Gram-negative central line-associated bloodstream infection incidence peak during the summer: a national seasonality cohort study

Abstract

Central line-associated bloodstream infections (CLABSI) cause increased morbidity, mortality, and hospital costs that are partially preventable. The phenomenon of seasonality among CLABSI rates has not been fully elucidated, but has implications for accurate surveillance and infection prevention trials. Longitudinal dynamic cohort of hospitals participating in hospital-wide and intensive care unit bloodstream infection surveillance for at least one full year over 2000 to 2014. Mixed-effects negative binomial regression analysis calculated the peak-to-low ratio between months as an adjusted CLABSI incidence rate ratio (IRR) with 95% confidence intervals (CI). Multivariate regression models examined the associations between CLABSI pathogens and ambient temperature and relative humidity. The study population included 104 hospital sites comprising 11,239 CLABSI. Regression analysis identified a hospital-wide increase in total CLABSI during July–August, with a higher gram-negative peak-to-low incidence rate ratio (IRR 2.52 [95% CI 1.92–3.30], p < 0.001) compared to gram-positive bacteria (IRR 1.29 [95% CI 1.11–1.48], p < 0.001). Subgroup analysis replicated this trend for CLABSI diagnosed in the intensive care unit. Only gram-negative CLABSI rates were associated with increased temperature (IRR + 30.3% per 5 °C increase [95% CI 17.3–43.6], p < 0.001) and humidity (IRR + 22.9% per 10% increase [95% CI 7.7–38.3), p < 0.001). The incidence and proportion of gram-negative CLABSI approximately doubled during the summer periods. Ambient temperature and humidity were associated with increases of hospital-acquired gram-negative infections. CLABSI surveillance, preventive intervention trials and epidemiological studies should consider seasonal variation and climatological factors when preparing study designs or interpreting their results.