Detecting chronic kidney disease in older people; what are the implications?

Background: National policy is focused on early identification, referral and management of chronic kidney disease (CKD) to prevent both progression to endstage renal failure and cardiovascular disease. However, the significance of identifying CKD in older people is unclear. Objective: To determine the frequency of CKD in older people using estimated glomerular filtration rate (eGFR), and its associations with morbidity and functional measures. Design: Observational cross-sectional analysis of baseline data from a large cluster randomised trial of health and social assessment of older people in the community. Setting: Included 53 general practices in Great Britain. Subjects: Subjects were people aged 75 and over, living in the community participating in the trial arm where systematic blood testing was undertaken. Methods: The response rate for participation at baseline assessment of those eligible was 73% (15,536/20,934), of whom 13,109 (86%) participants had a serum creatinine measured, and an eGFR derivable using the Modification of Diet in Renal Disease formula (MDRD) in ml/min/ 1.73 m². Key outcomes were the prevalence of CKD stages and their associations with morbidity and functional status. Results: prevalence of CKD was 56.1% (95% CI 55.3-57.0) for eGFR < 60, 17.7% for eGFR <45 (95% CI 17.1-18.4), and 2.7% (95% CI 2.4-2.9) for eGFR <30. It was higher in older ages, females, and those with cardiovascular comorbidity and doctor-diagnosed hypertension but not with diabetes. The strength of the association with measures of morbidity and functional impairment increased as eGFR fell, especially once the eGFR was <45. For example, the odds ratios in males for anaemia for an eGFR <30, 30-44 and 45-59 versus reference GFR <60 were 8.3 (5.1-13.7), 3.0 (2.1-4.2) and 1.2(0.8-1.7) respectively; similar figures for partial dependence on activities of daily living were 2.2 (1.4-3.3), 1.6 (1.2-2.1) and 1.0 (0.9-1.3) and for lack of physical activity 2.20 (1.39-3.48), 1.78 (1.37-2.32) and 1.10 (0.92-1.32). Conclusions: An eGFR <60 is very common in older people. An eGFR <45 identifies a smaller sub-group of older people with significant comorbidity, impaired functional state and a high risk of potentially reversible consequences such as anaemia. The benefits of identifying older people with an eGFR >45 need to be determined.