Glomerular filtration rate estimations in elderly cardiac rehabilitation patients

Background/Introduction Cardiovascular disease (CVD) patients admitted to exercise based cardiac rehabilitation (EBCR) often have multiple comorbidities potentially leading to chronic kidney disease (CKD). Cardiovascular index events, procedures and surgeries are also a main reason for acute kidney injury. Common equations for the calculation of estimated glomerular filtration rate (eGFR) may not be as accurate in elderly patients, due to low muscle mass or altered creatinine metabolism. This can lead to overestimation of kidney function, a challenge addressed by the berlin initiative study (BIS) eGFR equation derived specifically for patients aged 70 years and over. Purpose Our aim was to compare two eGFR calculation equations in a population of cardiac rehabilitation patients to detect possible differences in CVD patients ≥ 70 years. Methods This retrospective monocentric cohort study examined patients referred to a specialized rehabilitation center in Switzerland between December 2022 and June 2024 for inpatient EBCR. Kidney function was assessed by creatinine measurements at admission and discharge. In our population of patients aged ≥ 70 years, eGFR was calculated by applying the CKD EPI race-free formula (CKDEPIrf), as well as the creatinine-based BIS1 eGFR formula. Differences between eGFR estimations were shown with a Bland-Altman-plot. Results 271 patients were aged ≥ 70 years 267 had available creatinine measurements at admission, 270 at discharge. CKD was apparent as a diagnosis in 16.6 % of these patients. Admission CKDEPIrf eGFR was 66.8 ml/min/1.73m2 [quartiles 54.6 – 80.2] and 65.7 ml/min/1.73m2 [quartiles 52.4 – 80.4] at discharge. BIS1 eGFR calculations were lower with 55.1 ml/min/1.73m2 [quartiles 46.5 – 63.5] at admission and 53.5 ml/min/1.73m2 [quartiles 45.9 – 63.0] at discharge. The proportion of patients aged 70 years and over with CKDEPIrf eGFR values below 60 ml/min/1.73m2 was 36.9 % at discharge. This proportion was 66.1 % based on BIS1 eGFR equation results. Admission and discharge proportions were similar. Mean difference between the two equations was -11.14 ml/min/1.73m2 (95%CI -23.81 to 1.53). Although eGFR results remained comparable at the lower range, the agreement deteriorated with eGFR measurements over 50 ml/min/1.73m², as shown in the scatter, as well as in the Bland-Altman-plot (Figure 1 and Figure 2). Conclusion BIS1 equation application identified nearly twice the number of patients with eGFR values below CKD diagnosis threshold. Taking the low prevalence of CKD in this cohort at admission to EBCR into consideration, a thorough follow-up measurement of eGFR and albuminuria should be obligatory to confirm potential de novo CKD especially in elderly EBCR patients.