This is the final post in a series of posts on popular diets and their effects on PCOS symptoms. So far, we have covered ketogenic diets, the Mediterranean diet, low glycemic index/load diets, and bean-based diets.

Given the rising popularity of plant-based eating, I thought it was important to include this way of eating. Unfortunately, the literature is seriously lacking on the topic. For nutrition interventions, we can best trust the results of randomized control trials. In these trials, a large group of people are recruited. Then half of the group is randomly assigned to the diet of interest and the other half are assigned to the control group (usually just continuing to eat as they had previously). For a plant-based diet intervention, we’d ideally recruit a large group of meat eaters, and then randomly assign half of them to switch to a plant-based diet. We’d measure important metrics before and after the intervention period to see if anything changed in those people who switched diets. We’d also track the same metrics in the control group, in case there are factors that affect both groups that also affect outcomes. For example, maybe the intervention spans the holidays and both groups end up eating more food than normal. We can net out the effects of these external factors by comparing the changes across the two groups.

All of that said… there are no full randomized control trials considering the use of plant-based diets for PCOS! Just one lonely randomized control trial feasibility trial, discussed below.

The first two studies discussed below can only detect correlation, not causation. It is possible that other more impactful things - a penchant for eating vegetables, exercise routines, avoidance of processed foods, etc. - are correlated with being a vegan or vegetarian, and these other things are actually causing the observed differences in health outcomes.

So as we go through the existing studies, please keep this limitation in mind.

plant-based vs. western diets

The first study recruited 225 women with PCOS and 345 women without PCOS to serve as the control group (Shahdadian et al. 2019). All participants were recruited from the Isfahan Fertility and Infertility Center or the Shahid Beheshti Hospital of Obstetrics and Gynecology in Iran.

All participants were interviewed by a dietician to establish how often they ate foods from 36 food categories. Diets were then given three scores based on composition:

A Western Diet Score where consumption of foods like processed meats, fast foods, salty snacks, sugar, sweets, and desserts increased the participant’s score
A Plant-Based Diet Score where consumption of things like vegetables, nuts, and fruits increased the participant’s score
A Mixed Diet Score where things like refined grains and red meats increased the participant’s score

Note that none of these scores strictly identified a participant as a vegetarian or vegan.

After calculating each participant’s three scores, they ranked participants by score for each diet. They then compared the prevalence of patients with PCOS to the prevalence of patients without PCOS in each third of the rankings.

When just considering the Western Diet rankings, about 40% of those with PCOS fell into the top third of scores, while only about 27% were in the lowest third of scores. Conversely, 29% of those without PCOS were in the top third of scores and 37% in the lowest third of scores.

These data suggest that a Western Diet is correlated with prevalence of PCOS.

But… when we look at the Plant-Based Diet scores, 40% of women with PCOS are in the top third of scores, 25% in the middle third, and 34% in the lowest third.

There is no clear correlation between plant-based eating and prevalence of PCOS.

Lastly, when looking at the Mixed Diet scores, 45% of women with PCOS are in the top third of scores, 25% in the middle and 30% in the lowest third. Again, there’s no clear correlation between this way of eating and prevalence of PCOS.

In all three categories, PCOS prevalence was highest in the top third of scores, which doesn’t really help us figure out if plant-based diets are helpful for PCOS.

A comparison of biomarkers for Vegetarians vs. non-vegetarians

This next study recruited 200 women with PCOS and 400 without PCOS from the All India Institute of Medical Sciences located in New Delhi, India and the Sheri-Kashmir Institute of Medical Sciences located in Srinagar, India (Ganie et al. 2019).

Those participants from New Delhi were all vegetarians, and those from Srinagar were all meat-eaters. It is not clear why participants were recruited in this manner, and this study design is a bit problematic. Any difference found between vegetarians and non-vegetarians could be driven by the participants living in two different cities with access to different medical care, eating different kinds of foods beyond just being vegetarians or meat-eaters, or exposure to different levels or kinds of pollution.

All participants completed a food frequency questionnaire and provided information on what they had eaten during the three days prior.

Unlike the previous study, this study’s vegetarian group did not consume any meat or eggs.

The researchers then collected a wide range of biomarkers relevant to PCOS, and a comparison of these markers across women with and without PCOS is shown in the table below.

A summary of results from Ganie et al. (2019) comparing biomarkers for vegetarian and non-vegetarian women with PCOS. “No difference” indicates no statistically significant difference between the two groups. For those markers for which a statistically significant difference was observed, the group with the “better” level is reported. For some markers (e.g., adiponectin), a higher level is preferable while for others (e.g., hs-CRP), a lower level is preferable.

Interestingly, for two of the primary indicators of PCOS - menstrual cycles per year and the Ferriman-Gallway score for hirsutism (a more masculine pattern of facial and body hair usually resulting from excess androgens), no difference is observed. Similarly, there is no difference between groups for LH, FSH, or prolactin, three reproductive hormones.

They do observe that vegetarians have 16.7% lower total testosterone than non-vegetarians. However, given the lack of difference in Ferriman-Gallway score, I suspect that there may not be differences in free testosterone, which drives the primary symptoms of PCOS (acne, hirsutism, male pattern baldness). The researchers failed to measure free testosterone or sex hormone binding globulin, which binds to total testosterone and reduces free testosterone.

While no differences are observed across groups for blood pressure, vegetarians have 10.5% lower LDL cholesterol and 20.6% lower triglycerides, suggesting that vegetarian diets reduce risk of heart disease. However, non-vegetarians have lower hs-CRP, a marker of inflammation associated with coronary artery disease, so it is not clear which diet is actually better for cardiovascular health.

In terms of insulin resistance, there are no differences between groups for fasting glucose, fasting insulin, or HOMA-IR (a measure of insulin resistance), but vegetarians do have higher QUICKI (a measure of insulin sensitivity. If you’ve read the post on bean diets you might remember that beans, lentils, and split peas improve insulin sensitivity. It’s possible that the increased insulin sensitivity found here is due to consuming more legumes, and not necessarily due to avoiding meat.

Lastly, when considering several markers of inflammatory and anti-inflammatory compounds, non-vegetarians appear to have the advantage. On average, levels of resistin (linked to obesity and diabetes) is 42.3% lower for non-vegetarians, and adiponectin, an anti-inflammatory hormone, is about twice as high for non-vegetarians than vegetarians.

From these results, it is difficult to distinguish a clear “winning” diet, and as discussed above, these results come with some caveats. First, the respective groups might be consuming certain foods or engaging in certain behaviors that are bigger drives of these biomarkers than eating or avoiding meat. Increased consumption of legumes by vegetarians would be one example. Someone who eats meat and legumes would still experience the benefits of consuming legumes.

Second, the two diet groups also lived in different cities. New Delhi is a larger and more polluted city than Srinagar which may explain the differences in inflammatory markers better than differences in diet.

Low glycemic index vegan vs low-calorie diets for weight loss

Our last study recruited 18 overweight or obese women with PCOS between the ages of 18 and 35 who had been trying to conceive for at least 6 months (Turner-McGrievy et al. 2014). Participants were assigned to follow either a low glycemic index vegan diet or a low calorie (either 1200 or 1500 calories per day depending on starting weight) diet.

Participants followed their respective interventions for 6 months, with assessments taken at 0, 3 and 6 months.

At the 3-month mark, the vegan group had lost 1.8% of their body weight, while the low calorie group had lost 0% of their body weight, on average. However, by the 6-month mark, both groups weighed, on average, the same as they had at the start of the intervention.

The study also assessed participants’ emotional health, body hair concerns, infertility concerns, and menstrual regularity. No differences were found between those assigned to the vegan diet and those assigned to the low calorie diet.

Take home points

More research is needed on the use of a plant-based diet for PCOS. There is need for a large, randomized control trial that examines both vegetarian and vegan diets.

From the available evidence, it is unclear if a plant-based diet causes bigger improvements in important health metrics for women with PCOS compared to other options.

There is some evidence that vegetarian diets are correlated with lower LDL cholesterol and triglycerides and improved insulin sensitivity. It is unclear if this is due to the absence of meat consumption or the increased consumption of things like legumes and vegetables.

references

Ganie1 MO, Sahar T, Rashid A, Wani IA, Nisar S, Sathyapalan T, Vishnubhatla S, Ramakrishnan L, Parvez T and Geer I. 2019. Comparative Evaluation of Biomarkers of Inflammation Among Indian Women With Polycystic Ovary Syndrome (PCOS) Consuming Vegetarian vs. Non-vegetarian Diet Frontiers in Endocrinology 10 https://doi.org/10.3389/fendo.2019.00699

Turner-McGrievya GM, Davidson CR, Wingard EE, Billings DL. 2014. Low glycemic index vegan or low-calorie weight loss diets for women with polycystic ovary syndrome: a randomized controlled
feasibility study. Nutrition research 34:552–558 http://dx.doi.org/10.1016/j.nutres.2014.04.011

Shahdadian F, Ghiasvand R, Abbasi B, Feizi A, Saneei P, and Shahshahan Z. 2019. Association between major dietary patterns and polycystic ovary syndrome: Evidence from a case-control study. Applied Physiology, Nutrition, and Metabolism. 44:52–58 dx.doi.org/10.1139/apnm-2018-0145