Abstract 14045: Pulmonary Hypertension Phenotype and Incretin-Based Therapy in CardioMEMS-Guided Heart Failure Management
Abstract
Background: Incretin-based therapies (IBT) reduce major adverse cardiovascular events in diabetes, but their impact on survival in advanced heart failure (HF) managed with invasive hemodynamic monitoring is unknown. Pulmonary hypertension (PH) phenotype at CardioMEMS implantation reflects pulmonary vascular remodeling burden and may identify patients at the highest risk of death despite hemodynamic improvement. Methods: Single-center retrospective analysis of 250 consecutive HF patients undergoing CardioMEMS implantation with 2-year follow-up. The PH phenotype was classified during right heart catheterization (RHC) using the ESC 2022 criteria. The primary outcome was all-cause mortality at 2 years. IBT use included semaglutide (n=44) and tirzepatide (n=16); all users had diabetes. Multivariable Cox regression and 1:2 propensity score matching (PSM), adjusting for age, sex, EF, GFR, BMI, and PH phenotype, were used to evaluate the independent association of IBT with mortality. Results: Among 250 patients (mean age 70.4±11.8 years; 45.6% women), PH distribution was: no PH 20 (8%), pre-capillary PH 56 (22%), isolated post-capillary PH (IpcPH) 90 (36%), and combined pre- and post-capillary PH (CpcPH) 84 (34%). IBT, including semaglutide and tirzepatide, was used in 60 patients (24%) and was distributed evenly across PH phenotypes (p=0.68). Despite the greatest hemodynamic response, CpcPH had the highest 2-year mortality (47.6%). Mean 12-month PAD reduction was similar between IBT users and non-users (–5.5 vs –5.6 mmHg, p=0.80). IBT users had lower 2-year all-cause mortality than non-users (16.4% vs 49.7%). On multivariable Cox regression, IBT was independently associated with lower mortality (HR 0.36, 95% CI 0.19-0.69; p=0.002), confirmed in the PSM cohort (17.3% vs 52.0%; p=0.004). This association was directionally consistent across PH phenotypes (Figure 1). Conclusions: In CardioMEMS-guided HF management, IBT use was independently associated with lower 2-year all-cause mortality after adjustment for PH phenotype and propensity score matching. The highest mortality in CpcPH despite the greatest hemodynamic response suggests that filling pressure reduction alone may not overcome advanced pulmonary vascular remodeling. Similar PAD reduction between IBT users and non-users suggests that this survival association may not be explained solely by hemodynamic improvement. These findings support further prospective investigation of incretin-based therapies in advanced HF populations.