New-onset diabetes after hemodialysis initiation: impact on survival.

BACKGROUND: The incidence of new-onset diabetes after initiation of hemodialysis (NODAD) and its impact on survival is not known.

METHODS: We used data from the United States Renal Data System (USRDS) from January 2000 to December 2001, with at least 3 years of follow-up for this study. Patients aged 18-80 years were included. NODAD was defined as two Medicare institutional claims for diabetes in patients with no history of diabetes prior to starting hemodialysis (HD). Incidence (per 1,000 patient-years), prevalence (%) and hazard ratios for mortality in patients with NODAD were calculated.

RESULTS: There were 59,340 incident patients with no history of diabetes prior to starting HD, of which 3,853 met criteria for NODAD. The overall incidence and prevalence of NODAD were 20 per 1,000 patient-years and 7.6%, respectively. In a cohort of 444 patients without diabetes and documented glycosylated hemoglobin A1c, <6% prior to starting HD (from January 2005 and March 2006), at a mean follow-up of 4.7 +/- 2.6 months, 6.8% developed NODAD defined by two Medicare claims for diabetes after initiation of HD. NODAD was associated with a significantly increased risk of death as compared to non-diabetes patients (hazard ratio 1.20, 95% confidence interval 1.14-1.25).

CONCLUSION: The USRDS showed a high incidence of NODAD, associated with significantly higher mortality compared to those who did not develop NODAD. The mechanism of NODAD needs to be explored further in experimental and clinical studies.