Background and purpose: Impaired kidney function is associated with an increased risk of vascular events in acute stroke patients, when assessed by single measurements of estimated glomerular filtration rate (eGFR). It is unknown whether repeated measurements provide additional information for risk prediction. Methods: The MonDAFIS (Systematic Monitoring for Detection of Atrial Fibrillation in Patients with Acute Ischemic Stroke) study randomly assigned 3465 acute ischemic stroke patients to either standard procedures or an additive Holter electrocardiogram. Baseline eGFR (CKD-EPI formula) were dichotomized into values of < versus ≥60 ml/min/1.73 m2. eGFR dynamics were classified based on two in-hospital values as “stable normal” (≥60 ml/min/1.73 m2), “increasing” (by at least 15% from baseline, second value ≥ 60 ml/min/1.73 m2), “decreasing” (by at least 15% from baseline of ≥60 ml/min/1.73 m2), and “stable decreased” (<60 ml/min/1.73 m2). The composite endpoint (stroke, major bleeding, myocardial infarction, all-cause death) was assessed after 24 months. We estimated hazard ratios in confounder-adjusted models. Results: Estimated glomerular filtration rate at baseline was available in 2947 and a second value in 1623 patients. After adjusting for age, stroke severity, cardiovascular risk factors, and randomization, eGFR < 60 ml/min/1.73 m2 at baseline (hazard ratio [HR] = 2.2, 95% confidence interval [CI] = 1.40–3.54) as well as decreasing (HR = 1.79, 95% CI = 1.07–2.99) and stable decreased eGFR (HR = 1.64, 95% CI = 1.20–2.24) were independently associated with the composite endpoint. In addition, eGFR < 60 ml/min/1.732 at baseline (HR = 3.02, 95% CI = 1.51–6.10) and decreasing eGFR were associated with all-cause death (HR = 3.12, 95% CI = 1.63–5.98). Conclusions: In addition to patients with low eGFR levels at baseline, also those with decreasing eGFR have increased risk for vascular events and death; hence, repeated estimates of eGFR might add relevant information to risk prediction.
