Abstract Background Serratia marcescens is an opportunistic gram-negative pathogen increasingly associated with healthcare-associated infections. Data on outcomes and factors influencing mortality in S. marcescens bacteremia are limited. Methods We conducted a retrospective cohort study of all inpatients with an initial positive blood culture for S. marcescens at UK HealthCare from July 1, 2022, to December 31, 2024. Among 970 patients with bacteremia, 61 (6.3%) had S. marcescens. Clinical characteristics, management strategies, and outcomes were analyzed. Data were obtained through manual chart review and the Center for Clinical and Translational Science (CCTS) database. Predictors of in-hospital mortality were evaluated using univariate analysis. Results Among 61 patients with S. marcescens bacteremia, the median (IQR) age was 58 years (44, 68), and 58.1% were male. Over half of infections (53.2%) were hospital-acquired. Common comorbidities included COPD (45.2%), diabetes with complications (35.5%), congestive heart failure (CHF) (43.5%), and peripheral vascular disease (41.9%). ICU admission occurred in 43.5% of patients. The median hospital length of stay was 17 days (10, 35), with ICU length of stay of 8.4 days (3.8, 15.3). In-hospital mortality was 27.4%, and 90-day readmission occurred in 38.7% of patients. An Infectious Diseases (ID) consult was obtained in 54.8% of cases. ID involvement was significantly more frequent in non-ICU patients (77.1%, 27/35) compared to ICU patients (25.9%, 7/27; p 0.001). Mortality was substantially lower in patients who received an ID consult: 14.7% (5/34) vs 57.1% (16/28) without consult (p = 0.013). Other significant predictors of mortality included ICU admission (p = 0.039), qPitt score (p 0.001), Charlson Comorbidity Index (p = 0.042), and SOFA score (p 0.001). Conclusion Serratia marcescens accounted for a small proportion of bacteremia cases but was associated with high mortality. ID consultation was associated with a four-fold reduction in mortality yet was markedly underutilized in critically ill patients. These findings highlight a significant opportunity to improve outcomes through earlier and more consistent ID involvement, especially in the ICU setting. Disclosures All Authors: No reported disclosures
Burgess et al. (Thu,) studied this question.