Abstract
Background
Multi-drug-resistant organisms (MDRO) pose a growing burden, including in non-hospital settings. Delay in initiation of appropriate antimicrobial therapy (DAAT) upon admission to an acute care hospital is common and is associated with worse outcomes. The aim of this study was to develop a prediction score for MDRO infection upon admission, in order to improve patients’ outcomes and avoid misuse of broad-spectrum antimicrobials.
Methods
A retrospective case–control analysis was conducted at Assaf Harofeh Medical Center, Israel, comparing adult patients with MDRO infections diagnosed in the first 48 hours of hospitalization to patients presenting with non-MDRO sepsis (i.e., patients with microbiologically confirmed non-MDRO infection, or patients with non-microbiologically confirmed sepsis). MDROs were determined by clinical laboratory testing. Patients were identified over four consecutive months (August–December 2016). A multivariable logistic regression of predictors for MDRO infection upon admission was used to develop the prediction score.
Results
Ninety-five of 818 total patients (11.6%) had MDRO infection. The final score included 10 parameters: (1) home therapy (IV therapy, wound care, or specialized nursing care, 16 points), (2) routine (at least weekly) outpatient clinic visits in the past 3 months (15 points), (3) history (2 years) of past MDRO colonization (14 points), (4) any antibiotics in the preceding 3 months (12 points), (5) invasive procedure in the past 6 months (11 points), (6) elderly (≥65 years old, 10 points), (7) hemiplegia or paraplegia (8 points), (8) resident of long-term care facility (7 points), (9) severe sepsis (i.e., severe sepsis, septic shock, or multi-organ failure, 6 points), and (10) acute kidney injury (5 points). A cutoff of ≥24 points had a sensitivity of 90%, a specificity of 73% and an ROC AUC = 0.88 (figure).
Conclusion
This study presents the development of a new prediction score for MDRO infection upon admission, based on parameters that could easily be extracted at bedside for patients admitted with sepsis. A future prospective interventional study is needed in order to validate the score, captured at bedside electronically, in terms of improving patients’ outcomes.
Disclosures
All authors: No reported disclosures.
Session: 136. Healthcare Epidemiology: MDR-Gram Negative Infections
Friday, October 5, 2018: 12:30 PM