Background Pulmonary hypertension (PH) is common in patients with heart failure (HF) but, its relationship to symptoms and prognosis in a broad representative population of patients with HF has rarely been investigated. Methods We interrogated a large database of consecutive referral to a HF clinic serving a population of 600,000 people. Systolic HF (S-HF) was defined as a LVEF≤45% and Preserved systolic LV HF (PS-HF) was defined as LVEF >45% with an NT-proBNP value >400pg/mL. Right Ventricle Systolic Pressure (RVSP) was assessed by the measurement of the tricuspid regurgitation (TR) peak systolic velocity. Results Of 2135 patients referred, 1026 had S-HF, 354 had PS-HF and 533 had an accurate TR velocity measurement. Median (IQR) calculated RVSP for HF patients was 25 (18-35mmHg). Amongst these patients, 270 had S-HF and 143 had PS-HF. Mean age, prevalence of ischaemic heart disease and arrhythmia were 74+/-10, 52% and 44%. The maximal follow up was 88 months and the median follow up was 65 months. Amongst 1,380 patients with HF, the 413 with RVSP measurement had a worse prognosis than those who did not (log Rank=0.0001). Patients in the lowest quartile of RVSP (<17.5mmHg) had a similar prognosis to the subgroup of HF patients without TR measurement (16% and 17% at 2 years). Mortality increased as function of the increase in PH (log Rank=0.005). Quartile of RVSP predicted outcome similarly in patients with S-HF and PS-HF. Patients in the highest quartile of RVSP had a 2 year mortality of 38% if S-HF and 35% if PS-HF. On Cox proportional hazards multivariate analysis, COPD, Age, logNTproBNP, mean blood pressure and the highest RVSP quartile were independent predictors of outcome. Conclusion Patients with heart failure and a TR velocity indicating even mild pulmonary hypertension have a worse prognosis whether or not LV systolic function is impaired. Patients with more severe PH have an even worse outcome. PH is an independent predictor of outcome in HF.