Introduction

Vitamin D regulates calcium and phosphate metabolism in the body. The vitamin D receptor and vitamin D metabolising enzymes are expressed in a number of reproductive tissues including the ovary, endometrium, fallopian tube, placenta, hypothalamus and pituitary, suggesting that it may play an important role in human reproduction.1 The Endocrine Society defined vitamin D deficiency as serum 25-hydroxyvitamin D [25(OH)D] of less than 20 ng/mL (50 nmol/L) and vitamin D insufficiency as serum 25(OH)D at 21–29 ng/mL (50–72.5 nmol/L).2 Numerous reports have suggested that low vitamin D status is a global health problem affecting a high proportion of individuals throughout the world.3

Polycystic ovary syndrome (PCOS) is one of the most common reproductive endocrine disorders, affecting 8%–13% of reproductive age women.4 PCOS is associated with different degrees of reproductive and metabolic dysfunctions, including ovulatory dysfunction, hyperandrogenism, obesity, insulin resistance, hyperinsulinaemia and dyslipidaemia. Low vitamin D status has been found very prevalent among women with PCOS in many populations.5 6

A high level of serum anti-Mullerian hormone (AMH), which is exclusively produced by the granulosa cells of the preantral and small antral follicles in adult women, is found in women with PCOS.7 8 The high AMH level is related to the increased number of small antral follicles and increased AMH expression per granulosa cell, and is postulated to decrease the sensitivity of antral follicles to follicle-stimulating hormone thereby resulting in follicular arrest. The presence of a functional vitamin D response element in the promotor region of the AMH gene suggested that AMH signalling can be altered by vitamin D level.9

Vitamin D supplementation was found to increase soluble receptor for advanced glycation-end product levels and result in normalisation of serum AMH suggesting an improvement in folliculogenesis.10 In ovulation induction, serum vitamin D level was an independent predictor of live birth and ovulation in women with PCOS.11 12 In a systemic review and meta-analysis of randomised controlled trials, vitamin D supplementation significantly improved the occurrence of dominant ovarian follicles in women with PCOS, although no significant difference was shown in the frequency of regular menstrual cycles, probably due to the small sample size overall, and some studies were confounded by the use of metformin.13

There is some evidence demonstrating a causal link between low vitamin D levels and manifestation of PCOS by intervention studies. Two meta-analyses13 14 showed no significant improvement in insulin sensitivity, but a more recent meta-analysis and another randomised trial did show significant improvement in insulin sensitivity15 16 after vitamin D supplementation. A recent randomised trial revealed that vitamin D supplementation could reduce insulin resistance and liver enzymes in women with PCOS.17 In non-PCOS adults, one trial suggested that active vitamin D treatment may be useful in improving Homeostatic Model Assessment of beta cell function (HOMA-β) and reducing Homeostatic Model Assessment for Insulin Resistance (HOMA-IR) and risk of type 2 diabetes mellitus.18 A recent meta-analysis revealed that vitamin D supplementation could reduce fasting glucose and increase high density lipoprotein (HDL)-cholesterol in those with cardiovascular diseases.19 Another randomised trial in Saudi Arabic adolescents also suggested that vitamin D supplementation could reduce glucose, triglyceride, systolic blood pressure and incidence of metabolic syndrome, while increase the HDL-cholesterol level.20 Regarding the androgen profile, two meta-analyses14 21 showed no significant change in free androgen after vitamin D supplementation, but a more recent randomised trial16 showed significant reduction in androstenedione and increase in sex-hormone binding globulin (SHBG). The inconsistency among current randomised trials and meta-analyses can be due to the fact that majority of the existing studies had a small sample size,22 variable inclusion criteria and had a wide range of intervention dosage from 400 to 12 000 IU/day and treatment duration.13

Good evidence is still lacking and it is therefore important to fill in this gap by a double-blind randomised clinical trial with an adequately powered sample size to confirm the effect of vitamin D supplementation on improvement of ovulation, in addition to its metabolic benefits.

Objective and hypothesis

The objective of this study is to assess the effect of vitamin D supplementation in addition to letrozole on the ovulation rate and other reproductive, endocrine and metabolic outcomes in women with PCOS. The null hypothesis is that vitamin D supplementation does not improve ovulation and metabolic abnormalities in women with PCOS.

Methods and analysisTrial design

This is a randomised double-blinded and placebo-controlled trial conducted in five reproductive endocrinology clinics in Hong Kong at Queen Mary Hospital, Pamela Youde Nethersole Eastern Hospital, Kwong Wah Hospital, Princess Margaret Hospital and Queen Elizabeth Hospital.

Selection and withdrawal of subjects

The population for the trial will be women with PCOS.

Inclusion criteria

Subjects admitted to the study will fulfil all of the following criteria:

Premenopausal.

Aged 18–40 years.

Irregular long cycles (>35 days).

PCOS according to the Rotterdam criteria (presence of two out of three criteria) with exclusion of other aetiologies of anovulation.23

Oligo-anovulation.

Clinical (including acne, alopecia or hirsutism) or biochemical hyperandrogenism (above laboratory normal of total testosterone, that is >1.7 nmol/L or free androgen index >5).

Presence of 12 or more follicles on ultrasonography measuring 2–9 mm or ovarian volume ≥10 mm3.

Agree for transvaginal or transrectal ultrasound scan.

Exclusion criteria

Subjects should not be recruited in any of the following conditions:

Use of hormonal medication (including contraception) within 3 months prior to study inclusion except the use of a progestogen to induce withdrawal bleeding every 3 months.

History of any medical condition or medications that may predispose to vitamin D sensitivity, altered vitamin D metabolism and/or hypercalcemia, including active tuberculosis or current therapy for tuberculosis, sarcoidosis, history of renal/ureteral stones, parathyroid disease, renal or liver failure or current use of anti-convulsants.

Use of insulin-sensitising drugs, lipid lowering drugs or anti-hypertensives.

Anticipated use of the above medications in the coming 1 year.

Known type 2 diabetes mellitus.

Refusal to join the study.

Abnormal blood calcium level.

For those on supplements, we will ask them to stop their own supplements.

If the participant become pregnant during the study, they will be asked to stop the study drugs. They will be referred for antenatal care accordingly and the pregnancy outcome will be followed up.

Subject withdrawal criteria

Participation in the study is totally voluntary. The subjects can withdraw from the study at any time without giving any reasons and subsequently they will receive standard medical care.

Treatment of subjects

The schedule of enrolment, interventions and assessments is shown in figure 1.

Figure 1

Schedule of enrolment, interventions and assessments.

Baseline (day 2–5)

Transvaginal or transrectal ultrasound scan to assess the presence of polycystic ovary morphology that is, presence of 12 or more follicles on ultrasonography measuring 2– 9 mm or ovarian volume ≥10 mm3.23

Clinical measures include blood pressure, body height, body weight, waist circumference and hip circumference.

Fasting blood for AMH, total testosterone, SHBG, androstenedione, glucose, insulin, HbA1c, lipid profile and 25 (OH)D.

2-hour plasma glucose in a standard 75 g oral glucose tolerance test.

Liver enzymes.

Randomisation

On the day of ultrasound scan, recruited women will be randomly allocated into one of the following two groups: (1) vitamin D and (2) placebo as control group according to a computer-generated randomisation list prepared by a designated research nurse, who is not involved in clinical care and this trial.

Blinding

The women and the physicians will be blinded to the group allocation. Only a dedicated research nurse who is not involved in the patients’ recruitment, follow-up, outcome assessment and data analysis will know the group allocation.

Intervention

The vitamin D group will take the loading dose of Vitamin D3 (Vita-Thrive Cholecalciferol, Enexi Healthcare, UK) 50 000 IU once weekly for 4 weeks, followed by the maintenance dose of 50 000 IU once every 2 weeks for 52 weeks in total, whereas the placebo group will take placebo tablets with the same external appearance manufactured by the same company. This treatment dose was endorsed by the Endocrine Society with no evidence of toxicity.2

Follow-up

The subjects will be asked to chart their cycle length on a standardised menstrual chart, as well as any side effects of treatment.

Follow-up visits will be arranged at the 3rd, 6th, 9th and 12th months after the start of the intervention (vitamin D/placebo). Weekly blood test for progesterone with or without transvaginal ultrasonography which will be performed in those with fertility wish will be conducted at the 3rd, 6th, 9th and 12th months to document ovulation. 25(OH)D, calcium, phosphate and parathyroid hormone will also be measured at follow-up.

The same set of investigations as at baseline will be repeated at 6 and 12 months after commencement of treatment:

Clinical measures including blood pressure, body height, body weight, waist circumference and hip circumference.

Fasting blood for AMH, total testosterone, SHBG, androstenedione, glucose, insulin, HbA1c, lipid profile.

2-hour plasma glucose and insulin in standard 75 g oral glucose tolerance test (fasting glucose at 6 months).

Liver enzymes.

Transvaginal/transrectal ultrasound at 12 months after commencement of treatment.

The subjects will be asked to return the drug bottles together with all unused tablets (active/placebo) to the research nurse at the end of the study to verify drug compliance. Serum vitamin D levels will be measured and serve as a compliance parameter.

Adjunctive hormonal treatments

If they do not have periods 3 months after starting the study drugs, the subjects shall be advised to take progestogen to induce withdrawal bleeding.

In those who do not have regular menstrual cycles (defined as cycle length ≤35 days) or ovulation 6 months after the start of the intervention (vitamin D/placebo), letrozole 2.5 mg daily for 5 days per month will be administered. This is recommended as the first-line pharmacological treatment for anovulation secondary to PCOS.4 Ovarian response will be monitored by transvaginal ultrasound with blood test for serum progesterone level. The dose of letrozole will be increased up to 7.5 mg daily for 5 days if there is no ovulatory response to lower doses of letrozole.

Assessment of outcomes

The primary outcome is presence of ovulation defined as serum progesterone level >10 nmol/L.24

Secondary outcomes include:

Serum 25(OH)D.

Secondary reproductive outcomes including:

Rates of pregnancy (defined by positive urine or serum hCG test) and live birth (defined as delivery of a baby after 22 completed weeks of gestational age, or with birth weight of 500 g if the gestation age is unknown, which shows evidence of life).

Serum AMH levels and antral follicle count.

Androgen indices (as hyperandrogenism is a common phenotypic feature of PCOS):

Metabolic outcomes (as increased risks cardio-metabolic problems such as insulin resistance, dysglycaemia, dyslipidaemia, obesity, hypertension, systemic inflammation, coronary heart disease and fatty liver disease are well recognised in women with PCOS, and yet current data on the effect of vitamin D are inconsistent and of low quality as explained in the Background section):

Body mass index (BMI), waist circumference, hip circumference.

Blood pressure.

Fasting and 2 hour (post 75 g oral glucose tolerance test) glucose and insulin, HbA1c, HOMA-IR.

Lipid profile.

Liver enzymes.

Metabolic syndrome, as defined based on the Joint Interim Statement of the International Diabetes Federation Task Force of Epidemiology and Prevention, National Heart, Lung and Blood Institute American Heart Association, World Heart Federation, International Atherosclerosis Society and International Association for the Study of Obesity. Metabolic syndrome is diagnosed if at least three out of the following five criteria are met: (1) waist circumference ≥80 cm (Chinese women); (2) triglycerides ≥1.7 mmol/L; (3) HDL-cholesterol <1.29 mmol/L; (4) systolic blood pressure ≥130 mm Hg or diastolic blood pressure ≥85 mm Hg and (5) fasting glucose ≥5.6 mmol/L or known history of type 2 diabetes mellitus.25

Consistency of case management

The same standardised study protocol will be adopted in all the study centres. The clinicians who manage the women in the participating hospitals had all been trained in the same unit at Queen Mary Hospital, and are now adopting the same clinical management protocols in their respective units. The same lot of vitamin D/placebo tablets will be distributed to the study centres for this study. All blood tests involved in this study will be performed in a centralised laboratory. Regular and frequent communication among the participating centres will also ensure dissemination of updated information on recruitment and safety issues.

Patient and public involvement

Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.

StatisticsSample size calculation

The ovulation rate in Chinese women after taking letrozole is about 60%.26 Assuming the ovulation rate is increased to 80% after taking vitamin D, 91 women in each group will achieve at least 80% power to detect the difference of 20% between the groups with type I error of 0.05. Accounting for 20% dropouts, then 110 women per arm and 220 women are needed in total.

Data analysis

All statistical analyses will be performed using IBM SPSS software using intention-to-treat and per protocol analyses. Demographic features of the study groups will be compared. The primary outcome is the ovulation rate in each treatment group. Two-sample proportion test will be performed to estimate the group difference. Point estimate along with corresponding CI will be reported for the difference. There is no interim analysis. Chi-square test and Fisher’s exact test will be used for categorical variables. Analysis of variance and Mann-Whitney U test will be used to compare the continuous variables between groups depending on the normality. P values of <0.05 will be considered clinically significant. We will perform subgroup analyses based on the women’s age, BMI, antral follicle count and PCOS phenotype. The statistician will perform the analysis prior to unblinding.

Ethics and dissemination

Ethics approval was sought from the Institutional Review Board of the University of Hong Kong/Hospital Authority Hong Kong West Cluster, Hospital Authority Hong Kong East Cluster Research Ethics Committee, Hospital Authority Kowloon Central Cluster Research Ethics Committee and Hospital Authority Kowloon West Cluster Research Ethics Committee. All participants will provide written informed consent before randomisation. The research findings from this study will be submitted to scientific conferences and peer-reviewed journals for publication so as to disseminate the results to other researchers and clinicians working in the field.

Discussion

Ovulation is an important outcome measure as anovulatory infertility is one of the most common reasons for women with PCOS to consult the healthcare service. If it can be demonstrated that vitamin D supplementation can improve ovulation and other endocrine and metabolic outcomes, it will be a very valuable and cheap therapeutic intervention to achieve reproductive and long-term health benefits of women with PCOS. The live birth rate is not chosen as the primary outcome measure as women with PCOS who have no fertility wish can also be recruited.

The dosage of vitamin D used in the present trial is based on the Endocrine Society recommendation that adults who are vitamin D deficient be treated with 50 000 IU of vitamin D2 or vitamin D3 once a week for 8 weeks or its equivalent of 6000 IU of vitamin D2 or vitamin D3 daily to achieve a blood level of 25(OH)D above 30 ng/mL, followed by maintenance therapy of 1500–2000 IU/day.2 The dosage or cut-off required for non-skeletal health benefits is unknown. We used a shorter duration of 50 000 IU/week for 4 weeks instead of 8 weeks at the initial phase because although our own data showed a high prevalence of vitamin D deficiency in reproductive age women, not all recruited women were vitamin D deficient at baseline. The maintenance dose of 50 000 IU every 2 weeks used in our study is below the upper tolerable intake level with no evidence of toxicity.2

Trial status

The date of recruiting the first subject was on 26 January 2021 and 110 women were recruited up to the writing of this protocol paper. The planned study end date is 30 June 2025.