Computed tomography angiography accuracy in brain death diagnosis.

OBJECTIVE

The present study was designed to answer several concerns disclosed by systematic reviews indicating no evidence to support the use of computed tomography angiography (CTA) in the diagnosis of brain death (BD). Therefore, the aim of this study was to assess the effectiveness of CTA for the diagnosis of BD and to define the optimal tomographic criteria of intracranial circulatory arrest.

METHODS

A unicenter, prospective, observational case-control study was undertaken. Comatose patients (Glasgow Coma Scale score ≤ 5), even those presenting with the first signs of BD, were included. CTA scanning of arterial and venous vasculature and transcranial Doppler (TCD) were performed. A neurological determination of BD and consequently determination of case (BD group) or control (no-BD group) was conducted. All personnel involved with assessing patients were blinded to further tests results. Accuracy of BD diagnosis determined by using CTA was calculated based on the criteria of bilateral absence of visualization of the internal cerebral veins and the distal middle cerebral arteries, the 4-point score (4PS), and an exclusive criterion of absence of deep brain venous drainage as indicated by the absence of deep venous opacification on CTA, the venous score (VS), which considers only the internal cerebral veins bilaterally.

RESULTS

A total of 106 patients were enrolled in this study; 52 patients did not have BD, and none of these patients had circulatory arrest observed by CTA or TCD (100% specificity). Of the 54 patients with a clinical diagnosis of BD, 33 met the 4PS (61.1% sensitivity), whereas 47 met the VS (87% sensitivity). The accuracy of CTA was time related, with greater accuracy when scanning was performed less than 12 hours prior to the neurological assessment, reaching 95.5% sensitivity with the VS.

CONCLUSIONS

CTA can reliably support a diagnosis of BD. The criterion of the absence of deep venous opacification, which can be assessed by use of the VS criteria investigated in this study, can confirm the occurrence of cerebral circulatory arrest.Clinical trial registration no.: 12500913400000068 (clinicaltrials.gov).