Vitamin D before IVF raised blood levels and left live birth rates unchanged
An 876-participant double-blind trial across 24 Chinese fertility centres nearly doubled serum 25-OHD in women with PCOS. Live births were 52.0% against 50.2% on placebo.
Vitamin D deficiency is common in women with polycystic ovary syndrome, and observational studies have repeatedly linked low vitamin D status to poorer fertility treatment outcomes. Supplements are inexpensive, widely available and easy to justify on that reasoning. A randomised trial published in The BMJ tested the reasoning directly, and it did not hold.
The trial
The multicentre, double-blind, placebo-controlled trial enrolled 876 participants with polycystic ovary syndrome undergoing in vitro fertilisation at 24 fertility centres in China [s1]. Participants were randomised 1:1 to vitamin D 4,000 IU a day or placebo, taken before IVF for up to 90 days until the trigger day [s1]. The primary outcome was live birth after the first embryo transfer [s1].
Of 876 randomised, 865 were included in the modified intention-to-treat analysis: 435 in the vitamin D group and 430 in the placebo group [s1].
The supplement worked; the outcome did not move
Baseline mean serum 25-hydroxyvitamin D was 16.5 ± 7.2 ng/mL in the vitamin D group and 16.1 ± 6.7 ng/mL in the placebo group — low by most laboratory reference ranges [s1].
By the day of triggering, serum 25-OHD was 32.3 ± 11.2 ng/mL in the vitamin D group against 18.2 ± 7.6 ng/mL in the placebo group, an adjusted mean difference of 13.6 ng/mL (95% CI 10.9 to 16.3) [s1].
This is the part of a supplement trial that most often fails, and here it did not. The intervention roughly doubled circulating vitamin D and moved the average participant from deficiency into the sufficient range.
Live births occurred in 226 of the vitamin D group (52.0%) and 216 of the placebo group (50.2%), an adjusted risk ratio of 1.03 (95% CI 0.91 to 1.18) [s1]. The confidence interval includes both a small harm and a modest benefit, but the point estimate sits essentially at no difference.
Severe ovarian hyperstimulation syndrome — the most consequential complication of the process — occurred in three participants on vitamin D and six on placebo, an adjusted risk difference of −0.7% (95% CI −2.0% to 0.6%) [s1]. With nine events in total, that comparison is uninformative in either direction.
The authors' conclusion is stated plainly: although vitamin D 4,000 IU a day for up to 90 days increases serum 25-OHD, this does not translate into improved live birth rates after the first transfer in patients with polycystic ovary syndrome [s1]. The trial was registered as NCT04082650 [s1].
The journal published an accompanying editorial framed around vitamin D supplementation and fertility success [s2].
Why this design settles more than the observational literature
The association between low vitamin D and poor fertility outcomes is real in observational data. What that data cannot do is establish direction. Vitamin D status tracks with time spent outdoors, adiposity, general health, diet quality and socioeconomic position — all of which independently predict IVF success. A randomised comparison breaks that link by assigning the exposure at random.
The trial's negative result therefore does more than fail to find a benefit. Combined with the confirmed biochemical response [s1], it substantially weakens the causal interpretation of the observational association, at least for this population and this outcome.
Two limits are worth stating. The trial measured live birth after the first embryo transfer, not cumulative live birth across all transfers from a stimulation cycle [s1]. And it enrolled women with polycystic ovary syndrome specifically, at centres in one country [s1]; whether the same null holds in other infertility diagnoses is untested here.
The contrast with the small-trial literature
A useful comparison arrived a week earlier. A systematic review and meta-analysis in Frontiers in Medicine pooled 20 trials of vitamin D supplementation in gestational diabetes, involving 1,737 patients in total, from Chinese and English databases searched to September 2024 [s3].
It reported reductions in fasting glucose (standardised mean difference −1.01, p = 0.0002) and two-hour postprandial glucose (SMD −0.89, p = 0.0002), and large reductions in a long list of pregnancy outcomes: caesarean delivery (RR 0.68), preterm birth (RR 0.28), fetal distress (RR 0.17, p = 0.004), neonatal asphyxia (RR 0.22, p = 0.006), macrosomia (RR 0.34, p = 0.001) and neonatal hyperbilirubinaemia (RR 0.49, p = 0.001) [s3]. No significant differences appeared for excess amniotic fluid (RR 0.46, p = 0.10) or pre-eclampsia (RR 0.60, p = 0.38) [s3].
Those are not modest effects. A relative risk of 0.17 for fetal distress would be one of the largest treatment effects in obstetrics. The authors themselves counsel caution, citing substantial heterogeneity and methodological limitations in the included trials, and describe their results as exploratory and in need of confirmation [s3].
The two papers illustrate a pattern familiar across nutrition research: many small trials pooled together produce large effects; one large, adequately powered, double-blind trial produces none. The meta-analysis pooled 20 trials that shared 1,737 patients between them [s3]; the BMJ trial analysed 865 in a single protocol [s1]. Small trials with positive results are also more likely to be published than small trials without them.
What to watch
The open questions the BMJ trial leaves are specific: whether correcting deficiency earlier than 90 days before treatment changes anything, whether cumulative live birth across all transfers differs, and whether any effect exists in women with more severe deficiency than this cohort's average of about 16 ng/mL [s1]. None of those is answered, and none should be assumed.
This article describes research findings and is not guidance about taking any supplement.
Sources
- [s1] Vitamin D supplementation before in vitro fertilisation in women with polycystic ovary syndrome: multicentre, double blind, placebo controlled, randomised clinical trial. BMJ, 17 February 2026. https://doi.org/10.1136/bmj-2025-087438
- [s2] Vitamin D supplementation for fertility success. BMJ, 25 February 2026. https://doi.org/10.1136/bmj.s331
- [s3] Effects of vitamin D supplementation on glucose metabolism and pregnancy outcomes in GDM: a systematic review and meta-analysis. Frontiers in Medicine, 9 February 2026. https://doi.org/10.3389/fmed.2026.1743776
Sources
- Vitamin D supplementation before in vitro fertilisation in women with polycystic ovary syndrome: multicentre, double blind, placebo controlled, randomised clinical trial — BMJ , February 17, 2026
- Vitamin D supplementation for fertility success — BMJ , February 25, 2026
- Effects of vitamin D supplementation on glucose metabolism and pregnancy outcomes in GDM: a systematic review and meta-analysis — Frontiers in Medicine , February 9, 2026
More on
A genetics analysis ties screen time and mood swings to six gynecologic conditions
Using genetic variants as instruments, researchers found exercise and wellbeing lowered risk of PCOS, endometriosis and menstrual disorders, while screen time and mood swings raised it.
What actually manages PCOS, according to the international guideline
The 2023 global guideline puts lifestyle first for everyone, letrozole first for fertility, and the pill for cycles and excess hair — while rating much of its own evidence low to moderate quality.
What egg freezing actually delivers: age at freezing and egg count decide the odds
In two large clinics, the live birth rate per woman who returned to use her eggs was around 39-50% under 35 or 38, but far lower for older women — and most women who froze never came back to use them.
How much does fertility really fall with age? Less sharply than the scary numbers
In modern couples timing intercourse, an estimated 18% of women aged 35-39 failed to conceive within a year — not one in three. The most alarming figures come from models, not from watching real couples try.