Identifying predictors for bacterial and fungal coinfection on chest computed tomography in patients with Pneumocystis pneumonia

Background: Pneumocystis pneumonia (PCP) is a common opportunistic infection with high mortality in individuals with decreased immunity. Pulmonary coinfections with PCP are associated with poor prognosis. The study aims to identify radiological predictors for pulmonary coinfections in patients with PCP and risk factors for mortality. Methods: This is a retrospective, five-year study was conducted in a medical center, enrolling patients diagnosed with PCP, who received a chest computed tomography (CT) scan. The radio-logical findings and medical records of all participants were reviewed carefully by 2 indepen-dent doctors. Univariable and multivariable analysis was performed to identify radiological predictors for pulmonary coinfection and clinical risk factors for poor prognosis. Results: A total of 101 participants were included, of which 39 were HIV-infected and 62 were non-HIV-infected. In multivariable analysis, radiologic predictors on chest CT for coinfection with bacteria pneumonia included lack of ground glass opacity (adjusted odds ratio [aOR], 6.33; 95% confidence interval [CI], 2.03-19.77; p = 0.001) and presence of pleural effusion (aOR, 3.74; 95% CI, 1.27-10.99; p = 0.017). Predictors for fungal pneumonia included diffuse consolidation (adjusted OR, 6.27; 95% CI, 1.72-22.86; p = 0.005) and presence of pleural effusion (adjusted OR, 5.26; 95% CI, 1.44-19.17; p = 0.012). A significantly higher in-hospital mortality was associated with older age, recent corticosteroid exposure, cytomegalovirus coinfection, and acute respiratory failure. Conclusion: Early identification of pulmonary coinfections in PCP using radiological features on the CT scans, will enable appropriate treatment which is crucial to improve the prognosis. Copyright 2020, Taiwan Society of Microbiology. Published by Elsevier Taiwan LLC. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/bync-nd/4.0/).

Automated summary

Radiological predictors identified through chest CT scans can help differentiate pulmonary coinfections in patients with PCP, which is crucial for improving prognosis. Lack of ground glass opacity and presence of pleural effusion were predictors for bacterial pneumonia, while diffuse consolidation and presence of pleural effusion were predictors for fungal pneumonia.