To determine the efficacy, acceptability, and tolerability of second-generation antipsychotic drugs (SGAs) in anorexia nervosa.

We covered all randomized controlled trials (RCTs) comparing a pharmacotherapy with SGAs to placebo or no treatment in anorexia nervosa. Primary outcome was weight gain assessed by mean change in body mass index (BMI). Secondary outcomes were change of anorectic symptoms assessed by the Yale–Brown–Cornell Eating Disorders Scale (YBC-EDS) score and the Eating Disorders Inventory (EDI) score, and premature discontinuation of treatment. Using the random-effects model of Der-Simonian and Laired standardized mean differences based on Hedges's g and risk ratios were calculated.

Seven RCTs investigating olanzapine (N=4), quetiapine (N=2), and risperidone (N=1) with 201 subjects were included. There was no significant between-group difference in mean BMI change when comparing the pooled group of SGA drugs with the pooled control group (N=6, n=152; Hedges's g=0.17, 95% CI: -0.14-0.48; p=0.28). None of the examined single SGAs was significantly superior to placebo/no treatment in achieving BMI gain. We found no significant between-group differences for all secondary outcomes with the exception of the EDI total score which decreased statistically significant more in the pooled control group. The antipsychotic medication was well tolerated.

Because of lacking evidence SGAs cannot be generally recommended in anorexia nervosa although some individuals or subgroups of patients might respond to an antipsychotic medication. Further research is needed to identify which anorectic patients could probably benefit from treatment with antipsychotic drugs. The medication with SGAs was well accepted and tolerated in anorectic patients.