RESEARCH ARTICLE

Maternal dyslipidemia and risk for preterm

birth
Caitlin J. Smith1, Rebecca J. Baer2,3, Scott P. Oltman3,4, Patrick J. Breheny ID5, Wei Bao1,
Jennifer G. Robinson1, John M. Dagle6, Liang Liang7, Sky K. Feuer8, Christina
D. Chambers2, Laura L. Jelliffe-Pawlowski3,4, Kelli K. Ryckman ID1,6*
1 Department of Epidemiology, University of Iowa, Iowa City, Iowa, United States of America, 2 Department
of Pediatrics, University of California San Diego, La Jolla, California, United States of America, 3 California
Preterm Birth Initiative, University of California San Francisco, San Francisco, California, United States of
America, 4 Department of Epidemiology and Biostatistics, University of California San Francisco, San
a1111111111 Francisco, California, United States of America, 5 Department of Biostatistics, University of Iowa, Iowa City,
a1111111111 Iowa, United States of America, 6 Department of Pediatrics, University of Iowa, Iowa City, Iowa, United
a1111111111 States of America, 7 Department of Genetics, Stanford University, Stanford, California, United States of
a1111111111 America, 8 Department of Obstetrics, Gynecology & Reproductive Sciences, University of California San
Francisco, San Francisco, California, United States of America
a1111111111
* [email protected]

OPEN ACCESS Abstract

Citation: Smith CJ, Baer RJ, Oltman SP, Breheny Maternal lipid profiles during pregnancy are associated with risk for preterm birth. This study
PJ, Bao W, Robinson JG, et al. (2018) Maternal
dyslipidemia and risk for preterm birth. PLoS ONE
investigates the association between maternal dyslipidemia and subsequent preterm birth
13(12): e0209579. https://doi.org/10.1371/journal. among pregnant women in the state of California. Births were identified from California birth
pone.0209579 certificate and hospital discharge records from 2007–2012 (N = 2,865,987). Preterm birth
Editor: Zhong-Cheng Luo, Mount Sinai Health was defined as <37 weeks completed gestation and dyslipidemia was defined by diagnostic
System, University of Toronto, CANADA codes. Subtypes of preterm birth were classified as preterm premature rupture of mem-
Received: June 21, 2018 branes (PPROM), spontaneous labor, and medically indicated, according to birth certificate
Accepted: December 7, 2018
data and diagnostic codes. The association between dyslipidemia and preterm birth was
tested with logistic regression. Models were adjusted for maternal age at delivery, race/eth-
Published: December 21, 2018
nicity, hypertension, pre-pregnancy body mass index, insurance type, and education.
Copyright: © 2018 Smith et al. This is an open Maternal dyslipidemia was significantly associated with increased odds of preterm birth
access article distributed under the terms of the
Creative Commons Attribution License, which
(adjusted OR: 1.49, 95%CI: 1.39, 1.59). This finding was consistent across all subtypes of
permits unrestricted use, distribution, and preterm birth, including PPROM (adjusted OR: 1.54, 95%CI: 1.34, 1.76), spontaneous
reproduction in any medium, provided the original (adjusted OR: 1.51, 95%CI: 1.39, 1.65), and medically indicated (adjusted OR: 1.454, 95%
author and source are credited.
CI: 1.282, 1.649). This study suggests that maternal dyslipidemia is associated with
Data Availability Statement: The data used in this increased risk for all types of preterm birth.
analysis is owned by the State of California who
grants access through an application and approval
process. This process is open to any interested
researcher or other investigator who seeks access.
No special permission was granted for this project. Introduction
Interested researchers may apply for access to the
data at: https://www.cdph.ca.gov/Programs/CFH/ Preterm birth is defined as delivery prior to 37 weeks of completed gestation. The World
DGDS/Pages/cbp/default.aspx. Health Organization estimates that preterm birth affects 11% of pregnancies worldwide, repre-
Funding: This work was supported by the
senting nearly 15 million births in 2010 [1]. It is the second leading cause of death in children
California Preterm Birth Initiative within the under age 5 [1]. Despite decades of research into the causes of preterm birth, the biological
University of California, San Francisco causes of preterm birth remain largely unknown [2].

PLOS ONE | https://doi.org/10.1371/journal.pone.0209579 December 21, 2018 1 / 10

 Dyslipidemia and preterm birth

(UCSF7027075). Members of the California Normal pregnancy is accompanied by metabolic changes, particularly in carbohydrate and
Preterm Birth Initiative participated as authors on lipid metabolism. The benefit of these changes is presumably to increase circulating glucose and
this manuscript and are specifically noted. The
triglycerides to nourish the growing fetus. Changes in carbohydrate metabolism are bimodal, in
funder had no role in study design, data collection
and analysis, decision to publish, or preparation of which fasting plasma glucose is decreased in early pregnancy, and impaired glucose tolerance
the manuscript. occurs in late pregnancy [3]. Circulating lipids, including high density lipoprotein (HDL), low
density lipoprotein (LDL), total cholesterol, and triglycerides, increase throughout pregnancy,
Competing interests: The authors have declared
that no competing interests exist. with the greatest increase observed for triglycerides [3]. Although much research has been
devoted to glucose metabolism during pregnancy due to the risk of gestational diabetes mellitus
[4], increasing interest in lipid levels during pregnancy has revealed associations between mater-
nal lipid levels and adverse pregnancy outcomes, including preterm birth.
Many studies have investigated associations between maternal lipid levels during pregnancy
and risk for preterm birth, although the lipid components and magnitude of associations have
been inconsistent across studies [5–17]. One previous study investigated the association
between dyslipidemia, as defined by lipid levels in prenatal screening, and found increased
risks for preterm birth with mid-trimester hyperlipidemia in combination with elevated levels
of tumor necrosis alpha [8]. The present study investigates the association between a clinical
diagnosis of maternal dyslipidemia and subsequent preterm birth among pregnant women in
the state of California.

Materials and methods

Study population
Births were identified from California birth certificate and hospital discharge records from
2007–2012 (N = 2,962,434) as collected by the California Office of Statewide Health Planning
and Development. Records were linked approximately 9–12 months prior to delivery through
9–12 months post-delivery [18]. Inclusion criteria included singleton pregnancy, availability of
linked records, gestational age between 20–44 weeks and absence of severe hypertensive dis-
eases including hypertensive heart disease, hypertensive chronic kidney disease and secondary
hypertension. We excluded these women, as those forms of hypertension are not the primary
focus of our study and could confound the association with preterm birth. Dyslipidemia was
defined by the International Classification of Diseases and Related Health Problems (ICD-9)
codes 272.0–272.4. Specifically, these codes include pure hypercholesterolemia (ICD-9 272.0),
pure hyperglyceridemia (ICD-9 272.1), mixed hyperlipidemia (ICD-9 272.2), hyperchylomi-
cronemia (ICD-9 272.3), and other unspecified hyperlipidemia (ICD-9 272.4). We restricted
our dyslipidemia definitions to only those that occurred on a hospital admission at or prior to
the delivery date. Methods and protocols were approved by the Committee for the Protection
of Human Subjects within the Health and Human Services Agency of the State of California.
All data was de-identified and determined not to qualify as human subjects research by the
University of Iowa Institutional Review Board.

Core outcomes and statistical analysis

Preterm birth was defined as gestational age at delivery <37 weeks and term birth was defined
as gestational age at delivery �37 weeks, according to best obstetric estimate. Births were fur-
ther categorized into early preterm birth (<32 weeks), late preterm birth (32–36 6 =7 weeks)
and term birth (�37 weeks). Subtypes of preterm birth were classified as preterm premature
rupture of membranes (PPROM), spontaneous, and medically indicated, according to birth
certificate data or hospital discharge records as previously described [19]. Specifically, preterm
births with indication of premature rupture of membranes were classified as PPROM and

PLOS ONE | https://doi.org/10.1371/journal.pone.0209579 December 21, 2018 2 / 10

 Dyslipidemia and preterm birth

births with indication of preterm labor or tocolytic medication AND absence of PPROM were
classified as spontaneous. Births with absence of premature rupture of membranes, premature
labor and tocolytic medication AND a code for ‘medical induction’ or ‘artificial rupture of
membranes’ or cesarean delivery without such codes were classified as medically indicated.
All analyses were performed using Statistical Analysis Software (SAS) version 9.4 (SAS
Institute, Cary, North Carolina). The associations between dyslipidemia and preterm birth
were tested using logistic regression (PROC LOGISTIC). Dyslipidemia was modeled as a com-
posite variable and as individual diagnostic codes. The associations were tested without adjust-
ment and with adjustment for maternal age at delivery, hypertension (which included pre-
existing essential hypertension, gestational hypertension, pre-eclampsia or eclampsia), race/
ethnicity, BMI, insurance type, and education. Maternal age at delivery was analyzed as a linear
variable. Hypertension was coded as a binary variable and the absence of hypertension was
used as the referent group. Race/ethnicity was categorized as Black, Asian, Caucasian or His-
panic, and Caucasian was used as the referent group. BMI was categorized according to stan-
dard cut-points (underweight [<18.5], normal [18.5–24.9], overweight [25–29.9], or obese
[�30]), and ‘normal’ was used as the referent group [20]. Insurance type was categorized as
Medi-Cal, private, self-pay, or other, and ‘private’ was used as the referent group. Medi-Cal is
California’s Medicaid program, which provides health insurance and health care services for
low-income individuals. Education was categorized as <12 years, exactly 12 years (completion
of high school diploma), or >12 years, which was used as the referent group.
We considered maternal age at delivery as a potential confounder, wherein we hypothesized
that advanced maternal age would be associated with increased likelihood of diagnosis of dysli-
pidemia and an increased likelihood of delivering preterm [21–24]. We also considered BMI
as a potential confounder, in which overweight and obesity would be associated with increased
likelihood of diagnosis of dyslipidemia and increased likelihood of delivering preterm [23–26].
Other potential confounders including race/ethnicity, maternal age at delivery, hypertension
(includes both pre-pregnancy and pregnancy diagnoses), pre-pregnancy body mass index
(BMI), insurance type, and education [21–26]. These variables were available from birth certif-
icate records. Hypertension diagnoses were also confirmed by hospital discharge records.
Several supplemental analyses were performed. These included: 1) stratification of analyses
by BMI category to determine if BMI modifies the relationship between dyslipidemia and pre-
term birth; 2) consolidation of ICD-9 codes into cholesterol dyslipidemia and triglyceride dys-
lipidemia to determine if the type of dyslipidemia (cholesterol versus triglyceride) affects the
results; and 3) examination of the individual impact of confounders, including race/ethnicity,
hypertension, BMI, insurance type, maternal education and maternal age, on the association
between dyslipidemia and preterm birth. All statistical analyses were performed with statistical
power >99.9% to detect an odds ratio of 1.5.

Results
Demographic characteristics of the study population are presented in Table 1. The analysis
included 9,162 women with dyslipidemia and 2,953,272 women without dyslipidemia.
Women with dyslipidemia differed from women without dyslipidemia by race/ethnicity, BMI,
hypertension (pre-existing or onset during pregnancy), insurance status, education, and
maternal age at delivery (all at p<0.0001). Specifically, the group of women who had dyslipide-
mia included more Black women, were less likely to have a normal BMI, were more likely to
have hypertension, were less likely to be on Medi-Cal insurance, were more likely to have com-
pleted more than 12 years of education, and were slightly older than the group of women with-
out dyslipidemia.

PLOS ONE | https://doi.org/10.1371/journal.pone.0209579 December 21, 2018 3 / 10

 Dyslipidemia and preterm birth

Table 1. Demographic characteristics of study population.

Dyslipidemia (N = 9,162) No Dyslipidemia (N = 2,953,272)
Maternal Age at Delivery� 32.4 ± 5.97 28.3 ± 6.29
Race
Black 646 (7.8%) 157,917 (5.8%)
Asian 1,305 (15.7%) 365,274 (13.4%)
Caucasian 2,293 (27.5%) 770,805 (28.2%)
Hispanic 4,086 (49.0%) 1,963,803 (52.7%)
Missing (N = 218,664)
BMI
Underweight 190 (2.2%) 144,146 (5.2%)
Normal 2,278 (26.7%) 1,349,503 (49.0%)
Overweight 2,308 (27.1%) 701,674 (25.5%)
Obese 3,754 (44.0%) 558,825 (20.3%)
Missing (N = 199,124)
Insurance
MediCal 2,473 (27.0%) 1,427,199 (48.4%)
Private 6413 (70.1%) 1,366,516 (46.4%)
Self-Pay 58 (0.6%) 59,778 (2.0%)
Other 210 (2.3%) 95,014 (3.2%)
Missing (N = 4,765)
Education
<12 years 1,350 (15.3%) 707,119 (24.9%)
12 completed years 2,208 (25.0%) 755,196 (26.6%)
>12 years 5,292 (59.8%) 1,381,901 (48.6%)
Missing (N = 109,056)
Hypertension
Yes 2,447 (26.7%) 209,004 (7.1%)
Preterm
Yes 1,369 (14.9%) 209,717 (7.1%)
�
Data are presented as mean ± standard deviation. All other data are presented as N (%).

https://doi.org/10.1371/journal.pone.0209579.t001

Results of the traditional logistic regression analyses are presented in Table 2. Three differ-
ent outcomes are presented: preterm versus term, early and late preterm versus term, and
PPROM, spontaneous, and medically indicated versus term. Dyslipidemia was significantly
associated with preterm birth, both before and after adjusting for race/ethnicity, maternal age
at delivery, hypertension, BMI, insurance type, and education.
Results of the traditional logistic regression analyses, stratified by type of dyslipidemia, are
presented in Table 3. Hyperchylomicronemia (ICD-9 272.3) was not analyzed due to low sam-
ple size (N<10). Each type of dyslipidemia was significantly associated with preterm birth,
both before and after adjusting for race/ethnicity, maternal age at delivery, hypertension, BMI,
insurance type, and education.
Results of the traditional logistic regression analyses, stratified by BMI category, are pre-
sented in S1 Table. Within each BMI category, dyslipidemia was significantly associated with
preterm birth. After adjusting for maternal age at delivery, hypertension, race/ethnicity, insur-
ance type, and education, dyslipidemia was significantly associated with preterm birth among
normal weight, overweight, and obese women, but not among underweight women. Obesity

PLOS ONE | https://doi.org/10.1371/journal.pone.0209579 December 21, 2018 4 / 10

 Dyslipidemia and preterm birth

Table 2. Association between dyslipidemia and preterm birth.

Total population (N = 2,962,434)
Unadjusted OR (95% CI) Adjusted�
OR (95% CI)
Outcome 1a 2.30 (2.17, 2.44) 1.49 (1.39, 1.59)
Outcome 2b
<32 weeks vs. Term 2.97 (2.61, 3.37) 1.63 (1.41, 1.89)
32–36 weeks vs. Term 2.19 (2.06, 2.33) 1.46 (1.36, 1.57)
Outcome 3
PPROM vs. term 1.92 (1.69, 2.17) 1.54 (1.34, 1.76)
Spon. vs. term 2.43 (2.26, 2.62) 1.51 (1.39, 1.65)
Indicated vs. term 2.85 (2.55, 3.17) 1.45 (1.28, 1.65)
�
Adjusted for race, maternal age at delivery, hypertension, body mass index, insurance type, and education
a
Preterm birth (<37 weeks) vs. term birth (�37 weeks)
b
Early and late preterm birth

https://doi.org/10.1371/journal.pone.0209579.t002

itself was associated with a 1.6-fold increase in risk for medically indicated preterm birth com-
pared to normal BMI (OR: 1.61; 95%CI: 1.57, 1.65).
Results of the consolidation of ICD-9 codes into cholesterol dyslipidemia and triglyceride dys-
lipidemia are presented in S2 Table. Cholesterol dyslipidemia, which included pure hypercholes-
terolemia and mixed dyslipidemia, was significantly associated with preterm birth before and
after adjustment. Triglyceride dyslipidemia, which included pure hyperglyceridemia and hyperch-
ylomicronemia, was significantly associated with preterm birth before and after adjustment.
To investigate the individual impact of confounders, including race/ethnicity, hypertension,
BMI, insurance type, maternal education and maternal age, on the association between dyslipi-
demia and preterm birth, each confounder was individually added to the logistic regression
models (S3 Table). Adjusting for hypertension alone showed the greatest attenuation of the
association between dyslipidemia and preterm birth of all the individual confounders (OR:
1.53; 95%CI: 1.45, 1.63). Adjusting for other confounders did not affect the odds ratios com-
pared to the unadjusted models.

Discussion
In this retrospective cohort of 2.9 million pregnant women in California, maternal diagnosis
of dyslipidemia was significantly associated with increased risk for preterm birth. To the best
of our knowledge, this study represents the largest investigation of the association between
clinical dyslipidemia and risk for preterm birth done to date and it is the only study that we
know of to utilize hospital diagnostic codes to define dyslipidemia, which include both familial
and non-familial forms of dyslipidemia. The size and diversity of the study population allowed
for the investigation of the association between dyslipidemia and preterm birth, stratified by
subtypes of dyslipidemia.
Several previous studies have investigated associations between maternal lipid levels during
pregnancy and risk for preterm birth [5–17]. These studies varied in the lipid components
they measured, the gestational age at which they were measured, and fasting status, which may
explain their discordant findings. For example, of the seven studies that measured all four lipid
components [5–11], four studies failed to identify an association between individual lipid com-
ponents and risk for preterm birth. Of the three studies that measured only total cholesterol
(TC) [15–17] one identified a positive association between elevated TC and preterm birth and

PLOS ONE | https://doi.org/10.1371/journal.pone.0209579 December 21, 2018 5 / 10

 Dyslipidemia and preterm birth

Table 3. Association between types of dyslipidemia and preterm birth.

Pure Pure Mixed hyperlipidemia Other unspecified Maternal lipid disorder
hypercholesterolemia hyperglyceridemia (N = 379) hyperlipidemia (N = 6,088) before delivery (N = 6,816)
(N = 2,599) (N = 6,81)
UnaOR AOR� UnaOR AOR� UnaOR (95% AOR� UnaOR (95% AOR� UnaOR (95% AOR�
(95% CI) (95% CI) (95% CI) (95% CI) CI) (95% CI) CI) (95% CI) CI) (95% CI)
Outcome 1a 2.16 1.30 2.54 1.64 2.41 1.77 2.39 1.53 2.32 1.63
(1.93, (1.14, (2.07, (1.29, (1.82, 3.18) (1.29, (2.23, 2.57) (1.41, (2.17, 2.48) (1.50,
2.41) 1.47) 3.12) 2.09) 2.43) 1.66) 1.76)
Outcome 2b
<32 weeks vs. Term 2.92 1.40 3.43 2.07 2.94 2.03 3.07 1.67 2.89 1.79
(2.30, (1.07, (2.22, (1.26, (1.57, 5.50) (1.03, (2.63, 3.58) (1.40, (2.49, 3.36) (1.51,
3.70) 1.83) 5.31) 3.39) 3.99) 2.00) 2.13)
32–36 weeks vs. 2.04 1.28 2.39 1.56 2.32 1.72 2.28 1.50 2.23 1.59
Term (1.81, (1.11, (1.92, (1.20, (1.72, 3.14) (1.22, (2.12, 2.46) (1.38, (2.07, 2.39) (1.47,
2.30) 1.46) 2.99) 2.03) 2.42) 1.64) 1.73)
Outcome 3
PPROM vs. term 1.64 1.41 2.05 1.86 1.81 1.43 2.06 1.57 1.99 1.61
(1.28, (1.08, (1.31, (1.16, (0.97, 3.40) (0.71, (1.77, 2.39) (1.33, (1.73, 2.29) (1.38,
2.11) 1.83) 3.21) 2.99) 2.88) 1.86) 1.89)
Spon. vs. term 2.43 1.38 3.23 1.76 2.59 1.99 2.42 1.50 2.45 1.70
(2.11, (1.17, (2.53, (1.30, (1.81, 3.71) (1.33, (2.21, 2.65) (1.34, (2.25, 2.68) (1.54,
2.79) 1.61) 4.13) 2.40) 2.97) 1.67) 1.88)
Indicated vs. term 2.46 1.13 2.50 1.56 3.05 1.62 3.23 1.65 2.84 1.53
(1.98, (0.89, (1.63, (0.99, (1.82, 5.10) (0.88, (2.85, 3.66) (1.43, (2.50, 3.22) (1.32,
3.05) 1.44) 3.83) 2.46) 2.98) 1.91) 1.77)
�
AOR: Adjusted odds ratio including race, maternal age at delivery, hypertension, body mass index, insurance type, and education
a
Preterm birth (<37 weeks) vs. term birth (�37 weeks)
b
Early and late preterm birth

https://doi.org/10.1371/journal.pone.0209579.t003

two identified associations between both low and high TC and preterm birth. A recent meta-
analysis identified significant pooled associations between elevated TC, elevated TG and low
HDL and preterm birth [27]. All previous studies have used lipid levels as a continuous expo-
sure, although some categorized lipid levels by percentiles. However, this does not mean that
these studies sampled women who would have met criteria for dyslipidemia. Thus, our study
is unique in its use of a clinically significant exposure.
Of particular interest in the present study is the consistency of the magnitude of association
across all subtypes of preterm birth, after adjusting for potential confounders. These adjusted
odds ratios ranged from 1.4–1.6 (Table 2), providing strong and consistent evidence that women
with maternal dyslipidemia are approximately one-and-a-half times more likely to deliver preterm
than comparable women without dyslipidemia regardless of preterm birth subtype.
Dyslipidemia is often comorbid with obesity [26], and obesity has long been known to
increase the risk for pregnancy complications such as gestational diabetes mellitus and pre-
eclampsia [28]. Dyslipidemia severe enough to warrant a clinical diagnosis may be a marker
for more severely disturbed cardiometabolic milieu. Chronic dyslipidemia is accompanied by
inflammation, a hallmark of obesity, and acute inflammation triggers altered lipid metabolism
[29]. Interestingly, stratification by BMI category did not significantly alter the associations
between dyslipidemia and preterm birth among overweight or obese women (S1 Table). How-
ever, hypertension had the most affect in attenuating the association. Hypertension is known
to co-occur with dyslipidemia and increases the risk for preterm birth. This suggests that

PLOS ONE | https://doi.org/10.1371/journal.pone.0209579 December 21, 2018 6 / 10

 Dyslipidemia and preterm birth

dyslipidemia is associated with increased risk for preterm birth independent of obesity and
some of that risk is explained by the co-occurrence of hypertension.
A limitation of this study is the lack of information regarding dyslipidemia diagnostic prac-
tices. Heterogeneity exists among practitioners in terms of the degree of follow-up testing of
lipid levels. Thus, some women may have received a diagnosis after a single abnormal lipid
panel, with no repeat testing, while other women may have received a diagnosis following mul-
tiple abnormal panels. Some women with dyslipidemia may not have a diagnosis because they
have never had their lipid levels tested or they were treated before pregnancy and entered preg-
nancy with normal lipids. This type of non-differential misclassification would bias the results
toward the null. There are currently, no-evidence based standards for how to treat women
with dyslipidemia during pregnancy. The most common recommendations include lifestyle
changes, glycemic control and close follow-up of the pregnancy [23]. It is unlikely that women
were treated with cholesterol-lowering drugs such as statins or niacin, since these drugs are
contraindicated during pregnancy [30].
It should also be noted that an important limitation of the study is the lack of lipid level
information. Such data would have allowed for discrimination between familial, monogenic dys-
lipidemias, which are characterized by markedly abnormal lipid levels, and non-familial, poly-
genic dyslipidemias, which typically manifest as less drastic changes in lipid levels. However, a
Norwegian study of 895 women with familial hypercholesterolemia (FH) found no association
between FH and risk for adverse pregnancy outcomes, including preterm birth [31]. Thus, we
can infer that the association between pure hypercholesterolemia and preterm birth is driven by
the non-familial form, which may be exacerbated by the co-occurrence of hypertension. Addi-
tionally, assuming a prevalence of 1 in 250 for heterozygous FH [32], only ten women with pure
hypercholesterolemia would be expected to have FH in our study, which would likely not influ-
ence the results. Further, the detection of small differences in lipids between women who deliver
term and preterm is unlikely to be clinically meaningful. In contrast, dyslipidemia is a clinically-
validated medical condition that could be readily identified as a risk factor for preterm birth. We
were also limited to the accuracy of the data on both the hospital discharge record and the birth
certificate record, which could have introduced some bias in the estimates of our confounders.
There were large amounts of missing data for maternal age, obesity and education; however,
when these variables were considered individually, there was little difference between the unad-
justed and adjusted models. Therefore, missing data is unlikely to affect our conclusions.
In conclusion, dyslipidemia, as both an aggregate exposure and individual subtypes, was
significantly associated with a 1.5-fold increased risk for preterm birth after adjusting for
potential confounders. These findings suggest that dyslipidemia may be a potential factor in
the etiology of preterm birth, and may serve as a marker of increased risk for preterm birth.
The identification of dyslipidemia as a risk factor for preterm birth is impactful for several rea-
sons: 1) There are few known causal risk factors for preterm birth, as the causes of parturition
and preterm birth remain largely unknown, 2) dyslipidemia may be modified by lifestyle
changes and medication [30] and 3) severe dyslipidemia receiving a clinical diagnosis may be
easy to incorporate into clinical decision-making in the era of electronic medical records.
Findings from this study support lipid screening among women of reproductive age and addi-
tional studies are needed to determine if diagnosing and treating dyslipidemia early in preg-
nancy reduces the risk for preterm birth.

Supporting information
S1 Table. Association between dyslipidemia and preterm birth, stratified by BMI category.
(DOCX)

PLOS ONE | https://doi.org/10.1371/journal.pone.0209579 December 21, 2018 7 / 10

 Dyslipidemia and preterm birth

S2 Table. Analysis of consolidation of ICD-9 dyslipidemia codes.

(DOCX)
S3 Table. Analysis of the individual impact of confounders.
(DOCX)

Acknowledgments
Supported by the California Preterm Birth Initiative within the University of California, San
Francisco.

Author Contributions
Conceptualization: Caitlin J. Smith, Patrick J. Breheny, Wei Bao, Jennifer G. Robinson, John
M. Dagle, Liang Liang, Christina D. Chambers, Laura L. Jelliffe-Pawlowski, Kelli K.
Ryckman.
Data curation: Rebecca J. Baer, Scott P. Oltman.
Formal analysis: Caitlin J. Smith, Rebecca J. Baer.
Funding acquisition: Sky K. Feuer, Laura L. Jelliffe-Pawlowski.
Project administration: Sky K. Feuer, Laura L. Jelliffe-Pawlowski, Kelli K. Ryckman.
Supervision: Kelli K. Ryckman.
Writing – original draft: Caitlin J. Smith.
Writing – review & editing: Caitlin J. Smith, Rebecca J. Baer, Scott P. Oltman, Patrick J. Bre-
heny, Wei Bao, Jennifer G. Robinson, John M. Dagle, Liang Liang, Sky K. Feuer, Christina
D. Chambers, Laura L. Jelliffe-Pawlowski, Kelli K. Ryckman.

References
1. Blencowe H, Cousens S, Chou D, Oestergaard M, Say L, Moller AB, et al. Born too soon: the global epi-
demiology of 15 million preterm births. Reproductive health. 2013; 10 Suppl 1:S2.
2. Ferrero DM, Larson J, Jacobsson B, Di Renzo GC, Norman JE, Martin JN Jr., et al. Cross-Country Indi-
vidual Participant Analysis of 4.1 Million Singleton Births in 5 Countries with Very High Human Develop-
ment Index Confirms Known Associations but Provides No Biologic Explanation for 2/3 of All Preterm
Births. PloS one. 2016; 11(9):e0162506. https://doi.org/10.1371/journal.pone.0162506 PMID:
27622562
3. Hadden DR, McLaughlin C. Normal and abnormal maternal metabolism during pregnancy. Seminars in
fetal & neonatal medicine. 2009; 14(2):66–71.
4. Hartling L, Dryden DM, Guthrie A, Muise M, Vandermeer B, Donovan L. Diagnostic thresholds for gesta-
tional diabetes and their impact on pregnancy outcomes: a systematic review. Diabetic medicine: a jour-
nal of the British Diabetic Association. 2014; 31(3):319–31.
5. Alleman BW, Smith AR, Byers HM, Bedell B, Ryckman KK, Murray JC, et al. A proposed method to pre-
dict preterm birth using clinical data, standard maternal serum screening, and cholesterol. American
journal of obstetrics and gynecology. 2013; 208(6):472.e1–11.
6. Chatzi L, Plana E, Daraki V, Karakosta P, Alegkakis D, Tsatsanis C, et al. Metabolic syndrome in early
pregnancy and risk of preterm birth. American journal of epidemiology. 2009; 170(7):829–36. https://
doi.org/10.1093/aje/kwp211 PMID: 19713286
7. Emet T, Ustuner I, Guven SG, Balik G, Ural UM, Tekin YB, et al. Plasma lipids and lipoproteins during
pregnancy and related pregnancy outcomes. Archives of gynecology and obstetrics. 2013; 288(1):49–
55. https://doi.org/10.1007/s00404-013-2750-y PMID: 23400357
8. Jelliffe-Pawlowski LL, Ryckman KK, Bedell B, O’Brodovich HM, Gould JB, Lyell DJ, et al. Combined ele-
vated midpregnancy tumor necrosis factor alpha and hyperlipidemia in pregnancies resulting in early
preterm birth. American journal of obstetrics and gynecology. 2014; 211(2):141.e1–9.

PLOS ONE | https://doi.org/10.1371/journal.pone.0209579 December 21, 2018 8 / 10

 Dyslipidemia and preterm birth

9. Jin WY, Lin SL, Hou RL, Chen XY, Han T, Jin Y, et al. Associations between maternal lipid profile and
pregnancy complications and perinatal outcomes: a population-based study from China. BMC preg-
nancy and childbirth. 2016; 16:60. https://doi.org/10.1186/s12884-016-0852-9 PMID: 27000102
10. Mudd LM, Holzman CB, Catov JM, Senagore PK, Evans RW. Maternal lipids at mid-pregnancy and the
risk of preterm delivery. Acta obstetricia et gynecologica Scandinavica. 2012; 91(6):726–35. https://doi.
org/10.1111/j.1600-0412.2012.01391.x PMID: 22404756
11. Niromanesh S, Shirazi M, Dastgerdy E, Sharbaf FR, Shirazi M, Khazaeipour Z. Association of hypertri-
glyceridaemia with pre-eclampsia, preterm birth, gestational diabetes and uterine artery pulsatility
index. The National medical journal of India. 2012; 25(5):265–7. PMID: 23448624
12. Kramer MS, Kahn SR, Rozen R, Evans R, Platt RW, Chen MF, et al. Vasculopathic and thrombophilic
risk factors for spontaneous preterm birth. International journal of epidemiology. 2009; 38(3):715–23.
https://doi.org/10.1093/ije/dyp167 PMID: 19336437
13. Vrijkotte TG, Krukziener N, Hutten BA, Vollebregt KC, van Eijsden M, Twickler MB. Maternal lipid profile
during early pregnancy and pregnancy complications and outcomes: the ABCD study. The Journal of
clinical endocrinology and metabolism. 2012; 97(11):3917–25. https://doi.org/10.1210/jc.2012-1295
PMID: 22933545
14. Lei Q, Niu J, Lv L, Duan D, Wen J, Lin X, et al. Clustering of metabolic risk factors and adverse preg-
nancy outcomes: a prospective cohort study. Diabetes/metabolism research and reviews. 2016; 32
(8):835–42. https://doi.org/10.1002/dmrr.2803 PMID: 27037671
15. Edison RJ, Berg K, Remaley A, Kelley R, Rotimi C, Stevenson RE, et al. Adverse birth outcome among
mothers with low serum cholesterol. Pediatrics. 2007; 120(4):723–33. https://doi.org/10.1542/peds.
2006-1939 PMID: 17908758
16. Maymunah AO, Kehinde O, Abidoye G, Oluwatosin A. Hypercholesterolaemia in pregnancy as a pre-
dictor of adverse pregnancy outcome. African health sciences. 2014; 14(4):967–73. https://doi.org/10.
4314/ahs.v14i4.28 PMID: 25834509
17. Oluwole AA, Adegbesan-Omilabu MA, Okunade KS. Preterm delivery and low maternal serum choles-
terol level: Any correlation? Nigerian medical journal: journal of the Nigeria Medical Association. 2014;
55(5):406–10.
18. International Classification of Diseases Ninth Revision Clinical Modification: NIH National Cancer Insti-
tute; 2014 [Available from: https://wiki.nci.nih.gov/display/VKC/International+Classification+of
+Diseases+Ninth+Revision+Clinical+Modification.
19. Jelliffe-Pawlowski LL, Baer RJ, Blumenfeld YJ, Ryckman KK, O’Brodovich HM, Gould JB, et al. Mater-
nal characteristics and mid-pregnancy serum biomarkers as risk factors for subtypes of preterm birth.
BJOG: an international journal of obstetrics and gynaecology. 2015; 122(11):1484–93.
20. About Adult BMI Centers for Disease Control and Prevention: Centers for Disease Control and Preven-
tion; 2015 [Available from: https://www.cdc.gov/healthyweight/assessing/bmi/adult_bmi/index.html.
21. Ferre C, Callaghan W, Olson C, Sharma A, Barfield W. Effects of maternal age and age-specific pre-
term birth rates on overall preterm birth rates—United States, 2007 and 2014. MMWR morbidity and
mortality weekly report. 2016; 65(43):1181–84. https://doi.org/10.15585/mmwr.mm6543a1 PMID:
27811841
22. Jin H, Nicodemus-Johnson J. Gender and age stratified analyses of nutrient and dietary pattern associ-
ations with circulating lipid levels identify novel gender and age-specific correlations. Nutrients. 2018;
10(11):pii:E1760.
23. Wild R, Weedin EA, Wilson D. Dyslipidemia in pregnancy. Cardiology clinics. 2015; 33(2):209–15.
https://doi.org/10.1016/j.ccl.2015.01.002 PMID: 25939294
24. Baer RJ, McLemore MR, Adler N, Oltman SP, Chambers BD, Kupperman M, et al. Pre-pregnancy or
first-trimester risk scoring to identify women at high risk of preterm birth. European journal of obstetrics,
gynecology, and reproductive biology. 2018; 231:235–40. https://doi.org/10.1016/j.ejogrb.2018.11.004
PMID: 30439652
25. Bronstein JM, Wingate MS, Brisendine AE. Why is the U.S. preterm birth rate so much higher than the
rates in Canada, Great Britain, and Western Europe? International journal of health services: planning,
administration, evaluation. 2018; 48(4):622–40.
26. Klop B, Elte JW, Cabezas MC. Dyslipidemia in obesity: mechanisms and potential targets. Nutrients.
2013; 5(4):1218–40. https://doi.org/10.3390/nu5041218 PMID: 23584084
27. Jiang S, Jiang J, Xu H, Wang S, Liu Z, Li M, et al. Maternal dyslipidemia during pregnancy may increase
the risk of preterm birth: A meta-analysis. Taiwanese journal of obstetrics & gynecology. 2017; 56(1):9–
15.
28. Leddy MA, Power ML, Schulkin J. The impact of maternal obesity on maternal and fetal health. Reviews
in obstetrics & gynecology. 2008; 1(4):170–8.

PLOS ONE | https://doi.org/10.1371/journal.pone.0209579 December 21, 2018 9 / 10

 Dyslipidemia and preterm birth

29. Papoutsidakis N, Deftereos S, Giannopoulos G, Panagopoulou V, Manolis AS, Bouras G. Treating dys-
lipidemias: is inflammation the missing link? Medicinal chemistry (Shariqah (United Arab Emirates)).
2014; 10(7):643–52.
30. Stone NJ, Robinson J, Lichtenstein AH, Merz CNB, Blum CB, Eckel RH, et al. 2013 ACC/AHA Guide-
line on the Treatment of Blood Cholesterol to Reduce Atherosclerotic Cardiovascular Risk in Adults. A
Report of the American College of Cardiology/American Heart Association Task Force on Practice
Guidelines. 2013.
31. Toleikyte I, Retterstol K, Leren TP, Iversen PO. Pregnancy outcomes in familial hypercholesterolemia:
a registry-based study. Circulation. 2011; 124(15):1606–14. https://doi.org/10.1161/
CIRCULATIONAHA.110.990929 PMID: 21911783
32. Bouhairie VE, Goldberg AC. Familial hypercholesterolemia. Cardiology clinics. 2015; 33(2):169–79.
https://doi.org/10.1016/j.ccl.2015.01.001 PMID: 25939291

PLOS ONE | https://doi.org/10.1371/journal.pone.0209579 December 21, 2018 10 / 10