Early diagnosis of hollow viscus injury using intestinal fatty acid-binding protein in blunt trauma patients.

ABSTRACT

A delay in diagnosing hollow viscus injury (HVI) causes an increase in mortality and morbidity. HVI remains a challenge to diagnose, and there is no specific diagnostic biomarker for HVI. We evaluated the utility of intestinal fatty acid-binding protein (I-FABP) in diagnosing HVI in blunt trauma patients. Within a 5-year period, 93 consecutive patients with clinically suspected HVI at our trauma center were prospectively enrolled. The diagnostic performance of I-FABP for HVI was compared with that of other various parameters (physical, laboratory, and radiographic findings). HVI was diagnosed in 13 patients (14%), and non-HVI was diagnosed in 80 patients (86%). The level of I-FABP was significantly higher in patients with HVI than in those with non-HVI (P = 0.014; area under the curve, 0.71). The sensitivity, specificity, positive predictive value, and negative predictive value were 76.9%, 70.0%, 29.4%, and 94.9%, respectively (P = 0.003). However, all other biomarkers were not significantly different between the groups. Presence of extraluminal air, bowel wall thickening on computed tomography (CT), and peritonitis signs were significantly higher in patients with HVI (P < 0.05). Of 49 patients (52.7%) who had a negative I-FABP and negative peritonitis signs, none developed HVI (sensitivity, 100%; negative predictive value, 100%). This is the first study that demonstrated the diagnostic value of a biomarker for HVI. I-FABP has a higher negative predictive value compared to traditional diagnostic tests. Although the accuracy of I-FABP alone was insufficient, the combination of I-FABP and other findings can enhance diagnostic ability.