Sex-Disaggregated Analysis of Pneumonia in Patients with High-Risk Neutropenia.

Introduction: Pneumonia is a common and serious complication during high-risk neutropenia in patients with cancer. Even though sex differences were described in patients with infectious diseases or cancer in general, sex-specific analyses for pneumonia are lacking. This exploratory study aimed to compare epidemiology and outcome of pneumonia between men and women in this high-risk cohort.

Methods: Patient data were harmonized from four primary databases collected by our research group at two tertiary care centers in Germany. High-risk neutropenia was either defined by duration of neutropenia or assumed for autologous or allogeneic hematopoietic stem cell transplantation. All patients who developed pneumonia associated with febrile neutropenia were stratified by sex, and their characteristics during the first observed pneumonia case were compared. Additionally, all identified causative pathogens were stratified by sex and described.

Results: In total, 906 patients contributed 1,511 cases of high-risk neutropenia. Pneumonia occurred in 110/689 (16.0%) of cases in women and 132/822 (16.1%) of cases in men. Patient characteristics such as age, underlying disease, and risk factors like duration of neutropenia did not show significant differences. Intensive care unit treatment was needed by 15/97 (15.5%) women and 22/104 (21.2%) men, and the inhospital mortality was 5/98 (5.1%) in women and 12/113 (10.6%) in men, but this result did not reach statistical significance. Seventy-three causative pathogens were identified. Among them, Gram-positive pathogens were identified in three times as often in women (13/36 [36.1%]) than in men (5/37 [13.5%]; p < 0.001, q = 0.002). In contrast, fungi were identified twice as often in men (13/37 [35.1%]) than in women (7/36 [19.4%]; p < 0.001, q = 0.002).

Conclusion: Our exploratory study suggests that while pneumonia rates are similar in women and men, pathogen patterns differ and outcomes may be different. These findings should be verified prospectively and in larger cohorts.