P1440ASSOCIATION BETWEEN THE USE OF DRUGS FOR CONSTIPATION AND MORTALITY IN HEMODIALYSIS PATIENTS; FROM JAPAN-DIALYSIS OUTCOMES AND PRACTICE PATTERNS STUDY (J-DOPPS)

Abstract Background and Aims Hemodialysis patients have a high prevalence of constipation related to medication use, water restriction, insufficient dietary fiber intake, and reduced physical activity. Thus, these patients use drugs for constipation (DFC) more than five times that used by healthy individuals. In hemodialysis patients, two pathological factors, namely, reduced defecation function and DFC use itself, occur. These factors could be related to deteriorating life prognosis resulting from altered intestinal microbiota, gastrointestinal complications, or declining quality of life. In the general population, some reports have suggested that constipation and DFC use are risk factors of coronary artery disease, cardiovascular death, or all-cause death. However, whether such risk factors are appropriate for hemodialysis patients has not been clarified. This study aimed to investigate the association between DFC use and risk of death in hemodialysis patients. Method In this prospective cohort study based on the Japan-Dialysis Outcomes and Practice Patterns Study (J-DOPPS), we used J-DOPPS phase 1-5 (1998-2015) data. Hemodialysis patients enrolled in the J-DOPPS whose dialysis vintage was >3 months (n=12,217) were divided into two groups according to usage (DFC group;n=3,721) and non-usage (non-DFC group;n=8,496) of DFC. The primary endpoint was all-cause death. Secondary endpoints were deaths from infection, malignancy, and cardiovascular causes. Hazard ratio (HR) was calculated using multilevel Cox regression analysis with facility level. Potential confounders were adjusted by the inverse probability of treatment weighting using the propensity score. Rubin’s rule was used for combined data. Sensitivity analysis was conducted using instrumental variable method to assess the effect of unmeasured confounders. Missing data were imputed using multilevel multiple imputation repeated 20 times. Results DFCs were prescribed in 30.5% of patients at baseline. Over a median follow-up of 730 [418, 974] days, 1,240 deaths from any cause were noted. Overall data were analyzed first. Although DFC use was associated with increased mortality risk (adjusted HR, 1.12; 95% confidence interval [CI] 1.03, 1.21), proportional assumptions were not met (Figure). Then, we only analyzed data of patients with observation period >1.5 years. In this study, 8,345 patients had an observation period >1.5 years, DFCs were prescribed in 31.0% of patients at baseline, and 389 patients died from any cause. Marked difference was observed after 1.5 years from baseline (adjusted HR, 1.35; 95% CI, 1.17, 1.55). Risks for deaths from infections (adjusted HR 1.62; 95% CI 1.14, 2.29) and malignancy (adjusted HR 1.60; 95% CI, 1.08, 2.36) were higher in the DFC group, but no difference in cardiovascular death risk was found (adjusted HR 1.11; 95% CI, 0.91, 1.36). Conclusion In hemodialysis patients, DFC use was associated with an increased risk of death. Thus, it may be important to manage defecation habit without using DFC by modulating intake of dietary fibers, exercise, or use of medications that are less likely to cause constipation.