Citation

Egea-Guerrero JJ, Rodríguez-Rodríguez A, Gordillo-Escobar E, Fernández-Delgado E, Martínez-Roldán Á, Roldán-Reina Á, Durán-Martínez P, de Vega-Ríos E, Freire-Aragón MD, Vilches-Arenas A, Murillo-Cabezas F, Quintana-Díaz M. J. Head Trauma Rehabil. 2018; 33(1): 46-52.

Affiliation

NeuroCritical Care Unit (Drs Egea-Guerrero, Gordillo-Escobar, Fernández-Delgado, Martínez-Roldán, Roldán-Reina, Durán-Martínez, Freire-Aragón, and Murillo-Cabezas) and Emergency Department (Dr Rodríguez-Rodríguez), Virgen del Rocío University Hospital, IBIS/CSIC/University of Seville, Seville, Spain; Internal Medicine, La Princesa University Hospital, Madrid, Spain (Dr de Vega-Ríos); Department of Preventive Medicine and Public Health, Virgen Macarena University Hospital, University of Seville, Seville, Spain (Dr Vilches-Arenas); and Emergency Department, University Hospital La Paz, Madrid, Spain (Dr Quintana-Díaz).

Copyright

(Copyright © 2018, Lippincott Williams and Wilkins)

DOI

10.1097/HTR.0000000000000292

PMID

28195955

Abstract

BACKGROUND: The aim of this work was to validate the IMPACT (International Mission for Prognosis and Analysis of Clinical Trials in TBI) model in a Spanish cohort of patients with moderate-severe TBI (traumatic brain injury). SETTING: Two level I neurotrauma centers. PARTICIPANTS: Patients admitted to these hospitals between 2011 and 2014 with a diagnosis of TBI and a Glasgow Coma Scale score of 12 or less.

DESIGN: Prospective observational study. MAIN MEASURES: We collected prospectively the clinical variables included in the IMPACT models. Outcome evaluation was prospectively done at 6-month follow-up according to the Glasgow Outcome Scale.

RESULTS: A total of 290 patients were included in the study. Forty-seven patients (16.2%) died within 6 months post-TBI, and 74 patients (25.5%) had an unfavorable outcome. The Hosmer-Lemeshow test revealed that there was no difference between observed and predicted outcomes; hence, the 3 models displayed adequate calibration for predicting 6-month mortality or unfavorable outcome. The receiver operating characteristic curve indicated that the 3 models (Core, Extended, and Lab) could accurately discriminate between favorable and unfavorable outcomes, as well as between survival and mortality (P <.001).

CONCLUSION: The IMPACT model validates prediction of 6-month outcomes in a Spanish population of moderate-severe TBI. IMPACT Lab model is the one that presents a higher discriminative capacity. These results encourage the implementation of the IMPACT model as a prognostic tool in the management of patients with TBI.

Language: en