UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Prediction of maximum oxygen uptake in paraplegics and quadraplegics using multiple regression equations Rogers, Allen Robert James 1981

You don't seem to have a PDF reader installed, try download the pdf

Item Metadata

Download

Media
UBC_1981_A7_5 R64.pdf [ 4.75MB ]
[if-you-see-this-DO-NOT-CLICK]
Metadata
JSON: 1.0077394.json
JSON-LD: 1.0077394+ld.json
RDF/XML (Pretty): 1.0077394.xml
RDF/JSON: 1.0077394+rdf.json
Turtle: 1.0077394+rdf-turtle.txt
N-Triples: 1.0077394+rdf-ntriples.txt
Original Record: 1.0077394 +original-record.json
Full Text
1.0077394.txt
Citation
1.0077394.ris

Full Text

PREDICTION OF MAXIMUM OXYGEN UPTAKE IN PARAPLEGICS AND QUADRAPLEGICS USING MULTIPLE REGRESSION EQUATIONS B.P.E., The University of British Columbia, 1975 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF THE FACULTY OF GRADUATE STUDIES Department of Sport Science School of Physical Education and Recreation We accept this thesis as conforming to the required standard by ALLEN ROBERT JAMES ROGERS MASTER OF PHYSICAL EDUCATION in THE UNIVERSITY OF BRITISH COLUMBIA May 1981 lAllen Robert James Rogers, 1981 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 ^cAcc^^^J ~ S^^-^f ^ /^^^ f ^J^a**, The University of British Columbia 2075 Wesbrook Place Vancouver, Canada V6T 1W5 Date ABSTRACT Twenty physically disabled subjects performed a progressive continuous exercise protocol on a wheelchair ergometer, to maximum exertion. Cardio respiratory responses were monitored by means of direct ECG recording, and HR was reported for the last 30 seconds of each workload (WL). Expired gases were continuously sampled and analyzed for 15 second determinations of respiratory gas exchange variables. The last 30 seconds at each WL was averaged and assumed to be'representative of steady state responses to that WL. In addition, body weights, ages, and maximum breath ing capacities were recorded. Two subjects were deleted from the study due to incomplete data. The inability to equate structural and functional characteristics of quadraplegics and paraplegics necessitated the division of the total group into paraplegics (N=13) and quadraplegics (N=5). For each paraplegic subject, three submaximal WL and corresponding cardiorespiratory responses were chosen for multiple regression analysis on maximal oxygen uptake. The three workloads were selected on the basis of HR responses, i.e., those workloads where HR was found to be between 65% and 85% of maximum HR (max HR = 220 - age). The lowest of the three workloads and corresponding cardiorespiratory responses (CRR) for each sub ject were assigned to the LHR group while the highest workloads and CRR for each were assigned to the HHR group. The remaining WL and CRR for each subject was assigned to the MHR group. Mean HR responses for the ii LHR, MHR, and HHR groups were: 131, 139,, and 148, respectively. These means corresponded to approximately 70%, 75%, and 80% of maximum heart rate. 2 The squared multiple correlation coefficients (R ) adjusted for both 2 sample size and number of variables contributing to R , were found to be .8761, .9218, and .9094 for LHR, MHR, and HHR, respectively. The respec tive standard error of estimates were reported to be .1397, .1101, and .1195 or + CV% equal to 6.5%, 5.2%, and 5.6%. Cross validation was not performed due to the-small sample size. However, the adjusted press prediction gives an indication of how the equa tion may predict for subjects outside the experimental group. Prediction is performed in turn for each subject with the effects of that subject's data removed from the beta coefficients. The mean absolute errors reported were 22.2%, 11.8%, and 13.2% for the LHR, MHR, and HHR groups, respectively. Multiple regression analysis of the quadraplegic data was restricted to the WL and CRR at the fourth minute of the progressive continuous work-2 load protocol. The adjusted R for the prediction equation produced was .9992 with a standard error of estimate of .0058 L/min. The variables 2 contributing to the R were: Ventilation, V02 ml/kg min., and WL. It was concluded that: 1. Multiple regression analysis appeared to be a suitable method of develop ing accurate prediction equations for MV02 L/min., in paraplegics. The best equation being: MV02 L/min. = 5.85 + 1.70(VC02 L/min.) - .026(age) - .0015(WL) - .008(HR) - 3.42(RQ). 2. The accuracy of prediction of MV02 L/min. in paraplegics is increased with the increase in the physiological stress (as reflected in % maximum HR) . iii 3. Due to the limited sample size, no conclusions were reached regarding prediction of MV02 L/min. within the quadraplegic population. iv TABLE OF CONTENTS ABSTRACT ii LIST OF TABLES viLIST OF FIGURESACKNOWLEDGEMENT viii Chapter I INTRODUCTION 1 Statement of the Problem 3 Subproblem 4 JustificationLimitationsDelimitations 5 DefinitionsAbbreviations 7 II REVIEW OF LITERATURE 8 Description of the Population 8 Prediction of Maximum Oxygen Uptake 14 III METHODS AND PROCEDURES 2Subjects 2Data Collection 4 Data Analysis 7 IV RESULTS AND DISCUSSION 29 Results 2Discussion 41 V SUMMARY AND CONCLUSIONS 50 Summary 5Conclusions 1 Recommendations for Further Research - 51 v REFERENCES 53 APPENDIX A The Wheelchair Ergometer 58 APPENDIX B Individual Cardiorespiratory Responses to the Workload Protocol 63 APPENDIX C Correlation Matrices 71 vi LIST OF TABLES Table 1 A Summary of Investigations into the Accuracy of Predicting Maximum Oxygen Uptake Utilizing the Methods Indicated 15 2 Protocols for Continuous Increasing Workloads 26 3 Subject Characteristics 30 4 Means and Standard Deviation of Variable Corresponding to LHR, MHR, and HHR 1 5 Summary of Correlation Coefficients between Predictors and Criterion MV02 L/min 32 6 Summary of Multiple Regression Analysis of Paraplegic Data .... 34 7 Summary of Error Associated with the Prediction of MV02L from Prediction Equations for LHR, MHR, and HHR 47 8 Summary of Univariate Statistics for the 4th Minute of Exercise (Quadraplegics) 49 LIST OF FIGURES Figure 1 Comparison of MV02L predicted (unadjusted) vs Observed MV02L for LHR group 35 2 Comparison of MV02L predicted (unadjusted) vs Observed MV02L for MHR group 6 3 Comparison of MV02L predicted (unadjusted) vs Observed MV02L for HHR group 37 4 Comparison of MV02L predicted (adj. press) vs Observed MV02L for LHR group 8 5 Comparison of MV02L predicted (adj. press) vs Observed MV02L for MHR group 39 6 Comparison of MV02L predicted (adj. press) vs Observed MV02L for HHR group 40 7 Internal Resistance vs Weight in Wheelchair 68 Photograph of Wheelchair Ergometer 61,62 vii ACKNOWLEDGEMENT The author extends appreciation to the members of the committee, Dr. Ken Coutts [Chairman}, Dr. Stan Brown, Mr. Bert Halliwell, and Dr. S. Pinkerton, for their participation in this investigation. Special thanks to Mr. M. Walsh for his technical assistance in the testing session and Mr. Doug Dunwoody for his aid in the data resolution. Last, but not least, special thanks to my family who have endured much, allowing me to pursue my extended period of study. viii CHAPTER I INTRODUCTION Several studies have directly assessed the maximum aerobic capacities of paraplegics and quadraplegics (Knutsson et al., 1973; Zwiren and Bar-or, 1975; Wicks et al., 1977; Cameron et al., 1977; Ernes, 1977; Hjeltnes, 1977). Unfortunately, the data tells us little about the cardiorespiratory "fitness" of the individuals. Without norms, interpretations are limited. The data obtained by the investigations listed above, represent only a sub-population, i.e., those capable and motivated enough to perform exercise to maximum oxygen uptake. The results obtained from investigations (Heigenhauser et al., 1977; Zwiren and Bar-or, 1975) indicated that cardiorespiratory capacities of the inactive paraplegics are significantly lower than sedentary normals. This information supports the concern expressed by other investigators who have indicated a need to monitor the cardiorespiratory fitness of the physically disabled (Hjeltnes, 1977; Engel & Hilderbrandt, 1973; Marincek et al., 1977; Knutsson et al., 1973). For such tests to be suitable for the physically disabled population, the following criteria are suggested: 1. The test should be submaximal in nature. Although indications are that paraplegics and quadraplegics who are stabilized, can perform maximal tests, the usefulness of such tests is limited to this stabilized segment of the population. During the rehabilitation stage, aerobic testing 1 2 would prove extremely valuable. However, maximum exertion is ill advised since the subjects are not stabilized and prone to dizziness and fainting. In addition, maximum exertion is not recommended for the elderly. 2. The test should be easy to administer and its duration relatively short so that large population studies will not be overly laborious. 3. The test should be valid and reliable in its prediction of maximum oxygen uptake. Direct assessment methods do not meet the first two criteria above. Thus, indirect methods are considered to be more practical for the purpose of cardiorespiratory fitness assessment with the physically disabled popula tion. Several indirect tests for the prediction of maximum oxygen uptake are available to the normal population. A summary of the predictive capac ities of each is provided in Table 1 .(page 15). The most widely used indirect tests are those which employ extrapola tion of the plot of heart rate vs. oxygen uptake (Astrand & Ryhming, 1954; Margaria et al., 1965; Maritz et al., 1961). Serious problems exist with these methods. Most of the problems can be ascribed to three basic assump tions incorporated into the extrapolation methods. These are as follows: 1. A linear relationship exists between heart rate and oxygen uptake, at all levels of work. 2. There is one maximum heart rate value for all members of a given popula tion. 3. The mechanical efficiency for all subjects is about 23% (where V02 for submaximum workloads is not measured directly). The validity of assumptions 1 and 2 has been questioned by several investigators (Maritz et al., 1961; Flandrois & LaCour, 1971; Davies, 1968; Glassford et al., 1965; Wyndham, 1967; Wyndham, 1959; Rowell 3 et al., 1964). A coefficient of variation of about 4-5% in mechanical efficiency is reported (Shephard, 1977). Error of up to 20% (Hermiston & Faulkner, 1971) can be ascribed to violation of one or more of these basic assumptions. The unsatisfactory results obtained using the extrapolation, has led to the investigation of a large number of variables and their relationships to maximum oxygen uptake. Investigators have attempted to improve predic tion by incorporating more variables into the prediction procedure. This is accomplished through multiple regression analysis and multiple regression equations (Hermiston & Faulkner, 1971; Mastropaolo, 1970; Bonen et al., 1979; Fox, 1973; Jessup et al., 1974; Jette et al., 1976; 'Bell & Hinson, 1974). Table 1 summarizes the majority of the results of these investiga tions. The variability in maximum HR found within the paraplegic population, as well as the lack of information relating other physiological variables during exercise, suggests that extrapolation may not prove a viable method of predicting MV02L. Multiple regression appears to be most suitable for the development of maximum oxygen uptake prediction tests with the physically disabled. Statement of the Problem The purpose of this investigation is to determine which of certain structural and functional measures of physically disabled subjects (predic tors) are significantly related to maximum oxygen uptake (criterion) and to establish the greatest multiple correlation coefficient between the pre dictors and the criterion. 4 Subproblem To construct a multiple regression equation for the prediction of maximum oxygen uptake. Justification At present, no indirect cardiorespiratory fitness test is available for use with the physically disabled population. Direct tests require elaborate testing equipment, maximum effort from the subjects, and considerable time to administer. Population studies are not practical with these direct methods. In addition, direct methods are not recommended for unstabilized subjects nor for the elderly. In contrast, indirect tests require little equipment, take approximately six minutes to complete and only submaximal efforts from the subjects. Information regarding physiological changes, as a result of bed rest, a training regime, or rehabilitation procedures, can be made available through repeated testing. This feedback is extremely important when monitoring an individual's rehabilitation, to assist in further exercise prescription and as a motivational factor for the subject. Limitat ions 1. The results of the multiple regression analysis of the data collected on the thirteen paraplegic subjects may be inferred to subjects within the age range and level of spinal lesion associated with the sample inves tigated in the present study. 2. The small sample size of the quadraplegic group, does not allow justified inference to the population outside the experimental group. 3. The assumptions associated with the protocol for workload adjustments, i.e., the final 30 seconds at each workload (duration one minute) 5 represents a steady state. 'h. The equation(s) developed for the prediction of MV02L do not constitute a "fitness test." MV02L indicates aerobic power. In addition, cross validation was not performed. Delimitations The present study is delimited to: 1. male paraplegics and quadraplegics between the ages of 18 to 53 2. the variables (predictors) submitted to regression analysis 3. the wheelchair ergometer constructed for workload variations h. the Pearson-product-moment correlation analysis 5. the method of expression of aerobic capacity, i.e., MV02L. Definitions 1. Stabilized: A post rehabilitated state in which the subject's cardio vascular system has been returned to a state which reduces the chance of the orthostatic reaction. This is accomplished via an increase in constrictory vasomotor tone in the vessels serving the immobilized seg ments of the body. 2. Orthostatic reaction: A condition resulting from a loss of tone in blood vessels serving the upper and/or lower limbs. The condition may be further compounded by loss of the sympathetic innervation to the heart and resulting bradycardia. The results of the above is a hypo kinetic circulation of blood and blood pooling in the extremities. 3. Classifications for sporting events: Class IA—Upper cervical lesions with triceps non functional against gravity. A person with this disability cannot lift his arms above his head and cannot grip with his hands. Class IB—Lower cervicals with good triceps and strong finger flexors or extensors of functional value. A person with this disability can lift his arras above his head but not against resistance and can grip with some strength. Class IC—Lower cervicals with good triceps and strong finger flexors and extensors. No onterossei or lumbricals of functional value. A person with this disability can lift his arms above his head against resistance arid can grip firmly. Class II—Below Thl-Th5 inclusive. No balance when sitting. This person has weak abdominal and back muscles but has full use of his arms Class III—Below Th5-Thl0 inclusive. Ability to keep balance when sitting ignoring nonfunctional lower abdominal muscles (cannot act with out falling over if slightly pushed). Class IV—Thll-Thl3 inclusive. A person with this disability is affected from the hips down and in some cases will have some balance difficulties. Class V—Below L3-S5. This person is usually affected from the hip down and quite often only one leg. Class VI—This class for swimming competitions only. Below L5. This person is able to kick with some effect. 2 2 Adjusted R : R adjusted for both sample size and number of subjects in sample, i.e., adj R = 1 - K yr- Y J (N - m) where: N = size of the sample m = number of variables on the problem (N-m) = degree of freedom K = (1 - R2) 7 5. Adjusted Press Prediction: Predicted value for the case after removing the effects of that case removed from the regression coefficients. 6. Deleted Press Residual: The residual error for each case from predicting that case by means of the Adjusted Press Prediction. Abbreviat ions 1. MV02L: Maximum oxygen uptake in Liters/minute 2. MV02ml: Maximum oxygen uptake ml/kg.min. 3. V02L: The rate of oxygen uptake L/min. 4. V02ml: The rate of oxygen uptake ml/kg;min. 5. VC02L: The rate of carbon dioxide expired L/min. 6. VC02ml: The rate of carbon dioxide expired ml/kg-min. 7. Sub-RQ: Submaximum Respiratory Quotient 8. Vent.: Ventilation rate L/min. 9. HR: heart rate beats/min. 10. WIT.:-' Workload Kpm/min. CHAPTER II REVIEW OF LITERATURE Description of the Population Lesion of the spinal cord results in the loss of nervous supply to the segments of the body below the level of the lesion. Atrophy of the muscles normally served by the respective nerves, is a common observation. Body weights in paraplegics and quadraplegics have been reported to average about 80% of predicted values from height tables (Hjeltnes, 1977). The loss of innervation is not restricted to the muscle tissue. A lesion at any level results in the loss of central control of sympathetic outflow to parts of the body below that level. A lesion at the level of the splanchnic plexus, thoracic 6-7 vertebrae, results in the greater part of the body and blood vessels being deprived of the normal sympathetic vaso motor control (Wolf & Magora, 1976). The loss of the splanchnic outflow results in reduced capacities of the circulatory system to adapt to stress, i.e., changes in body position and/or exercise (Knutsson et al., 1973; Wolf & Magora, 1976). Erect posi tion causes an accumulation of blood in the lower extremities. Exercise of the upper limbs causes a vasodilation in these muscles, thus blood pools in the extremities. Redistribution of blood from the core to the working muscles, via a vasoconstriction in the abdominal organs, is absent or reduced. The decrease in peripheral resistance accompanied by the inability to redistribute blood, results in a reduced return to the heart. Increases 8 9 in cardiac output are limited and a fall in blood pressure occurs. Wolf and Magora (1976), investigated the effects of position change in relation to systolic blood pressure. Various levels of spinal lesions were observed. Eighteen men, ages 18-62 (3 quadraplegics, 5 high thoracic paraplegics, 7 low thoracic, and 3 lumbar paraplegics) were investigated. Systolic blood pressure decreased markedly in cervical and high thoracic patients in the erect position. Further decreases were seen during effort. Low thoracic and lumbar groups showed little change in systolic blood pres sure as a result of change in body position. There was:-marked decrease in systolic blood pressure during effort. With increased time following the spinal lesion, an increased toler ance to changes in body position occurs. A change in renin release has been suggested to explain the increased tolerance to vertical position (Knutsson et al., 1973; Guttman, 1946; Guttman, 1954; Jonason, 1947), -since increased renin release has been reported after a series of repeated changes in body position (Johnson et al., 1969). The increased tolerance to vertical position could not be accounted for by increased blood volume, as these are reported to be low in chronic paraplegia (Knutsson et al., 1973). The problem of circulatory adjustment to stress is further complicated with high spinal lesions, those above the thoracic 6-7 vertebrae. High spinal lesions deprive the upper and lower body of sympathetic outflow. The sympathetic acceleratory influence to the heart is reduced or.abolished (Knutsson et al., 1973; Freyschuss & Knutsson, 1969: Freyschuss, 1970; Wolf & Magora, 1976). Knutsson et al. (1973), reported subjects with lesions between cervical vertebrae 5 (C5) to thoracic vertebrae 3 (Th3) incomplete and complete at 10 Th4, to have maximum heart rates varying between 100 and 130 beats/minute. Wolf and Magora (1976), found in two patients with cervical lesions, the heart rate did not increase over 120 beats/minute. Only slight increases in heart rate were observed in patients with high thoracic lesions. Normal physiological responses of heart rate were found in both low thoracic and lumbar groups. Freyschuss and Knutsson (1969), observed in patients with complete cervical cord transections that heart rate increases normally seen during voluntary contraction in a non-paretic muscle group, were completely abol ished by atropin block. The increased heart rate response to effort to contract remained intact in healthy normals after atropin block (Freyschuss, 1970). Increased heart rate response must originate from the supraspinal centers and be elicited by an inhibition of vagal outflow to the heart, (Knutsson et al., 1973). He concluded that heart rate regulation in com plete cervical cord transections is attained by varying the vagal tone. Wicks et al. (1977), examined 72 athletes at the 1976 Olympiad for the physically disabled. The average maximum heart rates for paraplegics and quadraplegics with spinal cord lesions were: 182 + 13 beats/minute and 132 + 17 beats/minute, respectively. Paraplegics, victims of polio, did not differ from those with spinal injuries. However, quadraplegics, polio victims, had heart rates averaging 167 + 27 beats/minute as compared to the 132 + 17, found with paraplegics with spinal cord lesions. Nilsson et al. (1975), found in two subjects with high lesions (C6-7 and C7-Thl) had maximum heart rates of 165 and 150, respectively. Similar results were reported for one subject, age 24, with a spinal lesion at Th2, whose maximum heart rate was reported to be only 160. Eight other subjects (low thoracic) had normal maximum heart rate values reported. 11 Corbett et al. (1971), reported that heart rate response of quadra-plegics to head-up tilting was greater than that which could be explained by variation of vagal tone. The beta-receptor reflex acting through the isolated spinal cord has been suggested as an explanation for this and the well-known hyperflexia in patients with high spinal cord transections (Guttman, 1947; Pollock et al., 1951; Cunningham et al., 1953; Whitter-idge, 1954; Kurnick, 1956; Cole et al., 1967). The mode of increasing cardiac output during exercise is mainly via increased heart rate. Stroke volume has been found to increase only slightly. Hjeltnes (1977), investigated the cardiovascular adaptations to work of nine paraplegics, Th6-Thl2 lesions. Increases in cardiac output ranged from 54-1057° with a mean of 677o. Stroke volume increases accounted for 6-367= (mean 247=,) of the increased cardiac output. At an oxygen uptake of 1 liter/minute, the extrapolated values of stroke volume in paraplegics were 43-66ml. Corresponding values in seven healthy subjects were 59-100ml. In one subject, oxygen uptake increased 627., cardiac output increased 277>, while stroke volume decreased 127,. Lower stroke volumes may be accounted for by decreased venous return as a result of hypokinetic circulation. In higher lesions, the reduced sympathetic inotropic effect on the heart muscle, which normally results in greater force of cardiac contraction and lower end systolic volumes, may be a factor. Few studies have attempted to assess the maximum oxygen uptake capac ities of wheelchair subjects. Wicks et al. (1976), assessed maximum oxygen uptake capacities of 72 athletes. The findings were categorized on the basis of International Classifications, i.e., IA, IB, II, III, IV. The reported maximum capacities are: .15.9, 15.8, 24.0, 31.1, 39.0 ml/kg-min., 12 respectively. Cameron et al. (1976), examined the aerobic capacities of 42 athletes. Categorization was made on the basis of the type of sport participated in. Wheelchair track and swimming athletes had maximum oxygen uptakes in excess of 40 ml/kg-min. Skill athletes had the lowest maximum capacities, 24.4 + 6.2 ml/kg.min. Strength athletes had values slightly greater, 25.6 + 4.5 ml/kg-min. Non specialization in sport participation made it difficult to identify characteristic types for the above. Zwiren and Bar-or (1975), compared four groups: normal athletes (NA), normal sedentary (NS), wheelchair bound athletes (WA), and wheelchair bound sedentary (WS). The WS subjects were all with lesions below the Th7 level. Arm work was performed with no significant difference found between NA and WA (MV02ml). Significant differences were reported when maximum oxygen uptake was expressed as MV02L. The differences in lower body mass, between the two groups probably is an important factor in the interpretations of these results. The author suggested that WS and NS were different in terms of maximum oxygen uptake, even though a significant difference was not found. Nilsson et al. (1975), examined the aerobic capacities of 12 rehabili tated paraplegic subjects and the effects of training on their aerobic capac ities. Two subjects with cervical lesions had maximum oxygen uptakes aver aging 16.8 ml/kg-min. (pre-training). The remaining subjects with thoracic lesions, pre-training values averaged about 20.3 ml/kg-min. The subjects varied considerably in age and habitual physical activity levels. Only one of the subjects with a cervical lesion participated in a trainig pro gram. His maximum oxygen uptake increased 3.7 ml/kg-min. Physical training increased in the mean MV02ml, for the group with thoracic'lesions, to 26.9 ml/kg.min. Hjeltnes (1977), examined nine paraplegic subjects, eight with low thoracic lesions (Th6-Thl2) and one with a Th2 lesion. Subjects varied in age from. 17-46 years, mean 26.8. Maximum oxygen uptake ranged from I. 1 l/min. to 1.7 l/min. (20.8 ml/kg-min. to 36.6 ml/kg-min., mean 27.4 ml/kg-min.) Several investigators have reported the efficiency with which the subjects perform the work. Brubaker et al. (1979), reported mechanical efficiencies at three different work loads: .25, .33, and .50 watts/kg of body weight. The mechanical efficiencies reported are: 9.33, 10.55, and II. 49 per cent, respectively. Mechanical efficiencies at two different speeds were also investigated, corresponding to workloads of 2.0 and 3.0 kpm/min. Mechanic efficiencies were reported to be 11.29% and 9.62%,, respectively. Barr and Glaser (1977), reported mechanical efficiency to decrease with increased workload. The value decreased from 9% to 67o for workloads varying from 50 to 150 kpm/min. No mention was made of how load increases were achieved, i.e., by increases in speed or increases in resistance. Glaser, Young, and Suryaprasad (1977) investigated the mechanical efficiency of various methods of striding, i.e., normal-synchronous versus asynchronous technique. Mechanical efficiencies of 4.7% and 7.47o, respectively, were reported. Marincek and Vojko (1978), using arm eyeloergometry, reported mechan ical efficiencies of five subjects to vary from 16.1 to 20.77.= . These results agree with the findings of Bevergard et al. (1966), who reported mechanical efficiencies of 18 and 23% for arm and leg work, respectively. Nilsson et al. (1975), examined the effects of training on mechanical efficiency of 12 paraplegics. Arm cycloergometry was employed. Pre-trained values averaged: 16.0 + 1.9% at submaximal work of 300-370 kpm/min. and 18.3 + 2.9% at maximum effort. These values were increased to 18.3 + 2.9%. and 21.5 + 2.9%, respectively. The variation in magnitude of the reported mechanical efficiencies for the various studies may result from the use of different methods of calculation. In most cases the method was not reported making interpre tations difficult. Prediction of Maximum Oxygen Uptake Prediction of maximum oxygen uptake has been and continues to be the concern of many exercise physiologists. Several methods have been devel oped. Researchers have reported varying degrees of success with each. Table 1 lists the majority of the methods reported to date, and the accur acy of prediction using the various methods. It is not the purpose of this chapter to do an in-depth review of each method and/or the various studies which have looked at these. The reader is referred to Astrand and Rodahl (1970) and Davies (1968) for a more in-depth review of the prediction of maximum oxygen uptake by means of the extrapolation methods. The limitations of each have been well documented in these and other exercise physiology texts. An overview of the more popular methods will be made here with a more detailed look at prediction of maximum oxygen uptake via multiple regression equations. Astrand and Rhyming (1954) , '.initiated one method which utilized heart rate and the measurement of or estimated oxygen uptake at a submaximum work load, to predict maximum oxygen uptake values. A straight line is fitted between a "common'1' heart rate of 61 beats/min. (at zero oxygen consumption) and the measured heart rate at the oxygen consumption for a particular sub-Table 1. A Summary of Investigations into the Accuracy of Predicting Maximum Oxygen Uptake Utilizing the Methods Indicated Study* Sample distribution Particulars Error Astrand & Rhyming (1954) method Astrand & Rhyming (1954) Hermiston & Faulkner (1971) 28 Davies (1968) Glassford et al. (1965) 80 24 males males females females normals normals, ages 20-50 physically active males bike ergometry bike ergometry bike ergometry bike ergometry 900 kpm 1200 kpm 600 kpm 900 kpm MV02 determined by treadmill submaximal HR less than 140 (N=5) submaximal HR greater than 140 (N=23) prediction via nomogram bike ergometry submaximum HR 120-140 submaximum HR 140+ prediction via nomogram prediction via nomogram compared to: MV02 observed bike ergometry M0V2 observed treadmill "Study—Investigations which have tested the accuracy of the indicated methods Errors: All errors reported as mean % error of prediction + standard deviations . „^ sum of residual errors -, nria, . M ""Absolute percent variation + SD, i.e., ; , x iOO/o - N "''"'Coefficient of variation, i.e. observed MV02 standard error of estimate mean observed MV02' 100% 10.4%-6.IV 14.4%-9.4%-< 15% + 11 18% + 12 12% + 8 9% + 9 0% +20 Table 1, continued Study-Sample distribution Particulars Error Rowell et al. (1964) Joseph et al. (1973) DeVries & Klafs (1965) Verma (1977) Maritz-Wyndham (1967) method Davies (1968) Rowell et al. (1964) 10 10 12 20 80 10 nonathletes nonathletes athletes males, ages 20-30 16 males, ages 20-26 45 normals, ages 20-50 pre-training post-training prediction via nomogram MV02 via bike ergometry prediction via nomogram WL for all subjects 750 kpm MV02 via bike ergometry prediction via nomogram WL all subjects 900 kpm MV02 determined via bike ergometry prediction via nomogram . prediction via extrapolation of the line produced from two submaximum work rates (sub HR between 130-170) bike ergometry endurance athletes treadmill exercise ages 18-24 extrapolation to MAX HR of 195 sedentary males pre-training ages 20-30 post-training 27% + 7 14% + 7 6% + 4 5% + 9 9.3%*" 11.62 + .72 12% + 9 15% + 8 23% + 7 18% + 8 Table 1, continued Study* Sample distribution Particulars Error Verma (1977) 45 moderately bike ergometry active WL: 600, 750, and 900 best-fit line fitted to three points; extrapolated to 180 Max HR 14% + 1** Margaria et al. (1965) method Margaria et al. (1965) Davies (1968) 80 males and females ages 9-47 normals, ages 20-50 step test (30-40) cm bench) prediction via nomogram step test prediction via nomogram 1% + 6 10% + 7 Issekutz et al. (1962) method Issekutz et al. (1962) 24 males, ages 20-65 females, ages 55-65 all untrained bike ergometry change in RQ, ie., log RQ vs V02L produces a straight line .0% + 4.86 DeVries & Kalfs (1965) 16 males, ages 20-26 MV02 determined via bike ergometry 8% + 18 Joseph et al. (1977) 14 soldiers, ages 20-30 bike ergometry 18% + 11 Table 1, continued Study-' Sample distribution Particulars Error Shephard (1967) 10 sedentary progressive step test 07» +8.5 Fox (1973) method Fox (1973) 87 untrained college males MV02L = 6,300 - 19.26 (HR) HR: heart rate response to 150 watts (work rate during bike erg.) linear regression .01 + 7. 9.6%*** Hermiston & Faulkner (1971) method Hermiston & Faulkner (1971) 36 36 males, active males, inactive treadmill multiple regression separate equations for each group prediction accuracy increased over total group prediction equation anthropometric and cardiorespiratory variables 27, + 8 Mastropaolo (1970) method Mastropaolo (1970) 13 middle-aged males bike ergometry multiple regression anthropometric and cardiorespiratory variables 57> + 3 6.67*-Co 19 maximum workload. Maximum oxygen uptake is obtained from the extrapola tion of the straight line to a population maximum heart rate (195 beats/ min). Nomograms have been developed for-both step-tests and bike ergometry. Maritz et al. (1962) utilized four submaximal rates of work. V02L and HR for each workload were plotted and a straight line fitted to four pairs of values. Extrapolation was made to a maximum heart rate of 180 beats/min. Work was performed on a bike ergometer. Margaria et al. (1956), employed two rates of work (stepping up and down, on and .off a bench) which produced a heart rate between 100-150 beats/ min. Adjustments were incorporated for the very young and very old, i.e., three maximum heart rate lines are given in the nomogram to take into account the effects of age on maximum heart rate. All three of these extrapolation methods rely on assumptions: 1. A linear relationship exists between heart rate and oxygen uptake. 2. Inter-individual variations of heart rate about the population mean is sufficiently small for the population mean to be used as the maximum heart rate for all subjects. 3. The mechanical efficiency for all subjects is about 23% (when oxygen uptake for submaximum work is assumed). The validity of 1 and 2 has been questioned by many investigators, as noted in Chapter I. The coefficient of variation in mechanical efficiency of 4-5%, reported by Shephard (1977), is suggested to be 6% by Astrand and Rodahl (1970). The accuracy of prediction is dependent on a number of factors which reflect the validity of the above assumptions. Fitness levels are of prime concern. Davies (1968), notes that only in subjects with a high observed maximum oxygen uptake (where the decline in maximum heart rate tends to 20 compensate for the asymptotic nature of heart rate), does the procedure of extrapolation of the line HR versus V02 to a mean population pulse of 190, produce realistic results. Underestimation of maximum oxygen uptake is the trend for more sedentary subjects. Rowell et al. (1964), reported similar conclusions. In the case of younger subjects, any method which employs a maximum heart rate of 170 or 180 will underestimate the true maximum oxygen uptake. Age adjustments have been made in some cases (Astrand & Rodahl, 1970; Margaria, 1965). Prediction is also dependent on many environmental factors. Ambient temperature, humidity, and the partial pressure of oxygen (elevation) can all influence the physiological stress placed on the body and in so doing, influence the submaximal response (Astrand & Rodahl, 1970). Rowell et al. (1964), listed several factors that will cause heart rate to vary independent of oxygen uptake. These include physical condi tioning, elapsed time after previous meal, total circulating hemoglobin, the degree of hydration of the subject, and hydrostatically induced changes resulting from prolonged erect position. Issekutz et al. (1962), investigated an alternative method to extrapo lation, i.e., the change in the respiratory quotient. They reported the value: working RQ - .75, increased logarithmically with the workload and maximum oxygen uptake was reached when this value became equal to .40. Varying degrees of success have been reported using this method. Table 1 may be referred to for a summary of the various results. The variability in success found with the preceding methods, has led investigators to the use of more variables for the prediction of maximum oxygen uptake. Multiple regression equations have been developed by 21 several authors (Bell & Hinson, 1974; Bonen & Babineau, 1977; Fall et al., 1966; Fox, 1973; Hermiston & Faulkner, 1971; Jessup et al., 1974; Jette et al., 1976; Mastropaolo, 1970). Mastropaolo (1970), obtained a simple regression by stepwise multiple regression analysis. Thirteen middle aged men were exercised to maximum oxygen uptake. Submaximal and maximal heart rate, systolic blood pressure, expired volumes, expired carbon dioxide and oxygen were determined. In this study submaximum RQ and maximum oxygen uptake were found to be highly correlated, r=.89 and a standard error of estimate of .175. The addition of work rate raised the multiple correlation to .92 and decreased the stand ard error of estimate:to .156 L/min. The multiple regression equation devel oped from these two submaximum variables is as follows: MV02L = 11.158 -0.007(WL) - 4.517(RQ). The reported best prediction equation was: MV02L = 14.703 - 4.909(RQ) - 0.008(WL) - 0.004(blood pressure) + 0.018 (Vent) - 16.083(V02L), obtained at 600 kpm/min. The multiple regression correlation coefficient was reported to be .93 with a standard error of estimate of .172 L/min. Deviations from the true values ranged from -87« to + 11%, mean of +.37o, absolute mean of 5.47o with a standard deviation of 3%. Of the 13 subjects, 3 trained for 12 weeks and were tested again. Pre-training values resulted in an estimate between -57o and +470 of true MV02L, while post training prediction underestimated MV02L by 87o. It is noted that Mastropaolo reported the second multiple regression equation as the "Best." The standard error of estimate was greater and 2 the adjusted R would be considerably less than that reported for the equa tion using only RQ and WL. When small sample sizes are used, it is 2 important to report the adjusted R value. Small sample sizes tend to inflate the multiple correlation coefficient, as does the use of a large 22 number of variables. Hermiston and Faulkner (1971), looked at 25 anthropometric and cardiorespiratory submaximal variables. Data were collected on 60 men. The overall group was divided on the basis of physical activity levels into two interlocking groups, a physically active group (N=36) and a physically inactive group (N=36). Data on 12 border-line subjects were included in both groups. Prediction equations.; were developed for: total group data; as well as for each group. The regression equation for the total group did not provide an accurate prediction of maximum oxygen uptake, R=.54. Regression equations for each of the sub-groupings raised the R to .90 in each case. The percent error for prediction using either equation was reported as 2+8%. Fox (1973), found a multiple correlation of .78, utilizing body weight, height, and submaximal heart rate at a workload of 150 watts. This did not significantly predict MV02ml better than the use of heart rate response alone. The prediction equation (utilizing only the heart rate response during the fifth minute of exercise at a work rate of 150 watts) is as follows: MV02ml = 6300 - 19.26(HR), standard error of estimate: 246 ml/min. (7.8%,). Prediction of MV02ml made on a group of subjects taken from the literature, was not significantly different from measured values before and after training, or with age variation (X + SD : 3.13 + .43 L/min., measured: 3.1 + .36 L/min., predicted [r=.83]). This method does not rely on the premise that HR increases linearly with oxygen consumption and workload over the entire range of workloads to maximum effort. Jette et al. (1976), investigated the possibility that maximum oxygen uptake could be predicted from independent variables measured during the administration of the Canadian Home Fitness Test. Fifty-nine subjects, 23 ages 15-74 years, underwent the fitness test and progressive exercise tread mill test for direct determination of volitional maximum oxygen uptake. The following multiple regression equation was found to produce a multiple R of .905: MV02ml = 42.5 +• 16.6(V02L) - .12(Wt) - .12(post exercise heart rate) - .24(AG). V02L as used in this equation, represents the average oxygen cost for the last completed exercise stage of a steptest (table values). Bonen et al. (1979), investigated the use of multiple regression equa tions for the prediction of maximum oxygen uptake in boys, ages 7-15. Data were collected on 100 subjects. Prediction equations for MV02L were obtained from subjects height, V02L, and HR observed during the third minute of a treadmill walk, R = .95, CV = + 9.77o. When just subject height, weight, and age were used, similar results were obtained. MV02ml was also predicted from age, HR, VC02L, and VC02ml. Slightly better accuracy was the result, coefficient of variation equal to 8.47c Cross-validation on 39 boys (trained) resulted in a prediction error of about l-27» + 97„. The use of just age, height, and weight was found to underestimate both MV02L and : MV02ml. Jessup et al. (1974), incorporated the results of a 12-minute run, Astrand-Rhyming test, age, height, weight, diastolic blood pressure, and leg length to predict maximum oxygen uptake. Forty male volunteers were studied. The best prediction equation was: MV02L = 1.46 + 0.005(AG) -0.118(height) + 0.014(WT). + 0.007(diastolic blood pressure) + 0.099(leg length) + 0.232(12-minute run) + Astrand(0.345). The multiple correlation coefficient was reported as: 0.814 with a standard error of estimate equal to 0.188 l/min. CHAPTER III METHODS AND PROCEDURES Subjects Twenty male subjects volunteered to take part in the study. The group consisted of 5 quadraplegics and 15 paraplegics. Ages ranged from 17 to 53. All subjects were stabilized and had spent a minimum of six months in their wheelchairs. Two subjects were eventually rejected from the study for two reasons: 1) premature termination of the work session, i.e., RQ .86; 2) ECG was lost during the work session. Data Collection Testing took place in the Buchanan Fitness and Research Center, Univer sity of British Columbia. Heart rate was monitored by direct ECG utilizing an A'vionic 4000 cardiograph with oscilloscope and ST depression computer and display. Heart rate responses were measured during the final 15 seconds of each minute of both rest and exercise. Expired gases were continuously sampled and analyzed by a Bechman Metabolic Measurement Cart (BMMC) interfaced into a Hewlett Packard 3052A Data Acquisition system for 15 second determinations of respiratory gas exchange variables. Each subject reported to the lab on the day of testing. Body weights were measured in a variety of ways, depending on the subject. Lighter sub jects (unable to stand) were held by one of the testers and the total of 24 25 the two individuals was measured. The tester's weight was then subtracted from the total. With larger individuals, the weight scale was placed on a table. Subjects, with the aid of the testers^,lifted themselves out of their chairs.' onto the ;scale. Finally, subjects with very low spinal lesions or other incapacitation which left them able to stand (with assistance), were weighed standing on the scales. All body weights were assessed with subjects wearing pants or sweat pants less shoes and shirts. Maximum breathing capacities were assessed prior to the metabolic measures. A Collins 13.5 liter respirometer was used for this measurement. Two 12-second trials were permitted. The first usually serving/as a practice trial and the second as the recorded measure. In all cases, the best score was recorded. During a progressive continuous work session, cardiorespiratory data were collected. The protocol for load increases varied between quadraple-gics and paraplegics. Initial load increases were achieved by increases in resistance. Increases of 1 kg were applied each minute until values of 4.5kg and 7.5kg were achieved for quadraplegics and paraplegics, respec tively. Subsequent load increases were arrived at by increased speed. Initial speed was set at 20 rpm of the wheelchair wheels and increased 5 rpm/min. until the subject could no longer match the required work rate or the subject terminated the work bout. Table 2 illustrates the work proto col for the two groups. For each workload the last two of the four respiratory gas exchange determinations were averaged, i.e., the last 30 seconds at each workload. Heart rate was recorded during the last 15 seconds at each workload. Maxi mum oxygen uptake was reported as the highest of the 30 second averaged values. Work was performed on a wheelchair ergometer. A detailed Table 2. Protocols for continuous increasing workloads Paraplegics Quadraplegics Time resistance speed resistance speed 0-1 internal 20 rpm internal 20 rpm 1-2 internal 20 rpm internal 20 rpm 2-3 3.5 kg 20 rpm 3.5 kg 20 rpm 3-4 4.5 kg 20 rpm 4.5 kg 20 rpm 4-5 5.5 kg 20 rpm 4.5 kg 25 rpm 5-6 6.5 kg 20 rpm 4.5 kg 30 rpm 6-7 7.5 kg 20 rpm 4.5 kg 35 rpm 7-8 7.5 kg 25 rpm 4.5 kg 40 rpm 8-9 7.5 kg 30 rpm 4.5 kg 45 rpm 9-10 7.5 kg 35 rpm 4.5 kg 50 rpm 10-11 7.5 kg 40 rpm 11-12 7.5 kg 45 rpm 12-13 7.5 kg 50 rpm 13-14 7.5 kg 55 rpm 14-15 7.5 kg 60 rpm 27 description of this ergometer appears in Appendix A. Data Analysis Difficulty in equating workloads to functional and structural charac teristic of the combined group of paraplegics and quadraplegics, necessi tated the division of the total group. Two sub-groups were produced .para plegics (N=13) and quadraplegics (N=5). Three lines of data were selected for each paraplegic. Each line represented the cardiorespiratory responses to a workload, where the heart rate value was found to be between 65% and 85% of maximum heart rate (maxi mum heart rate = 220 - age). In cases where less than three values of sub-maximum HR were found within this range, the closest HR value(s) outside the limits, were added. Where more than three submaximum HR values fell within the limits, the WL and corresponding cardiorespiratory' values were deleted, using the-following criteria: 1. where two HR responses were found to be very close in numerical values, the lowest was deleted, 2. where the HR response was very close to the lower limit, it was deleted. Appendix B lists complete data sets for all subjects. The three lines of data selected from each subject's data set, are indicated. Each of the three lines were assigned to one of three groupings: LHR: data line corresponding to the lowest HR MHR: data line corresponding to the middle HR HHR: data line corresponding to the highest .HR Multiple regression analysis was carried out on each of the LHR, MHR, and HHR groups. The UBC Computing Center's:BMD P:2R (stepwise multiple regression) and BMD P: 9R ('.'Best" subset multiple regression) programs were employed for the analysis. 28 The data analysis for the quadraplegics was arranged differently. Prediction of maximum HR was not possible at the time of the present study and no research has been conducted to determine the range between resting heart rate and maximum heart rate where prediction of maximum oxygen uptake will be most accurate. Therefore, one WL and corresponding cardiorespira tory responses was selected from each subject's data set. This line of data represented the fourth minute of work for each subject. Stepwise mul tiple regression analysis was performed on this data, UBC Computing Center's P:2R. CHAPTER IV RESULTS AND DISCUSSION Results Twenty male subjects were tested. Two were deleted from the study for reasons noted in Chapter III. The final group of 18 physically disabled individuals were divided into two groups: paraplegics (N=13) and quadraple gics (N=5). Subjects structural characteristics appear in Table 3. Indi vidual physiological responses to the progressive continuous workload proto col, appears in Appendix B. Multiple correlation analysis was performed on the paraplegics' physio logical responses to each of three work intensities, (LHR, MHR, and HHR). Respective mean heart rates were approximately: 70%, 75%, and 80% of the predicted maximum heart rate for the'.group, i.e., 220 - the mean age of the group. Table 4 lists the mean values for the workloads and physiological responses to each work intensity. Correlation coefficients between the 11 predictors and the criterion at the three work intensities, are reported in Table 5. At the lowest work intensity, six variables were found to be significantly correlated to the criterion at the .05 level of significance. Only two variables, WL and V02L correlated significantly to ehe-.criterion, at the .01 level. The MHR intensity produced five variables correlated to MV02L (significant at the' .05rievel). Six variables were significant (.05 level) for the HHR inten sity. Three and four variables were significantly related to the criterion 29 30 Table 3. Subject Characteristics Subjects Age (yrs) Weight (kg) Type of Injury 01 38 79.4 lesion ThlO 02 38 78.0 lesion Th5-6 03 53 95.5 lesion L4-5 04 38 87.5 lesion C6 05 18 56.0 06 22 54.0 lesion C6-7 07 29 70.6 lesion Thl2-Ll 08 25 61.7 lesion C7 09 31 69.0 lesion LI 10 26 70.0 Polio 11 48 91.8 lesion Th5 12 22 62.1 lesion Th4-5 13 52 65.0 lesion ThlO 14 37 66.9 lesion C5 15 33 84.0 16 21 64.0 17 21 60.0 lesion C6-7 18 35 53.0 Polio Paraplegics Mean 34 72.0 Std Dev 11.5 13.Quadraplegics Mean 28.6 66.0 Std Dev 8.3 12.9 31 Table 4. Means and Standard Deviation of Variable Corresponding to LHR, MHR, and HHR LHR MHR HHR Variables Means Std Dev Means Std Dev Means Std Dev WL 421 102 451 122 529 123 HR 131 12 139 12 149 13 V02L 1.25 .303 1.37 .357 1.60 .399 V02ml 18.3 5.80 19.0 5.36 22.6 8.22 VC02L 1.12 .318 1.31 .416 1.60 .484 VC02ml 15.7 4.98 18.2 6.34 22.5 6.78 RQ .89 .09 .94 .11 1.00 .11 Vent L 29.8 7.59 33.4 10.1 43.4 11.8 Table 5. Summary of Correlation coefficients between predictors and criterion MV02 L/min •Work Intensity Wt MBC AG WL HR V02L V02ml VC02L VC02ml RQ Vent L LHR .013 .530 -.513 .816 .600 .873 .718 .714 .586 .065 .535 MHR .013 .530 -.513 .784 .667 .763 .767 .588 .552 -.051 .507 HHR .013 .530 -.513 .834 .525 .873 .718 .714 .586 -.066 .651 Correlation i coefficient required for sig nificance: 0. 553 @ .05; 0.684 @ .01 Quadraplegics .775 -.443 .665 -.515 .493 .774 -.402 .843 -.033 .499 .852 Correlation coefficient required for si gnificance: 0 .805 @ .10; .878 @ .05 ro 33 (.01 level) for the MHR and the HHR intensities, respectively. Complete correlation matrices appear in Appendix C. Multiple regression analysis revealed significant adjusted multiple 2 correlation coefficients (adj R ), between the five best predictors and the criterion. This was observed at all three work intensities. A sum mary of the results of the multiple regression analysis is provided in Table 2 6. At the two higher work intensities an increase in the adj R and a reduced standard error of estimate were observed. Significance of these differences was not determined. Figures 1-3 graphically illustrate the prediction accuracy of the multiple regression equations developed for the three work intensities. A comparison which may better indicate the accuracy of the prediction of MV02L for subjects outside the experimental group, is illustrated in Figures 4-6 (Observed vs Adjusted Press Predicted MV02L). Prediction is made, in turn for each subject with the effects of his data removed from the regression coefficients. Over all work intensities, the deleted press residual error was found to be greater than the residual error resulting from the unadjusted multiple regression equations. The greatest scattering of plots about the line of unity, was observed at the LHR intensity. Both the MHR and HHR equations produced much less scattering and did not appear to differ from each other. Table 7 summarizes the residual errors for each equation (page 47). The use of the UBC Computing Service's P:2R program for stepwise mul tiple regression analysis, revealed significant limitations of the stepping process. The stepping procedure does not always select the best combina tion of variables. Since all other steps are affected by the preceding steps and corresponding variables entered, certain variable combinations 34 Table 6. Summary of multiple regression analysis of paraplegic data Work Inten- Contribution Std Error sity Variables Coefficients to R2 R2(adj) of est. Sign. HHR Wt 0.0261173 0.323542 0.9094 0.1195 0.0002 MBC 0.0035206 0.135327 AG -0.0533642 0.325075 HR -0.0340337 0.150292 VC02ml 0.0428753 0.285904 Intercept 5.56273 MHR AG -0.0264823 0.083831 0.9218 0.1101 0.0001 WL -0.0015175 0.021833 HR -0.0080279 0.015104 VC02L 1.69754 0.202779 RQ -3.42073 0.311303 Intercept 5.84850 LHR Wt 0.023374 0.415916 0.8761 0.1397 0.0007 AG -0.0263043 0.202233 HR -0.0116148 0.031783 VC02ml 0.1083700 0.307697 RQ -2.85780 0.110529 Intercept 3.32762 1.00 1.50 2.00 2.50 3.00 PREDICTED V02 max (l/min) Figure 1. Comparison of MV02L predicted (unadjusted) vs observed MV02L for LHR group 36 MHR OBSERVED vs PREDICTED (unadjusted) 3.00+ 2.50 2.0 0+ 1.50+ 1. 00 OBSERVED V02 max (l/min) 1.00 2.50 1.50 2.00 PREDICTED V02 max (l/min) Figure 2. Comparison of MV02L predicted (unadjusted) vs observed MV02L for MHR group 3.00 37 HHR OBSERVED vs PREDICTED (unadjusted) 3.00+ 2.50 + 2.0 0+ 1.50 + 1.00 OBSERVED V02 max (l/min) 1.00 2. 50 1.50 2.00 PREDICTED V02 max (l/min) Figure J. Comprison of MV02L predicted (unadjusted) vs observed MV02L for HHR group 3.00 38 LHR OBSERVED vs PREDICTED (ADJ. Press) 3.00+ 2.50 + 2.0 0+ 1. 50 + 1.00 OESEFVED VC2 max (l/min) 1.00 1.50 2.00 2.50 PREDICTED V02 max (l/min) Figure k. Comparison of MV02L predicted (adj. press) vs observed MV02L for LHR group 3.00 39 MHR OBSERVED vs PREDICTED (ADJ. Press) 3.00+ OBSERVED V02, max (l/min) 2.50 + 2.0 0+ 1.50 + 1.00 1.00 2. 50 1.50 2.00 PREDICTED V02 max (l/min) Figure 5. Comparison of MV02L predicted (adj. press) vs observed MV02L for MHR group 3.00 HHR OBSERVED vs PREDICTED (ADJ. Press) PREDICTED V02 max (l/min) Figure 6. Comparison of MV02L predicted (adj. press) vs observed MV02L for HHR group 41 are not tested. This problem was noted with the LHR equation. The P:2R program's best five predictors included: Wt, AG, WL, HR, and V02L produced 2 2 an adj R =.8343 as compared to an adj R of .8761 reported in Table 6, a result of the P:9R program. The P:9R program tests all combinations of the variables in varying numbers. The only problem found with this program was the restrictions of the variables which can be submitted to the program for analysis. Vari ables which are highly intercorrelated with each other cannot all be sub mitted. An improper choice to delete one or more variables, may not allow the best subset to be determined. It was found necessary to make several runs with different variable subsets. Discussion Analysis of the correlation coefficients, at three work intensities, suggested increased accuracy in prediction may be associated with the higher work intensities, i.e., approximately 70-75% of predicted maximum heart rate. This result is supported by findings which dictate the protocols for submaximal tests currently applied to able bodied subjects (Davies, 1968; Astrand & Rhyming, 1954; Hermiston & Faulker, 1971). Many variables were found to be significantly correlated to maximum oxygen uptake (MV02L), however, not all appear in the prediction equation. When two or more variables are intercorrelated, the information contained in each is similar. The higher the intercorrelation coefficients, the greater the similarity. The addition of two or more of these intercorrelated vari ables to the prediction equation will have little effect on the multiple 2 correlation coefficient and may decrease the adjusted R value. During the stepping procedure of multiple regression analysis, once one of these 42 intercorrelated variables enters the equation the partial correlation coeffi cients between a remaining variable and the criterion falls, as does the F value to enter. Variables which are not necessarily significantly correlated to MV02L and not intercorrelated with other variables, may enter the equation and 2 contribute optimally to the adjusted R , i.e., MBC, Wt, AG, and RQ. It is well documented that the variables V02L, WL, and HR are signif icantly intercorrelated when workload is varied over a wide range of submaximal workloads (Astrand & Rodahl, 1970). These relationships provide the basis for the extrapolation methods of predicting MV02L for able bodied subjects (Astrand & Rhyming, 1954; Maritz et al., 1962; Margaria et al., 1965; Hermiston & Faulkner, 1971; Davies, 1968). Similar relationships have been reported for arm work by both physically disabled and able bodied sub jects (Wicks et al., 1973; Glaser et al.,:1978a, 1978b). However, a curvi linear relationship between V02L and workload has also been reported (Wicks et al., 1977; V0kac et al., 1975; Stenberg, 1967; Davies & SaNrgeant, 1974). This relationship is suggested to be due to the recruitment of trunk muscles to provide stabilization of the shoulders, allowing the sub jects to exert greater forces against the wheels of the chair (Vokac et al., 1975; Glaser et al., 1978, 1977; Wicks et al., 1977; Engel & Hilderbrandt, 1973). This produces a fall in mechanical efficiency at higher relative workloads. Ventilation, VC02L, and RQ tend to be curvilinearly related to workload, when the workload is varied over a wide range of submaximal workloads (Astrand & Rodahl, 1970; Davis et al., 1976; Wasserman et al., 1973). In theory, lower correlations should be found between these variables and variables which are linearly related to workload. The intercorrelation coefficients observed in this study do not appear to support the above. Very high 43 correlations are observed at all the three work intensities, i.e., VC02 (L or ml) vs Vent, Vent vs V02L. An explanation for these observations relates to the fact that corre lation analysis was performed on physiological responses to the same work intensity for each subject. The intercorrelation coefficients reflect how the relationships between the physiological responses to the particular work intensity vary over the 13 subjects. This differs from the determin ation of how the population's physiological responses are related to each other over a wide range of workloads, i.e., in this study correlation anal ysis is performed on three small segments of the continuum from rest to maximum exertion. Within the limits of each segment, linearity between variables may be found. Body weight was found to be nonsignificantly related to MV02L. This has been reported elsewhere.(Jette et al., 1976). The reverse has also occurred (Jessup et al., 1974; Bonen & Belcastro, 1977; Hermiston & Faulkner, 1971), the dif ference.-being:' related to the nature of the population in each study. In children and in young, healthy, lean subjects there is a good correlation between body weight and MV02L (Astrand, 1952). The distribution of body weights associated with the subjects of the present study, would suggest a heterogeneous sample and corresponding sig nificant relationship between body weight and MV02L. This was not found. The relationship appears to be obscured by the variability in the degree of atrophy of lower limb muscles, the varying degrees of obesity, and the age related deterioration of the oxygen transport system, which is not reflected in the dimensions of an individual (Astrand & Rodahl, 1970). Age was found to be negatively correlated with the criterion. This is a common observation (Astrand & Rodahl, 1970; Hermiston & Faulkner, 1971; 44 Jette et al., 1976). Where children are involved the trend may be reversed (Bonen & Belcastro, 1977). Respiratory quotient did not correlate significantly with the criter ion at any of the three work intensities. Varying results have been reported, regarding the use of submaximum RQ or the change in RQ in both simple regres sion (Rowell et al., 1964; Issekutz & Rodahl, 1961; DeVries & Klafs, 1965; Issekutz et al., 1962; Shephard, 1967; Joseph et al., 1972), and multiple regression equations (Hermiston & Faulkner, 1971; Mastropaolo, 1970). Rowell et al. (1964), concluded the use of submaximum RQ was limited since the level of training is a primary determinant in how RQ changes with V02 during submaximum work. No relationship between RQ and MV02L was reported. The use of submaximal RQ in multiple regression equations has lent support to the usefulness of this variable in prediction of maximum oxygen uptake. Mastropaolo (1970), reported a correlation coefficient of .89 between submaximum RQ and MV02L. Hermiston and Faulkner (1971), did not report a significant correlation between RQ and maximum oxygen uptake. However, in the two equations reported, the change in submaximum RQ alone were found to be important variables in each of the multiple regression equations. In the present study, the correlation coefficients between submaximum RQ and MV02L were nonsignificant. Nevertheless, submaximum RQ was incor porated into two of the three multiple regression equations. In each equa-2 tion, submaximum RQ contributes substantially to the R values. Maximum RQ values for subjects 02, 16, and 17 were noted to be consid erably higher than that normally seen at termination of a maximum work bout. Explanations for this observation may be related-to the observed greater proportion of white (glycolytic) muscle fibres in the muscles of the upper 45 limbs. This greater proportion of white fibres suggests that a greater rate of lactate may be produced by a given muscle mass. Hyperventilation, to blow off CC>2 from the body's bicarbonate buffering stores, occurs in an effort to buffer the lactate produced during exercise. Peak production at termination of the maximum work bout, may be proportionally greater to that of oxygen consumption normally found with leg work. Vokac et al. (1975), compared the physiological responses of able bodied male subjects to arm and leg work. Respiratory quotients were found to be significantly higher for arm work, V02L equal to 1.9 L/min. Respir atory frequency was reported to be higher and tidal volume lower during arm work, for the same pulmonary ventilation rate. Subjects were observed to synchronize breathing with stroking frequency. Blood lactate levels have been reported to be similar during both arm and leg work at maximum effort. Paraplegics reach maximum oxygen up take at a much lower V02L than is found with leg work. However, blood lactate levels are reported to be similar (Vokac et al., 1975). The VC02L should be equivalent to that found during leg work, while oxygen uptake is reported to be about 66% of that found during leg work. Since RQ is a simple ratio of VC02L to V02L, this ratio may theoretically be higher during arm work. Heart rate was found to be significantly correlated to MV02L at all work intensities. The numerical value of the correlation fell with increased work intensities. As noted previously, HR and many other variables are intercorrelated. The effects of this was noted in the small contribution 2 to the R values in each case. Subjects with spinal lesion below the thoracic 6-7 level, showed normal maximum HR values. This supports the observation of other investigators 46 (Knutsson et al., 1973; Wolf & Magora, 1976; Freyschuss & Knutsson, 1969; Freyschuss, 1970; Nilsson et al., 1975). Subject 02 with a lesion at the Th 5-6 level had an observed maximum heart rate of 189. This is very close to the predicted value for his age, i.e., 220 - age. Although the lesion is above the Th 6-7 level, the inter-individual variations with regard to the level of exit of spinal nerves, or the angle of the lesion may have left some nerves intact at that level, may account for this observation. Subjects 11 and 12, lesions Th 5 and Th 4-5, respectively, showed what appeared to be partial loss of the sympathetic stimulation to the heart (observed 161, predicted 172; observed 180, predicted 198, respectively). 2 The variables that contributed to the R values of each equation were similar. The logic determining the use of one variable rather than another appears to be mathematical in nature, rather than based on physiological principles. The P9R program lists many combinations of variables which are only slightly less accurate in the prediction of MV02L, as compared to the "Best" subset. The best equations developed at each intensity compared favourably with similar procedures applied to the able bodied population (Hermiston & Faulkner, 1971; Mastropaolo, 1970; Bonen & Babineau, 1977; Metz & Alexander, 1971; Bonen et al., 1979; Fox, 1973). The size of the sample 2 may cast some doubt on the validity of the R values. However, the adj 2 R values are greater and the standard errors of estimate are less than those reported for normal subjects (Bonen et al., 1977; Fox, 1973; Bell et al., 1974; Hermiston & Faulkner, 1971; Jessup et al., 1974; Mastropaolo, 1970; Jette et al., 1976). Cross validation of the resulting equations was not possible with 47 the sample size available to the study. The use of the deleted press residual gave some indication of the accuracy of prediction that may be found in the population outside the experimental group. A summary of these errors are available in Table 7. It is noted that the LHR intensity did not result in as accurate a prediction as either the MHR and HHR intensities. The absolute mean error was approximately double that found with the equations produced from the higher work intensities. Table 7. Summary of error associated with the prediction of MV02L from prediction equations for LHR, MHR, and HHR Mean Mean absolute Intensity Method Res idual Error CV HHR adj del 0.015 13.2% unad j 0.000 6.77, 5.67, MHR adj del 0.010 11.87, unad j 0.000 6.67, 5.27, LHR adj del 0.020 22.27, unadj 0.000 9.07, 6.57, The results of analysis of the quadraplej ;ic data must be viewed with the limitation of the sample size taken into consideration. The P:9R pro gram was not utilized for the multiple regression analysis. Difficulty was found with the selection of variables which could be eentered into the program. Stepwise multiple correlation analysis was performed and only these results were reported (Table 5). Correlation analysis suggested physiological relationships, i.e., linear relationships between physiological parameters, observed in able bodied as well as paraplegic subjects, may exist within the quadraplegic 48 sample. Significant correlation coefficients between HR and RQ, HR and V02ml, Vent and V02L, Vent and VC02L, were found (significant at .05 level). MBC was of particular interest in this group. Since the respiratory muscles of the abdomen and thoracic cage are innervated by nerves exiting from the spinal cord, loss of or reduced capacity to ventilate the lungs (as'.reflected in MBC), may reflect loss of nervous supply to the heart. In addition, other functional capacities may also be reflected in this measure. The results suggest significant relationships between MBC and RQ, MBC and HR, MBC and VC02L. Although these results must be viewed with the limitations specified, they do point to some very interesting differ ences in the relationships between MBC and certain physiological measures not reported for normals nor paraplegics. For the quadraplegics, the multiple regression equation developed for the fourth minute of the progressive work protocol (Table 8) produced 2 an adjusted R of .9992, with a standard error of estimate of .0058 l/min. The variables contributing to the equation were: Vent, V02ml, and WL. 2 Although the adjusted R is formulated to take into account both sample size and the number of variables used in the prediction, it is felt that the value is inflated. 49 Table 8. Summary of Univariate Statistics for the 4th Minute of Exercise (Quadraplegics) Variable Mean Standard Deviation MBC 130.4 16.1 WL 206.6 17.6 HR 105.0 18.4 V02L 0.606 0.057 V02ml 9.35 1.30 VC 02L 0.531 0.081 VC02ml 8.25 1.4RQ 0.886 0.087 Vent 20.1 5.1CHAPTER V SUMMARY AND CONCLUSIONS Summary The purposes of this investigation were, firstly, to determine which of certain structural and functional characteristics of quadraplegics and paraplegics were significantly related to MV02L. Secondly, to utilize multiple regression analysis to determine and combine the best five predic tors into a multiple regression equation for the prediction of MV02L. Preliminary investigation indicated that the two subgroups, quadra plegics and paraplegics could not be equated on either submaximal workloads or the physiological responses to these. The group of 18 physically disabled subjects were divided into two sub-groups for analysis. Analysis of three work intensities was carried out on the paraplegic data. These corresponded to: 70%, 757», and 807, of the mean maximum heart rate for the group. Increasing numbers of variables were found to be sig nificantly related to MV02L, with increased work intensities. Multiple regression analysis supported these findings. The two higher work inten sities produced more accurate prediction equations than the lowest intensity. The adjusted press absolute mean error, illustrated that the error for sub jects outside the experimental group will average 227o, with the LHR equation. This error was reduced to 13.2% and 11.8% for HHR and MHR, respectively. The analysis of the quadraplegic data was restricted to the 50 51 physiological responses to the WL at the fourth minute of the progressive continuous workload protocol. Due to the small sample size, interpretations and conclusions are not justified. However, some interesting relationships were reported. Conclusions 1. Multiple regression analysis appeared to be a suitable method of develop ing accurate prediction equations for MV02L, in paraplegic subjects. 2. Accuracy in prediction of maximum oxygen uptake in the paraplegic popu lation, is increased with the increase in physiological stress (as re flected in % of maximum heart rate) the subject is subjected to. 3. Due to the limited sample size, no conclusions were made regarding pre diction of MV02L within the quadraplegic population. Recommendations for Further Research The area of prediction of maximum oxygen uptake with the physically disabled population is virtually unexplored. The present study is the first of its kind. Thus the area is wide open. Some specific research is suggested: 1. Determination of a'valid method of estimating fat free weight, is required. Many of the subjects were overweight causing problems with the interpre tation of MV02ml and the correlation coefficients involving body weights. 2. It is recommended that the protocol for workload adjustment be modified to increase the duration to 2 minutes at each workload. 3. Validation of the prediction equations developed for the paraplegics is necessary. 4. The results of the analysis of the quadraplegic data has indicated promis ing results. This study should be continued with greater numbers of 52 quadraplegics, so that conclusions are possible. 5. Differences, if any should be determined for the prediction of maximum oxygen uptake with individuals who vary in the type of disability, i.e., polio, bone diseases, amputees, etc. 6. The above should be conducted for physically disabled females. 53 REFERENCES Astrand, P. 0. Experimental studies of physical work capacity in relation to sex and age. Munkgaard, Copenhagen (1952). Astrand, P. 0., & Rodahl, K. Textbook of work physiology. McGraw Hill Book Company, New York, St. Louis, San Francisco, London, Sydney, Toronto, Mexico, Panama, (1970). Astrand, P. 0., & Rhyming, I. A nomogram for calculation of aerobic capacity (physical fitness) from pulse rate during submaximal work. J. Appl. Physiol., 1954, 7, 218-221. Bar-or, 0., & Zwiren, L. D. Maximal oxygen consumption test during arm exercise—reliability and validity. J. Appl. Physiol., 1975, 38, 424-426. Barr, S. A., & Glaser, R. M. Physiological responses to wheelchair and bicycle activity. Fed. Proc., 19 , 36, 850. Bell, A. C., & Hinson, N. N. Prediction of maximal oxygen uptake in women twenty to forty years of age. J. Sport Med. Phys. Fit., 1974, 14, 208-212. Bevegard, S., Freyschuss, U., & Strandell, T. Circulatory adaptations to arm and leg exercise in supine and sitting position. J. Appl. Physiol., 1966, 21, 37-46. Bonen, A., & Belcastro, A. N. A rapid method for estimating maximal oxygen uptake for college age hockey players. Can. J. Appl. Sport Sci., 1977, 28, 27-33. Brubake, C. E., McLaurin, C. A., Gibson, J. D. , & Soos, T. Effect of speed and load on wheelchair propulsion. Med. Sci. Sports, 1979, 1_1, 112. Cameron, B. J., Ward, G. R., & Wicks, J. R. Relationship of type of train ing to Max 02 uptake and upper limb strength in male paraplegic athletes. Med. Sci. Sport, 1977, 9(1), 58. Cole, T. M., Kottke, F. J., Olson, M., Stradal, L., & Niederloh, B. S. Alterations of cardiovascular control in high spinal myelomalacia. Arch. Phys. Med. Rehabil., 1967, 48, 359-368. Corbett, J. L., Frankel, H. L., & Harris, P. J. Cardio-responses to tilting in.tetraplegic man. J. Physiol., 1971, 215, 411-431. Cunningham, D. J. C., Guttmann, L., Whitteridge, D., & Wyndham, C. H. Cardio vascular responses to bladder distension in paraplegic patients. J. Physiol., 1953, 121, 581-592. 54 Davies, C. T. M. Submaximal test for estimating maximum oxygen uptake. Commentary in: Proc. Int. Sympo. on Physical Activity and Cardio vascular Health. Canadian Med. Ass. J., 1968, 96_, 743-744. Davies, C. T. M., & Sargeant, A. J. Physiological responses to standard ized arm work. Ergonomics, 1974, 17, 41-49. Davis, J. A., Vodak, P., Wilmore, J. H., & Kurtz, P. Anaerobic threshold and maximal aerobic power for three modes of exercise. J. Appl. Physiol., 1976, 41, 544-550. DeVries, H. A., & Klafs, C. E. Prediction of maximal 02 intake from sub-maximal tests. J. Sports Med., 1965, 5, 207-214. Engel, P., & Hilderbrandt, G. Long-term spiroergometric studies of para plegics during the clinical period of rehabilitation. Paraplegia, '.. 1973, 11, 105-110. Ernes, C. Physical work capacity of wheelchair athletes. Res. Quart., 1977, 48, 209-212. Falls, H., Ismail, A. H., & MacLeod, D. F. Estimation of maximal oxygen uptake in adults from AAHPER youth fitness test items. Res. Quart., 1966, 37, 192-201. Flandrois, R., & La Cour, J. R. The prediction of maximal oxygen uptake in acute moderate hypoxia. Int. Z. Angew. Physiol., 1971, 29, 306-313. Fox, E. L. A simple, accurate technique for predicting maximal aerobic power in man. J. Appl. Physiol., 1973, 35, 914-916. Freyschuss, U. Comparison between arm work and leg work in exercise in women and men. Sc. J. Clinic. Lab. Invest., 1975, 3_5, 795-800. Freyschuss, U., & Knutsson, E. Cardiovascular control in man with trans verse cervical cord lesions. Life Sci. , 1969, 8, 421-424. Freyschuss, U. Cardiovascular adjustment to somatomotor activation. Acta. Physiol. Scand., 1970, Suppl. 342. Glaser, R. M., Laubach, L. L., Foley, D. M., Barr, S. A., Suryaprasad, A. G., & Burk, R. D. An interval training program for wheelchair users. Med. Sci. Sport, 1978, 10, 54. Glaser, R. M., Young, R. C., & Suryaprasad, A. G. Reducing energy cost and pulmonary stresses during wheelchair activity. Fed. Proc., 1977, 36, 580. Glaser, R. M., Stephen, B. P., Lloyd., L. L., & Agaram, S. G. A cardio pulmonary fitness test utilizing the wheelchair ergometer. Fed. Proc.,. 1978, 37, 429. 55 Glaser, R. M., Foley, F. M., Lloyd, L. L., Sawka, M. N., & Agaram, S. G. An exercise test to evaluate fitness for wheelchair activity. Paraplegia, 1979, 16, 341-349. Glaser, R. M., Sawka, M. N., Laubach, L. L., Suryaprasad, A. G., & Al-Samkari, 0. Wheelchair vs bicycle ergometry: Cardiorespiratory responses. Med. Sciv Sport, 1979, VL, 112. Glassford, R. G. , Baycroft, G. H. Y.., Sedgwick, A. W. , &MacNab, R. B. J. Comparison of maximal oxygen uptake values determined by prediction and actual methods. J. Appl. Physiol., 1965, 2_0, 509-51.3 Guttman, L. Rehabilitation after injuries to spinal cord and cauda eguina. Brit. J. Phys. Med., 1946, 9, 162-171. Guttman, L. In CIBA Foundation Symposium Peripheral Circulation in Man. Edited by Wolstenholme, G., & Freeman, J., p. 191. London: Churchill, 1954. Hermiston, R. T., & Faulkner, J. A. Prediction of maximal oxygen uptake by a stepwise regression technique. J. Appl. Physiol., 1971, 30, 833-837. Heigenhauser, G. H., Ruff, G. L., Miller, B., & Faulkner, J. A. Cardio vascular response of paraplegics during graded arm ergometry. Med. Sci. Sport, 1977, 8, 68. Hjeltnes, N. Oxygen uptake and cardiac output in graded arm exercise in paraplegics with low level spinal lesions. Scand. J. Rehab. Med., 1977, 9, 107-113. Issekutz, B., Birkhead, N. C., & Rodahl, K. Use of respiratory quotient in assessment of aerobic work capacity. J. Appl. Physiol;, 1962, 17, 47-50. Issekutz, B., & Rodahl, K. Respiratory quotient during exercise. J. Appl. Physiol., 1961, 16, 606-610. Jessup, G. T., Tolson, H., & Terry, J. W. Prediction of maximum oxygen uptake by stepwise regression technique. Am. J. Phys. Med., 1974, 53, 200-207. Jette, M. , Campbell, J., Mongeon, J., & Routhier, R. The Canadian home fitness test as a predictor of aerobic capacity. Can. Med. Assoc. J., 1976, 114, 680-683. Jocheim, K., & Strohkend, H. The value of particular sports of the wheelchair disabled in maintaining health of the paraplegic. Paraplegia, 1973, 11, 173-178. 56 Johnson, R. H., Smith, A. C, & Spalding, J. M. K. Blood pressure response to standing and to Valsalva manoeuvre: Independence of the two mechanisms in neurological disease including cervical cord lesions. Clin. Sci., 1969, 36, 77-86. Jonason, P. H. A. Discussion of treatment of persons with traumatic para plegia. Proc. Roy. Soc. Med., 1947, 40, 188. Knutsson, E., Lewenhaupt-Olsson, E., & Thorson, M. Physical work capacity and physical conditioning in paraplegic patients. Paraplegia, 1973, 11, 205. Kurnick, N. B. Autonomic hyperreflexia and its control in patients with spinal cord lesions. Ann. Intern. Med., 1956, 44, 678-686. Maritz, J. S., Morrison, J. F., Peter, J., Strydom, N. B., & Wyndham, C. H. A practical method of estimating an individuals maximum oxygen uptake. Ergonomics, 1962, 4, 97-122. Margaria, R., Aghemo, P., & Rovelli, E. Indirect determination of maximal oxygen consumption in man. J. Appl. Physiol., 1965, 2j3, 1070-1073. Marincek, C. R.'?T., & Vojko, V. Arm eyelo-ergometry and kinetics of oxygen consumption in paraplegics. Paraplegia, 1978, 15, 178-185. Mastropoalo, J. A. Prediction of maximum oxygen consumption in middle-aged men by multiple regression. Med. Sci. Sport, 1970, 2, 124-127. Metz, K. F., & Alexander, J. F. Estimation of maximal oxygen uptake predic tion in young and middle-aged males. J. Sport Med. Phys. Fit., 1969, 9, 17-22. Nakamura, M. D. Working.ability of the paraplegics. Paraplegia, 1973, 11, 182-193. Nilsson, S., Staff, P. H., & Pruett, E. D. R. Physical work capacity and the effect of training on subjects with long standing paraplegia. Scand. J. Rehab. Med., 1975, ]_, 51-56. Pollock, L. J., Boshes, B., Chor, H., Finkelman, I., Arieff, A. J., & Brown, M. Defects in regulatory mechanisms of autonomic function in injuries to spinal cord. J. Neurophysiol.', 1951, 14, 85. Rowell, L. B., Taylor, H. L., & Wang, Y. Limitations to prediction of maximal oxygen uptake. J. Appl. Physiol. , 1964, 1_9, 919-927. Shephard, R. J. Endurance fitness. University of Toronto Press, Toronto and Buffalo, 1977. Shephard, R. J. The prediction of maximum oxygen consumption using new progressive step test. Ergonomics, 1967, 10, 1-15. 57 Verma, S. S., Sen Gupta, J., & Malhotra, M. S. Prediction of maximum aerobic power in man. Eur. J. of Appl. Physiol., 1977, 36, 215-222. Voigt, E. D., & Bohn, D. Metabolism and pulse rate in physically handicapped when propelling a wheelchair up an incline. Scand. J. Rehab. Med., 1969, I, 101-106. Vokac, Z., Bell, H., Bautz-Holter, E., & Rodahl, K. Oxygen uptake/heart rate relationship in leg and arm exercise, sitting and standing. J. Appl. Physiol., 1975, 39, 54-59. Wasserman, K., Whipp, B. J., Koyal, S. N., & Beaver, W. L. Anaerobic threshold and respiratory exchange during exercise. J. Appl. Physiol., 1973, 35, 236-243. Whitteridge, D. Cardiovascular disturbances in paraplegics (Honyman Gillespie lecture). Edinburgh Med. J. , 1954, 6JL, 1-6. Wicks, J. R., Lymburner, K., Dinsdale, S. M., & Jones, N. L. The use of multistage exercise testing with wheelchair ergometry and armccrank-ing in subjects with spinal cord lesions. Paraplegia, 1977, 11, 252-261. Wicks, J. R., Head, E., Oldridge, N. B., Cameron, B., & Jones, N. L. Maximum oxygen uptake of wheelchair athletes competing at the 1976 Olympiad for the physically disabled. Med. Sci. Sport, 1977, 9, 58. Wolf, E., & Magora, A. Orthostatic and ergometric evaluation of cord-injured patients. Scand. J. Rehab. Med., 1976, 8, 93-96. Wyndham, C. M., Stydom, N. B., Maritz, J. S., Morries, J. P., Peter, J., & Potgieter, Z. U. Maximum oxygen intake and maximum heart rate during strenuous work. J. Appl. Physiol., 1959, 14, 927-936. Zwiren, L. D., & Bar-or, 0. Responses to exercise of paraplegics who differ in conditioning level. Med. Sci. Sport, 1975, _7, 94-98. 58 APPENDIX A The Wheelchair Ergometer 59 Wheelchair Ergometer The ergometer was designed so that each subject could utilize his own wheelchair. It consisted of three steel rollers 3 inches in diameter and 30 inches in length. The center roller was fixed in position while the outer two were suspended by springs. The two outer rollers served the purpose of accepting much of the weight of the subject and chair and to stabilize the chair against forward and backwards rocking during the stroking. The loading system of a Monarch bike was adapted and utilized to apply resistance to the middle roller. The friction strap which normally was placed around the flywheel of the bike was placed around the center roller of the wheelchair ergometer. Resistance was adjusted and read from the pendulum indicator as it would be with the normal bike ergometer. However, internal resistance had to be taken into account. Internal resistance is that within the roller system and the inden tation of the pneumatic tire caused by the rollers. This value is varied on both tire pressure and weight of the subject. This resistance was mea sured at a constant tire pressure of 50 pounds per square inch. The force required to cause movement of the wheelchair wheels was determined for a variety of weights placed in the wheelchair. Increasing weight was hung vertically at a right angle from the outer perimeter of one wheel until movement was initiated. Internal resistance was plotted against weight in the wheelchair. This produced a curvilinear line. For each subject internal resistance was read from the graph. Exercise resistances were achieved by adding the necessary resistance via the load adjuster, to the internal resistance. During exercise the wheelchair had to be further secured with small gauge chains to prevent rocking as the subjects stroked at higher workloads. The chains were secured around the front casters of the wheelchair and fastened to the wooden frame of the ergometer. -3- CM o 0"\ CN CN CO VO CM 00 OJ CM CM CM CM TH «H 63 APPENDIX B Individual Cardiorespiratory Responses to the Workload Protocol 64 Sub Wt MBC AG WL HR V02L V02ml VC02L VC02ml RQ Vent L MV02 L 01 79.3 168 38 110 189 208 231 228 280 352 *368 *409 *423 : 564 568 511 02 78.0 180 38 106 157 158 215 *269 *360 *420 595 579 600 03 95.5 250 53 138 117 106 226 089 0.491 06. 19 0.186 092 0. 724 09. 12 0.406 090 0. 638 08. 03 0.399 096 0. 788 09. 93 0.510 098 0. 757 09. 54 0.534 102 0. 826 10. 40 0.628 110 0. 961 12. 10 0.751 122 1. 135 14. 30 0.969 130 " 1. 278 16. 10 1.143 145 1. 532 19. 30 1.461 165 1. 858 23. 40 2.088 175 1. 898 23. 90 2.390 185 1. 842 23. 20 2.342 088 0. 440 05. 64 0.250 092 0. 499 06. 40 0.300 095 0. 548 07. 03 0.356 106 0. 693 08. 89 0.497 118 0. 771 09. 89 0.555 132 0. 970 12. 44 0.797 142 1. 009 12. 94 0.906 159 1. 225 15. 70 1.169 175 1. 440 18. 46 1.580 185 1. 774 22. 75 2.338 084 1. 030 10. 79 0.858 090 0. 945 09. 90 0.783 096 0.864 09. 05 0.640 105 1. 024 10. 72 0.784 03. 60 0.58 Oil. 10 1.912 05. 12 0.56 014. 70 05. 03 0.63 012. 80 06. 43 0.65 015. 70 06. 72 0.71 014. 90 07. 91 0.77 016. 90 09. 46 0.78 019. 40 12. 20 0.86 025. 25 14. 40 0.90 019. 00 18. 40 0.96 038. 00 26. 30 1.13 080. 00 30. 10 1.28 076. 00 29. 65 1.29 085. 00 03.20 0.57 010.40 1. 774 03.84 0.60 011.14 04.56 0.65 012.80 06.37 0.72 016.70 07.12 0.72 017.60 10.22 0.82 023.33 11.62 0.90 025.38 14.99 0.96 032.94 20.26 1.10 046.48 29.97 1.32 084.24 08.98 0.82 024.90 2. 158 08.20 0.76 023.80 06.70 0.75 021.60 08.21 0.76 024.90 ^Indicates lines of data selected for regression analysis 65 Sub Wt MBC AG WL HR V02L V02ml VC02L VC02ml RQ Vent L MV02 L 04 87.5 129 38 05 56.0 186 18 188 100 1. 290 13. 51 0.987 10.34 0.77 029. 71 275 125 1. 226 12. 84 1.012 10.60 0.82 029. 80 *318 125 1. 220 12. 78 1.003 10.50 0.82 028. 20 *340 130 1. 369 14.34 1.203 12.60 0.84 032. 15 *445 135 1. 662 17. 40 1.413 14.80 0.86 041. 50 511 140 1. 583 16. 60 1.375 14.40 0.86 042. 40 530 150 1. 891 19; 80 1.671 17.50 0.88 050. 40 500 160 2. 063 21. 60 1.900 19.90 0.92 056. 90 368 165 2. 158 22. 60 1.995 •20.90 0.92 061. 30 127 101 0. 630 07. 20 0.595 06.80 0.94 023. 60 127 102 0. 800 09. 14 0.744 08.50 0.90 026. 60 148 102 0. 656 07. 50 0.595 06.80 0.90 024.00 191 106 0.621 07. 10 0.586 06.70 0.94 021. 80 233 121 0. 770 08. 80 0.674 07.70 0.86 023. 80 292 130 1. 085 12. 40 1.083 12.38 1.01 037. 74 275 142 1. 251 14.30 1.444 16.50 1.16 046. 80 398 142 1. 108 12. 67 1.278 14.60 1.16 039. 70 072 111 0. 558 09. 97 0.402 07.19 0.72 012.90 105 116 0.611 10. 91 0.470 08.40 0.77 014. 90 148 125 0. 683 12. 20 0.538 09.60 0.78 016. 40 191 135 0.796 14. 22 0.633 11.30 0.80 019. 00 233 139 0. 900 16. 07 0.706 12.60 0.78 020. 03 292 143 0. 963 17. 20 0.773 13.80 0.80 022. 50 *350 145 1. 114 19. 90 0.907 16.20 0.82 024.80 *413 153 1. 170 20. 90 1.002 17.90 0.86 027. 00 *477 162 1. 428 25. 50 1.254 22.40 0.88 033. 40 556' 172 1. 579 28. 20 1.602 28.60 1.01 040. 70 636 181 1. 725 30. 80 1.860 33.21 1.08 048. 00 715 192 2. 005 35. 80 2.261 40.37 1.13 060. 00 795 192 2. 110 37. 69 2.486 44.40 1.18 068. 20 874 202 2. 285 40. 80 2.620 46.80 1.15 074.30 66 Sub Wt MBC AG WL HR V02L V02ml VC02L VC02ml RQ Vent L MV02 L 06 54.0 138 22 07 70.6 125 29 086 082 0. 442 08.18 0.380 07.04 0.86 010.10 0.792 081 087 0. 511 09.46 0.441 08.16 0.85 012.00 144 094 0. 513 09.50 0.454 08.40 0.89 012.80 233 099 0. 520 09.63 0.455 08.43 0.88 013.30 289 099 0. 759 14.05 0.657 12.16 0.88 018.00 326 104 0. 792 14.67 0.772 14.30 0.97 021.70 268 105 0. 750 13.90 0.801 14.84 1.11 034.40 091 090 0. 395 05.60 0.282 04.00 0.70 008.76 1.673 115 086 0. 504 07.14 0.266 05.18 0.72 Oil.12 152 097 0. 443 06.27 0.336 04.76 0.76 010.10 209 098 0.518 07.34 0.394 05.58 0.76 011.40 262 109 0. 591 08.37 0.459 06.50 0.78 013.80 284 115 0. 788 11.17 0.641 09.08 0.82 016.60 366 130 0. 870 12.33 0.768 10.88 0.89 020.40 443 132 1. Oil 14.23 1.005 14.23 1.00 026.10 504 161 1. 334 18.89 1.461 20. 70 1.09 039.60 598 175 1. 358 19.23 1.627 23.40 1.20 047.80 602 180 1. 553 22.14 1.920 27.20 1.22 063.40 738 181 1. 673 23.70 2.066 29.26 1.24 068.70 08 61.7 106 25 080 120 0.750 12.15 0.648 10.50 0.86 024.00 1.185 099 130 0.580 09.40 0.600 09.72 1.03 023.90 211 134 0.586 09.50 0.584 09.46 0.99 024.80 204 135 0.654 10.60 0.645 10.45 0.99 027.30 225 136 0.827 13.40 0.823 13.34 0.99 032.60 239 140 0.833 13.50 0.913 14.80 1.09 037.10 286 140 1.012 16.40 1.174 19.03 1.16 048.70 264 139 1.185 19.20 1.252 20.30 1.25 048.30 67 Sub Wt MBC AG WL HR V02L V02ml VC02L VC02ml RQ Vent L MVO2 L 09 69.0 206 31 10 70.0 243 26 154 095 0.545 06.90 0.321 04.06 0.60 011.30 2.280 175 101 0.664 08.40 0.389 04.92 0.68 012.40 218 101 0.742 09.39 0.465 05.89 0.62 014.80 267 111 0.890 11.26 0.590 07.47 0.66 017.90 353 118 0.059 13.40 0.717 09.08 0.68 021.40 438 131 1.270 16.07 0.940 11.90 0.74 027.80 *459 133 1.480 18.70 1.160 14.70 0.79 032.22 •-536 140 1.550 19.60 1.340 16.93 0.86 037.20 *592 146 1.720 21.80 1.580 20.00 0.92 044.90 727 160 1.710 21.60 1.596 20.20 0.94 045.60 636 167 1.960 24.80 1.830 23.20 0.94 055.60 693 176 2.280 28.80 2.430 30.78 1.08 077.50 149 085 0.728 10.40 0. 495 07.04 0.68 012.60 2.849 254 101 0.927 13.24 0. 666 09.52 0.72 018.10 300 101 0.987 14.10 0. 802 11.46 0.82 020.20 387 107 1.078 15.40 0. 854 12.20 0.80 021.10 417 113 1.211 17.30 1. 029 14.70 0.84 025.00 427 117 1.246 17.80 1. 106 15.80 0.89 026.80 481 121 1.435 20.50 1. 274 18.20 0.89 030.20 *527 135 1.582 22.60 1. 484 21.20 0.94 034.60 *573 154 1.834 26.20 1. 834 26.20 1.00 041.60 -670 165 2.219 31.70 2. 380 34.00 1.08 053.50 680 170 2.303 32.90 2. 555 36.50 1.11 058.50 793 180 2.583 36.90 3. 045 43.50 1.18 073.80 772 185 2.849 40.70 3. 346 47.80 1.18 083.10 68 Sub Wt MBC AG WL HR V02L V02ml VC02L VC02ml RQ Vent L MV02 L 11 91.8 106 48 136 088 0.863 09.40 0. 780 08. 50 0.90 021. 40 1.970 150 087 0.964 10.50 0.815 08. 88 0.85 022. 80 138 090 0.861 09.38 0.688 07. 50 0.89 018. 50 203 089 0.863 09.40 0.732 07. 97 0.85 018. 40 280 101 0.909 09.90 0.799 08. 70 0.88 020. 50 *356 111 0.991 10.80 0.909 09. 90 0.92 023. 30 *532 132 1.320 14.40 1.240 13. 50 0.94 031. 10 --404 148 1.560 17.00 1.625 17. 70 1.04 040. 80 400 161 1.970 21.50 2.313 25. 20 1.17 072. 00 12 62.1 198 22 13 65.0 156 52 139 088 0.766 12. 34 0.545 08.78 0.72 014.90 2.640 189 095 0.900 14.50 0.621 10.00 0.69 016.70 197 095 0.807 13. 00 0.602 09.70 0.75 015.70 226 098 0.851 13. 70 0.658 10.60 0.78 016.80 292 106 1.049 16. 90 0.795 12.80 0.76 017.77 388 116 1.186 19. 10 0.919 14.80 0.78 022.00 392 125 1.292 20. 80 1.059 17.06 0.82 024.90 434 129 1.534 24. 70 1.298 20.90 0.85 029.70 523 132 1.590 25. 60 1.453 23.40 0.92 033.50 *550 149 . 1.652 26. 60 1.528 24.60 0.94 035.00 *654 160 1.906 30. 70 1.913 30.80 1.02 045.00 *718 166 2.322 37. 40 2.496 40.20 1.08 061.10 757 179 2.360 38. 00 2.875 46.30 1.24 081.00 782 180 2.640 42. 50 3.173 51.10 1.22 092.00 089 087 0.748 11. 50 0.500 07.70 0.66 014.90 1.580 156 089 0.693 10. 60 0.533 08.20 0.76 013.90 160 093 0.783 12. 50 0.670 10.30 0.76 019.40 203 094 0.741 11. 40 0.658 10.12 0.89 018.40 290 102 1.060 17. 46 1.060 16.31 0.93 026.00 339 110 1.222 18. 80 1.313 20.20 1.09 034.20 *368 116 1.262 19. 40 1.378 21.20 1.09 036.20 69 Sub Wt MBC AG WL HR V02L V02ml VC02L VC02ml RQ Vent L MV02 L 14 66.9 150 37 15 84.0 150 33 16 64.0 129 21 *348 130 1. 398 21. 50 1.684 25. 90 1.21 045. 90 *409 137 1. 436 22. 10 1.801 27. 70 1.26 055. 50 438 145 1. 580 24. 30 1.944 29. 09 1.23 062. 00 089 068 0. 278 05. 76 , 0.268 04.00 0.70 013. 00 092 073 0. 506 07. 57 0.332 04. 96 0.66 014. 00 164 078 0. 530 07. 92 0.389 05. 82 0.74 016. 40 241 083 0. 656 09. 80 0.476 07. 42 0.76 020. 00 227 085 0. 763 11. 40 0.594 08. 88 0.78 023. 80 262 091 0. 727 10. 87 0.605 09. 05 0.83 025. 60 277 094 0. 777 11. 61 0.726 10. 86 0.94 031. 40 177 095 1. 024 15. 31 1.084 16. 20 1.06 050.80 143 086 0. 659 07. 84 0.512 06. 10 0.79 014. 90 124 083 0. 659 07. 84 0.521 06. 20 0.80 015. 40 175 092 0. 692 08.24 0.554 06. 60 0.81 016. 20 224 094 0. 756 09. 00 0.618 07. 36 0.83 018. 00 267 098 0. 916 10. 90 0.753 08. 96 0.83 021. 00 347 105 0.949 11. 30 0.774 09. 22 0.82 021. 10 377 112 1. 151 13. 70 0.943 11. 22 0.82 025. 00 515 118 1. 235 14.70 1.075 12. 80 0.88 029. 00 561 120 1. 445 17. 20 1.302 15. 50 0.90 035. 00 *582 130 1. 579 18. 80 1.478 .17. 60 0.94 039. 00 -592 133 1. 730 20. 60 1.630 19. 40 0.95 044.50 -704 145 1. 982 23. 60 1.966 23. 40 1.00 057. 00 861 168 2. 453 29. 20 2.755 32. 80 1.14 090. 00 850 180 2. 612 31. 10 3.167 37. 70 1.23 111. 50 085 080 0.387 06. 04 0.336 05. 25 0.87 010. 70 127 084 0. 454 07. 09 0.371 05. 80 0.82 012. 10 148 093 0. 561 08. 76 0.461 07. 20 0.82 014.30 191 102 0.531 08. 30 0.420 06. 56 0.79 013. 50 1.024 70 Sub Wt MBC." AG WL HR V02L V02ml VC02L VC02ml RQ Vent L MV02 L 17 60.0 129 21 18 53.0 134 35 233 118 0.730 11. 40 0.580 09. 07 0.79 017. 40 291 125 0.890 13. 90 0. 705 11. 02 0.79 020. 60 350 130 1.152 18. 00 1. 004 15. 69 0.87 027. 60 413 131 1.197 18. 70 1. 119 17. 42 0.94 031. 10 *477 144 1.264 19. 74 1. 196 18. 69 0.94 031. 90 -557 150 1.330 20. 80 1. 307 20. 42 0.98 035. 30 *636 165 1.754 27. 40 1. 843 28. 80 1.05 051. 00 716 182 1.933 30. 20 2. 500 39. 06 1.30 077. 10 795 202 2.195 34.30 3. 268 51. 06 1.50 124.00 072 075 0.332 05. 54 0. 245 04.08 0.74 020. 21 106 083 0.416 06. 93 0. 322 05. 36 0.88 014. 16 149 091 0.494 08. 23 0. 681 05. 99 0.78 016. 89 191 100 0.580 09. 66 0. 494 08. 24 0.86 017. 64 233 109 0.847 14.12 0. 868 14.47 1.03 027. 50 291 114 0.846 14. 10 0. 973 16. 21 1.16 033. 46 349 110 0.742 12. 36 1. 080 18. 00 1.46 028. 40 087 098 0.143 02. 70 0. 099 01. 87 0.69 005. 80 087 098 0.378 07. 14 0.251 04. 73 0.66 Oil. 60 148 107 0.514 09. 70 0. 343 06. 48 0.67 012. 60 191 113 0.659 12. 43 0. 463 08. 75 0.70 015. 72 *233 118 0.668 12. 61 0. 531 10. 02 0.80 016. 70 *291 131 0.960 18. 11 0.814 15. 36 0.85 025. 22 *350 135 1.090 20.56 1. 00 18. 86 0.92 030.00 413 170 1.510 28. 49 1. 535 28. 97 1.02 045. 50 477 175 1.600 30. 18 1. 866 35. 20 1.16 059. 12 557 180 1.780 33. 58 2. 120 40. 01 1.19 070. 86 636 180 1.716 32. 38 1. 913 35. 90 1.11 068. 20 APPENDIX C Correlation Matrices Correlation Matrix: LHR Group 72 123456 789 Wt MBC AG WL HR V02L V02ml VC02L VC02ml 1 1.000 2 0.115 1.000 3 0.632 -0.023 1. 000 4 0.218 0.439 -0. 220 1.000 5 -0.411 0.432 -0. 683 0.542 6 0.119 0.620 -0. 104 0.926 7 -0.517 0.205 -0. 560 0.602 8 0.081 0.456 -0. 017 0.889 9 -0.455 0.301 -0. 332 0.640 10 -0.069 -0.207 o. 187 0.372 11 0.187 0.503 0. 167 0.825 12 0.013 0.530 -0. 513 0.826 13 -0.656 0.289 -0. 796 0.409 1.000 0.528 1.000 0.740 0.624 1.000 0.405 0.947 0.660 1.000 0.559 0.737 0.864 0.834 1. 000 -0.068 0.370 0.425 0.642 0. 660 0.393 0.914 0.563 0.932 0. 695 0.600 0.784 0.571 0.673 0. 569 0.749 0.445 0.763 0.406 0. 721 10 11 12 13 RQ Vent L MV02L MV02ml 10 1.000 11 0.520s 1.000 12 0.065 0.535 1.000 13 0.106 0.232 0.713 1.000 73 Correlation Matrix: MHR Group 1 2 3 4 5 6 78 9 Wt MBC AG MV02L MV02ml WL HR V02L V02ml 1 1.000 2 0.115 1.000 3 0.632 -0.023 1. 000 4 0.013 0.530 -0. 513 1.000 5 -0.656 0.290 -0. 796 0.713 6 -0.028 0.195 -0. 588 0.784 7 -0.152 0.217 -0.576 0.667 8 0.276 0.375 -0. 102 0.763 9 -0.308 0.314 -0. 471 0.767 10 0.212 0.216 0. 005 0.588 11 -0.249 0.191 -0. 283 0.552 12 -0.017 -0.269 0. 179 -0.051 13 0.169 0.141 0. 055 0.507 1.000 0.526 1.000 0.630 0.713 1.000 0.332 0.801 0.658 1. 000 0.731 0.783 0.784 0. 809 1. 000 0.259 0.706 0.613 0. 950 0. 816 0.565 0.688 0.695 0. 790 0. 957 -0.036 0.279 0.282 0. 469 0. 506 0.211 0.634 0.526 0. 858 0. 730 10 11 12 13 VC02L VC02ml RQ Vent L 10 1.000 11 0.880 1.000 12 0.714 0.728 1.000 13 0.924 0.811 0.700 1.000 74 Correlation Matrix: HHR Group 1 2 3 4 5 6 7 8 9 wt • MBC AG MV02L MV02ml WL HR V02L V02ml 1 1.000 2 0.115 1.000 3 0.632 -0.023 1.000 4 0.013 0.530 -0.513 1.000 5 -0.656 0.289 -0.796 0.713 1.000 6 0.034 0.161 -0.528 0.834 0.537 1.000 7 -0.48,4 0.174 -0.878 0.525 0.716 0.603 1.000 8 0.007 0.481 -0.384 0.873 0.602 0.860 0.558 1.000 9 -0.525 0.316 -0.659 0.718 0.885 0.690 0.750 0.827 1.000 10 -0.118 0.328 -0.347 0.714 0.564 0.809 0.585 0.924 0.854 11 -0.506 0.219 -0.540 0.586 0.765 0.656 0.708 0.791 0.962 12 -0.284 -0.915 0.001 -0.066 0.100 0.244 0.286 0.315 0.423 13 -0.083 0.490 -0.244 0.651 0.497 0.539 0.460 0.694 0.609 10 11 12 13 VC02L VC02ml RQ Vent L 10 1.000 11 0.902 1.000 12 0.610 0.649 1.000 13 0.743 0.647 0.441 1.000 75 Correlation Matrix: Quadraplegic Group Wt MBC AG WL HR V02L V02ml VC02L VC02ml 1 1.000 2 0.012 1.000 3 0.823 0.375 1.000 4 -0.596 0.398 -0.224 1.000 5 -0.025 -0.978 -0.310 -0.247 1.000 6 0.458 -0.225 0.610 -0.459 0.261 1.000 7 -0.860 -0.197 -0.596 0.404 0.236 0.055 1.000 8 0.427 -0.854 0.160 -0.524 0.873 0.596 -0.091 1.000 9 -0.629 -0.729 -0.656 0.173 0.768 0.079 0.792 0.433 1.000 10 0.185 -0.906 -0.239 -0.293 0.912 0.035 -0.126 0.822 0.506 11 0.384 -0.648 0.338 -0.500 0.682 0.881 0.100 0.892 0.397 12 0.775 -0.443 0.665 -0.515 0.493 0.774 -0.402 0.843 -0.033 13 -0.218 -0.711 -0.417 0.049 0.807 0.534 0.590 0.705 0.843 10 11 12 13 RQ Vent L MV02L MV02ml 10 1.000 11 0.489 1.000 12 0.499 0.852 1.000 13 0.524 0.768 0.448 1.000 

Cite

Citation Scheme:

    

Usage Statistics

Country Views Downloads
China 25 18
United States 7 0
Germany 3 5
United Kingdom 1 1
City Views Downloads
Beijing 19 0
Shenzhen 6 18
Ashburn 3 0
Unknown 3 8
Mountain View 2 0
Atlanta 2 0
Lisburn 1 1

{[{ mDataHeader[type] }]} {[{ month[type] }]} {[{ tData[type] }]}
Download Stats

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.831.1-0077394/manifest

Comment

Related Items