|
In the September issue of GGH (Vol 19, No 3) you reviewed an interesting publication by Singhal et al
1 who studied the relation between infant feeding, early growth and insulin
resistance at age 13-16 years in individuals with a birth weight below 1,850 grams (which the authors labeled
preterm). In their study, insulin resistance was not associated with birth weight but with growth in the first
two weeks of postnatal life; thus, they concluded that Barker's hypothesis "can be reinterpreted as a postnatal
event". In our opinion, their data should be interpreted more cautiously, for the following reasons.
The first point is that selection bias is quite likely. The application of birth weight instead of gestational
age as inclusion criterion (< 1,850 grams) suggests that severely growth-retarded individuals born at term
are also included. Another point of our concern is that in both experimental groups there were considerable
numbers lost to follow-up (65-68%).
Secondly, conclusions with respect to insulin resistance in later life were drawn from a population aged 13-16 years.
In this age period there is a wide variation in pubertal stages, and during pubertal development insulin sensitivity
is decreased.2 Moreover, girls born small for gestational age have a tendency towards
early and rapid progression of puberty,4 and hyperandrogenism,4
which is accompanied by decreased insulin sensitivity. It is likely that many infants in the experimental groups had
a low weight for gestational age at term; thus, it is conceivable that they may have shown abnormalities in pubertal
onset and tempo, as well as in androgen metabolism.
Thirdly, although an earlier study of this research group (in the same population at age 7.5-8 years) suggested that
suboptimal nutrition, which may result in poor early postnatal growth, adversely effects neurodevelopmental outcomes,
little emphasis is put on the possible beneficial effects of nutrient-enriched preterm formulas.5
This suggests that discouraging early postnatal catch-up growth by restricted food intake in infants with a birth
weight below 1,850 grams is hard to justify.
The last and major point is that Singhal and colleagues have extrapolated their findings in individuals born preterm
to conclusions about the general population. The first two postnatal weeks of the experimental groups took place
halfway into the third trimester. In our opinion it cannot be automatically assumed that postnatal growth taking
place at an age that is normally spent in utero , can be considered equivalent to postnatal growth of a
term infant.
In conclusion, this interesting study has shown that slow early postnatal growth of preterm infants is associated with
low insulin resistance at adolescence. In our view it is uncertain whether these effects persist into adulthood and
whether early postnatal catch-up growth predisposes to insulin resistance only in preterm infants or also in those
born at term.
References
- Singhal A, et al. Lancet 2003;361:1089-1097.
- Amiel SA, et al. N Engl J Med 1986;315:215-219.
- Ibáñez L, et al. Pediatrics 2000;106:E72.
- Ibáñez L, et al. J Clin Endocrinol Metab 1998;83:3558-3562.
- Lucas A, et al. BMJ 1998;317:1481-1487.
MatijnFinken, MD
Anne Margriet Euser, MSc
Friedo W. Dekker, PhD
Jan Maarten Wit, MD, PhD
Leiden University Medical Center
Leiden, The Netherlands
Response: The comments of Dr. Finken and colleagues are welcome as they point to several possible
methodologic and interpretive flaws in the work of Singhal et al. Although we do not know the exact number of small
for gestational age neonates included in the cohort of subjects reported, it is likely that the majority were preterm
and appropriate for gestational age. If there is a potential way in which to prevent the development of insulin
resistance and the dysmetabolic syndrome, it should be explored. However, clearly, one would not want to jeopordize
optimal neural devlopment under any circumstances.
Allen W. Root, MD
|
Current GGH - Vol. 24 No. 1
print version (pdf)
Printer Friendly
Email Paper