Predicting renal function response to renal artery stenting.

Objective: The Cardiovascular Outcomes in Renal Atherosclerotic Lesions Trial found no benefit of renal artery stenting (RAS) over medical therapy, although it was underpowered to detect a benefit among patients with chronic kidney disease (CKD). A post hoc analysis demonstrated improved event-free survival after RAS for patients whose renal function improved by 20% or more. A significant obstacle to achieving this benefit is the inability to predict which patients' renal function will improve from RAS. The objectives of the current study were to identify predictors of renal function response to RAS.

Methods: The Veteran Affairs Corporate Data Warehouse was queried for patients who underwent RAS between 2000 and 2021. The primary outcome was improvement in renal function (estimated glomerular filtration rate [eGFR]) after stenting. Patients were categorized as responders if the eGFR at 30 days or greater after stenting increased by 20% or more compared with before stenting. All others were nonresponders.

Results: The study cohort included 695 patients with a median follow-up of 7.1 years (interquartile range, 3.7-11.6 years). Based on postoperative change in eGFR, 202 stented patients (29.1%) were responders, and the remainder (n = 493 [70.9%]) were nonresponders. Before RAS, responders had a significantly higher mean serum creatinine, lower mean eGFR, and higher rate of decline of preoperative GFR in the months before stenting. After stenting, responders had a 26.1% increase in eGFR, compared with before stenting (P < .0001), which remained stable during follow-up. In contrast, nonresponders had a progressive 5.5% decrease in eGFR after stenting. Logistic regression analysis identified three predictors of renal function response to stenting: (1) diabetes (odds ratio [OR], 0.64; 95% confidence interval [CI], 0.44-0.91; P = .013), (2) CKD stages 3b or 4 (OR, 1.80; 95% CI, 1.26-2.57; P = .001), and (3) rate of decline in preoperative eGFR per week before stenting (OR, 1.21; 95% CI, 1.05-1.39; P = .008). CKD stages 3b and 4 and the rate of decline in preoperative eGFR are positive predictors of renal function response to stenting, whereas diabetes is a negative predictor.

Conclusions: Based on our data, patients in CKD stages 3b and 4 (eGFR 15-44 mL/min/1.73 m) are the only subgroups with a significant probability of improved renal function after RAS. The rate of decline of preoperative eGFR over the months before stenting is a powerful discriminator of patients who are most likely to benefit from RAS. Specifically, patients with a more rapid decrease in eGFR before stenting have a significantly greater probability of improved renal function with RAS. In contrast, diabetes is a negative predictor of improved renal function, so interventionalists should be circumspect about RAS in diabetic patients.