Outcome of nonoperative management of blunt liver trauma.

OBJECTIVE To evaluate our experience with the management of blunt liver trauma at Riyadh Central Hospital, Kingdom of Saudi Arabia. METHODS The hospital records of 68 patients treated for blunt liver trauma at Riyadh Medical Complex over a 5-year period (1997 through to 2002) were reviewed retrospectively. Patients who were hemodynamically unstable or had peritonitis were treated by urgent laparotomy (operative group). The other group of patients were treated conservatively as they were hemodynamically stable (nonoperative group) in the intensive care unit (ICU). All patients had computed tomogram of the abdomen with oral and intravenous contrast. Injuries grades were classified according to American Association for the Surgery of Trauma (AAST). Follow up computed tomogram of the abdomen was performed in 1-2 weeks time. RESULTS A total of 68 patients were treated over a 5-year period. Thirty-three patients (49%) were treated by immediate surgery. Hepatorraphy was performed in 22 patients, non anatomical resection in 3 patients and liver packing in 2 patients to control bleeding. No active bleeding was found in 8 patients. Nonoperative group (n=35, 51%) were managed in the ICU for close monitoring. Surgically treated group had more patients with complex liver injury (30% versus 11%), required more units of blood (3.2 versus 2.1) but had a shorter hospital stay (9 days versus 12 days). Two patients for nonoperated group develop complications (biloma: one patient, infected hematoma: one patient), the nonoperative treatment failed in one patient who required laparotomy due to rebleeding from grade IV liver injury with failure rate of 3%. The only mortality (one patient) was in the operated group. CONCLUSION The nonoperative treatment is a safe and effective method in the management of hemodynamically stable patients with blunt liver trauma. The nonoperative treatment should be the treatment of choice in such patients whenever computed tomogram and ICU facilities are available. However, further studies with larger number of patients are needed to support our results.