1 INTRODUCTION

Cardiac rehabilitation is a multidisciplinary program recommended for patients with ST and non-ST-segment elevation acute coronary syndromes (ACS) and chronic coronary syndromes (class I, level of evidence A).1-4 Despite the proven benefit, the attendance rate for these programs is 34%, according to the EUROSPIRE V registry.5 This situation has worsened as a result of the COVID-19 pandemic.6 Cardiac telerehabilitation (CTR) has shown to be at least equally beneficial and cost-effective than center-based cardiac rehabilitation (CBCR).7-9

We developed a cardiac telerehabilitation system called Cardioplan, which consists of a web platform and a smartphone application that allows prolonged telemonitoring follow-up after minimal patient training. To validate this approach, a clinical trial was designed to compare a 10-month program of telerehabilitation with a conventional 8-week program of hospital cardiac rehabilitation. In addition to testing a previously unevaluated strategy, this study was conducted during a period of lockdown and restricted social mobility due to the COVID-19 pandemic and provides additional insight on what telerehabilitation can offer in this situation.

2 METHODS 2.1 Study design

This is a randomized controlled trial, with an intervention group that followed a CTR program in Hospital Arnau de Vilanova, and a control group that followed a CBCR program in University Hospital La Fe.

Patients with ACS willing to participate were randomly assigned (1:1) to either CTR or CBCR. Randomization was performed by an independent statistician. Seven blocks were randomly selected and the assignments were stored in seven packages of 10 closed envelopes. These packages were handed over to the principal investigator, who only opened them after recruitment of a new participant.

Each participant signed an informed consent form before participation and was notified of their group allocation after completing their baseline exercise test after 12 days of hospital discharge. Baseline measurements were done at this moment. All assessments were completed at the hospital Arnau de Vilanova. Patients were asked not to disclose their trial group to the faculty, including the investigators involved in questionnaire administration and cardiopulmonary stress testing (CPET) evaluations. Visits to primary care physicians and corresponding specialists were not impacted by the trial.

The trial was conducted according to the ethical principles of the Declaration of Helsinki. The study was approved by the ethics committee of Hospital Arnau de Vilanova and the Spanish Agency of Medicines and Medical Devices (484/14/EC) (Supporting Information Material 1). Each participant signed an informed consent form prior to before participation. This report follows the CONSORT-EHEALTH guidelines (Supporting Information Material 2).10

2.2 Study population

Patients were recruited by face-to-face assessment at discharge after ACS between May 28, 2019 and March 10, 2020, when recruitment was stopped due to the COVID-19 pandemic. Patient age was limited to 18–72 years old. All included patients had to meet low-risk criteria, left ventricular ejection fraction ≥ 50%, and have minimum smartphone usage skills. The main exclusion criteria were reduced mobility, pulmonary diseases, neoplasms, or cognitive impairment.

2.3 Control and experimental rehabilitation procedures

Both groups were given the same education. The target heart rate during exercise sessions was 60%–80% of the heart rate reserve based on the baseline treadmill test. During follow-up, patients were instructed to engage in recommended moderate physical activity guided by Borg's rating of perceived exertion scale of 12–14 (6–20 scale), as well as strength exercises twice a week. Warm-up, stretching, and resistance-band exercises were included in both groups.

2.4 Telerehabilitation group

A portion of hospital training, comprising 2 weeks with four supervised sessions of exercise, was completed. Physical activity consisted of walking down a corridor, adjusting their pace to attain a target heart rate as measured by their smartphone and heart rate monitor (Polar H7). The smartphone application guided participants through a daily exercise and data entry program for 10 months.

2.5 CBCR group

The CBCR program comprised 2 months of treatment with 16 sessions of supervised exercise. Physical activity consisted of routine workouts and aerobic cycling training.

2.6 Description of the comprehensive monitoring system The system, designed in cooperation with Trilema Salud (Valencia, Spain), consists of the following elements: 1.

A webpage that allowed personalized healthcare and tracking of patient adherence to recommendations with password-protected access. The healthcare team monitored seven variables based on a traffic light color code and communicated with patients if necessary.

2.

A smartphone application that allowed daily scheduling of exercise sessions; recording of subjective general condition, vital signs, and medication adherence; and gave a recommended diet. The exercise module tracked and recorded every exercise session and provided access to warm-up and stretching videos, a virtual educational classroom, and suggested websites. Access was password-protected to ensure confidentiality. Technical assistance in the case of sensor/system failure was provided through a call center.

2.7 Outcome measures

The primary outcome was an increase in self-reported physical activity in MET-min/week, derived from the IPAQ questionnaire. The main secondary outcome was an increase in the VO2max. Additional secondary outcomes included other CPET measurements, changes in laboratory parameters, anthropometric variables, adherence to the rehabilitation program, returning to work, adherence to a Mediterranean diet, psychological well-being, health-related quality of life, and smoking cessation.

2.8 Cardiopulmonary stress test

Symptom-limited CPET was performed after hospital discharge and at 10 months. Stress testing was based on a Bruce protocol using an ergospirometer (Jaeger, MS-CPX). Heart rate, blood pressure, 12-lead ECG, and breath-by-breath respiratory gas analyses were recorded. The test was assumed to be maximal in case of a respiratory gas exchange ratio (RER) > 1.1 or cardiac heart rate >85% of the maximal predicted heart rate. VO2 was defined as the maximal oxygen uptake during the final 30 s of the test. The final CPET was performed with or without beta-blockers, depending on their use at baseline.

2.9 Blood tests and lipid measurements

Blood samples were drawn at baseline and at 4 and 10 months follow-up and were subsequently analyzed in the hospital laboratory for a standard panel. Medications other than lipid-lowering therapy were allowed to be modified during the study period. If LDL cholesterol was above 100 mg/dl at 4 months, the treatment was modified, and the patient was excluded from the lipid substudy.

2.10 Body composition and waist circumference

Measurements at baseline, after 4 months of follow-up, and after 10 months of follow-up, including weight and visceral fat were assessed using a Tanita BC-602 scale, Japan. Waist circumference was measured midway between the costal border and the iliac crest.

3 QUESTIONNAIRES

Questionnaires were administered at baseline and at 10 months through computer-assisted face-to-face interviews to obtain higher response rates. The interviewer did not know the patient's rehabilitation group. The International Physical Activity Questionnaire (IPAQ) consists of seven questions about physical activity (intense, moderate, or walking) in the last 7 days, as well as the time spent sitting on a weekday. The MET-minute/week is calculated by multiplying the value of the exercise level (3.3, 4, or 8) by the duration in minutes of the daily activity and by the number of days per week that it is performed. The level of physical exercise can be categorized as low, moderate, or high.11, 12

The Prevention with Mediterranean Diet (PREDIMED) questionnaire includes 14 items. A score of 9 or more reflects good adherence.13

The Hospital Anxiety and Depression Scale (HADS) consists of 14 items divided into anxiety and depression subscales. The reference period was the previous week.14

The EQ-5D-5L questionnaire consists of two parts. The first part measures five aspects of health. The second part is a scale from 0 (worst state of health) to 100 (best state of health).15

3.1 Sample size

Based on the potential impact of the intervention, a greater increase in total MET-min/week derived from the IPAQ questionnaire in the experimental group was considered plausible. This gave rise to expected mean values 25% higher at the end of rehabilitation for the experimental group (5000 MET-min/week vs. 4000 for the CBCR group) (SD in both groups was considered equal to 1000).16, 17 A total of 30 patients in each group was calculated to provide 95% power at the 5% level of significance to detect a statistically significant difference between groups using the Mann–Whitney U test. A 12% loss to follow-up was estimated; therefore, 70 patients were planned to be enrolled in the study.

3.2 Statistical analysis

Outcomes were analyzed according to the intention-to-treat principle. To analyze treatment effects within groups (at 4 months or 10 months), we used the McNemar–Bowker test of symmetry for qualitative variables (McNemar test for dichotomous variables) or Student's t-test for paired samples for quantitative variables (Wilcoxon signed-rank test when parametric assumptions could not be assumed). The comparison of treatments between groups was carried out using Pearson χ2 test for qualitative variables (Fisher exact test for dichotomous variable) or Student's t-test for independent samples for quantitative variables (Mann–Whitney U test when parametric assumptions could not be assumed). The relationship between two variables was assessed by Pearson's correlation coefficient (Spearman's rank correlation coefficient when parametric assumptions could not be assumed). Two-sided exact p values were calculated whenever possible, and p ≤ 0.05 were considered statistically significant. Data were analyzed using IBM SPSS Statistics 22 and R 4.0.2 for Microsoft Windows.

4 RESULTS 4.1 Patients and program adherence

A total of 67 patients were enrolled in the study, but only 59 were included in the intention-to-treat analysis: 31 and 28 were randomized to the CTR and CBCR groups, respectively (CONSORT Flow chart, Figure 1). There were no significant differences between the groups at baseline (Table 1).

CONSORT patient flow diagram for nonpharmacologic treatment trials

Table 1. Baseline demographics, clinical characteristics, cardiopulmonary exercise testing parameters, and medication use CTR group (n = 31) CBCR group (n = 28) p Demographic characteristics Age (years), mean (SD) 57.5 (9.0) 54.7 (9.9) .266 Male, n (%) 27 (87.1%) 27 (96.4%) .356 Weight (kg), mean (SD) 81.6 (15.9) 84.5 (13.0) .447 Visceral fat (%), median (IQR) 12 (10–13) 12 (10–14) .663 BMI (kg/m2), mean (SD) 27.0 (4.2) 27.7 (3.5) .158 Waist circumference (cm), mean (SD) 98.4 (13.4) 101.3 (9.1) .357 HbA1c (%), mean (SD) 5.89 (0.46) 5.72 (0.55) .235 Total cholesterol (mg/dl), mean (SD) 127.1 (32.4) 120.6 (20.6) .389 HDL cholesterol (mg/dl), mean (SD) 46.6 (12.1) 46.0 (10.1) >.800 LDL cholesterol (mg/dl), mean (SD) 59.3 (27.1) 54.2 (16.8) .565 Triglycerides (mg/dl), median (IQR) 91 (77–121) 84 (72–118) .532 ApoB/apoA-I ratio, mean (SD) 0.62 (0.28) 0.56 (0.20) .381 LP(a) (mg/dl), median (IQR) 44 (11–89) 31 (17–64) .388 IPAQ questionnaire Total MET-min/week median (IQR) 1251 (693–2624) 1502 (896–3924) .186 Cardiopulmonary exercise testing HRmax (bpm), mean (SD) 136.5 (18.7) 137.5 (20.1) >.800 Exercise time (min), mean (SD) 7.14 (2.29) 7.69 (2.21) .354 RERmax, mean (SD) 1.17 (0.08) 1.22 (0.10) .049 VO2max (ml (kg min)), mean (SD) 23.94 (4.64) 23.68 (4.48) >.800 Cardiovascular risk factors Total (number), median (IQR) 2 (1–2) 2 (1–3) .500 Hypertension, n (%) 16 (51.6%) 13 (46.4%) .796 Smoking, n (%) 14 (45.2%) 18 (64.3%) .193 Dyslipidaemia, n (%) 14 (45.2%) 14 (50.0%) .797 Diabetes mellitus, n (%) 6 (19.4%) 6 (21.4%) >.800 Acute coronary syndrome .763 UA, n (%) 9 (29.0%) 6 (21.4%) NSTEMI, n (%) 9 (29.0%) 10 (35.7%) STEMI, n (%) 13 (41.9%) 12 (42.9%) Medication Dual antiplatelet therapy, % 30 (96.8%) 27 (96.4%) >.800 Statins, n (%) 31 (100.0%) 26 (92.9%) .221 B-blockers, % 20 (64.5%) 22 (78.6%) .264 ACE inhibitors/ARBs, % 22 (71.0%) 20 (71.4%) >.800 Systolic function LVEF (%), mean (SD) 62.6 (7.4) 62.0 (6.4) .709 Coronary arteries Coronary lesions, median (IQR) 1 (1–2) 1 (1–1) .225 Number of stents, median (IQR) 1 (1–2) 1 (1–1) .222 Abbreviations: BMI, body mass index; HbA1c, glycosylated hemoglobin; IQR, interquartile range; LVEF, left ventricular ejection fraction; NSTEMI, non-ST elevation myocardial infarction; RER, respiratory exchange ratio; STEMI, ST elevation myocardial infarction; UA, unstable angina; VO2max, maximal oxygen uptake.

Patients with at least one training session for both groups were included in the follow-up period. Eight patients (12%) were excluded: six in the CBCR group and two in the CTR group. A total of seven women were randomized (10.4%), four in the CTR group and three in the CBCR. All women in the CTR group attended the sessions, but one declined follow-up. In the CBCR group, two women did not start rehabilitation, and the third completed only three sessions. During the study period, three patients were readmitted. In the CBCR group, one patient with postinfarction angina received stenting in a new vessel. Two patients in the CTR group with unstable angina had known distal lesions in the stented vessel and were not amenable to revascularization. One patient was excluded due to refractory angina.

4.2 Primary outcome

The increase in total physical activity was significantly higher in the CTR group than in the CBCR group (median increase of 1726 METS-min/week for CTR vs. 636 METS-min/week for CBCR group, p = .045) (Figure 2). The correlation between MET-min/week and VO2max before and after rehabilitation was ρ = 0.224 (p = .091) and 0.311 (p = .018), respectively.

Changes in self-reported physical activity through the International Physical Activity Questionnaire (IPAQ) (Panel A). Changes in maximal oxygen uptake (VO2max) during cardiopulmonary exercise testing (Panel B). CBCR, center-based cardiac rehabilitation; CTR, cardiac telerehabilitation

4.3 Secondary outcomes 4.3.1 Cardiopulmonary exercise testing

In the CTR group, VO2max mean increase from baseline was 1.62 ml/(kg min) (95% CI: 0.56–2.69, p = .004) and in the CBCR group it was 0.60 ml/(kg min) (95% CI: −0.83 to 2.03, p = .40) (Figure 2). The most relevant data can be observed in the Table 2.

Table 2. Physical activity through the International Physical Activity Questionnaire (IPAQ) and cardiopulmonary exercise test (CPET) parameters at baseline and after the 10-month follow-up period CTR group (n = 30) CBCR group (n = 28) Baseline Final p1 Baseline Final p2 p12 IPAQ Total MET-min/week, m (IQR) 1251 (693–2624) 4031 (1875–5973) 0.005 1502 (896–3924) 2420 (1391–4997) 0.374 0.045 Walking, m (IQR) 743 (231–1386) 1287 (990–2079) <0.001 644 (330–1386) 1172 (743–1386) 0.070 0.350 Moderate activity, m (IQR) 140 (0–720) 580 (240–2400) 0.033 440 (0–2700) 540 (0–1020) 0.548 0.039 Vigorous activity, m (IQR) 0 (0–320) 620 (0–2400) 0.053 0 (0–240) 0 (0–1960) 0.313 0.484 Hours a day sitting, m (IQR) 7 (5–8) 6 (4–7) 0.147 6 (5–9) 7 (6–8) >0.800 0.215 Weekly energy expenditure, m (IQR) 1524 (773–3892) 5445 (2905–8772) 0.002 2207 (1334–8505) 3349 (1956–7491) 0.767 0.036 Effort level <0.001 0.343 0.031a Inactive/low, n (%) 6 (20.0%) 1 (3.3%) 6 (21.4%) 3 (10.7%) Moderate, n (%) 17 (56.7%) 6 (20.0%) 12 (42.9%) 13 (46.4%) High, n (%) 7 (23.3%) 23 (76.7%) 10 (35.7%) 12 (42.9%) CPET VO2max (ml/(kg min), mean (SD) (n1 = 31, n2 = 28) 23.9 (4.6) 25.6 (5.6) 0.004 23.7 (4.5) 24.3 (5.5) 0.398 0.240 VO2max (%), mean (SD) (n1 = 31, n2 = 28) 89.6 (12.2) 96.3 (15.5) 0.001 85.4 (15.5) 88.3 (17.8) 0.216 0.174 METS, mean (SD) (n1 = 31, n2 = 28) 6.76 (1.26) 7.29 (1.59) 0.001 6.88 (1.31) 6.98 (1.54) 0.698 0.131 VEmax (l/min), mean (SD) (n1 = 31, n2 = 28) 72.6 (18.0) 79.4 (20.6) 0.010 75.1 (21.9) 76.2 (22.1) 0.701 0.138 RERmax, mean (SD) (n1 = 31, n2 = 28) 1.17 (0.08) 1.20 (0.09) 0.091 1.22 (0.10) 1.18 (0.11) 0.131 0.030 Heart rate max (bpm), mean (SD) (n1 = 31, n2 = 28) 136.5 (18.7) 145.0 (17.2) <0.001 137.5 (20.1) 141.0 (16.2) 0.186 0.130 Heart rate max (%), mean (SD) (n1 = 31, n2 = 28) 84.2 (10.4) 90.0 (9.6) <0.001 83.6 (11.3) 85.4 (9.7) 0.192 0.035 Blood pressure max (mmHg), mean (SD) (n1 = 31, n2 = 28) 166.8 (21.7) 166.3 (27.8) >0.800 169.6 (27.7) 161.3 (25.4) 0.124 0.301 Cardiac output max (l/min), mean (SD) (n1 = 31, n2 = 28) 11.9 (2.8) 12.6 (3.2) 0.010 12.3 (2.8) 12.8 (3.1) 0.128 0.779 BRmax (b/min), mean (SD) (n1 = 29, n2 = 26) 24.8 (19.0) 26.4 (19.5) 0.655 23.5 (19.8) 32.4 (16.7) 0.028 0.180 Effort time (min), mean (SD) (n1 = 31, n2 = 28) 7.14 (2.29) 8.59 (2.60) <0.001 7.69 (2.21) 8.18 (2.32) 0.273 0.055 Treadmill speed (km/h), m(IQR) (n1 = 31, n2 = 28) 5.4 (4.0–6.6) 6.7 (5.4–6.7) <0.001 5.4 (5.3–6.7) 5.5 (5.4–6.7) 0.154 0.019 Elevation (%), m(IQR) (n1 = 31, n2 = 28) 14 (12–16) 16 (12–18) <0.001 14 (12–18) 14 (12–18) 0.218 0.006 Abbreviations: BR, breathing reserve; IQR, interquartile range; m, median; n1(n2), the sample size for the effect analysis in the CTR (CBCR) group; p1 (p2), change in the CTR (CBCR) group; p12, comparison of changes between the two groups; RER, respiratory exchange ratio; VEmax, maximal ventilation; VO2max (%), percentage of theoretical maximal oxygen uptake; VO2max, maximal oxygen uptake. 5 QUESTIONNAIRES

From the IPAQ questionnaire, the percentage of patients who reported a high level of effort at the end of the study period was significantly higher in the CTR group than in the CBCR group (76.7% vs. 42.9%, p = .031) (Table 2).

The HADS global score improved significantly from baseline in both groups, but the improvement was significantly greater in the CTR group than in the CBCR group (p = .015). The anxiety subscale showed a significantly greater effect in the CTR group (p = .006), whereas the depression subscale only improved significantly in the CTR group (p = .020).

The PREDIMED score improved significantly from baseline in both groups, with no differences between groups (p = .345). The percentage of patients reporting high adherenc