Effects of pimobendan on prolonging time to rehospitalization or frequency of rehospitalization in patients with heart failure: A retrospective cohort study using a medical administrative database (PREFER study)

Abstract

Background: As approximately 24% of patients with chronic heart failure are rehospitalized within 1 year and heart failure is aggravated by repeated hospitalizations, greater importance was attached to the prevention of hospitalization. Objective: The objective of this study was to investigate the influence of pimobendan on rehospitalization of patients with advanced heart failure using a Japanese medical administrative database. Methods: From January 2010 to February 2018, patients hospitalized two or more times for heart failure were selected for analysis. The primary endpoint was the incidence of hospitalizations for heart failure during the follow-up period, which was compared between pimobendan prescription and nonprescription groups after propensity score matching. Results: The total number of patients with heart failure included during the study period was 1 421 110 and we matched 276 patients in both groups. The incidence of rehospitalization throughout the period to completion of follow-up was 365.23/1000 people/yr (95% confidence interval [CI]: 327.78-402.69) in the pimobendan prescription group and 537.81/1000 people/yr (95% CI: 492.36-583.27) in the non-prescription group. The cumulative incidence at 365 days was significantly lower in the pimobendan prescription group (pimobendan prescription group: 35.4% (95% CI: 29.8-41.8), non-prescription group: 51.2% (95% CI: 45.1-57.7), (P < 0.001). The adjusted hazard ratio in the pimobendan prescription group was 0.556 (95% CI: 0.426-0.725, P < 0.001). Conclusion: Pimobendan was suggested to extend the time to rehospitalization for patients with advanced heart failure. It is necessary to verify the results of this study by performing a prospective study. In addition, the influence of pimobendan on general heart failure patients must be examined.