Outcomes and Recommendations of an Indian Expert Panel for Improved Practice in Controlled Ovarian Stimulation for Assisted Reproductive Technology

Table 1

Characteristics of included studies addressing the questions under investigation.

Authors, year	Methods	Patient population	Interventions/methods	Question under study
Sunkara et al., 2011	Analysis of the United Kingdom IVF Registry database from April 1991 to June 2008	400,135 IVF cycles with autologous oocytes and fresh transfers	To explore the association between the number of eggs and live birth outcome, a likelihood logistic model with live birth outcome as the dependent variable was fitted using a fractional polynomial to handle the number of oocytes as a continuous independent variable	1
Ji et al., 2013	Cohort study	Normogonadotropic women (,455) aged 18–34 years undergoing their first IVF treatment cycle with a long GnRH agonist protocol at a single institution, including one complete treatment round of a fresh transfer cycle and subsequent frozen transfer cycles	Patients were categorized into four groups according to the number of oocytes retrieved: 0–5, 6–10, 10–15, or ≥16 oocytes. LBR per fresh transfer and cumulative LBR were calculated per group and compared. Logistic regression analyses identified association between oocyte number and live birth outcome after adjusting for confounders	1
De Geyter et al., 2015	Analysis of the Swiss IVF Registry database from 1993 and 2012	>100,000 IVF cycles reporting delivery rates per fresh embryo transfer	Data extraction and analysis of IVF outcomes, including graphical displays with the crude outcome data	1
Ata et al.,   2012	Cohort study	990 patients undergoing IVF in 70 North American Centers between January 2010 and July 2011	PGS of human embryos by array CGH. Embryo biopsy was performed on day 3 or at blastocyst stage. The proportion of euploid embryos over embryos biopsied was calculated per cycle. Linear regression analysis was performed to assess the effect of cohort size on euploidy rate adjusted for the effect of female age	2
Lehert et al., 2010	Systematic review and meta-analysis of prospective randomized or quasi-RCT	16 studies involving 4,040 women aged 40 or less undergoing IVF or ICSI irrespective of use of GnRH agonists or antagonists in which hMG and r-hFSH for COS were compared	Comparison of number of retrieved oocytes (primary endpoint) between two modalities of gonadotropin treatment, using the random effects model. Evidence of superiority of one treatment was accepted when the results of the main analysis and the sensitivity analyses were consistent	3
Lehert et al., 2014	Systematic review and meta-analysis of prospective, randomized, parallel-, comparative-group trials	40 RCTs involving 6,443 patients aged 18-45 undergoing IVF or ICSI treated with GnRH analogues and r-hFSH plus r-hLH or r-hFSH alone for COS	Comparison of number of retrieved oocytes and CPR between two modalities of gonadotropin treatment (primary endpoints), using the random effects model. Ovarian response to treatment—normal or poor—was included as a covariate for subgroup analysis	3
Xu et al., 2012	Retrospective cohort study	11,055 consecutive patients undergoing IVF/ICSI using long GnRH agonist protocol and a subgroup of 4,021 patients participating in a FET program at a single institution from January 2002 and September 2011	OPRs assessed with logistic regression analysis according to 8 distinct P levels intervals on the day of hCG administration and compared among low (<4 oocytes retrieved), intermediate (5–19 oocytes retrieved), and high (≥20 oocytes retrieved) responders. P thresholds for a detrimental effect on cycle outcome were calculated	4
Griesinger et al., 2013	Retrospective combined analysis from six clinical trials comparing GnRH agonists and antagonists in COS	Normogonadotropic women (,866) up to 39 years of age undergoing COS with r-hFSH and GnRH antagonist	OPRs assessed with univariate and multivariate analyses according to serum P levels ≤1.5 ng/mL versus >1.5 ng/mL on the day of hCG administration and compared among low (1–5 oocytes retrieved), normal (6–18 oocytes retrieved), and high (>18 oocytes retrieved) responders	4
Requena et al., 2014	Retrospective cohort study	2,850 high responders (≥20 oocytes retrieved or estradiol levels ≥3000 pg/mL) undergoing IVF-ET or FET in 11 Spanish institutions from January 2009 to December 2011	Implantation and CPRs assessed with logistic regression analysis according to 5 distinct serum P levels intervals on the day of hCG administration	4
Venetis et al., 2015	Retrospective cohort study	3,296 patients undergoing IVF/ICSI and fresh ET in a single IVF center between 2001 and 2013	LBRs assessed with bivariate and multivariate analyses according to serum P levels ≤1.5 ng/mL versus >1.5 ng/mL on the day of hCG administration and compared among low (<6 oocytes), normal (6–18 oocytes), and high (>18 oocytes) responders	4

Questions under study are as follows. (1) What is the optimum number of oocytes that is associated with the highest live birth rates? (2) Is there a correlation between cohort size and embryo quality? (3) Does the choice of gonadotropins affect oocyte yield in IVF? (4) Should a “freeze-all” policy be adopted in all cycles with progesterone levels >1.5 ng/ml on day of hCG?
IVF: in vitro fertilization; LBR: live birth rates; PGS: preimplantation genetic screening; CGH: comparative genomic hybridization; RCT: randomized controlled trials; ICSI: intracytoplasmic sperm injection; hMG: human menopausal gonadotropin; r-hFSH: recombinant human follicle stimulating hormone; COS: controlled ovarian stimulation; r-hLH: recombinant human luteinizing hormone; CPR: clinical pregnancy rates; FET: frozen-embryo transfer; GnRH: gonadotropin releasing hormone; P: progesterone; OPR: ongoing pregnancy rate; ET: embryo transfer; hCG: human chorionic gonadotropin.