Interdialytic weight gain and survival in hemodialysis patients: effects of duration of ESRD and diabetes mellitus.

Background: Medical mortality determinants in end-stage renal disease (ESRD) patients treated with hemodialysis (HD) are well known. More recently, associations have been established between the dose of dialysis administered and patient survival. We showed in a prospective study that both dialyzer type and patient compliance with the dialysis prescription were independently associated with survival. Although several parameters of dialytic technique and patient compliance are associated with differential survival in patients with ESRD treated with HD, the association of interdialytic weight gain (IWG) with survival is unclear. No study has assessed the relationship between IWG and mortality in HD patients, controlled for multiple medical risk factors. The aim of our study was to determine whether IWG was associated with survival in patients with ESRD treated with HD, controlling for multiple medical and dialytic risk factors.

Methods: We prospectively conducted an observational, longitudinal, multicenter study of 283 urban HD patients to determine the relationship of IWG with several dialytic parameters and patient survival. Medical risk factors such as demographic indices and comorbid conditions were assessed. We studied Kt/V, the protein catabolic rate (PCR), serum albumin and anthropometric measurements, behavioral compliance indices, dialyzer characteristics, and serum electrolyte concentrations, and correlated these with IWG. In addition, the duration of dialysis was assessed in HD patients with and without diabetes mellitus. Cox proportional hazards models assessed the relative mortality risk of increased IWG, controlling for variations in medical comorbidity and other mortality determinants.

Results: The mean (+/- SD) age of our population was 54.6 +/- 14.1 years, and the mean time they were treated with HD was 30.4 +/- 46.9 months. The mean IWG was 1.54 +/- 0.71% dry wt/day. Correlations were found between increased IWG and younger age, and lower midarm circumference, and increased Kt/V, PCR, and serum potassium concentration. The mean follow-up period was 48.9 +/- 10.6 months. An increase in IWG was associated with a significantly increased relative mortality risk in diabetic ESRD patients treated with HD when variations in age, comorbidity, serum albumin concentration, and dialyzer type and site were controlled. There was, however, no association of increased mortality risk with increased IWG in the larger population of patients without diabetes. In further analyses, the increased mortality risk associated with increased IWG was found to be present only in patients with diabetes mellitus who had recently started HD therapy for ESRD.

Conclusion: IWG is correlated with several nutritional and dialytic variables and with parameters that predict survival in HD patients. Increased IWG is independently associated with decreased survival of diabetic ESRD patients treated with HD, after adjusting for variation in other medical risk factors. The population of incident diabetic HD patients is particularly susceptible to increased risk associated with increased IWG. The mechanisms underlying these results are obscure, but IWG might be associated with poorer survival in this population if it were linked to worsened hypertension, cardiovascular stress, or poorer glycemic control. Interventions to improve compliance with IWG in incident diabetic HD patients are warranted.