Ounali Jaffer MBBS, MRCP, Presenter: Nothing to Disclose

Dylan Lewis MBBS, Abstract Co-Author: Nothing to Disclose

Vanash Patel, Abstract Co-Author: Nothing to Disclose

Shirin Khorsandi, Abstract Co-Author: Nothing to Disclose

Nigel Heaton, Abstract Co-Author: Nothing to Disclose

William Bernal, Abstract Co-Author: Nothing to Disclose

Pauline Anne Kane MBBS, Abstract Co-Author: Nothing to Disclose

John Barr Karani MBBS, Abstract Co-Author: Nothing to Disclose

Currently, CT liver injury grade is used to predict the need for surgery following trauma. This, however, does not consider the increased use of procedures such as arterial angiography/embolisation or endoscopic biliary stenting. Our study aims to investigate whether CT injury grade can be used to predict the need for these minimally invasive procedures following blunt or penetrating liver injury.  

We retrospectively reviewed 114 consecutive patients with liver trauma from November 1998 to January 2009. The age, CT injury grade, mechanism of injury, associated injuries, type of surgical, radiological and endoscopic intervention, Charlson Score, length of hospital stay and survival data were recorded. Both univariate (Spearman’s Rank Correlation) and multivariate linear regression models were used to determine which variables could be predicted by CT injury grade. To assess mechanism of injury as a predictor, a multivariate logistic regression model was used.

The median age was 31 years (range 5–81 years) and they presented with CT injury grade II (n=23), grade III (n=47), grade IV (n=36), and grade V (n=8). Univariate analysis of CT liver injury grading demonstrated a positive correlation with need for hepatic angiography (ρ=0.42,p=<0.001), hepatic arterial embolisation (ρ=0.35,p=<0.001), endoscopic biliary stenting (ρ=0.18,p=0.05), length of hospital stay (ρ=0.37,p=<0.001), liver surgery (ρ=0.32,p=<0.001) and recurrent liver surgery (ρ= 0.28,p=0.015). A negative correlation was demonstrated with no intervention (ρ=-0.34,p=<0.001), survival (ρ=-0.23,p=0.014) and Charlson Score (ρ=-0.22,p=0.018). Multivariate analysis showed that hepatic angiography and death were significantly associated with higher CT injury grade (r=0.60,p=<0.001 and r=-0.32,p=<0.001 respectively). Hepatic angiography and liver surgery was 65 times (p=0.015) and 22 times (p=0.04) more likely in penetrating in comparison to blunt trauma.      

Predicting the need for intervention in liver trauma using CT injury grade should consider minimally invasive procedures as well as surgery. In patients with penetrating liver injury there should be a low threshold for using hepatic angiography in addition to surgery.

CT grading of post-traumatic liver injury correlates with escalating need for hepatic angiography; early recognition of this association may have profound implications for patient care. 

Jaffer, O, Lewis, D, Patel, V, Khorsandi, S, Heaton, N, Bernal, W, Kane, P, Karani, J, Do Liver Trauma Scoring Systems Predict Requirement for Angiographic Intervention and Patient Outcome?.  Radiological Society of North America 2011 Scientific Assembly and Annual Meeting, November 26 - December 2, 2011 ,Chicago IL. http://archive.rsna.org/2011/11004229.html