The Value of Structural Neuroimaging in First-Episode Psychosis and the Prevalence of Imaging Abnormalities and Clinical Relevance: A Real-World Observational Study.

: The usefulness of neuroimaging in patients with first-episode psychosis (FEP) remains controversial. The aim of this study was to assess the prevalence and types of structural abnormalities on neuroimaging in patients with FEP and identify the most frequently used imaging modalities in a real-world setting. A retrospective observational study based on a consecutive series of patients admitted to our institution with FEP was conducted. We analyzed the imaging tests performed, the presence of specific lesions, the degree of cortical atrophy (Global Cortical Atrophy, GCA scale), medial temporal atrophy (Medial Temporal lobe Atrophy, MTA scale) and non-specific white matter lesions (Fazekas scale). Descriptive and bivariate analyses were performed according to previously established age cut-offs. A total of 105 patients were included (median age: 36 years; 52.4% men). The most frequently used neuroimaging test was computed tomography (CT) (92.4%). GCA scores that were out of the age range were found in 32.4% of patients, being more frequent in those older than 65 years ( < 0.001). Out-of-range MTA scores were found in 36.2% of patients, especially in patients older than 75 years ( < 0.001). Out-of-range Fazekas scores were found in 4.3% of patients, especially in patients older than 70 years ( = 0.157). Finally, only one specific structural lesion (right frontal cavernoma) was identified in one patient (1%). Overall, at least one non-age-matched abnormality was found in 46.7% of patients. Although non-specific alterations not in accordance with age exist in a significant percentage of patients with FEP, the prevalence of specific lesions is very low. This suggests that neuroimaging tests could be restricted in patients with FEP, especially CT, due to the risks associated with ionizing radiation. However, further prospective and controlled studies are needed to validate our results.