Association of residual pulmonary hypertension with sudden cardiac death after hospitalisation for acute decompensated heart failure

Background Sudden cardiac death (SCD) remains a major cause of mortality in patients with heart failure (HF) despite advances in medical and device therapy. Risk stratification relies largely on left ventricular ejection fraction (LVEF), yet many SCD events occur outside conventional implantable cardioverter-defibrillator (ICD) eligibility criteria. Residual pulmonary hypertension (PH) after hospitalisation for acute decompensated HF (ADHF) predicts adverse outcomes, but its association with subsequent SCD remains unclear. Purpose We aimed to investigate whether residual PH at discharge is associated with post-discharge SCD in patients hospitalised for ADHF. Methods We analysed consecutive patients hospitalised for ADHF in a multicentre prospective cohort between 2006 and 2017. Patients with missing echocardiographic data, loss to follow-up, or in-hospital death were excluded. Residual PH was defined as peak tricuspid regurgitant velocity (TRV) >2.8 m/s at discharge. The primary outcome was SCD or appropriate ICD shock. Associations were evaluated using Fine–Gray competing-risk models treating non-SCD death as a competing event and cause-specific Cox models adjusted for the Seattle Proportional Risk Model (SPRM). TRPG was additionally analysed as a continuous variable. Prespecified subgroup analyses and interaction testing were performed according to LVEF category (≤40% vs >40%), HF aetiology (ischaemic vs non-ischaemic), discharge guideline-directed medical therapy (GDMT) use, and both in-hospital change and discharge levels of natriuretic peptides. Results Among 2,988 patients (median age 77 years, 40% female, median LVEF 45%), 41.5% had residual PH. During a median follow-up of 692 days (IQR 266–1,125), 173 patients (5.8%) experienced SCD or ICD shock. The cumulative incidence was higher in patients with residual PH (7.3% vs. 4.7%; Gray’s test p=0.013, Figure 1). After SPRM adjustment, residual PH remained independently associated with SCD or ICD shock (subdistribution HR (sHR) 1.49, 95%CI 1.11–2.00; p=0.009), consistent with cause-specific Cox analysis (HR 1.55, 95%CI 1.15–2.09; p=0.004). Each 10-mmHg increase in TRPG was independently associated with higher risk (adjusted sHR 1.08, 95%CI 1.03–1.13; p<0.001, Figure 2). The association was consistent across LVEF categories, HF aetiology, and discharge GDMT status, and was not significantly modified by in-hospital change or discharge levels of natriuretic peptides (all p for interaction >0.05). Conclusions Residual PH at discharge was independently associated with subsequent SCD after ADHF. Assessment of residual PH may improve post-discharge arrhythmic risk stratification beyond established clinical predictors and natriuretic peptide response.