Validation of the pulmonary hypertension connection equation for survival prediction in pulmonary arterial hypertension.

Abstract

The pulmonary hypertension connection (PHC) equation predicts contemporary survival in idiopathic, heritable, and anorexigen-associated pulmonary arterial hypertension (PAH). The aim of this study is to validate the PHC equation in a prospective PAH population cohort and compare its predictability with the French equation. We compared the rates of actual survival in patients prospectively followed for up to 3.5 years in four double-blind, randomized trials and their open-label extension studies with predicted survival calculated using the PHC equation [(P(t) = e((-A(x,y,z)t)), A(x,y,z) = e((-1.270-0.0148x + 0.0402y - 0.361z)), where P(t) is the probability of survival, t the time interval in years, x the mean pulmonary artery pressure, y the mean right atrial pressure, and z the cardiac index] and the French equation in patients with idiopathic, heritable, and anorexigen-associated PAH (n = 449). Mean age was 44 ± 15 years, 77% were women, and 80% had World Health Organization (WHO) functional class III/IV symptoms. The mean 6-min walk distance (6MWD) was 354 ± 95 m. The baseline hemodynamics were as follows: mean right atrial pressure 10 ± 6 mm Hg, mean pulmonary artery pressure 59 ± 15 mm Hg, and cardiac output 4.1 ± 1.5 L/min. The 1-, 2-, and 3-year Kaplan-Meier survival rates were 89%, 80%, and 70%, respectively; the nonadjusted survival rates were 91%, 87%, and 84%, respectively. The expected survival predicted by both the PHC and the French equations was similar to the actual observed Kaplan-Meier survival and was within its 95% confidence limits. The PHC equation also performed well when used in patients with WHO functional class III/IV, cardiac output < 4 L/min, or 6MWD < 380 m. Risk prediction equations (PHC and French) accurately predicted survival and may be useful for risk estimation in patients with idiopathic, heritable, and anorexigen-associated PAH in large cohort studies. Their use for survival prediction for individual patients needs further study.