#826 Assessing performance of creatinine and cystatin-c based estimating glomerular filtration rate equations in South Asian populations: a systematic review and meta-analysis

Le résumé fourni par la source

Abstract Background and Aims Chronic kidney disease (CKD) prevalence continues to rise and is predicted to be the 5th leading cause of mortality worldwide by 2040. It is common in people from South Asian backgrounds and may be associated with rapidly progressive disease. Accurate measurement of kidney function is essential to diagnose, manage and classify CKD. Widely used creatinine-based estimated glomerular filtration rate (eGFR) equations had minimal South Asian representation in their development cohorts and subsequent studies in South Asians and diaspora have demonstrated reduced accuracy related to non-GFR determinants, such as muscle mass and vegetarian diets. We aimed to compare performance of creatinine (Cr)-based, cystatin C (CysC)-based and combined (Cr and CysC) eGFR equations with measured GFR (mGFR) in South Asians and diaspora. Method Systematic review was undertaken according to Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines and included the following database searches from 1946 to 30th November 2024: PubMed/Medline, Embase (Ovid), Scopus, Web of Science and Cochrane Library for Systematic Reviews and Clinical Trials. Inclusion criteria: studies which included eGFR compared to mGFR and/or measures of accuracy (bias and 30% accuracy) in South Asian populations. Sample size, demographic and renal characteristics (endogenous and exogenous filtration markers, with assay methods, eGFR, mGFR, bias and 30% accuracy) were extracted. Aggregate data, for each study and by eGFR equation were reported as weighted (by sample size) mean ± standard deviation, 95% confidence intervals or standard error and illustrated using forest plot. Heterogeneity was calculated using Cochran's Q and I2 statistic. Data were analysed using Software R 4.2.2. Results 37 studies including 35 independent cohorts (n = 4725) and 16 eGFR equations (Cr = 9, CysC = 4, combined = 3) were included. Weighted mean age was 46.8 ± 14.6 years, BMI was 25.6 ± 5.3 kg/m2 and 48.4% were male (n = 2289). Half the cohort were of Indian heritage (n = 2389; 50.6%) and the majority were prospective kidney donors or healthy volunteers (n = 2569; 54.4%). Fewer had CKD (n = 1523; 32.2%). Weighted mean mGFR was 65.8 ± 35.0 mL/min/1.73 m2. Study characteristics are summarised in Table 1. Cr-based and combined eGFR estimates frequently overestimated mGFR. MDRD 4-variable was most commonly used with weighted mean bias 3.56 [95% CI 2.80 to 4.32] mL/min/1.73 m2 and P30 65.3% (15 cohorts). The most commonly used combined equation was CKD-EPI 2012 (5 cohorts), with weighted mean bias of −2.26 [95% CI −2.51 to −2.01] mL/min/1.73 m2 and P30 74.8%. CysC equations underestimated mGFR 44.4% of the time and had the lower reported aggregate bias. The CysC CKD-EPI 2012 equation was most commonly used (6 cohorts) with bias of -1.44 [95% CI −1.60 to −1.28] mL/min/1.73 m2 and P30 68.2%. The most accurate equation was the CysC EKFC with mean bias of 0.30 [95% CI −0.13–0.73] mL/min/1.73 m2 and P30 of 60% reported in one cohort (n = 412). Weighted mean bias with heterogeneity is illustrated in Fig. 1. Conclusion To our knowledge this is the largest meta-analysis of South Asian participants. CysC-based eGFR equations are more closely correlated with mGFR compared to Cr-based or combined eGFR equations. Cr-based equations tended to overestimate mGFR with greater variability, however accuracy (weighted mean P30 = 60.4 ± 0.4%) was lower than studies in European or US cohorts. Heterogeneity was high (100%), which is likely related to variability in the cohorts, biomarkers, assays and methodologies used. Other limitations include missing covariate data in studies (e.g. BMI), small sample sizes in cohorts outside of South Asia and few participants with CKD. Overestimation of kidney function may lead to delays in diagnosis or inappropriate administration of renally excreted medication, thus negatively impacting outcomes for high-risk groups. Use of cystatin-C eGFR equations should be considered when evaluating kidney function in South Asian populations, particularly if non-GFR creatinine-based determinants are present.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.