Severe liver injuries in a contemporary national trauma registry: In-hospital mortality remains high regardless of therapeutic approach

Background

Severe liver injuries remain a common cause of lethal uncontrolled cavitary bleeding. Therapeutic approaches include laparotomy, angioembolization, and observation, alone or in combination. We hypothesized that angioembolization use increased and liver injury in-hospital mortality decreased over time.

Methods

We queried the 2017–2022 Trauma Quality Improvement Program database for adults with severe liver injury (Abbreviated Injury Scale score = 4/5). Management within the first 24 hours was categorized as laparotomy only, laparotomy → angioembolization, angioembolization only, angioembolization → laparotomy, or observation (no surgical or interventional radiology procedures). Cox proportional hazards models were used to adjust in-hospital mortality for confounders.

Results

18,445 patients were managed by laparotomy = 42.7%; observation = 47.0%; laparotomy → angioembolization = 5.1%; angioembolization = 4.5%; angioembolization → laparotomy = 0.8%. The confounder-adjusted use of angioembolization (alone or in combination with laparotomy) increased over time (P = .002). In-hospital mortality (18.7%) remained stable over time (P = .63). Compared to laparotomy-only, all other therapeutic approaches were associated with a lower adjusted hazard ratio: angioembolization-only = 0.81 (0.79–0.83); angioembolization → laparotomy = .89 (0.81–0.97); laparotomy → angioembolization = 0.88 (0.84–0.92); observation = 0.84 (0.83–0.85). Patients with admission hypotension and grade 5 (vs 4) liver injury experienced similar results. Of all therapeutic approaches, only observation was associated with a significant in-hospital mortality increase over time (adjusted hazard ratio = 1.003 [1.0001–1.005]).

Conclusions

Severe liver injury mortality remains high and unabated over recent years. Management with angioembolization has increased over time and was associated with improved survival, even when employed in patients hypotensive on admission and with grade 5 liver injury. These data support an integrated approach to severe liver injuries, possibly optimized in a hybrid operating room environment.