Non-Operative Management of High Grade Blunt Hepatic Injury: Clinical Complications and the Role of Collateral Damage (NOMLI)

Background:

Modern approach to liver injured patients favours non-operative management of liver injury (NOMLI) including endovascular artery occlusion. Numerous studies over the past two decades have confirmed the feasibility of NOMLI in up to 95% of hemodynamically stable patients with blunt trauma mechanisms. To further improve the outcome of patients with hepatic injuries, investigations should focus on the overall morbidity and mortality of NOMLI. Purely hepatic-related complication rates in most series are low, ranging from 0-7% and parallel the grade of liver lesion. But the majority of patients included in those studies suffered low-grade liver injuries (LI). Regarding the safety of NOMLI in high grade lesions, the results may be biased. Complications in patients with high-grade LI are more frequent and their management is considerably more complex. Regarding the sparsely available literature focussing on the morbidity of NOMLI in high grade LI, hepatic-related complication rates of 11 and 13% have been described. Only the LI grade and the amount of packed red blood cell (PRBC) transfusion at 24 hours postinjury predicted hepatic-related complications. No data is available concerning the impact of collateral intra- and extra-abdominal damage on complications of NOMLI in those patients. Of note, up to 75% of patients with LI suffer from collateral intra- and extra-abdominal lesions. These injuries vary in their surgical importance but severe complications and to the LI independent laparotomies must be expected in a significant number of patients. Rates of such nonspecific laparotomies have been described in up to 20% of patients with hepatic trauma. Over the past decade, overall mortality of liver injured patients remained relatively constant, averaging between 10-15%. This mortality rate represents deaths of all causes in the clinical course of these complex trauma patients. There are studies which already suggested, that early deaths in patients with LI also were caused by uncontrolled bleedings from associated intra- and extra-abdominal injuries, and that most late deaths result from collateral head injuries and sepsis with multi-organ-failure (MOF). But no detailed data about the occurrence of extra-abdominal complications after NOMLI has been published so far.

Objective:

We first hypothesized that NOMLI can be safely achieved also in high-grade liver injured patients, the management of trauma patients with LI mainly consist of the treatment of collateral damages and their complications, and causes of death are in the majority of cases not liver related. A retrospective study involving 183 patients with blunt hepatic injuries was therefore carried out to investigate these hypotheses.

Methods:

The study was conducted at the Bern University Hospital, Switzerland between January 2000 and December 2006. An average of 286 (range, 204-344) multiple injured patients were treated in our level I trauma centre each year. Only blunt liver injured patients were included. All charts (including surgery and autopsy reports) were reviewed retrospectively. Demographic data collected included age, gender, mechanism of injury. Injury patterns were defined by the Abbreviated Injury Score (AIS) and Injury Severity Score (ISS). The grade of hepatic injury was determined by an experienced radiologist and by two experienced hepatobiliary surgeons in parallel based on contrast enhanced computed tomography (CT) scan findings (Siemens® Somatom Sensation 16) or by laparotomy according to the American Association for the Surgery of Trauma Organ Injury Scale for hepatic injuries. For the current study, high-grade injuries were considered grades 3 to 5. Other data collected included the number and types of surgical procedures, hepatic-related and overall complications, and causes of deaths. All patients were managed and resuscitated using the protocols outlined in the Advanced Trauma Life Support (ATLS) manual of the American College of Surgeons Committee on Trauma. NOMLI was