RESUMO
INTRODUCTION: Experience accumulated over the last decades suggests nonoperative management (NOM) of civilian gunshot liver injuries can be safely applied in selected cases. This study aims to compare the outcomes of selective NOM versus operative management (OM) of patients sustaining gunshot wounds (GSW) to the liver. METHODS: A registry-based retrospective cohort analysis was performed for the period of 2008 to 2016 in a Brazilian trauma referral. Patients aged 16-80 years sustaining civilian GSW to right-sided abdominal quadrants and liver injury were included. Baseline data, vital signs, grade of liver injury, associated injuries, injury severity scores, blood transfusion requirements, liver- and non-liver-related complications, length-of-stay (LOS), and mortality were retrieved from individual registries. RESULTS: A total of 54 patients were eligible for analysis, of which 37 underwent NOM and 17 underwent OM. The median age was 25 years and all were male. No statistically significant differences were observed between groups regarding patients' demographics, injury scores, grade of liver injury and associated lesions. NOM patients tended to sustain higher-grade injuries (86.5% vs 64.7%; p = 0.08), and failure of conservative management was recorded in two (5.4%) cases. The rate of complications was 48% with no between-group statistically significant difference. Blood transfusion requirements were significantly higher in the OM group (58.8% vs 21.6%; p = 0.012). The median LOS was seven days. No deaths were recorded. CONCLUSION: Patients with liver GSW who are haemodynamically stable and without peritonitis are candidates for NOM. In this study, NOM was safe and effective even in high-grade injuries.
RESUMO
A multicentre retrospective analysis of blunt gastric injuries from four trauma centres in Brazil was performed. From January 1982 to May 1996, 33 patients were found to have blunt gastric injury: 26 were male and the mean age was 26.2 years. The most frequent mechanism of injury was automobile versus pedestrian (17) followed by motor vehicle accidents (seven). Abdominal pain was the most common finding (29) and peritoneal signs were present in 14. There were 21 grade I, seven grade II, four grade III and one grade IV blunt gastric injuries. Simple suture was the treatment of choice in 24 patients, eight required no treatment, and only one patient underwent gastric resection. The liver and spleen were the most commonly associated injured organs. Two patients had isolated gastric injury. There were two gastric fistulae and both patients died. Overall morbidity and mortality was 12 (36%) and nine (27%), respectively. The ISS was higher in the non-survivor group (P=0.03) and the gastric organ injury score did not reach statistical significance when comparing survivors and non-survivors. We conclude that blunt gastric injury is uncommon and is associated with other injuries of greater magnitude, which generally influence mortality.