A systematic review and meta-analysis on the association between ICSI and chromosome abnormalities

Background: In the decade following the introduction of ICSI, a higher prevalence of de novo chromosome abnormalities, in particular sex chromosome and autosomal structural abnormalities, as well as inherited abnormalities was described in children conceived by ICSI compared to both naturally conceived (NC) children and children conceived by standard IVF. The explanation for the observed increase in prevalence is not clear and has been suggested to reflect parental factors (e.g. age or sperm quality) or to be a result of the ICSI procedure itself. Over the years, the procedure, as well as the patient group, and indications for ICSI treatment have changed. Objective and Rationale: The objective of this systematic review and meta-analysis was to assess the prevalence of chromosome abnormalities in ICSI pregnancies and children and to examine any potentially increased risk compared to standard IVF and NC. Search Methods: Pubmed, Embase, Cochrane Libraries and Web of Science up to October 2020 were searched. Primary outcome measures were overall chromosome abnormalities and de novo abnormalities (including sex chromosome abnormalities and autosomal abnormalities). The secondary outcome was inherited abnormalities. We followed the PRISMA guidelines and relevant meta-analyses were performed. Outcomes: The search included 4648 articles, out of which 27 met the inclusion criteria, and 19 were included in quantitative synthesis (meta-analyses). The prevalence of chromosome abnormalities varied considerably between studies, possibly explained by large differences in sample size and patient demographics. Only five studies were eligible for pooled analyses on adjusted data. All studies had a critical risk of bias. Results from pooled adjusted data showed no evidence of an increased risk of overall chromosome abnormalities when comparing ICSI to either standard IVF (aOR 0.75 (95% CI 0.41-1.38)) or NC (aOR 1.29 (95% CI 0.69-2.43)). In contrast, meta-analyses on unadjusted data showed an increased risk of overall chromosome abnormalities in ICSI compared to both standard IVF (OR 1.42 (95% CI 1.09-1.85)) and NC (OR 2.46 (95% CI 1.52-3.99)) and an increased risk of de novo abnormalities in ICSI compared to NC (OR 2.62 (95% CI 2.07-3.31)). Yet, based on a very low certainty of evidence, the conclusion remains, that no indication of an increased risk of chromosome abnormalities in ICSI offspring could be found. If an increased risk of chromosome abnormalities in selected ICSI offspring should exist, the absolute risk continues to be small. Wider Implications: This review provides an extensive overview of the existing evidence on the relationship between ICSI and chromosome abnormalities in the offspring. We highlight the need for well-designed large, prospective, controlled studies with systematic cytogenetic testing. Existing data are limited and, in many cases, marred by critical levels of bias.