Neural Responsivity to Food Cues in Patients With Unmedicated First-Episode Psychosis

IMPORTANCE Schizophrenia is associated with a reduced life expectancy of 15 to 20 years owing to a high prevalence of cardiometabolic disorders. Obesity, a key risk factor for the development of cardiometabolic alterations, is more prevalent in individuals with schizophrenia. Although obesity is linked to the altered reward processing of food cues, no studies have investigated this link in schizophrenia without the confounds of antipsychotics and illness chronicity. OBJECTIVE To investigate neural responsivity to food cues in first-episode psychosis without the confounds of antipsychotic medication or illness chronicity. DESIGN, SETTING, AND PARTICIPANTS A case-control study was conducted from January 31, 2015, to September 30, 2018, in London, United Kingdom, of 29 patients with first-episode psychosis who were not taking antipsychotic medication and 28 matched controls. MAIN OUTCOMES AND MEASURES Participants completed a food cue paradigm while undergoing a functional magnetic resonance imaging scan. Neural activation was indexed using the blood oxygen level-dependent hemodynamic response. The Dietary Instrument for Nutrition Education was used to measure diet, and the International Physical Activity Questionnaire was used to measure exercise. RESULTS There were no significant differences in age, sex, or body mass index between the 29 patients (25 men and 4 women; mean [SD] age, 26.1 [4.8] years) and 28 controls (22 men and 6 women; mean [SD] age, 26.4 [5.5] years). Relative to controls, patients consumed more saturated fat (t(46)=-3.046; P=.004) and undertook less high-intensity (U=304.0; P=.01) and low-intensity (U=299.5; P=.005) weekly exercise. There were no group differences in neural responses to food vs nonfood cues in whole-brain or region-of-interest analyses of the nucleus accumbens, insula, or hypothalamus. Body mass index was inversely correlated with the mean blood oxygen level-dependent signal in the nucleus accumbens in response to food vs nonfood cues in controls (R=-0.499; P=.01) but not patients (R=0.082; P=.70). CONCLUSIONS AND RELEVANCE Relative to controls, patients with first-episode psychosis who were not taking antipsychotic medication consumed more saturated fat and showed an altered association between body mass index and neural response to food cues in the absence of differences in neural responses to food cues. These findings highlight how maladaptive eating patterns and alterations in the association between body mass index and neural responses to food cues are established early in the course of schizophrenia.