TIMING OF INITIATION OF UREMIA THERAPY AND SURVIVAL IN PATIENTS WITH PROGRESSIVE RENAL-DISEASE

We conducted a prospective cohort study to detect any relationships be tween specific clinical features and laboratory indices at initiation of hemodialysis and long-term survival. One hundred and thirty-nine co nsecutive patients with chronic renal failure hospitalized to start ma intenance hemodialysis between January 1990 and December 1994 were enr olled, and follow-up was completed through December 1995. At baseline, subjects were assigned to one of five groups based on their major ind ication for initiation of hemodialysis. The indications were: (a) naus ea and vomiting; (b) severe weakness; (c) no major symptom (dialysis s tarted because of 'high' serum creatinine and blood urea nitrogen conc entrations); (d) volume overload, and (e) miscellaneous (angina, peric arditis, seizure, pruritus, and hyperkalemia). Blood urea nitrogen, se rum creatinine and serum albumin concentrations were measured once bef ore the first dialysis. The main outcome measure was death. The 139 st udy subjects included 77 women and 62 men comprising 116 Blacks (83%), 15 Hispanics (11%), and 8 Whites (6%) of mean age 54 +/- 15 years. Me an length of follow-up was 39 months. At baseline, mean blood urea nit rogen concentration was 121 +/- 38 mg/dl, mean serum creatinine concen tration was 12.6 +/- 5.2 mg/dl, and mean serum albumin concentration w as 3.5 +/- 0.62 g/dl. Forty-two subjects (30%) died during follow-up. Cox regression analysis showed that there was no significant associati on between mortality and any of the indicators evaluated (indication f or initiation of dialysis (p = 0.2), serum creatinine concentration (< 10 vs. greater than or equal to 10 mg/dl) (p = 0.8), blood ure nitrog en concentration (<100 vs, greater than or equal to 100 mg/dl) (p = 0. 68) and serum albumin concentration (<4 vs. greater than or equal to 4 g/dl) (p = 0.62). All analyses included adjustment for age and diabet es. We conclude that in patients with chronic renal failure, the clini cal features and laboratory indices used as guidelines for initiation of renal replacement therapy do not correlate with survival. Objective parameters that will permit initiation of dialysis at a time that wil l maximize survival in patients with chronic renal failure are needed.