Comparison of the yield of different screening approaches to detect chronic kidney disease

Abstract

Screening for chronic kidney disease (CKD) has been advised in high-risk populations. The present study aims to compare the yields of four approaches to select high-risk subjects for CKD screening, which are defined as follows: Approach 1, history of cardiovascular (CV) disease, diabetes mellitus or hypertension (=high CV risk); Approach 2, high CV risk or age >55 years; Approach 3, urinary albumin concentration (UAC) ≥20 mg/L; or Approach 4, UAC ≥10 mg/L at pre-screening. The study population is a sample of the general population of Groningen, the Netherlands (n = 3398). UAC was measured (nephelometry) in a first morning void urine sample collected at home and sent to a laboratory by post. Information on demographics and the presence of CV risk factors was obtained by a questionnaire. The presence of CKD was determined during examination at an outpatient clinic. At baseline, 12% of the subjects met the criteria of Approach 1, 33% of Approach 2, 8% of Approach 3 and 25% of Approach 4. CKD was diagnosed in 370 subjects (11%). Approach 2 detected the most CKD patients (sensitivity 65%), while Approach 3 resulted in the lowest number needed to screen (1.9). During a follow-up of 7 years, only the UAC pre-screening approaches detected CKD patients who had both significantly accelerated renal function loss and increased CV risk compared to subjects without CKD. Only 28% of CKD patients detected by the UAC approaches used antihypertensive/angiotensin-converting enzyme inhibitor treatment prior to screening. This study suggests that pre-screening based on UAC should be favoured in comparison to screening based on CKD risk factors to detect CKD patients at high renal and CV risk.