Children born after in vitro fertilization had a small but significant increase in the incidence of mental retardation in a nationwide Swedish study of 2.5 million births during a 25-year period, according to a report published in the July 3 issue of JAMA.
In contrast, the rate of autistic disorder was not increased among children born after IVF treatment, compared with those born after spontaneous conception.
However, when the data were broken down by type of IVF procedure, the use of one technique – intracytoplasmic sperm injection (ICSI) for paternal infertility – was associated with a small increase in the incidence of autistic disorder, said Dr. Sven Sandin of the Institute of Psychiatry, King’s College London, and his associates.
The latter findings may be particularly important in countries such as the United States where ICSI is often used even when the sperm sample is normal, "because of a presumed (but unproven) higher efficiency," they noted.
Dr. Sandin and his colleagues performed a prospective cohort study to test the hypothesis that IVF in general and ICSI in particular would be associated with an increased risk of mental retardation and autistic disorder. Both IVF and ICSI are known to raise the risk of perinatal complications and preterm birth, which in turn raise the risk of neurodevelopmental abnormalities.
IVF also has been linked with several specific neurological disorders including cerebral palsy, and Russell-Silver, Beckwith-Wiedemann, and Angelman syndromes. There also is concern that ICSI may allow fertilization with suboptimal sperm because it bypasses the natural selection of sperm; that it may physically damage the egg; and that it may contaminate the cytoplasm of the egg cell with culture media when the sperm is inserted, they said.
The study population comprised 2,541,125 children born in 1982-2007 and followed for a mean of 10 years. Among these children, 30,959 were born following an IVF procedure.
A total of 15,830 children had mental retardation, including 180 (1.1%) who were born after IVF. "Compared with offspring born following spontaneous conception, those born after an IVF procedure had a statistically significantly increased risk of mental retardation (relative risk, 1.18)," said Dr. Sandin, who is also in the department of medical epidemiology and biostatistics, Karolinska Institutet, Stockholm, and his associates.
However, the absolute difference in rates of mental retardation was small, at fewer than 7 cases per 100,000 person-years.
A total of 6,959 children had autistic disorder, including 103 (1.5%) who were born after IVF. Compared with children born after spontaneous conception, those born after any IVF procedure did not have a significantly increased risk of autism disorder.
However, this risk was significantly greater after one specific procedure, ICSI using surgically extracted sperm with fresh embryos, with an RR of 4.60, the investigators said (JAMA 2013;310:75-84).
The risk for mental retardation also increased in children born after ICSI using surgically extracted sperm with fresh embryos, with an RR of 2.35.
The risks for both mental retardation and autism disorder increased further in preterm births and in cases of multiple gestation.
In contrast, the risks for mental retardation and autistic disorder showed no association with IVF procedures regardless of whether blastocyst transfer, cleavage-stage transfer, frozen embryos, or fresh embryos were used.
Hormone stimulation is part of IVF, and some in the medical community have suggested that the use of hormones, not IVF per se, may account for any increased risk for autistic or other neurodevelopmental disorders. To control for this possibility, Dr. Sandin and his associates separately compared outcomes in children born to mothers who used only hormone therapy without any IVF procedure. They found no increase in risk for autistic disorder or mental retardation in this subset of patients.
To account for other factors that might contribute to neurodevelopmental risk, the investigators adjusted the data for parental age and psychiatric history. This had no effect on IVF-associated risks for mental retardation or autistic disorder. Duration of infertility also had no effect on these risks.
The researchers were unable to account for the number of embryos transferred during IVF, because that information was not available in the medical records before 2003.
They also had no information on parental educational attainment or socioeconomic status, which might potentially skew the study population toward more affluent couples who could afford multiple cycles of IVF. However, in Sweden, up to three IVF treatment cycles are free for childless women, so any bias of this type would likely be small, the researchers said.
"Our results should be applicable to most countries where IVF and ICSI are used. There are no major differences in equipment or laboratory work across countries," they added.