SESSION TITLE: Sepsis and Septic Shock
SESSION TYPE: Original Investigation Slide
PRESENTED ON: Monday, October 28, 2013 at 07:30 AM - 09:00 AM
PURPOSE: Numerous studies suggested that outcomes of patients presenting with infection (“PI”) are improved when antibiotics are administered as early as possible relative to the time of patient registration at emergency department (Tr). No prior studies have examined the relationships of outcomes to time of onset of initial symptoms to first antibiotics (Ts). We hypothesized that "Ts" would better predict outcomes than "Tr".
METHODS: All “PI” admitted to the hospitalist service were interviewed. “PI” were excluded if they had no clear source of infection or caregivers stopped antibiotics within the first 48 hours. The timing from emergency room registration, onset of symptoms, to first administration of antibiotics were recorded as (Tr and Ts respectively). Outcomes included length of stay, ICU transfer, and mortality.
RESULTS: 115 “PI” patients were included, mean age 64.6±17.2 years with 59% being male. The median time from onset of infection site-specific symptoms (±6 hours) to first antibiotic dose was 28±28 hours. "Ts" was most associated with length of stay, using linear regression model (R2=0.32, p<0.0001), including age (p=0.15), number of co-morbidities (p=0.63), "Tr" (p=0.86). There were too few ICU transfers (n=3) and mortalities (n=2) to permit meaningful analyses of outcomes. When patients were stratified into three risk groups - low (no co-morbidities and Ts<24 hours), medium (≧ 1 co-morbidity or Ts >24 hours), and high (≧1 co-morbidity and Ts>24 hours) - mean length of stay tracked groups. Low risk patients stayed 2.8±1.5 days, medium risk 4±2.8days, high risk 5.1±3.3days (ANOVA p=0.03).
CONCLUSIONS: These results suggest that the duration of site-specific infectious symptoms prior to first antibiotics is more highly correlated with length of stay than other variables studied.
CLINICAL IMPLICATIONS: While “Tr” is used for administration purposes and national standards, “Tr” ignores the impact of co-morbidities and trajectory of illness prior to admission, whereas “Ts” may be more pathophysiologically germane to outcomes.
DISCLOSURE: The following authors have nothing to disclose: Andrey Pavlov, Rostislav Muraviev, Ulysses Wu, Constantine Manthous
No Product/Research Disclosure Information