Survival Analysis of Hemodialysis Patients: A Single Center Study.
- Author:
Mi Rye SUH
1
;
Eun Bum LEE
;
Won Seok YANG
;
Soon Bae KIM
;
Su Kil PARK
;
Sang Koo LEE
;
Jung Sik PARK
;
Chang Gi HONG
Author Information
1. Department of Internal Medicine, College of Medicine and Asan Medical Center at University of Ulsan, Seoul, Korea. jspark@amc.seoul.kr
- Publication Type:Original Article
- Keywords:
Cause of death;
Hemodialysis;
Survival
- MeSH:
Biomarkers;
Cardiovascular Diseases;
Cause of Death;
Chungcheongnam-do;
Communicable Diseases;
Creatinine;
Diabetes Mellitus;
Diabetic Nephropathies;
Follow-Up Studies;
Hepatitis;
Humans;
Kidney Failure, Chronic;
Proportional Hazards Models;
Renal Dialysis*;
Retrospective Studies;
Sex Ratio;
Survival Analysis*;
Survival Rate
- From:Korean Journal of Nephrology
2002;21(4):636-644
- CountryRepublic of Korea
- Language:Korean
-
Abstract:
PURPOSE: In order to investigate the survival rate, causes of death, and the predictors of death in hemodialysis patients with chronic renal failure. METHODS: Among the 3,462 patients who started hemodialysis in Asan Medical Center between May 1989 and December 1999, we selected 674 patients who were closely observed at least for 3 months since they began hemodialysis as an initial replacement therapy for chronic renal failure. Demographic and clinical characteristics, including diabetes mellitus, hepatitis viral marker, and laboratory findings at the beginning of hemodialysis were retrospectively analyzed RESULTS: The mean age of the patients was 51.0+/-14.8 years, and their sex ratio was 1.4:1(M:F). The proportion of patients with diabetic nephropathy was 34.4%. The mean follow-up period was 34.2+/-28.1 months and 117 patients died(17%). The greatest number of patients were found to have died of cardiac problems(23%), followed by cerebrovascular disease(16.2%), infection(13.7%) and malignant disease(11.1%). The survival rate of the 674 patients was found to range from 1 to 10 years:96% at 1 year, 88% at 2, 73% at 5, and 54% at 10 years. In order to determine the independent predictors of death, we used Cox proportional hazards model. The predictors of death in hemodialysis patients were found to be (i) male(Relative risk(RR)=1.46, p=0.06), (ii) people over 60 years of age(RR=1.54, p=0.03), diabetes mellitus(RR=1.45, p=0.06), and (iii) low level of serum creatinine(RR=0.65, p < 0.001). CONCLUSION: Cardiovascular disease and infectious diseases were important causes of death in hemodialysis patients. Old age and low serum creatinine at the beginning of hemodialysis were significant predictors of death.
