Effectiveness of olanzapine in the treatment of anorexia nervosa: A systematic review and meta‐analysis

Abstract Objective Anorexia nervosa (AN) is a severe psychiatric disorder characterized by starvation and malnutrition, a high incidence of coexisting psychiatric conditions, and treatment resistance. The effect of pharmacotherapy has been controversial. Method A systematic review was conducted for evidence of an effect of olanzapine versus placebo in adults or its effect as adjuvant treatment of AN in adolescents. Results A total of seven articles (304 patients with AN) were identified. There were four double‐blind, randomized studies examining the effect of olanzapine in the treatment of AN. The mean difference in body mass index (BMI) at the end of treatment between olanzapine and placebo was 0.67 kg/m2 (95% confidence interval (CI) 0.15–1.18 kg/m2; p = 0.01; I 2 = 0%, p for heterogeneity < 0.79). The olanzapine groups showed a significant increase in BMI of 0.68 kg/m2 (95% CI 0.22–1.13 kg/m2; p < 0.001; I 2 = 0%, p for heterogeneity = 0.74) compared to the placebo groups. Only two studies examined the effect of olanzapine as adjuvant treatment in adolescents and showed an increase in BMI of 0.66 kg/m2 (95% CI −0.36 to 1.67 kg/m2; p = 0.21; I 2 = 11%, p for heterogeneity = 0.32). Discussion Olanzapine showed efficacy in the treatment of AN with an increased BMI at the end of treatment in adults. The effect of olanzapine as adjuvant treatment in adolescents remains unclear.

There is an extensive history of medication trials for the pharmacological treatment of AN, based on the initial observation that individuals with AN exhibit core symptoms that are believed to suggest the presence of a biological disturbance. Given the near delusional quality in AN of some of the beliefs around food, weight, and body image, together with rigidity, obsessionality, and intense anxiety characteristic of AN, antipsychotic medications have also been proposed as potential therapeutic agents for AN. Several randomized, double-blind, placebo-controlled trials demonstrated both a greater rate of weight gain and higher BMI at the end of treatment (Attia et al., 2011(Attia et al., , 2019Bissada et al., 2008).
For children and adolescents with AN, the major guidelines recommend family-based treatment. The treatment of choice for young adults and adults with AN is the Maudsley Anorexia Nervosa Treatment for Adults (MANTRA), Cognitive Behavior Therapy-Enhanced (CBT-E), and Specialist Supportive Clinical Management (SSCM), but none of these treatments seems to be superior (Jansingh et al., 2020). Adjunctive olanzapine treatment for AN in adolescents was also reported recently (Norris et al., 2011;Spettigue et al., 2018).
A meta-analysis and systematic review of olanzapine were conducted recently (Meftah et al., 2020;Murray et al., 2019). Due to the limited number of studies, the conclusion regarding olanzapine in the treatment of AN was still unclear. This systematic review attempted to identify all controlled clinical trials of olanzapine in AN patients and assessed the effect on weight gain. Adult and adolescent AN patients were analyzed separately.

Overview
The study protocol followed the Meta-analysis Of Observational

Search strategy and selection criteria
Three major databases (Medline, CHAHL, and Web of Science) were searched on June 1, 2021. The two reviewers independently extracted and recorded data for a predefined checklist including the following items: study characteristics (i.e., country and year of study), characteristics of the cohort, and outcomes. The following search formula was used: ((anorexia nervosa) OR (eating disorder)) and (olanzapine).
Two review authors (RH and HC) independently screened the titles and abstracts and carefully evaluated the full text to select eligible articles. In cases of discrepancy, they reached a consensus through discussion. Review articles and included original articles were hand-searched (RH and HC) for additional research papers that met the inclusion criteria.
No restrictions were placed on article types or publication language.
To be included, a study had to include: (1) patients with AN; (2) olanzapine treatment; and (3) weight gain after treatment could be calculated.

Outcomes
The primary outcome was gain in weight. In placebo-controlled studies, the difference in BMI at the end of treatment between olanzapine and placebo was examined. Increased BMI after treatment by olanzapine or placebo was also identified. In adolescents using olanzapine as adjuvant treatment, only increased BMI was analyzed, because only increased BMI was available in one study.

Quality assessment
Two reviewers independently assessed the methodological quality of selected studies using the Newcastle-Ottawa Scale quality assessment to evaluate the quality of observational studies (Stang, 2010). Disagreements among reviewers were discussed, with agreement reached by consensus.

Statistics
All analyses were performed using Review Manager version 5.3

Role of the funding source
There was no funding source for this study.

RESULTS
A total of 397 articles, including 395 articles through database searching and two articles by hand searching, were identified. There were 243, 83, and 7 articles left after removing duplication, screening, and full-article reading, respectively (Figure 1) Three studies identified the effect of olanzapine as adjuvant treatment in adolescents. Newcastle-Ottawa Scale scores ranged from 6 to 7.
After treatment with olanzapine, the mean difference in BMI at the end of treatment between olanzapine and placebo was 0.67 kg/m 2 (95% confidence interval (CI) 0.15-1.18 kg/m 2 ; p = 0.01; I 2 = 0%, p for heterogeneity = 0.79) (Figure 2). Although all of the above studies were double-blind, randomized studies, due to the small sample sizes of several studies, the initial BMI might have been different before treatment. The increase in BMI after treatment was also There was no obvious publication bias in all subgroup analyses on funnel plots (Supporting Information Figures S3-S5).

DISCUSSION
This article identified five double-blind, randomized studies, including 304 patients with AN, showing that there were obvious differences in BMI at the end of treatment between groups, and the olanzapine group had a significant increase in BMI compared with the placebo group. This finding provides strong support for the efficacy of pharmacotherapy in adult AN. In a previous meta-analysis, olanzapine showed a trend in the treatment of AN, but due to the limited sample size, the conclusion was blurred. (Dold et al, 2015;Rosager et al., 2021) In adolescents, although olanzapine as adjuvant treatment showed a trend to improving BMI, due to the limited number of participants enrolled, there was no significant difference. There was no heterogeneity in all subgroup analyses, which made the conclusion reliable.
An absolute cut-off in terms of low BMI is not stipulated, since several other factors warrant consideration, including the patient's age, sex, BMI before the occurrence of symptoms, and rapidity of weight loss; however, a low weight (e.g., BMI ≤17.5 kg/m 2 ) is usually observed in adults with AN. Olanzapine treatment resulted in an increase in BMI, 0.6 kg/m 2 , at the end of treatment. It was difficult to evaluate the absolute value of the increased BMI. The body weight changed more frequently in the olanzapine group than in the placebo group according to a study of 152 AN cases, and the mechanism of waves in BMI values during the treatment was still unclear (Attia et al., 2019).
Olanzapine alone might be insufficient in the treatment of AN. An improved BMI is important in AN patients; cognitive remediation therapy (CRT), family therapy, and other nonpharmacological interventions also play important roles in the treatment of AN (Attia & Walsh, 2009;Fisher et al, 2018;Gan et al, 2021;Tchanturia et al, 2017).

CONCLUSION
Olanzapine showed efficacy in the treatment of AN, with weight gain at the end of treatment in adults. Its effect as adjuvant treatment in adolescents remains unclear. More comparative studies are needed.

CONFLICT OF INTEREST
The authors declare no conflict of interest.

AUTHOR CONTRIBUTIONS
RH and HC were involved in data acquisition and drafting the manuscript. HC contributed to study conception. All authors performed data acquisition, analysis, interpretation, and drafting. HC involved in interpretation and critical revision. All authors provided final approval and take accountability.

PEER REVIEW
The peer review history for this article is available at https://publons. com/publon/10.1002/brb3.2498

DATA AVAILABILITY STATEMENT
The raw data are available by email on reasonable request to the corresponding author. E-mail: chinsmd@gmail.com