PREDICTORS OF MORTALITY IN LONG-TERM HEMODIALYSIS-PATIENTS WITH A LOW-PREVALENCE OF COMORBID CONDITIONS

Citation
Jjg. Delima et al., PREDICTORS OF MORTALITY IN LONG-TERM HEMODIALYSIS-PATIENTS WITH A LOW-PREVALENCE OF COMORBID CONDITIONS, Nephrology, dialysis, transplantation, 10(9), 1995, pp. 1708-1713
Citations number
21
Categorie Soggetti
Urology & Nephrology",Transplantation
ISSN journal
09310509
Volume
10
Issue
9
Year of publication
1995
Pages
1708 - 1713
Database
ISI
SICI code
0931-0509(1995)10:9<1708:POMILH>2.0.ZU;2-0
Abstract
Background. Total and cardiovascular mortality rates for haemodialysis patients are still high despite the continuous improvement of dialysi s technology. This trend may be a consequence of the increased number of elderly patients and patients with concurrent systemic diseases adm itted to dialysis programmes. The objective of the present investigati on was to determine the predictors of mortality more closely related t o uraemic status and to dialysis treatment by studying relatively youn g haemodialysis patients with a low prevalence of comorbid conditions using the Cox proportional hazards model. Methods. Seventy-four haemod ialysis patients with a low prevalence of extrarenal diseases (mean ag e 44.5 +/- 12.1 years, mean dialysis duration 51.3 +/- 36.1 months) we re submitted to extensive cardiological and clinical evaluation and fo llowed prospectively for a median period of 60.8 months (range: 5.1 - 79.8). There were only three diabetics and all patients denied previou s myocardial infarction. Results. At follow-up 23 patients had died, 1 0 of cardiac causes. Five-year survival rates were 70% for overall mor tality and 83% for cardiac mortality. Age over 44 years (relative risk 3.73; 95% confidence interval 1.35 - 10.26) and serum creatinine (inv erse correlation) (relative risk 0.73; 95% confidence interval 0.57 - 0.94) were shown to be independently associated with global mortality by the Cox proportional hazards analysis. Gender, race, dialysis durat ion, interdialytic weight gain, angina, NYHA classification, hypertens ion, LV hypertrophy, LV systolic and diastolic dysfunction, complex ve ntricular arrhythmias and altered myocardial perfusion tests were not significant predictors of overall mortality. There were crude associat ions between cardiac mortality and cardiothoracic and Sokolow indices (P < 0.01) and age, NYHA classification, systolic hypertension, LV wal l and septum hypertrophy, LV dilatation and complex ventricular arrhyt hmias (P < 0.05). Conclusion. In the absence of serious comorbid condi tions, the survival of patients on long-term haemodialysis is influenc ed by nutritional status (as indicated by lower serum creatinine) and age.