Effects of Telerehabilitation with Exercise as the Core Component on Peak Oxygen Uptake and Blood Pressure in Patients with Cardiovascular Disease: A Systematic Review and Meta-Analysis (Preprint)

Figures

• 

  Figure 1.. PRISMA 2020 flow diagram of the study selection process, showing records identified from the 5 databases and from other sources, duplicate records removed before screening, records screened and excluded, reports assessed for eligibility with reasons for exclusion, and the 13 studies included in the review. PRISMA: Preferred Reporting Items for Systematic Reviews and Meta-Analyses; RCT: randomized controlled trial.

• 

  Figure 2.. Risk of bias assessment of each of the 13 included randomized controlled trials of exercise-based remote rehabilitation using the Cochrane Risk of Bias tool version 2 (RoB 2) across 5 domains: randomization process, deviations from intended interventions, missing outcome data, outcome measurement, and selection of reported results [31-43].

• 

  Figure 3.. Summary of Cochrane Risk of Bias tool version 2 (RoB 2) judgments across 5 domains (randomization process, deviations from intended interventions, missing outcome data, outcome measurement, and selection of reported results) and overall bias ratings for 13 included randomized controlled trials of exercise-based remote rehabilitation. Overall, 15.4% (2/13) of studies were rated low risk, 30.8% (4/13) had some concerns, and 53.8% (7/13) were rated high risk of bias.

• 

  Figure 4.. Forest plot of the effects of exercise-based telerehabilitation vs usual care on VO₂ peak (mL/kg/min) in patients with cardiovascular disease, based on 10 randomized controlled trials (n=749) with intervention durations ranging from 8 to 24 weeks. Pooled analysis using a random-effects Hartung-Knapp-Sidik-Jonkman model showed a significant improvement in VO₂ peak (mean difference 2.58 mL/kg/min, 95% CI 1.16 to 4.00, t9=4.10, P=.003), with substantial heterogeneity (I²=74.45%) [31,33,34,36-41,43].

• 

  Figure 5.. Forest plot of the effects of exercise-based telerehabilitation vs usual care on systolic blood pressure (mm Hg) in patients with cardiovascular disease, based on 5 randomized controlled trials (n=295; intervention: n=148, control: n=147) published between 2008 and 2026. Pooled analysis using a random-effects Hartung-Knapp-Sidik-Jonkman model showed no statistically significant effect on systolic blood pressure (mean difference −1.80 mm Hg, 95% CI −7.42 to 3.81, t4=−0.89, P=.42), with moderate heterogeneity (I²=58.24%) [31,32,35,38,41].

• 

  Figure 6.. Forest plot of the effects of exercise-based telerehabilitation vs usual care on diastolic blood pressure (mm Hg) in patients with cardiovascular disease, based on 5 randomized controlled trials (n=295; intervention: n=148, control: n=147) published between 2008 and 2026. Pooled analysis using a random-effects Hartung-Knapp-Sidik-Jonkman model showed no statistically significant effect on diastolic blood pressure (mean difference −2.00 mm Hg, 95% CI −4.76 to 0.75, t4=−2.02, P=.11), with low to moderate heterogeneity (I²=36.98%) [31,32,35,38,41].

• 

  Figure 7.. Funnel plot assessing publication bias for VO₂ peak outcomes from 10 randomized controlled trials of exercise-based telerehabilitation in patients with cardiovascular disease. Egger linear regression test indicated statistically significant funnel plot asymmetry (P=.03), suggesting a possible small sample effect that should be considered when interpreting the pooled results.

• 

  Figure 8.. Leave-one-out sensitivity analysis for the 3 pooled outcomes. Each row shows the pooled mean difference and its 95% CI when the named trial is omitted. (A) VO₂ peak, 10 randomized controlled trials; (B) systolic blood pressure, 5 trials; (C) diastolic blood pressure, 5 trials. The dashed line marks the overall pooled estimate and the shaded band its 95% CI; the dotted line marks the line of no effect. All analyses use a random-effects model with the Sidik-Jonkman estimator and the Knapp-Hartung adjustment, as in Figures 4-6 [31-41,43].