Association between Hospital Volume and Early Mortality of Newly Diagnosed Acute Promyelocytic Leukemia: A Nationwide Database Study in Japan

Abstract

Background: Compared to prospective studies from high-volume hospitals, real-world data have shown higher rates of early death in patients with newly-diagnosed acute promyelocytic leukemia (APL). Previous reports suggested that in-hospital mortality seemed to be lower in high-volume hospitals in urban areas. We therefore hypothesized that hospital volume was associated with in-hospital mortality of newly diagnosed APL. To examine this association, we performed a nationwide database study in Japan. Study Design and Methods: We retrospectively identified patients with newly diagnosed APL who received induction therapy including all-trans retinoic acid (ATRA) between July 2007 and March 2018 from the Diagnosis Procedure Combination database. A multivariable logistic regression model was used to examine factors affecting in-hospital mortality. Independent variables included patient backgrounds (age, sex, body mass index, comorbidity index, and activity of daily living) and institutional factors, including hospital volume (≥5 or <5 newly diagnosed APL patients hospitalized during the study period in each hospital), location of hospital (urban or rural), educational hospital, the number of beds in each hospital, and the number of hematologists in each prefecture (obtained from The Japanese Society of Hematology). We further evaluated the association between the institutional factors and treatment practices using univariate analyses by chi-square test. Regarding treatment practices, immediate initiation of ATRA was defined as ATRA administration within one day from admission, and prompt initiation of conventional chemotherapy was defined as the start within seven days from admission. Results: We identified a total of 1,138 patients who received induction therapy including ATRA. During hospitalization, 195 (17%) patients died at a median of 11 days (interquartile range, 5-25). In the multivariable model, treatment at high-volume hospitals was significantly associated with lower in-hospital mortality (odds ratio, 0.60; 95% confidence interval, 0.39-0.91). In-hospital mortality was not significantly associated with the number of beds in each hospital, location of hospital, educational hospital, or the number of hematologists in the prefecture. We found that patients treated in high-volume hospitals were significantly more likely to receive immediate initiation of ATRA than those in low-volume hospitals (83% vs. 76%, p=0.004). Similarly, higher hospital volume was associated with a higher proportion of prompt conventional chemotherapy in addition to ATRA (55% vs. 41%, p<0.001). Discussion and Conclusions: The present study using a nationwide database showed that treatment at high-volume hospitals was significantly associated with lower in-hospital mortality in patients with newly diagnosed APL. In such high-volume hospitals, treatments were performed more promptly. Our results from real-world data may suggest the benefits of centralizing APL patients to high-volume hospitals.