The pitch is easy to assemble and almost never stated outright: obesity lowers fertility, these drugs treat obesity, therefore they improve the odds of a baby. Each clause is defensible on its own. The conclusion has never been tested, and the closest thing to a test that exists came out the other way.
Two separate questions get folded together here. One is what happens if someone conceives while taking a GLP-1, which is a pregnancy-exposure question answered in the pregnancy and contraception article. The other is whether taking one first makes conception more likely. This page is about the second.
The labeled washout is not symmetric, and the gap is not small
Read on DailyMed on September 15, 2026, the semaglutide labels give a number. Wegovy’s reproductive-potential section instructs that the drug be discontinued at least two months before a patient plans to become pregnant, to account for the long half-life of semaglutide. The Ozempic label says the same thing in different words, citing the long washout period.
The tirzepatide labels contain nothing equivalent. Zepbound’s and Mounjaro’s reproductive-potential sections consist entirely of the oral-contraceptive paragraph — the four-week barrier-method window after starting and after each dose increase — and no pre-pregnancy interval appears anywhere in either document. Both full labels were read to confirm that absence rather than inferred from the section heading.
That is a real and under-discussed difference between two weekly injections that are otherwise cross-shopped constantly, and it is not obviously a difference in the drugs. It is a difference in what each sponsor chose to write, and it means that how long to stop before trying is answered by the molecule in the pen rather than by the class. A planned two-month gap is also a planned interruption, and weight returns after withdrawal — which lands the regain squarely in the months when conception is being attempted.
What the reproductive toxicology actually recorded
Both labels report fertility studies in rats, and both report the same two directions. On the male side, neither found anything: the tirzepatide label states that no effects were observed on sperm morphology, mating, fertility or conception, and the semaglutide label states that no effects were observed on male fertility.
On the female side, both recorded a reduction. Semaglutide increased estrus-cycle length at all dose levels and reduced the number of corpora lutea. Tirzepatide increased the number of females with prolonged diestrus and decreased the mean number of corpora lutea, which in turn decreased implantation sites and viable embryos — again at all dose levels tested. Both sponsors attribute those findings to the pharmacological effect on food consumption and body weight rather than to a direct reproductive action.
That attribution is probably right, and it is also the point. If fewer ovulations are what an energy deficit produces in a rat, then the same mechanism that makes these drugs work is the mechanism the labels are explaining away. It is a caution about conceiving during active weight loss, and it is the opposite of the marketing direction.
The human fertility signal comes from one population
Every positive human fertility result in this class was generated in women with polycystic ovary syndrome, where restored ovulation is the expected mechanism, the trials are small, and conception was measured after the drug had been withdrawn. Those studies, their comparators and their designs are set out in the PCOS article and are not restated here.
A 2026 systematic review that set out specifically to integrate the assisted-reproduction data reached the boundary of that evidence and said so. It found preliminary clinical signals of improved reproductive function in PCOS, noted that animal work shows both beneficial effects on follicular growth and detrimental ones including oxidative stress, granulosa-cell death and uterine inflammation depending on context and dose, and concluded that controlled human research is needed particularly in non-PCOS patients and in IVF settings.[1]
So for the reader without PCOS — which is most people buying a GLP-1 — there is no human fertility evidence for these drugs at all. What exists instead is a literature on preconception weight loss by other means, and it is unexpectedly discouraging.
The randomized test of the premise
FIT-PLESE randomized 379 women with a body-mass index of 30 or above and unexplained infertility, at nine academic centers, to two 16-week preconception programs: an intensive arm targeting 7% weight loss through increased activity, meal replacements and orlistat, or a weight-neutral arm of increased activity alone. Both were then followed by the same standardized treatment — three cycles of ovarian stimulation with intrauterine insemination. The primary outcome was a healthy live birth, defined as a term infant of normal weight without major anomalies.[2]
The intervention worked on everything it was designed to move. Weight fell 6.6 ± 5.4% against 0.3 ± 3.2% (P < 0.001), and metabolic syndrome prevalence fell from 52.8% to 32.2% in the intensive arm while barely shifting in the control arm, from 53.6% to 49.4% (P = 0.003).[2]
It did not move the endpoint. Healthy live births occurred in 23 of 188 intensive participants (12.2%) against 29 of 191 standard participants (15.2%), a rate ratio of 0.81 (95% CI 0.48 to 1.34, P = 0.40). First-trimester pregnancy loss was higher in the intensive arm, at 33.3% against 23.7%, a rate ratio of 1.40 (95% CI 0.79 to 2.50) that did not reach significance. Gastrointestinal side effects were significantly more common on the intensive program. The authors’ own conclusion is the sentence to carry away: improvement in metabolic health may not translate into improved female fecundity.[2]
Neither of those point estimates is statistically significant, and a single trial does not establish harm. What it does establish is that the premise is untested rather than obvious, and that the only randomized attempt to prove it produced numbers pointing the wrong way on both the birth and the loss.
Two syntheses find the same split
A 2024 meta-analysis of randomized trials of preconception weight loss through lifestyle change, medication, or both, in women with overweight or obesity, pooled 16 studies and 3,588 participants for pregnancy, 13 studies and 3,329 for live birth, and 11 studies and 3,248 for miscarriage. Clinical pregnancy rose, at a risk ratio of 1.24 (95% CI 1.07 to 1.44, I² = 59%). Live birth did not, at 1.19 (95% CI 0.97 to 1.45, I² = 69%). Miscarriage overall did not move, at 1.17 (95% CI 0.79 to 1.74) — but among the subgroup undergoing fertility treatment it did, at 1.45 (95% CI 1.07 to 1.96).[3]
More people got pregnant; the same number went home with a baby; and among those in treatment, more pregnancies ended early. The authors state that their findings do not support a one-size-fits-all recommendation for weight loss immediately before conception.[3]
A Cochrane review of preconception lifestyle advice reaches the same place from a different corpus: seven randomized trials, 2,130 participants, and for advice on weight against routine care a live-birth risk ratio of 0.94 (95% CI 0.62 to 1.43, two trials, 707 participants, I² = 68%) on very low quality evidence, alongside a miscarriage risk ratio of 1.50 (95% CI 0.95 to 2.37) and a body-mass index reduction of 1.30 kg/m² (95% CI −1.58 to −1.02). Only one of the seven trials included male partners at all.[4]
The timing question, borrowed from surgery
Bariatric surgery is the only other intervention that produces this scale of weight loss before conception, and its guidelines have long advised postponing pregnancy for 12 to 24 months. A 2026 multicenter cohort stratified 156 pregnancies after metabolic bariatric surgery by interval — 41 conceived under 12 months, 71 between 12 and 24, and 44 after 24. It found no significant differences between the groups in maternal complications, mode of delivery, gestational age at birth, birthweight or Apgar scores, all P > 0.05, with gestational weight gain merely tending lower in the earliest group. Its authors argue for individualizing on postoperative weight stabilization and nutritional status rather than a fixed threshold.[5]
That is a 156-pregnancy retrospective cohort in a surgical population, not a drug trial, and the interval it examined starts at twelve months rather than at two. What transfers is the principle rather than the number: the relevant state is weight stability and nutritional adequacy, not a date on a calendar. How these drugs compare with surgery on every other axis is in the surgery comparison.
The census, and its controls
Stated plainly, so that the absences are not mistaken for oversights. No randomized trial has reported a live birth after GLP-1 exposure in a population without polycystic ovary syndrome. No trial has reported an embryo count, a fertilization rate or an implantation rate after pretreatment with either dominant molecule. No trial has compared conceiving after a two-month washout against conceiving after six.
The controls that make those absences credible rather than assumed: a 2026 systematic review of preconception, pregnancy and lactation exposure across this drug class screened four databases to September 2025 and included 36 studies, none of them a fertility trial in a non-PCOS population.[6] The review written specifically to integrate IVF data names controlled human research in non-PCOS and IVF settings as the outstanding need rather than as a completed literature.[1] And on the male side, both labels’ only fertility data are from rats — neither reports a human study of semen parameters, which is a gap visible in the labels themselves.
A European position statement on women with obesity across reproductive life sits on the other side of the argument and is worth quoting accurately for that reason: it states that weight loss of 5% to 10% over six months improves fertility, and, in the same document, that clinical data on the safety and efficacy of obesity medication during pregnancy or lactation are limited.[7]Six months of gradual loss and sixteen weeks of acute loss before an insemination cycle are not the same exposure, which is one honest way to reconcile that guidance with the trial results above. It is a reconciliation, not a finding.
What this means at an intake form and at a fertility clinic
A cash-pay telehealth service prescribes for weight. It is not running a conception plan, it is not coordinating with a reproductive endocrinologist, and it is not positioned to decide when to stop. Three questions belong to the prescriber holding the chart: which molecule is being used and therefore which washout instruction applies, whether conception will be attempted during active loss or after weight has stabilized, and what contraception is in place in the meantime — because restored ovulation is a documented effect in the one population where fertility has been studied.
Two practical notes for buyers. Most sellers dispense compounded semaglutide or tirzepatide, which is not FDA-approved and which the FDA does not review for safety, efficacy or quality before it is dispensed, so none of the labeled washout language above travels with the vial — see what a compounded vial contains. And how carefully an intake asks about conception plans varies by seller, which is recorded in the provider write-ups against the criteria in the methodology. Sellers that market to women specifically are collected on their own board, and prescribe the same medication as everyone else.
The honest summary is short. These drugs are not fertility treatments, nobody has tested them as one outside polycystic ovary syndrome, and the one randomized trial of the underlying premise found that losing the weight improved the metabolic picture without improving the birth. None of that is a reason to start or stop a medication, and every part of the timing belongs to a prescriber rather than to a checkout page.