DHEA for Female Infertility

DHEA is widely taken by women with poor ovarian reserve before IVF, but the trials say it doesn't raise the chance of a baby. A 2024 Cochrane review of 28 randomised trials, nearly all in poor responders, found DHEA likely makes little to no difference to live birth or ongoing pregnancy (Naik 2024).

The largest double-blind trial, in 821 poor responders at nine fertility centres, found live birth in 8.8% of women on DHEA and 9.0% on placebo (Wang 2022). ESHRE's 2019 guideline, since updated in 2025, concluded that DHEA is not recommended for improving IVF outcomes (ESHRE 2020).

The theory is that androgens made from DHEA make follicles respond better to stimulation drugs, so more eggs are collected and the chance of pregnancy rises (Naik 2024). DHEA does raise testosterone, by about 28 ng/dL on average across trials (Li 2020).

Among randomised trials, DHEA increased antral follicle count by about 1.2 (Zhang 2023), and a 2025 meta-analysis found it increased eggs collected by about 0.6 in women with diminished ovarian reserve (Conforti 2025).

A fraction of an extra egg is not a baby. In the same analysis of randomised trials that found more follicles, eggs retrieved, clinical pregnancy and live birth did not differ (Zhang 2023).

The outcome that matters is live birth, and on that the best evidence is null.

Most trials used 75 mg of DHEA a day, started at least 12 weeks before stimulation (Kara 2014; Narkwichean 2017) or before and during the IVF cycle (Wiser 2010). The largest trial gave DHEA or placebo for 4-12 weeks before IVF (Wang 2022).

No dose has been shown to improve live birth, so there is no evidence-based dose to recommend.

The 2024 Cochrane review pooled 9 trials (1,433 women) for live birth or ongoing pregnancy: OR 1.30, 95% CI 0.95 to 1.76, moderate certainty. For a woman with a 12% chance on placebo, that means a likely rate between 12% and 20%. DHEA likely also made little to no difference to clinical pregnancy and did not reduce miscarriage (Naik 2024).

In the largest double-blind trial, 821 poor responders took DHEA or placebo before IVF: 36 of 410 (8.8%) versus 37 of 411 (9.0%) had a live birth (relative risk 0.98, 95% CI 0.63 to 1.51), with no difference in eggs retrieved or pregnancy loss (Wang 2022). A meta-analysis limited to randomised trials also found no difference in live birth (Zhang 2023).

An early trial in 33 poor responders without a placebo reported live birth in 23.1% with DHEA versus 4.0% without (P = 0.05) (Wiser 2010). A 2023 meta-analysis mixing randomised and observational studies reported higher clinical and ongoing pregnancy rates but no difference in live birth (Wang 2023), and a network meta-analysis of 16 randomised trials ranked DHEA well for clinical pregnancy (Zhu 2023).

Wiser 2010 was small and unblinded, and Wang 2023 mixed in observational studies. The network meta-analysis used randomised trials, but its clinical pregnancy signal has not carried through to live birth in the Cochrane review or the 821-woman blinded trial, and the double-blind pilot of 60 women was null (Narkwichean 2017).

In 38 women with premature ovarian insufficiency, 12 months of DHEA at 25 mg three times a day did not restore regular periods or change AMH, FSH or antral follicle count (Wong 2018).

In the Cochrane review, testosterone pre-treatment likely improved live birth (OR 2.53, 95% CI 1.61 to 3.99) (Naik 2024), though ESHRE's 2019 guideline did not recommend testosterone either (ESHRE 2020). See our female infertility page and folate for female infertility for preconception basics.

Data on adverse events in the IVF trials were sparse, and the events reported were minor (Naik 2024). In other trials, side effects were mainly androgenic, such as acne (Crosbie 2007), and DHEA may lower HDL cholesterol in women (Qin 2020).

The effect of DHEA on multiple pregnancy is very uncertain (Naik 2024), and no safety data exist for taking it once pregnant.

In the trials, DHEA was taken before and sometimes during stimulation with standard IVF drugs (Wiser 2010), but interactions were not formally studied. Because DHEA is converted into testosterone and estrogen, it changes the hormone picture your clinic monitors (Li 2020; Zhu 2021).