Ultrasonography for the diagnosis of extra-cranial carotid occlusion - Diagnostic test accuracy meta-analysis

Background: The correct diagnosis of internal carotid artery (ICA) occlusion is crucial as it limits unnecessary intervention, whereas correct identification of patients with severe ICA stenosis is paramount in decision making and selecting patients who would benefit from intervention. We aimed to evaluate the accuracy of ultrasonography (US) in the diagnosis of ICA occlusion. Methods: We conducted a systematic review in compliance with the Preferred Reporting Items for a Systematic Review and Meta-analysis (PRISMA) of diagnostic test accuracy studies. We interrogated electronic bibliographic sources using a combination of free text and thesaurus terms to identify studies assessing the diagnostic accuracy of US in ICA occlusion. We used a mixed-effects logistic regression bivariate model to estimate summary sensitivity and specificity. We developed hierarchical summary receiver operating characteristic (HSROC) curves. Results: We identified 23 studies reporting a total of 5,675 arteries of which 722 were proven to be occluded by the reference standard. The reference standard was digital subtraction or cerebral angiography in all but two studies, which used surgery to ascertain a carotid occlusion. The pooled estimates for sensitivity and specificity were 0.97 (95% confidence interval (CI) 0.94 to 0.99) and 0.99 (95% CI 0.98 to 1.00), respectively. The diagnostic odds ratio was 3,846.15 (95% CI 1,375.74 to 10,752.65). The positive and negative likelihood ratio were 114.71 (95% CI 58.84 to 223.63) and 0.03 (95% CI 0.01 to 0.06), respectively. Conclusions: US is a reliable and accurate method in diagnosing ICA occlusion. US can be used as a screening tool with cross-sectional imaging being reserved for ambiguous cases.