efficacy and safety of adjunctive topiramate for...

Meta-analysis

Efficacy and safety of adjunctive topiramatefor schizophrenia: a meta-analysis ofrandomized controlled trials

Zheng W, Xiang Y-T, Xiang Y-Q, Li X-B, Ungvari GS, Chiu HFK,Correll CU. Efficacy and safety of adjunctive topiramate forschizophrenia: a meta-analysis of randomized controlled trials.

Objective: To systematically examine the randomized controlledtrial (RCT) evidence regarding efficacy and tolerability oftopiramate cotreatment with antipsychotics in schizophrenia-spectrum disorders.Methods: Random-effects meta-analysis of RCTs of topiramatecotreatment with antipsychotics vs. placebo/ongoing antipsychotictreatment in schizophrenia-spectrum disorders. Standardized orweighted mean difference (SMD/WMD), risk ratio (RR) �95%confidence intervals (CIs), and number needed to harm (NNH) werecalculated.Results: Across 16 RCTs (n = 934, duration = 11.8 � 5.6 weeks),topiramate outperformed the comparator regarding change/endpoint of total (SMD: �0.58, 95% CI: �0.82, �0.35,P < 0.00001), positive (SMD: �0.37, 95% CI: �0.61, �0.14,P = 0.002), negative (SMD: �0.58, 95% CI: �0.87, �0.29,P < 0.0001), and general symptoms (SMD: �0.68, 95% CI: �0.95,�0.40, P < 0.00001). Furthermore, topiramate was superiorregarding body weight (WMD: –2.75 kg, 95% CI: �4.03, �1.47,P < 0.0001), body mass index (BMI) (WMD: –1.77, 95% CI:�2.38, �1.15, P < 0.00001), triglycerides (P = 0.006), and insulinlevels (P < 0.00001). Superiority regarding psychopathology andbody weight/BMI was consistent across Chinese/Asian andWestern RCTs, double-blind and open designs, clozapine and non-clozapine cotreatment, augmentation and co-initiation RCTs, andhigher and lower quality RCTs. In meta-regression analyses,topiramate’s efficacy for total symptoms was moderated by shorterillness duration (P = 0.047), while weight loss was greater inprevention/co-initiation vs. intervention/augmentation RCTs(�4.11 kg, 95% CI: �6.70, �1.52 vs. �1.41 kg, 95% CI: �2.23,�0.59, P < 0.001). All-cause discontinuation was similar betweentopiramate and comparators (RR: 1.28, 95% CI: 0.91, 1.81,P = 0.16). While topiramate led to more concentration/attentiondifficulties (P = 0.03, NNH = 8, 95% CI=4–25), psychomotorslowing (P = 0.02, NNH = 7, 95% CI = 4–25), and paresthesia(P = 0.05, NNH = 2, 95% CI = 4–33), it led to less ≥7% weightgain (P = 0.0001, NNH = 2, 95% CI = 2–3) and constipation(P = 0.04, NNH = 9, 95% CI = 5–100) than the comparator.Conclusions: These results indicate that adjunctive topiramate toantipsychotics is an effective and safe treatment choice for symptomaticimprovement and weight reduction in patients with schizophrenia-spectrum disorders.

W. Zheng1, Y.-T. Xiang2,3,Y.-Q. Xiang4, X.-B. Li4,G. S. Ungvari5,6, H. F. K. Chiu7,C. U. Correll8,91The Affiliated Brain Hospital of Guangzhou MedicalUniversity (Guangzhou Huiai Hospital), Guangzhou,China, 2The National Clinical Research Center forMental Disorders, China & Beijing Anding Hospital,Capital Medical University, Beijing, China, 3Unit ofPsychiatry, Faculty of Health Sciences, University ofMacau, Macao, China, 4Beijing Anding Hospital, CapitalMedical University, Beijing, China, 5The University ofNotre Dame Australia/Marian Centre, Perth, WA,Australia, 6School of Psychiatry & ClinicalNeurosciences, University of Western Australia, Perth,WA, Australia, 7Department of Psychiatry, ChineseUniversity of Hong Kong, Hong Kong, China, 8Division ofPsychiatry Research, The Zucker Hillside Hospital,Northwell Health, Glen Oaks, NY, USA and 9HofstraNorthwell School of Medicine, Hempstead, NY, USA

Key words: antipsychotic; meta-analysis; metabolicabnormalities; topiramate; weight gain

Yu-Tao Xiang, Faculty of Health Sciences, University ofMacau, Avenida da Universidade, 3/F, Building E12,Taipa, Macau, China. E-mail: [email protected] orWei Zheng, Guangzhou Huiai Hospital, Guangzhou510370, China. E-mail: [email protected]

Accepted for publication July 28, 2016

385

Acta Psychiatr Scand 2016: 134: 385–398 © 2016 John Wiley & Sons A/S. Published by John Wiley & Sons LtdAll rights reservedDOI: 10.1111/acps.12631

ACTA PSYCHIATRICA SCANDINAVICA

Summations

• Combination strategies are widely used to enhance efficacy in treating schizophrenia and also aim atreducing adverse drug reactions.

• Topiramate’s efficacy for improving psychopathology and/or reducing body weight in schizophreniahas been explored with conflicting results.

• Based on the available data, adjunctive topiramate to antipsychotics is an effective and safe treatmentchoice for symptomatic improvement and weight reduction in patients with schizophrenia.

Considerations

• The sample size of included studies was relatively small and the information in most trials was lim-ited.

• The study duration was relatively short (2–24 weeks).

Introduction

Schizophrenia is a severe and frequently chronicpsychiatric illness with

studies, animal studies, meta-analyses, and system-atic reviews were excluded.

Outcomes

The coprimary outcome measures includedchange/endpoint difference in Positive and Nega-tive Syndrome Scale (PANSS)/Brief PsychiatricRating Scale (BPRS) total score and in bodyweight. Secondary outcomes included positive,negative and general psychopathology symptoms,body mass index (BMI), waist and hip circumfer-ence, waist-to-hip ratio, glucose and lipid parame-ters, all-cause and specific-cause discontinuation,and adverse effect frequencies. Whenever changescore and endpoint values of a continuous out-come were available, we preferred change data(unless skewed, i.e., SD >2 times the mean), asoften in smaller randomized trials the baseline val-ues differ biasing the analysis of endpoint results.

Selection of studies

PubMed, PsycINFO, Embase, Cochrane Librarydatabases, the Cochrane Controlled Trials Registerand Chinese databases (WanFang Database, Chi-nese Biomedical database, and China Journal Net)were searched electronically independently by twoauthors (WZ, XBL) identifying full-text papers forfurther detailed review. The search included allstudies published until June 2015, regardless oflanguage. The keywords are listed in Appendix S1.The electronic search was supplemented by usingthe ‘related article’ function. Bibliographies ofRCTs, meta-analyses, and systematic reviews weremanually searched for studies that were omitted inthe initial electronic search (44).

Data extraction

Two authors (WZ, XBL) independently extracteddata. Data presented only in graphs and figureswere extracted whenever possible, but includedonly if the two abstractors independently came tothe same conclusion. Authors were contacted toobtain missing information or for clarification. Incase of multicenter studies, relevant data from eachindividual study center were extracted separately(45).

Statistical methods

The meta-analysis was performed according to therecommendations of the Cochrane Collaborationusing RevMan (version 5.1.7.0) (http://www.cochrane.org). Intent-to-treat (ITT) data were

preferred over observed cases data. For continuousdata, standardized mean difference (SMD) �95%confidence intervals (CIs) was calculated for effi-cacy outcomes, while weighted mean difference(WMD) �95% CIs was used for body weight andBMI as well as metabolic outcomes due to theirhomogeneous scaling. For dichotomous data, riskratios (RRs) �95% CIs were computed. WhenRRs were significant, the number needed to treat(NNT) or number needed to harm (NNH) was cal-culated as the inverse of the risk difference. The I2

method assessed statistical heterogeneity (46). Arandom-effects model was used in all cases. Allanalyses were two-tailed, with alpha = 0.05.

In case of I2 ≥ 50% for the effect of total psy-chopathology, we conducted one sensitivity analy-sis removing the two studies (9, 47) with anoutlying effect size of less than �1.0 (i.e., morethan one standard deviation superiority of topira-mate). Furthermore, we conducted the following12 subgroup analyses in order to identify potentialmoderators or mediators of the two coprimaryoutcomes (total psychopathology and bodyweight) and two key secondary outcomes with clin-ical relevance (positive and negative symptoms) inorder to assess whether any of them could reduce/explain the observed heterogeneity of the findings:(i) double-blind vs. non-blinded studies; (ii)cotreatment with clozapine vs. non-clozapineantipsychotics; (iii) costarting with antipsychoticvs. augmenting antipsychotic; (iv) treatment dura-tion 150 mg since this dose thatcan be given with currently available tablets in clin-ical care fell in between the median (139 mg/day)and the mean dose (165 mg/day) in the meta-ana-lyzed studies; (x) illness phase, that is., first episodevs. chronic; (xi) Asian vs. non-Asian; and (xii) Chi-nese vs. non-Chinese studies. In order to assess theeffect of dose, we entered the two fixed dose topira-mate treatment arms from one study (27) sepa-rately, but assigned half of the total patientsrandomized to placebo to each topiramate arm inorder to not inflate the number of placebo-treatedpatients.

Additionally, using STATA (version 12.0)(http://www.stata.com), meta-regression analyseswere conducted for the two coprimary and two keysecondary outcomes to examine the moderatingeffect of the following: (i) patient age, (ii) percent

387

Adjunctive topiramate for schizophrenia

http://www.cochrane.orghttp://www.cochrane.orghttp://www.stata.com

males, (iii) treatment setting, and (iv) illness dura-tion, or the mediating effect of (v) trial duration or(vi) topiramate dosage.

Finally, publication bias was assessed using fun-nel plots, Egger’s intercept (48), trim-and-fill pro-cedure (49), and fail-safe method (50) usingcomprehensive meta-analysis (version 2) (http://www.meta-analysis.com).

Risk of bias assessment

Risk of bias was assessed with the Cochrane Riskof Bias tool, evaluating methods of randomsequence generation (selection bias), allocationconcealment (selection bias), blinding of partici-pants and personnel (performance bias), blindingof outcome assessment (detection bias), incompleteoutcome data (attrition bias), selective reporting(reporting bias), and other biases. Additionally,the methodological quality was assessed using theJadad scale (range = 1–5) (51). The grading of rec-ommendations assessment, development, and eval-uation (GRADE) system was also employed torate the quality of evidence and the strength of rec-ommendations of the meta-analysis (i.e., very low,low, moderate, or high) as recommended by theCochrane Collaboration (52, 53).

Results

Search results

The electronic search identified 368 potentially rel-evant publications; 172 in English and 196 in Chi-nese. After removing duplicates (N = 130),reviewing the titles or abstracts (N = 151) and fulltexts (N = 71), 16 RCTs (1, 9, 25–30, 47, 54–60)met the entry criteria for the meta-analysis (seeFig. S1).

Study characteristics

The 16 RCTs (Table S1) included 934 participants(range = 26–80, median = 60, topiramate: n =473, control group: n = 461) and lasted 11.8 � 5.6(range = 2–24, median = 12) weeks. Participantswere 34.8 � 6.7 (range = 16–65, median = 34.9)years old, and 55.6 � 23.2% (range = 0–100%,median = 53.0%) were male. Seven studies wereconducted in China (n = 452), 2 each in Iran(n = 112) and Korea (n = 126), and 1 each in India(n = 67), USA (n = 48), Finland (n = 26), Ger-many (n = 43), and Italy (n = 60). Altogether, 13RCTs (81.3%) reported PANSS/BPRS total scorechange/endpoint and 15 (93.8%) reported bodyweight data. Placebo was used as the comparator

in all but four open studies (25, 47, 57, 58) thatused continued antipsychotic monotherapy as thecomparator. The mean topiramate dose was164.9 � 70.4 mg/day (range = 50–300, median =139.0 mg/day).

Figure S1 summarizes the risk of bias assess-ment, with details provided in the Appendix S1section. The mean Jadad score of the 16 RCTs was3.6 � 0.8 (range = 2–5, median = 4) (Table S1).

The GRADE quality assessment revealed somelimitations in study design and consistency, butalso some strengths regarding large treatmenteffect, and no obvious indirectness or imprecisionin reporting of the results. Overall, the quality ofevidence ranged from ‘low’ (4%), via moderate(42%), to ‘high’ (54%) (Table S2).

Positive and Negative Syndrome Scale/Brief Psychiatric RatingScale total scores

Topiramate cotreatment outperformed the controlgroup regarding change/endpoint in PANSS/BPRS total score (trials = 12, n = 651, SMD:�0.58 (95% CI: �0.82, �0.35), P < 0.00001;I2 = 53%, Fig. 1a). Results for PANSS/BPRStotal score remained significant when two stronglyoutlying studies were removed (trials = 11,n = 539, SMD: �0.46 (95% CI: �0.67, �0.26),P < 0.00001; I2 = 25%).

Superiority of topiramate was confirmed in 23of the 26 analyzed subgroups (Table 1). Signifi-cance was diminished to a trend level only for stud-ies of mixed in- and out-patients (studies = 3,P = 0.10), studies with topiramate doses >150 mg/day (studies = 5, P = 0.08), and non-Asian (stud-ies = 3, P = 0.23) (Table 1). In exploratory meta-regression analyses, only shorter illness duration(P = 0.047) was significantly related to greaterPANSS/BPRS total change, but other variableswere not (trial duration: P = 0.93, age: P = 0.83,sex: P = 0.16, topiramate dosage: P = 0.44, in-patient status: P = 0.89).

The funnel plots of the studies were symmetricaland Egger’s test did not support publication bias(P = 0.87). The fail-safe method indicated that 162additional studies would be required to lead to anull finding.

Positive and negative symptoms and general psychopathology

Topiramate cotreatment outperformed the controlgroup regarding PANSS positive symptoms (trials= 8, n = 436, SMD: �0.37 (95% CI: �0.61,�0.14), P = 0.002; I2 = 32%, Fig. 1b), negativesymptoms (trials = 8, n = 436, SMD: �0.58 (95%CI: �0.87, �0.29), P < 0.0001; I2 = 54%, Fig. 1c),

388

Zheng et al.

http://www.meta-analysis.comhttp://www.meta-analysis.com

and general psychopathology (trials = 6, n = 345,SMD: �0.68 (95% CI: �0.95, �0.40), P < 0.00001;I2 = 34%, Table 2).

Superiority of topiramate for positive symptomswas confirmed in 15 of the 26 subgroups (with anadditional 5 subgroups having a P = 0.05–0.06)(Table 1). Significance was diminished to a trend-level finding in open studies (studies = 2,P = 0.06), augmentation studies (studies = 5,P = 0.05), those with female predominance

(≥60%) (study = 1, P = 0.05), lower quality (Jadadscore ≤3, studies = 2, P = 0.06), and multi-episodepatients (studies = 5, P = 0.05). Significance wasfurther lost in studies lasting 150 mg/day (stud-ies = 4, P = 0.15), studies without male/femalepredominance (studies = 3, P = 0.31), and non-Asian (studies = 2, P = 0.30) (Table 1). Inexploratory meta-regression analyses, none of the

(a)

(b)

(c)

Fig. 1. (a) Adjunctive topiramate vs. placebo for schizophrenia: Forest plot for Standardized Mean Difference (SMD) in Positiveand Negative Syndrome Scale (PANSS)/Brief Psychiatric Rating Scale (BPRS) Total Score (Change or Endpoint). (b) Adjunctivetopiramate vs. placebo for schizophrenia: Forest plot for Standardized Mean Difference (SMD) in Positive and Negative SyndromeScale (PANSS) positive symptoms. (c) Adjunctive topiramate vs. placebo for schizophrenia: Forest plot for Standardized Mean Dif-ference (SMD) in Positive and Negative Syndrome Scale (PANSS) negative symptoms

389


Table 1. Subgroup analysis and sensitivity analysis of the effect of moderator or mediator variables on PANSS or BPRS total score as well as positive and negative symptoms

Variables Subjects (studies) SMDs (95% CI) I2 (%) P-value for each subgroup P-value across subgroups

PANSS or BPRS total score (Change or Endpoint)1. Double-blind/rater masked 460 (10) �0.50 (�0.78, �0.21) 53 0.0006 0.09

Non-blinded studies 191 (3) �0.85 (�1.15, �0.55) 1

examined variables was significantly related to pos-itive symptom change (trial duration: P = 0.95,age: P = 0.44, sex: P = 0.99, topiramate dosage:P = 0.26, in-patient status: P = 0.25, illness dura-tion: P = 0.11).

Similarly, superiority of topiramate for negativesymptoms was confirmed in 21 of the 26 subgroups

(Table 1). Significance was lost only for open stud-ies (studies = 2, P = 0.11), studies with trial dura-tion 150 mg/day (studies = 6,P = 0.05), and non-Asian (studies = 4, P = 0.13)(Table 3). ‘Preventive’ topiramate cotreatment,beginning with antipsychotic initiation, as opposedto augmentation after significant weight gain hadoccurred, yielded larger effects (co-initiation: tri-als = 352, WMD: �3.48 kg, 95% CI: �5.41,�1.55 vs. augmentation: trials = 9, n = 361,WMD: �1.50, 95% CI: �2.71, �0.82; subgroupcomparison P < 0.001). In exploratory meta-regression analyses, none of the examined

Table 1. (Continued)

Variables Subjects (studies) SMDs (95% CI) I2 (%) P-value for each subgroup P-value across subgroups

5. Male predominance (≥60%)* 222 (4) �0.70 (�1.12, �0.27) 56 0.001 0.27Female predominance (≥60%) 53 (1) �0.14 (�0.68, 0.40) N/A 0.61No sex predominance 161 (3) �0.59 (�1.14, �0.04) 65 0.04

6. Jadad score ≥4 225 (6) �0.66 (�1.03, �0.29) 60 0.0004 0.36Jadad score ≤3 111 (2) �0.38 (�0.85, 0.09) 36 0.11

7. In-patients 208 (4) �0.59 (�1.09, �0.09) 67 0.02 0.20Out-patients 55 (1) �1.07 (�1.64, �0.50) N/A 0.0002In- and out-patients 173 (3) �0.46 (�0.80, �0.12) 17 0.008

8. Last observation carried forward 270 (5) �0.58 (�0.97, �0.18) 58 0.004 0.93Observed cases 166 (3) �0.60 (�1.12, �0.09) 63 0.02

9. Topiramate dosage >150 mg/day 229 (4) �0.43 (�0.76, �0.10) 32 0.01 0.29Topiramate dosage ≤150 mg/day 207 (4) �0.75 (�1.25, �0.26) 66 0.003

10. Illness phase: first episode 175 (3) �0.60 (�1.11, �0.09) 63 0.02 0.95Illness phase: multi-episode 261 (5) �0.58 (�0.98, �0.18) 58 0.005

11. Asian 345 (6) �0.62 (�0.96, �0.29) 57 0.0003 0.71Non-Asian 91 (2) �0.46 (�1.25, 0.33) 70 0.26

12. Chinese studies 166 (3) �0.60 (�1.12, �0.09) 63 0.02 0.93Non-Chinese studies 270 (5) �0.58 (�0.97, �0.18) 58 0.004

BPRS, Brief Psychiatric Rating Scale; N/A, Not applicable; PANSS, Positive and Negative Syndrome Scale; SMDs, standardized mean differences. Bolded P-values: P < 0.05.*Only 12 RCTs reported data on sex.

Table 2. Adjunctive topiramate for schizophrenia: Secondary outcomes

Secondaryoutcomes N* (RCTs) SMD (95% CI) I2 (%) P-value

PANSS generalsymptom subscore

345 (6) �0.68 (�0.95, �0.40) 34

variables was significantly related to body weightchange (trial duration: P = 0.39, age: P = 0.09,sex: P = 0.75, topiramate dosage: P = 0.09, in-patient status: P = 0.44).

The funnel plots of the studies were symmetricalfor body weight and Egger’s test did not supportpublication bias (P = 0.33). The fail-safe methodindicated that 303 additional studies would berequired to lead to a null finding.

In addition to body weight, topiramate was alsoassociated with significantly lower BMI (trials = 9,n = 407, WMD:–1.77, 95% CI: �2.38, �1.15,P < 0.00001; I2 = 74%, Fig. 2). Conversely, therewas no significant treatment group difference inwaist circumference (trials = 3, n = 161, P = 0.12),hip circumference (trials = 2, n = 115, P = 0.27),and waist-to-hip ratio (trials = 2, n = 156,P = 0.11) (Table 2).

Serum lipids

Topiramate was superior to placebo/antipsychoticmonotherapy regarding triglycerides (trials = 4,

n = 233, P = 0.006), but not total cholesterol (tri-als = 4, n = 233, P = 0.08), high-density lipopro-tein (HDL) cholesterol (trials = 4, n = 233,P = 0.12), and low-density lipoprotein (LDL)cholesterol (trials = 4, n = 233, P = 0.17)(Table 2).

Carbohydrate metabolism and other biochemical variables

Topiramate was significantly superior to placebo/antipsychotic monotherapy regarding fasting insu-lin (trials = 2, n = 122, P < 0.00001), but not fast-ing glucose (trials = 4, n = 233, P = 0.14) or leptin(trials = 3, n = 169, P = 0.86) (Table 2).

Treatment discontinuation

All-cause discontinuation was similar betweentopiramate and control groups (trials = 14,n = 792, RR: 1.28 (95% CI: 0.91, 1.81), P = 0.16,I2 = 0%; Fig. S3). Similarly, topiramate did notdiffer significantly from the control conditionregarding discontinuation due to intolerability

Fig. 2. Adjunctive topiramate vs. placebo for schizophrenia: Forest plot for Weighted Mean Difference (WMD) in Body Weightand Body Mass Index (Change or Endpoint)

392

Zheng et al.

(trials = 4, n = 239, P = 0.77) or inefficacy (tri-als = 3, n = 175, P = 0.67).

Antipsychotic-induced drug reactions

Topiramate was associated with significantly moreADRs (Fig. S4) than the control condition, includ-ing concentration/attention difficulties (P = 0.03,NNH = 8, 95% CI = 4–25), psychomotor slowing(P = 0.02, NNH = 7, 95% CI = 4–25), and pares-thesia (P = 0.05, NNH = 2, 95% CI = 4–33).Conversely, topiramate was associated with signifi-cantly less ADRs (Fig. S3) regarding sponta-neously reported constipation (P = 0.04,NNH = 9, 95% CI = 5–100), increased appetite(P = 0.005, NNH = 4, 95% CI = 3–9), weightgain (P = 0.002, NNH = 2, 95% CI = 1–2), andmeasured weight gain ≥7% (P = 0.0001,NNH = 2, 95% CI = 2–3). Somnolence, drymouth, nausea/vomiting, fatigue, headache, mem-ory difficulties, insomnia, asthenia, and sedation(Fig. S4) were not significantly different betweentopiramate and the control groups (P = 0.08 to0.95). Only one study (28) formally assessed cogni-tive effects, finding either no difference betweentopiramate and control (phonemic and semanticfluency, Wisconsin card sorting errors, and

categories), or an advantage of topiramate (Strooptest).

Discussion

This meta-analysis of 16 RCTs and 934 patientswith schizophrenia-spectrum disorders suggeststhat addition of topiramate to antipsychotictreatment, either as augmentation or co-initiated, was significantly superior to antipsy-chotic monotherapy regarding total, positive,negative, and general psychopathology symp-toms, as well as lower body weight, BMI,serum triglycerides, and fasting insulin. More-over, all-cause discontinuation was similar tothe control condition. However, topiramatecotreatment was associated with a greater fre-quency of concentration/attention difficulties (8patients on topiramate (13%) vs. 0% on pla-cebo), psychomotor slowing (25 patients ontopiramate (26%) vs. 13% on placebo), andparesthesia (27 patients on topiramate (28%) vs.10% on placebo). Conversely, however, topira-mate reduced the risk of spontaneously reportedconstipation and increased appetite, weight gain,and weight gain ≥7%. Notably, despite hetero-geneity of the beneficial effects on

Table 3. Subgroup analysis and Sensitivity analysis of the Effect of Mediator Variables on the Outcome ‘Body weight (Change or Endpoint)’

Variables Subjects (studies) WMDs (95% CI) I2 (%)P-value for each

subgroupP-value acrosssubgroups

1. Double-blind/rater masked 542 (12) �2.58 (�4.16, �1.01) 86 0.001 0.54Non-blinded studies 171 (3) �3.84 (�7.54, �0.14) 73 0.04

2. Prevention 292 (5) �4.11 (�6.70, �1.52) 95 0.002 0.05Intervention 421 (10) �1.41 (�2.23, �0.59) 0 0.0008

3. Cotreatment with clozapine 159 (4) �1.58 (�3.93, 0.67) 20 0.19 0.32Cotreatment with non-clozapine antipsychotics 554 (11) �2.98 (�4.45, �1.51) 88

psychopathology and body weight, results wereconsistent across the vast majority of the exam-ined subgroups and moderator/mediator vari-ables, indicating that the heterogeneity occurredpredominantly within the effective range andnot around a null finding.

Potential mechanisms of topiramate’s symp-tomatic efficacy include the following: (i) decreas-ing the presynaptic release of glutamate andexcessive glutamate neurotransmission via kai-nate receptors, resulting in a blunting of thehyperdopaminergia and prolonging stimulationof AMPA/KA receptors (1, 10, 61, 62); and (ii)inhibition of glutamate activity, attenuatingactivity at Na+ and Ca2+ channels, and aug-menting effects of the GABA-A receptors (55,63). Potential mechanisms for improvement inbody weight and metabolic parameters includethe following: (i) attenuation of appetite via (64),(ii) insulin-sensitizing effects resulting in thedirect enhancement of insulin action and leadingto lower glucose and insulin levels (64), and (iii)stimulation of the lipoprotein lipase in the adi-pose tissue and skeletal muscle, inhibiting car-bonic anhydrase enzymes involved in severalsteps of de novo lipogenesis in the mitochondriaand cell cytosol (65, 66).

Study strengths

This is by far the largest and most comprehen-sive meta-analysis of topiramate-antipsychoticcotreatment in patients with schizophrenia-spec-trum disorders, surpassing the most recent lar-gest meta-analysis of only 8 studies and 422subjects (7) by ≥100%. With 934 patients in ourmeta-analysis, we are approaching the n ≥ 1000that has been postulated to yield reliable andreproducible meta-analytic findings (67). Thenumber of studies and subjects allowed us toobtain robust effect size estimates and conduct aseries of previously unavailable subgroup andmeta-regression analyses. Moreover, ourenhanced power enabled us to provide a morefine grained assessment of increased anddecreased ADRs with topiramate cotreatmentand to show that, different from prior results(7), efficacy and weight gain benefits generalizedacross topiramate cotreatment of clozapine andnon-clozapine antipsychotics.

Study limitations

Several limitations of this meta-analysis need to beacknowledged. First, although it included 16 RCTsconducted in single centers, the sample size of each

study was relatively small and the information inmost trials was limited or incomplete, which mayreduce full generalizability of the findings. Second,the study duration was generally short(mean = 11.8 � 5.6, range = 2–24 weeks). Whileeffect sizes regarding efficacy and body weight andBMI were moderate, it is unclear to what degreeeffects plateau, might increase or regress to themean upon longer follow-up. Thus, effects beyond24 weeks require further study. Furthermore, qual-ity of RCTs varied and weight loss with topiramatecould have led to functional unblinding. Alto-gether, 31.3% of RCTs were of lower quality andonly 6% had detailed descriptions in preregisteredprotocols. Third, there was significant heterogene-ity of the primary outcomes suggesting the effect ofrelevant moderator and mediator variables. Theobserved heterogeneity may be because meta-ana-lyses combine results from trials that differ in theirmethodology and clinical samples. For these rea-sons, a random-effects model was employedthroughout to provide a conservative estimate oftreatment efficacy. Moreover, a series of subgroupand meta-regression analyses were run, including ameasure of study quality, showing uniformity ofthe results, which increases the confidence in theobserved findings. Furthermore, Egger’s test didnot reveal any obvious publication bias. Fourth,studies did not examine the relationship betweentopiramate doses and psychopathology, metabolicbenefits, or ADRs, which definitely warrants fur-ther evaluation. Fifth, only one small RCT (28)assessed the potential for adverse cognitive effectsof topiramate. This is a relevant omission, asresearch (68, 69) indicates that topiramate canimpair cognitive functioning, which clearly needsto be examined further.

Comparison with other studies

By adding more evidence, our review found a moreoptimistic view of topiramate augmentation inattenuating psychotic symptoms than prior meta-analyses (7, 10, 33) in that effect size estimates weremore robust with much more narrow 95% CIsthroughout and in that effect sizes generalized toclozapine and non-clozapine antipsychotics as wellas across a number of other relevant design,patient, and illness variables. Furthermore, ourmeta-analysis confirmed the benefits of topiramateaugmentation for weight reduction/slowing ofweight gain, consistent with the prior reviews andmeta-analyses (34–41). However, the prior reviewsand meta-analyses had been limited by the muchsmaller number of studies, fewer participants, andby the lack of inclusion of all the metabolic

394

Zheng et al.

variables, ADRs, and non-clozapine antipsy-chotics, as well as by a lack of quality/risk of biasassessment, and relevant subgroup and meta-regression analyses. Moreover, our meta-analysisshowed an advantage when topiramate was givenpreventively, that is, starting topiramate togetherwith the antipsychotic, compared to adding topira-mate after relevant weight gain had alreadyoccurred (�3.5 vs. �1.5 kg). Future studies shouldfurther pursue this finding in order to establish thebest time point when considering topiramate aug-mentation, be it at baseline, after 1–4 weeks thatcould help identify patients at greatest risk forearly and subsequent weight gain, or after a signifi-cant threshold of weight gain has passed. The sameexamination should be performed for the timing ofthe amelioration of antipsychotic metabolic side-effects.

Comparison with other drugs

Together with topiramate, metformin is the best-studied medication for antipsychotic-related car-diometabolic adverse effects. Across a number ofreviews and meta-analyses of up to 21 RCTs (8,35, 36, 38, 40, 70–90), metformin has also beenshown to effectively prevent or reduce weight gainin subjects receiving antipsychotics, with a magni-tude similar to that of topiramate, that is, approxi-mately 2.5–3 kg less weight gain compared to thecontrol condition over a pooled duration of about3 months. Across these studies, metformin wasassociated with a low ADR risk, except for gas-trointestinal side-effects that are often mild andtransient. Similarly, topiramate augmentation wasalso well tolerated and not associated with greaterdiscontinuation due to intolerability than the con-trol condition. To date, no head-to-head trialshave been published to directly compare topira-mate and metformin in antipsychotic-treatedpatients. Moreover, as many patients gain moreweight than 3 kg, the possibility of combining met-formin and topiramate that have different mecha-nisms for reducing antipsychotic-related bodyweight should be explored for patients in whommetformin or topiramate monotherapy only insuf-ficiently improved cardiometabolic health (62).

To conclude, this meta-analysis of 16 RCTsindicates that topiramate is a useful agent tocombine with antipsychotics in patients withschizophrenia to improve psychopathology andprevent/reduce weight gain, triglycerides, and insu-lin resistance. The recommended doses of adjunc-tive topiramate may be 150 mg/day or lower, butthe dose effects require further evaluation. Whenconsidering adjunctive topiramate treatment in

patients taking antipsychotics, monitoring of con-centration/attention difficulties, psychomotorslowing, and paresthesia is recommended.Although current data suggest topiramate as a rel-atively safe and effective adjunctive treatment withantipsychotics for schizophrenia, future RCTsshould examine illness phase-specific efficacy andlong-term maintenance effects of topiramate onpsychopathology, body weight, and metabolicparameters. Studies should examine topiramate’sdose–response relationship, formally assess rele-vant effects on cognition, and consider comparingthe efficacy and safety of single and combinedadjunctive treatment of topiramate and metforminin patients who failed behavioural interventionsfor antipsychotic-induced cardiometabolic adverseeffects.

Acknowledgements

The authors thank Jari Tiihonen, MD, for providing unpub-lished data for this meta-analysis.

Declaration of interest

Dr. Correll has been a consultant and/or advisor to or hasreceived honoraria from Alkermes, Forum, Gerson LehrmanGroup, IntraCellular Therapies, Janssen/J&J, Lundbeck,Medavante, Medscape, Otsuka, Pfizer, ProPhase, Sunovion,Supernus, Takeda, and Teva. He has provided expert testi-mony for Bristol-Myers Squibb, Janssen, and Otsuka. Heserved on a Data Safety Monitoring Board for Lundbeck andPfizer. He received grant support from Takeda.

The other authors declare no conflict of interests concerningthis article.

References

1. Tiihonen J, Halonen P, Wahlbeck K et al. Topiramate add-on in treatment-resistant schizophrenia: a randomized,double-blind, placebo-controlled, crossover trial. J ClinPsychiatry 2005;66:1012–1015.

2. Kane JM, Correll CU. Past and present progress in thepharmacologic treatment of schizophrenia. J Clin Psychia-try 2010;71:1115–1124.

3. Carbon M, Correll CU. Clinical predictors of therapeuticresponse to antipsychotics in schizophrenia. DialoguesClin Neurosci 2014;16:505–524.

4. Muscatello MR, Bruno A, De Fazio P, Segura-Garcia C,Pandolfo G, Zoccali R. Augmentation strategies in partialresponder and/or treatment-resistant schizophreniapatients treated with clozapine. Expert Opin Pharma-cother 2014;15:2329–2345.

5. Wolff-Menzler C, Hasan A, Malchow B, Falkai P,Wobrock T. Combination therapy in the treatment ofschizophrenia. Pharmacopsychiatry 2010;43:122–129.

6. Nelson JC. Combined treatment strategies in psychiatry.J Clin Psychiatry 1993;54(Suppl):42–49; discussion 55-46.

7. Correll CU, Maayan L, Kane JM, De Hert M, Cohen D.Efficacy for psychopathology and body weight and safetyof topiramate-antipsychotic cotreatment in patients withschizophrenia-spectrum disorders: results from a meta-

395


analysis of randomized controlled trials. J Clin Psychiatry2016;77:e746–e756.

8. Zheng W, Li XB, Tang YL, Xiang YQ, Wang CY, de LeonJ. Metformin for weight gain and metabolic abnormalitiesassociated with antipsychotic treatment: meta-analysis ofrandomized placebo-controlled trials. J Clin Psychophar-macol 2015;35:499–509.

9. Afshar H, Roohafza H, Mousavi G et al. Topiramate add-on treatment in schizophrenia: a randomised, double-blind, placebo-controlled clinical trial. J Psychopharmacol2009;23:157–162.

10. Veerman SR, Schulte PF, Begemann MJ, Engelsbel F, deHaan L. Clozapine augmented with glutamate modulatorsin refractory schizophrenia: a review and metaanalysis.Pharmacopsychiatry 2014;47:185–194.

11. Dewan VK. Re: weight change with antipsychotic use. CanJ Psychiatry 2001;46:660.

12. Dursun SM, Devarajan S. Clozapine weight gain, plus topi-ramate weight loss. Can J Psychiatry 2000;45:198.

13. Karslioglu EH, Karakilic H, Taner E, Cosar B. Topira-mate-induced psychotic exacerbation: case report andreview of literature. Int J Psychiatry Clin Pract2007;11:285–290.

14. Levy E, Margolese HC, Chouinard G. Topiramate pro-duced weight loss following olanzapine-induced weightgain in schizophrenia. J Clin Psychiatry 2002;63:1045.

15. Navarro V, Pons A, Romero A, Bernardo M. Topiramatefor clozapine-induced seizures. A J Psychiatry2001;158:968–969.

16. Schuler-Springorum M, Hebebrand J, Wehmeier PM, Rem-schmidt H. [Is topiramate effective for weight loss in neu-roleptic-induced obesity? 2 case reports]. Z KinderJugendpsychiatr Psychother 2002;30:51–58.

17. Wang CY. Topiramate-induced weight loss in schizophre-nia: a case report (in Chinese). SiChuan Mental Health2011;24:11.

18. Dursun SM, Deakin JF. Augmenting antipsychotic treat-ment with lamotrigine or topiramate in patients with treat-ment-resistant schizophrenia: a naturalistic case-seriesoutcome study. J Psychopharmacol 2001;15:297–301.

19. Levy E, Agbokou C, Ferreri F, Chouinard G, MargoleseHC. Topiramate-induced weight loss in schizophrenia: aretrospective case series study. Can J Clin Pharmacol2007;14:e234–e239.

20. Roy Chengappa KN, Levine J, Rathore D, Parepally H,Atzert R. Long-term effects of topiramate on bipolarmood instability, weight change and glycemic control: acase-series. Eur Psychiatry 2001;16:186–190.

21. Cates ME, Feldman JM, Boggs AA, Woolley TW, WhaleyNP. Efficacy of add-on topiramate therapy in psychiatricpatients with weight gain. Annals Pharmacotherapy2008;42:505–510.

22. Chengappa KN, Chalasani L, Brar JS, Parepally H, HouckP, Levine J. Changes in body weight and body mass indexamong psychiatric patients receiving lithium, valproate, ortopiramate: an open-label, nonrandomized chart review.Clin Ther 2002;24:1576–1584.

23. Tramontina S, Zeni CP, Pheula G, Rohde LA. Topiramatein adolescents with juvenile bipolar disorder presentingweight gain due to atypical antipsychotics or mood stabi-lizers: an open clinical trial. J Child Adolesc Psychophar-macol 2007;17:129–134.

24. Vieta E, Sanchez-Moreno J, Goikolea JM et al. Effects onweight and outcome of long-term olanzapine-topiramatecombination treatment in bipolar disorder. J Clin Psy-chopharmacol 2004;24:374–378.

25. Kim JH, Yim SJ, Nam JH. A 12-week, randomized, open-label, parallel-group trial of topiramate in limiting weightgain during olanzapine treatment in patients withschizophrenia. Schizophr Res 2006;82:115–117.

26. Jiang J. The control study of topiramate treatment inweight gain caused by olanzapine (in Chinese). ChinaModern Doctor 2013;51:78–80.

27. Ko YH, Joe SH, Jung IK, Kim SH. Topiramate as an adju-vant treatment with atypical antipsychotics in schizophre-nic patients experiencing weight gain. Clin Neurophar-macol 2005;28:169–175.

28. Muscatello MR, Bruno A, Pandolfo G et al. Topiramateaugmentation of clozapine in schizophrenia: a double-blind, placebo-controlled study. J Psychopharmacol2011;25:667–674.

29. Narula PK, Rehan HS, Unni KE, Gupta N. Topiramate forprevention of olanzapine associated weight gain andmetabolic dysfunction in schizophrenia: a double-blind,placebo-controlled trial. Schizophr Res 2010;118:218–223.

30. Nickel MK, Nickel C, Muehlbacher M et al. Influence oftopiramate on olanzapine-related adiposity in women: arandom, double-blind, placebo-controlled study. J ClinPsychopharmacol 2005;25:211–217.

31. Rezaei V, Mohammadi MR, Ghanizadeh A et al. Double-blind, placebo-controlled trial of risperidone plustopiramate in children with autistic disorder. Prog Neu-ropsychopharmacol Biol Psychiatry 2010;34:1269–1272.

32. Wozniak J, Mick E, Waxmonsky J, Kotarski M, Hantsoo L,Biederman J. Comparison of open-label, 8-week trialsof olanzapine monotherapy and topiramate augmentationof olanzapine for the treatment of pediatric bipolar disorder.J Child Adolesc Psychopharmacol 2009;19:539–545.

33. Cohen D, Correll CU. Second-generation antipsychotic-associated diabetes mellitus and diabetic ketoacidosis:mechanisms, predictors, and screening need. J Clin Psychi-atry 2009;70:765–766.

34. Choi YJ. Efficacy of adjunctive treatments added to olan-zapine or clozapine for weight control in patients withschizophrenia: a systematic review and meta-analysis. SciWorld J 2015;2015:970730.

35. Ellinger LK, Ipema HJ, Stachnik JM. Efficacy of met-formin and topiramate in prevention and treatment of sec-ond-generation antipsychotic-induced weight gain. AnnPharmacother 2010;44:668–679.

36. Gierisch JM, Nieuwsma JA, Bradford DW et al. Pharmaco-logic and behavioral interventions to improve cardiovas-cular risk factors in adults with serious mental illness: asystematic review and meta-analysis. J Clin Psychiatry2014;75:e424–e440.

37. Greg Deardorff O, Syed A, Ames CJ, Yaeger JS. Rani-tidine, metformin, and topiramate: managing weight gainin a clozapine-treated patient with schizoaffective disorder.Prim Care Companion CNS Disord 2014;16, doi: 10.4088/PCC.13l01598.

38. Maayan L, Vakhrusheva J, Correll CU. Effectiveness ofmedications used to attenuate antipsychotic-related weightgain and metabolic abnormalities: a systematic review andmeta-analysis. Neuropsychopharmacology 2010;35:1520–1530.

39. Mahmood S, Booker I, Huang J, Coleman CI. Effect of topi-ramate on weight gain in patients receiving atypicalantipsychotic agents. J Clin Psychopharmacol 2013;33:90–94.

40. Mizuno Y, Suzuki T, Nakagawa A et al. Pharmacologicalstrategies to counteract antipsychotic-induced weight gainand metabolic adverse effects in schizophrenia: a

396

Zheng et al.

http://dx.doi.org/10.4088/PCC.13l01598http://dx.doi.org/10.4088/PCC.13l01598

systematic review and meta-analysis. Schizophr Bull2014;40:1385–1403.

41. Whitney Z, Procyshyn RM, Fredrikson DH, Barr AM.Treatment of clozapine-associated weight gain: a system-atic review. Eur J Clin Pharmacol 2015;71:389–401.

42. Furlan AD, Pennick V, Bombardier C, van Tulder M. 2009updated method guidelines for systematic reviews in theCochrane Back Review Group. Spine 2009;34:1929–1941.

43. Moher D, Liberati A, Tetzlaff J, Altman DG. Preferredreporting items for systematic reviews and meta-analyses:the PRISMA statement. Ann Intern Med 2009;151:264–269, W264.

44. Varadhan KK, Neal KR, Lobo DN. Safety and efficacy ofantibiotics compared with appendicectomy for treatmentof uncomplicated acute appendicitis: meta-analysis of ran-domised controlled trials. BMJ 2012;344:e2156.

45. Higgins JP, Green S. Cochrane handbook for systematicreviews of interventions. UK: Wiley Online Library, 2008

46. Higgins JP, Thompson SG. Quantifying heterogeneity in ameta-analysis. Stat Med 2002;21:1539–1558.

47. Weng DL. The study of topiramate combined with olanza-pine in acute schizophrenia (in Chinese). Med InnovationChina 2011;08:25–26.

48. Egger M, Davey Smith G, Schneider M, Minder C. Bias inmeta-analysis detected by a simple, graphical test. BMJ1997;315:629–634.

49. Duval S, Tweedie R. Trim and fill: a simple funnel-plot–based method of testing and adjusting for publication biasin meta-analysis. Biometrics 2000;56:455–463(459).

50. Rosenthal R. The file drawer problem and tolerance fornull results. Psychol Bull 1979;86:638–641.

51. Jadad AR, Moore RA, Carroll D et al. Assessing the qual-ity of reports of randomized clinical trials: is blinding nec-essary? Control Clin Trials 1996;17:1–12.

52. Balshem H, Helfand M, Schunemann HJ et al. GRADEguidelines: 3. Rating the quality of evidence. J Clin Epi-demiol 2011;64:401–406.

53. Atkins D, Best D, Briss PA et al. Grading quality of evi-dence and strength of recommendations. BMJ2004;328:1490.

54. Behdani F, Hebrani P, Rezaei Ardani A, Rafee E. Effect oftopiramate augmentation in chronic schizophrenia: a pla-cebo-controlled trial. Arch Iran Med 2011;14:270–275.

55. Chen ZX, Xia ML, Xia WJ, Huang HG, Jin L. Topiramatefor prevention of olanzapine associated weight gain (inChinese). Modern Pract Med 2012;24:979–980.

56. Chengappa K, Kupfer DJ, Parepally H et al. A placebo-controlled, random-assignment, parallel-group pilot studyof adjunctive topiramate for patients with schizoaffectivedisorder, bipolar type. Bipolar Disord 2007;9:609–617.

57. Li B, Liu ZL, Zhou HS et al. Efficacy and safety of topira-mate combined with clozapine in the treatment ofschizophrenia (in Chinese). China J Health Psychol2012;20:964–965.

58. Liu JJ. The efficacy of topiramate in the treatment ofantipsychotic-induced weight gain and lipid disorders (inChinese). Modern Pract Med 2011;23:665–667.

59. Yang B, Zhang YT, Ye LX. Effect of treatment with topi-ramate on risperidone-induced weight mass variation andplasma leptin levels (in Chinese). Chinese J Nervous Men-tal Dis 2013;39:331–334.

60. Yang Y, Yang DL. Effect of treatment with topiramate onolanzpine-induced weight gain (in Chinese). ZheJiang ClinMed J 2011;13:1263–1265.

61. Deutsch SI, Rosse RB, Schwartz BL, Mastropaolo J. Arevised excitotoxic hypothesis of schizophrenia: therapeu-tic implications. Clin Neuropharmacol 2001;24:43–49.

62. Hahn MK, Cohn T, Teo C, Remington G. Topiramate inschizophrenia: a review of effects on psychopathology andmetabolic parameters. Clin Schizophr Relat Psychoses2013;6:186–196.

63. Shank RP, Gardocki JF, Streeter AJ, Maryanoff BE. Anoverview of the preclinical aspects of topiramate: pharma-cology, pharmacokinetics, and mechanism of action.Epilepsia 2000;41(Suppl 1):S3–S9.

64. Richard D, Ferland J, Lalonde J, Samson P, Deshaies Y.Influence of topiramate in the regulation of energy bal-ance. Nutrition 2000;16:961–966.

65. Astrup A, Caterson I, Zelissen P et al. Topiramate: long-term maintenance of weight loss induced by a low-caloriediet in obese subjects. Obes Res 2004;12:1658–1669.

66. Schneiderhan ME, Marvin R. Is acetazolamide similar totopiramate for reversal of antipsychotic-induced weightgain? Am J Ther 2007;14:581–584.

67. Trikalinos TA, Churchill R, Ferri M et al. Effect sizes incumulative meta-analyses of mental health randomized tri-als evolved over time. J Clin Epidemiol 2004;57:1124–1130.

68. Dag E, Dag ZO, Baydas G et al. Effects of lamotrigine andtopiramate on brain maturation and cognitive functions inoffspring of pregnant rats – preliminary study. Adv ClinExp Med 2014;23:691–698.

69. Rass O, Umbricht A, Bigelow GE, Strain EC, Johnson MW,Mintzer MZ. Topiramate impairs cognitive function inmethadone-maintained individuals with concurrentcocaine dependence. Psychol Addict Behav 2015;29:237–246.

70. Annamalai A, Tek C. An overview of diabetes managementin schizophrenia patients: office based strategies for pri-mary care practitioners and endocrinologists. Int J Endo-crinol 2015;2015:969182.

71. Baptista T, Kin NM, Beaulieu S, de Baptista EA. Obesityand related metabolic abnormalities during antipsychoticdrug administration: mechanisms, management andresearch perspectives. Pharmacopsychiatry 2002;35:205–219.

72. Baptista T, Zarate J, Joober R et al. Drug induced weightgain, an impediment to successful pharmacotherapy:focus on antipsychotics. Curr Drug Targets 2004;5:279–299.

73. Birt J. Management of weight gain associated withantipsychotics. Ann Clin Psychiatry 2003;15:49–58.

74. Bushe CJ, Bradley AJ, Doshi S, Karagianis J. Changes inweight and metabolic parameters during treatmentwith antipsychotics and metformin: do the data inform asto potential guideline development? A systematicreview of clinical studies. Int J Clin Pract 2009;63:1743–1761.

75. Citrome L, Vreeland B. Schizophrenia, obesity, andantipsychotic medications: what can we do? Postgrad Med2008;120:18–33.

76. Das C, Mendez G, Jagasia S, Labbate LA. Second-genera-tion antipsychotic use in schizophrenia and associatedweight gain: a critical review and meta-analysis of behav-ioral and pharmacologic treatments. Ann Clin Psychiatry2012;24:225–239.

77. Gierisch JM, Nieuwsma JA, Bradford DW et al. Interven-tions to improve cardiovascular risk factors in people withserious mental illness. Rockville, MD: Agency for Health-care Research and Quality, 2013.

78. Hasnain M, Fredrickson SK, Vieweg WV. Metformin forobesity and glucose dysregulation in patients withschizophrenia receiving antipsychotic drugs. J Psy-chopharmacol 2011;25:715–721.

397


79. Himmerich H, Schuld A, Pollmacher T. Weight gain duringtreatment with antipsychotics: clinical relevance, patho-physiology, and therapeutical strategies. Psychiatr Prax2004;31(Suppl 2):S233–S237.

80. Housel AK, Waterbury N, Argo TR. Can metformin orrosiglitazone reduce metabolic side effects associated withatypical antipsychotics? Issues Ment Health Nurs2009;30:803–805.

81. Khan AY, Macaluso M, McHale RJ, Dahmen MM, GirrensK, Ali F. The adjunctive use of metformin to treat or pre-vent atypical antipsychotic-induced weight gain: a review.J Psychiatr Pract 2010;16:289–296.

82. Martinez-Ortega JM, Funes-Godoy S, Diaz-Atienza F,Gutierrez-Rojas L, Perez-Costillas L, Gurpegui M. Weightgain and increase of body mass index among children andadolescents treated with antipsychotics: a critical review.Eur Child Adolesc Psychiatry 2013;22:457–479.

83. Mizuno Y, Suzuki T, Nakagawa A et al. Pharmacologicalstrategies to counteract antipsychotic-induced weight gainand metabolic adverse effects in schizophrenia: a system-atic review and meta-analysis. Schizophr Bull2014b;40:1385–1403.

84. Praharaj SK, Jana AK, Goyal N, Sinha VK. Metforminfor olanzapine-induced weight gain: a systematic reviewand meta-analysis. Br J Clin Pharmacol 2011;71:377–382.

85. Klein DJ, Saldana SN. Review: metformin is effective atattenuating the metabolic risks associated with use ofatypical antipsychotics. Evid Based Ment Health2011;14:55.

86. Lee YJ, Jeong JH. A systematic review of metformin tolimit weight-gain with atypical antipsychotics. J ClinPharm Ther 2011;36:537–545.

87. McIntyre RS, Alsuwaidan M, Goldstein BI et al. TheCanadian Network for Mood and Anxiety Treatments(CANMAT) task force recommendations for the manage-ment of patients with mood disorders and comorbid meta-bolic disorders. Ann Clin Psychiatry 2012;24:69–81.

88. Miller LJ. Management of atypical antipsychotic drug-induced weight gain: focus on metformin. Pharmacother-apy 2009;29:725–735.

89. Newall H, Myles N, Ward PB, Samaras K, Shiers D,Curtis J. Efficacy of metformin for prevention of weightgain in psychiatric populations: a review. Int Clin Psy-chopharmacol 2012;27:69–75.

90. Correll CU, Sikich L, Reeves G, Riddle M. Metformin forantipsychotic-related weight gain and metabolic abnor-malities: when, for whom, and for how long? A J Psychia-try 2013;170:947–952.

Supporting Information

Additional Supporting Information may be found in the onlineversion of this article:Figure S1. PRISMA flow diagram.Figure S2. Risk of bias summary.Figure S3. Adjunctive topiramate vs. placebo for schizophre-nia: Forest plot for all-cause discontinuation.Figure S4. Adjunctive topiramate vs. placebo for schizophre-nia: Forest plot for categorical adverse drug reactions.Table S1. Summary of studies and patients.Table S2. GRADE Analyses: Adjunctive Topiramate Con-trolled Randomized Trials for Schizophrenia.Appendix S1. Methods.

398

Zheng et al.

本文献由“学霸图书馆-文献云下载”收集自网络，仅供学习交流使用。

学霸图书馆（www.xuebalib.com）是一个“整合众多图书馆数据库资源，

提供一站式文献检索和下载服务”的24 小时在线不限IP

图书馆。

图书馆致力于便利、促进学习与科研，提供最强文献下载服务。

图书馆导航：

图书馆首页文献云下载图书馆入口外文数据库大全疑难文献辅助工具

http://www.xuebalib.com/cloud/http://www.xuebalib.com/http://www.xuebalib.com/cloud/http://www.xuebalib.com/http://www.xuebalib.com/vip.htmlhttp://www.xuebalib.com/db.phphttp://www.xuebalib.com/zixun/2014-08-15/44.htmlhttp://www.xuebalib.com/

efficacy and safety of adjunctive topiramate for...

Documents