Shorter mothers have shorter pregnancies

We assessed whether maternal height was associated with gestational age in a cohort of 294 children born at term. Increasing maternal height was associated with longer pregnancy duration (p = 0.002). Stratified analyses showed that the main effect on pregnancy length appears to occur among shorter mothers (<165 cm tall), whose pregnancies were ∼0.6 and ∼0.7 weeks shorter than pregnancies of mothers 165-170 cm (p = 0.0009) and >170 cm (p = 0.0002) tall, respectively. Further, children of shorter mothers were more likely to be born early term than those of average height (p = 0.021) and taller (p = 0.0003) mothers. Maternal stature is likely to be a contributing factor influencing long-term outcomes in the offspring via its effect on pregnancy length.


Introduction
Th e factors determining gestational length in humans have not been adequately described. In the case of preterm births, maternal nutrition in the periconceptional period has been described as a possible causative factor. Recently, maternal stature has been also been linked to preterm birth, with a number of studies showing that shorter women are at a greater risk of delivering preterm babies (Han et al. 2012, Myklestad et al. 2013.
However, no studies seem to have examined whether maternal height also aff ects length of gestation among children born at term, which extends across a range from 37 0/7 to 41 6/7 weeks ' gestation. In light of the above fi ndings, we aimed to assess for the fi rst time whether maternal height is associated with gestational age in a cohort of children born at term.

Methods
We assessed 157 boys and 137 girls who were the control cohort recruited for a study examining the impact of conception with ovarian stimulation drugs on growth and metabolism in childhood (Savage et al. 2012). All children were of New Zealand European ethnicity, relatively high socioeconomic status, naturally conceived, from singleton pregnancies and born at term (37 0/7 to 41 6/7 weeks ' gestation). Exclusion criteria were being born small-forgestational-age (birth weight Ͻ Ϫ 2 standard deviation scores (SDS)), or born to mothers with gestational diabetes, chronic illness or maternal drug use (including tobacco) during pregnancy.
Gestational age was determined by ultrasound scans performed Ͻ 20 weeks ' gestation. Maternal and paternal heights were measured using a Harpenden Stadiometer to the nearest mm. Birth weight data were transformed into SDS (Niklasson et al. 1991).
Associations with maternal height were assessed using linear regression mixed models in SAS v.9.3 (SAS Institute, Cary, NC, USA), which included family identifi cation number as a random factor. Models also accounted for infant ' s gender, birth order, maternal age and paternal height. Stratifi ed analyses were carried out by subdividing the study cohort into three groups: shorter ( Ͻ 165 cm), average (165 -170 cm) and tall ( Ͼ 170 cm) women. Pregnancy length was compared using multivariate models, and the proportion of children born at early term (37 0/7 -38 6/7 weeks ' gestation) was compared using Fisher ' s exact tests. Demographic data are provided as mean Ϯ standard deviation; other data are means and 95% confi dence intervals adjusted for confounders in multivariate models.
Th e Northern Y Regional Ethics Committee provided ethics approval. Written informed consent was obtained from parents/guardians, and verbal or written consent from each child.

Results
We studied 294 children (53.4% boys) born of 213 mothers and 213 fathers (as all siblings were born to the same mother and father). Mean maternal height was 167.3 Ϯ 6.1 (range: 151.4 -183.0) cm, slightly higher than the estimated average height of New Zealand women at 165 cm. Increasing maternal height was associated with longer pregnancy duration ( ß ϭ 0.036; p ϭ 0.002). There was also a progressive increase in birth weight SDS with increasing maternal height ( ß ϭ 0.045; p Ͻ 0.0001).
Stratifi ed analyses showed that the main eff ect on pregnancy length appears to occur among shorter mothers ( Ͻ 165 cm tall), whose pregnancies were ∼ 0.6 and ∼ 0.7 weeks shorter than those of mothers who were 165 -170 cm ( p ϭ 0.0009) and Ͼ 170 cm ( p ϭ 0.0002) tall, respectively ( Figure 1A). Further, 34% of children of shorter mothers were born at early term compared with 19% of average height ( p ϭ 0.021) and 12% of taller ( p ϭ 0.0003) mothers ( Figure 1B).
Paternal height was not associated with the gestational age ( p ϭ 0.67) or birth weight SDS ( p ϭ 0.68) of the off spring. We assessed whether maternal height was associated with gestational age in a cohort of 294 children born at term. Increasing maternal height was associated with longer pregnancy duration ( p = 0.002). Stratifi ed analyses showed that the main eff ect on pregnancy length appears to occur among shorter mothers (<165 cm tall), whose pregnancies were ~0.6 and ~0.7 weeks shorter than pregnancies of mothers 165-170 cm ( p = 0.0009) and >170 cm ( p = 0.0002) tall, respectively. Further, children of shorter mothers were more likely to be born early term than those of average height ( p = 0.021) and taller ( p = 0.0003) mothers. Maternal stature is likely to be a contributing factor infl uencing long-term outcomes in the off spring via its eff ect on pregnancy length.
Keywords: maternal stature , length of gestation , term , height , gestational age

Discussion
Th is study shows that maternal height is positively associated with pregnancy length across the term 5-week window. As previously discussed, a number of studies have shown an association between shorter maternal stature and preterm birth (Han et al. 2012, Myklestad et al. 2013. Our fi ndings have extended these observations, showing that maternal short stature is associated with shorter gestational length within term pregnancies, so that shorter mothers are more likely to deliver early-term babies.
Th e breakdown of term pregnancies into early term (37 0/7 -38 6/7 weeks ' gestation), full term (39 0/7 -40 6/7 weeks ' gestation) and late term (41 0/7 -41 6/7 weeks ' gestation) has been recently proposed (American College of Obstetrics and Gynecology, 2013). Studies have shown that early-term infants have greater mortality and morbidity in the neonatal period and early infancy than those born at full term (Zhang andKramer, 2009, Reddy et al., 2011). Th ere is also evidence from large studies that neurocognitive outcomes in childhood and adulthood are worse in those born at early term (MacKay et al. 2010, Yang et al. 2010a, Yang et al. 2010b. As a result, maternal stature is likely to be a contributing factor infl uencing long-term outcomes in the off spring via its eff ect on pregnancy length. Th e mechanisms underpinning the association of maternal stature and pregnancy length are still unclear. Ozaltin et al. pro-posed that mechanisms involved may be socioeconomic factors, undernutrition in utero and maternal anatomical constraints (Ozaltin et al. 2010). In our study, socioeconomic factors were not at play in light of the homogeneity of our cohort, and their relatively high socioeconomic status also means that undernutrition in utero is most unlikely to have been a factor.
Anatomical constraints are more likely to explain the observed diff erences in pregnancy length. Th is is hypothesis is supported by the strong positive association between off spring birth weight SDS and maternal height (but not paternal height). Further, the anatomical explanation is also supported by the lack of eff ect of paternal height on pregnancy length, as previously observed in another study (Myklestad et al. 2013). However, further studies are necessary before the mechanisms underpinning the eff ects of maternal stature on pregnancy length can be elucidated.

Declaration of interest:
Th is study was supported by Gravida: National Centre for Growth and Development (New Zealand). Th e authors have no fi nancial or non-fi nancial interests to declare that may be relevant to this submitted work. Th e funders had no role in study design, data collection and analysis, decision to publish or preparation of the manuscript. Figure 1. (A) Gestational age among term children born to mothers whose heights were Ͻ 165 ( n ϭ 88), 165 -170 ( n ϭ 105) and Ͼ 170 ( n ϭ 101) cm. Data are means and 95% confi dence intervals adjusted for confounding factors in the multivariate models. (B) Percentage of children born at early term (37 0/7 -38 6/7 weeks ' gestation) according to maternal height groups.