Assisted reproductive technology and risk of congenital malformations: a meta-analysis based on cohort studies

作者：Qin, Jiabi^*; Sheng, Xiaoqi; Wang, Hua; Liang, Desheng; Tan, Hongzhuan; Xia, Jiahui

来源：Archives of Gynecology and Obstetrics, 2015, 292(4): 777-798.

DOI：10.1007/s00404-015-3707-0

• Assisted reproductive technology
• in vitro fertilization
• Intracytoplasmic sperm injection
• Congenital malformation
• Birth defect
• Meta-analysis

To assess the association between assisted reproductive technology (ART) and risk of congenital malformations (CM) by conducting a meta-analysis of cohort studies. @@@ PubMed, Google Scholar, Cochrane Libraries and Chinese database were searched through August 2014 to identify studies that met pre-stated inclusion criteria. Either a fixed- or a random-effects model was used to calculate the overall combined risk estimates. Subgroup analysis was performed to explore potential heterogeneity moderators. @@@ Fifty-seven studies involving 119,874 infants conceived following ART and 1,212,320 infants conceived naturally were included in the analysis. The ART-conceived infants were associated with a higher risk of CM [relative risk (RR) = 1.33; 95 % confidence interval (CI) 1.24-1.43] when compared with those conceived naturally. When data were restricted to singleton births (RR = 1.38; 95 % CI 1.30-1.47), major CM (RR = 1.47; 95 % CI 1.29-1.68), matched/adjusted studies (RR = 1.37; 95 % CI 1.27-1.47) or high quality studies (RR = 1.40; 95 % CI 1.27-1.55), the increased risk of CM still existed in ART pregnancies. Additionally, an increased risk of CM was also found when the ART twin (RR = 1.18; 95 % CI 1.06-1.32) or multiple births (RR = 1.16; 95 % CI 1.05-1.27) were separately compared with spontaneously conceived twin or multiple births. Substantial heterogeneity was observed across studies (I (2) = 68, 44, 39, and 33 % for all infants, singletons, twins and multiples, respectively). Whether confounding factors were matched or adjusted, study quality and sample size as the first three of the most relevant heterogeneity moderators have been identified. No evidence of publication bias was observed (P > 0.10). @@@ The ART-conceived infants have a higher risk of CM compared with those conceived naturally. However, these estimates have to be viewed with caution because of heterogeneity.

• 出版日期2015-10
• 单位中南大学