Poster Information, Presented at the 2011 ASRM Conference

A conundrum: wheat and gluten avoidance and its implication with endometriosis patients

Author: D. Shepperson Mills
The Endometriosis and Fertility Clinic, London, United Kingdom.

Objective

To determine if a relationship between endometriosis symptomatology and the ingestion of wheat and/or gluten exists. The objective is to show that for women with endometriosis the link is relevant, and that removing wheat and /or gluten from the diet may have beneficial effects.

We also wished to record whether there was some similarity of symptoms if people re-introduced wheat to their diet

Design

This is a retrospective analysis of data from a cohort of 363 laparoscopically diagnosed endometriosis patients, of reproductive age, using Chi-squared tests. The cohort were self selecting in that they had to have attended the Endometriosis and Fertility Clinic and have completed at least three MYMOP (Measure Yourself Medical Outcome Profile) questionnaires. The women were then split into three groups based on their answers to the question regarding the avoidance of wheat.

Materials and Method

A few women attended the Endometriosis and Fertility Clinic having researched our work and had already avoided wheat. Some, after consultation and the appropriate tests, may be requested to undergo a one month exclusion diet cutting out, wheat or wheat and gluten. Group A would be those women who had been eating wheat and were advised to exclude wheat; Group B would be the cohort who had been excluding wheat prior to their first MYMOP; and Group C would include those who did not report having excluded wheat.

The Clinic-Aid database was searched and 363 women fell into the selective category, i.e. diagnosed with endometriosis; had attended at least three consultations; and had at least three completed MYMOP questionnaires. Further analysis put 204 of these into Group A; 114 into Group B; and 45 into Group C. Because of the different size of groups and the fact that there are three groups a Chi squared test was chosen to see if there were significant differences in the initial to final scores.

MYMOP Medical Audit

Measure Yourself Medical Outcome Profile is a medical audit facility devised by Dr Charlotte Paterson for the Medical Research Council UK. MYMOP shows promise as an outcome measure for primary care and for complimentary treatment. It is a patient generated measure within the consultation and helps the practitioner to be more patient centred. Practitioners find that MYMOP is practical and applicable to all patients with symptoms and that its use increased their awareness of the patient's priorities.

MYMOP aims to measure the outcomes that the patient considers the most important. It uses their own words, which avoids conflicts over diagnosis. They choose two symptoms that are important to them and one activity that is impacted by their condition and score then on a seven point scale, with 0 being as well as can be, to 6 as bad as can be. They also score their sense of wellbeing on the same scale how do you feel in yourself. These scores can be combined to give an indication of the impact their dis-ease has on their body and can display how this improves over time.

The MYMOP questionnaires have been recorded on our Clinic-Aid database, the people selected for this study based on the criteria above and the information exported anonymously to a spreadsheet where the appropriate analysis has been performed, as reported. Informed consent has been signed for the use of blinded data.

According to Dr Paterson's analysis of MYMOP a 0.8 improvement in scores demonstrates a statistically significant improvement

Results

Patients who avoided wheat displayed a greater improvement in wellbeing than those who did not avoid wheat. The initial mean scores were for Group A was 4.19, Group B was 4.02, and Group C was 4.02. Final mean scores were Group A 2.67, Group B 2.48 and Group C 2.97.

Figure 1

From the above graph it can be seen that all three groups show a significant benefit in their MYMOP scores, following nutritional counselling. However, the cohort in Group C, who do not avoid wheat, do not achieve the same reductions in scores over the three months, as do those in Groups A and B.

The initial scores were tested, using Chi squared test, showing that there was no significant difference between the initial scores in each of the groups at the outset. The difference in scores between the initial MYMOP score and the third MYMOP score was examined and broken down into ranges so that we could use the chi- squared test. The values -3 to 0, 0 to 1, 1 to 1.8, 1.8 to 2.4, 2.4 to 3, and 3 to 6, were used to give reasonable numbers in each sector as required by the Chi-squared test. This gave us the results in Table 1.

Table 1 (showing the breakdown of the differences)

This gave us 10 degrees of freedom, and a Chi-squared score of 20.29, which shows 0.05>P>0.02, or at least 95% confidence limits.

Figure 2

A number of people in each group failed to improve their scores between the initial score and the third score. Most of these women reported adverse external influences which impacted on their health, several had surgical procedures linked with their condition, some had external stressors to their lifestyle (partnership breakdowns, family bereavements, work pressures), some had eaten wheat and had severe adverse reactions, and a small number failed to report any reason for the increase in score. We can see the individual results in the following charts, one for each group. The light blue line denotes the initial MYMOP score for an individual and the third MYMOP score is shown in red, the purple line indicates the 0.8 reduction in score that denotes a significant improvement in the health of the patient. It can be seen that the patients in Group C fail to achieve the large improvements that forty-four patients gained from Groups A and B who do avoid wheat.

Figure 3

Figure 4

Figure 5

We also reviewed the individual responses for those in Groups A and B looking at the variety of symptoms that these women reported if they re-introduced wheat to their diet. In this study we examined the responses from all MYMOP questionnaires not just the first few results. As people use their own words in the MYMOP questionnaires a certain amount of liberty was used to flag the symptoms, for example to register a tick in the tiredness symptom a mention of fatigue, exhaustion or tiredness was accepted, similarly for the bloating symptom, bloats, swells, bloating or bloated was accepted. For some people the symptoms were consistently found in each MYMOP, while others had different symptoms in some of the MYMOPs. The other symptom was ticked if the women reported a symptom not in the list, there were a few unique symptoms which were felt unworthy of reporting.

Figure 6

Figure 6 above, shows the number of women reporting each symptom in the list.

Figure 7

Figure 7 shows the number of symptoms that where given by each of the women who reported a problem when they introduced wheat to their diet.

Conclusion

Nutritional Therapy can play an important role to the wellbeing of women who suffer from endometriosis. It has been shown to make a significant improvement to the health of a good majority of the women.

From the results it can be seen that there is a link between wheat and endometriosis in a significant number of women and that to make highly significant improvements in these women's health excluding wheat and gluten from their diet would be beneficial. A large number of those who avoided wheat, report adverse reactions when wheat is re-introduced into their diet. Adding to the body of evidence supporting the premise.

For the 95 women (26.17%) whose scores dropped more than 2.4 points, 90 came from Groups A and B (41.86%), and 5 came from Group C (11.11%), hence there is a good chance of getting much better, using nutritional therapy, if the patient avoids wheat for a period of at least two months.

Further research is needed to explain the linkage and understand the mechanisms involved.