1043. Antibiotic (ABX) De-escalation in Sepsis: A Retrospective Study in a Tertiary Care Medical Center with an Established Antimicrobial Stewardship Program (ASP)
Session: Poster Abstract Session: Stewardship: Improving Treatments
Friday, October 4, 2013
Room: The Moscone Center: Poster Hall C
  • Betito IDSA Poster FINAL-1.pdf (702.2 kB)
  • Background:

    De-escalation of prompt, broad spectrum empiric ABX therapy is recommended in the therapy of sepsis. The frequency of this de-escalation is not well described. The aim of this study is to determine the frequency and characteristics of ABX de-escalation in patients with sepsis at an 885-bed tertiary care medical center with an established ASP.


    Adult inpatients with sepsis or septic shock during 2011 were randomly selected. Records (n=180) were retrospectively reviewed for demographics; ABX regimen, duration and adequacy; reason for de-escalating or continuing ABX; culture results; in-house mortality; length-of-stay (LOS) and 30-day re-admission. Adequacy was defined as empiric ABX susceptibility of all cultured pathogens, and de-escalation as ABX discontinuation or narrow spectrum substitution. The proportion of patients de- escalated by 72 and 96 hrs was calculated. An adjusted de-escalation proportion was determined by subtracting from the total population patients who died prior to 96 hrs, or could not be de-escalated due to culture susceptibility results. 


    ABX were de-escalated in 43 (24%) and 87 (48%) patients by 72 and 96 hrs, respectively (Table). The adjusted 96 hour ABX de-escalation proportion was 66%. Cultures were obtained in 178 (99%) patients. The proportion of patients de-escalated was not significantly different for those with positive or negative cultures. The ABX de-escalation group had a shorter LOS and less in-hospital mortality. Reasons for not de-escalating were persistence of SIRS/fever (30), microbial susceptibility only to empiric therapy (18), and negative culture results (13).


    Selected Characteristics of De-escalation


    De-escalation (n=87)

     No. (%)

    No De-escalation (n=93)

    No. (%)


    Culture positive

    60 (69)

    59 (63)


    Adequate empiric ABX

    56 (93)

    42 (71)



    6 (7)

    44 (47)


    Mortality after 96 hrs

    6 (7)

    23 (32)


    Mean LOS




    30-day re-admission

    4 (5)

    6 (6)



    A modest number of patients in whom ABX were empirically started for sepsis were de-escalated by 96 hours (half between 72 and 96 hrs). Culture positivity did not appear to affect de-escalation decisions. Further data is needed to determine optimal de-escalation metrics for patients with sepsis.

    Shara Ann Betito, MD1, Christopher Ohl, MD2, John Williamson, PharmD3, James Johnson, PharmD3, James Beardsley, PharmD3 and Vera Luther, MD2, (1)Infectious Disease, Wake Forest Baptist Health, Winston Salem, NC, (2)IM-Section On Infectious Diseases, Wake Forest School of Medicine, Winston-Salem, NC, (3)Wake Forest Baptist Health, Winston-Salem, NC


    S. A. Betito, None

    C. Ohl, None

    J. Williamson, None

    J. Johnson, None

    J. Beardsley, None

    V. Luther, None

    Findings in the abstracts are embargoed until 12:01 a.m. PST, Oct. 2nd with the exception of research findings presented at the IDWeek press conferences.