Description of the condition
Polycystic ovary syndrome (PCOS), one of the most common female endocrine disorders, affects 5% to 10% of women of reproductive age (ESHRE/ASRM group 2008).The disorder is characterised by chronic anovulation (ongoing failure or absence of ovulation), clinical and biochemical hyperandrogenism (hirsutism, acne, elevated circulating androgens) and obesity (Chittenden 2009; Hart 2004; Mayer 2005). PCOS is a leading cause of subfertility (Boomsma 2008) and accounts for 90% of anovulatory subfertility (Balen 2007). Besides this, PCOS impairs endocrine and metabolic functions; 50% to 70% of women with PCOS have insulin resistance (IR) (Nestler 1997), which is far more than for women without PCOS. Women with PCOS are at greater risk of developing type 2 diabetes mellitus, metabolic syndrome (including hypertension, dyslipidaemia, impaired fibrinolysis and vasodilation), cardiovascular disease and endometrial cancer (Chittenden 2009; Lobo 2001; Mayer 2005). The cause of PCOS is still unclear (Dasgupta 2008; ESHRE/ASRM group 2008; Padmanabhan 2009). At present, most researchers claim that PCOS is the result of both genetic and environmental factors, but how these factors interact is unknown (Dasgupta 2008; Diamanti-Kandarakis 2005).
The quality of life (QoL) of women with PCOS is impaired. This is associated with specific features of PCOS (such as obesity, hirsutism, subfertility) and potential mental health problems including depression and allied disorders (McCook 2005). Depression may be an emotional obstacle to seeking medical advice for the PCOS, which may further worsen the syndrome (Herbert 2010). The prevalence of depression in women with PCOS is high; one study has shown it to be four times that of women without PCOS (Hollinrake 2007). Even without diagnosed depression or allied mood disorders, women with PCOS are more anxious and more likely to be in a depressive mood than women without PCOS (Barnard 2007; Benson 2009; Ching 2007; Jedel 2010). The cause of the high prevalence of depression in women with PCOS is not clear; possibly obesity and IR play a part (Himelein 2006; Kerchner 2009; Rasgon 2003).
Description of the intervention
Antidepressants are one of the most commonly used therapies for depression and allied mood disorders. According to their different mechanisms of action, antidepressants can be divided into five broad categories. These are: monoamine oxidase inhibitors (MAOI); tricyclic antidepressants (TCA); tetracyclic antidepressants (TeCA); selective serotonin reuptake inhibitors (SSRI) and serotonin-norepinephrine reuptake inhibitors (SNRI). Common side effects of antidepressants include thirst, nausea, low blood pressure, drowsiness, sexual dysfunction and weight gain (Tong 2009). Amenorrhoea has been reported as another side effect (McElroy 2009; Peritogiannis 2007). Hepatonecrosis has been reported as a serious side effect of MAOIs and has limited the use of this type of antidepressant (Tong 2009).
The effect of some antidepressants on weight is relevant to this review because being overweight or obese is common in women with PCOS. This symptom can independently reduce women's QoL (McCook 2005) and is likely to play a part in the greater prevalence of depression in women with PCOS (Himelein 2006). Sibutramine, an SSRI, must be distinguished from the other antidepressants. It is no longer used as an antidepressant and has been approved by the US Food and Drug Administration (FDA) as an antiobesity drug. Sibutramine inhibits the reuptake of 5-hydroxytryptamine (5-HT, also known as serotonin), resulting in a decrease in appetite, an increase in satiety (thereby naturally reducing calorie intake) and an increase in calorie expenditure that leads to a loss of weight (Halford 2005; Tziomalos 2009). Other SSRIs (fluoxetine, paroxetine, sertraline) have a similar effect of causing weight loss and have been approved for use in obese people (Epling 2003; Haddock 2002; Halford 2005; Halpern 2003). The weight loss effect of SSRIs might have an added beneficial effect on ameliorating the depressive symptoms of women with PCOS. However, this advantage cannot be found in other classes of antidepressants. The effect of MAOIs and SNRIs on weight is slight, and weight gain is an adverse effect of TCAs and TeCAs.
How the intervention might work
Ameliorating depressive symptoms might encourage women with PCOS with psychological problems to seek medical advice (Herbert 2010). Furthermore, it has been shown that weight loss is associated with beneficial effects on hormone levels, metabolism and the clinical features of PCOS (ESHRE/ASRM group 2008; Vrbikova 2009). So as well as improving depressive symptoms, antidepressants may have beneficial effects on endocrine, metabolic and reproductive functions in women with PCOS. All of these effects may lead to improvements in the QoL of women with PCOS.
In theory, antidepressants could benefit all women with PCOS. Women with PCOS are more likely to be anxious and in a depressive mood compared with women without PCOS, even without diagnosed depression or allied mood disorders (Jedel 2010). Metabolic problems are more common in obese women with PCOS but are also evident in lean women with PCOS (Bideci 2008; Shumak 2009; Veldhuis 2001; Zwirska-Korczala 2008). Antidepressants may, therefore, be effective in women with PCOS who do not have depression or allied mood disorders and are not obese or overweight. This review sets out to evaluate systematically both the beneficial and side effects of antidepressants in women with PCOS and to identify whether their use is most applicable to any specific subgroups of women.
Why it is important to do this review
It is important to determine whether antidepressants are indicated for treatment of PCOS, whether all classes of antidepressants can be used and whether antidepressant therapy is applicable for all women with PCOS or just those with depression. This review sets out to answer these questions, and to determine whether antidepressants have any advantages compared with other treatments for PCOS.
To evaluate the effectiveness and safety of antidepressants in treating depression and other symptoms in women with PCOS.
Criteria for considering studies for this review
Types of studies
Only randomised controlled trials (RCTs) studying the effectiveness of antidepressants for women with PCOS were included in this review. Quasi-RCTs and pseudo-RCTs were excluded. We planned to include cross-over trials if the data from the first phase were available.
Types of participants
1. Women (including adolescents) of reproductive age (aged 13-44 years) with PCOS were included.
2. PCOS was identified according to the Rotterdam Consensus (2003) (ESHRE/ASRM group 2004):
- olio-ovulation or anovulation;
- clinical or biochemical signs of hyperandrogenism;
- polycystic ovaries (determined by gynaecological ultrasound).
Two of the three features listed above must have been present and other aetiologies (congenital adrenal hyperplasia, androgen-secreting tumours, Cushing's syndrome) excluded.
3. Participants with PCOS both with and without depression were included. Because of the lack of unified diagnostic criteria, depression was identified according to criteria including the International Statistical Classification of Diseases and Related Health Problems 10th Revision (ICD-10) (WHO 1992), the Diagnostic and Statistical Manual of Mental Disorders, 4th Edition (DSM-IV) (APA 1994) or the Chinese Classification of Mental Disorders (CCMD) (PICMA 1995); and according to participant scores on validated tools such as the Beck Depression Inventory (BDI), Hamilton Rating Scale for Depression (HRSD), the Montgomery-Asberg Depression Rating Scale (MADRS), or the Hospital Anxiety and Depression Scale (HAD).
4. Both obese and non-obese women with PCOS were included. Obesity was identified according to body mass index (BMI): for Europeans and Americans BMI less than 18.5 is defined as lean, BMI 18.5 or greater to less than 25 as normal, BMI 25 or greater as overweight and BMI 30 of greater as obese; for Asians BMI less than 18.5 is defined as lean, BMI 18.5 or greater to less than 23 as normal, BMI 23 or greater as overweight and BMI 25 or greater as obese (WHO 2000).
5. Women with PCOS with and without IR were included. IR was identified according to homeostasis model assessment-insulin sensitivity index (HOMA-ISI) (Tara 2004).
Types of interventions
1. Antidepressant medications versus placebo, no treatment, oral contraceptives, metformin, lifestyle changes, antiobesity drugs, or surgery; with and without co-medications.
2. Antidepressant medication combined with other treatments versus other treatments
3. One antidepressant versus another antidepressant.
4. Comparison of different doses of the same antidepressant.
Types of outcome measures
(1) Depression and allied mood disorder scores
We planned to assess depression and allied mood disorders using the following hierarchy of tools (as reported by the included studies):
- the BDI;
- other validated generic scales;
- condition-specific scales.
(2) Quality of life
We planned to evaluate QoL using the following hierarchy of tools (as reported by the included studies):
- the Short Form-36 (SF-36);
- the Quality-of-Life Questionnaire for Women with Polycystic Ovary Syndrome (PCOSQ);
- other validated generic scales;
- other condition-specific scales.
(3) Adverse events
- Any adverse event: how many women had any adverse event.
- Specific adverse events: how many women had a specific type of adverse event.
(4) Reproductive factors
- Live birth rate (excluding assisted conception), defined as the number of women giving birth to living baby divided by the total number of women seeking pregnancy.
- Ovulation (number of ovulatory menstrual cycles).
- Menstrual regularity.
(5) Endocrine factors
Levels of total testosterone (tT), free testosterone (fT), luteinising hormone (LH), follicle-stimulating hormone (FSH), LH/FSH, sex hormone-binding globulin (SHBG), dehydroepiandrosterone sulphate (DHEA-S), oestrogens (E2) or progesterone (P).
(6) Metabolic factors
Oral glucose tolerance test (OGTT) (including fasting glucose, fasting insulin, glucose/insulin ratio, insulin sensitivity index (ISI), insulin resistance index (IRI) and homeostasis model assessment-insulin resistance (HOMA-IR)); and levels of total cholesterol (TC), high-density lipoprotein cholesterol (HDL-C), low-density lipoprotein cholesterol (LDL-C), triglyceride (TG), apolipoprotein A (Apo A) and apolipoprotein B (Apo B).
(7) Anthropometric factors
Weight, BMI, waist-to-hip ratio and the Ferriman-Gallwey score (for the evaluation of hirsutism).
Search methods for identification of studies
The literature search aimed to locate RCTs regardless of language or publication status (published, unpublished, in press or in progress).
1. We searched the Cochrane Menstrual Disorders and Subfertility Group (MDSG) Trials Register (inception to 8 June 2012) using the search strategy as follows: Keywords CONTAINS "PCOS" or "polycystic ovary syndrome" or "polycystic ovary syndrome" or Title CONTAINS "PCOS" or "polycystic ovary syndrome" or "polycystic ovary syndrome" AND Keywords CONTAINS "anti-depressants" or "antidepressants" or "selective serotonin re-uptake inhibitor" or "SSRI" or "SSRIs" or "serotonin-norepinephrine reuptake inhibitor" or "serotonin reuptake inhibitor" or "SNRI" or "TCA" or "monoamine" or "MOAI" or "TeCA" or "monoaminoxidase" or Title CONTAINS "anti-depressants" or "antidepressants" or "selective serotonin re-uptake inhibitor" or "SSRI" or "SSRIs" or "serotonin-norepinephrine reuptake inhibitor" or "serotonin reuptake inhibitor" or "SNRI" or "TCA" or "monoamine" or "MOAI" or "TeCA" or "monoaminoxidase".
2. The following databases were also searched (inception to 8 June 2012): the Cochrane Central Register of Controlled Trials (CENTRAL) (The Cochrane Library), MEDLINE, EMBASE, PsycINFO, Chinese National Knowledge Infrastructure (CNKI).
The MEDLINE search was combined with the Cochrane highly sensitive search strategy for identifying randomised trials that appears in the Cochrane Handbook of Systematic Reviews of Interventions (Version 5.1.0, Chapter 6, 6.4.11).The EMBASE search was combined with trial filters developed by the Scottish Intercollegiate Guidelines Network (SIGN).
3. We searched for ongoing or registered RCTs in the following databases: meta Register of Controlled Trials (controlled-trials.com), the National Institute of Health Clinical Trials register (clinicaltrials.gov) and the World Health Organization International Trials Registry Platform search portal (www.who.int/trialsearch/Default.aspx).
Searching other resources
1. Reference lists of relevant trials, reviews and textbooks were checked.
2. Experts and pharmaceutical companies in the field were contacted.
Data collection and analysis
Selection of studies
Two review authors (ZJ, WX) independently performed the searches and scanned the titles and abstracts of every record retrieved. All potentially relevant articles were investigated as full text. The review authors (ZJ, WX and XL) then independently assessed whether studies met the inclusion criteria. We planned to resolve disagreements by discussion. If there was not enough information in an article to make this decision, we planned to contact the original trial authors for further information.
Data extraction and management
Data were independently extracted by two review authors (ZJ, WX) using forms designed to follow The Cochrane Collaboration guidelines. Data on study characteristics and results, including methods, participants, interventions, outcomes and adverse events, were extracted. We planned to resolve disagreements by discussion with a third review author (XL).
Assessment of risk of bias in included studies
The methodological quality of the trials was assessed independently by two review authors (ZJ, WX). The risk of bias in the studies was assessed in six domains: 1. sequence generation; 2. allocation sequence concealment; 3. blinding of participants, personnel and outcome assessors; 4. completeness of outcome data; 5. selective outcome reporting; 6. other potential sources of bias.
We followed the detailed descriptions of these categories provided in the Cochrane Handbook for Systematic Reviews of Interventions (Version 5.1.0) (Higgins 2011).
Measures of treatment effect
We planned to analyse continuous data by using mean differences (MD), and dichotomous data by Peto odds ratios (OR), with 95% confidence intervals (CI). Ordinal data (for example, QoL scores) were to be treated as continuous data.
Unit of analysis issues
We planned that if any cross-over trials were included, only the first-phase data would be included.
Dealing with missing data
We planned to contact trial authors by telephone or email to ask for missing data. If suitable data were available, intention-to-treat (ITT) analysis was conducted. If suitable data were unobtainable, we planned to assume that participants with unreported outcomes had not had improvements for our primary outcome.
Assessment of heterogeneity
We planned to assess statistical heterogeneity using the Chi
Assessment of reporting biases
If there were 10 or more studies in an analysis, we planned to assess review-wide reporting bias by constructing a funnel plot.
We planned to pool data if studies were sufficiently similar, using Review Manager (Version 5.0.1) software (RevMan 2011). If pooling was not appropriate, we planned to perform only descriptive analysis. We planned to use a ﬁxed-effect model unless there was substantial heterogeneity, in which case a random-effects model would be used.
We planned to combine the data from primary studies in the following comparisons.
- Antidepressant versus placebo, no treatment, oral contraceptive or surgery (with or without co-medications) stratified by dose of antidepressant:
- low dose;
- high dose.
- Antidepressant versus lifestyle changes, stratified by:
- antidepressant versus diet;
- antidepressant versus exercise;
- antidepressant versus diet combined with exercise.
- Antidepressant versus metformin or antiobesity drug, stratified by dose of each:
- low dose of antidepressant;
- high dose of antidepressant;
- low dose of metformin or antiobesity drug;
- high dose of metformin or antiobesity drug.
- Antidepressant combined with other treatment versus other treatment, such as oral contraceptive, metformin, lifestyle changes, surgery, stratified by dose of antidepressant:
- low dose;
- high dose.
- One antidepressant versus another antidepressant, stratified by dose of both antidepressants.
- low dose;
- high dose.
- An antidepressant in a low dose versus the same antidepressant in a high dose.
Subgroup analysis and investigation of heterogeneity
If substantial heterogeneity was detected, we planned to explore it, based on the characteristics of study participants:
- women with PCOS having depression or allied mood disorders, or not;
- women with PCOS who were overweight or obese, or not;
- women with PCOS who had IR, or not;
- women diagnosed using differing criteria for PCOS.
We planned to perform sensitivity analyses in order to explore the influence of the following factors on review findings:
- study quality (restricting analysis to studies which reported adequate methods of allocation concealment and blinding, and which used ITT analysis);
- funding source: restricting analysis to studies without commercial funding.
Description of studies
Results of the search
A total of 85 references (excluding duplications) were identified by the electronic searching in June 2012, 74 in English and 11 in Chinese. Eleven studies were potentially eligible and were retrieved in full text. One study met our inclusion criteria. Ten studies were excluded. See study tables: Characteristics of included studies and Characteristics of excluded studies. We prepared a PRISMA flow diagram to describe the articles found from our searches (Figure 1).
|Figure 1. Study flow diagram.|
The included trial was a RCT conducted in Turkey (Karabacak 2004). This was a single centre two-arm parallel arm study.
A total of 16 women were enrolled. It was unclear whether they had psychological problems at baseline. The PCOS diagnostic criteria were a combination of oligo-ovulation or anovulation, clinical and biochemical signs of hyperandrogenism, and polycystic ovaries (determined by gynaecological ultrasound) (ESHRE/ASRM group 2004). The baseline characteristics among groups were comparable. The inclusion and exclusion criteria were listed in Characteristics of included studies. Dropouts and withdrawals had not been reported
The included study did not report any of the primary outcomes of the review. It reported the following secondary outcomes:
- endocrine outcomes: LH, FSH and LH/FSH were reported;
- metabolic outcomes: OGTT were reported.
For abbreviations see Table 1.
Ten studies were excluded for different reasons:
- 5/10 were not RCTs;
- 1/10 reported no comparisons of interest;
- 1/10 had participants without PCOS;
- 3/10 reported intervention without antidepressants.
Risk of bias in included studies
The included study did not report what methods were used for sequence generation or allocation concealment, and it was rated as at unclear risk of bias for these domains.
|Figure 2. Risk of bias graph: review authors' judgements about each risk of bias item presented as percentages across all included studies.|
|Figure 3. Risk of bias summary: review authors' judgements about each risk of bias item for each included study.|
Incomplete outcome data
The included study did not state how many women were analysed and we rated it as at unclear risk of attrition bias (Karabacak 2004).
Other potential sources of bias
No other potential source of bias was identified.
The risk of review-wide reporting bias could not be assessed because there were insufficient studies to do a funnel plot.
Effects of interventions
Because there was only one study included, we described the results in narrative form.
1 Antidepressants plus other treatments versus other treatments
1.1 Sibutramine versus fluoxetine
1.1.1 Depression and allied mood disorder scores
Depression and allied mood disorder scores were not reported in the included study (Karabacak 2004).
1.1.2 Quality of life (QoL)
QoL was not reported in the included study (Karabacak 2004).
1.1.3 Adverse events
Adverse events were not reported in the included study (Karabacak 2004).
1.1.4 Reproductive factors
Reproductive factors were not reported in the included study (Karabacak 2004).
1.1.5 Endocrine factors
Luteinising hormone (U/L)
LH increased after treatment in the fluoxetine treatment group, but decreased in the sibutramine treatment group. However these differences were not statistically significant (baseline 8.53 ± 6.96 vs. 7.28 ± 5.37, after treatment 9.42 ± 11.85 vs. 5.24 ±3 .85, P value > 0.05, 16 women) (Karabacak 2004).
Follicle-stimulating hormone (U/L)
The FSH level decreased slightly in both fluoxetine and sibutramine groups, but the differences in and between the groups were not statistically significant (baseline 4.44 ± 2.13 vs. 3.84 ± 1.64, after treatment 3.23 ± 1.74 vs. 3.68 ± 1.48, P value > 0.05, 16 women) (Karabacak 2004).
Luteinising hormone/follicle-stimulating hormone
LH/FSH increased in fluoxetine group and decreased in sibutramine group, but the differences in and between the groups were not statistically significant (baseline 2.08 ± 1.44 vs. 1.78 ± 1.05, after treatment 2.32 ± 1.74 vs. 1.57 ± 1.12, P value > 0.05, 16 women) (Karabacak 2004).
1.1.6 Metabolic factors
Oral glucose tolerance test
No significant differences between pre-treatment and post-treatment were observed for OGTT (P value > 0.05). Detailed information could not be obtained from primary authors (Karabacak 2004).
1.1.7 Anthropometric factors
Anthropometric factors were not reported in the included study (Karabacak 2004).
Summary of main results
This review did not find any studies that reported clinical outcomes in women with PCOS treated for depression. The single eligible study reported only laboratory test outcomes. No significant difference was found for any outcomes between sibutramine and fluoxetine.
Overall completeness and applicability of evidence
The applicability of the included study was very limited because none of the primary outcomes of this review were reported and it addressed only secondary laboratory outcomes. Attempts were made to contact the study author for more details, but no reply was received. The limited evidence applies only to women with PCOS. It was unclear whether women in the included study were depressed at baseline. Because of the severe limitations of the current evidence, the clinical question of this review cannot be answered.
Quality of the evidence
Only one very small RCT with 16 women was included in the present review. It reported no clinically relevant outcomes, and it was of very limited quality, with inadequate reporting of study methods.
Potential biases in the review process
We did not identify any potential biases in the review process.
Agreements and disagreements with other studies or reviews
We found no other systematic review of antidepressants for women with PCOS.
Implications for practice
There is no evidence on the effectiveness and safety of antidepressants in treating depression and other symptoms in women with PCOS.
Implications for research
Large and well-designed RCTs with double blinding should be conducted on the use of antidepressants for women with PCOS, as there is currently no evidence to guide clinical practice. Study authors should report methodology in detail, such as randomisation and allocation concealment methods.
We thank Jane Clarke (the ex-Managing Editor), Helen Nagels (Managing Editor), Marian Showell (Trials Search Coordinator) and the Cochrane Menstrual Disorders and Subfertility Group for their valuable assistance in preparing this manuscript.
Data and analyses
This review has no analyses.
Appendix 1. CENTRAL search strategy (The Cochrane Library)
1 exp Polycystic Ovary Syndrome/ (617)
2 Polycystic Ovar$.tw. (893)
3 PCOS.tw. (599)
4 PCOD.tw. (22)
5 stein leventh$.tw. (3)
6 (ovar$ adj2 sclerocystic).tw. (0)
7 (ovar$ adj2 degeneration).tw. (1)
8 PCO.tw. (288)
9 or/1-8 (1245)
10 exp antidepressive agents/ or exp antidepressive agents, second-generation/ or exp antidepressive agents, tricyclic/ (8608)
11 antidepress$.tw. (5575)
12 TCA.tw. (145)
13 monoamine$.tw. (701)
14 MAOI$.tw. (93)
15 TeCA$.tw. (10)
16 selective serotonin reuptake inhibitor$.tw. (947)
17 SSRI$.tw. (945)
18 serotonin norepinephrine reuptake inhibitor$.tw. (77)
19 SNRI$.tw. (56)
20 monoaminoxidase$.tw. (3)
21 serotonin uptake inhibitor$.tw. (91)
22 or/10-21 (12,020)
23 22 and 9 (1)
Appendix 2. MEDLINE search strategy (Ovid)
1 exp Polycystic Ovary Syndrome/ (8620)
2 Polycystic Ovar$.tw. (8437)
3 PCOS.tw. (4519)
4 PCOD.tw. (252)
5 stein leventh$.tw. (582)
6 (ovar$ adj2 sclerocystic).tw. (80)
7 (ovar$ adj2 degeneration).tw. (89)
8 PCO.tw. (3245)
9 or/1-8 (13,266)
10 exp antidepressive agents/ or exp antidepressive agents, second-generation/ or exp antidepressive agents, tricyclic/ (110,914)
11 antidepress$.tw. (40,676)
12 TCA.tw. (5824)
13 monoamine$.tw. (25,633)
14 MAOI$.tw. (897)
15 TeCA$.tw. (310)
16 selective serotonin reuptake inhibitor$.tw. (6664)
17 SSRI$.tw. (5749)
18 serotonin norepinephrine reuptake inhibitor$.tw. (356)
19 SNRI$.tw. (524)
20 monoaminoxidase$.tw. (141)
21 serotonin uptake inhibitor$.tw. (506)
22 or/10-21 (151,085)
23 22 and 9 (29)
24 randomized controlled trial.pt. (315,458)
25 controlled clinical trial.pt. (83,272)
26 randomized.ab. (230,747)
27 placebo.tw. (135,547)
28 clinical trials as topic.sh. (157,457)
29 randomly.ab. (169,531)
30 trial.ti. (98,696)
31 (crossover or cross-over or cross over).tw. (51,826)
32 or/24-31 (772,305)
33 exp animals/ not humans.sh. (3,656,215)
34 32 not 33 (713,180)
35 23 and 34 (6)
Appendix 3. EMBASE search strategy (Ovid)
1 exp ovary polycystic disease/ (14,506)
2 Polycystic Ovar$.tw. (11,443)
3 PCOS.tw. (6734)
4 PCOD.tw. (304)
5 stein leventh$.tw. (538)
6 (ovar$ adj2 sclerocystic).tw. (81)
7 (ovar$ adj2 degeneration).tw. (96)
8 PCO.tw. (2779)
9 or/1-8 (18,179)
10 exp antidepressant agent/ (273,673)
11 antidepress$.tw. (56,351)
12 TCA.tw. (7138)
13 monoamine$.tw. (27,950)
14 MAOI$.tw. (1145)
15 TeCA$.tw. (524)
16 selective serotonin reuptake inhibitor$.tw. (9189)
17 SSRI$.tw. (9272)
18 serotonin norepinephrine reuptake inhibitor$.tw. (561)
19 SNRI$.tw. (1058)
20 monoaminoxidase$.tw. (157)
21 serotonin uptake inhibitor$.tw. (582)
22 or/10-21 (304,679)
23 22 and 9 (433)
24 Clinical Trial/ (865,899)
25 Randomized Controlled Trial/ (322,492)
26 exp randomisation/ (58,269)
27 Single Blind Procedure/ (15,924)
28 Double Blind Procedure/ (109,025)
29 Crossover Procedure/ (33,967)
30 Placebo/ (198,958)
31 Randomi?ed controlled trial$.tw. (74,924)
32 Rct.tw. (9244)
33 random allocation.tw. (1144)
34 randomly allocated.tw. (17,121)
35 allocated randomly.tw. (1806)
36 (allocated adj2 random).tw. (706)
37 Single blind$.tw. (12,161)
38 Double blind$.tw. (127,705)
39 ((treble or triple) adj blind$).tw. (268)
40 placebo$.tw. (174,566)
41 prospective study/ (204,466)
42 or/24-41 (1,247,638)
43 case study/ (15,680)
44 case report.tw. (224,934)
45 abstract report/ or letter/ (832,694)
46 or/43-45 (1,068,768)
47 42 not 46 (1,212,774)
48 23 and 47 (46)
Appendix 4. PsycINFO search strategy (Ovid)
1 Polycystic Ovar$.tw. (192)
2 PCOS.tw. (103)
3 PCOD.tw. (5)
4 stein leventh$.tw. (2)
5 (ovar$ adj2 sclerocystic).tw. (1)
6 (ovar$ adj2 degeneration).tw. (0)
7 PCO.tw. (36)
8 exp endocrine sexual disorders/ (757)
9 or/1-8 (920)
10 exp antidepressant drugs/ or exp serotonin norepinephrine reuptake inhibitors/ or exp sertraline/ or exp monoamine oxidase inhibitors/ (29,143)
11 antidepress$.tw. (25,798)
12 TCA.tw. (437)
13 monoamine$.tw. (6375)
14 MAOI$.tw. (490)
15 TeCA$.tw. (6)
16 selective serotonin reuptake inhibitor$.tw. (4739)
17 SSRI$.tw. (3958)
18 serotonin norepinephrine reuptake inhibitor$.tw. (233)
19 SNRI$.tw. (330)
20 monoaminoxidase$.tw. (36)
21 serotonin uptake inhibitor$.tw. (184)
22 or/10-21 (45,709)
23 9 and 22 (21)
Appendix 5. CNKI search strategy (Chinese database)
1 polycystic ovary syndrome (8560)
2 polycystic ovary (7682)
3 ovary polycystic disease (30)
4 or/1-3 (9323)
5 antidepressant (12,765)
6 tricyclic antidepressant (1073)
7 monoamine oxidase inhibitor (2695)
8 serotonin uptake inhibitor (1052)
9 or/5-8 (12,854)
10 random* (966,377)
11 4 and 9 and 10 (11)
Last assessed as up-to-date: 8 June 2012.
Protocol first published: Issue 7, 2010
Review first published: Issue 5, 2013
Contributions of authors
Jing Zhuang: drafting the protocol; literature searching; selection of trials for inclusion or exclusion; assessing methodological quality of studies; data extraction, analysis and presentation; co-drafting of the review.
Xianding Wang: literature searching; selection of trials for inclusion or exclusion; assessing methodological quality of studies; data extraction, analysis and presentation; contacting primary authors; co-drafting of the review.
Liangzhi Xu: resolving disagreements regarding study selection; assessing methodological quality of studies; assistance with statistics; review revision.
Taixiang Wu: data analysis and presentation; assistance with statistics; review revision.
Deying Kang: data analysis and presentation; assistance with statistics; review revision.
Declarations of interest
The authors declared that no conflict of interest was involved in the analysis or interpretation of data, or in the writing of the systematic review.
Sources of support
- West China Second University Hospital, Sichuan University, China.
- Chinese Cochrane Center, West China Hospital, Sichuan University, China.
- No sources of support supplied
Differences between protocol and review
The outcome of depression was moved from secondary to primary outcome.
The objectives were changed to include improving QoL and depression.
Medical Subject Headings (MeSH)
Antidepressive Agents [*therapeutic use]; Cyclobutanes [therapeutic use]; Depression [*drug therapy]; Fluoxetine [therapeutic use]; Polycystic Ovary Syndrome [*psychology]; Randomized Controlled Trials as Topic
MeSH check words
* Indicates the major publication for the study