Body and liver fat content and adipokines in schizophrenia: a magnetic resonance imaging and spectroscopy study.

RATIONALE: Although antipsychotic treatment often causes weight gain and lipid abnormalities, quantitative analyses of tissue-specific body fat content and its distribution along with adipokines have not been reported for antipsychotic-treated patients.

OBJECTIVES

The purposes of the present study were to quantitatively assess abdominal and liver fat in patients with schizophrenia on antipsychotic treatment and age- and body mass index (BMI)-matched healthy controls and to evaluate their associations with plasma leptin and adiponectin levels.

METHODS

In 13 schizophrenia patients on antipsychotic treatment and 11 age- and BMI-matched controls, we simultaneously quantified visceral and subcutaneous fat content using T1-weighted magnetic resonance imaging and liver fat content by 1H magnetic resonance spectroscopy. Associations of tissue-specific fat content with plasma levels of leptin and adiponectin were evaluated.

RESULTS

Plasma adiponectin level (μg/mL) was not statistically different between groups (7.02 ± 2.67 vs. 7.59 ± 2.92), whereas plasma leptin level (ng/mL) trended to be higher in patients than in controls (11.82 ± 7.89 vs. 7.93 ± 5.25). The values of liver fat (%), visceral fat (L), and subcutaneous fat (L) were 9.64 ± 8.03 vs. 7.07 ± 7.35, 4.41 ± 1.64 vs. 3.31 ± 1.97, and 8.37 ± 3.34 vs. 7.16 ± 2.99 in patients vs. controls, respectively. Liver fat content was inversely correlated with adiponectin in controls (r =  - 0.87, p < 0.001) but not in patients (r =  - 0.26, p = 0.39). In both groups, visceral fat was inversely associated with adiponectin (controls : r =  - 0.66, p = 0.03; patients : r =  - 0.65, p = 0.02), while subcutaneous fat was positively correlated with leptin (controls : r = 0.90, p < 0.001; patients : r = 0.67, p = 0.01).

CONCLUSIONS

These findings suggest that antipsychotic treatment may disrupt the physiological relationship between liver fat content and adiponectin but does not essentially affect the associations of adiponectin and leptin with visceral and subcutaneous compartments.