J Am Heart Assoc. 2021 May 20:e020237. doi: 10.1161/JAHA.120.020237. Online ahead of print.
Background The frequency of the initial short-term decline in estimated glomerular filtration rate (eGFR), eGFR dip, following initiation of sodium-glucose cotransporter-2 inhibitors (SGLT2i) and its clinical implications in real-world practice are not clear. Methods and Results We built a cohort of 36 638 new users of SGLT2i and 209 025 new users of other antihyperglycemics. Inverse probability weighting was used to estimate the excess rate of eGFR dip, risk of the composite cardiovascular outcome of nonfatal myocardial infarction, nonfatal stroke, hospitalization for heart failure, or all-cause mortality, and risk of the composite kidney outcome of eGFR decline >50%, end-stage kidney disease, or all-cause mortality. In the first 6 months of therapy, compared with other antihyperglycemics, excess rates of eGFR dip >10% and eGFR dip >30% were 9.86 (95% CI: 8.83-11.00) and 1.15 (0.70-1.62) per 100 SGLT2i users, respectively. In mediation analyses that accounted for eGFR dipping, SGLT2i use was associated with reduced risk of cardiovascular and kidney outcomes (hazard ratio, 0.92 [0.84-0.99] and 0.78 [0.71-0.87], respectively); the magnitude of the association reduced by eGFR dipping was small for both outcomes. SGLT2i was associated with reduced risk of both outcomes in those with higher than average probability of eGFR dip >10% or 30%. Compared with discontinuation, continued use of SGLT2i at 6 months was associated with reduced risk of cardiovascular and kidney outcomes in those with no eGFR dip or eGFR dip ≤10%, in those with eGFR dip >10%, and in those with eGFR dip >30%. Conclusions The salutary association of SGLT2i with cardiovascular and kidney outcomes was maintained regardless of eGFR dipping; concerns about eGFR dipping should not preclude use, and occurrence of eGFR dip after SGLT2i initiation may not warrant discontinuation.