T159. STRUCTURAL BRAIN ABNORMALITIES IN SCHIZOPHRENIA IN ADVERSE ENVIRONMENTS: EXAMINING THE EFFECT OF POVERTY AND VIOLENCE IN SIX LATIN AMERICAN CITIES

BACKGROUND: Social and environmental factors such as poverty or violence, modulate the risk and course of schizophrenia, but how they affect the brain in patients with psychosis remains unclear. We here studied how they are related to brain structure in schizophrenia and healthy controls in Latin America, where these factors are large and unequally distributed. METHODS: This is an MRI multi-center study in patients with schizophrenia and healthy controls from six Latin American cities: Buenos Aires, Medellin, Mexico City, Santiago, Sao Paulo and Porto Alegre. Total and voxel-level gray matter volumes obtained from T1-weighted MRI images and their relationship with income and homicide rates were analyzed using a general linear model. RESULTS: 334 patients with schizophrenia and 262 controls were included. Income was differentially related to total gray matter volume in the two groups (P=0.006). Controls showed a positive correlation between total gray matter volume and income (R=0.14, P=0.02). Surprisingly, this relationship was not present in schizophrenia (R=-0.076, P=0.17). Voxel-level analysis confirmed that this interaction was widespread across the cortex. After adjusting for global brain changes, income was positively related to prefrontal cortex volumes only in controls. Conversely, the hippocampus in patients, but not in controls, was relatively larger in affluent environments. There was no significant correlation between environmental violence and brain structure. DISCUSSION: Our results highlight the interplay between the environment, particularly poverty, and individual characteristics in psychosis. This is particularly important for harsh environments such as those from low and middle-income countries: potentially less brain vulnerability (less gray matter loss) is sufficient to become unwell in adverse (poor) environments. The development of algorithms exploring clinically-useful information from structural brain images in psychosis should include representative samples from low and middle-income countries.