Chest CT Findings of Pneumocystis jirovecii Pneumonia in a Large Cohort of Patients Who Are Immunocompromised.

Purpose To assess chest CT imaging manifestations of Pneumocystis jirovecii pneumonia (PJP) in a large, well-characterized cohort of patients who were immunocompromised and determine whether CT findings differ by immunocompromise etiology. Materials and Methods This retrospective study identified patients with proven or probable PJP across a multicenter hospital system over a 7-year period. Available chest CT scans were reviewed for the presence and distribution of ground-glass opacity (GGO), consolidation, interlobular septal thickening, fibrosis, nodules (characterized by size and distribution), and other findings. Imaging features and their evolution at follow-up CT were compared across immunocompromise etiologies and with the overall cohort using the Fisher exact test and Kruskal-Wallis test, as applicable. Results Among 701 evaluated patients, 675 (median age, 65 years [IQR, 54-73 years]; 423 male) had chest CT examinations available for review. Immunocompromise etiologies included hematologic malignancy (32.7%), solid neoplasia receiving chemotherapy (22.5%), autoimmune disease (17.3%), solid organ transplant (13.0%), bone marrow transplant (8.3%), corticosteroid therapy (8.1%), HIV infection (7.6%), and miscellaneous causes (4.6%). Across all etiologies, diffusely distributed GGO without distinguishing features was the most common CT finding (96.7%), with a random axial distribution in 80.1% of patients and peripheral dominance in 14.2%. Consolidation was present in 51.1% of patients. PJP CT findings substantially overlapped across immunocompromise etiologies, with no imaging pattern specific to a particular etiology. Follow-up CT was available in 71.6% of patients (median interval, 31 days; range, 2-1344 days), demonstrating PJP improvement in 66.0%, worsening in 25.5%, no change in 9.1%, and new CT patterns in 11.8%. Respiratory coinfection was present in 31.9% of patients. Conclusion PJP most commonly manifested as diffusely distributed GGO on chest CT scans, regardless of the underlying cause of immunocompromise. CT findings did not distinguish between HIV- and non-HIV-related immunosuppression, and respiratory coinfection was common. Keywords: Pneumocystis jirovecii Pneumonia, Immunosuppression, Immunocompromise, CT, Ground-Glass Opacity Supplemental material is available for this article. © RSNA, 2026.