Original Investigation
Promotion of Home-Based Exercise Training as Secondary
Prevention of Coronary Heart Disease
A PILOT WEB-BASED INTERVENTION
Anna Torri, MD; Claudia Panzarino, MSc; Anna Scaglione, MD; Maddalena Modica, MPsych;
Bruno Bordoni, BPT; Raffaella Redaelli, MEng; Renata De Maria, MD; Maurizio Ferratini, MD
Background: Although cardiac rehabilitation (CR) is cost-
effective in improving the health of patients with coronary heart
disease (CHD), less than half of eligible CHD patients attend a
CR program. Innovative web-based technologies might improve
CR delivery and utilization. We assessed the feasibility and im-
pact on functional capacity and secondary prevention targets
of a long-term web-monitored exercise-based CR maintenance
program.
Methods: Low- to moderate-risk CHD patients were recruit-
ed at discharge from inpatient CR after a coronary event or
revascularization. We developed an interactive web-based plat-
form for secure home individual access control, monitoring,
and validation of exercise training. Of 86 eligible patients, 26
consented to participate in the study intervention (IG). Using a
quasi-experimental design, we recruited in parallel 27 eligible
patients, unavailable for regular web monitoring, who consented
to a follow-up visit as usual care (UC).
Results: Among IG, active daily data transmission was 100%
during month 1, 88% at month 3, and 81% at 6 months, with
sustained improvement in self-reported physical activity begin-
ning with the first week after discharge from inpatient CR (2467
[1854-3554] MET-min/wk) to month 3 (3411 [1981-5347]
MET-min/wk, P = .019). Both groups showed favorable chang-
es over time in lipid profile, ventricular function, distance walked
in 6 min, and quality of life. At 6 mo, IG achieved a significantly
higher proportion of cardiovascular risk factor targets than UC
(75 ± 20% vs 59 ± 30%, P = .029).
Conclusions: Our web-based home CR maintenance program
was feasible, well-accepted, and effective in improving physical
activity during 6 mo and achieved higher overall adherence to
cardiovascular risk targets than UC.
Key Words: cardiac rehabilitation • exercise training • secondary
prevention • web-based technology
C ardiac rehabilitation (CR) is an integral part of the
management of patients with coronary heart disease
(CHD) and has shown to be cost-effective in improving
physical and psychological health.1,2 Cardiac rehabilita-
tion involves a variety of components: clinical assessment,
Author Affiliation: Cardiac Rehabilitation Unit, IRCCS Santa Maria
Nascente, Fondazione Don Carlo Gnocchi, Milan, Italy.
Supplemental digital content is available for this article. Direct URL citation
appears in the printed text and is provided in the HTML and PDF versions of
this article on the journal’s Web site (www.jcrpjournal.com).
The authors declare no conflicts of interest.
Correspondence: Maurizio Ferratini, MD, Cardiac Rehabilitation Unit,
IRCCS Santa Maria Nascente, Fondazione Don Carlo Gnocchi, Via
Capecelatro 66, 20148 Milan, Italy (mferratini@dongnocci.it).
Copyright © 2018 Wolters Kluwer Health, Inc. All rights reserved.
DOI: 10.1097/HCR.0000000000000316
www.jcrpjournal.com
Copyright © 2018 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
exercise prescription, and patient education to achieve car-
diovascular risk targets.3 In CHD patients, CR improves ex-
ercise tolerance, hemodynamics, autonomic balance, anxi-
ety, depression, smoking, lipid profile, and quality of life.3,4
Exercise is an essential component of CR. Exercise-based
CR programs are safe5,6 and reduce morbidity and mortality
rates in patients with cardiovascular disease,4 while increas-
ing exercise capacity and fitness. Unfortunately, less than
half of eligible patients attend outpatient CR7 for a variety
of reasons: poor health status, transportation and parking
problems, familial and work commitments for younger
patients,8 advanced age, female gender,9 educational level,
socioeconomic status, psychological disturbances,10 and
lack of motivation. While formal exercise training may be
perceived as obtrusive, increased awareness of energy ex-
penditure linked to daily physical activity may facilitate the
incorporation of exercise and adoption of healthy practices
in CHD patients’ lives
Home CR protocols show comparable effectiveness to
hospital-based CR;11 no difference in overall or domain
quality of life and health care costs and superior adher-
ence in the home-based setting was found in 4 of 14 trials.1
Home programs may favor continued exercise, but lack of
control of volume, type, and intensity may undermine train-
ing effectiveness.
The need for innovative approaches to improve the de-
livery and utilization of CR, by devising different options
to suit patients’ preferences and needs, has been under-
scored.12,13 Web-based technologies are innovative tools for
training supervision and continuing reinforcement of CR
goals. We aimed to assess the feasibility and potential scal-
ability to a larger setting of a long-term, web-delivered, and
web-monitored exercise-based CR maintenance program
(CRMP) implemented after discharge from inpatient CR
and its impact on functional capacity and secondary pre-
vention targets as compared to usual care (UC).
METHODS
The study protocol was approved by the institutional Eth-
ics Committee of Fondazione Don Carlo Gnocchi and con-
formed to the principles set in the Declaration of Helsinki.
Written informed consent was obtained from all partici-
pants prior to data collection.
PATIENTS
The study was conducted at the Santa Maria Nascente Car-
diac Rehabilitation Program, where patients were admitted
for inpatient CR (average stay 3 wk) after surgical or percu-
taneous coronary revascularization or acute ischemic events
(myocardial infarction, heart failure).
Web-Based Home Exercise Training for CHD Secondary Prevention 1
 From October 2011 to January 2014, we consecutively
recruited patients discharged from our inpatient CR unit
who met the following inclusion criteria: <85 years old;
CHD diagnosis; low- to moderate-risk profile defined by
the absence of angina/ischemia on exertion; severe left
ventricular dysfunction or major ventricular arrhythmias;
living in the Lombardy region; and a personal computer
and internet connection available at home. Exclusion cri-
teria were uncontrolled hypertension (systolic blood pres-
sure >200 mm Hg; diastolic blood pressure >110 mm Hg),
severe valvular heart disease, arrhythmias or uncontrolled
angina, New York Heart Association (NYHA) classes III-
IV, uncontrolled diabetes, and neuromuscular or musculo-
skeletal issues limiting ability to exercise.
STUDY DESIGN
On admission to inpatient CR, patients performed a 6-min
walk test (6MWT) and received formal education on the
importance of physical exercise, risk factor target goals,
and the effects and schedule of prescribed drugs. During the
inpatient stay, patients underwent 30-min aerobic training
sessions twice daily using a stationary cycle or treadmill,
20 min of resistance training/calisthenics exercises on alter-
nate days using dumbbells and resistance bands, and used
breathing aids to improve ventilatory capacity.
Before discharge, all patients underwent anthropometric
assessment (body mass index and waist circumference), vi-
tal measurements (blood pressure, heart rate), blood lipids
and glucose profile and echocardiography (left ventricular
volumes and ejection fraction), performed a 6MWT, and
completed a general quality-of-life questionnaire (36-item
Short Form Health Survey [SF-36]).14
Upon discharge from inpatient CR, all patients were ad-
vised to continue the exercise program used during the inpa-
tient stay and to perform 30 to 40 min of aerobic exercise,
at an intensity corresponding to a light to moderately hard
perceived exertion at least 3 d/wk and resistance training ses-
sions of 20 min at 70% to 80% of maximal workload based
on 1 repetition maximum (ie, maximal amount of weight
that could be lifted for 1 repetition) using dumbbells or
stretch bands. We prescribed 3 sets of 4 exercises with 10-12
repetitions of each exercise to strengthen the major muscle
groups: dumbbell curls and lateral raises for upper body and
back squats and stiff leg deadlifts for lower body training.15
Overall, 86 CHD patients (33%) met eligibility criteria.
Reasons for exclusion were clinical contraindications to ex-
ercise, “oldest old” (25%), no computer or internet access
(23%) and expected poor compliance to follow-up due to
transportation problems (18%). Among eligible patients,
26 accepted to engage in regular web-monitoring of home
exercise and were recruited in the intervention group (IG).
Using a quasi-experimental design, we enrolled 27 patients
who declined regular web-monitoring but agreed to under-
go a clinical visit and repeat testing after 6 mo and were the
UC group (Figure 1).
STUDY INTERVENTION
The engineering unit at our institution prepared an inter-
active web-based infrastructure to allow individual access,
monitoring, and validation of exercise training performed
by participants at home. The website presented general in-
formation and educational materials on the importance of
exercise training, dietary regimen, and medication adher-
ence for patients and their caregivers.
Before inpatient CR discharge, IG received person-
al log-in user identification and password and they were
trained to use the interactive platform for physical activity
2 Journal of Cardiopulmonary Rehabilitation and Prevention 2018;00:1-6
Copyright © 2018 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Figure 1. Study flow chart. Abbreviation: CR, cardiac rehabilitation.
self-assessment and communication with health care per-
sonnel for 6 mo. They were invited to record and transmit
the type and amount of daily physical activity, which was
converted by the software to MET equivalents using ded-
icated tables (see Supplemental Digital Content Figure 1,
available at: http://links.lww.com/JCRP/A66). MET-min/wk
was calculated by the equation (MET-hr) × (min/60).
Attending physiotherapists checked weekly compliance
with medical prescriptions and workload performed, vali-
dated patient-recorded physical activity, and made custom-
ized feedback phone calls to adjust the training program,
as needed. IG patients received a 30-d post-inpatient CR
discharge virtual visit through web consultation with a car-
diologist, with measurement of body weight, waist circum-
ference, blood pressure, and medication reconciliation and
a further assessment at 90 d to present blood glucose and
lipid profile results.
The UC group was advised to continue the exercise train-
ing program learned during the inpatient stay and was in-
vited to a follow-up visit. Six months after inpatient CR dis-
charge, both groups were recalled to our outpatient clinic to
repeat postdischarge assessments.
STUDY ENDPOINTS
The primary endpoint was change in physical activity in
the IG during the follow-up versus baseline, expressed as
differences between mean MET-min/wk in the week before
month 3 versus the week after discharge (baseline). Second-
ary endpoints included feasibility, efficacy, and safety of the
home exercise training program.
FEASIBILITY
Feasibility was reported as the proportion of patients who
completed at least 80% of the prescribed web-based CR
maintenance program with the proportion determined by
the number of patients who accessed the study website at
1, 3, and 6 mo.
EFFICACY
Efficacy was assessed by the differences in outcomes between
the IG and UC groups at 6 mo versus baseline values. Con-
trol of the cardiovascular risk factor profile was determined
as achievement of target values at 6 mo for smoking (no
smoking); hypertension (blood pressure <140/90 mm Hg);
lipids (low-density lipoprotein [LDL] cholesterol <100
mg/dL); body mass index (<25 kg/m2 or a decrease >5%
www.jcrpjournal.com
from baseline); and glucose control for patients with dia- CHANGES IN IG AND UC
betes (HbA1c<53 mmol/mol). Overall achievement of a Changes from baseline to the 6-mo follow-up visit for pa-
healthy risk factor profile was calculated as the ratio of tar- tients in both groups are summarized in Table 3. Favorable
get values achieved divided by all applicable targets (4 for changes over time were found in both groups for lipid pro-
patients without diabetes and 5 for patients with diabetes) file, with decreased triglycerides and increased high-density
and expressed as a percentage. An additional outcome to be lipoprotein cholesterol; a smaller increase in LDL levels was
assessed was the change in submaximal exercise tolerance as observed in IG compared to UC. Conversely, body mass in-
determined by the change in meters walked for the 6MWT. dex, systolic blood pressure, glucose, and HbA1c levels in-
creased significantly in both groups.
SAFETY The proportion of patients who achieved recommended
Safety of the home program was determined by the number cardiovascular risk targets for smoking, blood pressure,
of major adverse cardiac events (MACEs) and additional and LDL cholesterol at 6 mo was greater in the IG group
medical clinic visits that occurred during the study. than in the UC group, but the differences were not signifi-
cant (Figure 3). Overall control of cardiovascular risk fac-
STATISTICAL ANALYSIS tors, expressed as the percentage of targets achieved, was
Sample size for the IG for the primary efficacy endpoint significantly greater in the IG group than in the UC group
was estimated at 26 based on the following assumptions:
mean MET-min/wk 1 wk after discharge from CR 2500; Table 1
1000 MET-min/wk increase after 3 mo of web-controlled Baseline Characteristics of Patientsa
exercise training (3500 MET-min/wk); common SD of 1200
MET-min/wk; 80% power; .05 alpha level; and a 10% Intervention Usual Care
dropout rate. (n = 26) (n = 27) P Value
Continuous variables are reported as mean ± SD, and Male 24 (92) 22 (82) .42
categorical variables as numbers and percentages. For Age, M ± SD, y 65 ± 10 67 ± 8 .41
categorical data, groups were compared by contingency Body mass index 25.7 ± 4 24.2 ± 2 .08
tables with the χ2 test or Fisher’s exact test, where appro-
priate. After checking the normality of the data using the Years of formal education 13 ± 5 9±4 .007
Shapiro-Wilk statistic, continuous data were analyzed using Currently working 15 (58) 5 (18) .005
either a 2-sided Student’s t test or analysis of covariance ad- Cardiovascular risk factors
justed by clinical variables different at baseline, or 2-sided Active or past smoker 7 (24) 6 (19) .75
Wilcoxon’s rank sum test or Friedmann’s test, as appropri- Diabetes 8 (32) 2(8) .039
ate. Significance set was at the .05 level. Obesity 5 (19) 1 (4) .10
Arterial hypertension 18 (69) 18 (69) 1.00
Dyslipidemia 18 (69) 20 (75) .76
RESULTS CHD history
Baseline clinical characteristics are shown in Table 1. The Myocardial infarction 7 (27) 14 (52) .09
IG patients were more often currently working (P = .005) Percutaneous coronary 6 (23) 8 (30) .76
and had a higher educational level (P = .007) and diabetes angioplasty
was more prevalent (P = .039) than in the UC group. Pre- Coronary artery bypass grafting 19 (73) 17 (63) .56
scription of β-blockers and statins at discharge was similar- Ischemic heart failure 2 (8) 1 (4) .61
ly high. Only 1 patient in the UC group had mild symptoms Medical history
(NYHA class II). Atrial fibrillation 0 1 (4) 1.00
During the intervention, 2 subjects dropped out: 1 in the Implantable cardioverter 1 (4) 1 (4) 1.00
IG group withdrew consent and 1 in the UC group was defibrillator
referred to outpatient rehabilitation after a hospitalization Chronic obstructive pulmonary 3 (12) 3 (11) 1.00
that occurred during follow-up. Data at the final 6-mo visit disease
were available for analysis in 25 IG and 26 UC patients. Chronic kidney dysfunction 2 (8) 1 (4) .61
Peripheral vessel disease 1 (4) 1 (4) 1.00
IG ADHERENCE AND EFFECTS Pharmacological treatment
During the 6-mo intervention period, IG had 752 web con- β-Blockers 26 (100) 24 (89) .24
tacts (average 30) through the dedicated chat and 70% had ACE inhibitors or angiotensin 12 (44) 19 (74) .09
at least 1 video conference with health care professionals receptor blockers
at the CR unit. Active daily data transmission was 100% Calcium channel blockers 9 (35) 5 (19) .22
during the first month, 88% at 3 mo, and 81% at 6 mo. Diuretics 14 (54) 12 (44) .78
Most patients (85%) completed at least 80% of the pre- Amiodarone 7 (27) 2 (8) .08
scribed intervention (ie, at least 4.8 months of interactive Statins 22 (85) 24 (89) .70
contacts with health care professionals). Nitrates 1 (4) 1 (4) 1
Self-reported physical activity improved from the first Echocardiographic measures
week after hospital-based CR discharge (2467 [1854-3554] LVEF, % 54 ± 9 51 ± 10 .33
MET-min/wk) to month 3 (3411 [1981-5347] MET-min/wk,
P = .019) and remained stable (3246 [1737-5281] MET-
LVEDV, mL 97 ± 10 99 ± 7 .84
min/wk) after 6 mo (P = .95 vs month 3) (Figure 2). Ac- LVESV, mL 49 ± 6 54 ± 5 .96
tivity levels further increased from months 1 to 3 in 50% 6MWD, m 393 ± 70 418 ± 72 .22
of patients and in 52% from 1 to 6 mo. The average time Abbreviations: ACE, angiotensin-converting enzyme; CHD, coronary heart disease; LVEDV,
spent and the proportion of patients who engaged in dif- left ventricular end-diastolic volume; LVEF, left ventricular ejection fraction; LVESV, left
ferent activities as determined from platform recordings is ventricular end-systolic volume; 6MWD, 6-min walk distance.
shown in Table 2. a
Data are reported as number (%) or mean ± SD.

www.jcrpjournal.com Web-Based Home Exercise Training for CHD Secondary Prevention 3

Copyright © 2018 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.

Figure 2. Time trend of self-reported physical activity in the intervention
group. Box plots represent 25th and 75th percentile, heavy black line is
the median, and whiskers are extreme values. Abbreviations: BMI, body
mass index; BP, blood pressure; HbA1c, glycated hemoglobin; LDL,
low-density lipoprotein.
(75 ± 20% vs 59 ± 30%, P = .029; mean difference 16%,
95% CI, 2–31).
Significant improvement with time was found in both
groups for left ventricular ejection fraction, distance walked
in 6 min, and SF-36 physical and mental component scores
(Table 3). Adherence to drug therapy at the final visit was
comparably high in IG vs UC: at the final 6-mo visit, 100%
vs 89% (P = .23), respectively, were still on β-blockers, and
85% vs 80%, respectively, were still on statins (P = .72),
respectively. No significant between-group differences
were observed in cardiovascular events (Table 4; P = .71),
while more UC than IG patients (8 vs 1) underwent a
self-requested unplanned cardiology visit during follow-up.
DISCUSSION
This study documents that a long-term home CRMP, fo-
cused on bidirectional communication via the internet and
web-supervised feasible training, is feasible, well-accept-
ed, safe, and effective in obtaining a sustained increase in
self-reported physical activity up to 6 mo and resulted in
higher overall adherence to CV risk targets than UC. The
underutilization of guideline recommended CR after a
coronary event or procedure is a reality all over Europe,7
hence alternative strategies12,16 to increase adherence to the
core elements of CR are of paramount importance.
Literature on web-based interventions has recent-
ly accrued with studies that have explored the impact of
Table 2
Type and Amount of Daily Life Activities Performed by
Patients in the Intervention Group
Minutes/
Patient/Wk
Walking 96
Bicycle/cyclette 78
Structured physical exercise 43
Housework 162
Climbing stairs 30
Swimming 26
Leisure activities (dancing, gardening, playing 19
with children, football)
4 Journal of Cardiopulmonary Rehabilitation and Prevention 2018;00:1-6
Copyright © 2018 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
internet-delivered programs on physical activity, quality
of life, and adherence to secondary prevention prescrip-
tions.17-24 Overall, telehealth CR obtained similar, or great-
er, positive effects on functional capacity and cardiovascular
risk factors control to center-based CR or UC. Variability
in intervention type and specifically platform characteris-
tics impacts substantially on results. In the meta-analysis by
Rawston et al,17 telehealth CR focused on exercise training
was more effective for increasing physical activity level ver-
sus UC, but had no impact on other outcomes.
The CRMP used in this study was started after discharge
from a 3-wk inpatient CR program and, hence, mainly re-
cruited more severe patients, who required inpatient sta-
bilization after coronary bypass grafting or an acute isch-
emic event. The program focused on self-reported physical
activity and lifestyle changes to achieve cardiovascular
risk targets, overall frequent personal virtual contacts be-
tween patients and health care professionals, and case
management by a physiotherapist. We used a 6-mo dura-
tion for the CRMP to extend the classical CR period of
4 to 8 wk and focused on increasing physical activity in
order to encompass long-lasting lifestyle changes. To raise
patient awareness on actually physical activity performed
in daily life and to maximize compliance, our exercise-
focused program was designed to be unobtrusive and relied
on self-reported measures. During the first 3 mo, physi-
cal activity significantly increased by 38%, documenting
effective uptake of the training program in real-life situa-
tions. Although the increase was sustained with time, the
average change at 6 mo versus baseline was lower, with
higher variability than expected, and did not achieve sta-
tistical significance, indicating that the study was probably
underpowered.
While some previous research enrolled subjects who
were not participating to CR at variable distances from an
acute event,19-21 similar to CardioFit18 or Telerehab III,22 we
recruited patients who had just completed inpatient CR
after an acute ischemic event. This study design implies
that the impact of the intervention would build on the ben-
efits achieved during inpatient CR and sustain adherence
to the core components of center-based CR, as previously
documented.21
Measures of functional capacity and quality of life im-
proved with time in both IG and UC groups. Our partici-
pants were older (mean age 66 ± 9 y) than those in previ-
ous reports (mean age 56 to 61 ± 9 y),18,19,21-23 except for 1
study.20 The enrollment of patients who were older and had
recently undergone coronary artery bypass grafting, exclud-
ed from most previous trials, 18,19,21-23 may explain the lower
level of physical activity performed in this study.
Most previous studies18,21,23 did not assess CV risk tar-
gets. Devi et al20 and Brough et al21 found no differences af-
ter the intervention in weight and body fat. Frederix et al22
found no differences with time within and between groups
on glucose and lipids. In our study, the difference in the
overall CV risk factor target score at 6 mo was statistically
significant. Despite a relative imbalance in overweight and
Percent overt diabetes among groups, a stronger focus on nutrition-
Compliant al aspects might have impacted more on weight and glucose
85 target values.
81
Our experience confirms the view of telehealth CR as
96
a powerful tool, complementary to traditional home- or
center-based programs, to widen the outreach of CRMP
69
and distribute specialist expertise from clinical centers to
100
any location with internet access. This approach may allow
23
delivery of interventions to geographically disadvantaged
27
remote areas but also to metropolitan areas where travel
requirements may limit access.
www.jcrpjournal.com
 Table 3
Changes in Clinical and Laboratory Findings From Baseline to the End of Studya,b
Intervention (n = 25) Usual Care (n = 26) P Value P Value P Value
Baseline 6 mo Baseline 6 mo Time Interaction Group

25.8 ± 4 26.6 ± 4 24.2 ± 2 25.0 ± 2

2
Body mass index, kg/m .001 .816 .017
Waist circumference, cm 103 ± 10 104 ± 11 99 ± 8 98 ± 10 .768 .549 .042
Systolic blood pressure, mm Hg 120 ± 15 125 ± 15 114 ± 13 129 ± 18 .001 .068 .856
Biochemical tests
Total cholesterol, mg/dL 145 ± 38 152 ± 27 142 ± 38 163 ± 33 .205 .017 .836
HDL cholesterol, mg/dL 31 ± 7 41 ± 9 33 ± 12 47 ± 13 .001 .003 .129
LDL cholesterol, mg/dL 90 ± 38 97 ± 20 84 ± 29 96 ± 30 .726 .011 .885
Triglycerides, mg/dL 121 ± 52 118 ± 55 124 ± 66 98 ± 46 .047 .160 .714
Glucose, mg/dL 107 ± 49 118 ± 29 94 ± 12 108 ± 23 .016 .637 .597
HbA1c, mmol/mol 38 ± 20 44 ± 17 31 ± 11 39 ± 9 .001 .584 .438
Echocardiographic measures
LV ejection fraction, % 54 ± 9 58 ± 10 51 ± 10 55 ± 10 .002 .786 .084
LV end-diastolic volume, mL 97 ± 55 97 ± 29 100 ± 39 104 ± 32 .724 .429 .506
LV end-systolic volume, mL 52 ± 31 50 ± 28 55 ±27 54 ±30 .587 .861 .460
6MWD, m 393 ± 70 430 ± 9 418 ± 72 465 ± 75 .001 .928 .448
36-item Short Form Health Survey
Physical component score 41 ± 10 44 ± 9 39 ± 9 47 ± 9 .003 .316 .345
Mental component score 44 ± 10 48 ± 8 42 ± 10 48 ± 8 .001 .690 .836
Abbreviations: HbA1c, glycated hemoglobin; HDL, high-density lipoprotein; LDL, low-density lipoprotein; LV, left ventricular; 6MWT, 6-min walk distance.
a
Data are reported as mean ± SD.
b
Analysis of covariance for repeated measures adjusted by presence of diabetes at baseline.

Web-supervised exercise training cannot be expected to
equally suit all CHD patients; overall one-third of CHD
patients discharged from our inpatient CR program and
living in Lombardy were eligible for the web-based CRMP.
Besides nonmodifiable reasons for ineligibility in the study,
25% of patients had no computer or internet access. Since
the digital divide is likely to shrink in the future, poten-
tially one-half of low- to moderate-risk CHD patients dis-
charged from inpatients CR might eventually be involved in
a web-based home CRMP. Of note, over half of our patients
were working full-time and would not have continued tra-
ditional center-based outpatient CR, while one-third ex-
pressly asked for the opportunity to continue the program.
Two-thirds of eligible patients consented to the interven-
tion, a proportion similar to the study by Reid et al18 and
higher than the 15.5% reported by Devi et al.20 The lower
long-term attrition rate (19%) in comparison with previ-
ous studies (range 22%-73%)18-24 may be attributed to the
motivational effect of frequent personal contacts with the
physiotherapist case manager, as well as to the focus on un-
obtrusive implementation of exercise training through daily
life activities. Contact with cardiologists, who supervised
laboratory findings and drug therapy, may have contrib-
uted to the better lipid profile and blood pressure target
adherence.
The impact of telehealth exercise CR on clinical outcomes
remains currently uncertain due to a small number of events
reported.17 The safety profile of our intervention was good;
no deaths, chest pain episodes, or ventricular arrhythmias
during exercise training were observed. MACE occurred in
15% and 11% in the IG and UC groups, respectively. In 2
patients with moderately to severely depression ventricular
function at baseline who were hospitalized for heart failure,
the program allowed, through anticipated recognition of
signs and symptoms of impending destabilization, an early
hospitalization.

Table 4
Hospital Admissions and Unplanned Medical Visits During
Follow-upa
Intervention Usual Care
(n = 26) (n = 27)
Hospital admissionsb
Heart failure 2 1
Percutaneous coronary angioplasty 1 1
Cerebrovascular event 1 …
Sustained arrhythmia … 1
Figure 3. Proportion of patients achieving cardiovascular risk targets at Atypical chest pain 1 …
the end of study in the intervention (white bars) and usual care (black Pneumonia … 1
bars) groups. P > .05 for all between-group comparisons. Abbre- Unplanned cardiology outpatient visits 1 8
viations: BP, blood pressure; BMI, body mass index; HbA1c, glycated
a
hemoglobin. Data are reported as number.
b
Overall P = .71.

www.jcrpjournal.com Web-Based Home Exercise Training for CHD Secondary Prevention 5

Copyright © 2018 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
 There are several limitations in our study that may lim-
it the generalizability of our findings. To maximize patient
safety, we enrolled only patients with low- to moderate-risk,
stable CHD and preserved or mildly reduced systolic func-
tion, without major comorbidities or persistent heart failure
signs or symptoms. Two-thirds of our participants had high
school or college education and most were men (86.7%).
Gender imbalance may be consistent with the prevalence of
CHD and the lower digital literacy among elderly women.
To offer an unobtrusive program, we did not objectively
document exercise performed and some participants may
have exaggerated self-reported activity.
An important bias of our study is self-selection to IG,
which resulted, as might be expected, in a lower prevalence
of active workers and lower educational level among UC.
A lower level of clinical attention to the UC group, when
compared to frequent contacts between IG and health care
personnel, is another confounder. Using an UC group with
more frequent follow-up contacts in future studies would
be important to distinguish the beneficial effects of a similar
CRMP versus increased attention alone.
CONCLUSIONS
Our web-based telehealth CRMP was feasible, well-accepted,
safe, and effective in obtaining improvement in physical
activity up to 6 mo and higher overall adherence to car-
diovascular risk target than UC. Innovative features of the
program with respect to classic center-based outpatient CR
were self-monitoring and validation of physical activity, ed-
ucation and support, supervision of exercise and frequent
interactions via the internet with health care professionals
may have contributed to sustained adherence.
Web-based exercise training programs hold the potential
to increase the proportion of subjects who may benefit from
CR after hospital discharge to patients who live far from
the CR center and those of working age, who for logistical
reasons, cannot access traditional CR.
ACKNOWLEDGMENT
This study was funded by Regione Lombardia Independent
Research Call #2010 GA 12755.
REFERENCES
1. Anderson L, Taylor RS. Cardiac rehabilitation for people with
heart disease: an overview of Cochrane systematic reviews. Co-
chrane Database Syst Rev. 2014;12:CD01127.
2. Wong WP, Feng J, Pwee KH, Lim J. A systematic review of eco-
nomic evaluations of cardiac rehabilitation. BMC Health Serv Res.
2012;12:243.
3. Piepoli MF, Corra U, Benzer W, et al. Cardiac Rehabilitation Sec-
tion of the European Association of Cardiovascular Prevention and
Rehabilitation. Secondary prevention through cardiac rehabilita-
tion: from knowledge to implementation. A position paper from
the Cardiac Rehabilitation Section of the European Association of
Cardiovascular Prevention and Rehabilitation. Eur J Cardiovasc
Prev Rehabil. 2010;17:1-17.
4. Heran BS, Chen JM, Ebrahim S, et al. Exercise-based cardiac reha-
bilitation for coronary heart disease. Cochrane Database Syst Rev.
2011;(7):CD001800.
5. Van Camp SP, Peterson RA. Cardiovascular complications of outpa-
tient cardiac rehabilitation programs. JAMA. 1986;256:1160-1163.
6 Journal of Cardiopulmonary Rehabilitation and Prevention 2018;00:1-6
Copyright © 2018 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
6. Franklin BA, Bonzheim K, Gordon S, Timmis GC. Safety of medi-
cally supervised outpatient cardiac rehabilitation exercise therapy:
a 16-year follow-up. Chest. 1998;114:902-906.
7. Kotseva K, Wood D, De Backer G, De Bacquer D; EUROASPIRE
III Study Group. Use and effects of cardiac rehabilitation in pa-
tients with coronary heart disease: results from the EUROASPIRE
III survey. Eur J Prev Cardiol. 2013;20: 817-826.
8. Jackson L, Leclerc J, Erskine Y, Linden W. Getting the most out of
cardiac rehabilitation: a review of referral and adherence predic-
tors. Heart. 2005;91:10-14.
9. Daniels KM, Arena R, Lavie CJ, Forman DE. Cardiac rehabili-
tation for women across the lifespan. Am J Med. 2012;125:937.
e1-937.e7.
10. Serber ER, Todaro JF, Tilkemeier PL, Niaura R. Prevalence
and characteristics of multiple psychiatric disorders in cardiac
rehabilitation patients. J Cardiopulm Rehabil Prev. 2009;29:
161-168.
11. Dalal HM, Zawada A, Jolly K, Moxham T, Taylor RS. Home
based versus centre based cardiac rehabilitation: Cochrane system-
atic review and meta-analysis. BMJ 2010;340:b5631.
12. Clark RA, Conway A, Poulsen V, Keech W, Tirimacco R, Tideman
P. Alternative models of cardiac rehabilitation: a systematic review.
Eur J Prev Cardiol. 2015;22:35-74.
13. Piotrowicz E, Baranowski R, Bilinska M, et al. A new model of
home based telemonitored cardiac rehabilitation in patients with
heart failure: effectiveness, quality of life and adherence. Eur J
Heart Fail. 2010; 12:164-171.
14. Apolone G, Mosconi P, Ware JE. Questionario sullo stato di sa-
lute SF-36. Manuale d’uso e guida all’interpretazione dei risultati.
Italian. 1997. Milan, Italy: Guerini e Associati.
15. Nelson ME, Rejeski WJ, Blair SN, et al.; American College of
Sports Medicine; American Heart Association. Physical activity
and public health in older adults: recommendation from the Amer-
ican College of Sports Medicine and the American Heart Associa-
tion. Circulation. 2007;116:1094-1105.
16. Huang K, Liu W, He D, et al. Telehealth interventions versus
center-based cardiac rehabilitation of coronary artery disease: a
systematic review and meta-analysis. Eur J Prev Cardiol. 2015;
22:959-971.
17. Rawstorn JC, Gant N, Direito A, Beckmann C, Maddison R. Tele-
health exercise-based cardiac rehabilitation: a systematic review
and meta-analysis. Heart. 2016;102:1183-1192.
18. Reid RD, Morrin LI, Beaton LJ, et al. Randomized trial of an in-
ternet-based computer-tailored expert system for physical activ-
ity in patients with heart disease. Eur J Prev Cardiol. 2012;19:
1357-1364.
19. Antypas K, Wangberg SC. An internet- and mobile-based tailored
intervention to enhance maintenance of physical activity after car-
diac rehabilitation: short-term results of a randomized controlled
trial. J Med Internet Res. 2014;16:e77.
20. Devi R, Powell J, Singh S. A web-based program improves physical
activity outcomes in a primary care angina population: random-
ized controlled trial. J Med Internet Res. 2014;16:e186.
21. Brough C, Boyce S, Houchen-Wolloff L, Sewell L, Singh S. Evalu-
ating the interactive web-based program, activate your heart, for
cardiac rehabilitation patients: a pilot study. J Med Internet Res.
2014;16:e242.
22. Frederix I, Hansen D, Coninx K, et al. Medium-term effec-
tiveness of a comprehensive internet-based and patient-specific
telerehabilitation program with text messaging support for car-
diac patients: randomized controlled trial. J Med Internet Res.
2015;17:e185.
23. Maddison R, Pfaeffli L, Whittaker R, et al. A mobile phone inter-
vention increases physical activity in people with cardiovascular
disease: results from the HEART randomized controlled trial. Eur
J Prev Cardiol. 2015;22:701-709.
24. Pfaeffli L, Whittaker R, Dixon R, et al. Acceptability of a
mobile health exercise-based cardiac rehabilitation interven-
tion: a randomized trial. J Cardiopulm Rehabil Prev. 2015;35:
312-319.
www.jcrpjournal.com