Hakkinen and K o m i 9 2 until the latter half of the training program. It should be noted that Hakkinen and K o m i 9 2 expressed the opinion that in training for longer durations (i.e. greater than 16 weeks), specific effects of heavy resistance strength training may be demonstrable by a worsening in isometric and dynamic force-time characteristics. Whether this worsening is due to decreases in the firing frequencies and/or changes in recruitment pattern of the motor units was not determined.92 It should be noted the lack of correlation between muscle reaction time and the strength has been documented in the literature.70> I 3 7 The observed slowing of PTT (from the sixth week) demonstrated by the proprioceptive training group may be secondary to a lack of appropriate progression of the training protocol. The proprioceptive training protocol was designed to ensure those subjects who were symptomatic (i.e. pain and swelling) in their A C L reconstructed knee could still participate. However, verbal feedback from the subjects in Group Two 143 indicated they felt the latter half of the proprioceptive training program to be not as challenging as the first half of the program. Secondary to the lack of challenge in the latter half of the proprioceptive training program, subjects may have lost gains made in the initial half of the program. Another possible factor responsible for the slowing of PTT (from the sixth week) demonstrated by the proprioceptive training group at the end of the 12 weeks may be the integration of speed and complex tasks into the proprioceptive training protocol. Although the integration of speed and complexity of tasks into proprioceptive training programs for A C L deficient and A C L reconstructed knees is often perceived to be essential and appropriate in the clinical setting, an emphasis on the precision of performing tasks may be more important for regaining normal neuromuscular control. 136 According to Kottkel36 ; there are limits to voluntary muscular control. A person can be aware of only one muscle action or position at a time. An individual training for coordination can shift his or her attention only two to three times per second. According to Kottkel365 this means that for proper training under conscious control, the activity must be limited in complexity and performed slower compared to the rate of performance of normal automatic activity. The ultimate goal of proprioceptive training is to achieve the coordination of appropriate muscle firing patterns during functional activities without the conscious awareness of the individual. Engrams, or automatic patterns, are only established by repetitions of precise patterns of activity hundreds of thousands . of times. 136, 138 Patterns inadequately practiced results in errors of performance due to the lack of inhibition of muscles not desired in the motor pattern. In addition, incorrectly performed patterns of activity during practice develop into engrams of those incorrectly performed patterns. Finally, there is a need for perpetual practice to maintain peak performance of any engram. Thus, the slowing of PTT observed in the proprioceptive training group at the end of the 12 weeks may not be due to the lack of progression in the latter half of the training program, but rather secondary to the inappropriate integration of speed and complexity of tasks. Since the latter half of the proprioceptive training protocol focused on these two components, the precision of performing motor tasks may have deteriorated, and a non-144 optimal training environment was created. Thus, although factors such as speed, force, and complexity of tasks need to be introduced into any coordination training program, these factors should only be integrated i f the precision of motor control is not compromised. Further evidence against early introduction of speed into proprioceptive training can be found in the neurophysiology literature. Collins et al.l 39 conducted a study examining our ability to detect muscle receptors' signals in different motor tasks and contexts. Although it has been demonstrated that muscle receptors play an important role in our conscious perception of movement, there is little information on our ability to detect their signals during different motor tasks. Collins et al.l 39 demonstrated there was a general attenuation of sensory feedback during movement, especially when the movement was rapid and large in amplitude. This raises questions regarding the role of muscle receptors in movement control, especially during large, rapid movements. Collins et al.l39 concluded that the extent to which the nervous system gates the different sensory modalities in the same way, or differentially, depending on the sensory demands of the task at hand, requires further investigators. Based on the findings of Collins et al.l 39, the early integration of speed into proprioceptive training protocols should be viewed with some caution. If the primary focus of A C L proprioceptive training protocols is to maximize the sensory information mediated by the joint and muscle receptors to dynamically stabilize the knee joint, performing tasks at slower speeds would appear to allow this goal to be achieved more effectively As mentioned previously, no published study examining the effectiveness of proprioceptive training in the A C L reconstructed knee has used isokinetic muscle torque as an outcome measure. Results of this study indicate that proprioception training may induce strength gains in the A C L reconstructed limb. A group by test occasion interaction effect was demonstrated on the following isokinetic variables: 1. Average concentric quadriceps torque of the A C L reconstructed limb. 2. Average eccentric quadriceps torque of the A C L reconstructed limb. 3. Average eccentric hamstring torque of the A C L reconstructed limb. 145 After 12 weeks of training, the proprioceptive group made greater improvements in these three isokinetic variables than the strength training group. The strength training group changed minimally after 12 weeks of training in these three isokinetic variables The lack of greater gains in isokinetic variables for the strength training group was unexpected. From the data sheets handed in by the subjects it was evident that heavier weights were being used as the 12 weeks of training progressed. However, these improvements were not reflected in the results of isokinetic strength testing at the end of the 12 weeks. Nevertheless, the results of this study is supported by Wojtys et al.l 10 Wojtys et al.l 10 a i s o found no significant increase in isokinetic strength in normal subjects who were placed on a 6 week isotonic training protocol. These investigators felt the lack of isokinetic change at the end of the 6 weeks in the isotonic group reflected the specificity of training and testing modes. Rasch and Morehouse 140 demonstrated strength gains from a 6 week training protocol in tests when muscles were employed in a familiar way, but little or no gain in strength was observed when unfamiliar test procedures were employed. The isokinetic strength gains observed in the proprioceptive training group were surprising. These results are in contrast to those of Wojtys et al.l 10 These investigators found a lack of isokinetic change in normal subjects who were placed on a 6 week agility, program, similar to that employed in this study. Possible explanations for the gains observed in Group Two are: 1. The main goal of the proprioceptive training program is to improve the neuromuscular function or the coordination of the affected lower extremity. The training of coordination is generally considered a volitional activity, during which, by trial and perception of results, an individual selects the muscular activity resulting in the desired performance. 138 Isokinetic testing, to a certain degree, relies on the coordination of the tested limb or joint. Thus, it is possible that the isokinetic changes observed in Group Two is secondary to improved coordination and better performance of the desired motor pattern compared to Group One. 146 2. The strength gains observed in Group Two can probably be attributed to improvements in neural activation rather than muscle hypertrophy. 141 it is quite possible that the proprioceptive training program contributed significantly to this neural activation involved in the early stages of strength gain. The role(s) neural mechanisms play in strength increase before muscle hypertrophy can be quite extensive. Yue and Cole 142 compared the maximal voluntary force production after a training program of repetitive maximal isomeric muscle contraction with force output after a training program that did not involve repetitive activation of muscle; that is, after mental training. Results indicated that strength increases can be achieved without repeated muscle activation (i.e. mental training). The investigators concluded that these force gains are secondary to the practice effects on central motor programming or planning. The results of this study contribute to the existing evidence for the neural origins of strength gains that occur before muscle hypertrophy. The following isokinetic variables were found to be significantly different after the 12 weeks of training for the A C L intact limb: 1. Average concentric torque of the quadriceps muscle. 2. Average eccentric torque of the hamstring muscle. Both the experimental groups made similar gains in the above two isokinetic variables after 12 weeks of training. These isokinetic gains can be attributed to the following: 1. The increased frequency of physical activity (i.e. prescribed training protocol three times a week) the subjects experienced during the 12 weeks. 2. The concept of cross education. 141 It has been suggested that the nature of cross education effect may entirely rest on neural factors presumably acting at various levels of the nervous system which could result in increasing the maximal level of muscle activation of the untrained limb. Many studies in the literature have examined the reliability and validity of functional testing. 123, 128, 129 Other studies have examined the correlational relationship between the ability to perform these functional tests and some factor(s).91, 147 143 However, we were unable to find studies examining the effects of training after A C L reconstruction on the performance of functional tests. The results of this study indicate that both strength and proprioceptive training have beneficial effects on functional ability (SLHD and one-legged timed hop) for both the A C L reconstructed and the A C L intact limbs. The following factors have been documented in the literature to have significant effects on functional ability: 1. Concentric and eccentric peak torque of the quadriceps muscle.91, 144 2. Concentric torque of the hamstring muscle. 143 3. Concentric and eccentric quadriceps work. 144 From the regressional analyses of this study, the following factors were found to have effects on functional ability (SLHD): 1. Average concentric quadriceps torque. 2. Average eccentric quadriceps torque. 3. Average concentric hamstring torque. 4. Average eccentric hamstring torque. Of the four isokinetic variables, average concentric hamstring torque had the most statistically significant correlational relationship with functional ability, as determined by SLHD. Thus, it is evident that isokinetic strength of the hamstring and quadriceps muscles have effects on SLHD performance. No dependent variables in this study were found to have an effect on the performance of the one-legged timed hop. These results are supported by the findings of the literature. Pincivero et al.l 43 reported that concentric quadriceps and hamstring strength have a significant contribution to the one-legged single hop for distance test. They also concluded that the hamstring muscles may play a more important role during the propulsive phase, thereby enabling normal subjects to jump further. Govett 9! and Wilk et a l . 1 4 5 also demonstrated a significant effect of the concentric peak torque of the quadriceps on the performance of the SLHD. Delitto et al.144 examined the relationship between isokinetic peak torque and work of the quadriceps to one-legged single hop for distance and vertical jump in subjects who had undergone A C L reconstruction. These investigators concluded that weaker 148 relationships between quadriceps peak torque and quadriceps work to SLHD and vertical exist in subjects following A C L reconstruction compared to active subjects without knee pathology. They felt that performance of the SLHD and vertical hop in A C L reconstructed subjects is related to factors other than isokinetic quadriceps peak torque and work. They concluded that other factors such as stability, proprioception and confidence in the knee may be more related to the performance of the two hops. However, no statistically significant correlations were found between functional hop tests and PTT in this study. The lack of relationship found between PTT and functional ability can be substantiated by Govett's^l results. Govett^l also found a lack of relationship between proprioception and functional ability (SLHD). It is difficult to assess the clinical relevance of the lack of relationship found between PTT and SLHD performance since PTT is only one component of the neuromuscular control involved in functional ability. In regards to the subjective assessment and the subjective analysis of physical function of the A C L reconstructed limb, no significant difference was found between the two experimental groups. Both groups demonstrated a significant improvement in both subjective scores. These results are in agreement with the findings of Beard et al.l09. Beard et al.l09 conducted a prospective, double-blinded, randomized, clinical trial to investigate the efficiency of two rehabilitation programs (strength training and proprioceptive training) for A C L deficient limbs. Both training groups demonstrated a significant increase in subjective score (Lysholm and Gillquist Knee Scoring Scale). Beard et al.l09 a j s o demonstrated that the change in subjective score was significantly greater in the proprioceptive group than the strength training group. The lack of significance found between the two experimental groups may be secondary to the different population involved in this study. In conclusion, peak hamstring torque time of the A C L reconstructed limb was influenced by both strength training and proprioceptive training. It is proposed that the two types of training influence PTT through different neuromuscular mechanisms. A group by test occasion interaction effect was found on isokinetic strength measures. The proprioceptive group demonstrated greater gains in the average concentric and eccentric 149 quadriceps torques and the average eccentric hamstring torque of the A C L reconstructed limb after 12 weeks of training, while minimal change was observed in the strength training group after 12 weeks. Both groups demonstrated improved functional ability in the lower extremities after 12 weeks of training. Both experimental groups also demonstrated significant gains in the subjective assessment and the subjective analysis of physical function of the A C L reconstructed limb. 150 Limitations Several limitations should be considered for this study. Threats to Statistical Conclusion Validity: 1. Sample Size: The sample size used in this study is small. Smaller samples are less likely to be a good representation of population characteristics, and, therefore, true differences between groups are less likely to be recognized. A power analysis0 was performed prior to the start of the study to estimate the sample size required. The power analysis indicated that a sample size of 10 subjects would provide sufficient power levels to detect a difference between groups and an interaction effect, but not for a difference between test occasions. Thus, a sample size of 10 subjects was used for this study. A second power analysis was performed after the completion of the study (Appendix A). Based on the results of the second power analysis, a sample size of 10 subjects provided an extremely high power level for the interaction effect, but very low power levels for the main effects (i.e. experimental group and test occasion). Thus, it is highly probable that the insignificant findings found between groups and test occasions are due to the small sample size used. However, the results of this study did demonstrate the interaction effect between group and test occasion. Statistically, an interaction effect (when present) must be interpreted before conclusions can be made for the main effects. Therefore, although this study failed to demonstrated significant main effects, it still provided important information regarding the effects of proprioceptive and strength training on PTT in the A C L reconstructed limb by demonstrating the interaction effect. Threats to Internal Validity: 1. Lack of a Control Group: Due the ethical reasons, a control group was not established in this study. Without a control group, however, it becomes difficult to be confident that the changes observed in the dependent variables are secondary to the prescribed intervention or are simply due to the passage of time (from surgery). c This power analysis was performed based on estimated values for the peak hamstring torque time variable. Of all the dependent variables in this study, we were most interested in the PTT variable. 151 2. Attrition: Significant attrition introduces bias by changing the composition of the sample. Attrition was controlled in this study by maintaining an encouraging environment. No subject was lost to attrition in this study. 3. Compliance: Compliance is always an issue when specific instructions must be carried out by subjects involved in a study. Compliance was controlled by the following means: a. The exercise sessions were supervised whenever possible. b. When a supervised session was not possible, subjects were asked to record the details of the session. This record was later collected by the investigator(s). 3. Selection: The subjects involved in this study may be highly motivated individuals who do not represent the average person with an A C L reconstructed knee. Subjects involved in this study were asked to commit approximately 3.5 to 4 hours a week to their exercise protocol. The average individual may not be motivated to commit this amount of time per week to rehabilitate their A C L reconstructed knee. Threats to External Validity 1. Surgical Technique: There are many surgical techniques used to reconstruct the torn A C L . In addition, the same surgical technique performed by different surgeons may have slightly different outcomes or effects. Thus, the conclusions of this paper are applicable to the surgical technique used in this study. 2. Measurement of PTT: Peak hamstring torque time of the A C L reconstructed limb was assessed in the seated position. Measuring peak hamstring torque time in the seated position may limit the ability to generalize the results to more dynamic conditions, such as walking and running. It is unclear at the present time whether PTT measured in the seated position correlate well with those measured in the standing position. Kiefer et al.l 46 reported that reproduction of passive joint position scores in sitting should not be compared to those performed in standing. In addition, the use of electromyograph while measuring PTT would have provided a more detailed description of the neuromuscular function of the A C L reconstructed limb. 152 Chapter Nine SUMMARY AND CONCLUSION Summary The anterior cruciate ligament is the most commonly injured ligament in sports. Despite extensive research to date, the management of A C L injuries is still not to the point where the function of the affected knee can be completely restored. The A C L plays a major role in maintaining the normal function of the knee joint. When the A C L is disrupted, the knee joint loses both an important mechanical restraint and a source of sensory input for joint proprioception and kinesthesia. Thus, decreased stability of the knee associated with functional disability is a common feature after A C L injuries. Despite the extensive research done on the sensory function of the A C L , it is unclear as to what degree the biomechanics of the knee joint and vulnerability of its ligaments are dependent on its innervation. In addition, although A C L reconstructions are now commonly performed in young athletes, the effects of surgical reconstruction on the proprioceptive function of the affected knee remain unclear. Current research findings on the effects of A C L reconstruction on knee proprioception provide no consensus. In addition, the clinical significance of statistical differences of proprioception as determined in research studies is undetermined. Currently, proprioceptive training is thought to be an important component in the rehabilitation of A C L injuries. However, the effectiveness of proprioceptive training is not elucidated in the literature. The main purpose of this study was to examine the effectiveness of proprioceptive training in the A C L reconstructed knee. Rather than choosing the more commonly used measures of proprioception and kinesthetic awareness (i.e. detection of passive joint motion, reproduction of passive joint position), a more dynamic measure of knee neuromuscular function was chosen. Peak torque time is a measure of volitional activity of muscle(s) and we propose it relies on the following factors: 1. The sensitivity of the muscle spindles. 2. Kinesthetic awareness of the knee joint. 3. The ability to recruit as many motor units as possible in the shortest amount of time. 153 4. The coordination of the limb to the specified task. It is the postulation of this study that there is sensory information originating from the A C L graft and joint afferents that affects the primary muscle spindle afferents via the gamma efferent muscle spindle system. Thus, the A C L graft may contribute to the continuous preparatory adjustment of muscular stiffness through the gamma efferent muscle spindle system. PTT can be regarded as the ability of the hamstring muscle to increase its stiffness at the time of a displacement (i.e. the forward movement of the dynamometer lever arm). As indicated in the review of the literature, functional joint stability is greatly determined by joint load caused by the contraction (or stiffness) of its surrounding muscles. Hence, PTT can be regarded as a dynamic measure of the neuromuscular function of the knee joint. A second purpose of the study was to determine the relative contribution of isokinetic strength and peak hamstring torque time to functional ability. Functional ability was determined by two functional hop tests: one-legged single hop for distance and one-legged timed hop. Two experimental groups of five subjects were compared. Group One (Strength Training Group) was placed on a 12 week strength training protocol for the lower extremities. Group Two (Proprioception Training Group) was placed on a 12 week proprioceptive protocol for the lower extremities. At the end of the 12 weeks, there was no significant difference in PTT between or within (between test occasions) the two experimental groups. A curvilinear relationship was observed between PTT of the A C L reconstructed limb and test occasion for both experimental groups. There was a significant group by test occasion interaction effect on PTT of the A C L reconstructed limb. The strength training group demonstrated a slowing of PTT while the proprioceptive group demonstrated an improvement in PTT at the end of 6 weeks. At the end of the 12 weeks, both experimental groups regressed toward their baseline PTT values, such as the strength training group demonstrated an improvement in PTT while the proprioceptive group demonstrated a slowing of PTT in the latter 6 weeks. It is proposed that the two types of training influence PTT through different neuromuscular mechanisms. 154 There was a significant group by test occasion interaction effect on isokinetic strength measures of the A C L reconstructed limb. The proprioceptive training group demonstrated greater gains in average concentric and eccentric quadriceps torques and average eccentric hamstring torque of the A C L reconstructed limb than the strength training group after the 12 weeks. The greater gains observed in the proprioceptive training group at the end of the 12 weeks are proposed to be secondary to improved coordination of the A C L reconstructed limb, improved neural activation of the A C L reconstructed limb, and better performance of the desired motor pattern on the KIN-COM rather than actual muscle hypertrophy. Both experimental groups demonstrated similar gains in the subjective assessment and the subjective analysis of physical function of the A C L reconstructed limb. Both experimental groups also demonstrated similar gains in the functional ability, as determined by the two functional hop tests, of the A C L reconstructed limb after 12 weeks of training. Regressional analyses found average concentric and eccentric torques of the quadriceps and hamstring muscle to have significant effects on the performance of the one-legged single hop for distance. No significant effects were found for any of the dependent variables on the performance of the one-legged timed hop. However, due to the small sample size used in this study, the ability to generalize the results of this study may be limited. 155 Conclusion Based on the findings of this study, the following conclusions (as pertaining to the A C L reconstructed knee) can be made based on the hypotheses set for this study: 1. Both training protocols had beneficial effects on the subjective assessment and the subjective analysis of physical function of the A C L reconstructed knee. 2. The proprioceptive training group made greater isokinetic gains after 12 weeks of training compared to the strength training group. However, these isokinetic gains observed in the proprioceptive training group are probably secondary to improved coordination, improved neural activation, and better performance of the desired motor pattern on the KIN-COM rather than actual muscle hypertrophy. 3. Both training protocols had beneficial effects on the performance of the two functional hop tests. 4. There was no significant difference between or within (between test occasion) the two experimental groups in PTT. There was, however, a significant group by test occasion interaction effect on PTT. A curvilinear relationship between PTT and test occasion was observed for both groups. Further investigations are necessary to clarify the trends observed. It is proposed that the two training protocols influence PTT through different neuromuscular mechanisms. Strength training is proposed to positively influence PTT by increasing FT/ST muscle area ratio and to negatively influence PTT by decreasing muscle spindle sensitivity. Proprioceptive training is proposed to positively influence PTT by improving coordination and improving neural activation and inhibition of the appropriate muscles. However, the effectiveness of proprioceptive training is dependent on appropriate progression and repetition. Early integration of speed, force, and complexity of tasks into proprioceptive training may decrease the effectiveness of training. Integration of these components should only occur if the precision is not compromised. 5. Average concentric and eccentric torques of the quadriceps and hamstring muscles had significant effects on the performance of the one-legged single hop for distance test. Average concentric hamstring torque had the most statistically significant correlational relationship with functional ability (SLHD). 156 Chapter Ten CLINICAL RELEVANCE AND RECOMMENDATIONS Clinical Relevance The purpose of conducting clinical studies is to gain insight into clinical practices. Currently, proprioceptive training is emphasized in the literature as an important component of rehabilitation after A C L injuries. However, the effectiveness of proprioceptive training in the A C L reconstructed limb is not well documented. In addition, there is little discussion in the literature regarding the appropriate progression of proprioceptive training after A C L injuries. The results of this study indicate that both strength training and proprioceptive training are beneficial for the A C L reconstructed limb. Proprioceptive training is more effective than strength training in restoring the coordination of the A C L reconstructed limb. However, to ensure the effectiveness of proprioceptive training, appropriate progression and repetition are essential. Early integration of speed, force, and complexity of tasks into proprioceptive training protocols may decrease the effectiveness of training. Integration of these components should only occur i f the precision of motor control during tasks is not compromised. 157 Recommendations The following recommendations are proposed for future studies: 1. This study only involved subjects with A C L reconstructed limbs. To determine the effects of A C L reconstruction on proprioception, it is recommended that future prospective studies involve subjects with A C L deficient limbs also. 2. This study measured dynamic neuromuscular function in the seated position. Measuring dynamic neuromuscular function in the seated position may limit the ability to generalize the results to more dynamic conditions, such as walking and running. Therefore, it is recommended that future studies measure dynamic neuromuscular function in the standing position. In addition, the use of electromyograph would provide a more detailed description of neuromuscular function. 3. This prospective study used a small sample size. A small sample size is less likely to be good a representation of population characteristics. In addition, the power level for the main effects in the study were very low. Therefore, it is the recommendation that future studies use a larger sample size. 158 R E F E R E N C E S 1. Small C, Waters JT J, Voight M . Comparison of two methods of measuring hamstring reaction time using the Kin-Corn isokinetic dynamometer. Journal of Orthopaedic & Sports Physical Therapy 1994; 19:335-40. 2. Shelbourne K D , Klootwyk TE, Carlo Mark SD. Ligamentous injuries. In: Griffin L Y , ed. Rehabilitation of the Injured Knee. St. Louis: Mosby-Year Book, Inc., 1995:149-164. 3. Jackson DW. Anatomy of the Anterior Cruciate Ligament. In: Jackson DW, Arnoczky SP, Frank CB, Woo SL, Simon T M , eds. The Anterior Cruciate Ligament: Current and Future Concepts. New York: Raven Press, 1993:5-21. 4. Kennedy JC, Alexander IJ, Hayes K C . Nerve supply of the human knee and its functional importance. American Journal of Sports Medicine 1982; 10:329-35. 5. Schutte MJ , Dabezies EJ, Zimny M L , Happel LT. Neural anatomy of the human anterior cruciate ligament. The Journal of Bone and Surgery 1987; 69-A:243-7. 6. Irrgang JJ. Modern trends in anterior cruciate ligament rehabilitation: Nonoperative and postoperative management. Clinics in Sports Medicine 1993; 12:797-813. 7. Ellison A E , Berg EE. Embryology, anatomy, and function of the anterior cruciate ligament. Orthopedic Clinics of North America 1985; 16:3-14. 8. Warwick R, Williams P. Gray's Anatomy. Norwich: Jarrold and Sons Ltd., 1973. 9. Arnoczky SP. Anatomy of the anterior cruciate ligament. Clinical Orthopaedics and Related Research 1983:19-25. 10. Dye SF, Cannon WD. Anatomy and biomechanics of the anterior cruciate ligament. Clinics in Sports Medicine 1988; 7:715-25. 11. Smith BA, Livesay GA. Biology and biomechanics of the anterior cruciate ligament. Clinics in Sports Medicine 1993; 12:637-70. 12. Amis A A , Dawkins GPC. Functional anatomy of the anterior cruciate ligament. The Journal of Bone and Joint Surgery 1991; 73-B:260-7. 13. Clark JM, Sidles JA. The interrelation of fiber bundles in the anterior cruciate ligament. Journal of Orthopaedic Research 1988; 8:180-188. 159 14. Arnoczky SP. Blood supply to the anterior cruciate ligament and supporting structures. Orthopaedics Clinics of North America 1985; 16:15-28. 15. Zimny M , Schutte M , Dabezies E. Mechanoreceptors in the human anterior cruciate ligament. Anat Rec 1986; 214:204-9. 16. Schultz RA, Miller DC, Kerr CS, Micheli L. Mechanoreceptors in human cruciate ligaments. The Journal of Bone and Joint Surgery 1984; 66A: 1072-1076. 17. Barrack RL, Skinner HB. The sensory function of knee ligaments. In: Daniel D, ed. Knee Ligaments: Structure, Function, Injury, and Repair. New York: Raven Press Ltd., 1990.. 18. Johansson H, Sjolander P, Sojka P. A sensory role for the cruciate ligaments. Clinical Orthopaedics and Related Research 1991; 268:161-78. 19. Sjolander P, Johansson H, Sojka P, Rehnholm A. Sensory nerve endings in the cat cruciate ligaments: a morphological investigation. Neuroscience Letters 1989; 102:33-38. 20. O'Connor BL, McConnaughey JS. The structure and innervation of cat knee menisci, and their relation to a "sensory hypothesis" of meniscal function. American Journal of Anatomy 1978; 153:431 -442. 21. Johansson H. Role of knee ligaments in proprioception and regulation of knee stiffness. Journal of Electromyography and Kinesiology 1991; 1:158-179. 22. Barrack RL, Lund PL, Skinner HB. Knee joint proprioception revisited. Journal of Sport Rehabilitation 1994; 3:18-42. 23. Johansson H, Sjolander P, Sojka P. Receptors in the knee joint ligaments and their role in the biomechanics of the joint. Biomedical Engineering 1991; 18:341-368. 24. Norkin C, Levangie P. The Knee Complex. Joint Structure and Function: A Comprehensive Analysis. Philadelphia: F. A . Davis Company, 1983:291-330. 25. Butler DL, Noyes FD, Grood ES. Ligamentous restraints to anterior-posterior drawer in the human knee. Journal of Bone and Joint Surgery 1980; 62-A:259-270. 26. Arms SW, Pope M H , Johnson RJ, Fischer RA, Arvidsson I, Eriksson E. The biomechanics of anterior cruciate ligament rehabilitation and reconstruction. American Journal of Sports Medicine 1984; 12:8-18. 160 27. Lutz GE, Stuart MJ , Sim FH, Scott SG. Rehabilitative techniques for athletes after reconstruction of the anterior cruciate ligament [published erratum appears in Mayo Clin Proc 1991 Jan;66(l):l 14]. Mayo Clinic Proceedings 1990; 65:1322-9. 28. Galway HR, Macintosh DL. The lateral pivot shift: A symptom and sign of anterior cruciate ligament insufficiency. Clinical Orthopaedics and Related Research 1980:45-50. 29. Beard DJ, Kyberd PJ, Fergusson C M , Dodd CA. Proprioception after rupture of the anterior cruciate ligament. An objective indication of the need for surgery? Journal of Bone & Joint Surgery British 1993; 75:311-5. 30. Torzilli PA, Deng X , Warren RF. The effect of joint-compressive load and quadriceps muscle force on knee motion in the intact and anterior cruciate ligament-sectioned knee. American Journal of Sports Medicine 1994; 22:105-12. 31. Noyes FR, Butler DL, Grood ES. Biomechanical analysis of human ligament grafts used in knee ligament repairs and reconstructions. Journal of Bone and Joint Surgery 1983; 66A:344-52. 32. Kuo C Y . Field measurements in snow skiing injury research. Journal of Biomechanics 1983; 16:609-624. 33. Denti M , Monteleone M , Berardi A , Panni AS. Anterior cruciate ligament mechanoreceptors. Histologic studies on lesions and reconstruction. Clinical Orthopaedics & Related Research 1994; 308:29-32. 34. Barrett DS. Proprioception and function after anterior cruciate reconstruction. Journal of Bone & Joint Surgery British 1991; 73:833-7. 35. Markolf K L , Bargar WL, Shoemaker SC, Amstutz HC. The role of joint load in knee stability. The Journal of Bone and Joint Surgery 1981; 63A:570-585. 36. Pope DF, Cole KJ , Brand RA. Physiologic loading of the anterior cruciate ligament does not activate quadriceps or hamstrings in the anesthetized cat. The American Journal of Sports Medicine 1990; 18:595-9. 37. Solomonow M , Baratta R, Zhou BH, et al. The synergistic action of the anterior cruciate ligament and thigh muscles in maintaining joint stability. American Journal of Sports Medicine 1987; 15:207-13. 38. Renstrom P, Arms SW, Stanwyck TS, Johnson RK, Pope M H . Strain within the anterior cruciate ligament during hamstring and quadriceps activity. American Journal of Sports Medicine 1986; 14:83-7. 161 39. Skinner HB, Wyatt MP, Hodgdon HA, Conard DW, Barrack RL. Effect of fatigue on joint position sense of the knee. Journal of Orthopaedic Research 1986; 4:112-8. 40. Barrack RL, Skinner HB, Buckley SL. Proprioception in the anterior cruciate deficient knee. American Journal of Sports Medicine 1989; 17:1-6. 41. Yasuda K, Sasaki T. Muscle exercise after anterior cruciate ligament reconstruction: Biomechanics of the Simultaneous isometric contraction method of the quadriceps and the hamstrings. Clinical Orthopaedics and Related Research 1987; 220:266-74. 42. Yasuda K, Sasaki T. Exercise after anterior cruciate ligament reconstruction: The force exerted on the tibia by the separate isometric contractions of the quadriceps or the hamstrings. Clinical Orthopaedics and Related Research 1987:275-83. 43. Durselen L, Claes L, Kiefer H. The influence of muscle forces and external loads on cruciate ligament strain. American Journal of Sports Medicine 1995; 23:129-36. 44. Beynnon BD, Fleming BC, Johnson RJ, Nichols CE, Renstrom PA, Pope M H . Anterior cruciate ligament strain behavior during rehabilitation exercises in vivo. American Journal of Sports Medicine 1995; 23:24-34. 45. O'Connor JJ. Can muscle co-contraction protect knee ligaments after injury or repair. The Journal of Bone and Joint Surgery 1993; 75-B:41-8. 46. Borsa PA. The effects of joint position and direction of joint motion on proprioceptive sensibility in anterior cruciate ligament-deficient athletes. The American Journal of Sports Medicine 1997; 25:336-340. 47. Lephart SM, Pincivero D M , Giraldo JL, Fu FH. The role of proprioception in the management and rehabilitation of athletic injuries. The American Journal of Sports Medicine 1997; 25:130-7. 48. Irrgang JJ, Whitney SL, Cox ED. Balance and proprioceptive training for rehabilitation of the lower extremity. Journal of Sport Rehabilitation 1994; 3:68-83. 49. Johansson H, Sjolander P, Sojka P. Activity in receptor afferents from the anterior cruciate ligament evokes reflex effects on fusimotor neurones. Neuroscience Research 1990; 8:54-59. 50. Krauspe R, Schmidt M , Schaible HG. Sensory innervation of the anterior cruciate ligament. The Journal of Bone and Joint Surgery 1992; 74-A:390-7. 162 51. Pitman MI, Nainzadeh N , Menche D, Gasalberti R, Song EK. The intraoperative evaluation of the neurosensory function of the anterior cruciate ligament in humans using somatosensory evoked potentials. Arthroscopy 1992; 8:442-7. 52. Madey SM, Wolff A J , Brand RA, Cole KJ . Characterization of A C L neural endings with wheat germ agglutination-horse radish peroxidase, Orthopaedic Research Society, Transactions of the 39th Annual Meeting, San Francisco, 1993. Vol . 18. 53. Sherrington CS. On the proprio-ceptive system, especially in its reflex aspects. Brain 1906;29:467-482. 54. Bastian HC. The "muscular sense"; its nature and localization. Brain 1888; 10:1-136. 55. Gandevia SC. Kinesthesia: roles for afferent signals and motor commands. In: Rowell L, Shepherd JP, eds. Handbook on integration of motor, circulatory, respiratory, and metabolic control during exercise. New York: American Physiological Society, Oxford University Press:128-172. 56. Barrack RL, Skinner HB, Brunet M E , Cook SD. Joint laxity and proprioception in the knee. The Physician and Sportsmedicine 1983; 11:130-5. 57. Grigg P. Peripheral neural mechanisms in proprioception. Journal of Sport Rehabilitation 1994; 3:2-17. 58. Edin BB, Abbs JH. Finger movement responses of cutaneous mechanoreceptors in the dorsal skin of the human hand. Journal of Neurophysiology 1991; 386:63-71. 59. Cohen D A D , Prud'Homme MJL, JF K. tactile activity in primate primary somatosensory cortex during active arm movements: correlation with receptive field properties. Journal of Neurophysiology 1994; 71:161-172. 60. Clark FJ, Horch K W , Bach SM, Larson GF. Contributions of cutaneous and joint receptors to static knee-position sense in man. Journal of Neurophysiology 1979; 42:877-888. 61. Edin B H , Johansson N . Skin strain patterns provide kinaesthetic information to the human central nervous system. Journal of Physiology 1995; 487:243-251. 62. Goodwin G M , McCloskey DI, Matthews PB. The contribution of muscle afferents to kinaesthesia shown by vibration induced illusions of movement and by effects of paralysing joint afferents. Brain 1972; 95:705-748. 163 63. Goodwin G M , McCloskey DI, Matthews PB. The persistence of appreciable kinesthesia after paralysing joint afferents but preserving muscle afferents. Brain Research 1972;37:326-9. 64. Grigg P, Finerman GA, Riley L H . Joint position sense after total hip replacement. The Journal of Bone and Joint Surgery 1973; 5 5 A: 1016-1025. 65. Morberg E. The role of cutaneous afferents in position sense, kinesthesia, and motor function of the hand. Brain 1983; 106:1-19. 66. Ferrell WR. The adequacy of stretch receptors in the cat knee joint for signalling joint angle throughout a full range of movement. Journal of Physiology 1980; 299:85-99. 67. Ferrell WR, Gandevia SC, McCloskey DI. The role of joint receptors in human kinaesthesia when intramuscular receptors cannot contribute. Journal of Physiology 1986;386:63-71. 68. Beard DJ, Kyberd PJ, O'Connor JJ, Fergusson C M , Dodd CAF. Reflex hamstring contraction latency in anterior cruciate ligament deficiency. Journal of Orthopaedic Research 1994; 12:219-28. 69. Corrigan JP, Cashman WF, Brady MP. Proprioception in the cruciate deficient knee. Journal of Bone & Joint Surgery British 1992; 74:247-50. 70. Ihara H, Nakayama A. Dynamic joint control training for knee ligament injuries. American Journal of Sports Medicine 1986; 14:309-15. 71. Shiraishi M , Mizuta H, Kubota K , Otsuka Y , Nagamoto N , Takagi K. Stabilometric assessment in the anterior cruciate ligament-reconstructed knee. Clinical Journal of Sport Medicine 1996; 6:32-9. 72. Zatterstrom R, Friden T, Lindstrand A , Moritz U . The effect of physiotherapy on standing balance in chronic anterior cruciate ligament insufficiency. American Journal of Sports Medicine 1994; 22:531-6. 73. Pope M H , Johnson RJ, Brown DW, Tighe C. The role of the musculature in injuries to the medial collateral ligament. The Journal of Bone and Joint Surgery 1979;61A:398-402. 74. Yasuda K, Erickson AR, Beynnon BD, Johnson RJ, Pope M H . Dynamic elongation behaviour in the medial collateral and anterior cruciate ligaments during lateral impact loading. Journal of Orthopaedic Research 1993; 11:190-198. 164 75. Petersen I, Stener B. Experimental evaluation of the hypothesis of ligamento-muscular protective reflexes III. A study in man using the medial collateral ligament of the knee joint. Acta Physiol. Scand. 1959; 48 (Suppl. 166):51-61. 76. Sojka P, Sjolander P, Johansson H, Djupsjobacka M . Influence from stretch-sensitive receptors in the collateral ligaments of the knee joint on the gamma muscle spindle systems of flexor and extensor muscles. Neuroscience Research 1991;11:55-62. 77. Sojka P, Johansson H, Sjolander P, Lorentzon R, Djupsjobacka M . Fusimotor neurones can be reflexly influenced by activity in receptor afferents from the posterior cruciate ligament. Brain Research 1989; 483:177-183. 78. Solomonow M , Baratta R, D'Ambrosia R. The role of the hamstrings in the rehabilitation of the anterior cruciate ligament-deficient knee in athletes. [Review]. Sports Medicine 1989; 7:42-8. 79. Barrett DS, Cobb A G , Bentley G. Joint proprioception in normal, osteoarthritic and replaced knees. The Journal of Bone and Joint Surgery 1991; 73-B:53-6. 80. O'Connor BL, Visco D M , Brandt K D , Myers SL, Kalasinski L A . Neurogenic acceleration of osteoarthrosis. The Journal of Bone and Joint Surgery 1992; 74A:367-376. 81. Baratta R, Solomonow M , Zhou B H , Letson D, Chuinard R, D'Ambrosia R. Muscular coactivation. The role of the antagonist musculature in maintaining knee stability. American Journal of Sports Medicine 1988; 16:113-22. 82. Draganich LF, Jaeger RJ, Kralj AR. Coactivation of the hamstrings and quadriceps during extension of the knee. Journal of Bone and Joint Surgery 1989; 71-A:1075-81. 83. Hagood S, Solomonow M , Baratta R, Zhou Bh, D'Ambrosia R. The effect of joint velocity on the contribution of the antagonist musculature to knee stiffness and laxity. American Journal of Sports Medicine 1990; 18:182-7. 84. Grabiner M D , Koh TJ, Miller GF. Further evidence against a direct automatic neuromotor link between the A C L and hamstrings. Medicine and Science in Sports and Exercise 1992:1075-1079. 85. Palmar I. Plastic surgery of the ligaments of the knee. Acta Chir Scand 1944; 91:37-48. 86. Freeman M A , Wyke B. Articular reflexes at the ankle joint. An electromyographic study of normal and abnormal influences of ankle joint 165 mechanoreceptors upon reflex activity in the leg muscles. British Journal of Surgery 1967; 54:990-1001. 87. Safran MR, Caldwell GL, Fu FH. Proprioception considerations in surgery. Journal of Sport Rehabilitation 1994; 3:103-115. 88. Tsujimoto K, Andrish JT, Kambic HE, Grabiner M , Wink C. An investigation of the neurohistology and biomechanics of A C L reconstruction in goats: A comparison of primary repair and augmentation versus primary reconstruction alone, Orthopaedic Research Society, 39th Annual Meeting, San Francisco, California, 1993. 89. Barrack RL, Skinner HB, Brunet M E , Cook SD. Joint kinesthesia in the highly trained knee. Journal of Sports Medicine 1984; 24:18-20. 90. MacDonald PB, Hedden D, Pacin O, Sutherland K. Proprioception in anterior cruciate ligament-deficient and reconstructed knees. American Journal of Sports Medicine 1996; 24:774-8. 91. Govett JR. The relative importance of proprioception, ligament laxity and strength on functional performance in the A C L deficient and A C L reconstructed knee. School of Human Kinetics. Vancouver: University of British Columbia, 1995. 92. Hakkinen K, Komi PV. Changes in neuromuscular performance in voluntary and reflex contraction during strength training in man. International Journal of Sports Medicine 1983;4:282-288. 93. Co FH, Skinner HB, Cannon WD. Effect of reconstruction of the anterior cruciate ligament on proprioception of the knee and the heel strike transient. Journal of Orthopaedic Research 1993; 11:696-704. 94. Ciccotti M G , Kerlan RK, Perry J, Pink M . A n electromyographic analysis of the knee during functional activities ~ Part II. The anterior cruciate ligament-deficient and -reconstructed profiles. The American Journal of Sports Medicine 1994; 22:651-8. 95. Tibone JE, Antich TJ, Fanton GS, Moynes DR, Perry J. Functional analysis of anterior cruciate ligament instability. The American Journal of Sports Medicine 1986; 14:276-284. 96. Wojtys E M , Huston LJ. Neuromuscular performance in normal and anterior cruciate ligament-deficient lower extremities. [Review] [84 refs]. American Journal of Sports Medicine 1994; 22:89-104. 166 97. Andriacchi TP, Birac D. Functional testing in the anterior cruciate ligament-deficient knee. Clinical Orthopaedics & Related Research 1993; 288:40-7. 98. Kalund S, Sinkjaer T, Arendt-Nielsen L, Simonsen O. Altered timing of hamstring muscle action in anterior cruciate ligament deficient patients. American Journal of Sports Medicine 1990; 18:245-48. 99. Branch TP, Hunter R, Donath M . Dynamic E M G analysis of anterior cruciate deficient legs with and without bracing during cutting. The American Journal of Sports Medicine 1989; 17:35-41. 100. Tibone JE, Antich TJ. Electromyographic analysis of the anterior cruciate ligament-deficient knee. [Review] [15 refs]. Clinical Orthopaedics & Related Research 1993; 288:35-9. 101. Shelbourne K D , Nitz P. Accelerated rehabilitation after anterior cruciate ligament reconstruction. American Journal of Sports Medicine 1990; 18:292-9. 102. Johnson RJ, Beynnon BD, Nichols CE, Renstrom P. Current concepts review: The treatment of injuries of the anterior cruciate ligament. The Journal of Bone and Joint Surgery 1992; 74-A: 140-51.' 103. Bynum BE, Barrack RL, Alexander A H . Open versus closed kinetic exercises after anterior cruciate ligament reconstruction. American Journal of Sports Medicine 1995;23:401-6. 104. MacDonald PB, Hedden D, Pacin O, Huebert D. Effects of an accelerated rehabilitation program after anterior cruciate ligament reconstruction with combined semitendinosus-gracilis autograft and a ligament augmentation device. American Journal of Sports Medicine 1995; 23:588-92. 105. Ohkoshi Y , Yasuda K, Kaneda K, Wada T, Yamanaka M . Biomechanical analysis of rehabilitation in the standing position. American Journal of Sports Medicine 1991;19:605-11. 106. Yack HJ, Riley L M , Whieldon TR. Anterior tibial translation during progressive loading of the ACL-deficient knee during weight-bearing and nonweight-bearing isometric exercise. Journal of Orthopaedic & Sports Physical Therapy 1994; 20:247-53. 107. De Carlo MS, Sell K E , Shelbourne K D , TE K . Current concepts on accelerated A C L rehabilitation. Journal of Sport Rehabilitation 1994; 3:304-318. 108. Noyes FR, Matthews DS, Mooar PA, Grood ES. The symptomatic anterior cruciate-deficient knee. Part II: The results of rehabilitation, activity 167 modification, and counselling on functional disability. The Journal of Bone and Joint Surgery 1983; 65-A: 163-74. 109. Beard DJ, Dodd CA, Trundle HR, Simpson A H . Proprioception enhancement for anterior cruciate ligament deficiency. A prospective randomised trial of two physiotherapy regimes. Journal of Bone & Joint Surgery British 1994; 76:654-9. 110. Wojtys E M , Huston LJ, Taylor PD, Bastian SD. Neuromuscular adaptations in isokinetic, isotonic, and agility training programs. The American Journal of Sports Medicine 1996; 24:187-192. 111. Walla DJ, Albright JO, McAuley E, Martin RK, Eldridge V, El-Khoury G. Hamstring control and the unstable anterior cruciate ligament-deficient knee. American Journal of Sports Medicine 1985; 13:34-9. 112. Giove TP, Miller SJ, Kent BE, Sanford TL, Garrick JG. Non-operative treatment of the torn anterior cruciate ligament. The Journal of Bone and Joint Surgery 1983; 65-A: 184-92. 113. Beard DJ, Fergusson C M . The conservative management of anterior cruciate ligament deficiency: A nationwide survey of current practice. Physiotherapy 1992;78:181-186. 114. Farrell M , Richards JG. Analysis of the reliability and validity of the kinetic communicator exercise device. Medicine and science in sports and exercise 1986; 16:44-49. 115. Tis L L , Perrin DH. Relationship between isokinetic average force, average torque, peak force and peak torque of the knee extension and flexor musculature. Isokinetics and Exercise Science 1994; 4:150-152. 116. Hoens A M , Strauss GR. The effect of deleting nonisokinetic phases of movement from isokinetic strength evaluations. Isokinetic and Exercise Science 1994; 4:96-103. 117. Jensen RC, Warren B, Laursen C, Morrisey M C . Static pre-load effect on knee extensor isokinetic concentric and eccentric performance. Medicine and Science in Sports and Exercise 1991; 23:10-14. 118. Kramer JF, Hi l l K , Jones IC, Sandrin M , Vyse M . Effect of dynamometer application arm length on concentric and eccentric torques during isokinetic knee extension. Physiotherapy Canada 1989; 41:100-106. 168 119. Wessel J, Baergen J, Baron D. Effect of minimum force and intercontraction pause on measurement of isokinetic torque of the knee extensors, CPA/APTA Congress Poster Presentation, Toronto, Ontario, June 1994. 120. Kellis E, Baltzopoulos V . Isokinetic eccentric exercise. Sports Medicine 1995; 19:202-222. 121. Kramer JF, Nusca D, Bisbee L, MacDermid J, Kempt D, Boley S. Forearm pronation and supination: Reliability of absolute torques and nondominant/dominant ratios. Journal of Hand Therapy 1994; 7:15-40. 122. Harding B, Black T, Bruulsema A , Maxwell B, Stratford P. Reliability of a reciprocal test protocol performed on the Kinetic Communicator: An isokinetic test of knee extensor and flexor strength. The Journal of Orthopaedic and Sports Physical Therapy 1988; 10:S8-S20. 123. Tegner Y, Lysholm J, Lysholm M , Gillquist J. A performance test to monitor rehabilitation and evaluate anterior cruciate ligament injuries. American Journal of Sports Medicine 1986; 14:156-9. 124. Risberg M A , Ekeland A. Assessment of functional tests after anterior cruciate ligament surgery. Journal of Orthopaedic & Sports Physical Therapy 1994; 19:212-7. 125. Fonseca ST, Magee D, Wessel J, Reid D. Validation of a performance test for outcome evaluation of knee function. Clinical Journal of Sport Medicine 1992; 2:251-256. 126. Barber SD, Noyes FR, Mangine RE, McCloskey JW, Hartman W. Quantitative assessment of functional limitations in normal and anterior cruciate ligament-deficient knees. Clinical Orthopaedics and Related Research 1990; 255:204-14. 127. Noyes FR, Barber SD, Mangine RE. Abnormal lower limb symmetry determined by function hop tests after anterior cruciate ligament rupture. American Journal of Sports Medicine 1991; 19:513-8. 128. Kramer JF, Nusca D, Fowler P, Webster-Bogaert S. Test-retest reliability of the one-leg hop test following A C L reconstruction. Clinical Journal of Sport Medicine 1992;2:240-243. 129. Booher LD, Hench K M , Worrell TW, Stikeleather J. Reliability of three single-leg hop tests. Journal of Sport Rehabilitation 1993; 2:165-170. 130. Tegner Y, Lysholm J. Rating systems in the evaluation of knee ligament injuries. Clinical Orthopaedics and Related Research 1985; 198:43-9. 169 131. Lysholm J, Gillquist J. Evaluation of knee ligament surgery results with special emphasis on use of a scoring scale. American Journal of Sports Medicine 1982; 10:150-3. 132. Wyatt MP, Edwards A M . Comparison of quadriceps and hamstring torque values during isokinetic exercise. The Journal of Orthopaedic and Sports Physical Therapy 1981;3:48-56. 133. Dietz V , Quintern J, Sillem M . Stumbling reactions in man: Significance of proprioceptive and pre-programmed mechanisms. Journal of Physiology 1987; 386:149-163. 134. Portney L G , Watkins MP. Foundations of Clinical Research: Applications to Practice. Norwalk: Appleton & Lange, 1993:722. 135. Gomez T, McConkey JP, Thompson P, Ratzlaff C, Dean E. Semitendinosus repair augmentation of acute anterior cruciate ligament rupture. Canadian Journal of Sports Science 1990; 15:137-42. 136. Kottke FJ. From reflex to skill: The training of coordination. Archives of Physical Medicine and Rehabilitation 1980; 61:551-561. 137. Huston LJ, Wojtys E M . Neuromuscular performance characteristics in elite female athletes. The American Journal of Sports Medicine 1996; 24:427-436. 138. Kottke FJ, Halpern D, Easton J K M , Ozel AT, Burrill CA. The training of coordination. Archives of Physical Medicine and Rehabilitation 1978; 59:567-572. 139. Collins DF, Cameron T, Gillard D M , Prochazka A. Muscular sense is attenuated when humans move. Journal of Physiology 1998; 508:635-643. 140. Rasch PJ, Morehouse LE. Effect of static and dynamic exercises on muscular strength and hypertrophy. Journal of Applied Physiology 1957; 11:29-34. 141. Moritani T, DeVries HA. Neural factors versus hypertrophy in the time course of muscle strength gain. American Journal of Physical Medicine 1979; 58:115-130. 142. Yue G, Cole KJ . Strength increases from the motor program: Comparison of training with maximal voluntary and imagined muscle contractions. Journal of Neurophysiology 1992; 67:1114-1123. 170 143. Pincivero D M , Lephart SM, Karunakara RG. Relation between open and closed kinematic chain assessment of knee strength and functional performance. Clinical Journal of Sport Medicine 1997; 7:11-16. 144. Delitto A , Irrgang JJ, Harner CD, Fu FH, Nessi S. Relationshiop of isokinetic quadriceps peak torque and work to one legged hop and vertical jump in A C L reconstructed subjects. Physical Therapy 1993; 73:S85. 145. Wilk K E , Romaniello WT, Soscia SM, Arrigo CA, Andrews JR. The relationship between subjective knee scores, isokinetic testing, and functional testing in the ACL-reconstructed knee. Journal of Orthopaedic & Sports Physical Therapy 1994;20:60-73. 146. Kiefer G, Forwell L, Kramer J, Birmingham T. Comparison of sitting and standing protocols for testing knee proprioception. Physiotherapy Canada 1998; 50:30-34. Appendix A POWER ANALYSIS 172 Power Analysis #1 A power analysis was performed prior to the start of the study to determine the number of subjects required. The power analysis performed was based on estimated values for the peak hamstring torque time variable. The following equations were used: 1. L A = n * q * the sum of (UJ-U.) sigma2* [ 1 +(q-1 )*rho] Where L A = The difference between test occasion. Where n = The number of subjects per cell. Where q = The number of test occasions. 2. L B = n * p * the sum of (uj-u.) sigma2*(l-rho) Where L B = The difference between groups. Where p = The number of groups. 3. LAB = n * the sum of (uii-ui-ui+u.) sigma2* (1-rho) Where LAB = The interaction effect between group by test occasion. Values for sigma and rho were estimated from those in the literature and the pilot study results. Changes in peak hamstring torque time over the 12 week period were also estimated from those in the literature. The results for L A , L B , andXAB were: L A = 0.92 when n is 5. Power is 0.1316. L B = 36.90 when n is 5. Power is 0.9998. LAB = 12.83 when n is 5. Power is 0.8900. Since the power levels were high for L B and LAB when n is 5, a sample size of 10 was chosen for this study. Power Analysis #2 A second power analysis was performed after the completion of the study to determine the actual power levels of the study. The results for L A , L B , and LAB were: L A = 1.95 when n is 5. Power is 0.2251. L B = 2.06 when n is 5. Power is 0.1929. LAB = 47.71 when n is 5. Power is >1.9999. Appendix B SUBJECTIVE RATING SCALES & SUBJECTIVE ASSESSMENT SHEET 174 Table B l . Modified Lysholm and Gillquist Knee Scoring Scale Limp (5 points) None 5 Slight or periodical 3 Severe and constant 0 Support (5 points) None 5 Stick or crutch 2 Weight-bearing impossible 0 Locking (15 points) No locking and no catching sensations 15 Catching sensation but no locking 10 Locking Occasionally 6 Frequently 2 Locked joint on examination 0 Instability (25 points) Never giving away 25 Rarely during athletics or other severe exertion 20 Frequently during athletics or other severe exertion (or incapable of participation) 15 Occasionally on daily activities 10 Often in daily activities 5 Every step 0 Pain (25 points) None 25 Inconstant and slight during severe exertion 20 Marked during severe exertion 15 Marked on or after walking more than 2 km 10 Marked on or after walking less than 2 km 5 Constant 0 Swelling (10 points) None 10 On severe exertion 6 On ordinary exertion 2 Constant 0 Stair-climbing (10 points) No problems 10 Slightly impaired 6 . One step at a time 2 Impossible 0 Squatting (5 points) No problems 5 Slightly impaired 4 Not beyond 90° 2 Impossible 0 175 Table B2. Tegner and Lysholm Activity Scale 10 Competitive sports 5. Work Soccer - national and international elite Heavy labor (e.g. building, forestry) 9. Competitive sports Competitive sports Soccer, lower divisions Cycling Ice hockey Cross-country skiing Wrestling Recreational sports Gymnastics Jogging on uneven ground at least twice 8. Competitive sports weekly Bandy 4. Work Squash or badminton Moderately heavy labor Athletics (jumping, etc.) (e.g. truck driving, heavy domestic work) Downhill skiing Recreational sports 7, Competitive sports Cycling Tennis Cross-country skiing Athletics (running) Jogging on even ground at least twice Motorcross, speedway weekly Handball 3. Work Basketball Light labor (e.g. nursing) Recreational sports Competitive and recreational sports Soccer Swimming Bandy and ice hockey Walking in forest possible Squash 2. Work Athletics (jumping) Light labor Cross-country track findings both Walking on uneven ground possible but recreational and competitive impossible to walk in forest 6. Recreational sports 1. Work Tennis and badminton Sedentary work Handball Walking on even ground possible Basketball 0. Sick leave or disability pension because of Downhill skiing knee problems Jogging, at least five times per week 176 THE EFFECTIVENESS OF PROPRIOCEPTION TRAINING IN THE ACL RECONSTRUCTED KNEE SUBJECT INFORMATION SHEET 1. Subject ID: 2. Date of Birth: 3. Age: 4. Sex: Female / Male 5. Height: 6. Weight: 7. Knee Involved: Left / Right, Dominant / Non-dominant 8. Date of Injury: 9. When were you first diagnosed with an A C L tear? 10. Date of Surgery: _ _ _ 11. Surgeon: 12. What was your physical activity level prior to this injury? (Use the Tegner and Lysholm Activity Scale) 13. Please describe the physical activities your were involved in prior to this injury. Type: . Frequency (per week): 14. Please describe the physical activities your are currently involved in. Type: Frequency (per week): 177 15. Does the above activities (question 14) cause pain and/or swelling in your A C L reconstructed knee? If so, please indicate how long these episodes (of pain and swelling) last for. 16. Please describe the physiotherapy (if any) you received for your injured knee prior to surgery. Please include the length of time (in weeks) your received physiotherapy and the frequency (per week) of your visits. 17. Please describe the physiotherapy (if any) you received after your surgery. Please include the length of time (in weeks) you received physiotherapy and the frequency of your visits (per week). 18. Please describe any other treatments (e.g. massage therapy) you have received for your injured knee. Please include the length of time (in weeks), frequency (per week) and indicate whether it was before and/or after surgery you received such treatment(s). 19. How was your anterior cruciate ligament injured (if known)? 20. Have you ever been diagnosed by a physician to have an arthritic condition involving your legs? If so, please indicate the date you were diagnosed and the type of arthritis. 178 21. Have you ever been diagnosed by a physician to have a neurological condition or disease (e.g. multiple sclerosis)? Is so, please indicate the date you were diagnosed and the condition. 22. To the best of your knowledge, do you have any cardiovascular, respiratory, systemic, metabolic condition that may limit your exercise tolerance? Is so, please explain. 23. Do you have pain and/or swelling in your A C L reconstructed knee during normal activities of daily living? If so, please indicate the frequency (per week) you experience such episodes. Appendix C TRAINING PROTOCOLS 180 Table C I : Exercise Protocol for Group One (Strength Training) * Week 1 - Week 12 1. Stationary bike. Cycle for 15 minutes at tension appropriate for individual. 2. Leg press exercises. 3 sets of 10 repetitions. Load will be 70% to 80% of one-repetition maximum (1-RM). 3. Hamstring curls. 3 sets of 10 repetitions. Load will be 70% to 80% of 1-RM. 4. Lunges. 3 sets of 10 repetitions. Load as appropriate for individual. 5. Calf raises. Load as appropriate for individual. 6. Half-squats. 3 sets of 10 repetitions. Load as appropriate for individual. 7. Hip abduction exercises. 3 sets of 10 repetitions. Load as appropriate for individual. 8. Hip adduction exercises. 3 sets of 10 repetitions. Load as appropriate for individual. 181 Table C2: Exercise Protocol for Group Two (Proprioceptive Group) Week 1 - Week 2 1. Stationary Bike. Cycle for 15 minutes at tension appropriate for individual. 2. Eccentric hamstring control using exercise tubing. Allow the use of ski poles for balance. 3 sets of 10 repetitions. 3. Circular wobble board. Two legs for 5 minutes with eyes opened. Rest. One leg for 2 minutes (2 repetitions). 4. Wheeled board. Sitting. Affected leg on the wheeled board. Eyes closed. Sudden forward force applied by investigator, subjects attempts to stop the board's movement as soon as possible by using their hamstring muscle. A session of 5 minutes. 5. Wheeled board. Standing. Affected leg on the wheeled board. Eyes closed. Sudden forward force applied by investigator, subjects attempts to stop the board's movement as soon as possible by using their hamstring muscle. A session of 5 minutes. 6. Fitter board. Both legs on the Fitter. Movement pattern is forwards and backwards. One session of 5 minutes with the right leg placed forward. One session of 5 minutes with the left leg placed forward. Eyes opened. Allow the use of the ski poles. 7. Step overs on a wooden box. Start on Week 2. Three patterns of movement. Each movement consists of 60 repetitions. 8. Carrioca. Both directions. 6 sets of 5 repetitions. Week 3 - Week 4 1. As above. 2. Eccentric hamstring control using exercise tubing. Increase to .4 sets of 10 repetitions. 3. Circular wobble board. Week 3: Affected leg standing for 2 minutes. Eyes opened. Forward 747 on the affected leg for 5 minutes. Eyes opened. Week 4: Eyes opened for all tasks. Forward 747 for 5 minutes. Reverse 747 for 5 minutes. Rectangular board. Both legs on the board. Balance is forwards and backwards. Eyes opened. One session of 2 minutes with the right leg forward. One session of two minutes with the left leg forward. 4. Wheeled board. Sitting. Week 3: 2 minute session. Week 4: Discontinue. 5. Wheeled board. Standing. Week 3: As above. Week 4: Eyes closed. 6. Fitter. Week 3: As above. Wean from ski poles if able. Week 4: Eyes closed. 7. As above. 8. As above. 9. Hopping patterns. Week 3: Side to side. Affected leg only. 50 repetitions. Week 4: Forward and backward. 2 legs. 20 repetitions. 10. Squat on square balance board. Week 5 - Week 6 1. As above. 2. As above. 3. Circular wobble board. Eyes closed for all tasks. Affected leg standing for 2 minutes. Forward 747 for 5 minutes. Reverse 747 for 5 minutes. Rectangular board. As above. 4. Discontinued. 5. As above. 6. Fitter. Affected leg standing for 5 minutes. Eyes opened. Allow use of ski poles. 7. As above. 8. Carrioca. Week 5: As above. Week 6: A drill session of 5 minutes. Each drill is approximately 35 seconds. 9. Hopping Patterns. As above. Week 5: Square pattern hop. Two legs. 20 repetitions. Week 182 6: Triangular pattern hops. Two legs. 20 repetitions in total. 10. As above Week 7 - Week 8 1. As above. 2. As above. 3. Circular wobble board. Eyes closed for all tasks. Forward 747 for 5 minutes. Reverse 747 for 5 minutes. Play catch for 5 minutes, standing on the affected leg only. Rectangular board. As above. Eyes closed for all tasks. 4. Discontinued. 5. As above. 6. Fitter. Affected leg standing for 5 minutes. Eyes opened. No poles. 7. As above. 8. As above. 9. Hopping patterns. As above. "X" pattern hops. 20 repetitions in total. 10. As above. Week 9 - Week 10 1. As above. 2. As above. 3. Circular wobble board. As above. Rectangular board. As above. Play catch for 5 minutes (balance is side to side). 4. Discontinued. 5. As above. 6. Fitter. As above. Eyes closed. Poles as necessary. Side to side movement (2 legs) with eyes opened for 5 minutes. No poles. 7. As above. 8. As above. 9. Hopping Patterns. As above. Week 9: 12 meter straight hop forwards (2 legs). Week 10: 12 meter straight hop backwards (2 legs). 10. As above. Week 11-Week 12 1. As above. 2. As above. 3. Circular wobble board. Discontinue above except for playing catch for 5 minutes. Affected leg standing. 3 positions, 5 seconds each, for a total of 5 minutes. Eyes closed. Rectangular wobble board. Discontinue above except for playing catch for 5 minutes. Play catch with balance forwards and backwards. One session of two minutes with the right leg forward. One session of two minutes with the left leg forward. 4. Discontinued. 5. As above. 6. Fitter. Week 11: As above. Week 12: Eyes closed for all tasks. 7. As above. 8. As above. 9. Hopping patterns. As above. Week 11: 12 meter zig-zag forward hop (2 legs). Week 12: 12 meter zig-zag backward hop (2 legs). 10. As above. Appendix D RELIABILITY OF SYMMETRY INDEXES 184 The following formula was used to calculate Intraclass Correlation Coefficient (ICC): ICC (3.*) = BMS -EMS BMS The following Intraclass Correlation Coefficients were calculated for absolute torque values and symmetry indexes: 1. Average concentric quadriceps torque of the A C L reconstructed limb = 0.97. 2. Average eccentric quadriceps torque of the A C L reconstructed limb = 0.91. 3. Average concentric quadriceps torque of the A C L intact limb = 0.97. 4. Average eccentric quadriceps torque of the A C L intact limb = 0.90. 5. Average concentric hamstring torque of the A C L reconstructed limb = 0.93. 6. Average eccentric hamstring torque of the A C L reconstructed limb = 0.96. 7. Average concentric hamstring torque of the A C L intact limb = 0.98. 8. Average eccentric hamstring torque of the A C L intact limb = 0.98. 9. Concentric symmetry index for the quadriceps = 0.85. 10. Eccentric symmetry index for the quadriceps = 0.62. 11. Concentric symmetry index for the hamstring = 0.57. 12. Eccentric symmetry index for the hamstring = 0.81. 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0077173/manifest

Comment

Related Items