The association between Cytomegalovirus co-infection with Pneumocystis pneumonia and mortality in immunocompromised non-HIV patients.

Introduction: Impact of Cytomegalovirus (CMV) co-infection pneumonia in non-HIV patients with Pneumocystis jirovecii pneumonia (PCP) is unclear.

Objectives: The aim of our study was to determine whether CMV co-infection is associated with an increased risk of mortality.

Methods: Our study was conducted at Ege University Hospital, Turkey. We used molecular assays to diagnose Pneumocystis jirovecii in respiratory samples, and CMV in both respiratory and blood samples. We compared morbidity and mortality stratified by CMV co-infection status.

Results: Between 2009 and 2015, 43 patients (mean age: 56.7 ± 15.3 years) were diagnosed with PCP. Only 3 of 43 patients had received PCP prophylaxis. We microbiologically confirmed CMV co-infection in 28 of 43 (65.1%) patients. Acute respiratory distress syndrome (ARDS) and requirement of mechanical ventilation were more common in the CMV co-infection group (P = .019 and P = .031 respectively), and duration of intensive care unit was also longer (P = .006). In univariate analyses, mortality at 30 days was higher in the CMV co-infection group as compared to the group with PCP alone (78.6% and 46.7% respectively; P = .046). In multivariate analyses, mortality was independently associated only with the presence of ARDS [OR: 6.22 95% CI 1.3-29.32] and the association with CMV co-infection was no longer significant [OR: 2.6 95% CI 0.49-13.72, P = .257].

Conclusion: The risk of mortality appears to be increased in the setting of CMV and PCP co-infection in HIV-uninfected immunocompromised patients. PCP prophylaxis use was lower than expected, suggesting low physician awareness of the risks of PCP in this population.