Introduction
Cardiac rehabilitation is a validated clinical tool to improve fitness in those with heart disease. Fitness has been defined to include multiple facets including one’s abilities, endurance, strength, and flexibility. In addition to directly measuring various physical fitness outcomes during cardiac rehabilitation, subjective patient questionnaires are often used to assess quality of life, emotional well-being, and perceptions of functional status in daily life. The questionnaires are used to evaluate emotional well-being because a comprehensive cardiac rehabilitation program focuses on both the physical and mental growth of the patient. Reference Gauthier, Curran, O’Neill, Alexander and Rhodes1 One of the questionnaires often used is the Duke Activity Status Index.
The Duke Activity Status Index has been shown to correlate significantly with functional status in adults and those with chronic medical conditions including chronic obstructive pulmonary disease and heart failure. Reference Fan, Lee, Frazier, Lennie and Moser2,Reference Grodin, Hammadah, Fan, Hazen and Tang3,Reference Carter, Holiday, Grothues, Nwasuruba, Stocks and Tiep4 In addition to correlating with functional status, the Duke Activity Status Index has also been shown to predict health-related outcomes in the adult population following cardiac surgeries. Reference Koch, Li, Lauer, Sabik, Starr and Blackstone5 Minimal studies have been published showing the correlation of the Duke Activity Status Index in the paediatric population and specifically in children and young adults with heart disease including CHD. This is important because patients with CHD often have poor perceptions of their fitness. Reference Bredy, Ministeri and Kempny6 Improving these perceptions of what activities they feel they can do, may result in a more positive mindset, potentially removing a barrier to the protective effects of exercise and fitness. Reference Curran, Losi and Pymm7
This study aimed to; (1) assess the ability of the Duke Activity Status Index to correlate with functional measures of fitness in children and young adults with cardiac disease including CHD and (2) evaluate for improvements in the perceived (qualitative questionnaires) and actual fitness measures following completion of cardiac rehabilitation.
Materials and methods
This is a single-site retrospective cohort study assessing for perceived fitness abilities using the Duke Activity Status Index and comparing it to other markers of physical function for those completing cardiac rehabilitation for the first time between September 2016 to September 2022. Data on enrollment dates, number of completed weeks and sessions, and programme completion outcomes were collected. Additional data collected from the chart review included patients’ demographic information and medical diagnoses. Exclusion criteria included missing Duke Activity Status Index data, no history of heart disease, and age less than 8 years old. If an individual completed multiple rounds of cardiac rehabilitation, the later encounters were excluded if the individual had previously completed at least 1 month of a cardiac rehabilitation program, or earlier encounters of less than 30 days were excluded if the individual completed more than 1 month in a cardiac rehabilitation program after that encounter.
Cardiac rehabilitation is a personalised exercise and lifestyle intervention program designed per the standard of the American Association of Cardiovascular and Pulmonary Rehabilitation and has been previously described. Reference Aronoff, Chin and Opotowsky8 Cardiac rehabilitation sessions took place at our hospital’s cardiac rehabilitation facility. The programme included one-on-one supervision by a trained exercise physiologist. Age-appropriate activities were incorporated into sessions to increase motivation and enjoyment during cardiac rehabilitation.
Cardiac rehabilitation outcomes were recorded at the first and last cardiac rehabilitation sessions. These included physical body measurements, performance measures, and psychosocial questionnaires. Physical body measurements including height and weight were recorded. Cardiopulmonary exercise testing was performed on a stationary cycle ergometer (Corival; Lode; Groningen, The Netherlands) with an individualised ramp protocol using breath-by-breath analysis (Ultima CardiO2; MGC Diagnostics; Saint Paul, MN, USA), as previously described. Reference Powell, Wittekind and Alsaied9 Criteria for a maximal effort exercise test were that 2 of the following 3 criteria were met: respiratory exchange ratio ≥1.10; maximal heart rate ≥85% of the age-predicted maximum (220-age in years); or maximal rating of perceived exertion >18 on a 6 to 20 Borg scale. Reference Borg10 Predicted peak oxygen consumption was calculated per Wasserman et al. and Cooper et al. equations. Reference Wasserman, Hansen and Sue11,Reference Cooper, Weiler-Ravell, Whipp and Wasserman12
Additional performance outcomes collected included the number of sit to stand repetitions within 30 s, sit and reach distance, the number of arm curls within 30 s with a constant weight, 6-minute walk test, and handgrip strength. Metabolic equivalents were calculated from the 6-minute walk distance Reference Saba, Goharpey, Attarbashi Moghadam, Salehi and Nejatian13 . A set of patient questionnaires were also collected at the first and last sessions.
The Duke Activity Status Index was designed to be a reliable tool to assess a patient’s physical functional status. Reference Hlatky, Boineau and Higginbotham14 The 12 questions are in a yes-no format and cover a wide range of activities. Scores are weighted based on the difficulty of the task in question with more difficult tasks having a higher score. Higher scores correlate with higher functional status, with a maximum score of 58.2 points. Reference Hlatky, Boineau and Higginbotham14 Question #10 was removed for patients <18 years old as it asks about sexual activity, making a maximum score of 52.95 for patients <18 years old. Normative data for the Duke Activity Status Index does not exist for paediatric patients, but adult studies have shown prognostic significance when the score is less than 26–34. Reference Wijeysundera, Beattie and Hillis15,Reference Mustafaoglu, Demir, Aslan, Sinan, Zeren and Kucukoglu16 Additionally, to account for absolute score differences between paediatric and adult patients, Duke Activity Status Index is also represented as a percentage of expected maximum points (i.e. Duke Activity Status Index % of total). Metabolic equivalents were calculated from the Duke Activity Status Index. Reference Hlatky, Boineau and Higginbotham14
Additional social and emotional surveys were provided and completed before and after cardiac rehabilitation including the Patient Health Questionnaire-9 screen for depression and the Short Form Health Inventory-36, which included the Mental Component Scoring (a combination of the vitality, social functioning, role-emotional, and mental health forms) and the Physical Component Scoring (a combination of the physical function, role-physical, bodily pain, and general health forms). Reference Kroenke, Spitzer and Williams17,Reference Lins and Carvalho18
Statistical analyses
Descriptive statistics are presented as median [interquartile range] for continuous variables and as n (frequency) for categorical variables. Baseline and final cardiac rehabilitation completion data were compared using a paired t-test. Groups were compared at baseline utilising the Student’s t-test for independent groups with equal variances and Welch’s t-test for unequal variances. All t-tests were two-sided with p value <0.05 considered significant. Univariable and multivariable analyses were then completed for the Duke Activity Status Index. Univariable analysis was performed with Pearson’s correlation coefficient. The selection criteria for the multivariable model were a p value of 0.1 for entry and 0.05 to remain in the model. Candidate predictors for multivariable modelling were cardiac rehabilitation physical activity outcomes and included peak oxygen consumption, 6-minute walk distance, arm curls, sit to stand, and dominant handgrip. Statistical analyses were performed using JMP®, Version 16 from SAS Institute Inc. (Cary, NC). Figures were created with Microsoft Word, Excel, and PowerPoint (Redmond, WA).
Results
A total of 199 individual cardiac rehabilitation enrollments were evaluated during the study period (Figure 1). There were 81 enrollments (41%) excluded from this study: 38 for having completed more than 1 month of cardiac rehabilitation program previously, 13 enrollments that were less than 30 days of cardiac rehabilitation but the individual completed longer than 30 days later, 6 for individuals aged less than 8 years of age, 13 for individuals without underlying cardiac disease, and 11 for incomplete Duke Activity Status Index data. No patients included in the analysis had multiple rounds of cardiac rehabilitation. Of the 118 patients (aged 20 years old [13.9–22.5], 53% male) included in the analysis, 33 (28%) completed at least 12 weeks of the programme. There were no significant age or sex differences between those who did and did not complete cardiac rehabilitation. Full demographics for the study are included in Table 1. Of note, 67 of the 118 patients had pre-cardiac rehabilitation maximal effort cardiopulmonary exercise testing included. In the 33 patients who completed at least 12 weeks of the programme, 26 patients completed a maximal effort cardiopulmonary exercise test both before and after cardiac rehabilitation (Table 1).
Flowsheet of inclusion of participants in the study. Not naïve to cardiac rehabilitation was defined as having completed at least 30 days of cardiac rehabilitation prior to starting a new cardiac rehabilitation session. For those who participated in programmes in which less than 30 days were completed, but a session of at least 30 days was completed later during the study period, then the prior sessions of less than 30 days were excluded as well. Of those included, those who completed at least 12 weeks of a cardiac rehabilitation program were assessed for completion of different components of the testing. The functional and subjective outcomes referred to in this flow chart include 6-minute walk test, functional strength assessments, and patient questionnaires.
Represented are the baseline demographics and outcomes for the entire cohort. Additionally, the demographics, baseline and final functional outcomes for the subset who completed the full cardiac rehabilitation program are included
Data are presented as a median[IQR] or absolute number (%). For comparison of the baseline and final cardiac rehabilitation outcomes, a paired t test was performed to compare before and after cardiac rehabilitation outcomes. P < 0.05 was considered significant.
CPET = cardiopulmonary exercise test; M = male; F = female; 2V = 2 ventricle; CHD = congenital heart disease; 1V= single ventricle, CM = cardiomyopathy; HF = heart failure; EP = electrophysiology; PH = pulmonary hypertension; SS = subglottic stenosis; 6MW = 6-minute walk; m = meters; reps=repetitions; cm = centimeters; HG = hand grip; VO2 = oxygen consumption; ml= milliliters; min = minute; kg = kilogram; HR = heart rate; bpm = beats per minute; Duke = Duke Activity Status Index; METS = metabolic equivalents; PHQ-9=Patient Health Questionnaire-9; MCS = mental component scoring; PCS = physical component scoring.
Analysis for the whole cohort includes only data obtained from their pre-cardiac rehabilitation forms and testing as the majority of the cohort did not complete a full cardiac rehabilitation program (Table 1). The median Duke Activity Status Index was 32.6 [21.5–48.8] and the average median Duke metabolic equivalents was 3.4 [2.7–4.5]. There were 58% (69/118) of patients with a baseline Duke Activity Status Score <34 and 41% (48/118) with a Duke Activity Status Score <26. Median peak oxygen consumption was 60.1 [49–72.8] % of predicted, or 20.8 [16–24] ml/kg/min. The entire cohort trended towards having mild depression based on the median Patient Health Questionnaire-9 score of 5.5 [2–8.3]. Reference Kroenke, Spitzer and Williams17
Among the subset who completed cardiac rehabilitation (n = 33), there was a statistically significant increase in Duke Activity Status Index (33.2 [24.2–46.7] versus 41.9 [31.5–53]; p = 0.005) and Duke Activity Status Index % of total (50.9 [45.0–84.1] versus 78.1 [54.3–100.0] %, p = 0.0002) (Supplemental Figure 1). In addition, there was a significant increase in patients who scored >34 (39% [13/33] versus 70% [23/33], p = 0.01), but there was not a significant increase in those who scored>26 (67 [22/33] versus 82% [27/33], p = 0.2) on the Duke Activity Status Index. The median change in Duke Activity Status Index was 11.2 [0–34.4]. On functional assessment, there was a significant increase in 6-minute walk distance (405.4 [310–490] versus 488.7 [383.5–563] m; p < 0.0001), dominant hand grip (11.2 [7.0–16.1] versus 12.9 [7.3–17.8] kg; p = 0.006), sit to stand repetitions (15.5 [12–19.5] versus 21.8 [17–27.5] repetitions; p < 0.0001), arm curls (20.1 [17–23.8] versus 25.1 [19.3–30] repetitions; p < 0.0001), and sit and reach distance (38.1 [29.2–43.2] versus 42.9 [33.850.8] cm; p = 0.005). Though most data from the cardiopulmonary exercise testing suggested a trend towards improvement, no marker was statistically significant. There was improvement in other patient questionnaire forms including the Duke Activity Status Index metabolic equivalents (3.4 [2.9–4.2] versus 3.9 [3.3–4.6]; p = 0.0002), Patient Health Questionnaire-9 (6.2 [3–8.5] versus 4.5 [3.5–6.5]; p = 0.02), MCS (48.6 [44.3–53.7] versus 52.5 [48.5–57.4]; p = 0.003), and PCS (40.8 [31.9–47.9] versus 44.2 [40.7–51.9]; p = 0.003) (Table 1).
The results of Pearson’s correlation analysis for Duke Activity Status Index are summarised in Table 2. Percent of predicted peak oxygen consumption (r = 0.49, p < 0.0001), 6-minute walk distance (r = 0.45, p < 0.0001), 6-minute walk metabolic equivalents (r = 0.45, p < 0.0001), sit to stand repetitions (r = 0.48, p < 0.0001), dominant hand grip (r = 0.48, p < 0.0001), and physical component score (r = 0.56, p < 0.0001) were all associated with Duke Activity Status Index (Figure 2). On multivariable analysis, the percent of predicted peak oxygen consumption and dominant hand grip strength remained associated with the Duke Activity Status Index (Table 2).
Correlations with line of best fit for Duke Activity Status Index percent of predicted normal and 6-minute walk distance (2 A ), percent of predicted peak oxygen consumption (2B ), dominant hand grip (2C ), and PCS (2D ). Univariable analysis was performed with Pearson’s correlation coefficient. p-value<0.05 was considered significant.
Results of both the univariable and multivariable analyses for Duke Activity Status Index percent of total
Univariable analysis was performed with Pearson’s correlation coefficient. Candidate predictors for multivariable modeling included peak oxygen consumption, 6-minute walk distance, arm curls, sit to stand, and dominant handgrip. p value<0.05 was considered significant.
BMI = body mass index; VO2=oxygen consumption; HR = heart rate; 6MW = 6-minute walk; METS = metabolic equivalents; HG = hand grip; MCS = mental component scoring; PCS = physical component scoring; PHQ-9=Patient Health Questionnaire-9.
Duke Activity Status Index was strongly associated with the Duke Activity Status Index metabolic equivalents (r = 0.86, p < 0.0001), and the Duke Activity Status Index metabolic equivalents were associated with the 6-minute walk metabolic equivalents (r = 0.48, p < 0.0001).
There were no associations between the change in Duke Activity Status Index (absolute and percent of predicted) and the objective measures of fitness obtained in this study.
When focused only on patients with CHD (n = 62), the median Duke Activity Status Index was 32.5 [24.2–50.7], and the average median Duke Activity Status Index percent of total was 61.3 [41.6–87.1] %. Median peak oxygen consumption was 64.0 [52.8–74.0] % of predicted, or 18.7 [15.0–24.8] ml/kg/min. The Duke Activity Status Index remained associated with percent of predicted peak oxygen consumption (r = 0.52, p = 0.0003), 6-minute walk distance (r = 0.49, p < 0.0001), sit to stand repetitions (r = 0.49, p < 0.0001), dominant hand grip (r = 0.49, p < 0.0001), and physical component score (r = 0.53, p < 0.0001).
Discussion
This study compared perceived measures of fitness (using the Duke Activity Status Index) to actual measures of fitness in a population of youth and young adults with cardiac disease including many quantitative measures collected during cardiac rehabilitation. The Duke Activity Status Index was associated with 6-minute walk distance, the 6-minute walk test metabolic equivalent changes, and the Duke Activity Status Index metabolic equivalents in youth and young adults with heart disease. Additionally, the Duke Activity Status Index was also independently associated with peak oxygen consumption and maximal dominant hand grip. Lastly, though this cohort had a low cardiac rehabilitation completion rate, multiple measures of actual and perceived fitness increased in those who completed at least 12 weeks of cardiac rehabilitation. That resulted in a 31% increase in the number of patients who scored over 34 on the Duke Activity Status Index with this value having prognostic significance in adults with heart disease. Reference Wijeysundera, Beattie and Hillis15
Cardiac rehabilitation improves objective measures of fitness. Reference Aronoff, Chin and Opotowsky8,Reference Powell, Wittekind and Alsaied9,Reference Opotowsky, Rhodes and Landzberg19,Reference Rhodes, Curran and Camil20 In addition to again demonstrating such objective benefit, this study shows that cardiac rehabilitation also improves subjective perceptions of fitness assessed by the increase in the Duke Activity Status Index. Patients with CHD often have poor perceptions of their fitness. Reference Bredy, Ministeri and Kempny6,Reference Kamphuis, Ottenkamp and Vliegen21 Improving perceptions of fitness is important as youth with positive mindsets tend to have healthier exercise habits and higher levels of fitness have been associated with reduced morbidity and mortality in CHD. Reference Baceviciene, Jankauskiene and Emeljanovas22–Reference Diller, Dimopoulos and Okonko26 Our study showed that the Duke Activity Status Index was associated with multiple measures of fitness and that through exercise therapy both actual and perceived fitness can improve. This differs from previous research that has shown no correlation with subjective measures of fitness and exercise capacity. Reference Burns, Olson, Kazmucha, Balise, Chin and Chin27 Potential explanations for the differences in findings between our cohort and the study from Burns R. et al (2010) include a lower power in our cohort, different subjective questionnaires for each study, and our cohort being an entirely cardiac rehabilitation cohort that generally had quite poor fitness. Additionally, the improvement of the Duke Activity Status Index above previously shown cut-offs may infer prognostic protections; however, this should be confirmed in larger studies. Reference Wijeysundera, Beattie and Hillis15 On the other hand, the improvement in in Duke Activity Status Index may be secondary to either neuromuscular adaptation to exercise or could be a reflection of patient motivation to perform, which may be supported by the lack of improvement in the peak oxygen consumption following cardiac rehabilitation.
As perceptions correlate to actual fitness, mental training should be prioritised as much as physical training for CHD patients in cardiac rehabilitation. Reference Curran, Losi and Pymm7 The patients who have a healthier sense of self are better prepared to have improvements in physical activity, quality of life, and overall physical and mental health. Unfortunately, children and adults with CHD are at increased risk for many different forms of mental health difficulties including anxiety and depression. Reference Gonzalez, Kimbro and Cutitta28–Reference Kovacs, Brouillette and Ibeziako30 This highlights the importance of regular mental health screening and interventions in this population. Cardiac rehabilitation may have a unique role to play in addressing some of these mental health concerns, provided that mental health screening and mental fitness training are integrated into the programme. Reference Denniss, Sholler, Costa, Winlaw and Kasparian31,Reference Wang, Hay, Clarke and Menahem32 This is supported in our cohort by the significant improvements in the Patient Health Questionnaire-9 screen and the Mental Component Score following cardiac rehabilitation. These improvements could possibly be even greater should there be further integration of psychological services with the cardiac rehabilitation program, and this should be evaluated in future studies.
The Duke Activity Status Index is validated in adult heart failure patients. Reference Fan, Lee, Frazier, Lennie and Moser2–Reference Carter, Holiday, Grothues, Nwasuruba, Stocks and Tiep4,Reference Mustafaoglu, Demir, Aslan, Sinan, Zeren and Kucukoglu16 This study shows that the questionnaire can also be informative for children and young adults with heart disease. It correlates with 6-minute walk distance and 6-minute walk metabolic equivalent just like in other non-CHD populations. Reference Mustafaoglu, Demir, Aslan, Sinan, Zeren and Kucukoglu16,Reference Bagur, Rodés-Cabau and Dumont33,Reference Hamilton and Haennel34 It also correlates with indexed peak oxygen consumption similar to other studies that have shown an association with peak oxygen consumption. Reference Carter, Holiday, Grothues, Nwasuruba, Stocks and Tiep4,Reference Mustafaoglu, Demir, Aslan, Sinan, Zeren and Kucukoglu16,Reference Bagur, Rodés-Cabau and Dumont33 There were associations with the Duke Activity Status Index and maximum handgrip seen in our cohort, providing even more evidence that this fitness perception questionnaire correlates with multiple other aspects of actual fitness. This survey has been advocated to be used to risk-stratify adult patients before cardiac surgery. Reference Mustafaoglu, Demir, Aslan, Sinan, Zeren and Kucukoglu16 As this survey is quick and easy to administer, there may be a role in administering the Duke Activity Status Index as a screener for low fitness in heart disease clinics. Should a patient score abnormally low, that could then in turn trigger further investigations (such as a cardiopulmonary exercise test) and possibly even referrals to cardiac rehabilitation.
There were several limitations to this study other than the inherent limitations of a single-site retrospective cohort. This is a selected sample in which there was a minority of patients who completed a full cardiac rehabilitation program and also underwent cardiopulmonary exercise testing. Those who complete cardiac rehabilitation and return for exercise testing are likely different from those who do not, thus introducing potential sampling bias. Another limitation includes no assessment of activity level outside of cardiac rehabilitation, as no exercise prescriptions or accelerometers were given in this study, so patients may have had different levels of exercise stimuli. Future studies could assess whether the Duke Activity Index Status is associated with future morbidity and mortality in this population of children and young adults with heart disease. Lastly, the Duke Activity Status Index is a validated instrument in adults but has not been validated in paediatric populations nor has it been validated when the question regarding sexual performance is removed. This should be researched in future studies.
In conclusion, cardiac rehabilitation improves perceptions of fitness as well as objective measures of fitness in children and young adults with heart disease. The Duke Activity Status Index is associated with multiple functional outcomes including 6-minute walk distance, hand grip, and peak oxygen consumption, and may help assess perceptions of fitness in those youth and young adults with heart disease.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/S1047951124025939.
Acknowledgements
The authors would like to acknowledge The Heart Institute and the Cardiopulmonary Exercise Laboratory at Cincinnati Children’s Hospital for their support in this project.
Financial support
This research received no specific grant from any funding agency, commercial, or not-for-profit sectors. AWP and ARO were supported by the Heart Institute Research Core at Cincinnati Children’s Hospital.
Competing interests
The authors have no relevant financial or non-financial interest to disclose.
Ethical standard
The authors assert that all procedures contributing to this work comply with the ethical standards of the relevant national guidelines on human experimentation and with the Helsinki Declaration of 1975, as revised in 2008, and has been approved by institutional committees (Institutional Review Board at Cincinnati Children’s Hospital Medical Center).
Open access
