Intrauterine hCG Does Not Improve IVF Live Birth Rates

The proposed benefit did not translate into better outcomes

Human chorionic gonadotropin, or hCG, has a central role in early pregnancy and embryo–endometrium signaling. That biology created a plausible rationale for placing hCG inside the uterus shortly before embryo transfer: perhaps local exposure would make the endometrium more receptive, improve implantation and ultimately increase the chance of a live birth.

The new meta-analysis found that this additional procedure did not improve the outcomes that matter to patients. Intrauterine hCG before transfer produced no benefit in live birth and no benefit in clinical pregnancy when compared with embryo transfer without intrauterine hCG. Those concordant findings are important. A treatment that does not increase either the intermediate pregnancy outcome or the definitive live-birth outcome lacks evidence of clinical value.

The result also helps resolve uncertainty left by earlier aggregate-data reviews. A 2018 Cochrane review included 17 randomized trials involving 4,751 women and found the evidence difficult to interpret because of low or very low certainty, substantial variation among studies and signals of benefit confined to selected analyses. The newer work revisited the question using participant-level rather than publication-level data.

This changes the counseling emphasis. The question is no longer whether a biologically plausible intervention might help implantation in theory. It is whether adding an intrauterine procedure before transfer measurably increases the probability of taking home a baby. The individual-patient-data evidence says it does not.

Why individual-patient data matter

The analysis synthesized randomized comparisons of patients undergoing in vitro fertilization or intracytoplasmic sperm injection followed by embryo transfer. The intervention was intrauterine hCG administered before transfer; the comparator was transfer without that intervention, including control or placebo care as defined by the contributing trial.

Rather than relying only on summary results reported in each publication, the investigators assembled and harmonized records for individual trial participants. This approach can apply consistent outcome definitions, check the integrity of trial data and evaluate whether treatment effects differ according to participant or treatment characteristics. It is particularly useful when the literature contains small trials, inconsistent reporting or subgroup claims that cannot be tested reliably from published averages.

Live birth was the decisive outcome. Clinical pregnancy was also assessed, and it likewise did not improve. The absence of benefit across both endpoints argues against interpreting a transient implantation or biochemical-pregnancy signal as proof that the procedure works. For fertility interventions, changes in surrogate outcomes can be clinically misleading if they do not persist through pregnancy and delivery.

Because the evidence arose from randomized comparisons, it is better suited to causal assessment than an observational study of clinic outcomes would be. The meta-analysis nevertheless remains a synthesis of the available trials, not a new randomized trial. Its reliability depends on the design, execution and completeness of the contributing studies and on which investigators were able to provide usable participant-level records.

What the evidence supports in IVF practice

Routine use is not supported. Clinics evaluating IVF add-ons should prioritize live birth, cumulative live birth and harms rather than biological rationale, implantation markers or uncontrolled before-and-after experience. On that standard, intrauterine hCG before embryo transfer does not demonstrate added value.

For counseling, the practical message can be direct: current participant-level randomized evidence has not shown that the procedure increases the chance of clinical pregnancy or live birth. That wording avoids claiming that benefit is impossible under every future protocol while making clear that patients should not be asked to pay, accept inconvenience or undergo another uterine intervention on the expectation of improved success.

The finding is also relevant to consent and pricing. Labeling something an “add-on” does not reduce the need for evidence, particularly when patients are making time-sensitive decisions under financial and emotional pressure. If a clinic continues to use intrauterine hCG in a research setting, the investigational status and uncertainty should be explicit, and outcomes should be evaluated under an appropriately designed protocol.

The meta-analysis does not challenge established uses of hCG elsewhere in fertility treatment. It addresses local administration into the uterine cavity before embryo transfer. Conclusions about that procedure should not be extended to the use of systemic hCG for final oocyte maturation or other recognized components of assisted reproduction.

Nor does the result imply that embryo transfer technique is unimportant. Transfer quality, embryo characteristics, uterine factors and laboratory performance remain relevant to IVF outcomes. The narrower conclusion is that adding intrauterine hCG immediately before transfer has not improved live birth or clinical pregnancy in the randomized evidence synthesized here.

Uncertainty that still needs attention

Individual-patient-data meta-analysis can strengthen a review, but it cannot eliminate weaknesses in the original trials. Protocols may differ in hCG amount, timing, embryo stage, fresh versus frozen transfer, patient prognosis and control procedures. Small contributing trials may also have limited power to identify uncommon adverse events.

Availability of participant-level records is another consideration. If data could not be obtained from every eligible randomized trial, the resulting cohort may not represent the entire published evidence base. Harmonizing outcomes across trials can reduce inconsistency, but it cannot fully correct differences in how pregnancy, live birth or follow-up were originally documented.

Claims about subgroups require restraint. A lack of overall benefit does not mathematically exclude a small effect in a narrowly defined population, but subgroup findings are credible only when prespecified, adequately powered and consistent across studies. The available evidence does not establish a patient group for whom intrauterine hCG should be offered as effective care.

Longer-term and less commonly reported outcomes also deserve attention, including cumulative live birth across all embryos from a retrieval, miscarriage, ectopic pregnancy, multiple pregnancy, procedure-related discomfort and cost. None of these considerations, however, reverses the central practice conclusion: an add-on without improvement in live birth or clinical pregnancy should not be presented as a way to increase IVF success.

Questions clinicians ask

Should intrauterine hCG remain on an IVF add-on menu?

The evidence does not support offering it as a routine success-enhancing add-on. If it is studied further, patients should be told that the individual-patient-data meta-analysis found no improvement in clinical pregnancy or live birth and that participation is research rather than established care.

Does this finding apply to hCG used for triggering ovulation?

No. The analysis concerns hCG placed inside the uterus before embryo transfer, not systemic hCG used for final oocyte maturation or ovulation triggering. Clinicians should distinguish the route, timing and purpose clearly so that patients do not discontinue other prescribed parts of an IVF protocol.

Could a specific dose, transfer stage or patient subgroup still benefit?

A benefit in every conceivable subgroup cannot be excluded, but no clinically actionable subgroup benefit has been established. Dose- or stage-specific signals from small studies should not override the overall live-birth result unless they are confirmed in adequately powered, prespecified randomized comparisons.

What outcome should clinics use when evaluating similar add-ons?

Live birth, preferably cumulative live birth, should carry the most weight because it reflects the goal of treatment. Clinical pregnancy can provide supportive information, but implantation or biochemical-pregnancy signals alone are insufficient when they do not lead to more live births.

References

1. Intrauterine human chorionic gonadotropin administration before embryo transfer — National Library of Medicine (PubMed), 2026
2. Intrauterine administration of human chorionic gonadotropin (hCG) for subfertile women undergoing assisted reproduction — Cochrane Database of Systematic Reviews, 2018
3. Infertility and Fertility — Eunice Kennedy Shriver National Institute of Child Health and Human Development, n.d.