Myoinositol Versus Metformin in the Treatment of Polycystic Ovarian Syndrome: A Systematic Review

Polycystic ovarian syndrome (PCOS) is a widespread, complex, and multi-system hormonal disorder that occurs in women of reproductive age. The wide variation in practice in the treatment of PCOS is a direct consequence of the lack of sufficient evidence on alternative treatment strategies, as well as a poor understanding of the disorder itself. The aim of our systematic review was to assess the therapeutic advantages and adverse effects of metformin (MET), a standard treatment modality, with myoinositol (MI), a recent substitute that may be used alone or in combination with other remedies to treat PCOS. A literature search was done using PubMed Central, PubMed, Medline, Cochrane, Science Direct, and Google Scholar. Studies were limited to those published in English between 2012 and 2022 that focused on the management of PCOS with both MET and MI. The systematic review complied with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 guidelines. Using standard quality assessment tools, two reviewers independently assessed the content of the incorporated studies. Three meta-analyses, eight randomized controlled trials (RCTs), and one non-randomized non-controlled trial (NN-RCT) were deemed eligible. Following extensive analysis, we found that MET and MI are comparable in their effects on clinical, hormonal, and biochemical profiles. MI, however, had a better safety profile and tolerance due to minimal side effects compared to MET. These results demonstrate the potential role of MI as a novel asset in the armamentarium in the management of PCOS.

sensitizers have routinely been used as a treatment. Metformin (MET) is the most prevalent insulin sensitizer, typically employed in treating type 2 diabetes, and as an off-label drug in non-diabetic women with PCOS [11]. MET has been shown to impart numerous metabolic and reproductive benefits, including weight reduction, ameliorating IR and androgen levels, and restoring regular menstrual cyclicity and ovulation [12,13]. However, its use is known to be accompanied by considerable side effects, such as nausea, vomiting, and gastrointestinal discomfort [14]. The dissatisfactory compliance noted with MET prompted the discovery of innovative techniques to treat PCOS.
One of the latest advancements in the treatment of PCOS is myoinositol (MI), a naturally occurring substance that has been studied in the last decade because of its insulin-sensitizing effects and broad safety profile [15,16]. Multiple clinical trials have been carried out to evaluate MI's efficacy in addressing metabolic and reproductive symptoms of PCOS women [17,18].
Various studies have been done on MET, D-chiro inositol (DCI, a stereo-isomer of inositol, such as MI), and MI. However, there are insufficient studies that compare the efficacy of these drugs either alone or in combination in treating PCOS. For this reason, we have decided to systematically review those studies to compare these two drugs, MET and MI, for their efficacy and safety in PCOS patients.

Review Methods
We conducted and reported this systematic review in accordance with the Preferred Reporting Items for Systematic Review and Meta-Analyses (PRISMA) 2020 checklist [19].
The PRISMA flow chart is illustrated in Figure 2 [19].

Inclusion and Exclusion Criteria
Our review included full-text peer-reviewed papers written in English that were published between 2012 and 2022 with an available abstract and complete, unrestricted text. The studies included only females of all ages from all over the world, representing individuals diagnosed with PCOS according to the Rotterdam, the National Institutes of Health (NIH), or the Androgen Excess and Polycystic Ovary Syndrome (AE-PCOS) society criteria [20,21]. The participants of the studies received treatment with either MET or MI (used alone or in combination with either D-chiro inositol or folic acid) or both.
Any unpublished articles, studies without access to the full text, studies without a method section, review papers, animal or cell culture studies, case reports, grey literature, and research published before 2012 were excluded. We also excluded studies involving pregnant or lactating patients, other endocrinopathies causing similar symptoms, patients already on other drug treatments for PCOS, treatments combined with other drugs or supplements (excluding D-chiro inositol and folic acid), papers that covered fertility induction medication and pregnancy as a primary outcome, and studies with a low-quality appraisal.

Data Extraction
The selection process was carried out by two independent reviewers. Search queries were developed for PubMed, PubMed Central, Medline, Cochrane, ScienceDirect, and Google Scholar, which helped to streamline the identification of studies. Records deemed ineligible were deleted using the database's automated tools. Subsequently, duplicates were deleted using the EndNote program. The screening process involved reading titles and abstracts of articles and selecting those that could be relevant to our study. A manual search was conducted to locate the reports' full text. Articles for which full texts could not be retrieved were excluded. Ultimately, 861 records were examined, and 142 articles were included for full-text screening, eligibility, and quality assessment.

Quality Assessment and Analysis of the Studies
We assessed the quality of the articles using quality assessment measurement tools. Cochrane risk of bias was employed for the quality assessment of randomized controlled trials (RCTs) [22]. For systematic reviews and meta-analysis, the assessment of multiple systematic reviews (AMSTAR 2) checklist was used [23]. For non-randomized clinical trials and observational studies, the Joanna Briggs Institute (JBI) critical appraisal tool was utilized [24]. Only studies where both reviewers had a 95% accord and had more than 60% in terms of low-risk bias have been included. Any differences in opinions were resolved through discussion between the other authors and, if necessary, consultation with our mentor. This review has a total of 12 articles.
The risk assessment summary for selected studies is shown in Tables 2-4.

Literature Search
Our search strategy yielded 164 possibly relevant articles, out of which 79 had full-text availability. A further 49 papers were dismissed owing to noncompliance with the eligibility criteria. Another 18 papers were excluded from the research on full-text screening. An additional seven reports were deemed ineligible using quality assessment tools. In the end, 12 articles were accepted.

Study Characteristics
The systematic review includes eight RCTs, three meta-analyses, and one non-randomized clinical trial (NN-RCT). All of the articles employed MET and MI as interventions for treating PCOS. The 12 papers were published between 2015 and 2022, and study participants originated from three countries, i.e., India [25,26,28,[32][33][34][35][39][40][41][42][43], Iran [40][41][42], and Italy [40][41][42]. Treatment duration ranged between 12 weeks and six months. Eight papers studied the efficacy of both interventions on anthropometric measures, such as weight, waist-to-hip ratio (WHR), waist circumference, hip circumference, and body mass index (BMI). Seven papers incorporated studies on the influence of the interventions on clinical outcomes, including menstrual cycle abnormalities, hirsutism, acne, and ovarian morphology on ultrasonography (USG). Ten papers study their effect on the hormonal profile, which include serum testosterone, luteinizing hormone (LH), folliclestimulating hormone (FSH), dehydroepiandrosterone (DHEA), androstenedione, sex hormone binding globulin (SHBG), prolactin (PL), and progesterone. The effect on biochemical measures, such as fasting blood sugar (FBS), post-meal blood sugar (PMBS), insulin, glucose-to-insulin ratio, lipid profile, and the Homeostatic Model Assessment for Insulin Resistance (HOMA-IR), was analyzed in 10 papers. Adverse events (AE) were assessed in six papers. It is crucial to note that, currently, there is no single highly effective treatment for PCOS. On account of the complex and multifaceted nature of the syndrome, it is bestaddressed case by case with an individually tailored treatment plan.
The essential characteristics of the incorporated studies are highlighted (   The results of the outcome variables of included studies are displayed ( Table 6) below.

Results Conclusions
Angik et al. (2015) [26] Overall, 37.73% achieved regular cycles, 28.57% with MI, and 48% with MET. The decrease in the mean mFG score of hirsutism and acne was statistically significant in both groups; however, the reduction was not substantial between the two groups. Similar outcomes were noted in BMI, WHR, FBS, PMBS, post-meal insulin, testosterone, LH, LH/FSH ratio, mean ovarian volume, and antral follicle count (AFC). The mean fasting insulin decrease in MI was statistically significant; in the MET group, the decline was insignificant. The HOMA-IR index decrease was substantial in the MI group, but the decrease was not significant in the MET group. Between the two groups, the HOMA-IR decrease was substantial. Overall, 88% attained regular cycles, 87% with MI, and 90% after MET treatment. The results were comparable in both groups. In both groups, there was a statistically significant decrease in patients having PCOM at the end of 24 weeks compared to baseline. For the MI cohort, the percentage decreased from 83% to 13%. The metformin group's rate fell from 80% to 17%. At the end of 24 weeks, there was more significant progress in the number of patients having normal ovaries on USG compared to baseline values with MI than MET (70% vs. 63%). A comparable response was noted with both drugs in reducing the mFG score. Both groups displayed a statistically significant weight reduction compared to baseline values. There was more weight reduction with MI than with MET, but the difference was not statistically significant. There was slightly more improvement in BMI with MI than with MET, but the difference was not statistically significant. In the MI Group, the improvement in acne was 31%.
In the MET group, the improvement was 32%. The incidence of various AEs was higher in the MET group than the MI group, apart from menorrhagia, which was observed only with MI and dizziness, which was comparable in both groups. The MET cohort displayed a greater incidence of nausea, diarrhea, abdominal cramps, flatulence, and generalized weakness. A statistically substantial improvement was observed in insulin resistance, glucose/insulin ratio as well as HOMA-IR in both groups. The hormonal parameters demonstrated statistically significant improvements in both drugs concerning changes in FSH, LH, LH /FSH ratio, and testosterone levels. Lipid profiles also improved for both MI and MET. However, on comparison, the difference was not statistically significant.
There was considerable improvement in the biochemical profile of PCOS patients with both MET and MI.
Nabi et al.
(2018) [32] In the MI group, 53.1% of subjects attained cycle regularity, while in the MET group, 41.9% achieved regular cycles. The decrease in the number of follicles was more prominent with MI than with MET, while the decrease in ovarian volume was almost the same in both groups.
There was a substantial drop in body weight and WHR in both groups, but between the two, it was more significant in the MI group. The reduction in BMI was greater in the MET group. However, for all three indices, the differences were not statistically significant. There were notable drops in the mFG score for both groups as well, however, the fall in the MI group was more effective than for the MET group. Fasting insulin dropped from 13.90 ± 6.88 SD to 11.66 ± 6.05 SD in the MI group and from 12. AEs with MET were nausea and diarrhea, with reports of a severe entity in some cases, abdominal pain, lactic acidosis, and generalized weakness. With MI, they were nausea, mild diarrhea, and menorrhagia. MET and MI displayed no differences in short-term hormone changes.
Due to its better tolerance, MI is more suitable for improving the androgenic and metabolic profiles in PCOS women.
In the MI + DCI group, the menstrual cycle length decreased by seven days, the percentage of women with regular menstrual cycles increased by 20%, and the hirsutism score dropped by 6 points. In the MET group, the menstrual cycle length decreased by seven days, the percentage Both the combination of MI Thalamati (2019) [39] of women with regular menstrual cycles increased by 12%, and the hirsutism score dropped by 1 point. Insulin resistance in both groups demonstrated a statistically significant improvement as assessed with glucose-insulin ratio and HOMA-IR. The MI + DCI group saw an increase of 1.20 in the glucose-insulin ratio and a decrease of 1.32 in the HOMA-IR. In contrast, the MET group showed an increase of 1.03 in the glucose-insulin ratio and LH -MET had a significantly more effective outcome on LH than MI. LH/FSH -the LH/FSH ratio in the MI group was significantly higher than that of the MET group. Serum testosterone -Substantial heterogeneity was detected among the studies. The level of testosterone in the MET group was significantly higher than that observed in the MI group. In other words, MI was more effective than MET. Prolactin -There was evidence of a high heterogeneity among the studies. The level of prolactin in the MI group was significantly higher than that of the MET group. DHEA -DHEA in the MET group was significantly higher than that of the MI group. Therefore, MI was more effective than MET in modulating hyperandrogenism. 17-OH-P-MET was more effective than MI in reducing the 17-OH-P. Other characteristics -Other indices, such as FBS, FSH, estradiol, ovarian volume, and insulin, were estimated to be lower in the MET group than in the MI group. However, there was no statistically significant difference between the effectiveness of the two treatments.

Discussion
To determine and compare the safety and efficacy of MET and MI in PCOS, we addressed the outcomes of anthropometric, hormonal, biochemical, and clinical indices as indicators of efficacy and the development of AEs as a gauge of their safety profile.  [28]. Overall, the results demonstrate that, either alone or in combination with other modalities, MI is beneficial in menstrual cycle regularization.

Efficacy of MET and MI
Ovarian morphology: Five papers addressed changes in ovarian morphology. Both MI and MET are effective in improving ovarian morphology. Angik et al. observed a similar decrease in mean ovarian volume and AFC in both groups [26]. Similarly, Nehra et al. noted an equivocal reduction in polycystic ovarian morphology (PCOM) in both treatment groups [33]. They also concluded that there was a more substantial improvement in the number of subjects having normal ovaries on USG compared to baseline values with MI than MET (70% vs. 63%). In the paper by Nabi et al., the reduction in the number of follicles was considerably more with MI than with MET, while the decrease in ovarian volume was almost the same in both groups [32]. Likewise, on ultrasound, Agarwal et al. found that ovaries achieved better morphology with MI in volume and stromal thickness [25]. Still, the number of follicles decreased similarly for both interventions. Ovarian volume was estimated to be lower in the MET group than in the MI group; however, the difference was not statistically significant according to the results from the study by Kutanaei et al. [41]. MI seems to play a substantial role in normalizing ovarian morphology and recent studies suggest it could be a key factor in improving oocyte and embryo quality.
Anthropometric measures: Statistically significant declines were seen in BMI in both MI and MET, but the difference between the two was not substantial in studies by Angik [26,43,34,32,40,42]. Nabi et al. found that the fall in mean BMI was more remarkable in the MET group [32]. This is in contrast to the study by Bahadur et al., in which they noted a more significant decrease in the MET+MI+DCI group [28]. This can likely be attributed to the difference in interventions utilized as the study measured MET against the combination of MET, MI, and DCI.
Nehra et al. observed that there was a greater reduction in weight with MI as compared to MET, but the difference was not statistically significant [34]. Nabi et al. also noted that, although the drop in body weight was significant in both groups, it was more significant in the MI group [32]. However, again, the difference between the two was not statistically significant.
Similar observations were made in the WHR. Angik et al., Nabi et al., and Zhang et al. all noted significant reductions in the WHR, but the difference between MET and MI groups was not statistically significant [26,32,42]. The study by Bahadur et al. displayed no considerable change in WHR in the groups [28].
Traditionally, the first-line treatment for obese and overweight PCOS patients included lifestyle modifications, such as a hypocaloric diet and physical activity. If patients were found to be resistant to weight loss through lifestyle interventions, then MET would be prescribed as adjuvant therapy. The results of this review show that MI may be considered an equally effective alternative.
Hirsutism and acne: The studies incorporated in this review assessed hirsutism using the mFG score. Most studies described using the global acne score in grading acne. Angik et al. found that the decrease in the mean modified Ferriman-Gallwey (mFG) score of hirsutism and acne was statistically significant in both groups; however, the change was not significant between the two groups [26]. Nehra et al. also noted a comparable response with both drugs in reducing mFG scores [33]. In the MI group, the improvement in acne was 31%. In the MET group, the improvement was 32% [33]. In the study by Nabi et al., it was observed that the mFG score was reduced from 4.66 ± 4.06 SD to 3.56 ± 3.29 SD in the MI group and from 4.94 ± 4.05 SD to 3.87 ± 3.24 SD in the MET group; the fall in MI was more significant than MET, but the difference was not statistically significant [32]. Thalamati found that the hirsutism score decreased by one point in the MET group and six points in the MI+DCI group [39]. Statistically, a significant difference was observed in the mean global acne score of the MET+MI+DCI group's study by Bahadur et al. [28].
Hirsutism and acne are both distressing symptoms for PCOS patients. Currently, these manifestations are addressed through the use of oral contraceptives and androgen blockers, such as spironolactone, and are accompanied by their share of side effects. As the results demonstrate, the use of MI as a natural and safe alternative could be explored further.
Hormonal parameters: Angik et al. observed a significant and comparable decrease in testosterone, LH, and LH/FSH ratio [26]. Awalekar et al. and Nehra et al. also noted significant reductions in the LH/FSH ratio for both interventions [35,43]. In addition, Nehra et al. demonstrated a significant reduction in testosterone for both treatment groups [35]. Nabi et al. recorded that results for LH were not significant; however, the LH/FSH ratio fall was more marked in the MET group than in the MI group. The fall in free testosterone for MI was more significant than for MET. Changes in serum progesterone value were more effective with MI than with MET [32]. No difference between MET and MI was found in testosterone, androstenedione, and SHBG levels in the study conducted by Facchinetti et al. [40]. Thalamati [28]. The analysis by Kutanaei et al. found that both supplements improved estradiol and SHBG. However, the LH level, LH/FSH ratio, and prolactin levels were only enhanced by MET. MET was also more effective than MI in reducing the 17-OH progesterone. A significant decrease in the DHEA and testosterone levels, as circulating androgens, was observed in PCOS patients who received MI compared to those who received MET [41]. No significant differences between MI and MET were observed in the improvement of testosterone and SHBG in the study by Zhang et al. [42].
Both MI and MET have demonstrated their effectiveness in ameliorating the hormonal imbalances associated with PCOS. Current practice involves the use of oral contraceptives for hormonal regulation; however, given the multiple side effects associated with its use, alternate therapies such as MI should be explored as a possibility.
Dysglycemia: Angik et al. observed a significant and comparable decrease in FBS, PMBS, and post-meal insulin. The mean fasting insulin decrease in MI was statistically significant, while in the MET group, it was not. The HOMA index displayed a substantial reduction in the MI group, whereas the decrease was not significant in the MET group [26]. In contrast, Awalekar et al. found that the mean HOMA index reduced from 25.85 to 15.21 (p=0.000) for the MET group and was highly significant statistically, but the reduction for MI was not [43]. Nehra  Lipid profile: According to Nehra et al., the lipid profile displayed an improvement in both groups. However, in the comparison, no statistically significant difference was observed [35]. On the other hand, Agarwal et al. found that the lipids (total cholesterol, high-density lipoprotein, low-density lipoprotein) improved with MET [25]. Bahadur et al. noted a significant difference in mean cholesterol, mean high-density lipoprotein, and mean low-density lipoprotein values in the MET+MI+DCI group. No significant difference between the two groups was seen in mean triglyceride [28]. The study by Zhang et al. showed that MI could lower triglyceride levels compared to MET. For total cholesterol, no significant differences were observed between MI and MET. High-density lipoprotein and low-density lipoprotein showed no significant differences [42].

Side Effects of MET and MI
The studies in our review reported superior safety profiles for MI compared to MET in terms of AEs, such as lactic acidosis, generalized weakness, nausea, menorrhagia, dizziness, diarrhea, abdominal cramps, and flatulence. Angik [26,33,32,40,25,42]. MET treatment's most reported side effects were generalized weakness and gastrointestinal side effects, such as nausea, abdominal cramps, diarrhea, and flatulence. The only reported AE with MI were nausea, dizziness, and menorrhagia; however, the incidence was very low. Hence, it can be concluded that MI is better tolerated and has a broader safety profile compared to MET. This also results in better patient compliance with MI.

Limitations
There are certain limitations to our study, among which the primary limitation was a dearth of high-quality clinical studies comparing the treatment of PCOS with MET and MI. It is necessary to consider the biases of the included studies as they may have influenced the accuracy of the results. Certain biases, such as the lack of well-designed controlled trials, a lack of double blinding, small sample size, variations in doses and combinations of the drugs, short study durations, and the use of various criteria for describing PCOS patients, can influence the effects and reported results of the study.

Conclusions
Our systematic review revealed that MI could be an effective alternative or adjunctive treatment for PCOS patients undergoing conventional drug therapies. MI, a natural nutraceutical and an insulin sensitizer, is comparable to MET in treating PCOS for clinical, hormonal, lipid, glycemic, and insulinemic benefits. Additionally, MI treatment has minimal side effects and better tolerance than MET treatment. Given the increased rate of prevalence of PCOS worldwide, there is potential for further clinical studies with bigger sample sizes, longer study duration, different dosages of drugs, and studies employing a combination of other medications used in the standard treatment of PCOS.

Conflicts of interest:
In compliance with the ICMJE uniform disclosure form, all authors declare the following: Payment/services info: All authors have declared that no financial support was received from any organization for the submitted work. Financial relationships: All authors have declared that they have no financial relationships at present or within the previous three years with any organizations that might have an interest in the submitted work. Other relationships: All authors have declared that there are no other relationships or activities that could appear to have influenced the submitted work.