Our results suggest that chronic hypertension is a strong determinant of preterm SGA, independent of other factors, including maternal age, BMI, parity, and smoking status. Young and older maternal ages as well as recurrent early spontaneous abortions also conferred increased risk for preterm SGA, with modest or no effects on term SGA risk. These factors may involve an inadequate vascular response to pregnancy associated with abnormal placentation and may represent a pathogenesis distinct from that leading to term SGA.
An association between chronic hypertension and SGA risk has been reported in smaller studies(12
) butVillar et al, recently reported no increased risk for SGA among non-smoking women with chronic hypertension.(14
) The reason could be that his study was carried out in developing countries where other risk factors may be more important such as poor nutrition or infection (26
) . Kramer, et al reported that chronic hypertension was related to both term and preterm SGA among a large hospital-based cohort in Canada using birth weight growth standards(27
) but it is well accepted that preterm birth weights do not represent the weight of the entire fetal population. The reason for being born preterm may well impact fetal growth.(20
The evidence relating young or older maternal age to SGA risk has been conflicting, perhaps due to the heterogenous nature of SGA.(27
) Our data indicate that the effects of both young and old age are dominantly related to preterm SGA risk. It is likely that older age affects the ability of the maternal vasculature to adequately adapt to the demands of pregnancy and placentation, resulting in decreased placental perfusion.(31
) Results for women <20 years of age may be confounded by unmeasured social factors since many risk factors for SGA cluster in young pregnant women.(32
) However, it is also possible that implantation and placentation may be compromised in young woman due to biologic immaturity, perhaps when combined with poor nutritional status.(33
There is evidence that subfertility, or the causes of subfecundity, are associated with adverse pregnancy outcomes, including preterm birth (34
), perinatal loss (37
), preeclampsia (38
) and small for gestational age births.(34
) Our results extend this by providing evidence that recurrent early pregnancy losses were associated with increased risk of preterm but not term SGA; a long waiting time to pregnancy was related to preterm and term SGA. Thus, placental dysfunction may be implicated in some cases of SGA related to subfecundity but this warrants further study. Our data also provide intriguing evidence that subfertility may interact with smoking status such that risk of SGA is elevated only among non-smokers. It is possible that the profound effects of smoking eclipse the perhaps more subtle relationship of other factors on risk for SGA. Alternatively, smoking may be on the causal pathway, or be a collider, in the paths leading to subfertility,(39
) early spontaneous pregnancy loss, (40
) as well as SGA.
Our finding that smoking, parity and underweight status have similar effects on both term and preterm SGA suggest that these factors likely influence fetal growth in the second half of pregnancy. While vascular capacity expands dramatically in the first 20 weeks of pregnancy, nutritional needs are paramount later in gestation. (43
) Our results suggest that smoking, parity and underweight status may affect the maternal capacity to provide adequate nutritional substrate to meet the large cell division that occurs in the second and third trimester.
We have previously reported that chronic hypertension conferred a 3.4-fold increase in risk of preeclampsia among nulliparous women, and risk of preterm preeclampsia increased 5.4-fold. (44
) Here we show a similar 5-fold elevation among women with definite hypertension for preterm SGA. Taken together, these results demonstrate a strong and convincing relationship between chronic hypertension and risk for both preeclampsia and SGA, especially for the more severe subtypes of each condition. We also found that, in contrast to preeclampsia, SGA risk was inversely related to BMI providing support for the possibility that in the presence of more profound vascular damage, high BMI may predispose to preeclampsia. (45
Important strengths of our study include the large and well-characterized nature of the Danish National Birth Cohort that allows for investigation of relatively rare pregnancy outcomes, such as term and preterm SGA. The cohort was established within a tax paid public health care system where more than 99% of women access prenatal care and where standardized and validated diagnostic as well as outcome information are available in national registers.(15
Limitations to our study include the fact that the majority of the population in Denmark is caucasian, and results may well be different in other ethnic groups. We relied on self-report of some study variables, although we validated these data with diagnostic or other confirmatory information when available. For example, self-reported pre-existing hypertension was classified as definite when women also reported taking anti-hypertensive medication and there was high agreement between self-reported abortion information when compared to those in the Hospital Discharge Register. However, we were unable to evaluate actual blood pressure measures. Although we were not able to distinguish cases of transient hypertension during pregnancy, the link between this pregnancy complication and SGA risk is equivocal making it an unlikely source of confounding. BMI was based on self-reported height which tends to be overestimated, and self-reported weight which tends to be underestimated. (47
) Despite this, a validation study of over 5000 women in the DNBC indicated that 91.4% of women were allocated to the appropriate BMI category based on self-reported height and weight suggesting any bias in these data is minimal. (48
Our results indicated that chronic hypertension without superimposed preeclampsia increased the risk of preterm SGA substantially. In addition, young or older maternal age and recurrent early spontaneous abortions increased risk for preterm SGA, suggesting that these factors may disrupt fetal growth very early in gestation perhaps due to abnormal placentation. In contrast, smoking, parity and low maternal pre-pregnancy BMI had similar effects on preterm and term SGA suggesting a later pregnancy pathogenesis.