Recent advances in the treatment of pulmonary arterial hypertension provide a rational basis for earlier, noninvasive diagnosis of pulmonary arterial hypertension. However, the reliability of transthoracic echocardiography, plasma BNP levels, and other parameters for the diagnosis of pulmonary arterial hypertension remains unclear. Thus, the purpose of this study was to determine the utility of these modes of investigation for the prediction of pulmonary arterial pressure as compared with the current gold standard, Swan-Ganz catheterization. Among 46 PAH patients, 37 had connective tissue diseases, while the remainder had primary pulmonary arterial hypertension, chronic pulmonary thromboembolism, and interstitial pneumonitis. Systolic pulmonary arterial pressure calculated by transthoracic echocardiography was significantly correlated with systolic pulmonary arterial pressure measured using a Swan-Ganz catheter (r = 0.51, P < 0.01). Plasma BNP concentration did not correlate with systolic pulmonary arterial pressure (r = 0.10, NS) in the overall patient population. However, when we excluded left ventricular heart failure and left ventricular hypertrophy, BNP concentration was correlated with systolic pulmonary arterial pressure (r = 0.508, P < 0.05). Among other variables tested, ECG electrical axis was correlated with pulmonary arterial pressure (r = 0.46, P < 0.05) but uric acid, lactate dehydrogenase, %DLCO, enhanced IIp sound, and pulmonary artery enlargement on chest x-ray did not correlate with pulmonary arterial pressure. These data suggest that echocardiography is the noninvasive modality of choice for the assessment of pulmonary arterial hypertension. Plasma BNP level also predicts pulmonary arterial pressure, when left ventricular heart failure and cardiac hypertrophy are excluded.