Abstract 12547: Incidence of Pulmonary Hypertension in the Echocardiographic Referral Population

Background: Incident rates (IRs) for pulmonary hypertension (PH) using diagnostic data rather than diagnostic codes (DC) have not been reported. We examined PH IRs, risk factors, and mortality among patients (pts) referred for echocardiography (TTE) in a national healthcare system and a separate tertiary care center. Methods: Pulmonary artery systolic pressure (PASP) estimates were extracted from the Veterans Affairs (1999-2020) and Vanderbilt (1994-2020) systems. PH was defined as PASP >35 mmHg. Individuals free of PH at baseline were followed through incident PH, death, or censored at end of study. Secondary outcome was mortality rate (MR) after incident PH. We used Poisson regression to estimate IRs and cox models for hazard ratios. Results: We identified 245,067 veterans (60 years [IQR 55-64], 94% male, 20% Black) and 117,526 VUMC pts (60 years [IQR 46-70], 46% male, 11% Black) free of PH at baseline. Only 22% of pts with TTE-based PH had a DC for PH. PH IRs and adjusted hazard were substantially higher in pts with HF or COPD and highest in pts with both ( Table 1 ). MR after incident PH was highest in pts with both HF and COPD and lowest in pts without either (Figure 1A ). MRs increased markedly at PASP values >45mmHg ( Figure 1B ). Independent risk factors for incident PH included older age (HR 1.07; 95%CI 1.06-1.09), male sex (HR 1.24; 95%CI 1.17-1.30), black race (HR 1.10; 95%CI 1.07-1.13), and cardiometabolic comorbidities. Conclusions: PH IRs estimated by raw diagnostic data are higher than DC-based IRs and strongly associated with HF and COPD. PASP >45mmHg at the time of incident PH is associated with high mortality.