Response to Letter Regarding Article, “Evidence of Arteriolar Narrowing in Low-Birth-Weight Children”
We would like to thank Drs Degraeuwe and Been for their considered response to our recent report1 linking birth parameters, particularly low birth weight, to narrower arteriolar caliber in 6-year-old children, independent of blood pressure and other confounders. After conducting further analyses to test their suggestions, we report the following findings.
Degraeuwe and Been recommended that as birth weight, birth length, and head circumference are first and foremost related to gestational age, this parameter should be included continuously rather than categorically in the regression. After reanalyzing our data and adjusting for gestational age in weeks, for each SD decrease in birth weight, the decrease in arteriolar caliber was marginally greater (1.74 μm, P=0.005) than when using the categorical World Health Organization definition of prematurity (<37 weeks; 1.66 μm, P=0.003).
In attempting to disentangle the effects of gestational age and reduced intrauterine growth, which is reflected by low birth weight, they also recommended that each birth parameter be expressed as standard deviation scores of the respective means per week of gestational age (Z scores). After reanalyzing with this approach, the birth weight association in the regression was essentially unchanged (1.67 μm, P=0.0008), whereas the birth length (now 0.74 μm, P=0.158) and head circumference (now 0.76 μm, P=0.138) associations became nonsignificant, supporting the contention by Degraeuwe and Been that birth weight, as the major measure of placental insufficiency, should be predominantly affected.
Lastly, they recommended applying individualized fetal growth curves to the analysis, while accounting for all known birth-weight determinants: gestational age, ethnicity, infant gender, parity/birth rank, maternal height, and pre-gravid weight.2 We agree that this approach would be ideal, but collecting detailed maternal antenatal data with the limitations of our large population survey was not feasible. This could be addressed in future studies.
Understanding the relative contributions of birth weight and gestation to long-term health outcomes, including microvascular growth, is critical, given their differing mechanisms and potential for intervention.3 Degraeuwe and Been emphasized that birth weight should be the predominant outcome from fetal growth restriction and were supported by the Z score analyses of our data that they suggested, confirming that low birth weight had the major effect on retinal arteriolar narrowing at 6 years of age. The Z score approach can also separate infants with low birth weight due to prematurity who had not had time to reach their growth potential.
In support of this, we previously compared the contributions of low birth weight and prematurity to blood pressure levels in this cohort.4 When both birth weight and gestational age were modeled together, only birth weight remained significantly and inversely associated with blood pressure; each 1-kg decrease was associated with a 1.8-mm Hg increase (95% confidence intervals 0.1, 3.4, P=0.04) in systolic blood pressure.
Our findings generally agree with the idea posited by Degraeuwe and Been that low birth weight (after accounting for the proportion resulting from shorter gestation) may be the principal marker of early life disadvantage that underlies the fetal origins of adult diseases,5 including altered microvascular structure and its potential links to hypertension in later years. These lasting effects of low birth weight are particularly relevant given the increasing numbers of very-low-birth-weight babies now surviving into adulthood.
Mitchell P, Liew G, Rochtchina E, Wang JJ, Robaei D, Cheung N, Wong TY. Evidence of arteriolar narrowing in low-birth-weight children. Circulation. 2008; 118: 518–524.
Liew G, Wang JJ, Mitchell P. Which is the better marker for susceptibility to disease later in life: low birthweight or prematurity? Arch Dis Child. 2008; 93: 450.
Barker DJ, Bagsby SP. Developmental antecedents of cardiovascular disease: a historical perspective. J Am Soc Nephrol. 2005; 16: 2537–2544.