Clinical significance of lymphocytopenia in patients hospitalized with pneumonia caused by influenza virus

Recently, lymphocytopenia has been evaluated as an independent biomarker of mortality in hospitalized patients diagnosed with community-acquired pneumonia (CAP) [1, 2]. In these studies, patients recruited were affected by CAP due to different etiology, and no specific differences have been observed in viral and bacterial etiology distribution. On the other hand, in influenza virus infection, lymphocytopenia has been identified as a risk factor for bacterial superinfections [3], determining a worse prognosis. We would like to evaluate the potential role of lymphocytopenia as a prognostic factor in patients with pneumonia caused by influenza virus.

For this purpose, we performed a retrospective, observational study on patients hospitalized in a hospital in Rome with pneumonia due to influenza virus. Between January and April 2019, we observed 38 patients with either CAP (29 patients) or hospital-acquired pneumonia (9 patients) due to influenza virus (defined by the presence of fever, symptoms and signs of pneumonia syndrome, new onset of pulmonary infiltrates on chest X-rays or CT scans, and influenza virus detection on respiratory specimens). Consistent with already published data [4], the rate of nosocomial infections found was high and requires substantial improvement of early diagnosis and infection control strategies. Focusing on the lymphocyte count at the onset of infection, with the adoption of a previously reported cutoff value of 724 lymphocytes/μl [2], 23 patients were considered as affected by lymphocytopenic influenza virus pneumonia (L-IP) and 15 by a non-lymphocytopenic influenza virus pneumonia (NL-IP). As shown in Table 1, in comparison with NL-IP, patients with L-IP were more commonly affected by COPD (p = 0.046), they more frequently required admission to the intensive care unit (p = 0.002) and invasive mechanical ventilation (p = 0.031) and presented a higher SOFA score at the time of diagnosis (p = 0.004); they also experienced more frequently secondary bacterial and fungal pulmonary superinfections. As shown in Table 1, secondary bacterial pathogens were often multiple resistant nosocomial organisms as carbapenem-resistant Acinetobacter baumannnii and Corynebacterium striatum and methicillin-resistant Staphylococcus aureus. Notably, fungal pathogens were represented not only by Aspergillus fumigatus but also by Pneumocystis jiroveci. If the former association is well known, to our knowledge, only one case report has been published on P. jiroveci superinfection in an immunocompetent host affected by influenza virus [5]. Analyzing only the influenza CAP patients, the results were similar: SOFA score at the time of diagnosis was higher in patients with L-IP (p = 0.013) and they experienced more frequently respiratory failure requiring oxygen support (p < 0.001) and IMV (p = 0.045) (Table 1). Moreover, all the episodes of superinfection were experienced by lymphocytopenic patients.

Table 1 Baseline characteristics, severity, and microbiology of pulmonary superinfections and outcomes

In our experience, although we evaluated a small sample, a more severe course of a disease might be expected in episodes of L-IP. Under these circumstances, even otherwise immunocompetent patients seem to be at increased risk for opportunistic pulmonary superinfections. Based on the abovementioned, L-IP would require close clinical monitoring for these potentially fatal infectious complications possibly including anti-Aspergillus prophylaxis.