Introduction
Individuals with CHD particularly affecting the pulmonary valve and right ventricular outflow tract may require pulmonary valve replacement in adolescence or young adult years after primary corrective surgery. Reference Batra, McElhinney and Wang1–Reference Yang, Wattenbarger and Taylor9 Indications for pulmonary valve replacement include imaging biomarkers, electrocardiogram changes, and clinical symptoms. Exercise intolerance, when not due to extracardiac disease, is considered an indication for pulmonary valve replacement; however, data regarding the exertional benefit following pulmonary valve replacement are sparse. Reference Batra, McElhinney and Wang1–Reference Schoonbeek, Takebayashi and Aoki11 Cardiopulmonary exercise testing found mixed improvement in percent predicted VO2 and VE/VCO2 following pulmonary valve replacement across several studies. Reference Batra, McElhinney and Wang1,Reference Eyskens, Reybrouck and Bogaert4–Reference Hasan, Lunze and Chen6,Reference Sabate Rotes, Johnson, Burkhart, Eidem, Allison and Driscoll8,Reference Babu-Narayan, Diller and Gheta12–Reference Sandeep, Cheng and Luo15 The mixed data in these metrics demonstrate the need for a reliable measure to capture changes in exercise capabilities following pulmonary valve replacement.
Accelerometry provides a potential solution for a new outcome measure of exercise tolerance. Accelerometry has recently been utilised in several studies as a measure of daily physical activity levels in both individuals with heart failure and individuals with CHD. Reference Banks, Rosenthal and Manlhiot16–Reference Pozehl, McGuire and Duncan22 Potential advantages of accelerometry include utilisation in a remote, free-living environment and effort independence. This contrasts with standard measures such as cardiopulmonary exercise testing, which are usually effort dependent and must be performed in a clinical space. Reference Batra, McElhinney and Wang1,Reference Sabate Rotes, Johnson, Burkhart, Eidem, Allison and Driscoll8,Reference Babu-Narayan, Diller and Gheta12–Reference Müller, Engelhardt, Fratz, Eicken, Ewert and Hager14
In this study, participants undergoing pulmonary valve replacement were assessed for improvements in a variety of clinical metrics including electrocardiogram, echocardiogram, cardiac MRI (cardiac MRI), cardiopulmonary exercise testing, and accelerometry. Our hypotheses were (1) accelerometry measured physical activity would increase following pulmonary valve replacement and (2) preoperative features (e.g., electrocardiogram and imaging metrics) would associate with the degree of improvement in activity level.
Methods
Participants and clinical measures
This study was approved by the Vanderbilt Institutional Review Board (IRB 211522). All procedures were performed in compliance with relevant laws and institutional guidelines. Privacy rights of human subjects were observed. Eligible participants were identified at time of scheduling pulmonary valve replacement and then were contacted separately to determine interest in participating in the study. Participant demographics and reason for procedure—including relevant surgical history—were documented at time of enrollment (Table A1). Individuals under the age of 13 were not eligible to participate. Participants over the age of 18 provided written informed consent, while participants under the age of 18 provided assent and parent or guardian informed consent. As available, participant preoperative and postoperative echocardiogram and electrocardiogram results were collected from the electronic medical record, and the variables selected for analysis are listed in the supplement (Table A2). Participants underwent either percutaneous or surgical valve replacement. Valve type and length of hospitalisation following replacement were collected via participant chart review.
Accelerometry
Triaxial accelerometers (ActiGraph Link GT9X, Ametris LLC, Pensacola, FL) were utilised as a surrogate marker of daily preoperative physical activity. Prior to pulmonary valve replacement, participants were instructed to wear one accelerometer on their dominant wrist and one ankle accelerometer for 7 days, 24 hours a day, only removing to bathe. Post-pulmonary valve replacement accelerometer measurements were completed 3–6 months after pulmonary valve replacement, during which participants were again asked to wear an accelerometer on both wrist and ankle for 7 days, 24 hours a day. Compliance of wear was computed, taking the fraction of time accelerometer worn out of the 7-day period. Proportions of total activity spent in sedentary, light, and moderate-to-vigorous physical activity were computed as well as the number of activity counts, measured as vector magnitudes, per minute of wear. Accelerometer derived measures of entropy, jerk, and mean frequency were computed as described by Miller et al. Reference Miller, Lohse and Bland23 Additional accelerometer parameter information is available in supplement (Supplementary Material 1).
Quality of life survey
During both periods of accelerometry wear, participants also completed a self-report questionnaire assessing tolerance of daily physical activities. Questions concerning frequency of exercise-induced symptoms (including shortness of breath or fatigue) were scored using a response scale of never, rarely, sometimes, and frequently, while questions concerning challenges with daily activities (including self-care, housework, and general pain and discomfort) were scored using a scale of no problems, moderate problems, or extreme problems. Level of symptom severity was measured based on survey responses using the New York Heart Association classification system. Reference Caraballo, Desai and Mulder24
Statistical analysis
Wilcoxon signed rank tests were performed to assess any significant differences in continuous variables pre-and post-pulmonary valve replacement. A subgroup analysis was performed for participants with exercise intolerance symptoms as indication for pulmonary valve replacement. The subgroup was stratified by degree of severity based on baseline New York Heart Association classification. Spearman correlations were computed for associations between accelerometer-derived measures and clinical endpoints. Correlation strength was defined with r < 0.40 as weak, 0.40 < r < 0.70 as moderate, and r > 0.70 as strongly correlated as previously described. Reference Schober, Boer and Schwarte25 A threshold of p < 0.05 was used for significance. Paired t tests were run to assess significant changes in average score for participant-reported survey items.
Results
A summary of participant demographic information is available in the supplement (Table A1). A total of 18 participants who underwent either transcatheter pulmonary valve replacement or surgical pulmonary valve replacement and completed both the pre-procedure and post-procedure portions of the study were included. At baseline assessment, the cohort’s median age was 26.5 [16.0, 35.0] years, and median body surface area was 1.73 [1.46, 1.93] m2 (Table 1). Twelve (67%) participants were male, and the most common primary diagnosis was tetralogy of Fallot (10/18, 59%). The majority of patients (89%) underwent transcatheter pulmonary valve replacement. Impaired exercise tolerance was reported in 50% (9/18) as a symptom at baseline, with none reporting the symptom’s persistence at clinic follow-up post-procedure.
Median differences in key outcome measures (N =18)
Table 1. Long description
The table presents a comparison of median differences in key outcome measures for 18 participants who underwent either transcatheter or surgical pulmonary valve replacement. It includes pre-procedure and post-procedure median values with interquartile ranges for various measures such as VMs per minute for wrist and ankle, wear compliance for wrist and ankle, FOT sedentary, FOT low intensity, FOT MVPA, entropy, average jerk, mean frequency, QRS duration, LVEF percentage, and peak gradient PS in millimeters of mercury. The table also includes p-values to indicate statistical significance. Notable trends include a decrease in VMs per minute for the wrist and peak gradient PS post-procedure, and a significant change in wear compliance for the ankle.
Accelerometer-derived measures and clinical outcomes before and after pulmonary valve replacement.
For both ankle and wrist accelerometer wear across pulmonary valve replacement, median participant compliance was greater than 96% (Table 1). Paired median differences in key accelerometry and clinical assessment measures from pre-and post-pulmonary valve replacement were non-significant (Table 1). Sub-analyses isolating cohorts specifically reporting impaired exercise tolerance at baseline similarly showed no significant median differences (n = 9, Table A3) following pulmonary valve replacement.
Accelerometry measures of fraction of time spent in each activity level were highly correlated and all significant, as were measures of movement quality (entropy, jerk, and mean frequency). Most correlations between accelerometry measures and other clinical measures, including indexed right ventricular end-diastolic volume, QRS duration, left ventricular ejection fraction, and peak gradient pulmonary stenosis, were non-significant (Table A4). The correlation, however, between mean frequency and peak gradient pulmonary stenosis was moderately strong and positive (Table A4).
Most mean differences in patient-reported survey items pre-and post-pulmonary valve replacement were nonsignificant. Changes in score on the item “Engagement in Strenuous Activity” were highly variable. There was a significant decrease in the average score for the items “Shortness of Breath with Strenuous Activity,” “Extent Symptoms Impact Daily Activities,” and “Challenges with Activities of Daily Living” (Table 2).
Mean differences in patient-reported questionnaire items (N =18)
Table 2. Long description
The table presents mean differences in patient-reported questionnaire items for 18 patients. It includes 11 rows and 3 columns. The columns are labeled Question, Average score change, and p-Value. The questions cover various aspects such as shortness of breath, engagement in strenuous activity, distance able to walk, and challenges with daily living. Notable trends include a significant decrease in average scores for shortness of breath in strenuous activity, the extent symptoms impact daily activities, and challenges with activities of daily living. The p-values indicate the statistical significance of these changes. Row 1: Shortness of breath on stairs, Average score change: -0.500, p-Value: 0.132. Row 2: Engagement in strenuous activity, Average score change: 0, p-Value: 0.999. Row 3: Shortness of breath in strenuous activity, Average score change: -0.444, p-Value: 0.042. Row 4: Distance (Blocks) able to walk in normal conditions, Average score change: 0.500, p-Value: 0.095. Row 5: Shortness of breath in sedentary activity, Average score change: -0.167, p-Value: 0.331. Row 6: Extent symptoms impact daily activities, Average score change: -0.833, p-Value: 0.014. Row 7: Challenges with self-care, Average score change: -0.111, p-Value: 0.331. Row 8: Challenges with activities of daily living, Average score change: -0.333, p-Value: 0.029. Row 9: Regular pain or discomfort, Average score change: -0.111, p-Value: 0.495. Row 10: Anxiety or depression, Average score change: -0.056, p-Value: 0.668.
All survey items except for distance walked are reverse scored, with lower scores reflecting more affirmative.
Discussion
In this prospectively enrolled single centre investigation of participants undergoing pulmonary valve replacement, evidence of postoperative improvement in exercise tolerance was mixed. We did not find a statistical difference in physical activity following pulmonary valve replacement using accelerometry metrics. Pulmonary valve replacement led to median reductions in peak gradient pulmonary stenosis following the procedure (Table 1); however, this did not confer an increase in quantitative activity measures. We observed 50% of the cohort recorded impaired exertional tolerance as a symptom prior to pulmonary valve replacement with all reporting resolution of this symptom following pulmonary valve replacement. This qualitative improvement is important, despite the lack of quantitative benefit.
Patient-described clinical symptoms are a key feature in the algorithm of determining when to perform pulmonary valve replacement. Reference Geva, Wald and Bucholz26 While exertional symptoms can be related to cardiovascular and non-cardiovascular aetiologies in individuals with CHD, the resolution of subjective symptoms within 6 months of pulmonary valve replacement leads the authors to hypothesise that pulmonary valve replacement was the primary contributor to patient benefit. Quality of life surveys in adults with tetralogy of Fallot, an anatomical population commonly requiring pulmonary valve replacement, demonstrate lower scores in domains related to physical activity compared to healthy controls; therefore, optimisation of subjective symptoms related to physical activity is a key outcome to consider. Reference Zachos, Pappa and Milaras27 A meta-analysis including 48 studies and 3,118 patients with tetralogy of Fallot demonstrated a pooled reduction in New York Heart Association classification following pulmonary valve replacement, providing further evidence of the subjective benefit for postoperative patients. Reference Ferraz Cavalcanti, Sá and Santos28
Despite the evidence of qualitative improvement, literature regarding quantitative improvement in exercise capacity following pulmonary valve replacement has been mixed. Previous studies analysing cardiopulmonary exercise testing data following pulmonary valve replacement showed mixed results with change in participant percent predicted VO2 max and VE/VCO2. Reference Frigiola, Tsang and Bull5,Reference Hasan, Lunze and Chen6,Reference Sabate Rotes, Johnson, Burkhart, Eidem, Allison and Driscoll8,Reference Ho, Schamberger, Hurwitz, Johnson, Sterrett and Ebenroth13,Reference Müller, Engelhardt, Fratz, Eicken, Ewert and Hager14,Reference Geva, Gauvreau and Powell29 In the two studies where participant percent predicted VO2 max significantly increased following pulmonary valve replacement, changes in respiratory exchange ratio were non-significant. Reference Hasan, Lunze and Chen6,Reference Müller, Engelhardt, Fratz, Eicken, Ewert and Hager14 The significant change in percent predicted VO2 max also did not correspond to the significant change in VE/VCO2 across studies. In this study, there were no significant improvements in the accelerometer-measured average fraction of activity time spent in more strenuous activity and total activity counts following pulmonary valve replacement. When considered with previous studies using cardiopulmonary exercise testing, these findings may further support the conclusion that exercise parameters do not significantly improve in individuals following pulmonary valve replacement. These mixed results between qualitative and quantitative measurements, in our own study, present a challenge in understanding if these data reflect a lack of true relationship or a need for more innovative exertional metrics to quantify improvement.
Accelerometer measurements have intrinsic limitations, particularly in data captured using the sensors placed on the ankle. The vector magnitudes per minute for participant ankles were less than participant wrists at both the pre-pulmonary valve replacement and post-pulmonary valve replacement timepoints (Table 1). Given that compliance of both wrist and ankle accelerometers had a median of 96% or greater at both timepoints, and that counts were indexed to the amount of wear time, it is less likely that compliance with the device was a major limitation (Table 1).
The representation of the study’s sample may have been further hindered by its range in baseline participant symptoms and age. Exertional symptoms were reported in 50% of the cohort at baseline survey; therefore, not all individuals were symptomatic prior to pulmonary valve replacement. Perhaps restricting inclusion criteria to those who were symptomatic and increasing sample size may have provided a cohort more likely to demonstrate a measured benefit. While a minimum age of 13 years was set, the 75th percentile of participants analysed was 35 years old (Table A1). Although analyses were paired with participants in order to offset the random effects of this between-subjects variance, participant age may have been a mediating factor in the degree of improvement in exercise tolerance following pulmonary valve replacement. Further study would be required with a larger cohort to explore this potential mediation across specific age subgroups. Finally, limited availability of pre and post-pulmonary valve replacement participant cardiopulmonary exercise testing data hindered the ability to compare changes in accelerometry with other validated clinical measures. Only 6 participants had cardiopulmonary exercise testing data available at baseline, and none of the participants had post-pulmonary valve replacement cardiopulmonary exercise testing data to demonstrate changes. With mixed improvements across studies in cardiopulmonary exercise testing endpoints following pulmonary valve replacement, the availability of participant cardiopulmonary exercise testing data would have been a crucial reference point for the lack of significance in accelerometer-measured changes after pulmonary valve replacement.
Accelerometers have gained popularity in recent years as a novel outcome measure in CHD studies. Reference McCrindle, Williams and Mital19,Reference McKillop, McCrindle, Dimitropoulos and Kovacs20, Reference Voss, Duncombe, Dean, de Souza and Harris30,Reference Zaqout, Vandekerckhove, Michels, Bove, François and De Wolf31 Given the focus on exertional symptoms as an indication for pulmonary valve replacement, Reference Geva, Wald and Bucholz26 these patients, and particularly those with tetralogy of Fallot, have been perceived as a population in which accelerometers may have utility. While our sample size was modest, it was adequate to detect a clinically significant improvement in accelerometer measures following pulmonary valve replacement. Lack of this signal suggests a different modality may better quantify activity changes and correlate with our detected subjective improvement.
Conclusion
Subjective impaired exertional intolerance resolved following pulmonary valve replacement; however, quantitative measure of physical activity by accelerometry did not show significant changes. While physical accelerometry has the potential to serve as a free-living physical capacity measure, our data suggest investigation into alternative quantitative measures may be required given the discrepancy with qualitative data.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/S1047951126113493.
Acknowledgements
The authors would like to thank the participants for their involvement in the study.
Financial support
Dr Sunthankar is supported by a 2018 Burroughs Wellcome Fund Physician-Scientist Institutional Award to Vanderbilt University (1018894). Dr Sunthankar is supported by the National Institute of Child Health and Human Development (NICHD) Maternal and Pediatric Precision in Therapeutics (MPRINT) (P50HD106446). Dr Soslow is supported by the National Heart, Lung, and Blood Institute (K23HL123938, R56HL141248, and R01HL167969 (Soslow)), the Food and Drug Administration (FDA) (1R01FD006649 (Silver Spring, MD) (Soslow)), and National Center for Advancing Translational Science (UL1-TR002243).
Competing interests
Dr Soslow has received a grant from Ametris, LLC. to conduct an observational study, however no associated data was used in development of this manuscript.
Ethical standard
The authors assert that all procedures contributing to this work comply with the ethical standards of the Common Rule and with the Helsinki Declaration of 1975, as revised in 2008, and has been approved by the Vanderbilt Institutional Review Board (IRB 211522).
Appendix
Demographics of cohort included for analysis (N = 18)
Anthropometrics summarised at baseline (pre-PVR).
PVR = pulmonary valve replacement.
Summary of variables analysed from participant clinical measures
Parameters were captured both retrospectively (pre-PVR) and prospectively (post-PVR). Preoperative cardiac MRI data were also collected for participants with available scan, but was excluded from analysis due to limited data capture (n = 8 participants).
LVEF = left ventricular ejection fraction; MRI = magnetic resonance imaging.
Median differences in participants indicated for PVR by exercise intolerance (N = 9)
Median differences in accelerometer-derived measures and clinical outcomes from before and after pulmonary valve replacement for those who exercise intolerance were reported as an indication for PVR. The analysis was stratified by symptom severity, with NYHA class < 3 established as less severe exercise intolerance and ≥ 3 as more severe.
LVEF = left ventricular ejection fraction; MVPA = moderate-to-vigorous physical activity; NYHA = New York Heart Association; PVR = pulmonary valve replacement.
Correlations of key outcome measures (N = 18)
Key for significance: ns: P > 0.05, *: P < 0.05, **: P < 0.01, ***: P < 0.001.
MVPA = moderate-to-vigorous physical activity.
Open access
