Abstract | March 2, 2021

Utilization of a Burn Sepsis Algorithm: A 5 Year Analysis

Presenting Author: Jeffrey W. Williams, PA-C, Wake Forest Baptist Medical Center, Burn Center, Winston-Salem, NC

Co-authors: Christopher Craig, PA-C, Wake Forest Baptist Health, Winston Salem, NC, Anju Saraswat, MD, Wake Forest Baptist Health, Winston Salem, NC, J. Kevin Bailey, MD, Wake Forest Baptist Health, Winston Salem, NC, & James Holmes, MD, Wake Forest Baptist Health, Winston Salem, NC.

Learning Objectives

  1. Examine the data from 5 years of utilizing a Burn Code Sepsis Algorithm;
  2. Discuss the ongoing difficulty early identification of sepsis in burn injured patients.

Introduction: After the first 24 hours, the major cause of death in burn patients is multiple organ dysfunction/failure syndrome. It is preceded by infection in 83% of burn patients, with reported septic mortality up to 65%. Since the early recognition and treatment of infection has been shown to decrease mortality from sepsis, we implemented a multidisciplinary algorithm designed to rapidly identify septic adult burn patients.

Methods: Adult (≥18 y) admissions between 7/1/2014 – 6/30/2019 were identified from our registry, and all initial sepsis screens were evaluated in the EMR. Patients were screened clinically at least BID and were considered a “positive” screen if MAP 102.2F; HR >120; RR >28 (or RR >10% of ventilator set rate if set rate is >24 bpm). A positive screen prompted lab work to include CBC, BMP, procalcitonin (PCT), and lactic acid (LA), per protocol. If PCT>3.0 ng/ml or LA>2.0 mmol/L, or both thrombocytopenia and hyperglycemia were present, a “Burn Code Sepsis” was initiated and included cultures, a CXR, and empiric antibiotics. A patient was then formally considered “septic” (i.e. – infected), if the cultures were positive or the CXR demonstrated an infectious process.

Results: There were 1,523 admissions during the 5-year period, and 228 initial positive screens. Of the 228 patients with positive screens, 159 (70%) were infected. There was a significant difference in PCT level between patients with and without infections, while no difference was noted for LA, WBC, platelets, temperature, glucose, age, %TBSA burned, or time to triggering a positive screen between the groups (Table). Defining PCT ≥3.0 ng/mL as being positive for sepsis demonstrated a 76% PPV and a 36% NPV for PCT alone.

Conclusions: Recognition of sepsis continues to be difficult in burn patients. PCT may have a role in the early detection of sepsis. Further research is warranted.

Image 1

References and Resources

  1. ABA Consensus Conference to Define Sepsis and Infection in Burns https://verbrennungsregister.net/docs/Definition%20Sepsis%20in%20Burns.pdf
Download Abstract
Posted in: Burn Medicine101
Back to Abstracts