Study oversight

This minimal ovarian stimulation trial was designed by investigators at the Academic Medical Center (Amsterdam, The Netherlands) and the New Hope Fertility Center (NHFC; New York, NY). All the participants received IVF treatments in a single center (NHFC). The noninferiority trial was approved by the Institutional Review Board of New York Downtown Hospital and was registered before its start at clinicaltrials.gov : NCT 00799929.

The trial protocol was approved by a protocol review committee and a data and safety monitoring board, both appointed by NHFC and by the institutional review boards of the New York Downtown Hospital and the Biomedical Research Alliance of New York. Randomization, monitoring, and data analysis were coordinated at the Academic Medical Center in the Netherlands.

Study design and patients

Women were recruited between February 2009 and August 2013 via printed and online media. Participants were responsible to pay for their own medications; they were provided free services (ie, oocyte retrieval and embryo transfer) as compensation for participation. Inclusion criteria included women between 18-38 years of age, having normal menstrual cycles, undergoing first IVF treatment and having infertility diagnoses of male, unexplained, and tubal factors. Exclusion criteria included preexisting medical conditions (such as diabetes mellitus, hypertension, hypothyroidism, and hyperprolactinemia), a body mass index of <18.5 or >32 kg/m ² , and day-3 FSH >12 mIU/mL. Screening tests before initiation of treatment included complete blood count, varicella and rubella titer, Papanicolaou smear, syphilis, HIV 1 and 2, hepatitis B, hepatitis C, chlamydia, gonorrhea, prolactin, thyroid-stimulating hormone, cycle day 3 FSH and estradiol. Ultrasound testing was performed at baseline, and women with any submucosal or large intramural fibroid tumors that required surgery were excluded from the study.

Informed consent and randomization

After we obtained written informed consent, women were allocated randomly to mini-IVF or conventional IVF at a 1:1 ratio. The randomization was done at the start of the cycle with the use of sequentially numbered opaque envelopes that had been prepared on the basis of a computer-generated list at the Academic Medical Center in the Netherlands and sent to NHFC in New York. Women and medical staff were not blinded for treatment allocation. Outcome data were not disclosed to participants or investigators until completion of the trial.

Mini-IVF

After oral contraceptive pill pretreatment for 10-14 days, adequate suppression was confirmed with an estradiol level of <75 pg/mL. Minimal ovarian stimulation was started with an extended regimen (from day 3 until the day before triggering) of CC (50 mg/d orally) in conjunction with gonadotropin injections (Bravelle and/or Menopur, Ferring, Parsippany, NJ; Follistim, Merck, White House Station, NJ; or Gonal F, EMD Serono, Rockland, MA) starting on cycle days 4-7 with 75-150 IUs daily. No hypothalamic-pituitary suppression was performed, and the final maturation of oocytes was induced by a GnRHa nasal spray (Synarel nasal spray, Pfizer, New York, NY) when the lead follicle reached a diameter of ≥18 mm ( Figure 1 ). Oocyte retrieval was performed most often with local anesthesia; follicular flushing was performed as needed. Retrieved oocytes were fertilized by conventional IVF or ICSI, as indicated, and subsequently cultured until the blastocyst stage. All blastocysts were vitrified with the CryoTop method (Kitazato Biopharma, Fuji, Japan). A single thawed blastocyst was transferred in a subsequent natural or artificially prepared cycle with oral Estrace (Actavis Pharma, Inc, Parsippany, NJ).

Schematic diagrams of mini–in vitro fertilization and conventional in vitro fertilization protocols

In the artificially prepared frozen embryo transfer of the mini–in vitro fertilization protocol, oral estradiol treatment was started on day 3 and was given daily; progesterone treatment was added on day 13 onward to the estradiol pills.

GnRHa , gonadotropin-releasing hormone agonist; HCG , human chorionic gonadotropin; ICSI , intracytoplasmic sperm injection; IVF , in vitro fertilization; US , ultrasound.

Zhang et al. Mini-IVF vs conventional IVF. Am J Obstet Gynecol 2016 .

Conventional IVF

Conventional ovarian stimulation consisted of a long GnRHa protocol with mid-luteal down-regulation (Leuprolide Acetate, Teva, Sellersville, PA) followed by stimulation with daily gonadotropins injections (Bravelle and/or Menopur, Ferring; Follistim, Merck; or Gonal F) at a dose of 150-300 IUs daily starting in the early follicular phase (cycle day 3). The final maturation of oocytes was induced with the standard human chorionic gonadotropin (Novarel, Ferring; Pregnyl, Merck; or Ovidrel, EMD Serono, Rockland, MA), rather than GnRHa, when at least 2 follicles reached ≥18 mm. Oocyte retrieval was performed mostly with general anesthesia because of the presence of a large number of mature follicles. Retrieved oocytes were fertilized by conventional IVF or ICSI according to the same indications as in the mini-IVF protocol and subsequently cultured until the blastocyst stage. Two or 1 blastocysts (the latter if 2 embryos were not available or if there was a medical contraindication for double embryo transfer) were transferred in the fresh cycle. Remaining supernumerary blastocysts were vitrified, and 2 thawed blastocysts or 1 blastocyst (if 2 were not available) were transferred in subsequent naturally or artificially prepared cycles with oral Estrace.

Outcomes

The primary outcome was cumulative live birth per randomly assigned woman (which included fresh and subsequent frozen embryo transfers [FET]) within a time horizon of 6 months. Secondary outcomes were clinical pregnancy rate; OHSS; multiple pregnancy rate; gonadotropin usage; the number of retrieved, mature, and fertilized oocytes; implantation rate; cancellation rate, and failed fertilization. A clinical pregnancy was defined as at least 1 intrauterine sac at 6 weeks gestation; live birth was defined as a child born after 22 weeks of gestation or who weighed at least 500 g.

Sample size calculation and statistical analysis

Sample size calculation was based on an expected cumulative live birth rate of 65% in women who were ≤38 years old after conventional IVF based on US Society for Assisted Reproductive Technology registry data from 2007 because recruitment started in 2009. A noninferiority margin of –10% was considered a clinically significant difference. Thus, 564 women were needed to assure that, with a power of 80%, the lower limit of a 1-sided 95% confidence interval (CI) was within a prespecified border of –10%.

The effectiveness of mini-IVF vs conventional IVF was expressed as a risk ratio for cumulative live birth, with the corresponding 95% CI. The risk difference and 95% CI for live birth and the relative risks and 95% CI for all binary secondary outcomes were calculated. For continuous outcomes, data were expressed as means ± standard deviation, and the difference between both arms was compared with the use of t -test. Chi-square and Fisher exact tests were used for categoric data, as appropriate. The analysis of all outcomes was on an intention-to-treat basis. A probability value of <.05 was considered statistically significant. SPSS software (version 22.0; SPSS Inc, Chicago, IL) was used to perform all statistical analyses.