Does Intracytoplasmic sperm injection (ICSI) improve live birth rate when compared to conventional in vitro fertilisation (IVF) in non-male factor infertility? A systematic review and meta-analysis

Abstract Objective Initially developed for sperm abnormalities, the indications of intracytoplasmic sperm injection (ICSI) have progressively expanded beyond male factor infertility, but the benefit of ICSI over conventional in vitro fertilisation (IVF) in terms of live birth rate (LBR) in non-male factor infertility has not been formally demonstrated in the literature. The aim of this systematic review and meta-analysis was to determine whether the use of ICSI improves LBR as compared to IVF in non-male factor infertility cases. Evidence review This prospectively registered systematic review and meta-analysis was guided by PRISMA guidelines. PubMed, Embase, Cochrane Library were searched for relevant literature reported between January 01, 2004 and March 15, 2020. Only articles conducted in couples without obvious male factor infertility according to the authors were included. Live Birth Rate (LBR) per cycle and/or per transfer was used as main outcome. Fertilisation rate, cycles with total fertilisation failure, and clinical pregnancy rate were also recorded as secondary outcomes. Study selection, bias assessment and data extraction were performed by two independent reviewers. The methodological quality of the selected studies was assessed by two independent reviewers using the Cochrane Handbook methods and by the adapted Newcastle-Ottawa Quality Assessment Scale for cohort studies. The risks of bias were assessed using ROBIN-1 tools. Risk ratio and 95% confidence interval (CI) were estimated using random effect model. Results Among the 1,760 potentially eligible studies, 61 were fully assessed and 21 were included in the meta-analysis, corresponding to almost 1,000,000 assisted reproductive technology cycles. Among the selected studies, 17 reported LBR per cycle, 11 reported LBR per transfer, and 7 among them reported both outcomes. One was a prospective randomised controlled trial, while others were either retrospective cohort (n=17) or case-control studies (n=4). Individual studies’ sample size ranged from 60 to 569,605 cycles. Meta-analysis showed that IVF provided a significantly higher LBR per cycle when compared to ICSI (RR=1.10; 95%CI=1.02-1.18). LBR per transfer tended to be higher when IVF was used as compared to ICSI, but this difference barely reached statistical significance (RR 1.18, 95% CI 1.00-1.38, p=.05). Subgroup analysis performed in poor responders showed no significant difference between IVF and ICSI concerning LBR per cycle or LBR per transfer. Secondary analyses demonstrated a lower fertilisation rate, but comparable clinical pregnancy and fertilisation failure rate with IVF as compared to ICSI. Sensitivity analyses led to similar results and conclusions. Conclusion Despite its limitations, mainly related to the inclusion of a majority of retrospective studies, this meta-analysis synthesizes and reinforces the existing literature and provides further evidence that ICSI does not improve LBR per cycle / per transfer as compared to IVF in couples undergoing assisted reproductive technology for non-male infertility.