Inositol supplementation and body mass index: A systematic review and meta‐analysis of randomized clinical trials

Abstract Background Inositol is a sugar‐alcohol and recognized as a key component of cell membrane phospholipids. It has crucial role in the cell signaling pathways and contribute to improving glycemic responses. Although some earlier studies have revealed the effect of inositol mediating glucose uptake by improving insulin sensitivity, the benefit of inositol supplementation in patients with overweight and obesity is not completely understood. This study aimed to assess the impact of inositol supplementation on body mass index (BMI) through a systematic review and meta‐analysis of controlled clinical trials. Methods A systematic search was performed to August 2021 in the following databases: PubMed‐Medline, Embase, Web of Science and Scopus. Fifteen controlled clinical trials investigating the effect of inositol on adult's BMI were finally included in the study. A random‐effects model was employed to estimate the effect size. Subgroup analysis was performed by dose, duration, age, type of inositol. Meta‐regression was used to investigate presence of any linear relationship. Begg's and Egger's tests were carried out to detect small study effect. Results The results of pooled analysis showed that inositol supplementation significantly decreased BMI scores (WMD = −0.41 kg/m2; 95% CI: −0.78, −0.04; p = 0.028). Subgroup analysis was performed to identify the source of heterogeneity among studies (I 2 = 73.9%, p < 0.001), demonstrating supplementation duration, baseline BMI, mean age of participants, type of inositol and dosage were potential sources of heterogeneity. The effect of intervention was more clinically significant in participants with polycystic ovary syndrome (PCOS) and overweight/obesity. Inositol in the form of myo‐inositol (MI) had stronger effect on BMI reduction. Conclusion The meta‐analysis suggests that oral inositol supplementation has positive effect on BMI reduction. Inositol supplementation could be considered as an adjunct treatment to improve body mass index.


| INTRODUCTION
Inositol (1,2,3,4,5,6-cyclohexanehexol) is a sugar-alcohol with sixcarbon ring structure and has nine stereoisomers in nature namely myo-inositol (MI) and Di-chiro-inositol (DCI). In addition, there is a 3-O-methyl form of DCI (D-Pinitol) which is a natural derivate of inositol. 1 While there is an endogenous production from glucose-6phosphate, the major part of inositol comes from the exogenous sources like citrus fruits, beans, artichokes and most of other plants. 2 Inositol is mostly represented in the free form or phosphatidylinositol (PI) form from animal sources and in the InsP6 form from the plant sources. 3 Two reactions are needed to produce MI from glucose-6phosphate. First, isomerization of glucose-6-phosphate to inositol-3-phosphate by the NADH-dependent, cytosolic inositol 3phosphate synthase; and second, dephosphorylation of inositol 3phosphate to MI by inositol-monophosphatase (IMPase) for further use. 3 It has been revealed that, dietary MI supplementation in vivo is beneficially associated with lipid metabolism, bone formation, skeletal muscle metabolism, reproduction, nerve function, and brain actions. 4 Finally, MI is degraded in the kidney in mammalians. 5 Inositol deficiency occurs when food-dependent intake, de novo synthesis, and intestinal and cellular uptake are reduced and catabolism and excretion are increased. 6 Despite wide clinical usages of MI, there is still scarce information on the MI safety and side effects. A review study that assessed the safety of MI supplementation in clinical and non-clinical studies reported that MI caused mild gastrointestinal side effects in highest dose of MI (12 g/day). 7 A review study concluded that 2 g/day MI twice a day is the best MI regimen to improve metabolic pathways. 8 Italian Diabetes National Societies (SID) considered 4 g/day of MI as a therapeutic agent in the treatment of gestational diabetes mellitus. 8 It must be noticed that inositol is categorized as generally recognized as safe (GRAS) supplement by the FDA. 9 Inositol also is a key component of cell membrane phospholipids in the form of inositol triphosphate (IP3) playing a pivotal role in the G-protein-dependent pathways. 10 The inositol-dependent pathways transduce the signal of some hormones including vasopressin, thyrotropin-releasing hormone (TRH), thyroid-stimulating hormone (TSH), angiotensin II, and GnRH. 11 Also, inositol acted as an osmolyte in some tissues, such as kidney medulla and brain, where osmolarity has a pivotal biological meaning. 12 In addition, inositol (either in the DCI or MI isoform) has been reported to mediate glucose uptake by improving insulin sensitivity. Insulin-mimetic effects of MI or its isomers seem to be derived from inositol phosphoglycan (IPG) which contains MI or DCI. 13 IPG acts as a second messenger in insulin signaling pathways and plays a vital role in the activation of enzymes involved in glucose uptake and utilization. 14 Different rates of IPG imbalance have been observed in patients with diabetes and obesity. 15 Pinitol is recognized to be a mediator of the insulin signaling route and to be contained in glycosyl phosphatidil-inositol protein anchors. 16 Moreover, studies have reported an increase in the number and quality of oocytes (immature eggs in the ovary) following the use of MI supplementation in women who underwent in vitro fertilization. 17 There are some nutritional supplements that are advantageous in weight loss. A relatively new supplement which has attracted a lot of attention MI and DCI. 18,19 Recently, several studies have reported the effectiveness of DCI and MI in improving the complications associated with obesity in patients with polycystic ovary syndrome (PCOS) as well as metabolic syndrome and those who are suffering from diabetes. [19][20][21] It seems that inositol beneficial effect on obesity and the associated disorders may be related to its involvement in insulin signaling and improving insulin sensitivity.
Overall, the beneficial effects of inositol supplementation on BMI has not been completely understood and therefore, the purpose of this systematic review and meta-analysis is to address this issue, discuss the possible mechanisms and open a new window for further studies.

| Search strategy
The current study was performed according to the guiding principle of PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses). 22 The study was registered in the PROSPERO (Ref. Relevant publications were identified, and their related articles and citations were also scanned. Besides, the reference lists of all identified articles were examined to avoid missing any eligible study.

| Study selection and inclusion and exclusion criteria
Initially based on the PICO (Participants = adults, Intervention = inositol supplementation, Comparison = placebo or control group, Outcome = Changes in body mass index), the following principles were used to include studies: (i) controlled clinical trials with either parallel or cross-over design with inositol supplementation in adults (ii) investigation of the effects of any type of inositol on BMI, (iii) providing sufficient information at the baseline and end of the trial on BMI in both inositol and control groups. Studies were excluded if they: (i) were observational studies, (ii) were studies without control group, (iii) supplemented inositol concomitant with other ingredients, and (iv) were conducted on pregnant or lactating women.

| Data extraction
Data were extracted from eligible studies by two independent reviewers (Azadeh Dehghani and Nima Radkhah) using a specially developed data extraction form according to the selection criteria.
The extracted information included the description of the study, participants, intervention (dose and duration of supplementation with inositol) and study results based on the determined aforementioned outcomes. In terms of insufficient information, authors were requested by an e-mail for providing numerical data and further explanation in the case of any ambiguity, if possible. Any disagreements and doubts were resolved based on consensus discussions with the third reviewer (Meysam Zarezadeh). All extracted data were reappraised by the senior authors.

| Quality assessment
Cochrane Collaboration's risk of bias tool was employed to assess bias in the included studies, comprising inadequacy of sequence generation, improper blinding and allocation concealment, not addressing the dropouts (incomplete outcome data), selective outcome reporting, and other potential sources of bias. Based on the Cochrane Handbook recommendations, a judgment of "yes" indicated a low risk of bias, while "no" indicated a high risk of bias.
Labeling an item as "unclear" indicated an unclear or unknown risk of bias. Two reviewers (Azadeh Dehghani and Amir Hossein Faghfouri) evaluated the quality of included studies and any discrepancy was resolved through consensus with third reviewer (Meysam Zarezadeh).

| Statistical analysis
The current meta-analysis was performed based on a random-effects model using restricted maximum likelihood method (REML) if the amount of heterogeneity was high. The I 2 statistics and Cochrane Q test were used to assess the heterogeneity in which I 2 > 50% and p < 0.1 were defined as high between-study heterogeneity. 23 If the amount of heterogeneity was not significant, fixed-effects model was applied. Effect size was estimated following meta-analysis of mean differences (MD), and their respective standard deviations (SD). In the cases that mean � SE, median (range) and median (Q25-Q75) were reported, mean � SD was estimated using appropriate statistical equations. Due to the identical scale (kg/m 2 ) across studies, effect size was presented as weighted mean difference (WMD) as the units of studies variable were identical. Impact of moderator variables including mean age of participants, dosage, duration of supplementation, and sample size on the final effect size was evaluated using meta-regression analysis based on the presence of any linear relationship. In addition, subgroup analyses were conducted based on the duration of the supplementation, BMI at the baseline, mean age of the study participants, inositol dosage and population groups to clarify the potential sources of heterogeneity and the different diagnoses of the related complications. Sensitivity analysis was performed to assess the influence of each study removal on the overall effect size using leave-one-out method. For evaluation of small study effect, Begg's adjusted rank correlation and Egger's regression asymmetry tests were performed. 24   three studies were on pre-diabetic, diabetic, and healthy participants.

| Risk of bias assessment
According to Cochrane handbook, the results of quality assessment of included studies are presented in Figure 2.

| Effect of inositol on BMI
The results of pooled meta-analysis showed that inositol supple-  -391 population groups were identified as potential sources of heterogeneity following subgroup analyses (Table 2). In addition, based on subgroup analysis, inositol supplementation had more diminishing effect on BMI scores in individuals with overweight and obesity (Table 2). In addition, it seems that MI supplements has stronger lowering effect on BMI compared with other forms of inositol ( Table 2). Moreover, the diminishing effect of inositol on BMI was more considerable in participants with PCOS with overweight and obesity (Table 2). Additionally, Subgroup analysis revealed that intervention for less than 12 weeks and at a dosage <1000 mg had more clinically significant effect. Supplementation in participants with age >40 was more efficacious (

| DISCUSSION
In the meta-analysis of 15 controlled clinical trials, inositol supplementation was associated with reductions in BMI. Subgroup analysis showed that intervention for less than 12 weeks and at a dosage In a short three-armed study conducted by Chirania et al., the effects of metformin (1500 mg), MI (1 g/day), or the mixture of both were compared in infertile women with PCOS. There were remarkably further improvements in symptoms and hormonal parameters, and weight loss in both the second and third groups. The conclusion was that MI could be suggested for women with PCOS. 40 Of note, DCI supplementation was also associated with a decreases in diastolic and systolic blood pressures, serum levels of triglyceride, total cholesterol and circulating insulin, improved glucose tolerance and reduced serum androgen concentrations. 18 Remarkably, inositol has been characterized as a second messenger system resulting in enhanced metabolic enzymes activity due to its insulin-like properties. 19,41 Regarding possible mechanisms, as previously articles stated, DCI oral supplementation may ameliorate glucose metabolism and partially restore peripheral insulin sensitivity 42 and oocyte quality, decrease hyper-androgenism and adjust menstrual cycles ovulation and hirsutism. 39,43 GLUT4, an insulin-sensitive glucose transporter, has a wellknown role in controlling insulin-motivated glucose transport into skeletal muscle and adipose tissue, so that connecting insulin to its receptor leads the translocation of GLUT4 from intracellular storage vesicles to the cellular surface via protein phosphorylation cascade through the phosphatidylinositol-3 kinase/Akt signaling pathway. [44][45][46] An imperfection in glucose transport efficiency, containing GLUT4 expression and function, results in insulin resistance. 47 The benefits of an oral chemically synthesized PI for lowering plasma glucose level, increasing insulin sensitivity and boosting glucose disposal capacity has attracted a lot of attention in experimental animals and humans with impaired glucose metabolism. 42,48,49 There was a significant between-study heterogeneity in the present study which might overshadow the conclusions, however, analysis showed that supplementation was more effective in the short term of less than 12 weeks and at the dosage of less than 1000 mg. One possible explanation is the fact that inositol, as postprandial and nocturnal glycemic control, is more effective in the shorter time and is not effective in the long term due to changes in glucose and fat levels or their mechanisms. [50][51][52] In addition, more sensible positive effect of inositol in individuals suffering PCOS, might be due to higher levels of fat and hyper-androgenism. 53,54 Regarding its greater effectiveness at the age >40, it might be due to its effect on increasing lipoprotein particle size. 43,55,56 Although the overall results of the present study showed that the effect of inositol in lowering BMI was statistically significant, however, the magnitude of this effect is clinically small and may not be to different findings. Fifth, high between-study heterogeneity could be considered as one of the limitations of the study, although the sources of it were identified. Also, participants were of different races, and this may play a role in the effect of inositol, which this results did not consider racial and genetic differences.

| CONCLUSION
This study elucidated the significant reducing effect of inositol on BMI in adults younger than 30 years as well as in individuals with PCOS, although the clinical magnitude of this effect may not be much considerable. However, inositol supplementation could be administered as an adjunct therapy to improve anthropometric indices and glycemic responses.