Effects of Monthly Dose and Regular Dosing of Intravenous Active Vitamin D Use on Mortality Among Patients Undergoing Hemodialysis

Abstract

To determine if apparent protective mortality benefits of intravenous active vitamin D in patients undergoing hemodialysis extend across all groups defined by dialysis duration; if higher monthly dose and dosing regularity are associated with reduced mortality; and if intravenous active vitamin D use is associated with reduced cardiovascular, infectious, and cancer-related mortality. Retrospective cohort study. Centers for Medicare and Medicaid Services End-Stage Renal Disease database. A total of 193,830 patients undergoing hemodialysis during 1999-2000, of whom 94,208 (48.6%) were taking intravenous active vitamin D in the baseline period. Time-varying Cox proportional hazards models were used to assess the effects of monthly vitamin D dose and dosing regularity over 3-month intervals on risk of all-cause and cause-specific death, by dialysis duration groups (<1 yr, 1 to <3 yrs, 3 to <5 yrs, and >or=5 yrs from dialysis initiation). Models were adjusted for baseline characteristics, time-varying hospital days, monthly epoetin alfa dose, mean hemoglobin level, and urea reduction ratio in the 3-month intervals. Maximum follow-up time was 5.25 years. Adjusted all-cause mortality risk was reduced 7-17% among patients receiving vitamin D each month of the 3-month interval, with the highest reduction among patients with shorter dialysis duration. However, regular vitamin D dosing did not show consistent benefit across dialysis duration groups for cardiovascular, infectious, cancer, or other (all deaths not attributable to cardiovascular disease, infection, or cancer) mortality. Mortality benefits of intravenous vitamin D cannot be easily explained by currently proposed biologic mechanisms. Randomized controlled trials are needed to show definitively whether intravenous vitamin D can reduce all-cause and cause-specific mortality in patients undergoing dialysis compared with placebo.