Back
Sodium-glucose co-transporter-2 (SGLT2) inhibitors slow kidney disease progression and reduce cardiovascular risk in patients with CKD. Frailty is a state of increased vulnerability due to aging-related decline in physiological reserve and is common in CKD. Without randomized evidence, perceived altered risk-benefit profile may lead to under-treatment of frail patients. We sought to derive a frailty indicator and use it to assess the risk-benefit profile of empagliflozin in CKD according to levels of frailty in EMPA-KIDNEY.
EMPA-KIDNEY was a double-blind, placebo-controlled randomized trial which compared empagliflozin 10 mg once daily with placebo among 6609 patients with CKD (Clinicaltrials.gov: NCT03594110). Eligible patients had an eGFR of 20 to <45; or 45 to <90 ml/min/1.73m2 with a urinary albumin-to-creatinine ratio (uACR) ≥200 mg/g.
“Risk of hospitalization during follow-up” at baseline was used as a frailty indicator. Multivariable logistic regression models adjusted for age, sex and region assessed the association of all potential predictor variables with recorded hospitalization (first event) using forward stepwise selection and likelihood ratio tests with significance threshold P<0.01. Model performance was assessed using the area under the receiver operating characteristic curve (AUROC) for all-cause hospitalization and subsequently validated for death from any cause. Participants were categorized into approximate thirds of predicted risk of hospitalization (%), and the top third dichotomized. The effects of allocation to empagliflozin versus placebo on the trial’s pre-specified efficacy and safety outcomes were assessed using Cox regression models adjusted for age, sex, region, eGFR, uACR and diabetes status. Tests for trend across risk of hospitalization subgroups were used to assess for any evidence that frailty modifies treatment effects.
Median (Q1-Q3) follow-up was 2.0 (1.5-2.4) years, during which time median predicted risk of hospitalization was 27% (Q1-Q3: 18-40%). The strongest predictors of hospitalization were the log transformation of N-terminal prohormone of brain natriuretic peptide (baseline median [Q1-Q3] 160 [69-419] ng/L); poor mobility (based on EQ-5D-5L) and the presence of diabetes. The final model which additionally included eGFR and other comorbidities adequately predicted risk of hospitalization (AUROC [95% CI] 0.70 [0.69-0.71]) and death from any cause (AUROC 0.80 [0.78-0.82]). Overall, efficacy and safety event rates were higher in frailer patients. Empagliflozin reduced the risk of the primary composite outcome of kidney disease progression or cardiovascular death by 28% (hazard ratio 0.72, 95% CI 0.64-0.82), with no significant difference in relative effects by baseline level of frailty (P for trend=0.60, Figure). However, absolute benefits for the primary outcome increased with increasing frailty (P for trend <0.001, Figure). Safety outcomes were infrequent with no evidence of trend to suggest increased relative risk in frailer patients allocated to empagliflozin, including no excess of symptomatic dehydration or fractures.
Empagliflozin safely lowered risk of progression of CKD or death from cardiovascular causes among a broad range of patients with CKD, irrespective of baseline frailty. Absolute benefits for the primary outcome were greater in frailer patients.