The Renal Composite Benefit of Sodium Glucose Co-Transporter 2 Inhibitors Should Ideally Be Assessed Based on a Standardised Definition: A Meta-Analysis of Randomised Controlled Trials

Abstract

(1) Background: Chronic kidney disease (CKD) is extremely common against the backdrop of type 2 diabetes (T2D), accounting for nearly 30–40% of cases. The conventional management strategy relie predominantly on metabolic control and the renin–angiotensin–aldosterone system (RAAS) blockage. In the last decade, sodium glucose cotransporter 2 inhibitors (SGLT-2is) have emerged as the leading molecules preventing the development of, as well as retarding, the progression to CKD. Although the evidence in support of SGLT-2is is overwhelming, the definition of renal composite outcome in the trials varied considerably. The aim of the present meta-analysis was to explore the robustness of the renal composite benefits using a uniform definition. (2) Methods: A web-based search was conducted using the Cochrane Library to identify the relevant articles for meta-analysis. RStudio (1 July 2022, Build 554) software was used to conduct the meta-analysis. Hazard ratio (HR) was the effect size used to estimate the renal composite benefit, and prediction interval was used to detect heterogeneity. In view of the differing baseline characteristic of the trials as well as different molecules used, a random effects model was used. (3) Results: There were 12 trials including 78,781 patients, identified using the search strategy, and a five-point Cochrane risk-of-bias was used to assess quality of the publications. In the overall estimation (irrespective of the definition used for the renal composite) the HR was 0.68 (95% CI 0.60–0.76, prediction interval: 0.48–0.95) in favour of SGLT-2is, devoid of heterogeneity. While using a uniform definition of eGFR ≥ 40%decline, ESKD, or renal death, the HR was 0.64 (95% CI 0.53–0.78); using eGFR ≥ 50%decline, ESKD, or renal death the HR was 0.75 (95% CI 0.59–0.97); and with doubling of serum creatinine, renal replacement therapy, or renal death, the HR was 0.67 (95% CI 0.55–0.83) in favour of SGLT-2is. However, significant heterogeneity was encountered with all these three definitions. (4) Conclusion: There is a need to analyse the renal outcomes using a uniform definition in future trials. The presence of heterogeneity might disappear with the pooling of larger number of trials. However, if heterogeneity persists, we need to identify other clinical or laboratory attributes (in addition to SGLT-2is) responsible for the positive renal outcomes.

Keywords

• SGLT-2is
• renal composite
• eGFR decline
• meta-analysis
• prediction interval
• uniform definition