In a previous report, we showed that nutritional status and especially serum albumin had great predictive value for death in chronic hemodialysis patients, whereas blood pressure did not. In the present study, we analyzed the causes of death in consideration of the relationship between serum albumin and blood pressure. A total of 1,243 Okinawan patients (719 males, 524 females) undergoing hemodialysis in January 1991 were followed up through the end of 1995. Three hundred forty-two of the patients died, 45 received transplants, and 12 were transferred by the end of the follow-up period. The total duration of observation was 5,110.3 patient-years. Blood pressure as well as clinical and laboratory variables were determined immediately prior to the first dialysis session in January 1991. The crude death rate was 40.0% when the diastolic blood pressure (DBP) <70 mm Hg, 35.0% at 70 to 79 mm Hg, 25.0% at 80 to 89 mm Hg, 25.0% at 90 to 99 mm Hg, and 13.0% at >100 mm Hg. The death rate showed an inverse correlation with DBP. DBP showed a significant positive correlation with serum albumin (r = 0.137, P < 0.001) and age (r = -0.325, P < 0.0001). The adjusted odds ratio (95% confidence interval) of death was 0.84 (0.71 to 0.99) with 10 mm Hg increments in DBP when the reference DBP was less than 69 mm Hg. Low DBP may be a manifestation of malnutrition and/or cardiovascular disease in chronic hemodialysis patients. Target DBP levels may be higher levels in chronic hemodialysis patients than the general population.

No abstract

No abstract

We retrospectively surveyed all of the available medical records of 404 (191 females and 213 males) chronic dialysis patients, of whom 16 (4%) had insulin-dependent diabetes mellitus (IDDM) and 388 (96%) non-insulin-dependent diabetes mellitus (NIDDM). The patients were among 2,214 dialysis patients in Okinawa, Japan, of whom 443 were diabetic. The patients entered a large population-based dialysis registry. The mean duration from the diagnosis of diabetes mellitus (DM) to dialysis was 181.6 months in the IDDM patients and 150.4 months in the NIDDM patients. The NIDDM patients were classified into four subgroups according to their status when DM was first suspected. The duration from the diagnosis of DM until the onset of dialysis treatment was significantly shorter than in any other subgroup or in the IDDM subgroup with major vascular disease (131.9 months). Otherwise, the course of renal disease in NIDDM patients was similar to that in IDDM individuals. Most of our dialysis patients with DM had NIDDM. In most of the NIDDM patients, the diagnosis had been delayed for several years for unknown reason. However, if diagnosed early, NIDDM shows a clinical time course until dialysis similar to that of IDDM. Whether NIDDM patients contract chronic renal disease at an equal incidence to that of IDDM patients and the fraction of all diabetic patients accepted for chronic dialysis remain to be determined.