In vitro fertilization and childhood retinoblastoma

Abstract

Aims

To estimate the frequency of childhood retinoblastoma among children conceived by in vitro fertilization (IVF).

Methods

Using the United Kingdom-based General Practice Research Database (GPRD), we identified all live births, cases of retinoblastoma and IVF procedures occurring between January 1989 and December 2001.

Results

We identified 0 cases of retinoblastoma among 176 children conceived by IVF (Risk = 0/176, one-sided 97.5% CI 0, 0.02) and 24 cases of retinoblastoma diagnosed before age 5 years among 358 094 children not conceived by IVF (6.7 cases per 100 000 births [95% CI 4.5, 10]).

Conclusions

These data provide some reassurance that children born as a result of IVF are not at markedly increased risk of retinoblastoma.

Introduction

As the number of women using assisted reproductive technology, specifically, in vitro fertilization (IVF) increases, there is a need to investigate potential adverse health effects in children conceived by IVF. A study by Klip et al.[1] showed that after an average of 6 years of follow-up, children conceived by IVF or other related fertility techniques (n = 9484) were not at increased risk of childhood malignancies compared with children not born after IVF (n = 7532) (RR = 0.8, 95% CI 0.3, 2.3). More recently, Moll et al.[2] suggested that children born after IVF have a large elevated risk of retinoblastoma (RR = 7.2, 95% CI 2.4, 17.0), a cancer of the eye that is generally diagnosed in early childhood and often requires enucleation. Their estimate was based on a comparison of five cases of retinoblastoma born after IVF in the Netherlands between November 2000 and February 2002 with the expected number of retinoblastoma cases following IVF. The authors used sex-specific and age-specific retinoblastoma incidence rates and an estimate of the number of children born after IVF in the Netherlands, 1%[3], to calculate the expected number of cases. Given the marked difference in these two findings and the dearth of studies examining this association, we used data collected in the United Kingdom-based General Practice Research Database (GPRD) to examine the association between IVF and retinoblastoma.

Methods

The GPRD has previously been described in detail [4]. Briefly, since 1987, over three million residents in the United Kingdom (UK) have been enrolled with general practitioners (GPs) who use office computers originally provided by Value Added Medical Products (VAMP Health) and who agreed to provide data for research purposes to the GPRD. The GPs received 12 months of instruction on the standardized recording of medical information and they agreed to supply anonymized information to academic researchers on an ongoing basis. The information recorded includes patient characteristics, medical procedures (child birth), clinical diagnoses, and hospitalizations. Children born to women in the database are assigned a family identification number (ID) in addition to a personal identification number so that information on family members are linked.

We reviewed the GPRD for the period January 1, 1989 until December 31, 2001 and identified all live births. Among these live births we then identified all cases of retinoblastoma diagnosed before age 5 years. Using the family ID variable, we determined whether any of the cases of retinoblastoma were conceived by IVF. For the same period, we also determined the number of women who underwent at least one IVF procedure. For those women, we determined the number of children conceived by IVF. We provide frequency estimates and 95% confidence intervals for the number of cases of retinoblastoma occurring over the follow-up period, and separately, for the number of live births conceived by IVF.

Results

We identified 358 270 live births occurring between January 1, 1989 and December 31, 2001 and among these, there were 24 cases of retinoblastoma. Fifty-eight percent of the retinoblastoma cases were male, 54% were diagnosed before 1 year of age and the average age at birth of the 24 mothers was 28.2 years (SD 4.5) (Table 1). All cases were alive at the end of follow-up. Based on these data, we estimate a frequency of 6.7 cases of retinoblastoma per 100 000 births (95% CI 4.5, 10), or 1 in 15 000 births, which is similar to an estimate reported by Koudstaal et al.[5]. Over the follow-up period, the frequency of retinoblastoma cases increased from 3.9 per 100 000 live births (1989–92) to 11.5 per 100 000 live births (1997–2001).

Table 1.

Frequency of retinoblastoma per 100 000 live births, age at diagnosis and mean age of the mother at birth in the GPRD, 1989–2001

Year	Number of cases	Mean age (years) of mother at birth (SD)	Diagnosed before 1 year of age (%)	Number of live births	Frequency of retinoblastoma (per 100 000 live births) and 95% CI
1989–92	5	28.4 (6.7)	0.60	129 830	3.9 (1.6, 9.0)
1993–96	8	26.5 (4.0)	0.38	133 401	6.0 (3.0, 11.8)
1997–2001	11	29.3 (3.8)	0.64	95 489	11.5 (6.4, 20.6)
Total	24	28.2 (4.5)	0.54	358 270	6.7 (4.5, 10.0)

For the same period, 1398 women had a total of 2091 IVF attempts. During the follow-up period, the proportion of live births conceived by IVF increased from 2.5 per 10 000 births (1989–92) to 2.9 per 10 000 births (1993–96) to 11 per 10 000 births (1997–2001) (Table 2). Moll et al.[2] estimated that 1–1.5% of births in the Netherlands (100–150 per 10 000 births) are secondary to IVF, an estimate considerably higher than that which we observed in the UK. In the GPRD, 176 live births occurred within 10 months of an episode of IVF indicating that the procedure successfully produced a pregnancy resulting in a live birth in 8.4% of attempts.

Table 2.

Number of total births and births resulting from IVF in the GPRD, 1989–2001

Year	Number of live births	Number of births due to IVF	Proportion of total births due to IVF (per 10 000 births)
1989–92	129 830	32	2.5
1993–96	133 401	39	2.9
1997–2001	95 489	105	11.0
Total	358 270	176	4.9

Discussion

In our series none of the 24 children with retinoblastoma was born to women who had undergone IVF (0/176, 1-side 97.5 CI 0, 0.02). In contrast to the findings by Moll et al.[2], these data provide some reassurance that children born after IVF are not at markedly increased risk of retinoblastoma. Increased surveillance of children born after IVF is warranted to ensure that children conceived by IVF are not at increased risk for birth defects, childhood cancers and developmental disorders.