Comparing the Denver criteria sets for blunt trauma: a retrospective study of cases in Edmonton, Alberta

Abstract

Objective: To determine whether a more conservative Denver criterion set could reduce unnecessary CT angiography (CTA) studies when screening for blunt cerebrovascular injury (BCVI) following blunt trauma. Methods: Following ethics approval, a retrospective chart review of 447 consecutive patients undergoing emergency CTA at two large teaching hospitals was conducted to determine the presence of risk factors for each Denver criterion set. Imaging studies of adults conducted between January 2016 and June 2020 containing sufficient clinical information for accurate classification were included in the study. Specificity, sensitivity, and predictive values were calculated. A two-sided Fisher exact test was used to evaluate the association between each iteration of the Denver criteria and the presence of BCVI. Results: The specificities of the Original, Modified, and Expanded Denver criteria were 43.58%, 34.32%, and 24.85%, respectively. Positive-predictive values (PPV) followed a different trend, with respective values of 2.77%, 3.06%, and 2.78%. Sensitivity and negative-predictive values (NPV) were found to be 100% for each criterion set. Being positive for a criterion set, and the presence of BCVI, was statistically significant for the original Denver criteria (p = 0.021, n = 443), but not the modified (p = 0.100, n = 345) or expanded Denver criteria (p = 0.202, n = 333). Conclusion: Use of the modified and expanded Denver criteria leads to the overuse of cerebrovascular imaging on patients suffering blunt force trauma. Advances in knowledge: The original Denver criteria may more appropriately identify subjects for further evaluation with CTA than the current standard, while retaining diagnostic efficacy for BCVI.