Remote Patient Monitoring in Heart Failure: A Systematic Review, Meta-Analysis, and Trial Sequential Analysis

Whether the cumulative evidence for remote patient monitoring (RPM) in heart failure (HF) is robust to sequential monitoring, and whether trials report geographic access modifiers, remains uncertain. We conducted a systematic review, meta-analysis, and trial sequential analysis (TSA) of 65 randomized controlled trials (RCTs) (59 poolable; approximately 23,000 participants) identified through a search of PubMed/MEDLINE, Cochrane CENTRAL, ClinicalTrials.gov, and the WHO International Clinical Trials Registry Platform (ICTRP) from inception through February 15, 2026, encompassing structured telephone support (STS) (15 trials), non-invasive telemonitoring (TM) (33 trials), and invasive hemodynamic monitoring (11 trials). Random-effects meta-analysis used restricted maximum likelihood (REML) with the Hartung-Knapp-Sidik-Jonkman (HKSJ) adjustment, and certainty of evidence was rated using the Grading of Recommendations, Assessment, Development, and Evaluations (GRADE) framework. RPM significantly reduced all-cause mortality (ACM) (risk ratio (RR): 0.911, 95% confidence interval (CI): 0.842-0.985; P=0.021; I²=0%; k=41; number needed to treat (NNT) 104 per year; prediction interval: 0.840-0.988), and trial sequential analysis suggested that the accrued evidence exceeded the required information size under a 15% relative risk reduction assumption, supporting a stable mortality signal. HF hospitalization was reduced (RR: 0.781, 95% CI: 0.710-0.859; P<0.001; k=39; number needed to treat 18 per year), although the prediction interval crossed 1.0 (0.586-1.040), indicating that in some clinical settings, the effect may be attenuated. All-cause hospitalization was not significantly reduced (RR: 0.959, 95% CI: 0.892-1.031; k=28). No statistically significant interaction by RPM modality was detected for any primary outcome (all-cause mortality P-interaction=0.80; HF hospitalization P-interaction=0.14). GRADE certainty was moderate for all-cause mortality and low for HF hospitalization, downgraded primarily for suspected publication bias and inconsistency. A descriptive geographic access analysis revealed that only 2 of 59 poolable trials reported formal rural-versus-urban subgroup analyses, precluding firm conclusions about whether RPM differentially benefits geographically underserved populations. Within these limitations, remote patient monitoring appears to reduce all-cause mortality and HF hospitalization compared with usual care across diverse modalities, while signaling persistent gaps in heterogeneity reporting and equity-focused subgroup data.