Pediatrics and Neonatology (2018) 59, 147e153

Available online at www.sciencedirect.com

ScienceDirect

journal homepage: http://www.pediatr-neonatol.com

Original Article

Association of preterm birth and small for

gestational age with metabolic outcomes
in children and adolescents: A
population-based cohort study from Taiwan
Yu-Ting Huang a,c, Hsiang-Yu Lin a,c, Chung-Hsing Wang b,c,
Bai-Horng Su a,c,*, Che-Chen Lin d

a
Department of Neonatology, China Medical University Children’s Hospital, Taichung, Taiwan
b
Department of Genetics and Metabolism, China Medical University Children’s Hospital, Taichung,
Taiwan
c
School of Medicine, China Medical University, Taichung, Taiwan
d
Management Office for Health Data, China Medical University Hospital, Taichung, Taiwan

Received Jan 22, 2017; received in revised form Jun 21, 2017; accepted Jul 18, 2017
Available online 25 July 2017

Key Words Background: Previous studies have identified preterm birth and/or small for gestational age
metabolic outcome; (SGA) as risk factors for features of the metabolic syndrome, including high blood pressure, in-
preterm birth; sulin sensitivity and atherosclerosis, occurring later in life, with controversial results. We con-
small for gestational ducted this population-based cohort study to investigate metabolic outcomes in those with
age (SGA) former preterm birth and/or SGA status in Taiwan.
Methods: Data were obtained from Taiwan’s universal National Health Insurance Research
Database. From 1996 to 2004, 37,119 preterm infants, 3386 SGA infants, and 162,020 matched
controls were included. We investigated the risk of the metabolic disease, including hyperten-
sion, diabetes, and hyperlipidemia, which had been recorded by the end of 2008.
Results: The preterm and SGA cohort, combined into one, had a significantly increased risk of
developing metabolic disorders when compared with the comparison cohort (HR Z 2.46, 95%
CI Z 2.02e3.01). We observed that children with former preterm and SGA status in Taiwan had
a higher risk of developing hypertension (HR Z 3.24, 95% CI Z 1.58e6.67), Type 1 diabetes
mellitus (HR Z 1.80, 95% CI Z 1.05e3.07), Type 2 diabetes mellitus (HR Z 2.49, 95%
CI Z 1.98e3.14), and hyperlipidemia (HR Z 2.14, 95% CI Z 1.29e3.52).
Conclusion: Our study revealed the risk of metabolic disease in those with preterm birth

* Corresponding author. Department of Neonatology, China Medical University Children’s Hospital, 2, Yuh-Der Road, Taichung, 404,
Taiwan. Fax: þ886 4 2203 2798.
E-mail address: subh1168@gmail.com (B.-H. Su).

http://dx.doi.org/10.1016/j.pedneo.2017.07.007
1875-9572/Copyright ª 2017, Taiwan Pediatric Association. Published by Elsevier Taiwan LLC. This is an open access article under the CC BY-
NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
148
and/or SGA. Further studies with a longer duration of follow-up are required to confirm if
there is a tendency for the metabolic syndrome to develop in this study cohort.
Copyright ª 2017, Taiwan Pediatric Association. Published by Elsevier Taiwan LLC. This is an
open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/
by-nc-nd/4.0/).
1. Introduction
Preterm births (less than 37 completed weeks’ gestation)
account for almost 10% of all births worldwide, represent-
ing more than 15 million births globally every year.1 The
development of neonatal intensive care units and
advancement in the quantity of facilities and care skills for
prematurity in recent decades has dramatically improved
the prognosis regarding both morbidity and mortality for
very low birth weight (VLBW) infants (<1500 g). More than
50% of babies born at 24 weeks gestation, and 70%e90% of
VLBW infants of GA 28 weeks generally survive longer.
This amelioration has led to the growth of a population with
former preterm status and related comorbidities.2,3
A number of studies have identified preterm birth and/
or small for gestational ages (SGA) as risk factors for fea-
tures of the metabolic syndrome in later life, including high
blood pressure4e6 and insulin resistance.7,8 However, some
studies indicated a non-negative impact of preterm birth on
atherosclerosis,9 and some found a modest relation be-
tween arterial stiffness and gestational age and/or birth
weight,10 while others declared no evidence that reduced
insulin sensitivity in adulthood was associated with
prematurity.11
Because metabolic diseases increase the risk of heart
disease and cerebrovascular disease, we conducted this
population-based cohort study to investigate the associa-
tion between former preterm birth and/or SGA status and
subsequent metabolic outcomes in Taiwan and to empha-
size the importance of preventing or delaying any compo-
nent of the metabolic syndrome in this cohort.
2. Methods
2.1. Data source
Taiwan National Health Insurance (Taiwan NHI) is an offi-
cial, nationwide single-payer social health insurance pro-
gram organized by the National Health Insurance
Administration (NHIA) in 1995. This compulsory insurance
program covered nearly 98% of the population of 23 million
of Taiwan until 1998.
In this study, we utilized the database of children to
establish the population. The materials were obtained
from an NHI program from 1996 to 2008, with a randomly
selected sample, which represented half of all insured
children (age < 18 years) in Taiwan. The database con-
tained annual claim data, such as a registry of benefi-
ciaries, files of outpatients and inpatients, prescription
records, and records of other medical services. In this
database, all registries of disease were recorded from
Y.-T. Huang et al
outpatients’ and inpatients’ data, according to the Inter-
national Classification of Diseases, Ninth Revision (ICD-9).
In order to protect the privacy of the insureds, all personal
information was encrypted with proxy identification codes
before it was released to the research team. This study
was approved by the Ethics Review Board of the China
Medical University (the approval number is CMU-REC-101-
012).
2.2. Study population
This was a retrospective population-based cohort study. We
collected data on preterm infants (ICD-9-765) and SGA in-
fants (ICD-9-764, and excluding 764.2 and 764.9) as case
cohorts (preterm and SGA cohort) from 1996 to 2004. The
preterm & SGA cohort was also categorized into the two
following sub-cohorts: a preterm sub-cohort, comprising
infants with code ICD-9-765, and an SGA sub-cohort, con-
sisting of infants with code ICD-9-764 and excluding pre-
mature (ICD-9-765) infants. The comparison cohort was
selected from the normal birth infants and frequency
matched by gender, urbanization, and parental occupation
at 1:4 ratios. In this study, we intended to investigate the
risk of the major components of metabolic disease. The
constituents of metabolic disease were defined here as
individually diagnosed hypertension (ICD-9-401-404), type 1
diabetes mellitus (type 1 DM, ICD-9250.01, 250.03, 250.11,
250.13, 250.21, 250.23, 250.31, 250.33, 250.41, 250.43,
250.51, 250.53, 250.61, 250.63, 250.71, 250.73, 250.81,
250.83, 250.91, 250.93), type 2 DM (ICD-9250.00, 250.02,
250.10, 250.12, 250.20, 250.22, 250.30, 250.32, 250.40,
250.42, 250.50, 250.52, 250.60, 250.62, 250.70, 250.72,
250.80, 250.82, 250.90, 250.92), and hyperlipidemia (ICD-9-
272). Hypertension was also classified by the presence or
absence of amlodipine treatment. The two study cohorts
were followed up from each infant’s birthday and termi-
nated under the following circumstances: the individual
withdrew from the insurance program, the first constituent
event of metabolic disease took place, or the record
reached December 31, 2008.
We extracted the methods to define urbanization as
outlined by Liu.12 Liu used population density (people/
km2), the proportion of the population with different
educational levels, the proportion of elder people, the
proportion of agricultural workers, and the number of
physicians per 100,000 people to classify the residents’
urbanization level. Level 1 expressed the highest urbani-
zation level and level 4þ indicated the lowest. The
parental occupations were divided into three groups ac-
cording to working status: 1) white collar: long indoor
working hours, 2) blue collar: long outdoor working hours,
and 3) other: retirees, low-income people, etc.
Preterm birth and metabolic outcomes in children and adolescents 149

2.3. Statistical analysis
To present the structure of the study cohort, we demon-
strated the mean value and standard deviation (SD) of age
at each endpoint and numbers and percentages for cate-
gorical demographic factors. To assess the differences in
distribution between the preterm plus SGA cohort and
comparison cohort, we applied a t-test to age and a chi-
square test to categorical variables. The incidence den-
sity for each study cohort was calculated as the number of
metabolic disease events divided by the sum of follow-up
years and expressed per 10,000 person-years. The cumu-
lative incidence curves for each cohort were measured by
the KaplaneMeier method and the differences in these
incidence curves were further assessed by log-rank test. To
demonstrate the association between preterm and/or SGA
infants and the risk of metabolic disease, the single vari-
able and multivariable Cox proportional hazards model was
used to measure the hazard ratios (HRs) and 95% confidence
intervals (CIs).
Statistical analysis was performed with SAS 9.4 software
(SAS Institute, Cary, NC, USA). The cumulative incidence
curve was also drawn with SAS 9.4. A two-sided p-value of
less than 0.05 was considered to indicate significance.
3. Results
In this study, we analyzed the data from 37,119 preterm
infants, 3386 SGA infants, and 162,020 matched controls
(Table 1). There was no significant difference between the
preterm plus SGA cohort and the comparison cohort
(p > 0.05), distributed by gender, urbanization, and
parental occupation. At the study endpoint, the mean age
of the comparison cohort was 0.5 years greater than that of
the preterm plus SGA cohort (8.6 years vs. 8.1 years,
p < 0.0001).
The incidence of developing metabolic disease in the
comparison cohort was 1.87 per 10,000 person-years, and
that in the preterm plus SGA cohort was 4.60 per 10,000
person-years (Table 2). In the preterm and SGA cohorts, the
incidence of metabolic disease was around 2-fold and 3-fold
greater than that in the comparison cohort (4.39 and 6.81
vs. 1.87 per 10,000 person-years, respectively). After
adjustment for age, gender, urbanization, and parental
occupation, the preterm plus SGA cohort had a significantly
increased risk of developing metabolic disease compared
with the comparison cohort (HR Z 2.46, 95%
CI Z 2.02e3.01). Preterm and SGA infants had a 2.35-fold
(HR Z 2.35, 95% CI Z 1.90e2.89) and a 3.67-fold
(HR Z 3.67, 95% CI Z 2.33e5.78) increased risk of devel-
oping metabolic disease, respectively, compared with
normal infants.
Fig. 1 demonstrates the cumulative incidence curves of
the comparison cohort, the preterm plus SGA cohort, the
preterm cohort, and the SGA cohort. The cumulative inci-
dence curves over 12 years’ follow-up of the preterm plus
SGA, preterm, and SGA cohorts were significantly higher
than that of the comparison cohort (all p < 0.0001).
Table 3 shows the independent effects of preterm plus
SGA, preterm, and SGA on the individual and combined risk
of the components of metabolic disease. Relative to the
comparison cohort, all three subgroups had a higher inci-
dence rate of combined metabolic parameters (HR Z 2.46,
95% CI Z 2.02e3.01; HR Z 2.35, 95% CI Z 1.90e2.89;
HR Z 3.67, 95% CI Z 2.33e5.78, respectively). The pre-
term plus SGA cohort had an association with increased risk
of developing hypertension (HR Z 3.24, 95%
CI Z 1.58e6.67) and the preterm cohort had a 3.28-fold
higher risk of hypertension (HR Z 3.28, 95%
CI Z 1.57e6.87). There was a 2.49-fold increased risk of
type 2 DM (HR Z 2.49, 95% CI Z 1.98e3.14), a 1.80-fold risk
of type 1 DM (HR Z 1.80, 95% CI Z 1.05e3.07), and a 2.14-
fold risk of hyperlipidemia (HR Z 2.14, 95%
CI Z 1.29e3.52) relative to the comparison cohort. We
attempted to obtain data on prescription orders for Amlo-
dipine for hypertension. The result revealed that few hy-
pertensive children had been treated with Amlodipine (two

Table 1 Comparison of demographics between 4 cohorts of preterm, small for gestational age (SGA), preterm & SGA and
controls.
Variable Preterm SGA Preterm& SGA Controls p-value
N Z 37,119 N Z 3386 N Z 40,505 N Z 162,020
N (%) N (%) N (%) N (%)
Mean age of 8.04 (3.2) 8.7 (3.0) 8.1 (3.2) 8.6 (2.6) <0.0001
follow-up years (SD)a
Sex
Female 16,420 (44.2) 1901 (56.1) 18,321 (45.2) 73,284 (45.2) >0.99
Male 20,699 (55.8) 1485 (43.9) 22,184 (54.8) 88,736 (54.8)
Urbanization
1 (highest) 10,647 (28.7) 1059 (31.3) 11,706 (28.9) 46,824 (28.9) >0.99
2 11,054 (29.8) 1022 (30.2) 12,076 (29.8) 48,304 (29.8)
3 6927 (18.7) 552 (16.3) 36,976 (22.8) 9244 (22.8)
4 (lowest) 8491 (22.9) 753 (22.2) 9244 (22.8) 36,976 (22.8)
Parental occupations
White collar 21,509 (57.9) 1958 (57.8) 23,467 (57.9) 93,868 (57.9) >0.99
Blue collar 9572 (25.8) 864 (25.5) 10,436 (25.8) 41,744 (25.8)
Others 6038 (16.3) 564 (16.7) 6602 (16.3) 26,408 (16.3)
a
t test.
150 Y.-T. Huang et al

Table 2 Incidence rates and relative risks of metabolic disease for 4 cohorts of preterm, small for gestational age (SGA),
preterm & SGA and controls by multivariate Cox proportional hazards regression model.
Event PYs IR Crude HR Adjusted HR
95% CI 95% CI
Controls 260 1,391,969 1.87 reference reference
Preterm 131 298,540 4.39 2.35 (1.91, 2.90) 2.35 (1.90, 2.89)
SGA 20 29,366 6.81 3.63 (2.30, 5.72) 3.67 (2.33, 5.78)
Preterm & SGA 151 327,906 4.60 2.47 (2.02, 3.02) 2.46 (2.02, 3.01)
Model adjusted by sex, urbanization and parental occupation.
Metabolic disease: any one of hypertension, diabetes mellitus, or hyperlipidemia.
PYs: person-years; IR: incidence rate, per 10,000 person-years.

status. Although the SGA infants had a trend toward a

higher risk of metabolic disease than normal birth infants,
there was no significant difference in the development of
metabolic disease between the two groups if their parental
occupations were not white collar or if they lived in a rural
area.

4. Discussion

To our knowledge, our study is the largest nationwide,

Figure 1 Cumulative incidence in being metabolic disease
for cohorts of comparison and preterm & small for gestational
age (SGA) (upper panel); and for cohorts of comparison, pre-
term, and SGA (lower panel).
in the control group and one in the study group), and there
was no significant difference between the control and study
groups.
Table 4 demonstrates the effects of preterm and/or SGA
infant cohorts on metabolic disease risk stratified by de-
mographic status including gender, urbanization, and
parental occupation. Relative to the normal birth infant
group, the risk of metabolic disease was unanimously
increased for the preterm plus SGA infant group as a whole
and in the preterm cohort alone under each demographic
population-based study with longitudinal cohorts focusing
primarily on childhood and prophase adolescence. This
study shows that preterm and/or SGA children in Taiwan
have a significantly higher overall rate of developing hy-
pertension, DM, and hyperlipidemia in the early years of
life, regardless of gender, urbanization, and parental
occupation. We also found a significant increase in the risk
of metabolic disease if their parental occupations were
white collar, or they lived in rural areas. This phenomenon
could be interpreted by the inverse relationship between
physical activity and metabolic disease.13
Most literature discussing the risk of subsequent meta-
bolic syndrome in infants with premature gestational age
and/or substandard birth weight focuses on late adoles-
cence or adult life.5,14,15 Prematurity and growth restric-
tion due to an adverse intrauterine environment have been
associated with an increased risk of elevated blood pres-
sure in adults. The pathogenesis of essential hypertension
remains incompletely understood. In the past few years,
the role of SGA owing to maternal preeclampsia in the
development of essential hypertension in early adolescence
has been critically reviewed.16 SGA, as well as prematurity
independent of birth weight, may contribute to the devel-
opment of primary hypertension in adulthood. Babies that
are small at birth are more likely to have higher blood
pressure readings during adolescence, and continue to be
hypertensive as adulthoods.17 Impaired intrauterine growth
leading to impaired renal development has been proposed
as a possible mechanism to explain the association between
low birth weight and subsequent hypertension.18,19 How-
ever, a systematic review of 55 studies suggests a modest
inverse association between birth weight and subsequent
hypertension.20 In contrast, in a large cohort study, high
newborn and maternal weight were stronger predictors of
adult blood pressure than low birth weight.21 There might
be no unifying hypothesis for early origins of cardiovascular
disease. The fetal origins hypothesis along with rapid
Preterm birth and metabolic outcomes in children and adolescents 151

Table 3 Incidence rates and relative risks of types of metabolic disease for 4 cohorts of preterm, small for gestational age
(SGA), preterm & SGA and controls by multivariate Cox proportional hazards regression analysis.
Type of event Controls Preterm & SGA Preterm SGA
Event IR Event IR HR (95% CI) Event IR HR (95% CI) Event IR HR (95% CI)
Overall 260 1.87 151 4.60 2.46 (2.02, 3.01) 131 4.39 2.35 (1.90, 2.89) 20 6.81 3.67 (2.33, 5.78)
Hypertension 17 0.12 13 0.40 3.24 (1.58, 6.67) 12 0.40 3.28 (1.57, 6.87) 1 0.34 2.83 (0.38, 21.3)
Hypertension 2 0.01 1 0.03 2.10 (0.19, 23.1) 1 0.03 2.28 (0.21, 25.1) 0 0.00 e
þ amlodipine
Diabetes
Type 2 DM 197 1.42 116 3.54 2.49 (1.98, 3.14) 101 3.38 2.38 (1.87, 3.03) 15 5.11 3.65 (2.16, 6.18)
Type 1 DM 45 0.32 19 0.58 1.80 (1.05, 3.07) 17 0.57 1.76 (1.01, 3.08) 2 0.68 2.12 (0.51, 8.74)
Hyperlipidemia 46 0.33 23 0.70 2.14 (1.29, 3.52) 19 0.64 1.94 (1.14, 3.32) 4 1.36 4.02 (1.44, 11.2)
Model adjusted by sex, urbanization and parental occupation.
PYs: person-years; IR: incidence rate, per 10,000 person-years.

Table 4 Incidence rates and relative risks of metabolic disease stratified by demographic factors for 4 cohorts of preterm,
small for gestational age (SGA), preterm & SGA and controls by multivariate Cox proportional hazards regression analysis.
Demographic factors Controls Preterm & SGA Preterm SGA
Event IR Event IR HR (95% CI) Event IR HR (95% CI) Event IR HR (95% CI)
Sex
Female 94 1.48 68 4.55 3.06 (2.24, 4.18) 58 4.36 2.94 (2.12, 4.08) 10 6.07 3.99 (2.08, 7.67)
Male 166 2.19 83 4.65 2.12 (1.63, 2.76) 73 4.41 2.01 (1.53, 2.65) 10 7.76 3.51 (1.85, 6.64)
Urbanization
Urban 184 2.25 108 5.57 2.48 (1.96, 3.14) 91 5.18 2.30 (1.79, 2.95) 17 9.37 4.31 (2.62, 7.09)
Rural 76 1.33 43 3.21 2.42 (1.67, 3.52) 40 3.26 2.46 (1.68, 3.61) 3 2.67 2.02 (0.64, 6.40)
Parental occupations
White collar 173 2.16 97 5.13 2.37 (1.85, 3.04) 82 4.76 2.19 (1.69, 2.85) 15 8.81 4.25 (2.50, 7.20)
Blue collar 52 1.41 33 3.81 2.71 (1.75, 4.19) 31 3.92 2.81 (1.80, 4.38) 2 2.62 1.78 (0.43, 7.33)
Others 35 1.56 31 4.04 2.57 (1.50, 4.42) 18 3.81 2.43 (1.38, 4.29) 3 6.37 3.98 (1.22, 13.0)
Model adjusted by sex, urbanization and parental occupation.
PYs: person-years; Incidence rate: incidence rate, per 10,000 person-years.

postnatal “catch-up” in body weight may contribute to
some extent.22
Prematurity may also increase the risk of hypertension
via decreased glomerulogenesis, which is independent of
birth weight.4,23 This phenomenon may thus explain our
results of no significant increase in the risk of hypertension
in the SGA cohort but a significant increase in the preterm
plus SGA cohort and preterm cohort. A meta-analysis of 10
studies with 3083 individuals from eight countries reported
an association of prematurity with adolescent and adult
blood pressure (measured at the average age of 18 years).4
Those who were born prematurely had higher systolic blood
pressures with statistical significance (2.5 mmHg), regard-
less of weight. The results could be partly attributed to
altered programming of nephron numbers, which are
adversely affected by developmental stressors, such as
maternal, fetal, and childhood under- or over-nutrition.14
The findings mentioned above provided inconclusive evi-
dence of birth weight and/or prematurity contributing to
blood pressure levels in later life.
In the current study, we attempted to correlate the
severity of hypertension with prescription orders of Amlo-
dipine. We used the prescription code for Amlodipine
because it is the most frequently prescribed antihyper-
tensive agent. The results revealed that few hypertensive
children were treated with Amlodipine, possibly indicating
that hypertension in most of the children in this cohort
could be well- controlled by non-pharmacological inter-
vention, such as exercise, weight reduction, and diet con-
trol in this relatively short-term follow-up.
Preterm birth was linked with a modestly increased risk
of diabetes (both type 1 and 2 diabetes mellitus) in young
Swedish adults. These findings reveal the important public
health issues, given the increasing number of infants sur-
viving preterm birth and the huge morbid burden of dia-
betes.24 Small-for-age babies are more likely to have
metabolic abnormalities associated with hypertension or
later-developing coronary disease, accompanied by
increased insulin resistance,25 diabetes mellitus, and
hyperlipidemia,26 especially in those with higher abdominal
(visceral) adiposity.
Those with VLBW had significantly higher blood pressure,
higher fasting insulin, 2-h insulin, and 2-h glucose concen-
trations as well as a higher HOMA-IR index reflecting
increased insulin resistance, compared with young adults
who had been born at term, implying a higher risk for
progression of type 2 diabetes and cardiovascular mortality.
These differences were neither attributable to adult body
size nor composition, nor to fat distribution. Thus, VLBW
appeared to be associated with signs of insulin resistance
152
and impaired glucose regulation in early adulthood. How-
ever, some have suggested that young adults with VLBW
would appear to benefit from early-targeting preventive
interventions, such as prevention of rapid weight gaining
during childhood.27,28 The correlation between low birth
weight at term and adverse profile of later insulin meta-
bolism and glucose regulation has been validated.29 Sub-
jects with intrauterine growth retardation tended to have
more severe impaired insulin sensitivity than subjects with
appropriate birth weight at gestational age. This difference
was foreseen in the subjects’ prepubertal childhood30,31 or
later in their young adulthood.32,33
Prematurity (with both appropriate and small for
gestational age) has been proven to have reduced in insulin
sensitivity and secretion capacity, with or without early
accelerated or poor postnatal growth.8,34 In our study, we
analyzed the risks of type 1 DM and type 2 DM individually
due to the different pathophysiology, and observed a higher
incidence both of type 1 DM and type 2 DM in the preterm
plus SGA cohort and the preterm cohort than the compar-
ison cohort, which was not only consistent with previously
study findings but also expanded our knowledge of adverse
effects of former SGA status on insulin sensitivity. This
result may be explained from an epigenetic standpoint,
which needs to be verified by further investigation of the
molecular mechanism.
A meta-analysis consisting of 27 studies revealed that
adults who had been born premature had significantly
higher systolic blood pressure, diastolic blood pressure, 24-
h ambulatory systolic blood pressure and low-density lipo-
protein levels than their at term-born counterparts. Their
findings were also in accordance with ours in that the
experimental groups had less favorable blood pressure and
lipid profiles than the control group.35
This study had some limitations. Firstly, the ICD-9-based
registry data base provided qualitative but not quantitative
data. Therefore, some information was not available,
including the severity of impaired glucose metabolism,
dyslipidemia, or hypertension. Secondly, the diagnosis and
coding of hypertension, diabetes, and hyperlipidemia are
performed according to patents’ clinical symptoms and
signs and thus are highly specific. However, pediatric hy-
pertension and hyperlipidemia are even underdiagnosed
because blood pressure and lipid profiles are rarely
measured in children. Lastly, relatively short follow-up
times may only reveal the tendency toward a higher cu-
mulative risk of metabolic disease in the preterm and/or
SGA groups compared with the control group, as indicated
in Fig. 1. In further studies, we should continue to observe
the tendency for metabolic syndrome to develop in these
cohorts.
At presents, premature neonates may differ both in
mortality and morbidity from those born one to two de-
cades ago, as reported in our study and others. Fortified
nutrition and improved care quality have changed the
growth rates of premature infants, possibly resulting in
subsequent metabolic disease. Our study revealed the risk
of metabolic disease in the preterm and/or SGA cohort, and
to our knowledge, this is the first population cohort study in
Taiwan to point out the importance of preventing or
delaying any component of metabolic syndrome in this
cohort, mainly with lifestyle modifications, including
Y.-T. Huang et al
exercise, weight reduction, and diet control. Further
studies with longer follow-up times are required to confirm
if there is a tendency for the metabolic syndrome to
develop in this study cohort.
Conflicts of interest
The authors have no conflicts of interest relevant to this
article.
Acknowledgement
This study is supported in part by Taiwan Ministry of Health
and Welfare Clinical Trial and Research Center of Excel-
lence (MOHW104-TDU-B-212-113002), Bureau of Health
Promotion (DOH99-HP-1205), China Medical University
Hospital, Academia Sinica Taiwan Biobank Stroke Bio-
signature Project (BM104010092).
References
1. Howson CP, Kinney MV, Lawn JE. March of Dimes, PMNCH, save
the children, WHO: born too soon. In: The global action report
on preterm birth. Geneva: World Health Organization; 2012.
2. Saigal S, Doyle LW. An overview of mortality and sequelae of
preterm birth from infancy to adulthood. Lancet 2008;371:
261e9.
3. Su BH, Hsieh WH, Hsu CH, Chang JH, Lien R, Lin CH. Neonatal
outcomes of extremely preterm infants from Taiwan: com-
parison with Canada, Japan, and the USA. Pediat Neonatol
2015;56:46e52.
4. de Jong F, Monuteaux MC, van Elburg RM, Gillman MW,
Belfort MB. Systematic review and meta-analysis of preterm
birth and later systolic blood pressure. Hypertension 2012;59:
226e34.
5. Johansson S, Iliadou A, Bergvall N, Tuvemo T, Norman M,
Cnattingius S. Risk of high blood pressure among young men
increases with the degree of immaturity at birth. Circulation
2005;112:3430e6.
6. Norman M. Preterm birth e an emerging risk factor for adult
hypertension? Semin Perinatol 2010;34:183e7.
7. Hovi P, Andersson S, Eriksson JG, Järvenpää AL,
Strang-Karlsson S, Mäkitie O, et al. Glucose regulation in young
adults with very low birth weight. N Engl J Med 2007;356:
2053e63.
8. Hofman PL, Regan F, Jackson WE, Jefferies C, Knight DB,
Robinson EM, et al. Premature birth and later insulin resis-
tance. N Engl J Med 2004;351:2179e86.
9. Singhal A, Kattenhorn M, Cole TJ, Deanfield J, Lucas A. Preterm
birth, vascular function, and risk factors for atherosclerosis.
Lancet 2001;358:1159e60.
10. Cheung YF, Wong KY, Lam BC, Tsoi NS. Relation of arterial
stiffness with gestational age and birth weight. Arch Dis Child
2004;89:217e21.
11. Willemsen RH, Leunissen RW, Stijnen T, Hokken-Koelega AC.
Prematurity is not associated with reduced insulin sensitivity in
adulthood. J Clin Endocrinol Metab 2009;94:1695e700.
12. Liu CY, Hung YT, Chuang YL, Chen YJ, Weng WS, Liu JS, et al.
Incorporating development stratification of Taiwan townships
into sampling design of large scale health interview survey. J
Health Manag 2006;4:1e22.
13. Stensvold D, Nauman J, Nilsen TI, Wisløff U, Slørdahl SA,
Vatten L. Even low level of physical activity is associated with
reduced mortality among people with metabolic syndrome, a
Preterm birth and metabolic outcomes in children and adolescents
population based study (the HUNT 2 study, Norway). BMC Med
2011;9:109.
14. Li S, Zhang M, Tian H, Liu Z, Yin X, Xi B. Preterm birth and risk
of type 1 and type 2 diabetes: systematic review and meta-
analysis. Obes Rev 2014;15:804e11.
15. Luyckx VA, Brenner BM. Birth weight, malnutrition and kidney-
associated outcomesda global concern. Nat Rev Nephrol 2015;
11:135e49.
16. Ferreiral I, Peeters LL, Stehouwer CD. Preeclampsia and
increased blood pressure in the offspring: meta-analysis and
critical review of the evidence. J Hypertens 2009;27:1955e9.
17. Davies AA, Smith GD, May MT, Ben-Shlomo Y. Association be-
tween birth weight and blood pressure is robust, amplifies with
age, and may be underestimated. Hypertension 2006;48:
431e6.
18. Hoy WE, Hughson MD, Bertram JF, Douglas-Denton R, Amann K.
Nephron number, hypertension, renal disease, and renal fail-
ure. J Am Soc Nephrol 2005;16:2557e64.
19. Bergvall N, Iliadou A, Johansson S, de Faire U, Kramer MS,
Pawitan Y, et al. Genetic and shared environmental factors do
not confound the association between birth weight and hy-
pertension: a study among Swedish twins. Circulation 2007;
115:2931e8.
20. Huxley R, Neil A, Collins R. Unravelling the fetal origins hypoth-
esis: is there really an inverse association between birthweight
and subsequent blood pressure? Lancet 2002;360:659e65.
21. Filler G, Yasin A, Kesarwani P, Garg AX, Lindsay R, Sharma AP.
Big mother or small baby: which predicts hypertension? J Clin
Hypertens (Greenwich) 2011;13:35e41.
22. Singhal A, Lucas A. Early origins of cardiovascular disease: is
there a unifying hypothesis? Lancet 2004;363:1642e5.
23. Hughson MD. Low birth weight and kidney function: is there a
relationship and is it determined by the intrauterine environ-
ment? Am J Kidney Dis 2007;50:531e4.
24. Crump C, Winkleby MA, Sundquist K, Sundquist J. Risk of dia-
betes among young adults born preterm in Sweden. Diabetes
Care 2011;34:1109e13.
25. Phillips DI, Barker DJ, Hales CN, Hirst S, Osmond C. Thinness at
birth and insulin resistance in adult life. Diabetologia 1994;37:
150e4.
153
26. Barker DJ, Hales CN, Fall CH, Osmond C, Phipps K, Clark PM.
Type 2 (non-insulin-dependent) diabetes mellitus, hyperten-
sion and hyperlipidaemia (syndrome X): relation to reduced
fetal growth. Diabetologia 1993;36:62e7.
27. Nijpels G. Determinants for the progression from impaired
glucose tolerance to non-insulin-dependent diabetes mellitus.
Eur J Clin Invest 1998;28:8e13.
28. DECODE Insulin Study Group. Plasma insulin and cardiovascular
mortality in non-diabetic European men and women: a meta-
analysis of data from eleven prospective studies. Dia-
betologia 2004;47:1245e56.
29. Newsome CA, Shiell AW, Fall CH, Phillips DI, Shier R, Law CM. Is
birth weight related to later glucose and insulin metabolism? A
systemic review. Diabet Med 2003;20:339e48.
30. Hofman PL, Cutfield WS, Robinson EM, Bergman RN, Menon RK,
Sperling MA, et al. Insulin resistance in short children with
intrauterine growth retardation. J Clin Endocrinol Metab 1997;
82:402e6.
31. Arends NJ, Boonstra VH, Duivenvoorden HJ, Hofman PL,
Cutfield WS, Hokken-Koelega AC. Reduced insulin sensitivity
and the presence of cardiovascular risk factors in short pre-
pubertal children born small for gestational age (SGA). Clin
Endocrinol (Oxf) 2005;62:44e50.
32. Leger J, Levy-Marchal C, Bloch J, Pinet A, Chevenne D,
Porquet D, et al. Reduced final height and indications for in-
sulin resistance in 20 year olds born small for gestational age:
regional cohort study. BMJ 1997;315:341e7.
33. Jaquet D, Gaboriau A, Czernichow P, Levy-Marchal C. Insulin
resistance early in adulthood in subjects born with intrauterine
growth retardation. J Clin Endocrinol Metab 2000;85:1401e6.
34. Bazaes RA, Alegrı́a A, Pittaluga E, Ávila A, Iñiguez G, Mericq V.
Determinants of insulin sensitivity and secretion in very-low-
birth-weight children. J Clin Endocrinol Metab 2004;89:
1267e72.
35. Parkinson JR, Hyde MJ, Gale C, Santhakumaran S, Modi N.
Preterm birth and the metabolic syndrome in adult life: a
systematic review and meta-analysis. Pediatrics 2013;131:
e1240e63.