Remote monitoring of automated peritoneal dialysis reduces mortality, adverse events and hospitalizations: a cluster-randomized controlled trial

Abstract

Background: Remote monitoring (RM) of patients on automated peritoneal dialysis (APD) prevents complications and improves treatment quality. We analyzed the effect of RM-APD on mortality and complications related to cardiovascular disease, fluid overload and insufficient dialysis efficiency.

Methods: In a cluster-randomized, open-label, controlled trial, 21 hospitals with APD programs were assigned to use either RM-APD (10 hospitals; 403 patients) or conventional APD (11 hospitals; 398 patients) for the treatment of adult patients starting PD. Primary outcomes were time to first event of: (i) Composite Index 1 comprising all-cause mortality, first adverse events and hospitalizations of any cause, and (ii) Composite Index 2 comprising cardiovascular mortality, first adverse event and hospitalizations related to cardiovascular disease, fluid overload and insufficient dialysis efficiency. Secondary outcomes were time to first event of individual components of the two composite indices, and rates of adverse events, hospitalizations, unplanned visits and transfer to hemodialysis. Patients were followed for a median of 9.5 months. Primary outcomes were evaluated by competing risk analysis and restricted mean survival time (RMST) analysis.

Results: While time to reach Composite Index 1 did not differ between the groups, Composite Index 2 was reached earlier (ΔRMST: -0.86 months; P = .02), and all-cause mortality [55 vs 33 deaths, P = .01; sub-hazard ratio (sHR) 1.69 (95% confidence interval 1.39-2.05), P < .001] and hospitalizations of any cause were higher in APD group than in RM-APD as were cardiovascular deaths [24 vs 13 deaths, P = .05; sHR 2.44 (95% confidence interval 1.72-3.45), P < .001] and rates of adverse events and hospitalizations related to cardiovascular disease, fluid overload or insufficient dialysis efficiency. Dropouts were more common in the APD group (131 vs 110, P = .048).

Conclusions: This randomized controlled trial shows that RM may add significant advantages to APD, including improved survival and reduced rate of adverse events and hospitalizations, which can favorably impact the acceptance and adoption of the therapy.

Keywords: adverse events; automated peritoneal dialysis; cluster randomized control trial; mortality; remote patient monitoring.

Graphical Abstract

Figure 1:

Cumulative incidence rates of Composite Index 1. (A) First event of either all-cause mortality, adverse events and hospitalization of any cause; and the first event of the three components of index 1, all-cause mortality (B), adverse events (C) and hospitalization of any cause (D) for patients receiving APD and RM-APD, respectively. Inserts show results of competing risk analysis with sHR and 95% CI for these outcomes in patients using APD and RM-APD, respectively, after adjusting for age, sex and center. Results of RMST analysis with RM-APD as compared with APD at 5, 10 and 15 months for are shown as ΔRMST (95% CI) months.

Figure 2:

Cumulative incidence rates of Composite Index 2. (A) First event of either cardiovascular mortality, or adverse events and hospitalizations caused by cardiovascular events for patients receiving APD and RM-APD, respectively; and the first event of the components of Composite Index 2, cardiovascular mortality (B), adverse events (C) and hospitalizations (D) of cardiovascular origin. Inserts show results of competing risk analysis with sHR and 95% CI for these outcomes in patients using APD and RM-APD, respectively, after adjusting for age, sex and center. Results of RMST analysis with RM-APD as compared with APD at 5, 10 and 15 months are shown as ΔRMST (95% CI) months.