Objective: The objective of this study was to determine whether reduced lung diffusing capacity for carbon monoxide (DLCO) predicts the presence of pulmonary hypertension (PH) in heterogeneous group of patients.
Methods: Consecutive patients who underwent pulmonary function and transthoracic echocardiographic testing within a 6-month period were retrospectively identified by chart review. Right ventricular systolic pressure (RVSP) was measured using Doppler echocardiography. PH was defined as a RVSP >30 mm Hg. Patients were divided into 2 major groups: those whose RVSP was < or =30 mm Hg and those whose RVSP was >30 mm Hg. PH was classified as mild (RVSP = 31-40 mm Hg), moderate (RSVP = 41-60 mm Hg) and severe (RVSP >60 mm Hg).
Results: A total of 398 patients were entered into the study: 264 (66.3%) with PH and 134 (33.7%) without PH. Patients with PH were older, had a lower mean percentage of predicted values for forced vital capacity (%FVC), for DLCO (%DLCO) and for DLCO corrected for alveolar volume (%DLCO/VA) than those without PH. Multivariate analysis identified age and %FVC as independent predictors of PH [odds ratios and 95% confidence intervals of 1.038 (1.020-1.056) and 0.972 (0.955-0.988), respectively]. There was a weak, but significant negative correlation between RVSP and %DLCO (r = -0.205, P = 0.001), but there was no correlation between RVSP and %DLCO/VA. Neither %DLCO nor %DLCO/VA was found to be independent predictors of PH.
Conclusion: In a heterogeneous group of patients, %DLCO and %DLCO/VA do not independently predict the presence of PH.