Survival in primary pulmonary hypertension. Validation of a prognostic equation.

Abstract

BACKGROUND The prognosis of patients with primary pulmonary hypertension (PPH) remains a major problem for the planning and assessment of therapeutic interventions. The objectives of this study were (1) to characterize mortality in a Mexican population of patients with PPH and to investigate factors associated with survival and (2) to test the applicability in this population of the prognostic equation proposed by the US National Institutes of Health study on PPH. METHODS AND RESULTS A dynamic cohort of patients with PPH at our institution were enrolled between June 1977 and August 1991 and prospectively followed at regular intervals through September 1992. Measurements at diagnosis included hemodynamic and pulmonary function variables in addition to information on demographic data and medical history. The response to vasodilator treatment was also analyzed. The estimated median survival of the group was 4.04 years (95% confidence interval, 2.98 to 5.08 years). Variables associated with poor survival (univariate analysis) included an elevated mean right atrial pressure, a decreased cardiac index, and a decreased mixed venous PO2. A reduced forced vital capacity and the absence of vasodilator treatment were also associated with poor survival. A multivariate Cox proportional-hazards regression analysis was used to assess the adjusted hazard ratios, hence the relative contributions of the variables controlling for confounding. Reduced forced vital capacity and cardiac index and increased right atrial pressure were still significantly associated as risk factors for survival in patients with PPH. Survival as computed by the equation correlated with real survival of PPH patients with positive predictive values of 87%, 91%, and 89% at 1, 2, and 3 years, respectively. The equation, however, was relatively unable to predict deaths in our population, in part because of the strict limits of poor prognosis. CONCLUSIONS Mortality in PPH is largely associated with hemodynamic variables that assess right ventricular function. The proposed prognostic equation had a high sensitivity and a relatively low specificity to predict survival in our PPH population. To improve this specificity it may be necessary to increase the limits of poor prognosis as defined by the equation.