Blastocyst morphology is associated with the incidence of monozygotic twinning in assisted reproductive technology

An increased incidence of monozygotic twinning after a blastocyst transfer has been previously reported in assisted reproductive technology treatment. It is uncertain whether this phenomenon is due to the extended culture time, culture medium, or inherent blastocyst parameters.

This study aimed to investigate the association between blastocyst parameters (in vitro culture time, blastocyst stage, and inner cell mass and trophectoderm grading) and the incidence of monozygotic twinning after assisted reproductive technology.

This was a retrospective cohort study employing data from a multicenter, large, electronic database from 4 academic hospitals. All clinical pregnancies after a single blastocyst transfer between January 2014 and February 2020 were included. Blastocyst morphology was evaluated based on the Gardner grading system, considering the blastocyst stage, and inner cell mass and trophectoderm grading (grades A, B, and C). Monozygotic twinning was defined as ≥2 fetal heartbeats in a single gestational sac or 2 gestational sacs with sex concordance at birth. The multivariable predicted marginal proportions from logistic regression models were used to compute adjusted relative risks for the association between blastocyst parameters and the incidence of monozygotic twinning.

The overall monozygotic twinning rate was 1.53% (402 of 26,254 cases). The monozygotic twinning was not associated with the culture time in vitro (day 5 vs day 6) or blastocyst stage (early, blastocyst, expanded, hatching, and hatched). Alternatively, monozygotic twinning was associated with lower inner cell mass grading (B vs A: adjusted relative risk, 1.67 [95 % confidence interval, 1.28-2.25]; C vs A: adjusted relative risk, 1.98 [95% confidence interval, 1.18-3.11]) and higher trophectoderm grading (B vs C: adjusted relative risk, 1.38 [95% confidence interval, 1.03-1.92]; A vs C: adjusted relative risk, 2.14 [95% confidence interval, 1.45-3.20]). The incidence of monozygotic twinning was the lowest in blastocysts with grade A inner cell mass and grade B or C trophectoderm (0.82%, as the reference) and the highest in blastocysts with grade B or C inner cell mass and grade A trophectoderm (2.40%; adjusted relative risk, 2.62; 95% confidence interval, 1.60-4.43). The incidence of monozygotic twinning in blastocysts with consistent inner cell mass or trophectoderm grading was somewhere in between (both A: 1.58%; adjusted relative risk, 1.86 [95% confidence interval, 1.23-3.04]; both B or C: 1.59%; adjusted relative risk, 1.84 [95% confidence interval, 1.29-2.90]).

Higher risk of monozygotic twinning was associated with blastocyst morphology specific to those blastocysts with loosely arranged inner cell mass cells combined with tightly packed trophectoderm cells.