Pulmonary Artery Diameter Predicts Mortality in Community-Acquired Pneumonia

The impact of pneumonia on pulmonary hemodynamics may be underestimated by available risk scores, such as CURB-65 and the Pneumonia Severity Index (PSI). We evaluated the prognostic utility of pulmonary artery diameter (PAD) in pneumonia.

We retrospectively screened 1,887 CT pulmonary angiography scans. Pneumonia was defined as a pulmonary infiltrate accompanied by clinical or laboratory evidence of infection. Patients' characteristics and outcomes were extracted from medical records. PAD was measured using three methods (PAD1-3) and tested for association with mortality.

A total of 243 pneumonia cases were included. PAD1-3 showed strong inter-method agreement (ICC=0.88, P<0.0001). PAD1 showed the most consistent association with mortality and was widely applicable. PAD1 was larger in non-survivors at 30 days (29.6 vs. 26.4 mm, P=0.0003), 90 days (28.5 vs. 26.4 mm, P=0.002), and one year (28.0 vs. 26.4 mm, P=0.01). Thirty-day mortality increased stepwise across PAD1 tertiles (3.8%, 13.9%, and 22.8%, P=0.001), as did 90-day and one-year mortality. PAD1 was an independent predictor of 30-day mortality (HR 1.11 per mm, 95% CI 1.02-1.22, P = 0.02). PAD1, CURB-65, and PSI showed comparable performance in predicting 30-day mortality (ROC-AUC=0.70-0.73).

PAD is independently associated with mortality in pneumonia and demonstrates comparable predictive performance to CURB-65 and PSI for 30-day mortality. These findings support incorporating vascular imaging markers into risk stratification scores.