Introduction: There has been limited data in comparison of Pneumocystis pneumonia (PCP) outcomes and course of illness among non-HIV immunocompromised cohorts. This study aims to assess the manifestations and outcomes of PCP in diverse immunocompromised populations. The findings may be helpful for identifying risk factors of destructive consequences in this life-threatening fungal infection.
Methods: We retrospectively reviewed and collected data including clinical presentations, course of illness, comorbidities, laboratory data, imaging studies, and treatment of patients diagnosed with PCP at University Health Network, Toronto, Canada, from January 1, 2011, to December 31, 2021. The study outcome was a composite variable including 21-day ICU admission and/or 28-day all-cause mortality following PCP diagnosis. Risk factors of the composite outcomes were preliminarily identified by univariate analysis methods, then we used multivariable logistic regression analysis and estimated adjusted OR (aOR) to determine the concurrent effects of covariates.
Results: During the specific time frame, a total of 182 immunocompromised patients including 19 (10.4%) HIV infection, 54 (29.7%) hematologic malignancies, 32 (17.6%) hematopoietic stem cell transplantation, 33 (18.1%) solid tumors, 26 (14.3%) solid organ transplantation, 12 (6.6%) autoimmune diseases, and 6 (3.3%) other immunosuppressive conditions fullfilled the PCP diagnostic criteria. The median age was 59.5 (interquartile range [IQR], 45.5-69.5 years) with 112 male (61.5%). 83 (45.6%) of them were on chronic steroid treatment at diagnosis. Prolonged (4 weeks at the minimum) lymphopenia and neutropenia were present in 115 (64.6%) and 27 (15.2%) patients, respectively. Overall, 44/182 patients (24.2%) developed composite outcomes with 39 ICU admissions (21.4%) and 17 (9.3%) deaths. In univariate analysis, significant risk factors were SOT [13 (29.5%) vs 13 (9.4%), p < 0.001], chronic liver disease [15 (34.1%) vs 9 (6.5%), p < 0.001], prolonged lymphopenia [37 (84.1%) vs 78 (56.5%), p < 0.001], hypoxemia [24 (54.5%) vs 55  (39.9%), p = 0.037], pulmonary nodules [18 (40.9%) vs 95 (68.8%), p = 0.001], and pleural effusion [14 (31.8%) vs 13 (9.4%), p < 0.001]. Finally, SOT (aOR 3.8, IQR 1.1-13.8), prolonged lymphopenia (aOR 8.8, IQR 2.1-37.4), and chronic liver disease (aOR 6.3, IQR 1.8-22.2) were statistically significant predictive variables of the composite outcomes from multivariate analysis.
Conclusions: SOT recipients appear to be at a greater risk of combined ICU admission and mortality outcomes compared to other non-SOT immunocompromised patients infected with PCP. Prolonged lymphopenia and chronic liver disease were also correlated with unfavorable composite outcomes in these cohorts.