Two large nationwide mega cohort analyses demonstrated a large survival benefit associated with RM utilization (ALTITUDE⁷ and MERLIN¹⁰). Moreover, the extent of benefit correlated with the degree of utilization. These large retrospective analyses were system specific and lacked clinical details. Possibly more motivated patients with fewer comorbidities were more likely to receive and implement RM, and these characteristics contributed to clinical outcome irrespective of RM utilization. However, when linked to patient demographics, the Patient RElated Determinants of ICD Remote Monitoring (PREDICT RM) study¹¹,¹² confirmed that RM patients had a lower risk of mortality and hospitalization compared with conventionally managed patients (hazard ratio [HR]: 0.67; 95% confidence interval [CI]: 0.64-0.71; P < 0.0001; and HR: 0.82; 95% CI: 0.80-0.84; P < 0.0001, respectively).

Studies inclusive of differing RM technologies (eg, the prospective EFFECT¹³ and the retrospective analysis by Portugal et al¹⁴) suggested that RM improved clinical outcomes (death and cardiovascular hospitalizations). These studies, however, were small. The latest nationwide cohort study,¹⁵ with a large number of patients (N = 92 566) with implanted devices from any manufacturer, demonstrated that patients using RM had a lower adjusted risk of all-cause hospitalization, at an HR of 0.82 (95% CI: 0.80-0.84; P < 0.001), and a shorter mean length of hospitalization (5.3 vs 8.1 days; P < 0.001) during follow-up. RM was associated with a $3703 reduction in hospitalization cost per patient-year (P < 0.001) including all device types and a $7358 reduction in hospitalization cost for ICD patients (43% lower, P < 0.001). This large cohort is noteworthy because it proves that, if deployed appropriately, RM has the power to reduce the hospitalization and cost associated with heart failure, no matter which manufacturer and device are selected.

Large prospective randomized controlled trials have yielded mixed results. In Clinical Evaluation of Remote Notifications to Reduce Time to Clinical Decision (CONNECT), patients randomized to RM had reduced length of hospitalizations.¹⁶ Influence of Home Monitoring on Mortality and Morbidity in Heart Failure Patients with Impaired Left Ventricular Function (IN-TIME)⁶ was the first to show the survival benefit of RM among patients randomized to daily automatic multiparameter data transmission (using Home Monitoring) compared with conventional follow-up. However, hospitalization was unchanged.

Subsequent randomized trials did not confirm these effects.¹⁷ Thus, remote management based on thoracic impedance in Medtronic devices did not change outcome in the OptiLink trial.⁹ The Monitoring Resynchronization Devices and Cardiac Patients (MORE-CARE) trial⁸ testing RM among CRTD patients with CareLink inaudible alerts showed no difference. In Remote Management of Heart Failure Using
Implantable Electronic Devices (REM-HF),¹⁸ there was no impact of weekly RM in outcome during a median follow-up period of 2.8 years. The RESULT study is an ongoing randomized study including three brands (St. Jude Medical, Biotronik, and Medtronic) to assess the composite outcome of death for any reason or hospitalization for cardiovascular reasons.

Meta-analyses yield differing results (although they do not include the latest trials). One meta-analysis¹⁹ included 4932 patients from seven randomized controlled RM trials and showed that overall clinical outcomes were comparable to conventional follow-up in terms of all-cause mortality (odds ratio [OR]: 0.83; P = 0.285), cardiovascular mortality (OR: 0.66; P = 0.103), and hospitalization (OR: 0.83; P = 0.196). However, survival benefit was noted in the three trials only featuring RM with daily automatic multiparameter data transmission, that is, Home Monitoring (OR: 0.65; P = 0.021). This was confirmed by the latest meta-analysis of three trials (TRUST, ECOST, and IN-TIME) using Home Monitoring presented in the European Society of Cardiology 2016. This analysis demonstrated Home Monitoring reduced all-cause death (P = 0.037) and the composite end points of worsening heart failure hospitalization and all-cause or worsening heart failure death (P = 0.007).

Another meta-analysis of 11 randomized controlled studies conducted up to the year 2015²⁰ showed that RM was associated with a reduction in the total number of visits (planned, unplanned, and emergency room; relative risk: 0.56; 95% CI: 0.43-0.73; P < 0.001) but no effect on cardiac hospitalization or total and cardiac mortality (because only IN-TIME reported positive survival effect).

Where do these inconsistent results lead us? First, it is difficult to generalize because there were notable differences in both patient populations studied and RM technologies used, and, importantly, in alert mechanisms and reaction pathways among these trials. For example, REM-HF enrolled fewer New York Heart Association (NYHA) Class III/IV patients than IN-TIME or OptiLink. The negative results of the MORE-CARE study are thought to be caused by the lower mortality rate in its control group compared with that of IN-TIME⁶ (4.8% vs 8.7%), whereas the high comorbidity rate (two-thirds of hospitalization cases) mitigated the impact of RM.

IN-TIME, employing a RM technology with daily transmissions with high reliability,²¹ instituted a treatment algorithm in response to RM alerts (without thoracic impedance) that was strictly applied. In contrast, in OptiLink, which enrolled similar patients⁹ but relied on changes in impedance, alerts were transferred to the physician in only 76% of cases (1324 out of 1748 alerts). In the remaining 24%, alert transmission was unsuccessful because patients were not online. Even when successfully conveyed, actual medical intervention was conducted in only 30% (529/1748) of all alerts, and for 19% (250/1324) of transmitted alerts, patients were not contacted by physicians. Setting of the threshold of thoracic impedance to trigger alert was not standardized, and medical intervention was at physicians’ discretion. This means that the alert threshold applied may not have had sufficient predictive value for heart failure decompensation, leading to misdirected preemptive therapy.