Pancreatic Stone Protein Predicts Sepsis in Severely Burned Patients Irrespective of Trauma Severity: A Monocentric Observational Study

Ann Surg. 2021 Dec 1;274(6):e1179-e1186. doi: 10.1097/SLA.0000000000003784.


Objective: The burn victim's inherent state of hyperinflammation frequently camouflages septic events delaying the initiation of targeted intensive care therapy. Accurate biomarkers are urgently needed to support sepsis detection before patients' clinical deterioration.

Summary of background data: Evidence on the usefulness of pancreatic stone protein (PSP) as a powerful diagnostic and prognostic marker in critically ill patients has recently accumulated.

Methods: Analysis of biomarker kinetics (PSP, routine markers) was performed on 90 patients admitted to the Zurich Burn Center between May 2015 and October 2018 with burns ≥15% total body surface area with regard to infection and sepsis (Sepsis-3) over a 14-day time course.

Results: PSP differentiated between sepsis, infection and sterile inflammation from day 3 onward with an area under the curve of up to 0.89 (P < 0.001), therefore, competing with procalcitonin (area under the curve = 0.86, P < 0.001). Compared to routine inflammatory biomarkers, only PSP demonstrated a significant interaction between time and presence of sepsis - signifying a steeper increase in PSP levels in septic patients as opposed to those exhibiting a nonseptic course (interaction P < 0.001). Event-related analysis demonstrated tripled PSP serum levels within 72 hours and doubled levels within 48 hours before a clinically apparent sepsis.

Conclusion: PSP is able to differentiate between septic and nonseptic patients during acute burn care. Its steep rise up to 72 hours before clinically overt deterioration has the potential for physicians to timely initiate treatment with reduced mortality and costs.

Publication types

  • Observational Study
  • Research Support, Non-U.S. Gov't

MeSH terms

  • Adult
  • Biomarkers / blood*
  • Burns / complications*
  • Critical Illness
  • Female
  • Humans
  • Lithostathine / blood*
  • Male
  • Middle Aged
  • Predictive Value of Tests
  • Prognosis
  • Sepsis / blood*


  • Biomarkers
  • Lithostathine