Sie sind auf Seite 1von 9


21, 2017



Preterm Birth and Risk of Heart Failure

Up to Early Adulthood
Hanna Carr, BS,a Sven Cnattingius, MD, PHD,a Fredrik Granath, PHD,a Jonas F. Ludvigsson, MD, PHD,b,c
Anna-Karin Edstedt Bonamy, MD, PHDa,d


BACKGROUND In small clinical studies, preterm birth was associated with altered cardiac structure and increased
cardiovascular mortality in the young.

OBJECTIVES The goal of this study was to determine the association between preterm birth and risk of incident heart
failure (HF) in children and young adults.

METHODS This register-based cohort study included 2,665,542 individuals born in Sweden from 1987 to 2012 who
were followed up from 1 year of age to December 31, 2013. The main study outcome was diagnosis of HF in the National
Patient Register or the Cause of Death Register. The association between preterm birth and risk of incident HF was
analyzed by using a Poisson regression model. Estimates were adjusted for maternal and pregnancy characteristics,
socioeconomic status, and maternal and paternal cardiovascular disease.

RESULTS During 34.8 million person-years of follow-up (median 13.1 years), there were 501 cases of HF. After exclusion
of 52,512 individuals with malformations (n ¼ 196 cases), 305 cases of HF remained (0.88 per 100,000 person-years).
Gestational age was inversely associated with the risk of HF. Compared with individuals born at term ($37 weeks’
gestation), adjusted incidence relative risks for HF were 17.0 (95% confidence interval [CI]: 7.96 to 36.3) after extremely
preterm birth (<28 weeks) and 3.58 (95% CI: 1.57 to 8.14) after very preterm birth (28 to 31 weeks). There was no risk
increase after moderately preterm birth (32 to 36 weeks) (relative risk: 1.36; 95% CI: 0.87 to 2.13).

CONCLUSIONS There was a strong association between preterm birth before 32 weeks of gestation and HF in
childhood and young adulthood. Although the absolute risk of HF is low in young age, our findings indicate that preterm
birth may be a previously unknown risk factor for HF. (J Am Coll Cardiol 2017;69:2634–42) © 2017 by the American
College of Cardiology Foundation.

B etween 5% and 13% of all live births occur

before term (<37 weeks of gestation) (1,2).
Although prematurity is still the main cause
of neonatal death globally, high-income countries
medical needs of these subjects, and for developing
appropriate preventive measures.
Results from previous studies suggest that survi-
vors of preterm birth are at increased risk of hyper-
have experienced dramatic increases in survival rates tension, stroke, and cardiovascular mortality but not
in preterm infants over the past few decades (2,3). ischemic heart disease (4–8). To the best of our
Knowledge about how the burdens of prematurity knowledge, the association between preterm birth
may be carried into later life in these steadily growing and risk of heart failure (HF) has not previously been
generations of new survivors is important for explored. HF in children and young adults is an un-
improving neonatal care, for meeting the future usual but dangerous condition with high mortality

Listen to this manuscript’s

audio summary by From the aClinical Epidemiology Unit, Department of Medicine Solna, Karolinska Institutet, Stockholm, Sweden; bDepartment of
JACC Editor-in-Chief Medical Epidemiology and Biostatistics, Karolinska Institutet, Stockholm, Sweden; cÖrebro University Hospital, Örebro, Sweden;
Dr. Valentin Fuster. and the dDepartment of Women’s and Children’s Health, Karolinska Institutet, Stockholm, Sweden. This study was funded by the
Swedish Research Council for Health, Working Life and Welfare (Dr. Bonamy, 2010-0643), Swedish Society for Medical Research
(Dr. Bonamy), Stockholm County Council (Dr. Bonamy, clinical research appointment), the Swedish Heart and Lung Foundation
(Dr. Bonamy, 20160578), and the Karolinska Institutet Distinguished Professor Award (Dr. Cnattingius). All other authors have
reported that they have no relationships relevant to the contents of this paper to disclose.

Manuscript received October 14, 2016; revised manuscript received February 19, 2017, accepted March 20, 2017.
JACC VOL. 69, NO. 21, 2017 Carr et al. 2635
MAY 30, 2017:2634–42 Preterm Birth and Risk of HF

rates (9,10). Congenital heart disease and cardiomy- covers all hospitalizations in Sweden from ABBREVIATIONS

opathies, particularly idiopathic dilated cardiomyop- 1987 onward, and information on hospital- AND ACRONYMS

athy, are the main causes of HF at young age (11–13). based outpatient care is included from 2001.
CI = confidence interval
Incidence data for pediatric HF are scarce but were The Cause of Death Register provides infor-
HF = heart failure
estimated to be 0.87 per 100,000 person-years in a mation on causes and dates of death in
ICD = International
study in the United Kingdom and Ireland of HF Sweden from 1961 (24). The Multi-Generation
Classification of Diseases
caused by cardiac muscle disease (14). Between 1987 Register was created in 2000, and it includes
RR = relative risk
and 2006, the incidence of HF among young adults in individual index-persons born after 1932 who
Sweden increased by 50%, and the proportion of were alive in 1961 and links them to their parents (25).
cardiomyopathies as an underlying cause of HF Information on educational level was collected from
increased from 15% to 25% (15). the Swedish Register of Education (26). Date of
Preterm birth entails exposure of the immature emigration was retrieved from the Register of the
infant heart to extrauterine conditions. Evidence Total Population (27).
from animal models and small studies of preterm EXPOSURES. Data on the main exposure (i.e., gesta-
infants shows that preterm birth interferes with tional age at birth) were retrieved from the Medical
normal cardiac development in the neonatal period Birth Register and categorized into 22 to 27 weeks
(16–19). In a cardiac imaging study of adults born (extremely preterm), 28 to 31 weeks (very preterm), 32
preterm, ventricular mass in adulthood was seen to to 36 weeks (moderately preterm), and $37 weeks
increase with lower gestational age at birth. Preterm (term). Since the early 1990s, all pregnant women in
birth was also associated with further alterations in Sweden are offered a diagnostic ultrasound scan in
cardiac structure and function (20). the early second trimester, usually between weeks 17
SEE PAGE 2643 and 20, and >96% accept (28). When no information
on ultrasound dating of pregnancy was available, the
We hypothesized that preterm birth is associated
last menstrual period was used for assessing gesta-
with an increased risk of later HF. In a nationwide
tional age.
Swedish cohort study including >2.6 million live
Data on size at birth were calculated as deviation
births, we investigated the association between
from the estimated weight for gestational age and
gestational age at birth and risk of incident HF in
sex, based on the Swedish reference curve for intra-
childhood and young adulthood.
uterine growth (29). Individuals were categorized as
very small (<2 SD), small (2 SD to <1 SD), appro-
priate (1 SD to 1 SD), large (>1 SD to 2 SD) or very
large (>2 SD). These data were also used to statisti-
cally correct for the possibility that an association
based cohort study included 2,665,542 individuals
between preterm birth and later HF is confounded by
born in Sweden and registered in the Medical Birth
low birth weight for gestational age, a proxy for poor
Register between 1987 and 2012 (Figure 1). Individuals
fetal growth.
were followed up from 1 year of age until death,
emigration, first diagnosis of HF or ischemic heart OUTCOMES. The primary outcome was a diagnosis of

disease, or end of study (December 31, 2013), which- incident HF without a previous diagnosis of ischemic
ever came first. Start of follow-up was set to 1 year of heart disease in the National Patient Register or the
age to avoid measuring HF as an immediate compli- Cause of Death Register. The International Classifi-
cation during neonatal care. cation of Diseases (ICD)-9th revision (ICD-9; used
A unique personal identity number given to all between 1987 and 1996) and 10 (ICD-10; introduced in
Swedish residents allows for comprehensive 1997) were used to define HF (ICD-9 code 428 and
cross-linking with other national registries (21). The ICD-10 code I50) and ischemic heart disease (ICD-9
caregivers are required by law to contribute informa- codes 410 to 414 and ICD-10 codes I20 to I25).
tion to these registries. The Medical Birth Register was OTHER VARIABLES. From the Medical Birth Register,
started in 1973 and covers >98% of all births in Sweden we included information on maternal factors such
(22). Since 1982, it is based on copies of standardized as age at delivery, country of birth, singleton or
clinical record forms used in all antenatal care clinics multiple pregnancy, diagnosis of hypertension, pre-
and delivery and neonatal wards in the country, and it eclampsia, diabetes mellitus, or gestational diabetes.
contains data on both mother and infant. The National Data on maternal smoking in the Medical Birth
Patient Register contains data on patient diagnoses Register were divided into 2 groups according
and medical and surgical procedures (23). The registry to information collected at the first antenatal visit,
2636 Carr et al. JACC VOL. 69, NO. 21, 2017

Preterm Birth and Risk of HF MAY 30, 2017:2634–42

STATISTICAL ANALYSIS. The association between

F I G U R E 1 Study Population
preterm birth and HF was assessed in a Poisson
regression analysis. Relative risks (RRs) (adjusted
Live births in Sweden 1987-2012, n=2,714,789 incidence rate ratios) and corresponding 95% confi-
dence intervals (CIs) for each gestational age category
were modeled by using log (person-years at risk) as an
Missing personal identification number, n=22,828
offset. Covariates for the adjusted models were cho-
Missing information on gestational age, n=3,217
Missing or unreasonable birth weight, n=8,896 sen on the basis of an association with our main
outcome at a level of p #0.20. Birth year period (1987
to 1995, 1996 to 2003, and 2004 to 2012), attained age
Death <1 year of age, n=9,399 during follow-up (in 5-year intervals), maternal age
Emigration <1 year of age, n=3,881
Diagnosis of heart failure <1 year of age, n=753 and education, subject sex, and birth weight were
Diagnosis of ischemic heart disease <1 year of age, n=81 included as covariates in the first adjusted model (or
Did not reach 1 year of age before 31 Dec 2013, n=192
gestational age in analyses of the association between
birth weight for gestational age and risk of HF). In the
Final study population, n=2,665,542 second model, we also adjusted for maternal or
paternal HF or ischemic heart disease. Adjusted
models 1 and 2 were applied both before and after
Flow chart of inclusions and exclusions in the study. excluding individuals with malformations. Because of
the large number of missing data on maternal smok-
ing among preterm individuals, we adjusted for
usually in weeks 8 to 12 of pregnancy. This informa- maternal smoking habits in a separate model,
tion has been routinely collected since 1983, but data including individuals with complete data on all
are sometimes missing, particularly for mothers of covariates.
preterm individuals. Information on highest attained Incidence rates for HF were also calculated on the
level of maternal education was retrieved from the basis of gestational age at birth and attained age at
Education Register and was categorized as #9 years, time of diagnosis (Online Table 1). Using a Poisson
10 to 12 years, or $13 years of education. regression model, we estimated unadjusted incidence
Individuals born with malformations that could rate ratios for the same age intervals, also presented
possibly correlate with risk of HF were identified by in Online Table 1. All data were analyzed by using SAS
searching the Medical Birth Register and the National version 9.4 software (SAS Institute, Inc., Cary, North
Patient Register for 1 of the following ICD-9 or ICD-10 Carolina).
diagnoses: malformations of the circulatory system ETHICS. Ethical permission was obtained from
(745.0 to 747.9 or Q20 to Q28), congenital malforma- the Regional Ethical Vetting Board in Stockholm
tion syndromes due to known exogenous causes not (Etikprövningsnämnden) with the registration
classified elsewhere (759.8 or Q86), other specified number 2011/195-31/2.
congenital malformation syndromes affecting multi-
ple organ systems (743.0/755/756.0/756.7/756.8/757.1/ RESULTS
758.6/759.8 or Q87), other congenital malformations
not classified elsewhere (759.0 to 759.9 or Q89), or Among 2,665,542 individuals included in the study,
chromosomal abnormalities (758.0 to 758.9 or Q90 to 156,879 (5.9%) were born preterm; 5.14% were
Q99). Children with a diagnosis of patent ductus moderately preterm, 0.56% very preterm, and 0.18%
arteriosus (747A or Q25.0) were not excluded from extremely preterm (Online Table 2). Preterm
our analysis because this condition is very common individuals were more often low birth weight for
after preterm birth and is a possible mediator of HF. gestational age than individuals born at term.
By using the Multi-Generation Register, it was Mothers of preterm individuals were more likely to be
possible to trace the registered father for almost 99% of younger (#19 years of age) or older ($35 years of age),
individuals in the cohort. This approach enabled a to have lower levels of education, to be smokers, and
search for both maternal and paternal diagnoses of HF of non-Nordic origin. Maternal pregnancy complica-
or ischemic heart disease or death from HF or ischemic tions and multiple pregnancies were more common in
heart disease in the National Patient Register and Cause women with preterm births.
of Death Register, using the aforementioned ICD-9 During follow-up (beginning at 1 year of age),
and ICD-10 codes plus the earlier ICD version 8 codes there were 501 cases of HF. Three of these cases were
for HF (428) and ischemic heart disease (410 to 414). deaths caused by HF. Total time of follow-up was
JACC VOL. 69, NO. 21, 2017 Carr et al. 2637
MAY 30, 2017:2634–42 Preterm Birth and Risk of HF

34.8 million person-years, yielding an incidence of

T A B L E 1 Cohort Characteristics in Relation to Outcome: Children Born in Sweden
1.4 per 100,000 person-years. After exclusion of 1987–2012, Follow-Up From 1 Yr of Age
52,512 individuals born with malformations (as
Unadjusted IRR
specified in the Patients and Methods section), there No. of Incidence (95% CI)
were 305 cases of HF in 34.2 million person-years of Total % Cases Rate* for HF

follow-up (incidence 0.89 per 100,000 person-years). All subjects 2,665,542 100.0 501 1.44 –

The median individual time of follow-up for all Subject characteristics

subjects was 13.1 years (IQR: 6.1 to 20.1 years).
Female 1,296,690 48.6 211 1.25 1.00 (reference)
Cohort characteristics in relation to outcome are
Male 1,368,852 51.4 290 1.62 1.30 (1.09–1.56)
presented in Table 1. Individuals diagnosed with HF
Birth period
were more often male than the healthy population. 1987–1995 1,009,608 37.9 381 1.77 1.72 (1.25–2.35)
Mothers of children later diagnosed with HF were 1996–2003 711,370 26.7 77 0.84 0.81 (0.56–1.18)
more often smokers and had lower levels of educa- 2004–2012 944,564 35.4 43 1.03 1.00 (reference)
tion. Maternal and paternal HF or ischemic heart Malformations

disease was also more frequent in subjects with HF. Yes 52,512 1.98 196 30.7 34.5 (28.8–41.2)
No 2,613,030 98.0 305 0.89 1.00 (reference)
Patent ductus arteriosus
incidence rates of HF in relation to gestational age Yes 8,840 0.33 45 43.6 33.2 (24.4–45.1)
at birth are presented in Table 2. Incidence rates of No 2,656,702 99.7 456 1.31 1.00 (reference)
HF were inversely related to gestational age at birth. Maternal and pregnancy characteristics
Preterm birth was associated with an increased risk Age

of HF across all 3 categories of prematurity, and risks #19 yrs 38,255 1.44 12 2.15 1.39 (0.78–2.50)
20–24 yrs 389,956 14.6 85 1.46 0.95 (0.73–1.22)
increased with decreasing gestational age.
25–29 yrs 861,936 32.3 187 1.55 1.00 (reference)
After exclusion of individuals with major congen-
30–34 yrs 868,124 32.6 143 1.34 0.87 (0.70–1.08)
ital malformations and adjustment for maternal 35–39 yrs 415,752 15.6 54 1.15 0.75 (0.55–1.01)
characteristics, subject sex, birth period, and birth $40 yrs 91,519 3.43 20 2.01 1.30 (0.82–2.07)
weight for gestational age, the risk of HF was 17 times Education
higher in subjects born extremely preterm, and 3.6 #9 yrs 264,007 10.0 76 2.17 1.75 (1.34–2.29)

times higher in subjects born very preterm, compared 10–12 yrs 1,203,184 45.6 245 1.45 1.17 (0.97–1.42)
$13 yrs 1,174,185 44.5 176 1.24 1.00 (reference)
with subjects born at term (Table 2). Additional
Missing data 24,166 4 – –
adjustment for parental cardiovascular disease only
Smoking habits (at first antenatal visit)
minimally attenuated relative risks for HF. Adjusting
Nonsmoker 2,152,861 85.5 361 1.36 1.00 (reference)
for maternal smoking habits did not alter the Smoker 366,709 14.6 102 1.66 1.22 (0.98–1.52)
described associations (Online Table 3). There was no Missing data 145,972 38 – –
significant increase in risk of HF for subjects born Country of birth
moderately preterm in the adjusted models. Sweden 2,194,530 82.7 418 1.41 1.00 (reference)

The median age at diagnosis of HF was 16.5 years Other Nordic country 71,440 2.69 19 1.75 1.11 (0.85–1.44)
Other 387,934 14.6 62 1.56 1.23 (0.78–1.95)
(IQR: 5.2 to 19.7 years). Online Table 1 displays age-
Missing data 11,638 2 – –
specific incidence rates for HF in relation to gesta-
Hypertensive disease
tional age at birth. Incidence rates dropped after the No 2,572,282 96.5 484 1.44 1.00 (reference)
first 5 years of life and then rose again after 16 years of Hypertension 16,310 0.61 2 1.02 0.71 (0.18–2.83)
age. Subjects born before 32 weeks of gestation had Preeclampsia 76,950 2.89 15 1.54 1.07 (0.64–1.79)
the highest incidence rates of HF across all age cate- Diabetes
gories. The same pattern was seen after excluding No 2,630,853 98.7 498 1.45 1.00 (reference)
Diabetes mellitus 10,666 0.40 1 0.71 0.49 (0.12–1.95)
subjects born with malformations.
Gestational diabetes 24,023 0.90 2 0.76 0.53 (0.07–3.76)
Multiple pregnancy 71,997 2.70 17 1.88 1.32 (0.81–2.14)
OF HF. We also found an association between low Maternal or paternal HF or ischemic heart disease
birth weight for gestational age and increased risk of Maternal 17,981 0.67 16 4.61 3.28 (1.99–5.40)
HF (Table 3). Compared with infants born with Paternal 65,682 2.47 24 1.90 1.34 (0.89–2.03)
appropriate birth weight for gestational age, those Missing data 21,058 – – –

born very small for gestational age (>2 SDs below the
*Events per 100,000 person-yrs.
mean) had a 3-fold risk of subsequent HF in the CI ¼ confidence interval; HF ¼ heart failure; IRR ¼ incidence rate ratio.
unadjusted analysis. After excluding subjects with
2638 Carr et al. JACC VOL. 69, NO. 21, 2017

Preterm Birth and Risk of HF MAY 30, 2017:2634–42

T A B L E 2 Associations Between Gestational Age at Birth and Incident HF: Unadjusted and Adjusted IRRs (95% CIs) for Incident HF in
Relation to Gestational Age at Birth

Unadjusted IRR Adjusted Model 1: IRR Adjusted Model 2: IRR

N No. of Events Incidence Rate* (95% CI) (95%CI)† (95% CI)‡

All subjects, N ¼ 2,665,542

Gestational age at birth
<28 weeks 4,845 11 20.1 15.0 (8.25–27.3) 13.0 (7.08–23.8) 12.9 (7.06–23.7)
28–31 weeks 14,951 9 4.71 3.52 (1.82–6.80) 2.60 (1.33–5.09) 2.60 (1.33–5.08)
32–36 weeks 137,083 42 2.32 1.73 (1.26–2.38) 1.54 (1.11–2.12) 1.54 (1.11–2.12)
$37 weeks 2,508,663 439 1.34 1.00 (reference) 1.00 (reference) 1.00 (reference)
Subjects with malformations excluded, n ¼ 2,613,030
Gestational age at birth
<28 weeks 4,219 7 14.2 16.8 (7.95–35.7) 17.1 (8.00–36.4) 17.0 (7.96–36.3)
28–31 weeks 13,656 6 3.38 4.01 (1.79–9.01) 3.57 (1.57–8.11) 3.58 (1.57–8.14)
32–36 weeks 131,522 21 1.20 1.43 (0.92–2.23) 1.36 (0.87–2.12) 1.36 (0.87–2.13)
$37 weeks 2,463,328 271 0.84 1.00 (reference) 1.00 (reference) 1.00 (reference)

*Events per 100,000 person-yrs. †Model 1: adjusted for maternal age and education, subjects’ period of birth, attained age during follow-up, sex, and birth weight for
gestational age. ‡Model 2: in addition to the factors noted in model 1, also adjusted for maternal or paternal HF or ischemic heart disease.
Abbreviations as in Table 1.

malformations and taking maternal characteristics pressure may cause organ damage such as left ven-
and gestational age into account in the adjusted tricular hypertrophy (31). Thus, it is plausible that an
models, this association was weakened and no longer elevation of blood pressure may contribute to the
significant. There was no indication of confounding increased risk of HF observed in preterm individuals
by parental cardiovascular disease. in our study.
Ischemic heart disease is a major cause of HF in
DISCUSSION adults (32) but comprises a very small proportion of
HF in children and adolescents (15,33). Thus far, there
PRINCIPAL FINDINGS. This registry-based cohort have been no data confirming a link between preterm
study of >2.6 million children and young adults birth and ischemic heart disease (6,34,35). In the
found that preterm birth was associated with an present study, individuals with ischemic heart dis-
increased risk of incident HF, also after adjustment ease as first event were censored and no longer
for birth weight for gestational age and potential contributed risk time in the study. Thus, ischemic
confounders. The RR was inversely related to gesta- heart disease is an unlikely explanation for the
tional age at birth (Central Illustration). Individuals observed association between preterm birth and HF.
born extremely preterm and very preterm faced a 17- Instead, cardiomyopathies, including the diagnosti-
fold and >3-fold increased risk of HF, respectively; cally broad “idiopathic dilated,” are considered a
corresponding risk was not significantly increased for principal cause of HF in the younger population
those born moderately preterm. A very low birth (15,33). In individuals born preterm, such heart mus-
weight for gestational age (>2 SDs below the mean) cle disease could be the result of cardiac remodeling
was also associated with an increased risk of HF, but after preterm birth.
this risk increase was not significant after adjustment The current understanding of cardiac develop-
for potential confounding factors. ment is that cardiomyocytes proliferate until late
POTENTIAL MECHANISMS. The mechanisms by gestation and switch to an adult hypertrophic
which preterm birth may influence subsequent risk of growth mode shortly after birth (36–38). Animal
HF in childhood and young adulthood remain elusive. models show that the immature cardiomyocytes of
A review of existing evidence concluded that in- the preterm heart adapt to extrauterine conditions
dividuals born preterm have slightly higher resting through structural remodeling, which may have an
systolic blood pressure in early adulthood, which may impact on future cardiac function (16,17). In a small
increase their risk of developing hypertension (4,5). echocardiographic study of preterm infants, there
Hypertension is, in turn, 1 of the most important were signs of delay in maturation of the myocardium
risk factors for developing HF in adults, also in at 28 days of age (18). The same study found signs of
the absence of ischemic heart disease (30,31). left ventricular diastolic dysfunction and greater
In childhood, even mild, untreated elevation of blood dependence on atrial contraction in preterm infants.
JACC VOL. 69, NO. 21, 2017 Carr et al. 2639
MAY 30, 2017:2634–42 Preterm Birth and Risk of HF

T A B L E 3 Associations Between Birth Weight for Gestational and Incident HF: Crude and Adjusted IRRs (95% CIs) for Incident HF in
Relation to Birth Weight for Gestational Age

Unadjusted IRR Adjusted Model 1: IRR Adjusted Model 2: IRR

N No. of Events Incidence Rate* (95% CI) (95% CI)† (95% CI)‡

All subjects, N ¼ 2,665,542

Birth weight for gestational age
Very small 73,937 44 4.33 3.36 (2.45–4.61) 2.69 (1.94–3.73) 2.66 (1.92–3.70)
Small 366,036 87 1.80 1.40 (1.10–1.78) 1.31 (1.03–1.67) 1.25 (0.98–1.60)
Appropriate 1,794,551 301 1.29 1.00 (reference) 1.00 (reference) 1.00 (reference)
Large 338,333 56 1.27 0.99 (0.74–1.31) 1.00 (0.75–1.33) 1.00 (0.75–1.34)
Very large 92,685 13 1.09 0.84 (0.48–1.47) 0.84 (0.48–1.47) 0.84 (0.48–1.47)
Subjects with malformations excluded, n ¼ 2,613,030
Birth weight for gestational age
Very small 70,378 17 1.75 2.07 (1.26–3.40) 1.61 (0.97–2.68) 1.59 (0.95–2.65)
Small 357,453 48 1.02 1.20 (0.88–1.65) 1.13 (0.82–1.55) 1.06 (0.76–1.47)
Appropriate 1,762,768 194 0.84 1.00 (reference) 1.00 (reference) 1.00 (reference)
Large 332,025 40 0.92 1.09 (0.78–1.54) 1.11 (0.79–1.56) 1.12 (0.79–1.57)
Very large 90,406 6 0.51 0.61 (0.27–1.37) 0.62 (0.28–1.40) 0.62 (0.27–1.40)

*Events per 100,000 person-yrs. †Model 1: adjusted for maternal age and education, subjects’ period of birth, attained age during follow-up, sex, and birth weight for
gestational age. ‡Model 2: in addition to the factors noted in model 1, also adjusted for maternal or paternal HF or ischemic heart disease.
Abbreviations as in Table 1.

In addition to the major circulatory transition that (40). We speculate that such alterations of cardiac
occurs at birth, the preterm heart is often exposed to function in preterm individuals may be a conse-
conditions that increase cardiac workload (e.g., pat- quence of the combination of interrupted normal
ent ductus arteriosus), leading to important left-to- cardiac development and postnatal exposure to
right shunting and bronchopulmonary disease with circulatory challenges, and HF could ultimately be
the risk of pulmonary hypertension (39). A cardiac an expression of this.
imaging study of preterm infants found that patent
ductus arteriosus was associated with significantly STUDY STRENGTHS. Strengths of the current study
increased end-diastolic volumes and increased left include the very large cohort and the registry-based
ventricular mass (19). However, it is unclear to what nondifferential follow-up. Sufficient statistical po-
extent such changes are reversed after ductal wer allowed us to examine the risk of HF even among
closure. extremely preterm infants, although they only
Evidence of modulation of cardiac structure and comprised 0.18% of our cohort. Dividing gestational
function has also been found in older survivors of age into 4 categories also enabled us to show a strong
preterm birth. One study included 102 individuals dose–response relationship between low gestational
born preterm who underwent cardiovascular mag- age at birth and later HF. We were able to control for
netic resonance imaging at 20 to 39 years of age; many confounding factors, including maternal and
they were compared with control subjects born at pregnancy characteristics, heritability of heart dis-
term (20). Ventricular mass in young adulthood ease, and birth weight for gestational age. We had
increased with decreasing gestational age at birth. information on congenital malformations, and
Higher systolic blood pressure in preterm in- including or excluding children with malformations
dividuals could not alone explain this finding, as the yielded essentially the same results among in-
increase in ventricular mass was disproportionate dividuals born extremely preterm and very preterm.
relative to any elevation in blood pressure. There The National Patient Register and Cause of Death
were no observable differences in left ventricular Register were used to ascertain HF. After excluding
ejection fraction between the 2 groups, but both subjects born with congenital malformations, we
stroke volume and end-diastolic volume were lower found an incidence of 0.88 per 100,000 person-years,
in individuals born preterm. Preterm-born in- which is almost identical to that of a British-Irish
dividuals also had reduced diastolic myocardial study (0.87 per 100,000 person-years) (14). The val-
relaxation. Moreover, their right ventricular function idity of the HF diagnosis is high in Swedish registers,
was compromised, and 6% had a right ventricular with a positive predictive value of 82% and even 95%
ejection fraction below clinical reference values for those with a primary diagnosis of HF (41).
2640 Carr et al. JACC VOL. 69, NO. 21, 2017

Preterm Birth and Risk of HF MAY 30, 2017:2634–42

C E N T R A L IL LU ST R A T I O N Risk of HF in Childhood and Young Adult Age in Relation to

Gestational Age at Birth

Carr, H. et al. J Am Coll Cardiol. 2017;69(21):2634–42.

Subjects with malformations excluded, n ¼ 2,613,030. Incidence rate ratios (95% confidence intervals) adjusted for maternal age and education,
subjects’ periods of birth, attained age during follow-up, sex, birth weight for gestational age, and maternal and paternal cardiovascular disease.
HF ¼ heart failure.

STUDY LIMITATIONS. The limitations of our study are effect of prematurity. We attempted to investigate
typical for registry-based research. Children born at this topic further by looking closer at age at diagnosis,
low gestational age are generally subject to closer and we found that the excess incidence of HF in those
medical follow-up, especially during their first years of born very or extremely preterm was higher at 1 to 5
life. Thus, we cannot rule out that surveillance bias has years of age compared with later. This outcome could
influenced our results. To avoid measuring HF as a also be explained by those extravulnerable to disease,
direct complication in the neonatal period, all in- having already developed HF by the time they reach
dividuals diagnosed with HF before 1 year of age were the age of the expected increase in incidence (i.e.,
excluded. We were unable to investigate if and how a depletion of susceptible effect). Another possibility is
hemodynamically significant patent ductus arteriosus that subjects born in earlier birth years, thus
and its treatment relate to risk of later HF because of contributing longer follow-up, are not representative
the low number of HF cases in the lowest gestational of more recent preterm births. In summary, given the
ages and the probable underreporting of patent ductus age profile of our cohort, this study captured effects
arteriosus diagnosis in the national registries, of prematurity in childhood and adolescence but not
compared with other prospective national cohorts effects later in adulthood.
(42). The same limitation applies to the use of ante- Our results show a strong association between
natal corticosteroids, diagnosis of bronchopulmonary gestational age and risk of HF. However, the total
dysplasia, and duration of mechanical ventilation, number of cases of HF in our material is small (501 cases
which may all be factors that are potential mediators of in >2.6 million individuals), with only 11 cases among
the association between preterm birth and later risk of those born extremely preterm. Any minor change in
HF. the number of cases would thus influence effect size.
HF can be difficult to diagnose in a young patient, The great risk increase for HF we have observed in the
and reduced cardiac function may remain silent and most preterm subjects may be decreased but not easily
unreported for a long time (13,43). If this is the case, silenced by such alterations. Also, we have excluded
our findings may be a late reflection of a more direct all cases of HF before 1 year of age (n ¼ 753), which
JACC VOL. 69, NO. 21, 2017 Carr et al. 2641
MAY 30, 2017:2634–42 Preterm Birth and Risk of HF

could have influenced the association between pre- cardiac health in survivors of extremely and very
term birth and HF. Longitudinal follow-up would bring preterm birth.
us closer to the true nature of this relationship, but as
most survivors of extremely preterm birth are still ADDRESS FOR CORRESPONDENCE: Dr. Hanna Carr,
young, this scenario will not be possible for some time Clinical Epidemiology Unit, T2, Karolinska University
yet. Furthermore, the outpatient section of the Na- Hospital, Solna, 171 76 Stockholm, Sweden. E-mail:
tional Patient Register was not established until 2001,
which further limited our possibility of observing
patients over time or investigating underlying causes PERSPECTIVES
of HF in this particular cohort (23).
preterm birth are at higher risk of developing hypertension and
cardiovascular mortality in young adulthood than those born at
This study found a strong association between
term. The relationship between gestational age also applies to the
preterm birth and risk of incident HF in children and
risk of developing HF.
young adults. The increase in risk was inversely
related to gestational age at birth, although absolute
TRANSLATIONAL OUTLOOK: Further research is needed to
risks were low. Considering the rising number of
elucidate the mechanisms compromising cardiac function after
individuals surviving preterm birth and the potential
preterm birth and evaluate interventions to improve long-term
consequences of early onset of reduced cardiac
cardiovascular health in these individuals.
function, the problem may grow with time. There
may be a need for closer follow-up and assessment of


1. Lee AC, Katz J, Blencowe H, et al. National and 9. Massin MM, Astadicko I, Dessy H. Epidemiology 17. De Matteo R, Blasch N, Stokes V, Davis P,
regional estimates of term and preterm babies of heart failure in a tertiary pediatric center. Clin Harding R. Induced preterm birth in sheep: a
born small for gestational age in 138 low-income Cardiol 2008;31:388–91. suitable model for studying the developmental
and middle-income countries in 2010. Lancet effects of moderately preterm birth. Reprod Sci
10. Rossano JW, Kim JJ, Decker JA, et al. Preva-
Glob Health 2013;1:e26–36. 2010;17:724–33.
lence, morbidity, and mortality of heart failure-
2. Blencowe H, Cousens S, Chou D, et al. Born too related hospitalizations in children in the United 18. Hirose A, Khoo NS, Aziz K, et al. Evolution of
soon: the global epidemiology of 15 million pre- States: a population-based study. J Card Fail 2012; left ventricular function in the preterm infant.
term births. Reprod Health 2013;10 Suppl 1:S2. 18:459–70. J Amer Soc Echocardiogr 2014;28:302–8.

3. Liu L, Oza S, Hogan D, et al. Global, regional, 19. Broadhouse KM, Finnemore AE, Price AN, et al.
11. Rossano JW, Shaddy RE. Heart failure in chil-
and national causes of child mortality in 2000-13, Cardiovascular magnetic resonance of cardiac
dren: etiology and treatment. J Pediatr 2014;165:
with projections to inform post-2015 priorities: an function and myocardial mass in preterm infants: a
updated systematic analysis. Lancet 2015;385: preliminary study of the impact of patent ductus
430–40. 12. Kaufman B, Lin K, Patel A, Naim M, Shah M, arteriosus. J Cardiovasc Magn Reson 2014;16:54.
Shaddy R. Cardiac failure. In: Hoffman J, Moller J,
20. Lewandowski AJ, Augustine D, Lamata P, et al.
4. de Jong F, Monuteaux MC, van Elburg RM, editors. Pediatric Cardiovascular Medicine. Oxford,
Preterm heart in adult life: cardiovascular mag-
Gillman MW, Belfort MB. Systematic review and UK: Wiley-Blackwell, 2012:1021–31.
netic resonance reveals distinct differences in left
meta-analysis of preterm birth and later systolic
13. Wong CM, Hawkins NM, Jhund PS, et al. Clin- ventricular mass, geometry, and function. Circu-
blood pressure. Hypertension 2012;59:226–34.
ical characteristics and outcomes of young and lation 2013;127:197–206.
5. Johansson S, Iliadou A, Bergvall N, Tuvemo T, very young adults with heart failure: the CHARM 21. Ludvigsson JF, Otterblad-Olausson P,
Norman M, Cnattingius S. Risk of high blood programme (Candesartan in Heart Failure Assess- Pettersson BU, Ekbom A. The Swedish personal
pressure among young men increases with the ment of Reduction in Mortality and Morbidity). identity number: possibilities and pitfalls in
degree of immaturity at birth. Circulation 2005; J Am Coll Cardiol 2013;62:1845–54. healthcare and medical research. Eur J Epidemiol
14. Andrews RE, Fenton MJ, Ridout DA, Burch M. 2009;24:659–67.
6. Ueda P, Cnattingius S, Stephansson O, New-onset heart failure due to heart muscle dis- 22. National Board of Health and Welfare. Swedish
Ingelsson E, Ludvigsson JF, Bonamy AK. Cere- ease in childhood: a prospective study in the Medical Birth Register. A summary of content
brovascular and ischemic heart disease in young United Kingdom and Ireland. Circulation 2008;117: and quality. 2003. Available at: http://www.
adults born preterm: a population-based Swedish 79–84.
cohort study. Eur J Epidemiol 2014;29:253–60. Accessed February 15, 2017.
15. Barasa A, Schaufelberger M, Lappas G,
7. Crump C, Sundquist K, Sundquist J, Swedberg K, Dellborg M, Rosengren A. Heart 23. Ludvigsson JF, Andersson E, Ekbom A, et al.
Winkleby MA. Gestational age at birth and mor- failure in young adults: 20-year trends in hospi- External review and validation of the Swedish
tality in young adulthood. JAMA 2011;306: talization, aetiology, and case fatality in Sweden. National Inpatient Register. BMC Public Health
1233–40. Eur Heart J 2014;35:25–32. 2011;11:450.

8. Koupil I, Leon DA, Lithell HO. Length of 16. Bensley JG, Stacy VK, De Matteo R, Harding R, 24. National Board of Health and Welfare. Method
gestation is associated with mortality from cere- Black MJ. Cardiac remodelling as a result of pre- report: cause of death statistics (in Swedish). Avail-
brovascular disease. J Epidemiol Community term birth: implications for future cardiovascular able at:
Health 2005;59:473–4. disease. Eur Heart J 2010;31:2058–66. 2010/2010-4-33. Accessed February 15, 2017.
2642 Carr et al. JACC VOL. 69, NO. 21, 2017

Preterm Birth and Risk of HF MAY 30, 2017:2634–42

25. Ekbom A. The Swedish Multi-Generation Blood Pressure in Children and Adolescents. Butler AS, editors. Preterm birth causes, conse-
Register. Methods Mol Biol 2011;675:215–20. Pediatrics 2004;114:555–76. quences, and prevention. Institute of Medicine
(US) Committee on Understanding Premature
26. Statistics Sweden. Evaluation of the Swedish 32. Haydock PM, Cowie MR. Heart failure: classi-
Birth and Assuring Healthy Outcomes. Washing-
Register of Education. Available at: fication and pathophysiology. Medicine 2010;38:
ton, DC: National Academies Press, 2007.
se/en_/Finding-statistics/Publishing-calendar/Show- 467–72.
detailed-information/?publobjid¼3299. Accessed 40. Lewandowski AJ, Bradlow WM, Augustine D,
33. Hsu DT, Pearson GD. Heart failure in children:
February 15, 2017. et al. Right ventricular systolic dysfunction in
part I: history, etiology, and pathophysiology. Circ
young adults born preterm. Circulation 2013;128:
27. Statistics Sweden. Tables on the population in Heart Fail 2009;2:63–70.
Sweden 2009. Available at:
34. Zoller B, Sundquist J, Sundquist K, Crump C.
statistik/_publikationer/be0101_2009a01_br_be0110 41. Ingelsson E, Arnlov J, Sundstrom J, Lind L. The
Perinatal risk factors for premature ischaemic
tab.pdf. Accessed February 15, 2017. validity of a diagnosis of heart failure in a hospital
heart disease in a Swedish national cohort. BMJ
discharge register. Eur J Heart Fail 2005;7:787–91.
28. Swedish Agency for Health Technology Open 2015;5:e007308.
Assessment and Assessment of Social Services. 42. Fellman V, Hellstrom-Westas L, Norman M,
35. Kaijser M, Bonamy AK, Akre O, et al. Perinatal et al. One-year survival of extremely preterm
Routine ultrasound examination during pregnancy.
risk factors for ischemic heart disease: disen- infants after active perinatal care in Sweden. JAMA
Available at:
tangling the roles of birth weight and preterm 2009;301:2225–33.
birth. Circulation 2008;117:405–10.
during-pregnancy/. Accessed February 15, 2017. 43. Kantor PF, Lougheed J, Dancea A, et al.
36. Rudolph AM. Myocardial growth before and Presentation, diagnosis, and medical manage-
29. Marsal K, Persson PH, Larsen T, Lilja H,
after birth: clinical implications. Acta Paediatr ment of heart failure in children: Canadian Car-
Selbing A, Sultan B. Intrauterine growth curves
2000;89:129–33. diovascular Society guidelines. Can J Cardiol
based on ultrasonically estimated foetal weights.
Acta Paediatr 1996;85:843–8. 37. Walsh S, Ponten A, Fleischmann BK, Jovinge S. 2013;29:1535–52.

30. Lloyd-Jones DM, Larson MG, Leip EP, et al. Cardiomyocyte cell cycle control and growth esti-
Lifetime risk for developing congestive heart fail- mation in vivo—an analysis based on cardiomyocyte
nuclei. Cardiovasc Res 2010;86:365–73. KEY WORDS cardiovascular disease,
ure: the Framingham Heart Study. Circulation
epidemiology, neonatology, pediatrics, risk
2002;106:3068–72. 38. Bergmann O, Zdunek S, Felker A, et al. factor
Dynamics of cell generation and turnover in the
31. National High Blood Pressure Education
human heart. Cell 2015;161:1566–75.
Program Working Group on High Blood Pressure in
Children and Adolescents. The Fourth Report on 39. Behrman RE, Butler AS. Mortality and acute A PPE NDI X For supplemental tables, please
the Diagnosis, Evaluation, and Treatment of High complications in preterm infants. In: Behrman RE, see the online version of this article.