While a medicated Intrauterine Insemination (IUI) is cheaper than IVF ($3,000 vs $20,000) live birth rates remain stubbornly low (often 10% per cycle). Some clinicians have wondered whether performing an endometrial scratch during an IUI cycle might improve rates of success.
As background, endometrial scratch is relatively painless (no anesthesia needed) and has been sporadically adopted during IVF with the belief the scratch incites a reaction that makes the woman’s endometrium more receptive to an embryo, increasing implantation and pregnancy rates. That said, not all doctors are sold on its effectiveness and some think it may increase the risk for Asherman Syndrome in a small subset of patients.
What’s New
A team of researchers analyzed the 8 best-run studies on the impact of endometrial scratch on medicated IUI cycles (nearly 2,000 IUIs studied). The goal was to see if success rates improved as defined by “on-going pregnancy” (confirmed on ultrasound at week 12 of pregnancy).
What The Study Showed
IUI success rates with endometrial scratch effectively doubled from 10% to 20% and, after adjusting for a few variables, improved the odds of success by 2.3x.
There was no indication the approach led to a higher rate of twins, miscarriage (measured at 20 weeks on ultrasound) or any other potential issues.
Importantly, the endometrial scratch had to be performed during that cycle as an endometrial scratch performed in the preceding cycle showed no benefit.
Limitations
The work suffers from the fact that there was a fair amount of difference between these studies in terms of the types of patients they treated (unexplained, male factor etc), the drug protocols the woman were placed on and the endometrial scratch technique used. Investigators considered their findings to be “promising”, but strongly underscored there is still the need for a well-designed, large, prospective trial to confirm their observations. What's more, the study stopped short of analyzing live birth rates, which we consider to be the true mark of success.