Diagnostic yield of computed tomography of the brain in first episode psychosis

Brain computed tomography (CT) is inconsistently recommended worldwide in the investigative algorithm of patients presenting with first episode psychosis (FEP). The objective of this study is to investigate the clinical efficacy of brain CT in patients presenting with FEP without neurological signs in a major metropolitan teaching hospital.

The CT brain scan reports of 237 consecutive patients, for which the imaging requests or reports provided a history of FEP but no focal neurological signs, were retrospectively identified within a 6-year period in a 750-bed tertiary referral teaching hospital using the radiology information system text-search function (170 male, 67 female; mean age, 28.3 years). All reports were authored or approved by consultant radiologist. They were reviewed for the presence of any lesion that could cause psychosis and hence alter clinical management. Minor neuroradiological abnormalities were also noted. Hospital Ethics Committee registration and approval were obtained and patient informed consent was not required.

No focal brain lesion potentially responsible for the psychosis or focal lesion requiring surgical intervention was identified in any patient. Findings unable to be directly linked to the psychosis such as evidence of small vessel ischaemic disease, arachnoid cysts, cerebral atrophy, and normal variants were present in 17.6% of patients (45 of 237 studies), none of which led to an alteration of clinical management.

The results of this study postulate that brain CT should not be universally performed in the initial assessment of patients with first episode psychosis without neurological signs.