The relationship between lipid profiles in pregnancy and preterm delivery: a systematic review

Biomed Press An Open Access Publisher © 2018 The Authors. Published by the BioMedPress under the terms of the Creative Commons Attribution License http://creativecommons.org/licenses/ by/4.0/, which permits unrestricted use, provided the original author and source are credited. Background: High-density lipoprotein (HDL), low-density lipoprotein (LDL), triglyceride (TG), and total cholesterol (TC) are associated with maternal hormonal changes which can lead to pregnancy complications. High lipid levels during pregnancy may be accompanied by an increased risk of spontaneous preterm labor. The purpose of this review was to determine the relationship between lipid profiles in pregnancy and preterm delivery. Methods: A systematic search was performed in PubMed, EMBASE, Web of Science, Scopus, and Science Direct databases to find cross-sectional, case-control, and cohort studies on the associations between lipid profiles in pregnancy and preterm delivery. The search results were limited to papers published during 2000-2018. The quality of the selected studies was assessed based on the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) checklist. Finally, a total of 25 eligible high-quality papers were reviewed. Results: A high maternal TG level in the first trimester of pregnancy was identified as a significant risk factor for preterm delivery. Low TG, HDL-cholesterol (HDL-C) and LDL-cholesterol (LDL-C) levels were uniquely associated with an increased risk of preterm delivery. High TC, TG, and LDL-C levels increased the risk of spontaneous preterm labor. A further association was found between high TG and TC levels and an elevated risk of preterm labor before 34 weeks of gestation. The risk of preterm labor before 34 weeks and during the 34th-37th weeks of gestation increased with increasing TG and TC levels. Moreover, overweight women who had a delivery before 34 weeks exhibited significant increases in TC and LDL concentrations during early pregnancy. Conclusion: Lipid metabolism undergoes complicated changes in pregnancy. Lipid levels (LDL, HDL, TC, and TG) above or below the normal range during pregnancy increase the risk of preterm delivery. Considering the significance of maternal lipid metabolism in the incidence of preterm delivery, lipid profile screening in early pregnancy can largely contribute to the prevention of preterm delivery. Further studies are warranted to meticulously examine the effects of lipid profile on preterm delivery. Abstract Received: 17 May 2018 Accepted: 17 June 2018 Published: 25 August 2018 Aghaie et al. Biomedical Research and Therapy 2018, 5(8): 2590-2609


Background
Preterm delivery is one of the main causes of pre-and postnatal morbidity and mortality in developed countries. Preterm delivery accounts for 75-90% of neonatal and prenatal mortalities and serves as the main cause of short-and long-term neonatal defects [1]. About 13 million spontaneous preterm deliveries occur around the world each year. The incidence rates differ in different countries [2] and have been reported as 2.9% in Iran [3]. Preterm labor is associated with increased respiratory disorders, pulmonary edema, and impaired glucose tolerance in the newborn [4]. Premature infants may develop a number of short-and long-term conditions including respiratory distress syndrome, bronchopulmonary dysplasia, apnea of prematurity, hyperbilirubinemia, nosocomial infections, immunodeficiency, patent artery duct, intraventricular hemorrhage, bacterial and fungal sepsis, precocious retinopathy of prematurity, necrotizing enterocolitis, food intolerance, mental retardation, blindness, and hearing loss [5]. While approximately 10-15% of highly premature neonates show symptoms of neurodegenerative defects, such as cerebral palsy, more complications including learning disability and emotional problems may be seen in 50% of these infants [1].
Despite advancements in neonatal care, premature infants are at higher risk for mortality and those who survive may experience neurodevelopmental delay [6]. A variety of factors, some of which have not been clearly identified, may be responsible for spontaneous preterm delivery. Some studies have reported relationships between risk factors for maternal cardiopulmonary complications and undesirable pregnancy outcomes, such as low gestational age and low birth weight. Associations have also been established between HDL, LDL, TG, and TC levels and maternal hormonal changes, and in turn, pregnancy problems and changes in the fetal development process [7,8]. While the exact mechanisms through which these changes in the maternal plasma lipid profile affect fetal development and birth are not clear, recent studies have highlighted the significant biochemical and physiological effects of cholesterol on the function of fetal membranes and neonatal neurodevelopment.
The transfer of maternal cholesterol to the fetus through the umbilical cord might justify these effects. A previous study reported that low birth weight infants had lower HDL levels compared to their term counterparts [9,10]. There is still controversy over the association between maternal lipid profile and preterm delivery. In several studies, both high and low HDL, LDL, TG, and TC levels almost doubled the risk of spontaneous preterm delivery [11]. Some studies, on the other hand, have used maternal serum screening and rejected any relationships between the highest/lowest lipid levels and preterm labor [12]. A systematic review found that the risk of spontaneous preterm delivery could be increased by high TG levels, but was not affected by HDL-C and LDL-C levels [13]. Meanwhile, Heida et al. (2017) reported lower TC and LDL-C levels in women with spontaneous preterm delivery, but could not establish a significant relationship between undesirable lipid levels and preterm delivery [14]. Since changes in fat metabolism during pregnancy can be associated with adverse pregnancy outcomes, this review aimed to determine the association between lipid profiles during pregnancy and preterm delivery.

Methods (a) Search strategy
The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines were followed in the report of this study [15]

(b) Inclusion and exclusion criteria
Cohort and case-control studies published in English or Persian during 2000-2018 were included if they evaluated the relationship between lipid profiles during pregnancy and preterm labor (from the 28-37 th week of gestation) in mothers who had no particular medical conditions (preeclampsia or diabetes), and gave birth to a singleton infant through vaginal delivery. Case report studies of infants with low Apgar score (<7), women with multifetal pregnancy, and studies with incomplete information were excluded.

(c) Study selection
After the initial search, 1149 studies were extracted and independently evaluated by two authors to determine eligibility. Any cases of disagreement were resolved through a consensus. As a result, 434 irrelevant or duplicate articles were excluded. The titles and abstracts of the remaining 715 studies were then assessed and 545 more papers were excluded. Evaluating the full texts of the remaining 170 articles resulted in the exclusion of 145 ineligible articles and confirmed the eligibility of 25 papers (Figure 1).

(d) Quality assessment
The adherence of quantitative studies to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) checklist [16,17] was evaluated as a measure of their quality. Studies were classified as high-, medium-, and low-quality if they adhered to all seven items, six items, and two or more items of the STROBE, respectively.

(e) Data extraction
Study selection and validity assessment were independently performed by two authors and cases of disagreement were resolved by consulting a third researcher. The required data included the first author's name, publication year, study design, country of study, sample size, gestational age at sampling and delivery, outcomes, odds ratio (OR), risk ratio (RR), mean, and standard deviation (SD) of TC, TG, HDL and LDL. The data were extracted from all studies and entered into the analysis.

Results
All the 25 studies selected after the search process met the inclusion criteria and showed high quality according to their adherence to the STROBE checklist ( Figure 1). As seen in Tables 1, 2, 3, 4, 5, 6 and 7, the sample size of the selected studies ranged between 80 and 9938 women and provided a pool of 31833 women. These women were of different ages and were selected from different countries including USA (n=8), China (n=4), Iran (n=2), Japan (n=2), Netherlands (n=2), Turkey (n=1), India (n=1), Greece (n=1), Nigeria (n=1), Finland (n=1), Canada (n=1), and Spain (n=1). The selected studies had cohort or case-control designs (n=17 and n=8, respectively). Evaluating the results of the selected studies suggested high maternal TG level in the first trimester of pregnancy as an important and significant factor in preterm delivery. Low TG, HDL-C and LDL-C levels were exclusively related to the increased risk of preterm delivery. High TC, TG and LDL-C levels were also associated with an elevated risk of spontaneous preterm labor. A further association was observed between high TG and TC levels and increased risk of preterm labor before the 34 th week of pregnancy and during the 34 th -37 th weeks. Moreover, considerably high TC and LDL levels were found in overweight women who had a delivery before 34 weeks Tables 8 and 9.   [14] Netherlands cohort 165 Grace (2017) [28] USA cohort 715 Qiu (2017) [29] China case-control 300 Li (2015) [30] China case-control 378 Ghodke (2017) [

Discussion
Given the significance of preterm delivery, we reviewed the results of high-quality studies on the associations between lipid levels during pregnancy and preterm birth. Our review indicated high maternal TG level in the first trimester of pregnancy as a significant risk factor for preterm delivery. In a case-control study, Catov et al. (2007) reported an association between lipid profile and preterm labor, i.e. high TG and cholesterol levels increased the risk of preterm labor before the 34 th week and during the 34 th -37 th weeks of gestation. Overweight women who delivered before 34 weeks had particularly higher levels of cholesterol and LDL. TG levels were also increased in underweight women and these women showed lower TG responses during the first half of pregnancy [18]. Vrijkotte et al. (2012) investigated maternal lipid profile in early pregnancy and its association with pregnancy complications and concluded that high first trimester TG levels could be a significant risk factor for preterm delivery [19]. Yonezowa et al. (2009) examined umbilical cord lipids in preterm infants and confirmed low TG levels in preterm infants. They, thus, identified 34 weeks of pregnancy as a critical period for TG metabolism and reported considerable increments in umbilical cord TG levels during the 32 th -34 th weeks [20]. Ghodake et al. (2017) established an association between early pregnancy TG levels and preterm delivery. However, no such relations were found in cases of TC, LDL, and HDL levels [21]. Catov et al.
(2017) examined the heterogeneity of lipoproteins during pregnancies leading to preterm delivery and showed that preterm labor was seen in women with high TG levels. However, LDL-C had no direct relation with preterm delivery [22]. Likewise, Niromanesh et al. (2012) underlined the positive association between hypertriglyceridemia and preterm delivery [23].
We also found evidence of the relationship between low TG, HDL-C, and LDL-C levels and increased risk of preterm delivery. Indeed, elevations in TC, TG, and LDL-C levels were also associated with an increased risk of spontaneous preterm labor. According to Chen et al. (2017), despite differences in lipid levels of different races and ethnicities, higher HDL-C levels during early pregnancy universally increased the risk of spontaneous preterm labor [24]. Catov et al.
(2014) identified high HDL levels as a risk factor for spontaneous preterm labor [25]. Kramer et al. (2009) investigated the vasculopathic and thrombophilic risk factors for preterm delivery and highlighted the contribution of reductions in maternal lipids to the incidence of preterm labor. They further noted the associations between spontaneous preterm delivery and elevated TC and LDL levels. Their findings also suggested the protective role of high HDL levels in preterm delivery [26]. Kaser et al. (2001) evaluated lipid levels in infants and reported small for gestational age (SGA) neonates to have lower HDL, LDL, and TC levels, compared to appropriate for gestational age (AGA) infants [27]. Moreover, Heida et al. (2017) found no correlations between spontaneous preterm delivery and unfavourable lipid levels or cardiovascular biochemical risk factors in mothers [14]. Grace et al. (2017) examined the relationship between maternal serum lipoproteins and gestational age, and reported LDL and HDL concentrations to be independently related with gestational age at preterm delivery [28]. Qiu et al. (2017) studied maternal serum lipid profiles and homocysteine in late pregnancy and showed higher TC and LDL levels in mothers with preterm delivery than in those with term delivery. In contrast, mothers with term delivery had higher HDL levels compared to those with preterm delivery [29].
Moreover, Li et al. (2015) investigated the relationships between lipid levels and singlenucleotide polymorphisms of APOE, ABCA1, and HMGCR genes in women with preterm labor. They found that HDL, TG, and LDL levels decreased in mothers with preterm delivery. They also established relationships between maternal TG and LDL levels and HMGCR genotype, and concluded that HMGCR might be a suitable gene for preterm delivery [30]. Wang et al. (2017) found associations between elevations in first trimester TG, TC, LDL, and HDL levels and risk of preterm labor.
On the other hand, low LDL, TG, and TC levels and a high HDL level during pregnancy might have a protective role against preterm delivery [31]. Maymunah    -A dramatic increase in cord blood TG was observed during 32-34 weeks and the development of respiratory distress syndrome seemed to be inhibited afterwards.
-Higher LDL-C levels were seen in preterm infants than in term neonates.
-LDL-TG levels were significantly lower in preterm neonates than in term neonates.
-Preterm and term neonates had no differences in terms of HDL-C and HDL-TG levels.
Chen 2017 -A more favorable lipid profile and thus better cardiovascular protection was observed in African-American women.
-Higher HDL-C and apoA1 increased the risk of SPTD and this association was independent of ethnicity. SD: 0.69 -Elevations in early pregnancy TC, TG, and LDL-C levels, along with a reduction in first trimester HDL-C, increased the incidence of adverse pregnancy outcomes.
-Low early pregnancy TC, TG, and LDL-C levels, along with a high HDL-C level, might prevent the incidence of adverse pregnancy outcomes.
-The risk of PTB was higher in women with metabolic syndrome in early pregnancy.     [33]. Additionally, women with preterm labor reportedly had lower HDL-C concentrations [34]. Our review indicated associations between high TG and TC levels and increased risks of preterm delivery before 34 weeks and at 34-37 weeks of pregnancy. Catov et al. (2010) presented evidence of an association between pre-pregnancy cholesterol levels and the risk of preterm birth. This association was independent of age, race, parity, BMI, blood pressure, and physical activity during labor. There was also a relationship between cholesterol and labor before 34 weeks in women with normal blood pressure. No associations could be established between TG, HDL, and LDL concentrations, with preterm labor [35]. Steffen et al. (2007) examined changes in the cholesterol metabolism gene in infants born between 22 and 36 weeks and their mothers. They concluded that changes in the cholesterol metabolism gene in mothers and infants were associated with preterm labor and low birth weight [36]. These results indicate that genetic markers can help identify mothers who are at risk for dyslipidemia and preterm labor.

Study
Toleikytei (2011) did not detect any differences in the risk of preterm labor between women with hypercholesterolemia and other women of childbearing age [37]. Edison et al. (2007) highlighted the involvement of low TC levels in the pathogenesis of preterm delivery. They explained that poor nutritional status contributed to low TC levels which, in turn, increased the risk of undesirable pregnancy outcomes and preterm labor [38]. Liu-x (2011) studied the effect of pre-pregnancy BMI on premature labor and concluded that high levels of TC may be accompanied by other risk factors, such as unfavourable lifestyle and dietary habits [39]. A study on the risk factors for spontaneous preterm delivery in adolescents [40] revealed that high BMI and elevated TC levels during pregnancy could have a protective effect against spontaneous premature labor in teen pregnancies [40].
According to our review, overweight women with preterm labor (before 34 weeks) presented with significantly higher TC and LDL concentrations during early pregnancy. Chatzi et al. (2009) examined the relationship between metabolic syndrome in pregnancy and the risk of preterm delivery. They showed that high early pregnancy cholesterol increased the risk of preterm delivery [41]. Shoji et al. (2014) stated that umbilical cord lipid profiles widely varied depending on maternal and infant weights and gestational age. Premature fetal development may affect lipid metabolism and atherogenic indices. A critical period for neonatal TG metabolism is 34 weeks during which premature infants have lower TG levels [42]. Late pregnancy TG concentrations have been stated to be independently associated with a reduced risk of SGA [43]. TC and LDL levels in AGA infants tend to be higher than those of SGA neonates [43]. Another study further reported that TC and LDL levels were higher in AGA infants than in SGA neonates (48). These observations signify the importance of maternal lipid profile screening for preventing pregnancy complications.

Conclusion
Lipid metabolism undergoes complicated changes in pregnancy. Lipid levels (LDL, HDL, TC and TG) above or below the normal range during pregnancy would increase the risk of preterm delivery. Considering the significance of maternal lipid metabolism in the incidence of preterm delivery, lipid profile screening in early pregnancy can largely contribute to the prevention of preterm delivery. Further studies are warranted to meticulously examine the effects of lipid profile on preterm delivery.