D enver screening protocol for blunt cerebrovascular injury reduces the use of multi‐detector computed tomography angiography

Background Blunt cerebrovascular injury ( BCVI ) occurs in 0.2–2.7% of blunt trauma patients and has up to 30% mortality. Conventional screening does not recognize up to 20% of BCVI patients. To improve diagnosis of BCVI , both an expanded battery of screening criteria and a multi‐detector computed tomography angiography ( CTA ) have been suggested. The aim of this study is to investigate whether the use of CTA restricted to the D enver protocol screen‐positive patients would reduce the unnecessary use of CTA as a pre‐emptive screening tool. Methods This is a registry‐based study of blunt trauma patients admitted to Auckland City Hospital from 1998 to 2012. The diagnosis of BCVI was confirmed or excluded with CTA , magnetic resonance angiography and, if these imaging were non‐conclusive, four‐vessel digital subtraction angiography. Results Thirty (61%) BCVI and 19 (39%) non‐ BCVI patients met eligibility criteria. The D enver protocol applied to our cohort of patients had a sensitivity of 97% (95% confidence interval ( CI ): 83–100%) and a specificity of 42% (95% CI : 20–67%). With a prevalence of BCVI in blunt trauma patients of 0.2% and 2.7%, post‐test odds of a screen‐positive test were 0.03 (95% CI : 0.002–0.005) and 0.046 (95% CI : 0.314–0.068), respectively. Conclusions Application of the CTA to the D enver protocol screen‐positive trauma patients can decrease the use of CTA as a pre‐emptive screening tool by 95–97% and reduces its hazards.