• 
• 

Three Part Question

In [a patient with an anterior abdominal stab wound] does [CT scanning] reliably detect [intra-abdominal injury including peritoneal or hollow viscus perforation].

Clinical Scenario

A 35-year-old male presents to the Emergency Department with an anterior abdominal stab wound (AASW). He is haemodynamically stable; you are unsure what the best method of investigation is to detect any significant intra-abdominal injury including hollow viscus perforation.

Search Strategy

ICE Healthcare Databases (1985-2016) including: AMED, PubMED, BNI, EMBASE, HBE, HMIC, Medline, PsycINFO, CINAHL. Search terms included (anterior abdominal stab).ti,ab OR (abdominal stab investigations).ti,ab.

Search Outcome

81 papers found. Abstracts were reviewed for relevance of which 16 were duplicates, 55 were irrelevant and 3 were of insufficient quality for inclusion. 6 papers were relevant and of sufficient quality and are included in this BET. Relevant papers are shown in the table below.

Relevant Paper(s)

Author, date and country	Patient group	Study type (level of evidence)	Outcomes	Key results	Study Weaknesses
Biffl WL et al 2011 United States	160 adult haemodynamically stable patients with AASW enrolled from 4 centres (2008-2010). Management was guided by a previously stipulated algorithm which did not include CT, Biffl et al, 2009. 47/160 underwent CT scan, which was counted as a deviation from protocol. Pregnant and incarcerated patients were excluded. 29% had multiple stab wounds.	Multi-centre prospective observational study	Summary of CT values for therapeutic laparotomy:	Sensitivity 89% Specificity 82% PPV 53% NPV 93%	CT contrast administration protocol was not standardised in all patients. Self-reporting by surgeons regarding therapeutic value of surgery. Limited size of patient group. Possible under reporting of patients discharged after negative local wound exploration (LWE) and CT. Possible bias as study set out to validate a protocol which did not include CT.
			Non-therapeutic laparotomy rate for CT	31%
			Immediate discharge based on negative CT	34% (16 pts) [despite 60% of the group having no CT findings]
Plackett TP et al 2011 United States	1,961 patients with AASW (1992 – 2008). Paediatric and adult patients. 47 patients in the delayed laparotomy group (268 pts) underwent CT	Multi-centre retrospective study	Negative findings on laparotomy in CT group	12.2% (5 pts)	Due to long study period and multiple individual factors, the effect of CT scanning can not be definitively measured. Small group of patients undergoing CT.
			Negative findings on laparotomy in non-CT group	21.9% (58 pts)
Biffl WL et al 2009 United States	359 adult patients with AASW (2006-2007). 138 (50%) of haemodynamically stable, asymptomatic patients used CT as primary decision-making tool.	Multi-centre prospective observational study	Summary of CT values for therapeutic laparotomy:	Pre-test probability 24% Sensitivity 77% Specificity 73% PPV 47% NPV 91% Non-therapeutic laparotomy rate for CT 24%	Management not controlled prospectively. Management based on individual/ institutional protocol and not defined. Therapeutic assessment of surgery subjectively decided by surgeon. Unequal enrolment. No long-term data
Lee GL et al 2015 Germany	118 patients with AASW (2004-2012), 10 patients straight to laparotomy due to instability. 108 patients had CT scans.	Retrospective study	Diagnostic performance of CT in detecting abdominal stab injuries:	Sensitivity 94.2% Specificity 66.7% PPV 98.8% NPV 28.6%	Small study group. Cannot clearly correlate need for operation and CT findings as no consistent criteria.
Berardoni NE et al 2011 United States	98 patients (paediatric and adult) with AASW who underwent CT.	Retrospective review	Diagnostic performance of CT in detecting intra-abdominal injury:	Sensitivity 96% Specificity 97% PPV 93% NPV 99%	Small study group. No acknowledgement of limitations of the study in discussion.
			Diagnostic performance of CT for determining need for surgical intervention:	Sensitivity 93% Specificity 93% PPV 70% NPV 99% Accuracy 93%
Salim A et al 2006 United States	156 consecutive haemodynamically stable AASW patients, 2004-2006. All patients underwent serial clinical examinations +/- CT scan.	Prospective Observational Study	Value of CT for clinical outcomes (need for laparotomy, uneventful discharge without laparotomy, or return to hospital for adverse event)	Sensitivity 100% Specificity 81% PPV 42% NPV 100%	CT scans were obtained from only 43% of the patients and at discretion of surgeon. Only single contrast CT used. Outcome did not specifically report peritoneal penetration. No comparison on length of stay.
			Overall accuracy for determining intervention	84%

Comment(s)

This search revealed three prospective studies and three retrospective studies with no randomised control trials comparing CT to other methods of investigation. The studies excluded patients with haemodynamic instability. One prospective study was protocol-based, where CT was considered a deviation from protocol - this study was included due to the fact that moderate numbers of CTs were performed, despite the protocol. Different clinical outcomes were reported amongst the studies including detection of intra-abdominal injuries and therapeutic and non-therapeutic laparotomy values.

Within the haemodynamically stable population with AASW, sensitivity of CT scanning ranged from 77– 100%. Specificity is harder to assess as not all the studies assessed the number of false negatives, but in those that did it was found to be between 66.7 – 100%. Despite the high rate of true negative results, a negative CT did not automatically result in patient discharge; Biffl (2011) reported only 34% of patients with a negative CT scan were discharged home. However, Berardoni et al reported that in the 80% of patients with a negative CT scan who were admitted, mostly this was for psychiatric reasons, intoxication or associated injuries. Non-therapeutic laparotomy rate following CT ranged from 9-30%, which was similar to other methods of investigation reported by Biffl (2009), including diagnostic peritoneal lavage (31%), and local wound exploration (54%). All studies commented on the difficulty in attributing clear correlation between CT scan results and need for theatre due to variability in surgeon preferences. Consistency in the decision to request a CT was also variable, making it difficult to understand what factors influenced the decision to undertake a CT scan. These studies suggest CT scanning is a useful method of investigation and can be a valuable addition to clinical examination and judgement in the management of a haemodynamically stable AASW patient. However, from the available evidence, CT is not sensitive enough to definitively rule out peritoneal perforation or hollow viscus injury. In addition to clinical factors, it is likely that organisational factors e.g. skill set and availability of resources should also be taken into consideration. More evidence is needed including defined patient selection protocol and randomised controlled trials to definitively recommend CT scanning as an optimum investigation of AASW.

Editor Comment

retained for Trauma

Clinical Bottom Line

Although evidence is limited, it appears CT scanning can be useful as an adjunct to clinical examination and judgement in assessing the need for laparotomy in haemodynamically stable patients, as well as decreasing the number of non-therapeutic laparotomies. However, a negative CT is not sensitive enough to definitively rule out peritoneal perforation or hollow viscus injury.

References

1. Biffl WL, Kaups KL, Pham TN, Rowell SE, Jurkovich GJ, Burlew CC, Elterman J, Moore EE. Validating the Western Trauma Association algorithm for managing patients with anterior abdominal stab wounds: a Western Trauma Association multicenter trial. J Trauma 2011;71(6):1494–1502.
2. Plackett, Timothy P. DO; Fleurat, Jonathan MD; Putty, Brad MD; Demetriades, Demetrios MD, PhD; Plurad, David MD. Selective Nonoperative Management of Anterior Abdominal Stab Wounds: 1992:2008 Journal of Trauma-Injury Infection & Critical Care 2011; 70(2):408-414.
3. Biffl WL, Kaups KL, Cothren CC, et al. Management of patients with anterior abdominal stab wounds: A Western Trauma Association multicenter trial. J Trauma 2009;66:1294-1301.
4. Lee GJ, Son G, Yu BC, Lee JN, Chung M. Efficacy of computed tomography for abdominal stab wounds: a single institutional analysis. European Journal of Trauma and Emergency Surgery 2015; vol. 41 (no. 1); p. 69-74.
5. Berardoni NE, August DL, Kopelman TR, et al. Use of computed tomography in the initial evaluation of anterior abdominal stab wounds. American Journal of Surgery 2011; vol. 202 (no. 6); p. 690-696.
6. Salim A, Sangthong B, Martin M, et al. Use of computed tomography in anterior abdominal stab wounds: Results of a prospective study. Archives of Surgery 2006; vol. 141 (no. 8); p. 745-750