Studies of Tissue Perfusion Failure at LAC+USCMC and the Incorporation of the Results into a National Trauma Database

Abstract

During the first year of this grant, an extensive retrospective database was constructed with emphasis on early hemodynamic studies of patients with severe trauma. The goal is to produce a comprehensive, objective, unbiased database that can be independently mined by members of the combat-casualty-care community. The database is expected to provide insight into the resuscitative outcome of combat casualties and civilian trauma victims. One of the major goals of this project is to identify physiological parameters that are strong predictors of outcome or reflect the need for further therapy. A total of 689 patients are currently in the main database. These patients were monitored with state-of-the-art non-invasive sensors as well as with invasive techniques and procedures. The database is considered very high in quality and utility because the diagnostic measurements are comprehensive, and the prevalence of non-survivors is high. Approximately 180 additional patient histories were processed during the first project year, which represents an increase of -40% in the number of trauma patients studies compared to the previous year. Additional data are required for the development of a large validation group to test predictor-guided models. An initial assessment of outcome predictors indicates that a mathematical "nearest neighbor" analysis is the best technique for predicting survival/non-survival, whereas PtcO(sub 2)/FIO(sub 2) has emerged as the best single parameter predictor. The overall goal is to identify patients who are at risk in the earliest stage of the therapeutic process, adjust therapies to improve outcomes, and promptly determine whether the new therapy will lead to survival. In this regard, analyses of hemodynamic patterns provide an important decision-making aid.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Oct 01, 2002
Accession Number
ADA411609

Entities

People

  • Frank T. Djuth
  • Howard Belzberg
  • William C. Shoemaker

Organizations

  • University of Southern California

Tags

Communities of Interest

  • Biomedical

DTIC Thesaurus Topics

  • Blood
  • Computers
  • Data Mining
  • Databases
  • Health Services
  • Hemorrhagic Shock
  • Information Science
  • Insensitive Explosives
  • Medical Personnel
  • Operating Systems
  • Physiological Monitoring
  • Wound Infections
  • Wounds And Injuries

Fields of Study

  • Medicine

Readers

  • Oncology
  • Systems Analysis and Design
  • Trauma or Military Medicine