Computed tomography angiography as a confirmatory test for the diagnosis of brain death

OBJECTIVE For a diagnosis of brain death (BD), ancillary testing is performed if patient factors prohibit a complete clinical examination and apnea test. The American Academy of Neurology (AAN) guidelines identify cerebral angiography (CA), cerebral scintigraphy, electroencephalography, and transcranial Doppler ultrasonography as accepted ancillary tests. CA is widely considered the gold standard of these, as it provides the most reliable assessment of intracranial blood flow. CT angiography (CTA) is a noninvasive and widely available study that is also capable of identifying absent or severely diminished intracranial blood flow, but it is not included among the AAN's accepted ancillary tests because of insufficient evidence demonstrating its reliability. The objective of this study was to assess the statistical performance of CTA in diagnosing BD, using clinical criteria alone or clinical criteria plus CA as the gold-standard comparisons. METHODS The authors prospectively enrolled 22 adult patients undergoing workup for BD. All patients had cranial imaging and clinical examination results consistent with BD. In patients who met the AAN clinical criteria for BD, the authors performed CA and CTA so that both tests could be compared with the gold-standard clinical criteria. In cases that required ancillary testing, CA was performed as a confirmatory study, and CTA was then performed to compare against clinical criteria plus CA. Radiographic data were evaluated by an independent neuroradiologist. Test characteristics for CTA were calculated. RESULTS Four patients could not complete the standard BD workup and were excluded from analysis. Of the remaining 18 patients, 16 met AAN criteria for BD, 9 of whom required ancillary testing with CA. Of the 16 patients, 2 who also required CA ancillary testing were found to have persistent intracranial flow and were not declared brain dead at that time. These patients also underwent CTA; the results were concordant with the CA results. Six patients who were diagnosed with BD on the basis of clinical criteria alone also underwent CA, with 100% sensitivity. For all 18 patients included in the study, CTA had a sensitivity of 75%, a specificity of 100%, a positive predictive value of 100%, and a negative predictive value of 33%. CONCLUSIONS Clinical examination with or without CA remains the gold standard in BD testing. Studies assessing the statistical performance of CTA in BD testing should compare CTA to these gold standards. The statistical performance of CTA in BD testing is comparable to several of the nationally accepted ancillary tests. These data add to the growing medical literature supporting the use of CTA as a reliable ancillary test in BD testing.