This analysis was done in a consecutive series of 419 patients with potentially reversible acute renal failure (ARF) requiring dialysis treated during the 12 year period January 1977 to December 1988. The aims were to describe the relationship between morbidity, mortality, and the year of admission, to evaluate the effects of various follow-up periods and stratification on main prognostic factors, and to report long term survival.
Results: 1) There was a significantly higher (p < 0.001) number of organ failures/patient in the last 6 year period compared with the first 6 year period. 2) Early (day 90) mortality was 46% and did not change significantly during the 12 year period. 3) Etiology, age of the patient, and the severity and the progression of the underlying disease were the most important factors determining outcome. 4) Time of censorship and stratification of the patients influenced prediction models. 5) Estimated 5 year survival rate was 52% (95% confidence limits 44-60%) in patients with a medical etiology, and 28% (95% confidence limits 20-36%) in patients with a surgical etiology.
Conclusions: 1) Prognosis did not change significantly during the 12 year study period, and was almost solely dependent on the severity and progression of the underlying disease process. 2) Changed censoring time and stratification were important tools for enhancing the yield of information from the study. 3) The time(s) and method(s) of risk scoring, stratification(s) of patients, the time(s) of censorship on vital status, and the method for statistical analysis have to be identical if results of treatment for ARF from different centers should be compared.