Time to Development of SIRS Impacts Outcomes of Patients Recognized with Sepsis on the Wards

SESSION TITLE: Sepsis & Septic Shock SESSION TYPE: Original Investigation Slide PRESENTED ON: Sunday, October 29, 2017 at 01:30 PM - 03:00 PM PURPOSE: Most septic patients admitted to a hospital are screened in the emergency department. These patients receive treatment with antibiotics and fluids based on the recognition of infection and laboratory values. However, if they are identified on the wards, outcomes based on three-hour-bundle management are not well studied. METHODS: We performed a retrospective study of adult patients admitted to the inpatient wards at a tertiary care academic center in central Texas from June 2014 until December 2016 who met sepsis screen. We implemented a sepsis alert system with twice a day nursing screening for SIRS criteria and activation of a sepsis team led by a PA or physician. Multivariable logistic and negative binomial regression assessed the effect of covariates on mortality, organ failure (MV, RRT, NIV, vasopressors), and hospital length of stay (LOS). qSOFA (score >1 because <5% met qSOFA >2) and SIRS scores were compared in predicting the outcomes controlling for other variables in the regression models. RESULTS: Of 1,016 patients screened, 576 were included,. The mean age was 56 (SD, 19), 54% were female and mean Charlson comorbidity score of 2.9 (SD 2.5). 28 day mortality was 6% and hospital LOS was 8.9 days (SD 8.9). The time to recognition of positive SIRS criteria was 64.9 hours (SD 125.0) and notification of sepsis provider of 1.92 hours (SD 3.6). Most common source of infection were respiratory (43%) and GU (19.6%). 20% developed organ dysfunction (2012 surviving sepsis campaign). A sepsis team intervened 58% of the time and primary team 42%. 96% of patients had a SIRS ≥2, while only 5% of patients had a qSOFA ≥2. Three-hour-bundle compliance was 12.5% and antibiotic administration within 3 hours occurred in 51.4%. 8.5% of patients were transferred to the ICU and 429 (74.5%) patients were discharged home. The risk of combined mortality and organ failure increased by 57% if time to SIRS development doubled since admission to the wards (P<0.001).Patients evaluated with the qSOFA of equal and greater were more likely to experience the combined outcome of organ dysfunction/mortality (OR 2.3, CI=1.2-4.3 P=0.008), while SIRS score was not a significant predictor of the outcome. The length of stay and home discharge were also negatively impacted by time to SIRS (RR:1.19 CI=1.05-1.35, P <0.0001; OR 0.63, CI= 0.54-0.74, P< 0.0001). Mortality/organ failure was reduced in patients receiving antibiotics in less than 3 hours (OR 0.47, CI 0.23-0.95, P=0.03) and increased in those who received more than 30 ml/kg IVF (OR 2.8, CI 1.7-6.9, P=0.008). CONCLUSIONS: Time to development of sepsis based on SIRS after admission to the wards inversely impacted mortality, organ failure, length of stay and home discharge. Of the three-hour-bundle, the early administration of antibiotics improved outcomes, however receiving 30 ml/kg or 1 liter of fluid had a negative effect on this population. CLINICAL IMPLICATIONS: This study details the use of the three-hour-bundle in a unique population and reveals that its use in patients who become septic after admission does not improve mortality. DISCLOSURE: The following authors have nothing to disclose: Kirill Lipatov, Chhaya Patel, Justin Price, Thomas Delmas, Angela Birdwell, Tasnim Lat, Matthew Crowe, Andrew Widmer, Angela Hochhalter, Angela Eklund, Heath White, Alfredo Vazquez-Sandoval, Lori Murdoch, Ludmyla Ploskanych, Kiumars Zolfaghari, Shekhar Ghamande, Alejandro Arroliga No Product/Research Disclosure Information