Abstract Background Remote monitoring (RM) of implantable cardioverter defibrillators (ICD) has been widely adopted as the standard follow-up strategy [1]. Observational data [2,3] has suggested morbidity and mortality reduction associated with RM compared to traditional in-person ICD follow-up, however, randomised controlled trials (RCTs) demonstrate heterogeneous results. Purpose To assess the impact of RM of ICDs on morbidity and mortality through a systematic review and meta-analysis of RCTs comparing RM versus in-person follow-up. Methods PubMed, Embase, Cochrane Central Register of Controlled Trials (CENTRAL) and reference lists were searched for RCTs evaluating RM compared to in-person follow-up of patients with ICDs or cardiac resynchronisation therapy with defibrillator (CRT-D) devices. Meta-analyses of all-cause mortality, cardiovascular (CVD) hospitalisation and inappropriate ICD shock rates were performed. Results Literature search identified 1011 relevant publications, out of which 15 RCTs [4-18] involving a total of 7579 patients (13 RCTs reporting mortality rate, 7 RCTs reporting CVD hospitalisation rate, 6 RCTs reporting inappropriate ICD shock rate) met the inclusion criteria. Pooled analyses demonstrated that RM is not associated with a significant reduction in all-cause mortality (OR 0.95; 95% CI 0.78-1.16; p=0.60)(Figure 1, panel a) or odds of CVD hospitalisation (OR 0.95; 95% CI 0.72-1.25; p=0.70) (Figure 1, panel b). RM was, however, shown to be associated with significantly lower odds of receiving an inappropriate ICD shock compared to the control group (OR 0.63; 95% CI 0.46-0.88; p=0.006) (Figure 1, panel c). The IN-TIME study (the only RCT that independently reached statistical significance in terms of RM effect on mortality) enrolled patients exclusively with a Biotronik ICD or CRT-D; this raised the question as to whether the choice of vendor had an impact on mortality. Our subgroup analysis of the Biotronik device studies demonstrated that there was no significant all-cause mortality benefit (OR 0.87, 95% CI 0.59-1.27; p=0.46). Conclusions This meta-analysis of 15 RCTs including 7579 patients suggests that RM of ICDs is associated with a reduction in inappropriate ICD shocks, without a significant reduction in mortality or CVD hospitalisation when compared to in-person follow-up. Further studies comparing different RM processing and alert response models are warranted.