Elsevier

The Lancet

Volume 354, Issue 9190, 6 November 1999, Pages 1579-1585
The Lancet

Articles
Deliveries and children born after in-vitro fertilisation in Sweden 1982–95: a retrospective cohort study

https://doi.org/10.1016/S0140-6736(99)04345-7Get rights and content

Summary

Background

In-vitro fertilisation is an effective treatment for infertility, but there is concern about the health of children. We investigated, in a retrospective-registry study, malformations, cancers, and deaths in the complete Swedish in-vitro-fertilisation birth cohort compared with the general population.

Methods

We collected data from all in-vitro-fertilisation clinics in Sweden and compared the obstetric outcomes of babies (n=5856) born between 1982 and 1995 with all babies born in the general population (n=1 505 724) during the same period, according to data from the Swedish Medical Birth Registry and the Registry of Congenital Malformations. We investigated the incidence of childhood cancer through the Swedish Cancer Registry. Data were stratified for maternal age, parity, previous subfertility, year of birth, and multiple of pregnancies.

Findings

Multiple births occurred in 27% of pregnancies compared with 1% in the control group. In the in-vitrofertilisation group, more babies were born preterm (<37 weeks) than controls (30·3 vs 6·3%) and more had low birthweights (<2500 g, 27·4 vs 4·6%). The perinatal mortality was 1·9% in the in-vitro fertilisation group and 1·1% in the controls. For in-vitro-fertilisation singletons, the risk ratios, adjusted for year of birth, for very preterm birth (<32 weeks) and very low birthweight (<1500 g) were 3·54 (95% CI 2·90–4·32) and 4·39 (3·62–5·32), respectively. Malformations occurred in 5·4% of all babies in the in-vitrofertilisation group (1·39 [1·25–1·54]), and the rates of neural-tube defects and oesophageal atresia were higher than those in the controls. There was no increase in childhood cancer in the in-vitro-fertilisation group.

Interpretation

A high frequency of multiple births and maternal characteristics were the main factors that led to adverse outcomes, and not the in-vitro-fertilisation technique itself. The clinical practice of in-vitro-fertilisation needs to be changed to lower the rate of multiple pregnancy.

Introduction

More than 20 years have passed since the first baby was born after in-vitro fertilisation.1 A 1995 world report showed that each year more than 40 000 babies are born worldwide after in-vitro fertilisation.2 In Sweden, the number of births after this treatment accounts for 1·6% of all neonates.

Perinatal outcome after in-vitro fertilisation has remained under scrutiny by the medical profession and the general public. Outcomes differ from those after spontaneous conception, mainly because of the high frequency of multiple pregnancies.3 Most countries report an incidence of multiple births of 20–30% after assisted conception, which leads to 30–40% of all in-vitro-fertilisation babies being born as a result of multiple pregnancies, compared with 2–3% in the general population.2 Treatment of infertility by assistedreproduction technology is thought to be an important contributing factor to the increase in multiple births seen in many developed countries in the past few decades.4 This increase has raised concerns, not only about the social consequences but also because multiple-birth babies are at higher risk of perinatal mortality and morbidity and of long-term sequelae (eg, cerebral palsy) than singletons.5

Studies have also shown that in-vitro-fertilisation singleton pregnancies have higher complication rates than those in the general population. In particular, the rate of preterm deliveries is higher and there is an increased risk of babies being small for gestational age.6, 7 Mothers who achieve a pregnancy after in-vitro fertilisation are generally older and have a lower parity than mothers who conceive their children naturally, which confers a higher risk of maternal and perinatal complications. One study has also shown that infertility contributes to increased obstetric risks.8

The introduction of new technologies has raised concerns that offspring may be at increased risk, especially of malformations. No evidence has yet shown that cryopreservation is an additional risk factor for perinatal outcome9, 10 but studies suggest an increased risk of chromosomal anomalies, mainly of the sex chromosomes, after intracytoplasmic sperm injection.11, 12 Since many perinatal outcomes—malformations and perinatal mortality—occur with a low frequency, a large study group is needed to investigate the risks with sufficient power. We aimed to assess the outcome of the complete Swedish in-vitro-fertilisation birth cohort from 1982 to 1995 compared with all births in Sweden during the same period, from data from existing registries on birth, malformation, cancer, and death.

Section snippets

Study group

The study group included all children born in Sweden after in-vitro fertilisation, from the first child born in 1982 to the last child born before April, 1995. Individual data on women after in-vitro fertilisation were, with permission from the Swedish Data Inspection Board, requested by the National Board of Health and Welfare from all 14 in-vitro-fertilisation clinics in Sweden.

In Sweden, every individual has a unique identification number that is widely used by health-care and social-service

Identified deliveries

4634 deliveries (in 4353 women) after in-vitro fertilisation were reported from the clinics, 4517 (97·5%) of which could be identified in the Swedish Medical Birth Registry. 77 deliveries had not been reported to this registry but were identified in the official birth registry of Statistics Sweden. Thus, only 40 (0·9%) deliveries of those reported from the clinics could not be identified as delivered. The number of deliveries reported from the clinics was checked against the cumulative annual

Discussion

In-vitro fertilisation in Sweden resulted in the birth of nearly 6000 children from the procedure's introduction in 1982 to 1995. Most of these children were born healthy. Our retrospective registry study of these babies has several unique features: it was a large controlled study; it was an investigation of all children born after in-vitro fertilisation in Sweden and not a sample; the various sources of data (reports from all in-vitro fertilisation centres, the Swedish Medical Birth Registry,

References (29)

  • P Lancaster

    High incidence of preterm birth and early losses in pregnancy after in vitro fertilisation

    BMJ

    (1985)
  • R Derom et al.

    The epidemiology of multiple births in Europe

  • B Petterson et al.

    Twins, triplets, and cerebral palsy in births in Western Australia in the 1980s

    BMJ

    (1993)
  • P Doyle et al.

    Preterm delivery low birth weight and small-for-gestational weight in liveborn singleton babies resulting from in-vitro fertilisation

    Hum Reprod

    (1992)
  • Cited by (616)

    View all citing articles on Scopus
    View full text