Revista Colombiana de Cirugía (Dec 2009)
Evaluación de las escalas ISS y NISS en trauma penetrante grave Evaluation of the ISS and NISS injury severity scores in patients with severe penetrating trauma
Abstract
Introducción. Existe un amplio uso de la escala Injury Severity Score (ISS) como factor predictor de mortalidad en trauma. Algunos reportes han mostrado que es pobre factor predictor en pacientes con lesiones penetrantes graves. Recientemente, se ha usado el New Injury Severity Score (NISS) y se ha comparado con el ISS en trauma cerrado grave y trauma penetrante. Objetivo. Evaluar el comportamiento del NISS y del ISS en pacientes con trauma penetrante grave llevados a laparotomía de control de daños y en laparotomía convencional no de control de daños. Materiales y métodos. Se incluyeron pacientes adultos del Registro DAMACON. El NISS fue calculado retrospectivamente. Mediante un modelo de regresión logística se construyó la curva ROC (receiver operating characteristics) y el área bajo la curva (AUC) para comparar la capacidad discriminatoria de ambas escalas. Resultados. Se incluyeron 214 pacientes, 93 (43,4%) necesitaron laparotomía de control de daños. Entre los grupos, no hubo diferencias en la edad (32,3±10 años Vs. 30,3±10, p=0,12); ni en el sexo masculino (92,5% Vs. 90,1%, p=0,45). La gravedad del trauma por Revised Trauma Score (RTS) fue mayor en el grupo de laparotomía de control de daños comparado con el grupo de laparotomía convencional no de control de daños (6,21,9 Vs. 7,1±1,2, p=0,0022) al igual que la mortalidad a 30 días (35,5% Vs. 3,3%, pIntroduction. Despite the wide use of ISS as a mortality predictive score in trauma, several reports have shown that ISS is a poor predictor of outcome especially in patients with severe penetrating injuries. More recently, the use of New ISS (NISS) has been compared with ISS in blunt severe trauma and mild penetrating trauma. The purpose of this study was to evaluate the performance of NISS and ISS in patients with severe penetrating trauma, both in damage control (DC) and not damage control (no-DC) surgery. Methods. Consecutive adult patients with penetrating trauma and surgery were identified over a 6-years period (2003-2008) in our prospective trauma single center registry (DAMACON). NISS was estimated retrospectively. Logistic regression was used to construct ROC curves and areas under the curve (AUC) in order to compare the discriminative capacity among scores by type of surgical approach. Results. A total of 214 patients with penetrating trauma, 93 (%) needed DC surgery. Mean age for DC group was of 32.3±10 years old, RTS was 6.2±1.9, and 30 days mortality was of 35.5%. Compared with non-DC patients with mean age of 30.3±10.9 years old, RTS was 7.1±1.2 and 30 days mortality of 3.3%. NISS and ISS AUC were 0.86 and 0.72 respectively (P=0.01), for non-DC group; and 0.71 and 0.76 in DC group. Conclusion. NISS showed an improved performance for mortality prediction over ISS in the subgroup of patients with non-DC surgery for penetrating trauma, but not in DC setting. Furthermore, overall performance of the scores was sub-optimal in this setting. More efforts have to be made to identify new prognostic factors for this subgroup of very ill patients.
