Comparison of atriocaval shunting with perihepatic packing versus perihepatic packing alone for retrohepatic vena cava injuries in a swine model

Background Retrohepatic vena cava (RVC) injuries are technically challenging and often lethal. Atriocaval shunting has been promoted as a modality to control haemorrhage from these injuries, but evidence from controlled studies supporting its benefit is lacking. We hypothesised that addition of an atriocaval shunt to perihepatic packing would improve outcomes in our penetrating RVC injury swine model. Methods After a survivable atriocaval shunting model was refined in 4 swine without an injury, 13 additional female Yorkshire swine were randomised into either perihepatic packing and atriocaval shunt (PPAS, n = 7) or perihepatic packing alone (PP, n = 6) treatment arms prior to creating a standardised, 1.5 cm stab wound to the RVC. Haemodynamic parameters, intravenous fluid, and blood loss were recorded until mortality or euthanisation after 4 h. Statistical tests used to test differences include the Wilcoxon rank sums test, Fisher exact test and analysis of covariance. A p -value ≤0.05 was considered statistically significant. Results Immediately before and after RVC injury, no difference in temperature, cardiac output, heart rate, mean arterial pressure or mean pulmonary artery pressure was detected (all p > 0.05) between the two groups. While the RVC injury did affect measures parameters in PPAS swine over time, haemodynamic compromise and blood loss were not significantly greater in PPAS than PP swine. Survival time was significantly different with all PPAS swine dying within 2 h (mean survival duration 39 (SD 58) min) while all 6 PP swine survived the entire 4 h study period. Conclusions While perihepatic packing alone slowed haemorrhage to survivable rates during the 4 h study period, atriocaval shunt placement led to rapid physiologic decline and death in our standardised, penetrating RVC model.