2013. Blood Culture Contamination in the Emergency Department: A Risk Factor Analysis

Background Blood cultures (BCx) guide treatment for hospitalized patients, yet contaminated BCx lead to clinical uncertainty, impacting care. The Clinical and Laboratory Standards Institute (CLSI) recommends contamination rates should be <3%, yet our Emergency Department (ED) rate is consistently above this benchmark. Reasons for this are unclear, thus it is imperative to investigate potential risk factors for BCx contamination. Methods We performed a retrospective case–control risk factor analysis of patients with BCx collected in our ED between 2014 and 2018. Contaminated BCx were identified by the microbiology laboratory per American Society of Microbiology recommendations. Demographics, comorbidities, and clinical characteristics were evaluated in patients with false-positive/contaminated BCx (cases) and patients with negative BCx (controls). Potential risk factors identified in univariate analysis were included in a logistic regression model. Unadjusted and adjusted analyses were performed using SAS 9.4. Results 25,668 BCx from 13,782 patients were included in analysis. 20,907 BCx from 11,266 (82%) patients were negative, 2,856 BCx from 1,504 (11%) patients were true positives, and 1,905 BCx from 1,012 (7%) patients were contaminated. Yearly ED contamination rates ranged from 5.0–9.3%. Collector contamination rates varied, though 38 (19%), 75 (35%), and 7 (3%) of 209 collectors had a contamination rate <3%, ≥ 10%, and ≥ 20%, respectively. Significant patient-specific risk factors identified in univariate analysis are listed in the attached table along with adjusted analysis. Conclusion In our analysis, we identified that older age, African American race, higher BMI, COPD, paralysis, and presenting in septic shock independently increases risk of having a contaminated BCx. Difficulty obtaining venipuncture in patients with these risk factors, often requiring multiple collection attempts, likely leads to decreased sterile technique. It is imperative to have a process assuring sterile technique in these high-risk individuals to minimize consequences associated with having a false-positive BCx result in these high-risk patients. Additionally, variable collector contamination rates seen in this study highlight the necessity for frequent technique in-service training. Disclosures All authors: No reported disclosures.