Diagnostic accuracy of computed tomography angiography (CTA) for diagnosing blunt cerebrovascular injury in trauma patients: a systematic review and meta-analysis

Abstract

OBJECTIVES: Previous literature showed that the diagnostic accuracy of computed tomographic angiography (CTA) is not equally comparable with that of the rarely used golden standard of digital subtraction angiography (DSA) for detecting blunt cerebrovascular injuries (BCVI) in trauma patients. However, advances in CTA technology may prove CTA to become equally accurate. This study investigated the diagnostic accuracy of CTA in detecting BCVI in comparison with DSA in trauma patients. METHODS: An electronic database search was performed in PubMed, EMBASE, and Cochrane Library. Summary estimates of sensitivity, specificity, positive and negative likelihood, diagnostic odds ratio, and 95% confidence intervals were determined using a bivariate random-effects model. RESULTS: Of the 3293 studies identified, 9 met the inclusion criteria. Pooled sensitivity was 64% (95% CI, 53-74%) and specificity 95% (95% CI, 87-99%) The estimated positive likelihood ratio was 11.8 (95%, 5.6-24.9), with a negative likelihood ratio of 0.38 (95%, 0.30-0.49) and a diagnostic odds ratio of 31 (95%, 17-56). CONCLUSION: CTA has reasonable specificity but low sensitivity when compared to DSA in diagnosing any BCVI. An increase in channels to 64 slices did not yield better sensitivity. There is a risk for underdiagnosis of BCVI when only using DSA to confirm CTA-positive cases, especially in those patients with low-grade injuries. KEY POINTS: • Low sensitivity and high specificity were seen in identifying BCVI with CTA as compared to DSA. • Increased CTA detector channels (≤ 64) did not lead to higher sensitivity when detecting BCVI. • The use of CTA instead of DSA may lead to underdiagnosis and, consequently, undertreatment of BCVI.