Skip to main navigation Skip to search Skip to main content

Predicting the mortality of necrotizing fasciitis with blood pressure and white blood cell count

  • Wen Chih Fann
  • , I. Jen Chiang
  • , Cheng Ting Hsiao
  • , Yu Cheng Hong
  • , I. Chuan Chen

Research output: Contribution to journalArticlepeer-review

Abstract

Aim: Even with modern care, necrotizing fasciitis (NF) is a life-threatening emergency disease. The present study attempts to identify simple clinical admission characteristics or laboratory tests to differentiate between high- and low-mortality risk groups in patients with NF. Patients and Methods: The records of 272 patients who received debridement were reviewed. The X2-test (or the Mann-Whitney U-test) and C4.5 decision tree were utilized to analyse 23 clinical characteristics and laboratory tests. The main outcome measure was in-hospital mortality. Results: The overall mortality rate was 17 per cent. Three independent predictors of mortality - white blood cell (WBC) count; left shift in the differential WBC count, with an increase in immature neutrophils (band form); and hypotension - were determined using the C4.5 decision tree. From these predictors, a decision tree was produced to classify patients with NF into high- and low-mortality risk groups. The accuracy of the C4.5 decision tree with cross-validation was 84.2 per cent (95 per cent confidence interval: 80.3-88.1 per cent). Conclusion: Using routine blood pressure measurements and simple laboratory tests of WBC count and differentials, clinicians can rapidly identify patients with high-mortality risk.

Original languageEnglish
Pages (from-to)103-108
Number of pages6
JournalSurgical Practice
Volume16
Issue number3
DOIs
Publication statusPublished - Aug 2012

Keywords

  • Decision tree
  • Mortality
  • Necrotizing fasciitis
  • Risk factor

ASJC Scopus subject areas

  • Surgery

Fingerprint

Dive into the research topics of 'Predicting the mortality of necrotizing fasciitis with blood pressure and white blood cell count'. Together they form a unique fingerprint.

Cite this