Persistent lymphocytopenia does not increase nosocomial infection risk in the ICU

ABSTRACT Background Lymphocytopenia is frequent in critically ill patients and has been associated with an increased risk of nosocomial infections and death in the ICU. Immunotherapies to promote recovery of lymphocyte counts have therefore been proposed. However, it is unknown if lymphocytopenia is a direct cause of ICU-acquired infections and death, or merely a marker of disease severity. We set out to study the prevalence, temporal evolution, and clinical correlates of lymphocytopenia in ICU patients, and estimate the attributable risk of lymphocytopenia in ICU-acquired infections. Methods We assessed the association between persistent lymphocytopenia (absolute lymphocyte counts <1×10^9/L on day 4) and ICU-acquired infections using multivariable competing risk Cox-regression analyses. Results Among 2302 patients admitted to a Dutch tertiary ICU having sepsis, trauma, or major surgery between 2011 and 2018, persistent lymphocytopenia was observed in 980 (42.6%) subjects. Lymphocyte counts remained relatively stable during early ICU admission, and the median duration of lymphocytopenia was 3 (IQR 1-6) days among exposed patients. ICU-acquired infections occurred in 239 (18.1%) patients without and 214 (21.8%) patients with persistent lymphocytopenia (p=0.03). However, in multivariable survival analysis persistent lymphocytopenia was not associated with infection occurrence, either directly (adjusted cause-specific HR 1.08, 95% CI, 0.90–1.31) or indirectly (subdistribution HR 1.09, 95% CI, 0.91–1.32). Sensitivity analyses did not alter these findings. Conclusion Persistent lymphocytopenia was not associated with a higher incidence rate of nosocomial infections in critically ill patients. This challenges the rationale for using absolute lymphocyte counts as a therapeutic target to prevent ICU-acquired infections.