UBC Faculty Research and Publications

The feasibility of assessing swallowing physiology following prolonged intubation after cardiovascular… Skoretz, Stacey A; Yau, Terrence M; Granton, John T; Martino, Rosemary Nov 21, 2017

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

Item Metadata


52383-40814_2017_Article_199.pdf [ 645.51kB ]
JSON: 52383-1.0360658.json
JSON-LD: 52383-1.0360658-ld.json
RDF/XML (Pretty): 52383-1.0360658-rdf.xml
RDF/JSON: 52383-1.0360658-rdf.json
Turtle: 52383-1.0360658-turtle.txt
N-Triples: 52383-1.0360658-rdf-ntriples.txt
Original Record: 52383-1.0360658-source.json
Full Text

Full Text

RESEARCH Open AccessThe feasibility of assessing swallowingphysiology following prolonged intubationafter cardiovascular surgeryStacey A. Skoretz1,2,6*, Terrence M. Yau3, John T. Granton4,5 and Rosemary Martino6,7AbstractBackground: Dysphagia following prolonged intubation after cardiovascular (CV) surgery is common occurring in 67%of patients; however, this population’s swallowing physiology has never been prospectively evaluated using standardizedmethods. Hence, prior to conducting a larger study, our primary objective was to determine the feasibility of assessingswallowing physiology using instrumentation and validated interpretation methods in cardiovascular surgical patientsfollowing prolonged intubation.Method: From July to October 2011, we approached adults undergoing CV surgery at our institution whowere intubated > 48 h. Those with a tracheostomy were excluded. Videofluoroscopic swallowing study (VFS)and nasendoscopy were completed within 48 h after extubation. Feasibility measurements included recruitmentrate, patient participation, task completion durations, and the inter-rater reliability of VFS measures using the intraclasscorrelation coefficient (ICC). VFSs were interpreted using perceptual rating tools (Modified Barium SwallowMeasurement Tool for Swallow Impairment™© and Penetration Aspiration Scale) and objective displacementmeasurements (hyoid displacement and pharyngeal constriction ratio).Results: Of the 39 patients intubated > 48 h, 16 met inclusion criteria with three enrolled and completing the VFS. Allrefused nasendoscopy. Across all VFSs, rating completion time ranged from 14.6 to 51.7 min per patient with ICCs forVFS scales ranging from 0.25 (95% CI − 0.10 to 0.59) to 0.99 (95% CI 0.98 to 0.99).Conclusions: This study design was not feasible as recruitment was slow, few patients participated, and no patientagreed to all procedures. We discuss necessary methodological changes and lessons learned that would generalize tofuture research.Keywords: Swallowing, Dysphagia, Intubation, Cardiovascular, Feasibility, Speech-language pathology, VideofluoroscopyBackgroundDysphagia occurs in two thirds of all patients who areintubated for 48 h or more following cardiovascular(CV) surgery [1, 2]. Not only does post-extubationdysphagia increase hospitalization costs [3]; patients withthe disorder are at greater risk of pneumonia, reintuba-tion, and death [4].What is known about dysphagia frequency and swallow-ing physiology following prolonged intubation after CVsurgery is limited by study design and the quality of theavailable literature [5]. Dysphagia incidence, as reported todate, is variable given these limitations. Available literatureincludes retrospective studies that rely on chart audit dataand heterogeneous assessment methods [5]. Furthermore,many studies have inherent bias risk as they lack (1) asses-sor blinding to clinical data and/or outcomes and (2)validated rating methods for instrumental assessments [5].To date, no study has analyzed swallowing physiologyusing standardized ratings in consecutively enrolled pa-tients following prolonged intubation after CV surgery.Hence, the feasibility of a rigorous research paradigm thatincludes CV surgical patients from intensive care as wellas blinded raters using validated tools has yet to be* Correspondence: sskoretz@audiospeech.ubc.ca1School of Audiology and Speech Sciences, University of British Columbia,#421-2177 Wesbrook Mall, Vancouver, BC V6T 1Z3, Canada2Department of Critical Care, University of Alberta, 2-124 Clinical SciencesBuilding, Edmonton, AB T6G 2B7, CanadaFull list of author information is available at the end of the article© The Author(s). 2017 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, andreproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link tothe Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver(http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.Skoretz et al. Pilot and Feasibility Studies  (2017) 3:62 DOI 10.1186/s40814-017-0199-7determined. Our primary objective was to determine thefeasibility of using validated and objective interpretationmeasures for videofluoroscopy in conjunction with nasen-doscopy to assess swallowing and upper airway physiologyon prospectively enrolled CV surgery patients followingprolonged intubation. Our secondary objective was to ex-plore the tolerability and impact of this study on patientsand nursing practice. These findings will be used to in-form a future large-scale study to systematically evaluatethe incidence of dysphagia and comprehensively assessthe swallowing physiology of this patient population.MethodsParticipantsFrom July 1 to October 31 2011, patients at our institu-tion intubated for > 48 h following CV surgery wereapproached for study participation and consented by anindependent research assistant. Eligible patients includedthose > 18 years of age who were extubated within theprevious 48 h. We excluded patients with a history ofdysphagia, tracheotomy, head/neck cancer, or neuro-logical disorders including stroke or seizures. Also ex-cluded were patients deemed inappropriate for studyparticipation by the attending medical team, includingthose with reduced consciousness, medical instability orwho were nil per os (NPO) due to gastroenterologiccomplications. Our institutional research ethics boardapproved this study, and all participants or surrogatedecision-makers provided written informed consentprior to participation.Study processWithin 48 h following extubation, the patient’s swallowwas assessed using a videofluoroscopic swallowing study(VFS) conducted by a speech-language pathologist blindedto all clinical data. In addition, these patients wereapproached to undergo flexible nasendoscopy by otolaryn-gology to assess for upper airway pathology. At the com-pletion of instrumental testing, we invited the patient andattending nurse to provide anonymous feedback on thestudy process using a self-administered impact question-naire. We measured process and resource feasibilityincluding recruitment rate (aim of ≥ two patients/week inorder to meet a convenience sample size of N = 30), num-ber of eligible patients, losses to enrollment, patient par-ticipation with all instrumental protocols, task completiontimes, and the inter-rater reliability of the VFS measures(aim of ICC ≥ 0.60).Videofluoroscopic swallow studyImagingFluoroscopy was conducted using a Toshiba UltimaxSystem MDX-8000A (Toshiba America Medical SystemsInc., Tustin, CA) in the lateral position. Using continu-ous pulse, the image was captured in uncompressedform using a TIMS 2000 DICOM system (Forest Im-aging, Chelmsford, MA) at a rate of 30 frames per sec-ond. Image collimation allowed for views of the anteriorlip margins, superior aspect of the nasal passages, pos-terior margins of the cervical vertebrae, and cervicalesophagus. A visible scalar with known dimensions (aquarter) was placed submentally in the image field forcalibration of image magnification.Videofluoroscopic swallow study procedureEach patient was presented with various fluid and foodtextures combined with E-Z-EM barium contrast agents(E-Z-EM Inc., Lake Success, NY) according to institu-tional standard practice and preparation. The trials werepresented as follows: (1) thin liquid with diluted liquidPolibar Plus barium sulfate suspension (47% w/v;3 × 5 ml boluses by spoon, 2 × 15 ml boluses by cup),(2) applesauce mixed with powdered barium (28% w/w;3 × 5 ml boluses by spoon), (3) ½ Peak Frean digestivecookie coated with barium paste (60% w/w), and (4) se-quential cup sips of the thin liquid barium dilution (47%w/v; 100-ml maximum). If the patient exhibited aspir-ation of any texture, a larger bolus of that texture wasnot administered. The testing sequence was discontin-ued in its entirety if aspiration of applesauce occurred.For all trials, except for sequential cup sipping, a cued-swallow paradigm was used with instruction to hold thebolus in the oral cavity and then swallow on command.Videofluoroscopic swallow study measures FollowingVFS completion, we conducted two types of evaluationsusing (i) standardized VFS rating tools and (ii) objectivedisplacement measurements.i. Standardized VFS rating tools. Each VFS was scoredin its entirety using two standardized, validated, andreliable tools: the Modified Barium SwallowImpairment Profile (MBSImp™©; Northern SpeechServices, Gaylord, MI) [6] and the PenetrationAspiration Scale (PAS) [7]. These tools arecomplementary, measuring swallowing impairmentand airway protection respectively.One MBSImp™© certified rater (SAS) completed theMBSImp™© for each bolus administration while blindedto clinical data. The MBSImp™© is a standardized toolused to quantify severity of oral and pharyngeal impair-ment through assessment of 17 physiologic components.Each component has a rank-ordered scoring system,ranging from a three- to a five-point scale, with increas-ing scores indicating greater impairment. A priori, weexcluded two components, pharyngeal constriction andSkoretz et al. Pilot and Feasibility Studies  (2017) 3:62 Page 2 of 10esophageal clearance, the former, as it requires an anter-ior posterior fluoroscopic view, which cannot be ob-tained from this patient population given their restrictedmobility at the time of testing and, the latter, as it wasbeyond the scope of this study. For each patient, eachbolus administration was scored individually accordingto published guidelines [6], with the patient receivingthe highest, most impaired score if a texture or bolusvolume was not administered due to safety concerns.Two independent raters (SAS and RM) were blindedto each other and all clinical data conducted PAS scor-ing for each bolus administration. The PAS [7] is aneight-point ordinal severity scale scoring the depth ofairway invasion by the bolus, whether it is expelled fromthe airway as well as any patient reaction. It ranges fromone (material does not enter the airway) to eight (mater-ial enters the airway, passes below the vocal folds, andno effort is made to eject).ii. Displacement measurements. Two independentraters (SAS and RM) blinded to each other and allclinical data, conducted frame selection anddisplacement measurements for each 5 and 15-mlbolus using ImageJ (National Institutes of Health,Bethesda, MD) focusing on two domains: hyolaryngealexcursion and pharyngeal constriction [8–10]. Thesespecific features are associated with successfulpropulsion of food or fluid through the pharynxinto the esophagus thereby preventing laryngealpenetration, tracheal aspiration, and/or pharyngealresidue [9–15].Hyoid excursion during each swallow was calculatedaccording to two techniques: (1) as an absolute trajec-tory measurement in millimeters (mm) quantifying theanterior-superior displacement of the hyoid bone fromits rest position [10, 16] and (2) as an internally scaledcalculation of separate anterior and superior hyoidmovement using the patient’s cervical vertebrae as a ref-erence [9, 17]. We utilized two fluoroscopy images perbolus for both techniques: one with the hyoid at restand one at maximal anterior-superior displacement. Thehyoid “rest” frame was defined differently for each tech-nique: (1) during bolus hold prior to swallow initiationfor the absolute measurements and (2) during the lowesthyoid position following swallow completion for thescaled measurements. Frame selection, anatomical tra-cing specifications, and displacement calculations forboth techniques were completed according to previouslypublished methods [9, 10, 13, 17].We measured pharyngeal constriction during the swal-low using ImageJ through a pharyngeal constriction ratio(PCR) calculation [10, 13, 16]. For this measurement, wedefined the bolus hold frame (PAHOLD) as the holdduring the 5-ml thin liquid bolus. The maximumpharyngeal contraction frames (PAMAX) were defined asthe frame with maximal pharyngeal contraction duringeach of the following: 5- and 15-ml thin liquid and 5-mlpureed boluses. Following pharyngeal space tracing foreach frame, we calculated the pharyngeal constrictionratio by dividing the pharyngeal area (cm2) of the se-lected frames (PAMAX/PAHOLD) [10, 13, 16].Study impact questionnaires and patient variablesFollowing VFS completion, the patient and attendingnurse were asked to complete the study impact ques-tionnaires (Table 1). The questionnaires utilized a 5-point Likert-like scale rating patient comfort, studyimpact on nursing workload and perceived study valuewhile providing an open-ended section for comments.Once completed, we instructed the patient and/or nurseto deposit their questionnaire in a locked box placed onthe hospital unit. Following unblinding, the first author(SAS) recorded patient variables such as demographics(age, gender) and operative information (surgery type,intubation duration). The first author (SAS) entered andanalyzed all data following VFS rating completion andunblinding to clinical data.Scoring and statistical analysesFor each patient, we scored each bolus administrationindividually with both the MBSImp™© and PAS. We de-rived overall impression impairment scores (OI) for eachof the MBSImp™© components for each patient accord-ing to published standards, namely (1) levying the max-imum score should a bolus texture or volume not beadministered due to safety concerns and (2) using thepatient’s worst score for each component regardless oftexture or volume [6]. We derived the total oral impair-ment (components 1–6) and total pharyngeal impair-ment (components 7–12 and 14–16) scores for eachpatient by the summation of each OI score from the ap-propriate oral and pharyngeal components. We dichoto-mized PAS scores as either normal (PAS scores 1 or 2)or abnormal (PAS scores 3 to 8) airway protection foreach bolus administration [18, 19] and summarizedthese data as frequency counts.We reported the approximate duration in minutes forpatient transport to medical imaging, patient VFSparticipation, and task completion (VFS preparation,study archiving). Completion time for (1) frame selectionand deriving displacement measurements, (2) MBSImp™©,and (3) PAS ratings were reported in minutes as meanswith standard deviations (SD). We summarized question-naire responses descriptively and according to frequencycounts. Across each bolus administration as describedabove, we explored the single measures inter-observeragreement using absolute agreement two-way mixedSkoretz et al. Pilot and Feasibility Studies  (2017) 3:62 Page 3 of 10intraclass correlation coefficient (ICC) with 95% confi-dence intervals (CI) for (1) the PAS (i.e., classificationagreement across the eight-point scale), (2) frame selec-tion (i.e., selection agreement for the following frames:rest/hold, maximum hyoid displacement, and maximumpharyngeal constriction), and (3) displacement measure-ments (i.e., hyoid trajectory, scaled hyoid displacement,and pharyngeal constriction ratio). These ICC estimateswere crossed two-way analyses of variances adjusting forrepeated measures. Statistical analyses were conductedusing IBM SPSS version 22 (IBM Corporation, Armonk,NY). Radar plots were created using OriginPro 9.1 (Origi-nLab Corporation, Northampton, MA).ResultsPatient recruitment and characteristicsDuring the 4-month study period, 39 patientsrequired intubation for more than 48 h (Fig. 1). Ofthose, 16 met the inclusion criteria with threepatients agreeing to participate for an approximaterecruitment rate of one patient every 5 weeks. Of theremaining 13 patients, six declined participation, sixwere not approached due to institutional and oper-ational restrictions (i.e., no weekend coverage anddiagnostic imaging suite downtime), and one patientwas transferred off-service.Our study consisted of two males and one femalewith ages ranging from 37 to 71 years. Intubation du-rations for patient 1 (P1), patient 2 (P2), and patient3 (P3) were 53.3, 49.3, and 82.5 h respectively. Otherdemographic and medical information was collectedbut not included here to preserve patient anonymityin light of the small sample. At the time of study par-ticipation, all patients were in intensive care requiring24-h one-to-one nursing care. All enrolled patientsrefused nasendoscopy.Table 1 Study impact questionnairesQuestion Rating scale1 2 3 4 5Patient questionnaire How did you find the x-ray swallow test? Easy SomewhateasyAdequate SomewhatdifficultVery difficultHow did you find the nasendoscopy?Overall, how was your experience with this research?How was this form to complete?Nursing questionnaire In your opinion, to what degree did this study affect thedelivery of patient care?Not atallVery little A little Somewhat A lotHow did the participation of this study fit into your daily tasks? Easily SomewhateasilyAdequately WithdifficultyWith greatdifficultyHow were you able to accommodate being part of thevideofluoroscopic swallow study?How was this form to complete? Easy SomewhateasyAdequate SomewhatdifficultVery difficultComments:Intubated >48h (n=39)Excluded (n=23)Exclusion diagnoses (n=10)Expired (n=6)Tracheotomy (n=7)Unable to enroll (n=13)Imaging restrictions (n=2)Weekend extubation (n=4)Patient transfer (n=1)Declined participation (n=6)Assessed for eligibility (n=39)Invited (n=16)EnrollmentAssessmentConsented and enrolled (n=3)Follow-upInstrumental assessments(n=3) Videofluoroscopy (n=3)Nasendoscopy (n=0; all declined)Self-administered impact questionnaires (n=3)Completed questionnaire (n=1)Unable to complete (n=2)Discharged (n=1)Medical instability (n=1)Fig. 1 Study enrollmentSkoretz et al. Pilot and Feasibility Studies  (2017) 3:62 Page 4 of 10Videofluoroscopic swallow studyThe videofluoroscopic swallow study (VFS) was com-pleted 30, 25, and 37 h following extubation on P1, P2,and P3 respectively. All consistencies and volumes wereadministered to P1 and P2. For P3, the VFS protocolwas discontinued following two of the three pureed(5 ml) boluses due to safety concerns. All patients wereNPO prior to the VFS. Artifacts on the radiographicimage included a central line (P1 and P2) and an in situnasogastric feeding tube (P3).For each VFS, patient preparation and transporttime to and from the fluoroscopy suite ranged from30 to 60 min. Prior to patient arrival, room andequipment preparation required an average of 10 min.While in the imaging suite, patient setup and the VFSwere completed within 15 min. Following the comple-tion of each VFS, data transfer and study archivingtook approximately 1 h. A total of three studypersonnel were involved in the VFS directly: an x-raytechnician, a speech-language pathologist (SAS), and aresearch assistant. The patient’s nurse and clinicalspeech-language pathologist were also in attendance.Videofluoroscopic measuresVideofluoroscopic measurements were completed forall three patients. The time to complete frame selec-tion and measurement for each method ranged fromless than 2 to 5.9 min per bolus administration. Thetotal time to complete each measurement across alltargeted bolus volumes ranged from 14.6 to51.7 min for each VFS. For the displacement mea-sures, the frame selection ratio ICC ranged from0.98 (0.96 to 0.99) to 0.99 (0.98 to 0.99). Contras-tively, the displacement measurement ICC rangedfrom 0.25 (− 0.10 to 0.59) to 0.90 (0.76 to 0.96). TheICC for the PAS was 0.92 (0.83 to 0.96). Each VFSmeasure and its corresponding task completion timeand inter-rater reliability (as applicable) is summa-rized in Table 2.Standardized VFS rating toolsP3 received the maximum overall impression (OI) scoreacross all individual MBSImp™© oral and pharyngealcomponents (Fig. 2a, b). As a result, P3 received thehighest total oral and total adjusted pharyngeal scores of22 and 26 respectively, exhibiting the most severe swal-lowing impairment of all patients in our study. Of thetwo remaining patients, P1 received a total oralimpairment score of nine as compared to three for P2,with greater impairment on lip closure, bolus prepar-ation, bolus transport, and the initiation of thepharyngeal swallow. Conversely, P2 received a totalpharyngeal impairment score of six as compared to fivefor P1, with greater impairment on laryngeal vestibuleclosure and pharyngeal stripping wave components. Ofthe remaining pharyngeal components, P1 and P2 differedonly on tongue base retraction with OI scores of two andone respectively. Of the oral components, all three pa-tients exhibited impairment on oral residue and initiationof the pharyngeal swallow. Additionally, all patients exhib-ited impairment on the following pharyngeal components:pharyngoesophageal segment opening, tongue base retrac-tion, and pharyngeal residue. Across all bolus administra-tions, airway protection of P1 and P2 was rated as normal(PAS scores ≤ 2) whereas the airway protection of P3 wasrated abnormal (PAS scores ≥ 3). Clinically, P1 and P2began regular texture diets following VFS completionwhile P3 remained NPO with enteral feeding via nasogas-tric feeding tube.Displacement measurementsHyoid displacements are presented in Table 3. In P3, wefound the smallest median (interquartile range (IQR))absolute hyoid displacement [10, 16]: 8.3 mm (2.3) and8.3 mm (1.7) with thin (5 ml) and pureed (5 ml) texturesrespectively. Contrastively, in P1, we found the largestmedian (IQR) displacements 12.7 mm (1.7) and14.8 mm (2.1) respectively. Absolute displacementmedian values increased with both increasing volume asTable 2 Task completion time and inter-rater reliability according to VFS measureVFS measure Completion timea ICC (95% CI)Per bolus administration Per VFS Frame selection ratio MeasurementMBSImp™© 5.9 (0.8) 51.7 (16.5) N/A N/APAS 1.9 (0.9) 17.3 (10.1) N/A 0.92 (0.83 to 0.96)Absolute hyoid displacementb 5.2 (2.5) 38.7 (21.5) 0.99 (0.98 to 0.99) 0.90 (0.76 to 0.96)Scaled hyoid displacementb 1.9 (0.5) 14.6 (4.8) 0.98 (0.96 to 0.99) 0.26 (− 0.05 to 0.52)Pharyngeal constrictionb 4.5 (1.0) 33.3 (10.1) 0.99 (0.98 to 0.99) 0.25 (− 0.10 to 0.59)Note. VFS videofluoroscopic swallow study, CI confidence interval, MBSImp™© Modified Barium Swallow Impairment Profile, N/A not applicable, PAS PenetrationAspiration ScaleaReported in minutes; mean (SD)bDisplacement measurements conducted on 5 and 15-ml bolus volumes onlySkoretz et al. Pilot and Feasibility Studies  (2017) 3:62 Page 5 of 10well as increasing texture viscosity for P1 and P2whereas median values for P3 remained consistentregardless of stimulus change. Scaled anterior and super-ior hyoid displacement measurements (%C2–4 distance)[9, 17] were similarly patterned with the (1) smallestmedian values regardless of texture for P3 and (2) largestmedian values for P1 with thin fluid boluses.When comparing PCR measurements across patientsaccording to bolus texture and volume (Fig. 3), we mea-sured the smallest median (IQR) pharyngeal constrictionabFig. 2 a MBSImp™© oral components across patients. Note. LipC = lip closure, TC = tongue control, BP = bolus preparation, BT = bolus transport,OR = oral residue, IPS = initiation of pharyngeal swallow; scoring, 0 = normal, ≥ 1 = impairment. b MBSImp™© pharyngeal components acrosspatients. Note. SPE = soft palate elevation, LE = laryngeal elevation, HM = hyoid movement, EM = epiglottic movement, LVC = laryngeal vestibuleclosure, PSW = pharyngeal stripping wave, PESO = pharyngoesophageal segment opening; TBR = tongue base retraction; PR = pharyngealresidue; scoring: 0 = normal, ≥ 1 = impairmentTable 3 Hyoid displacement measurements according to patientCase Absolute hyoid displacement (mm) Scaled hyoid displacement (%C2–4 distance)Anterior hyoid displacement Superior hyoid displacementThin liquid Pureed Thin liquid Pureed Thin liquid Pureed5 ml 15 ml 5 ml 5 ml 15 ml 5 ml 5 ml 15 ml 5 mlP1 12.7 (1.7) 14.2 (3.2) 14.8 (2.1) 41.6 (18.7) 52.4 (6.0) 40.8 (30.7) 18.5 (25.8) 31.0 (10.9) 36.9 (25.9)P2 9.8 (1.8) 9.8 (1.6) 9.9 (2.1) 23.1 (41.0) 27.4 (8.1) 24.3 (9.4) 27.9 (57.8) 33.6 (17.0) 36.7 (30.8)P3 8.3 (2.3) NT 8.3 (1.7) 17.9 (12.2) NT 22.5 (5.8) 13.9 (14.5) NT 17.8 (9.5)Note. Values are reported as median (IQR)IQR interquartile range, NT not testedSkoretz et al. Pilot and Feasibility Studies  (2017) 3:62 Page 6 of 10ratio with thin fluid boluses (5 ml) for P1 (0.02 [0.03])and the largest with P3 (0.09 [0.09]) for the same bolustexture and volume. Within patients across bolus textureand volumes, PCR values decreased with increasing vis-cosity for P2 and P3 with the inverse measured for P1.Study impact questionnairesPatient comfort questionnairesOne patient completed the questionnaire. Ratings were“adequate” for VFS and overall study participation and“easy” for questionnaire completion. This patient wouldhave preferred more information during the consentprocess about the instrumental assessment. Of the tworemaining patients, one was medically incapable of com-pleting the questionnaire and one was discharged priorto completion.Workload impact questionnairesAll attending nurses, three cardiovascular intensive careunit (CVICU) nurses and one nursing student, com-pleted the questionnaire. All nurses reported the studyfit well with the daily tasks and the three felt VFS ac-commodation was “easy” or “adequate”. Half of thenurses reported the study’s impact on patient care deliv-ery was minimal. Nursing suggestions for study modifi-cations included (1) further advance notice regarding theprocedure timing in order to allow sufficient lead timeto ready the patient for transport and (2) frequent con-firmation with the nursing staff regarding the patient’shemodynamic stability and inotropic support require-ments prior to scheduling transport.DiscussionEvidenced by a slow recruitment rate with no patientagreeing to all procedures, our study as designed was notfeasible. While we are the first to conduct a study of thisnature with this patient population, the low enrollmentand missing data were devastating. This experience hasafforded us the opportunity to revisit our methodologyprior to conducting our larger study. Despite these fail-ures, we were able to minimize bias risk throughinstrumental assessment on all enrollees, swallow studyinterpretation using validated rating scales, and assessorblinding throughout the study. Our findings demonstratethat for the success of future large-scale studies, we willmaintain select aspects of our design in an effort toreduce unnecessary bias; however, changes need beconsidered to enhance recruitment and consent. Forexample, changes to the enrollment criteria, diagnos-tic methods, and interpretation measures may meetthese goals.Although we targeted the patient group at the highestrisk of developing dysphagia [1, 2], our recruitment ratewas slow: a rate of less than one patient per month. Atour current rate, meeting a hypothetical sample size of100 would take 10 years. We suggest (1) revising our def-inition of prolonged intubation to include those intubatedfor 24 h or longer, (2) increasing research staff availabilityover weekends, (3) increasing the post-extubation instru-mental assessment window from 48 to 72 h, and (4)expanding the study to include multiple centers. Thesechanges must be considered carefully as they may (1)increase the study cost, (2) increase burden on the institu-tion and staff, and (3) require larger sample sizes for par-ticular outcome comparisons. For example, increasing thepost-extubation instrumental assessment window couldbe problematic for a study also targeting dysphagia inci-dence and/or swallow outcome comparisons given thatswallowing physiology changes rapidly following extuba-tion. It has been reported that swallow timing may im-prove within 48 h following extubation [20] withaspiration frequency decreasing within 8 h [21]. As aresult, a broader range in post-extubation assessmenttimes would require patient stratification thereby ne-cessitating a larger sample size.Fig. 3 Pharyngeal constriction ratio by patient according to volumeSkoretz et al. Pilot and Feasibility Studies  (2017) 3:62 Page 7 of 10We incorporated aspects into our study that would pro-vide information regarding study acceptance and consent.Our low enrollment limited our ability to make inferencesregarding study impact on either nursing or patients; how-ever, the responses can provide insight on how to improvestudy process and potentially increase our consent rate.All four nursing staff responded to our impact question-naires and rated our protocol favorably. In contrast, onlyone of the three patients completed their questionnaires.Two patients did not respond due to discharge and med-ical instability respectively. Our single patient respondentsuggested that in order to improve the consenting process,the instrumental assessment should be explained morefully. Given the limited enrollment and all patients refus-ing nasendoscopy, the study protocol should be altered toinclude only one instrumental procedure. For future stud-ies, we will review the consenting process not limited tothe information presented to the participants. Nursingcommented on timing issues regarding transport to radi-ology. We hypothesize that a bedside instrumental swal-lowing assessment may be favored by both nursing andpatient. It would eliminate patient transport outside of theICU and in so doing reduce (1) the number of staff re-quired to conduct the assessment, (2) the time spenttransporting the patient, and (3) reduce patient burden.Moving forward, we recommend the conduct of struc-tured interviews with potential study participants includ-ing patients, caregivers, and nursing staff. This wouldenable investigators to have a greater understanding of theprocedures with which they are willing to participate,when they would be willing to participate and their prefer-ences for the consenting process.This is the first study to provide a detailed assessmentof swallowing physiology in this population using a validand reliable interpretation method of videofluoroscopy.All three patients exhibited swallowing impairmentacross multiple areas. However, while the swallowing im-pairments found in our study were similar to those re-ported previously [22, 23], we did note an incongruencebetween the MBSImp™© scores for P1 and P2 with whatwas observed clinically. Both patients were able to con-sume regular texture diets without difficulty. As a result,in conjunction with valid and objective interpretationmethods for videofluoroscopy, we recommend that fu-ture research also incorporate a means by which to de-termine clinically relevant dysphagia.To the best of our knowledge, we are the first to reporton swallowing kinematics on recently extubated patientsfollowing CV surgery. When comparing inter-rater agree-ment across all displacement measures, caution needs tobe taken as these estimates are based on only three pa-tients. The reliability for frame selection across all mea-sures and absolute hyoid displacement measurement wasexcellent [24]. In contrast, the reliability was poor for bothscaled hyoid displacement measurement as well aspharyngeal constriction ratio. This may be due to our verysmall sample size or patient factors. Due to the anatomicallocation of a radiographic artifact either along the cervicalspine (i.e., central line) or in the pharynx (i.e., nasogastrictube) across each patient, it may be that these measureswere most susceptible thereby accounting for the highvariability between raters. During our reliability training,our sample studies did not include these artifacts. For thispopulation specifically, displacement and pharyngeal con-striction are important measures due to the relative in-activity of the laryngeal and pharyngeal musculatureduring intubation and the potential effect of the endo-tracheal tube on swallow function [25, 26]. Addressingmeasurement reliability prior to enrollment using fluoros-copy samples with similar radiographic artifacts would notonly be prudent but critical.This feasibility study, which by definition is descriptiveand exploratory in nature, has several additional limita-tions. Not only was this study conducted for a very lim-ited duration, we were unable to approach all potentialparticipants for study participation. This limited our en-rollment resulting in an inability to report dysphagia fre-quency. In addition, the small sample size precludedstatistical comparisons of our VFS measures, with thereported swallowing characteristics unlikely to be repre-sentative of this population as whole. Our patients’ acu-ity restricted our VFS assessments as all patientspresented with either a central line or nasogastric feed-ing tube.ConclusionsIn conclusion, while we were able to conduct VFSssuccessfully on our participants shortly after extubationfollowing CV surgery, our inability to successfullyexecute our complete study protocol and slow enroll-ment rate rendered our design not feasible. We have,however, learned lessons throughout the process whichinform the changes necessary for future feasibility stud-ies and protocols. Prior to the conduct of future largeprospective studies, investigators should engage both pa-tients and staff to gain a greater understanding of theprocedures with which they are willing to participateand to improve the consenting process. Other designchanges may include patients intubated for 24 h or morewhile allowing for instrumental swallowing assessmentsthroughout the entire seven-day week. Also, patientburden can be minimized with only one instrumentalprocedure and depending on study objectives, a bedsideinstrumental protocol such as fiberoptic endoscopicevaluation of swallowing (FEES) may be easier toimplement for acute patient populations. However, dueto the limited knowledge surrounding this population’sswallowing physiology, some future studies should stillSkoretz et al. Pilot and Feasibility Studies  (2017) 3:62 Page 8 of 10incorporate VFS. This instrumentation permits kine-matic, temporal, and event sequence measurement. Inorder to maintain objective interpretation, it is crucial touse standardized rating tools and inter-rater reliabilitycalibrated a priori to ensure consistency in displacementmeasurements. Although dysphagia incidence andswallowing physiology of this population still remainselusive, our study provides the necessary foundation forfuture investigations focused on dysphagia frequencyand swallow characteristics in post-extubation patients.AbbreviationsBP: Bolus preparation; BT: Bolus transport; CA: California; CI: Confidenceinterval; CV: Cardiovascular; CVICU: Cardiovascular intensive care unit;EM: Epiglottic movement; FEES: Fiberoptic endoscopic evaluation ofswallowing; HM: Hyoid movement; ICC: Intraclass correlation coefficient;ICU: Intensive care unit; IPS: Initiation of pharyngeal swallow;IQR: Interquartile range; LE: Laryngeal elevation; LipC: Lip closure;LVC: Laryngeal vestibule closure; MA: Massachusetts; MBSImp™©: ModifiedBarium Swallow Measurement Tool for Swallow Impairment; MD: Maryland;MI: Michigan; N/A: Not applicable; NPO: Nil per os; NT: Not tested; NY: NewYork; OI: Overall impression impairment scores; OR: Oral residue; P1: Patient1; P2: Patient 2; P3: Patient 3; PAS: Penetration Aspiration Scale;PCR: Pharyngeal constriction ratio; PESO: Pharyngoesophageal segmentopening; PR: Pharyngeal residue; PSW: Pharyngeal stripping wave;RM: Rosemary Martino; SAS: Stacey A. Skoretz; SD: Standard deviation;SPE: Soft palate elevation; TBR: Tongue base retraction; TC: Tongue control;VFS: Videofluoroscopic swallowing studyAcknowledgementsWe gratefully acknowledge the contributions of our research assistants:Loredana Cuglietta (LC) and Beatrix Reichardt (BR).FundingSAS was supported by an Ontario Graduate Scholarship from the OntarioMinistry of Training, Colleges and Universities. TMY holds the Angelo &Lorenza DeGasperis Chair in Cardiovascular Surgery Research. JTG is aphysician with the Division of Respirology and interdepartmental Division ofCritical Care with the University Health Network. RM holds a CanadaResearch Chair (Tier II) in Swallowing Disorders.Availability of data and materialsAll data generated or analyzed during this study are included in thispublished article except for medical information that may potentially identifythe participants. This information was collected but withheld due to theunanticipated limited enrollment in our study.Authors’ contributionsSAS designed the study, analyzed and interpreted the data, and drafted themanuscript. RM was a major contributor to the design, data analysis,interpretation, and manuscript. TMY and JTG made contributions to thestudy design, methods, and interpretation of results. All authors read andapproved the final manuscript.Ethics approval and consent to participateThis study was approved by the University Health Network Research EthicsBoard in Toronto, Ontario (REB file#: 09-1000-AE). All participants or, if unable,their surrogate decision makers provided written informed consent for studyparticipation and publication.Consent for publicationThe patients participating in this study, or their surrogate decision makers,signed our institutional consent form.Competing interestsThe authors declare that they have no competing interests.Publisher’s NoteSpringer Nature remains neutral with regard to jurisdictional claims inpublished maps and institutional affiliations.Author details1School of Audiology and Speech Sciences, University of British Columbia,#421-2177 Wesbrook Mall, Vancouver, BC V6T 1Z3, Canada. 2Department ofCritical Care, University of Alberta, 2-124 Clinical Sciences Building,Edmonton, AB T6G 2B7, Canada. 3Division of Cardiovascular Surgery,University Health Network, Toronto General Hospital, 200 Elizabeth Street,Toronto, ON M5G 2C4, Canada. 4Division of Critical Care, Toronto GeneralHospital, 200 Elizabeth Street, Toronto, ON M5G 2C4, Canada. 5Division ofRespirology, Toronto General Hospital, 200 Elizabeth Street, Toronto, ON M5G2C4, Canada. 6Department of Speech-Language Pathology, University ofToronto, 160-500 University Ave, Toronto, ON M5G 1V7, Canada. 7Division ofHealth Care and Outcomes Research, Krembil Research Institute, UniversityHealth Network, 399 Bathurst Street, Main Pavilion 11-331, Toronto, ON M5T2S8, Canada.Received: 7 February 2017 Accepted: 26 October 2017References1. Barker J, Martino R, Reichardt B, Hickey EJ, Ralph-Edwards A. Incidence andimpact of dysphagia in patients receiving prolonged endotracheal intubationafter cardiac surgery. Can J Surg. 2009;52(2):119–24.2. Skoretz SA, Yau TM, Ivanov J, Granton JT, Martino R. Dysphagia andassociated risk factors following extubation in cardiovascular surgicalpatients. Dysphagia. 2014;29(6):647–54.3. Ferraris VA, Ferraris SP, Moritz DM, Welch S. Oropharyngeal dysphagiaafter cardiac operations. Ann Thorac Surg. 2001;71(6):1792–5. discussion17964. Macht M, Wimbish T, Clark BJ, Benson AB, Burnham EL, Williams A, Moss M.Postextubation dysphagia is persistent and associated with poor outcomesin survivors of critical illness. Crit Care. 2011;15(5):R231.5. Skoretz SA, Flowers HL, Martino R. The incidence of dysphagia followingendotracheal intubation: a systematic review. Chest. 2010;137(3):665–73.6. Martin-Harris B, Brodsky MB, Michel Y, Castell DO, Schleicher M, Sandidge J,Maxwell R, Blair J. MBS measurement tool for swallow impairment—MBSImp:establishing a standard. Dysphagia. 2008;23(4):392–405.7. Rosenbek JC, Robbins JA, Roecker EB, Coyle JL, Wood JL. A penetration-aspiration scale. Dysphagia. 1996;11(2):93–8.8. Molfenter SM, Steele CM. Use of an anatomical scalar to control for sex-based size differences in measures of hyoid excursion during swallowing.J Speech Lang Hear Res. 2014;57(3):768–78.9. Steele CM, Bailey GL, Chau T, Molfenter SM, Oshalla M, Waito AA, ZorattoDCBH. The relationship between hyoid and laryngeal displacement andswallowing impairment. Clin Otolaryngol 2011; 36(1):30-36.10. Leonard R, Kendall K. Dysphagia assessment and treatment planning: ateam approach. 2nd ed. San Diego, CA: Plural Publishing; 2007.11. Kim Y, McCullough GH. Maximum hyoid displacement in normalswallowing. Dysphagia. 2008;23(3):274–9.12. Jacob P, Kahrilas PJ, Logemann JA, Shah V, Ha T. Upper esophagealsphincter opening and modulation during swallowing. Gastroenterology.1989;97(6):1469–78.13. Leonard R, Rees CJ, Belafsky P, Allen J. Fluoroscopic surrogate forpharyngeal strength: the pharyngeal constriction ratio (PCR). Dysphagia.2011;26(1):13–7.14. Yip H, Leonard R, Belafsky PC. Can a fluoroscopic estimation ofpharyngeal constriction predict aspiration? Otolaryngol Head Neck Surg.2006;135(2):215–7.15. Setzen M, Cohen MA, Perlman PW, Belafsky PC, Guss J, Mattucci KF, DitkoffM. The association between laryngopharyngeal sensory deficits, pharyngealmotor function, and the prevalence of aspiration with thin liquids.Otolaryngol Head Neck Surg. 2003;128(1):99–102.16. Leonard RJ, Kendall KA, McKenzie S, Gonçalves MI, Walker A. Structuraldisplacements in normal swallowing: a videofluoroscopic study. Dysphagia.2000;15(3):146–52.17. Steele, CM. Hyoid Displacement Calculation. http://steeleswallowinglab.ca/srrl/best-practice/hyoid-movement/. Accessed 21 June 2012.Skoretz et al. Pilot and Feasibility Studies  (2017) 3:62 Page 9 of 1018. Daggett A, Logemann J, Rademaker A, Pauloski B. Laryngealpenetration during deglutition in normal subjects of various ages.Dysphagia. 2006;21(4):270–4.19. Allen JE, White CJ, Leonard RJ, Belafsky PC. Prevalence of penetration andaspiration on videofluoroscopy in normal individuals without dysphagia.Otolaryngol Head Neck Surg. 2010;142(2):208–13.20. de Larminat V, Montravers P, Dureuil B, Desmonts JM. Alteration inswallowing reflex after extubation in intensive care unit patients. Crit CareMed. 1995;23(3):486–90.21. Burgess GE 3rd, Cooper JR Jr, Marino RJ, Peuler MJ, Warriner RA 3rd. Laryngealcompetence after tracheal extubation. Anesthesiology. 1979;51(1):73–7.22. Hogue CW Jr, Lappas GD, Creswell LL, Ferguson TB Jr, Sample M, Pugh D, BalfeD, Cox JL, Lappas DG. Swallowing dysfunction after cardiac operations.Associated adverse outcomes and risk factors including intraoperativetransesophageal echocardiography. J Thorac Cardiovasc Surg. 1995;110(2):517–22.23. Partik BL, Scharitzer M, Schueller G, Voracek M, Schima W, Schober E,Mueller MR, Leung AN, Denk DM, Pokieser P. Videofluoroscopy ofswallowing abnormalities in 22 symptomatic patients after cardiovascularsurgery. AJR Am J Roentgenol. 2003;180(4):987–92.24. Shrout PE, Fleiss JL. Intraclass correlations: uses in assessing rater reliability.Psychol Bull. 1979;86(2):420–8.25. Postma GN, McGuirt WF, Butler SG, Rees CJ, Crandall HL, Tansavatdi K.Laryngopharyngeal abnormalities in hospitalized patients with dysphagia.Laryngoscope. 2007;117(10):1720–2.26. Thomas R, Kumar EV, Kameswaran M, Shamim A, al Ghamdi S, MummigattyAP, Okafor BC. Post intubation laryngeal sequelae in an intensive care unit.J Laryngol Otol. 1995;109(4):313–6.•  We accept pre-submission inquiries •  Our selector tool helps you to find the most relevant journal•  We provide round the clock customer support •  Convenient online submission•  Thorough peer review•  Inclusion in PubMed and all major indexing services •  Maximum visibility for your researchSubmit your manuscript atwww.biomedcentral.com/submitSubmit your next manuscript to BioMed Central and we will help you at every step:Skoretz et al. Pilot and Feasibility Studies  (2017) 3:62 Page 10 of 10


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



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"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items