Prognosis and progression in chronic kidney disease

Abstract

Chronic kidney disease (CKD) is a life threatening condition with high risk of pre-term death and need for dialysis. The population prevalence of CKD has been estimated to 10-13% for both Europe and the USA. We investigated if analgesic use, occupational lead exposure and other patient characteristics were related to decline in kidney function in a population-based cohort of 920 patients with CKD. We also studied how survival related to timing of dialysis initiation.

The inclusion took place in Sweden between May 20 1996 and May 31 1998. All Swedish-born patients, 18-74 years old, who had been tested with a serum creatinine ≥300μmol/l for men and 250μmol/l for women for the first time were included and interviewed. At three occasions, during 1996-1998, age and sex-matched controls were included from the general population for comparison. Through linkages with the Swedish Renal Registry, the Swedish Population and Cause of Death Registry, and the patients medical records we followed these patients until either death, start of renal replacement therapy (RRT [dialysis or transplantation]) or Dec 31 2005.

At the end of the follow-up 756 patients had initiated RRT while 46 were still alive and without RRT. After one and three years, the proportion of patients alive and without RRT was 64% and 29%, respectively. The mean unadjusted decline in estimated glomerular filtration rate (eGFR) was 9.0ml/min/1.73 m2 per year, while the median decline was 5.1ml/min/1.73 m2 per year. Younger patients, those with higher blood pressure, and more albuminuria had a faster decline in glomerular filtration rate. However, patients with regular acetaminophen or aspirin use progressed with -5.1/-4.4 ml/min/1.73m2 compared to -5.3/-5.1 ml/min/1.73m2 among non-regular users. There was no difference in progression rate among patients with a high lifetime cumulative analgesic use compared to non-exposed. We could not detect any significant risk of CKD with occupational lead-exposure (OR 0.97, 95% CI 0.7-1.4). Neither did patients who had been occupationally exposed to lead differ in progression rate or risk for RRT compared to those who had never been lead exposed. Mortality was high both before and after dialysis, one and five-year survival was 97% and 61%. Compared to the general population the adjusted standardized mortality ratio was 8.3 (95% CI 7.5-9.2). Patients with diabetic nephropathy, low body mass index, and high co-morbidity score, blood pressure, and albuminuria had a significantly higher risk of death.

Mortality increased by eGFR and progression rate. The HR (death) for eGFR <7.5 ml/min/1.73 m2 was 4.65 (95% CI 1.28, 9.49) compared to non-RRT patients with eGFR 7.5-10ml/min/1.73 m2. After dialysis start mortality increased further; the HR for patients who had started dialysis was 2.64 (95% CI 1.80, 3.89) relative to patients who had not yet started dialysis. However, timing of dialysis initiation was not associated with survival. The HR for patients initiating dialysis with an eGFR <7.5 ml/min/1.73 m2 was 0.84 (95% CI 0.64-1.10) relative to those who started dialysis at higher eGFR. In summary, most patients with Stage 4 or 5 CKD progress to RRT, and the factors that most importantly affect the progression rate are age, blood pressure and proteinuria. Mortality is very high in this population, relates to eGFR and does not improve after initiation of dialysis.