Background: Current management guidelines for pulmonary arterial hypertension (PAH) recommend a treatment choice based primarily on World Health Organization (WHO) functional class. This study was designed to assess how the incorporation of readily obtained clinical and test-based information may significantly improve the prediction of outcomes over functional class alone.
Methods: Clinical and hemodynamic variables were assessed in 484 consecutive patients presenting with WHO group 1 PAH. The primary outcome measure was time to all-cause mortality over 5 years from the index presentation (data available in all). Follow-up was censored at the time of lung or heart/lung transplant in 21 patients or at 5 years. Predictors of mortality were assessed sequentially using Cox models, with the step-wise incorporation of clinical variables, echocardiographic, and catheterization findings. Results were further compared with the REVEAL (Registry to Evaluate Early and Long-term Pulmonary Arterial Hypertension Disease Management) prediction score.
Results: Overall median survival was 237 weeks (95% CI, 196-266), corresponding to 1-year, 3-year, and 5-year survival rates of 81.1% (77.0, 84.7), 61.1% (56.5, 65.3), and 47.9% (43.2, 52.4), respectively. The prediction of mortality was improved incrementally by incorporating clinical and echocardiographic measures with a concordance index (c-index) of 0.84 compared with that of 0.60 with functional class alone. The REVEAL prediction score was validated independently in this cohort to predict both 1-year and 5-year mortality. It had a prediction c-index of 0.71.
Conclusions: The integration of routine PAH clinical (predominantly noninvasive) parameters predicts long-term outcome better than functional class and, hence, should be incorporated into medical management decisions.