A retrospective, observational cohort study showed that remote monitoring (RM) added to clinic visits reduced hospitalization, hospital length of stay (LOS), and healthcare costs in patients implanted with any device for cardiac rhythm management (CRM), according to Jonathan P. Piccini, MD, Duke University Medical Center, Durham, North Carolina, USA.

A total of 92 566 patients (mean age, 72 years; 63% men) who had a CRM device implanted between April 1, 2008, and March 31, 2013, were included in the study; of these, 34 259 were in the RM plus clinic visit arm, and 58 307 were in the clinic visit only arm. Patients without any clinic follow-up or whose first clinic or RM follow-up was > 4 months after implant were excluded. Each clinic determined the type of follow-up for each patient. The data source was a commercial and Medicare supplemental health insurance database. All outcomes were adjusted using a boosted logistic regression propensity score that included 22 pre-implant comorbidities and age, sex, and geographic location.

More patients who received an implantable cardioverter defibrillator (ICD; 49%) or cardiac resynchronization therapy defibrillator (CRT-D; 51%) were in the RM arm than the pacemaker (PM; 29%) or CRT pacemaker (CRT-P; 27%) groups. The Charlson Comorbidity Index was 3.1 and 3.2 in the RM and no-RM arms, indicating the patients had a similar degree of comorbidity. In the RM arm vs no-RM arm, more patients had a history of heart failure (49.9% vs 44.9%) and ventricular arrhythmia (24.3% vs 15.7%), and fewer patients had atrial fibrillation (42.7% vs 46.7%) and prior cerebrovascular disease (28.1% vs 33.2%). The first clinic visit was 64 and 63 days after the implant in the RM and no-RM arms. The follow-up interval was ≤ 4 months in > 75% of patients.

The primary outcome of all-cause hospitalization for all device types was significantly lower with RM vs no RM (HR, 0.82; 95% CI, 0.80 to 0.84; P < .001). The mean LOS was 5.3 days with RM and 8.1 days with no RM (P < .001). All-cause hospitalization was lower with all device types in the RM vs no-RM arm, but Dr Piccini noted that the magnitude of this difference was greater with ICD (HR, 0.74; 95% CI, 0.71 to 0.77; P < .001) and CRT-D (HR, 0.72; 95% CI, 0.67 to 0.77; P < .001), compared with a pacemaker (HR, 0.83; 95% CI, 0.81 to 0.86; P < .001) and a CRT pacemaker (HR, 0.84; 95% CI, 0.69 to 1.03; P = .089).

Hospitalization costs were significantly lower ($12 423 vs $8720; P < .001) with RM vs no RM for all devices (by 30% per patient-year) and for each type of device (by 31% to 45% per patient-year). The greatest difference in hospitalization cost was observed for CRT-D, where hospitalization cost per patient-year was nearly $10 000 (45%) lower with RM vs no RM, followed by a nearly $7000 (43%) reduction for ICD with RM vs no RM.

RM vs no RM was associated with reduced hospitalization for heart failure in patients with a history of heart failure (HR, 0.76; 95% CI, 0.71 to 0.81; P < .001) and for stroke in patients with atrial fibrillation (HR, 0.78; 95% CI, 0.67 to 0.91; P < .001). The shorter mean LOS for stroke hospitalization (2.9 vs 3.3 days with RM and no RM) translated into a 44% reduction in mean costs per patient-year. Rehospitalization for heart failure was lower with RM at 30, 90, and 180 days for all devices and for each type of device; there was a 7% absolute difference in patients with a CRT in the RM vs no-RM arms.

The limitations of being a retrospective analysis, which can show correlation but not causality, and evaluating only hospital costs are balanced against the strengths of being a large nationwide cohort that was adjusted for 22 clinical conditions and the inclusion of all device manufacturers.

The investigators estimated that based on the study results, for every 100 000 patient-years, RM would be associated with 9810 fewer hospitalizations, 119 000 fewer hospital days, and $370 270 000 lower hospital payments. Confirmation in a broader trial is needed.