Maternal age and diabetes in childhoodBMJ 2010; 340 doi: https://doi.org/10.1136/bmj.c623 (Published 24 February 2010) Cite this as: BMJ 2010;340:c623
- Edwin A M Gale, professor of diabetes
Two recent meta-analyses of early life influences on the risk of early onset type 1 diabetes largely confirm what previous studies have shown.1 2 The first found that childhood onset diabetes is associated with increasing birth weight, equivalent to a 7% increase in risk for every 1000 g in weight,1 and the second reported that caesarean section increases the risk by around 20%.2 A third more recent pooled analysis of five cohort studies and 25 case-control studies provides formal confirmation that the risk of childhood onset diabetes increases with maternal age: 5% for each five years of age.3 The studies included were heterogeneous, so interpreting the results is a challenge.
Why should the age of the mother at delivery influence the risk of diabetes in the child? There is much speculation, but the main explanations are that biological programming of the child is in some way affected by the age of the mother, or perhaps the father, and that the difference is caused by environmental or social confounders encountered before birth or in early childhood. Older women may, for example, be more likely to have complicated pregnancies and require caesarean section. Statistical adjustment for this and other potential confounders did not, however, affect the outcome in the pooled analysis described above.3
Other conditions, including childhood cancer, are influenced by maternal age. A recent pooled analysis found an odds ratio of 1.08 (95% confidence interval 1.06 to 1.10) per five year increase in maternal age for the 10 most frequently diagnosed childhood cancers. Possible explanations include germline mutations (more common in sperm than in oocytes); alterations in oocyte gene expression as a result of promoter DNA methylation or age related hormonal changes; more prolonged maternal exposure to environmental carcinogens; and other unexamined potential confounders.4
In contrast, asthma in childhood seems to be associated with younger maternal age. A postal questionnaire of 16 000 adults in northern Europe reported an 8% reduction in the risk of childhood asthma (adjusted odds ratio 0.92, 0.88 to 0.97) for each five year increment in maternal age, which persisted after adjustment for known confounders.5 This provides yet another example of the curious reciprocal association between type 1 diabetes and atopic disorders. Both have increased in parallel over the past 50 years, especially in affluent populations of European descent, yet atopy reduces the risk of type 1 diabetes and diabetes reduces the risk of asthma.6 This reciprocity has been explained in terms of the Th1-Th2 paradigm. Type 1 diabetes is characterised by T helper type 1 (Th1) CD4 positive T cells and the signature cytokine interferon γ, whereas atopic disorders are characterised by Th2 cells and interleukins IL-4, IL-5, and IL-13.7 The axis is more complex than initially believed, but the polarity between Th1 and Th2 disorders seen at a population level remains highly suggestive.6 7
Infectious disease eliminated up to a third of the children born to our ancestors within the first few years of life, but recent generations have faced a much less challenging environment. One consequence, according to the hygiene hypothesis, is that an increasing number of children will fail to develop a healthy “mixed” pattern of immune response and will instead remain as outliers at either end of the Th1-Th2 spectrum.8 Hence the parallel rise in both types of immune disorder. But does the age of the mother influence the immune potential of the child, and if so, how? Once again we are left clutching at straws.
The rise of childhood type 1 diabetes is largely unexplained. The increase can be traced back to the middle of the 20th century, if not earlier, with a linear trend and an approximate doubling time of 20 years.9 10 Prospective studies from birth confirm the dominant role of HLA associated susceptibility in conferring individual risk, such that early onset is associated with the highest risk alleles.11 Paradoxically, it is equally clear that the global increase in type 1 diabetes must be driven by as yet unidentified non-genetic influences. To take one example, the increase in childhood type 1 diabetes over the past 50 years has largely been restricted to a subset of the population with HLA associated susceptibility, yet the ratio of intermediate to high risk alleles has increased substantially over the same period, implying that the environment has become progressively more favourable for the development of diabetes.12
Consistent with this interpretation, the increase in childhood onset diabetes has in some cases been mirrored by a reduced incidence in young adults, which suggests that we may be witnessing a “spring harvest” of genetically susceptible individuals rather than an overall increase in the lifetime incidence of type 1 diabetes. In all events, it is clear that the trend for women to delay their pregnancies into their 20s or 30s has made a modest contribution towards the rise of childhood onset type 1 diabetes, but is in no way responsible for it.3
Cite this as: BMJ 2010;340:c623
Competing interests: The author has completed the Unified Competing Interest form at www.icmje.org/coi_disclosure.pdf (available on request from the corresponding author) and declares (1) no financial support for the submitted work from anyone other than his employer; (2) No financial relationships with commercial entities that might have an interest in the submitted work; (3) No spouse, partner, or children with relationships with commercial entities that might have an interest in the submitted work; (4) No non-financial interests that may be relevant to the submitted work.
Provenance and peer review: Commissioned; not externally peer reviewed.