CLINICAL EVENT REDUCTIONS IN HIGH-RISK HYPERTENSION PATIENTS TREATED WITH RENAL DENERVATION: 10-YEAR PROJECTIONS BASED ON 3-YEAR FOLLOW-UP FROM THE GLOBAL SYMPLICITY REGISTRY

Abstract

Objective: Clinical event rates over 3 years have been published for patients with uncontrolled hypertension treated with radiofrequency renal denervation (RDN) in the Global SYMPLICITY Registry (GSR). We estimated 10-year clinical event reductions following RDN vs. a simulated control group without RDN, for two high-risk subgroups and the full GSR cohort. Design and method: A Markov model was used to compare projected 10-year clinical events [stroke, myocardial infarction (MI), cardiovascular death (CVD), heart failure (HF), all-cause death (ACD), and a composite of major adverse cardiac events (MACE, calculated as sum of stroke, MI, and CVD)] for RDN patients vs. a hypothetical control for the high atherosclerotic vascular disease risk (ASCVD) subgroup (n = 737; 69 ± 8 years), the type-2 diabetes mellitus (T2DM) subgroup (n = 1,007; 64 ± 10 years), and for all GSR patients (n = 2,651; 61 ± 12 years). The simulated control assumed maintenance of baseline office systolic blood pressure (oSBP) over time. The model was calibrated to 3-year stroke and MI events reported in the GSR and used published meta-regression data to calculate risk reduction based on cohort-specific changes in oSBP from baseline. Relative risks (RRs), events avoided, and numbers needed to treat (NNTs) were calculated at 3 and 10 years, along with the ratio of 10- vs. 3-year projected events avoided. Results: 10-year MACE events were 35.8% vs. 48.5% (-12.7%, RR = 0.74) for RDN vs. control for all GSR patients, 37.8% vs. 52.4% (-14.6%, RR = 0.72) for the high ASCVD risk cohort, and 40.5% vs. 54.0% (-13.5%, RR = 0.75) for the T2DM cohort. Across all three cohorts, events avoided were highest for stroke and lowest for MI. 10-year NNTs for MACE were comparable between the three cohorts, estimated between 7 to 8. The ratio of model-projected MACE at 10- vs. 3-years was 5.20, 5.46, and 5.03 for the full GSR, high ASCVD risk, and T2DM cohorts, respectively. Conclusions: Model-based projections based on 3- year events observed in the GSR suggest meaningful event reductions after treatment with radiofrequency RDN at 10 years that might be up to 5-times higher than the projected events avoided at 3 years under the assumption of maintained treatment effect versus baseline.