Confirmatory tests in the diagnosis of brain death: comparison between SPECT and contrast angiography

OBJECTIVE: Cerebral blood flow tests have increasingly been advocated for the confirmation of brain death (BD). Four-vessel angiography has been considered the most reliable investigation in the diagnosis of BD for >30 yrs, but it is invasive. (99m)Tc-HMPAO SPECT provides noninvasive, multiplanar imaging of brain tissue perfusion. The aim of this study was to check the reliability of SPECT compared with contrast angiography. DESIGN: Prospective, blind study. SETTING: Neurointensive care unit of a university hospital. PATIENTS: Consecutive clinically brain dead patients with flat electroencephalogram. INTERVENTIONS: BD was diagnosed according to Italian law. (99m)Tc-HMPAO SPECT and four-vessel angiography were performed in the same session; the rater of each investigation ignored the results of the other. Blood pressure, Sp(O2), and P(ECO2) were monitored throughout the study: any episode of hypoxia or hypotension caused exclusion of the patient from the study. MEASUREMENTS AND MAIN RESULTS: Twenty brain dead patients were enrolled. The cause of BD was head injury in seven cases (35%), subarachnoid hemorrhage in seven (30%), spontaneous hemorrhage in one (10%), brain tumors in two (10%), stroke in two (10%), and thrombosis of the sagittal sinus in one (5%). Both angiography and SPECT confirmed BD in 19 of 20 patients: angiography showed the absence of filling of intracranial arteries, while SPECT showed a picture of "empty skull." For the remaining patient, angiography showed slight and late filling of left vertebral, basilar, and posterior cerebral arteries, while SPECT showed faint traces of uptake in the posterior fossa on the right side and on the midline. For this patient, the tests were repeated 48 hrs later, and both showed the arrest of intracranial circulation, thus confirming BD. CONCLUSIONS: Our results confirm the reliability of SPECT in the diagnosis of BD; because SPECT is noninvasive, it is a good candidate for the "gold standard" of diagnosis.