94. Clinical Correlates of Cdc's Hospital-onset Adult Sepsis Event Surveillance Definition and Association with Reportable Healthcare-associated Infections. (31st December 2020)

• Record Type: Journal Article
• Authors: Page, Brady; Klompas, Michael; Chan, Christina; Filbin, Michael; Dutta, Sayon; McEvoy, Dustin; Rhee, Chanu
• Is Part Of: Open forum infectious diseases. Volume 7:Number 1(2020) Supplement
• Page Start: S178
• Page End: S178
• Publication Date: 2020-12-31
• Subjects: Communicable diseases -- Periodicals; Medical microbiology -- Periodicals; Infection -- Periodicals; 616.9
• DOI: 10.1093/ofid/ofaa439.404
• Languages: English
• ISSNs: 2328-8957

Abstract

Abstract: Background: Hospital-onset (HO) sepsis is associated with substantial mortality but is not tracked or reported by most hospitals. CDC's Adult Sepsis Event (ASE) definition may facilitate standardized surveillance but little is known about the clinical correlates of HO-ASEs and their association with currently reportable healthcare-associated infections (HAIs). Methods: In this retrospective study of all adult patients admitted to an academic medical center between June 2015–2018, we assessed the overlap between HO-ASEs and HAIs reported to the National Healthcare Safety Network (NHSN) and reviewed a random subset of 110 HO-ASE cases to determine their clinical correlates. Results: The cohort included 168, 249 hospitalized patients, including 2, 139 (1.3%) with HO-ASE and 2, 133 (1.3%) with NHSN HAIs. Amongst the 2, 139 HO-ASE patients, 480 (22.4%) had ≥1 HAI: 8.1% VAE, 6.2% CLABSI, 6.1% C.difficile, 3.1% CAUTI, 1.3% MRSA bacteremia, and 0.8% SSI. HO-ASE was associated with higher in-hospital mortality rates than HAIs (28.6% vs 14.6%, p< 0.001). HO-ASE associated mortality was high even when NHSN-reportable HAIs were absent (26.5%) whereas NHSN-reportable HAI mortality was relatively low when HO-ASE was absent (8.4%). Amongst the 110 reviewed HO-ASE cases, 102 (93%) were possible or confirmed infections, most commonly pneumonia (39%, of which 35% were ventilator-associated), non-C.difficile intra-abdominal infections (15%), febrile neutropenia (14%), urinary tract infection (7%, of which 88% were catheter-associated), and skin/soft tissue infection (7%). Most (86%) infections flagged by HO-ASEs were acquired in the hospital rather than the community. The most common non-infectious events flagged by HO-ASE were pulmonary edema and periprocedural blood loss associated with blood cultures and empiric antibiotics. Conclusion: CDC's hospital-onset ASE definition accurately identifies patients with nosocomial sepsis who have very high mortality rates and are generally not captured by currently reportable HAI metrics. Routine hospital-onset ASE surveillance could provide a broader window into serious nosocomial infections, identify new targets for prevention, and further improve outcomes for hospitalized patients. Disclosures: All Authors : No reported disclosures