PURPOSE

Liver is the most frequently injured organ in abdominal trauma. Today non-operative management (NOM) is considered as the standard of care in hemodynamically stable patients, with or without the adjunct of angioembolisation (AE). This systematic review assesses the incidence of complications in patients who sustained liver injuries and were treated with simple clinical observation. Given the differences in indications of treatment and severity of liver trauma and acknowledging the limitations of this study, an analysis of the results has been done in reference to the complications in patients who were treated with AE.

METHODS

A systematic literature review searched "liver trauma", "hepatic trauma", "conservative management", "non operative management" on MEDLINE (via PubMed), Cochrane Central Register of Controlled Trials databases, EMBASE, and Google Scholar, to identify studies published on the conservative management of traumatic liver injuries between January 1990 and June 2020. Patients with traumatic liver injuries (blunt and penetrating) treated by NOM, described at least one outcome of interests and provided morbidity outcomes from NOM were included in this study. Studies reported the outcome of NOM without separating liver from other solid organs; studies reported NOM complications together with those post-intervention; case reports; studies including less than 5 cases; studies not written in English; and studies including patients who had NOM with AE as primary management were excluded. Efficacy of NOM and overall morbidity and mortality were assessed, the specific causes of morbidity were investigated, and the American Association for the Surgery of Trauma classification was used in all the studies analysed. Statistical significance has been calculated using the Chi-square test.

RESULTS

A total of 19 studies qualified for inclusion criteria were in this review. The NOM success rate ranged from 85% to 99%. The most commonly reported complications were hepatic collection (3.1%), followed by bile leak (1.5%), with variability between the studies. Other complications included hepatic haematoma, bleeding, fistula, pseudoaneurysm, compartment syndrome, peritonitis, and gallbladder ischemia, all with an incidence below 1%.

CONCLUSION

NOM with simple clinical observation showed an overall low incidence of complications, but higher for bile leak and collections. In patients with grade III and above injuries, the incidence of bile leak, collections and compartment syndrome did not show a statistically significant difference with the AE group. However, the latter result is limited by the small number of studies available and it requires further investigations.