Elevated Mid-trimester 4-hour Postprandial Triglycerides for Predicting Late-onset Preeclampsia and Gestational Hypertension: a Prospective Screening Study


 Background: Abnormal maternal lipid concentrations are associated with increased risk of preeclampsia. However, previous studies mainly focused on fasting lipid concentrations, scarce data have been published on the relationship between postprandial triglyceride (TG) concentrations in the second trimester and the risk of preeclampsia. Our aim is to evaluate the potential of triglyceride (TG) concentrations at the time of oral lipid tolerance test (OLTT) measurement in the second trimester to predict preeclampsia and to elucidate the lipid metabolic changes related to these diseases. Methods: This is a prospective cohort study of Pregnant women at 12-24 weeks of gestation undergone an OLTT in a university affiliated hospital between May 2019 and January 2020. Data were stratified into binaries according to the OLTT results. The receiver operating characteristic (ROC) curve analysis was conducted to determine the optimal cut-off points of TG, HDL-C, LDL-C, sd-LDL, FFA, and BG for predicting preeclampsia.Results: 438 pregnant women were recruited to undergo an OLTT at 12-24 weeks of gestation. Among these, 24 women developed preeclampsia and 414 women remained normotensive. Women who subsequently developed preeclampsia had higher concentrations of 4-h postprandial TG than those who remained normotensive. In the linear logistic regression analyses of potential confounding factors, mid-trimester 4-h postprandial TG concentrations at the time of OLTT measurement were significantly higher in preeclamptic cases than in controls.Conclusions: Dyslipidemia in the second trimester of pregnancy, particularly postprandial hypertriglyceridemia, appears to be associated with an increased risk of preeclampsia. Mid-trimester 4-h postprandial TG concentration at the time of OLTT measurement may be a potential predictive marker of preeclampsia.

trimester was associated with the development of preeclampsia, overall suggesting that hypertriglyceridemia may be involved in the development of preeclampsia (4). Increasingly, the role of lipids has been recognized as critically important in vascular risk modi cation, and it is possible, therefore, that lipids in pregnancy has an impact on placental vascular development, which may be critical in the development of preeclampsia (5).
During pregnancy, the lipid concentrations increase gradually, resulting in a physiological hyperlipidemia state. Previous studies mainly focused on fasting lipid concentrations, neglected that we're in a postprandial state in most of the day (6, 7). Postprandial dyslipidemia can re ect the decreased ability of pregnant women to metabolize and clear lipids, which occurs before fasting dyslipidemia. However, only a few studies have characterized the relationship between postprandial triglyceride (TG) concentrations in the second trimester and the risks of preeclampsia.
The postprandial response to an oral lipid tolerance test (OLTT) is a better metabolic challenge to emulate nutrient overload in comparison to an oral glucose tolerance test (OGTT). OLTT is used for screening dyslipidemia (8). Here, we tested the hypothesis that elevated postprandial TG concentrations are correlated with an increased likelihood of preeclampsia in Chinese women. We aimed to determine whether pregnant women with abnormal OLTT results during the second trimester are at a high risk of developing preeclampsia.

Participants and ethical statement
This prospective cohort study enrolled 506 pregnant women who attended the prenatal examination clinics in Shanghai General Hospital between May 2019 and January 2020. In the present analysis, we excluded one woman who had a miscarriage in the second trimester, and 67 women who met the following exclusion criteria: (1) multiple pregnancy; (2) unnatural pregnancy; (3) polycystic ovarian syndrome, thyroid disease, diabetes, heart disease, or hyperlipidemia; (4) combined with autoimmune diseases; (5) combined with intrahepatic cholestasis of pregnancy or viral hepatitis; and (6) fasting TG concentrations > 5.6 (mmol/L). The remaining 439 women were included in the nal analysis (Fig. 1).
These subjects were recruited to undergo an OLTT at 12-24 gestational weeks. A uni ed standard fat meal referring to the guide was provided for the subjects; the calorie of which was approximately 680 kcal, consisting of a 50-g egg, 30-g butter, 100-g two pieces of toast bread and 250-ml whole milk. The energy proportion of fat, protein, and carbohydrate were 59%, 12% and 29%, respectively (8). TG, free fat acid (FFA), small dense low-density lipoprotein (sd-LDL) were measured by blood sampling at fasting, and TG was measured at 4 h after the fat meal. Each enrolled pregnant woman underwent a 75-g OGTT at 24-24 weeks of gestation. In addition, fasting lipid concentrations, including serum total cholesterol (TC), TGs, high-density lipoprotein-cholesterol (HDL-C), and low-density lipoprotein-cholesterol (LDL-C) as well as insulin, alanine aminotransferase (ALT), and aspartate aminotransferase (AST) concentrations were measured. Women were diagnosed with GDM if one or more of the following plasma glucose values during the 75-g OGTT at 24-28 gestational weeks were met or exceeded: 0 h, 5.1 mmol/L; 1 h, 10.0 mmol/L; and 2 h, 8.5 mmol/L. Each patient provided informed consent, and the research was conducted in compliance with the Declaration of Helsinki. This trial has been registered in Chinese Clinical Trial

Grouping
All pregnant women were categorized into the following binaries according to the OLTT results at the midtrimester: low binary [fasting TG (FTG) < 2.4 mmol/L(median); 4 h, TG < 3.47 mmol/L(median)] and high binary (FTG ≥ 2.4 mmol/L; 4 h, TG ≥ 3.47 mmol/L) groups. The women were also categorized into the following two groups according to the blood pressure levels and the presence or absence of proteinuria: normal and preeclampsia groups (Fig. 1).

Data collection
The following maternal characteristics were assessed: age at delivery, height, body weight before pregnancy, blood pressure, and insulin treatment during pregnancy.

Statistical analysis
Data were expressed as the mean ± standard deviation for normally distributed variables and as the median with the interquartile range for skewed data. To determine the differences between the groups, we conducted an analysis of variance or the Kruskal-Wallis test for continuous variables and a chi-square test for categorical variables. Least signi cant difference tests were used to perform pairwise comparisons between two groups. The receiver operating characteristic (ROC) curve analysis was conducted to determine the optimal cut-off points of TG, HDL-C, LDL-C, sd-LDL, FFA, and BG for predicting preeclampsia. Each optimal cut-off point was assessed by searching for the maximum value of sensitivity + speci city-1 (Youden index). The area under curve (AUC) was calculated to evaluate the predictive powers. Logistic regression analysis was performed to determine whether the elevated postprandial TG concentrations during mid-gestation were independently associated with preeclampsia.
Statistical signi cance was set at p < 0.05. Statistical analyses were performed using SPSS (IBM Corp., Armonk, NY).

Results
Demographic and metabolic characteristics of all patients in the second trimester who were categorized into binaries according to TG concentrations at the time of OLTT measurement There were signi cant differences in the metabolic levels during pregnancy between the high and low fasting TG groups. The high fasting TG binary group had a higher pre-pregnancy body mass index (BMI) (22.39 ± 2.11 versus 20.82 ± 1.78 kg/m 2 , p < 0.01), higher fasting PG concentrations (4.71 ± 0.46 versus 4.59 ± 0.35 mmol/L, p = 0.03), higher HOMA-IR (1.60 (1.14-2.23) versus 1.23 (1.01-1.66), p < 0.01), greater neonatal weight (3318.93 ± 418.02 versus 3260.50 ± 415.58 g, p = 0.03), and lower HDL-C concentrations (1.84 ± 0.35 versus 2.05 ± 0.39 mmol/L, p < 0.01) than the low fasting TG binary group (Table 1). .60 ± 0.34 mmol/L, p = 0.01), and lower HDL-C concentrations (1.83 ± 0.38 versus 2.07 ± 0.39 mmol/L, p < 0.01), and lower HDL-C concentrations (2.95 ± 1.00 versus 3.14 ± 0.76 mmol/L, p = 0.04) than the low 4-h TG binary group (Table 2). In addition, the prevalence rates of preeclampsia were 1.83% and 9.09% (p = 0.015) in the two groups (4-h TG < 3.47 mmol/L and 4-h TG ≥ 3.47 mmol/L, respectively) ( Table 2). Maternal plasma lipid and lipoprotein concentrations are shown in Table 3. Women who subsequently developed preeclampsia had higher plasma glucose and 4-h postprandial TG concentrations and HOMA-IR than the normotensive group. Associations between maternal lipid pro le and preeclampsia Table 4 presents the optimal cut-off points of maternal lipid concentrations in the second trimester for predicting preeclampsia. Four-hour postprandial TG concentrations predicting preeclampsia had the strongest predictive power with the largest AUC [0.680 (95% con dence interval [CI]: 0.505-0.854)]. The optimal cut-off points for second-trimester FTG and 4-h postprandial TG in predicting preeclampsia were ≥ 2.93 and 3.92 mmol/L, respectively. Besides, the optimal cut-off points for LDL-C, FFA, sd-LDL, and HDL-C in predicting preeclampsia were ≥ 1.8, 1.46, 29.085 and ≤ 2.42 mmol/L, respectively. Moreover, only high 4-h postprandial TG in the second trimester was signi cantly associated with the morbidity of preeclampsia (p = 0.045).  Table 5). In addition to the elevated 4-h TG concentrations, we also found that an abnormal OGTT result (OR: 3.869, 95% CI: 1.179-12.697, p = 0.026) was a risk factor for preeclampsia. However, no association was found between other factors and preeclampsia.

Discussion
Early identi cation of women at a high risk of preeclampsia might enable potential prophylactic treatment to reduce or avoid the onset of symptoms (9,10). Preeclampsia is more common and has a lower detection rate (11). Predictive models for preeclampsia have employed a combination of maternal characteristics as well as biochemical and biophysical markers before 14 weeks of gestation to predict the syndrome at 30-60% sensitivity (10)(11)(12)(13)(14).
Our ndings are consistent with the results of the few available prospective cohort studies on maternal non-fasting plasma lipid and lipoprotein concentrations in preeclamptic and normotensive pregnancies (15,16). We observed an association between maternal dyslipidemia in the second trimester, particularly 4-h postprandial hypertriglyceridemia, and the subsequent risk of preeclampsia. We evaluated both the FTG and 4-h postprandial TG concentrations at the time of OLTT measurement to show that 4-h postprandial TG concentrations might be an independent predictive biomarker for preeclampsia, which might be more meaningful compared to the FTG concentrations (Table 5). This is in line with the results of a comprehensive study (15) showing the associations between non-fasting plasma TG at 18 weeks of gestation and the risk of preeclampsia. In another prospective cohort study (16), a linear increase in preeclampsia risk was observed with increasing non-fasting plasma TG concentrations before 16 weeks' gestation. However, fasting TG concentrations in the second trimester were not associated with the development of preeclampsia, which was inconsistent with Oya's cohort study ndings (OR: 1.1 (1.1-1.2), p = 0.004) (17).
Although, the temporality, wherein 4-h postprandial hypertriglyceridemia clearly precedes the onset of preeclampsia, leads us to generate the hypothesis that we may be able to change the natural history of the disease if we intervene early by lowering the postprandial TG concentrations. Maternal lipids offer a potential therapeutic target, and it is worth noting that the current treatments for diabetes in pregnancy, including dietary and lifestyle changes as well as pharmacotherapy, all alter not only maternal glucose but also maternal lipids (3,18,19). Before such an intervention, it would be important to de ne the normal postprandial triglyceride concentrations in pregnancy and correctly identify women with lipid metabolic disturbances who could bene t most from this therapy.
High postprandial TG concentration is positively correlated with insulin resistance, and insulin resistance exists in preeclampsia patients (Tables 2 and 3). Changes in TG metabolism and clearance are mediated by insulin resistance. Women with elevated lipid concentrations are more likely to have preexisting endothelial dysfunction that worsens as a result of the physiological burden of pregnancy; this condition may be further exacerbated by increased maternal vascular in ammation (20). This was supported with evidence from a murine model, which suggested that, in pregnancy, excessive hyperlipidemia may lead to an accentuated in ammatory response, oxidative stress, and subsequent dysfunction of cerebral arteries, compromising cerebral percussion and contributing to preeclampsia-related neurological complications (21).
The sensitivity of the prediction of preeclampsia using 4-h postprandial TG concentration at the time of OLTT measurement in this study (72.7%) was markedly higher than that of previously published prediction of preeclampsia based on clinical markers such as mean arterial pressure, maternal age, and uterine artery pulsatility index in a cohort of women at 11 + 0 to 13 + 6 gestational weeks (38.5%) (22) and in another nulliparous cohort (37%) (23). However, the sensitivity at 10% false positive rate of prediction of preeclampsia is low (18.2%).
The major strength of our study was its prospective design with complete follow-up of 438 women with OLTT measurement. This study design more accurately re ects the predictive power of 4-h postprandial TG compared to a case-control design. Our high participation (86.6%) and follow-up (99.5%) rates minimized the likelihood of selection bias, which may affect the reported results. Second, the fasting and 4-h postprandial blood lipid concentrations obtained after a uni ed fat meal in the second trimester suggested that the disordered plasma lipid concentrations precede preeclampsia. The uni ed standard fat meal and blood sampling collection time can avoid lipid data deviation. Third, the exclusion of pregnant women with high risk factors for metabolic syndrome made the results more convincing. Fourth, compared to the fetus in the rst trimester, the fetus in the second trimester grows rapidly and needs more energy. The mother gradually switches to a catabolic condition resulting in an increased breakdown of lipid. Therefore, detecting postprandial lipid concentration changes (the ability of lipid metabolism) in the second trimester is more sensitive. Finally, we used the logistic regression to adjust for a number of confounders to obtain a conclusion.
Several important limitations must be considered when interpreting the results of our study. First, our study included a limited number of cases, which is di cult to overcome given the low incidence of these disorders. Our relatively small number of patients with preeclampsia hindered us from making inferences from some of our analyses. Further studies involving multiple centers and a larger number of patients are necessary in order to validate our ndings. In addition, although we compared many potential confounders, we cannot exclude the possibility of the other confounding factors from unmeasured covariates.

Conclusions
In conclusion, the results of our study served to highlight the importance of 4-h postprandial TG at the time of OLTT measurement in the second trimester as a possible predictive marker of preeclampsia.
Moreover, dyslipidemia, particularly 4-h postprandial hypertriglyceridemia, may be of etiologic and pathophysiologic importance in women who developed preeclampsia. Our data may contribute to the development and evaluation of behavioral and medical interventions aimed at reducing the occurrence of these disorders. On the other hand, preeclampsia appears to be a multi-etiological syndrome with heterogeneous biologic pathways.

Declarations
Ethics approval and consent to participate