Abstract
Purpose
Dyslipidemia has been linked to adverse pregnancy outcomes in observational studies. This study aimed to explore how variations in lipid levels during the first trimester might influence early pregnancy loss (EPL).
Methods
Blood samples from pregnant women were analyzed to examine the relationship between EPL and lipid metabolism using logistic regression and restricted cubic splines (RCS). Sensitivity analysis was conducted to verify the robustness of the results.
Results
Elevated low-density lipoprotein cholesterol (LDL-C) and total cholesterol (TC) levels at most times of 4–9 weeks of gestation were associated with a higher risk of EPL, regardless of whether the control group was successful pregnancy or live birth. Specifically, taking the successful pregnancy group as a control example, increased EPL risks were observed in the highest quartile of plasma TC at 4 weeks (OR = 2.18, 95%: 1.14–4.21) and 7 weeks (OR = 4.30, 95%: 1.87–9.93) of pregnancy. Significant EPL risks were also noted in the third (Q3) and fourth (Q4) quartiles of LDL-C at 4 weeks (Q3, OR = 2.98, 95%: 1.47–6.08; Q4, OR = 2.66, 95%: 1.27–5.55) and 7 weeks (Q3, OR = 3.12, 95%: 1.44–6.73; Q4, OR = 5.17, 95%: 2.14–12.49). High TC levels (> 3.25–3.78 mmol/L) and high LDL-C levels (> 1.92–2.04 mmol/L) were linked to an increased risk of EPL compared to lower levels of TC (≤ 2.91–3.05 mmol/L) and LDL-C (≤ 1.64–1.75 mmol/L).RCS analysis further confirmed this finding that plasma TC and LDL-C levels at 4 and 7 weeks of gestation may have a linear relationship with the risk of EPL. By the way, triglyceride levels at 6 and 8 weeks of gestation were associated with a higher risk of EPL, whereas high-density lipoprotein cholesterol (HDL-C) levels at 5 and 9 weeks of gestation have a completely opposite relationship with EPL risk.
Conclusions
Elevated cholesterol levels during the first trimester are associated with an increased risk of early pregnancy loss, emphasizing the need for lipid monitoring during pregnancy and even before pregnancy.
Supplementary Information
The online version contains supplementary material available at 10.1007/s00404-024-07893-5.
Keywords: Dyslipidemia, Cholesterol, Pregnancy loss
What does this study add to the clinical work
| Our study suggests that elevated cholesterol levels during the first trimester are associated with an increased risk of early pregnancy loss. Therefore, monitoring and managing cholesterol levels during this critical period could potentially mitigate this risk. |
Introduction
Pregnancy loss (PL) is generally defined as the spontaneous end of fetal development before 24 weeks of gestation [1]. About 30% of pregnant women experience PL before the 12th week of gestation [2, 3], but this rate drops significantly after 12 weeks, falling to around 1% [4]. Various factors can contribute to PL, such as uterine abnormalities, genetic issues, hormonal imbalances, and immune system irregularities [1, 5, 6]. Nevertheless, 50% to 70% of couples who experience PL do not have a clear cause [5].
Metabolic changes in blood lipids during pregnancy are among the most consistent alterations observed. Epidemiological data indicate that maternal dyslipidemia, marked by decreased high-density lipoprotein cholesterol (HDL-C) and increased total cholesterol (TC), low-density lipoprotein cholesterol (LDL-C), and triglyceride (TG) levels, may lead to pregnancy complications such as preeclampsia and preterm birth [7–9]. Additionally, a secondary analysis of a randomized controlled trial indicated that elevated pre-pregnancy TC, LDL-C and TG levels were linked to poorer pregnancy and live birth outcomes [10]. In addition, high LDL-C and low HDL-C levels before conception are linked to PL before 28 weeks [11]. This highlights the importance of lipid metabolism in maternal health and pregnancy outcomes.
Despite the common occurrence of early pregnancy loss (EPL), research on its connection to blood lipid profiles is sparse. Our study seeks to fill this gap by examining the relationship between early pregnancy blood lipid levels and EPL risk.
Methods
Study population
From August 2020 to June 2022, pregnant women aged 18 to 45 were recruited at The Second Hospital of Lanzhou University in China for a study on pregnancy loss. Ethical approval was granted (Ethics Approval No. NO.2023A-553). Out of 1152 patients initially with plasma lipid testing during pregnancy, 1102 were included in the study (see Fig. 1). Inclusion criteria were: age over 18 years and having had tests for plasma TC, TG, HDL-C, and LDL-C. Exclusion criteria included loss to follow-up, pregnancy termination due to fetal anomalies, molar pregnancy, ectopic pregnancy, and late pregnancy loss (beyond 12 weeks). None of these patients included had serious heart disease, liver disease, kidney disease, high blood pressure, or diabetes.
Fig. 1.
Flowchart of the inclusion and exclusion criteria for the study population
Pregnancy outcomes were categorized as follows: Early Pregnancy Loss (EPL), the case group, defined as pregnancy loss before 12 weeks of gestation; Live Birth (LB), the control group, defined as delivery at 37 weeks of gestation or later; and Successful Pregnancy (SP), the control group, which includes both live births and preterm births, defined as deliveries at 28 weeks of gestation or later. Biochemical pregnancy, characterized by a positive pregnancy test followed by a negative test without an ultrasound assessment [12], was included to enhance the applicability of the findings. Pregnancy outcomes were gathered through participant interviews, telephone follow-ups, and medical records.
Measurements of lipide indicators
Participant information, including age, body mass index (BMI), education level, ethnicity, history of prior pregnancy loss, and previous live births, was collected. BMI was calculated by dividing weight (in kilograms) by the square of height (in meters) (kg/m2). Participants provided blood samples at different times during their pregnancies after fasting overnight. Blood samples were collected at The Second Hospital of Lanzhou University and analyzed using the VITROS 5600 Integrated System (Johnson & Johnson, United States) to measure TC, TG, LDL-C, and HDL-C. The reference ranges for plasma lipid levels at the hospital were: TC 2.30–5.20 mmol/L, TG 0.56–1.70 mmol/L, LDL-C 1.20–3.30 mmol/L, and HDL-C 1.03–1.55 mmol/L. Abnormal lipid levels were defined as: TC levels ≥ 5.2 mmol/L, TG levels ≥ 1.7 mol/L, HDL-C levels < 1.0 mmol/L and LDL-C levels ≥ 3.4 mmol/L [13].
Statistical analysis
Categorical variables were summarized using counts (n) and percentages (%), while continuous variables were expressed as medians with interquartile ranges. The Mann–Whitney U test was used to compare two groups due to the non-normal distribution of continuous variables. Fisher's exact test or the Chi-squared test was applied to examine categorical variables based on their frequencies. The concept of a gestational week was elucidated through an illustrative example, wherein a gestational week of 4 weeks was defined as the period from week 4 + 0 day to week 4 + 6 days of gestation.
Logistic regression was employed to estimate adjusted odds ratios (OR) with 95% confidence intervals (CI) to assess the association between risk of EPL and plasma lipid concentrations at different gestational weeks. Plasma lipid levels were categorized into quartiles based on the distribution in the control group. This stratification was applied in all sub-analyses, including those based on age, BMI status, and number of pregnancy losses. To explore potential non-linear relationships between plasma lipid levels at various gestational weeks and the likelihood of EPL, we used an adjusted restricted cubic splines (RCS) model with four knots.
A sensitivity analysis was conducted to verify the stability of our results. We examined the relationship between blood lipid levels measured at 4 through 10 weeks of gestation and EPL. Additionally, we used LB and SP groups as control groups to further investigate the association between blood lipids at different gestational weeks and EPL. All statistical analyses were performed using R version 4.2.2, with a significance level set at p < 0.05 for all tests.
Results
Table 1 presents the demographic characteristics and lipid concentrations of women with measurements taken at 4 and 7 weeks of gestation. Notable differences in TC and LDL-C levels were observed between successful pregnancy (SP) and early pregnancy loss (EPL) patients.
Table 1.
Demographic characteristics and lipid concentrations of people, stratified by case–control
| Demographic | 4 weeks of gestationa | 7 weeks of gestationb | ||||
|---|---|---|---|---|---|---|
| EPL (n = 100) | SP (n = 349) | ρ-value | EPL (n = 81) | SP (n = 525) | ρ-value | |
| TC, mmol/L | 3.64 (3.17, 4.03) | 3.38 (3.05, 3.78) | < 0.01 | 3.60 (3.21, 3.94) | 3.25 (2.91, 3.75) | < 0.01 |
| TG, mmol/L | 0.90 (0.72, 1.17) | 0.89 (0.65, 1.20) | 0.42 | 1.21 (0.95, 1.65) | 1.12 (0.88, 1.44) | 0.08 |
| LDL-C, mmol/L | 2.27 (1.95, 2.53) | 2.05 (1.75, 2.39) | < 0.01 | 2.15 (1.85, 2.51) | 1.92 (1.62, 2.26) | < 0.01 |
| HDL-C, mmol/L | 1.31 (1.12, 1.46) | 1.28 (1.11, 1.46) | 0.56 | 1.31 (1.07, 1.49) | 1.24 (1.10, 1.43) | 0.49 |
| Age, years | 31.0 (29.0, 34.0) | 30.0 (28.0, 32.0) | < 0.01 | 31.0 (28.0, 34.0) | 30.0 (28.0, 32.0) | 0.15 |
| BMI, kg/m2 | 22.4 (20.3, 24.1) | 21.6 (19.8, 24.0) | 0.19 | 23.00 (21.0, 24.0) | 22.0 (20.0, 24.0) | 0.11 |
| Age stratified, years | 0.03 | 0.26 | ||||
| ≤ 30 | 43 (43.0%) | 196 (56.2%) | 38 (46.9%) | 281 (53.5%) | ||
| 30–35 | 41 (41.0%) | 122 (35.0%) | 31 (38.3%) | 195 (37.1%) | ||
| > 35 | 16 (16.0%) | 31 (8.9%) | 12 (14.8%) | 49 (9.3%) | ||
| BMI stratified, kg/m2 | 0.89 | 0.38 | ||||
| < 18.5 | 4 (4.0%) | 13 (3.7%) | 3 (3.7%) | 22 (4.2%) | ||
| 18.5–23.99 | 34 (34.0%) | 132 (37.8%) | 25 (30.9%) | 200 (38.1%) | ||
| 24–27.99 | 10 (10.0%) | 41 (11.7%) | 8 (9.9%) | 73 (13.9%) | ||
| ≥ 28 | 4 (4.0%) | 10 (2.9%) | 3 (3.7%) | 19 (3.6%) | ||
| Unknow | 48 (48.0%) | 153 (43.8%) | 42 (51.9%) | 211 (40.2%) | ||
| Education level | 0.26 | 0.05 | ||||
| Less than high school | 6 (6.0%) | 15 (4.3%) | 2 (2.5%) | 35 (6.7%) | ||
| High school | 6 (6.0%) | 28 (8.0%) | 8 (9.9%) | 41 (7.8%) | ||
| More than high school | 31 (31.0%) | 140 (40.1%) | 23 (28.4%) | 211 (40.2%) | ||
| Unknow | 57 (57.0%) | 166 (47.6%) | 48 (59.3%) | 238 (45.3%) | ||
| Ethnicity | 0.52 | 0.09 | ||||
| Minority | 5 (5.0%) | 16 (4.6%) | 1 (1.2%) | 25 (4.8%) | ||
| Han | 44 (44.0%) | 176 (50.4%) | 36 (44.4%) | 273 (52.0%) | ||
| Unknow | 51 (51.0%) | 157 (45.0%) | 44 (54.3%) | 227 (43.2%) | ||
| Previous live births | 0.96 | 0.43 | ||||
| NO | 89 (89.0%) | 310 (88.8%) | 67 (82.7%) | 455 (86.7%) | ||
| YES | 11 (11.0%) | 39 (11.2%) | 14 (17.3%) | 70 (13.3%) | ||
| Number of pregnancy loss | 0.77 | 0.93 | ||||
| 0 | 13 (13.0%) | 46 (13.2%) | 10 (12.3%) | 60 (11.4%) | ||
| 1 | 20 (20.0%) | 87 (24.9%) | 22 (27.2%) | 145 (27.6%) | ||
| 2 | 39 (39.0%) | 128 (36.7%) | 28 (34.6%) | 198 (37.7%) | ||
| ≥ 3 | 28 (28.0%) | 88 (25.2%) | 21 (25.9%) | 122 (23.2%) | ||
Bold values indicate significant differences between group comparisons, i.e., p-values less than 0.05
Values are median (inter-quartile range) or number of individuals (%)
EPL: early pregnancy loss; SP: successful pregnancy; BMI: body mass index; TC: total cholesterol; TG: triglyceride; LDL-C: low-density lipoprotein cholesterol; HDL-C: high-density lipoprotein cholesterol
a,bWomen who had lipid measurements at 4 and 7 weeks of gestation
As shown in Fig. 2, there were significant differences in TC and LDL-C levels between the early pregnancy loss group and the successful pregnancy group from weeks 3 to 12 of gestation.
Fig. 2.
Differences (A, C) and changing trajectory (B, D) in plasma TC and LDL-C concentrations between women with early pregnancy loss and successful pregnancy in the first trimester. TC, total cholesterol; LDL-C, low-density lipoprotein cholesterol
As depicted in Fig. 3 and Supplementary Table 1, compared to the SP control group, LDL-C concentrations at 4–9 weeks of gestation were linked to an increased risk of EPL. Additionally, TC levels at 4–7 and 9 weeks of gestation were associated with a higher risk of EPL. TG levels at 8 weeks of gestation were linked to a higher risk of EPL. HDL levels showed a negative association with EPL risk at 5 weeks but a positive association at 9 weeks.
Fig. 3.
Forest plot of association between early pregnancy loss and lipid concentrations measured at 4 to 10 weeks of gestation using successful pregnancy as a control group. EPL: early pregnancy loss; SP: successful pregnancy; CI: confidence interval; OR odds ratio; aOR: adjusted odds ratio; TC: total cholesterol; TG: triglyceride; HDL-C: high-density lipoprotein cholesterol; LDL-C: low-density lipoprotein cholesterol. aWomen who had lipid measurements at 4 to 10 weeks of gestation. bSuccessful pregnancy as the control group. cRaw odds ratio. dOdds ratios were adjusted by age, BMI, education, ethnicity, previous live birth, number of previous pregnancy loss
Figure 4 and Supplementary Table 2 indicate that, using live birth (LB) as a control group, LDL-C levels at 4–9 weeks of gestation were associated with a higher risk of EPL. Higher TC levels between 5 and 9 weeks and TG levels at 6 and 8 weeks of gestation were associated with increased EPL risk. Conversely, HDL-C levels at 5 weeks were linked to lower EPL risk, but at 9 weeks, they were linked to higher EPL risk.
Fig. 4.
Forest plot of association between early pregnancy loss and lipid levels measured at 4 to 10 weeks of gestation using live birth as a control group. EPL: early pregnancy loss; LB: live birth; CI: confidence interval; OR odds ratio; aOR: adjusted odds ratio; TC: total cholesterol; TG: triglyceride; HDL-C: high-density lipoprotein cholesterol; LDL-C: low-density lipoprotein cholesterol. aWomen who had lipid measurements at 4 to 10 weeks of gestation. bLive birth as the control group. cRaw odds ratio. dOdds ratios were adjusted by age, BMI, education, ethnicity, previous live birth, number of previous pregnancy loss
We investigated the link between plasma lipid concentrations and EPL by categorizing the lipid levels measured at 4 and 7 weeks of gestation into quartiles. Table 2 summarizes the risk of EPL associated with these plasma lipid levels. An increase of 1 mmol/L in TC levels at 4 and 7 weeks of pregnancy was associated with a 58% (aOR: 1.58, 95% CI 1.08, 2.30) and 105% (aOR: 2.05, 95% CI 1.42, 2.95) higher likelihood of EPL, respectively. At 4 weeks of gestation, statistically elevated OR was evident for women with plasma lipid levels measured during this period. Specifically, the covariate-adjusted OR was 2.18 (95% CI 1.14–4.21) when comparing the highest quartile (Q4) (> 3.78 mmol/L) of TC with the lowest quartile (Q1) (≤ 3.05 mmol/L). Similarly, at 7 weeks of gestation, the OR adjusted for covariates were 3.26 (95% CI 1.40–7.58) and 4.30 (95% CI 1.87, 9.93) when comparing the third (Q3) (3.25–3.75 mmol/L) and fourth (Q4) (> 3.75 mmol/L) quartiles of TC with the first quartile (Q1) (≤ 2.91 mmol/L). Consistent with these findings, results from RCS analysis further indicated that there may be a linear relationship between plasma TC levels at 4 and 7 weeks of pregnancy and the risk of EPL (Fig. 5A and E).
Table 2.
Risk of early pregnancy loss based on baseline lipid concentrations (mmol/L) tested at 4 weeks of gestation and 7 weeks of gestation
| Lipid categories | 4 weeks of gestationa | 7 weeks of gestationb | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Strata | EPL n, (%) |
SP n, (%) |
OR (95%CI) | aORc (95%CI) | Strata | EPL n, (%) |
SP n, (%) |
OR (95%CI) | aORc (95%CI) | |
| TC | 1.62 (1.14, 2.30) | 1.58 (1.08, 2.30) | 1.94 (1.38, 2.73) | 2.05 (1.42, 2.95) | ||||||
| Q1 (≤ 3.05) | 19 (19.0) | 90 (25.8) | 1.00 | 1.00 | Q1 (≤ 2.91) | 8 (9.9) | 132 (25.1) | 1.00 | 1.00 | |
| Q2 (3.05 ~) | 16 (16.0) | 86 (24.6) | 0.88 (0.43, 1.82) | 0.79 (0.37, 1.68) | Q2 (2.91 ~) | 14 (17.3) | 132 (25.1) | 1.75 (0.71, 4.31) | 1.70 (0.68, 4.25) | |
| Q3 (3.38 ~) | 24 (24.0) | 86 (24.6) | 1.32 (0.68, 2.58) | 1.23 (0.61, 2.48) | Q3 (3.25 ~) | 27 (33.3) | 135 (25.7) | 3.30 (1.45, 7.53) | 3.26 (1.40, 7.58) | |
| Q4 (> 3.78) | 41 (41.0) | 87 (24.9) | 2.23 (1.20, 4.14) | 2.18 (1.14, 4.21) | Q4 (> 3.75) | 32 (39.5) | 126 (24.0) | 4.19 (1.85, 9.44) | 4.30 (1.87, 9.93) | |
| p trend: < 0.01 | p trend: < 0.01 | p trend: < 0.01 | p trend: < 0.01 | |||||||
| TG | 1.10 (0.77, 1.57) | 1.08 (0.75, 1.57) | 1.30 (0.89, 1.90) | 1.26 (0.84, 1.89) | ||||||
| Q1 (≤ 0.65) | 90 (25.8) | 17 (17.0) | 1.00 | 1.00 | Q1 (≤ 0.88) | 15 (18.5) | 135 (25.8) | 1.00 | 1.00 | |
| Q2 (0.65 ~) | 88 (25.2) | 31 (31.0) | 1.86 (0.96, 3.61) | 1.71 (0.88, 3.37) | Q2 (0.88 ~) | 21 (25.9) | 130 (24.8) | 1.45 (0.72, 2.94) | 1.57 (0.76, 3.23) | |
| Q3 (0.89 ~) | 84 (24.1) | 28 (28.0) | 1.76 (0.90, 3.46) | 1.60 (0.80, 3.20) | Q3 (1.12 ~) | 16 (19.8) | 128 (24.4) | 1.12 (0.53, 2.37) | 1.90 (0.51, 2.34) | |
| Q4 (> 1.20) | 87 (24.9) | 24 (24.0) | 1.46 (0.73, 2.90) | 1.43 (0.70, 2.90) | Q4 (> 1.44) | 29 (35.8) | 131 (25.0) | 1.99 (1.02, 3.88) | 1.91 (0.95, 3.83) | |
| p trend: 0.39 | p trend: 0.44 | p trend: 0.07 | p trend: 0.14 | |||||||
| HDL-C | 1.21 (0.56, 2.61) | 1.09 (0.47, 2.50) | 1.35 (0.58, 3.13) | 1.23 (0.51, 2.98) | ||||||
| Q1 (≤ 1.11) | 25 (25.0) | 91 (26.1) | 1.00 | 1.00 | Q1 (≤ 1.10) | 22 (27.2) | 134 (25.6) | 1.00 | 1.00 | |
| Q2 (1.11 ~) | 18 (18.0) | 87 (25.0) | 0.75 (0.38, 1.48) | 0.59 (0.29, 1.19) | Q2 (1.10 ~) | 15 (18.5) | 129 (24.6) | 0.71 (0.35, 1.43) | 0.67 (0.33, 1.38) | |
| Q3 (1.28 ~) | 33 (33.0) | 83 (23.9) | 1.45 (0.80, 2.63) | 1.30 (0.69, 2.48) | Q3 (1.24 ~) | 19 (23.5) | 133 (25.4) | 0.87 (0.45, 1.68) | 0.83 (0.42, 1.63) | |
| Q4 (> 1.46) | 24 (24.0) | 87 (25.0) | 1.00(0.53, 1.89) | 0.91 (0.46. 1.78) | Q4 (> 1.43) | 25 (30.9) | 128 (24.4) | 1.19 (0.64, 2.22) | 1.05 (0.55, 2.00) | |
| p trend: 0.52 | p trend: 0.64 | p trend: 0.49 | p trend: 0.75 | |||||||
| LDL-C | 1.98 (1.29, 3.03) | 2.04 (1.29, 3.24) | 2.17 (1.43, 3.30) | 2.41 (1.55, 3.77) | ||||||
| Q1 (≤ 1.75) | 14 (14.0) | 91 (26.1) | 1.00 | 1.00 | Q1 (≤ 1.64) | 9 (11.1) | 143 (27.3) | 1.00 | 1.00 | |
| Q2 (1.75 ~) | 14 (14.0) | 83 (23.9) | 1.10 (0.49, 2.44) | 1.05 (0.46, 2.40) | Q2 (1.64 ~) | 16 (19.8) | 121 (23.1) | 2.10 (0.90, 4.92) | 1.98 (0.83, 4.70) | |
| Q3 (2.04 ~) | 38 (38.0) | 89 (25.6) | 2.78 (1.41, 5.47) | 2.98 (1.47, 6.08) | Q3 (1.92 ~) | 38 (46.9) | 199 (38.0) | 3.03 (1.42, 6.47) | 3.12 (1.44, 6.73) | |
| Q4 (> 2.39) | 34 (34.0) | 85 (24.4) | 2.60 (1.31, 5.18) | 2.66 (1.27, 5.55) | Q4 (> 2.25) | 18 (22.2) | 61 (11.6) | 4.68 (1.99, 11.02) | 5.17 (2.14, 12.49) | |
| p trend: < 0.01 | p trend: < 0.01 | p trend: < 0.01 | p trend: < 0.01 | |||||||
Bold values indicate that lipid levels were still significantly correlated with early pregnancy loss after adjusting for variables such as age, BMI, education, ethnicity, previous live birth, number of previous pregnancy loss, i.e., p-value less than 0.05
EPL: early pregnancy loss; SP: successful pregnancy; CI: confidence interval; OR: odds ratio; aOR: adjusted odds ratio; TC: total cholesterol; TG: triglyceride; HDL-C: high-density lipoprotein cholesterol; LDL-C: low-density lipoprotein cholesterol; TC: total cholesterol; TG: triglyceride; HDL-C: high-density lipoprotein cholesterol; LDL-C: low-density lipoprotein cholesterol.
a,bWomen who had lipid measurements at 4 and 7 weeks of gestation
cOdds ratios were adjusted by age, BMI, education, ethnicity, previous live birth, number of previous pregnancy loss
Fig. 5.
Spline analyses of plasma lipid concentrations and early pregnancy loss at 4 (A: TC; B: TG; C: HDL-C; D: LDL-C) and 7 (E: TC; F: TG; G: HDL-C; H: LDL-C) weeks of gestation. Logistic regression was adjusted by age, body mass index, education, ethnicity, previous live birth, number of previous pregnancy loss. Odds ratios are indicated by solid lines and 95% CIs by shaded areas
Table 2 also shows that each 1 mmol/L increase in plasma LDL-C levels measured at 4 and 7 weeks of gestation was associated with an 104% (aOR: 2.04, 95% Cl 1.29, 3.24) and 141% (aOR: 2.41, 95% Cl 1.55, 3.77) increase in the probability of EPL, respectively. Notably, statistically significant elevated OR was observed in women with plasma lipid levels measured at 4 weeks of gestation. Comparing Q3 (2.04–2.39 mmol/L) and Q4 (> 2.39 mmol/L) of LDL-C with Q1 (≤ 1.75 mmol/L) showed adjusted OR of 2.98 (95% CI 1.47–6.08) and 2.66 (95% CI 1.27–5.55), respectively. Similarly, at 7 weeks of gestation, comparing Q3 (1.92–2.25 mmol/L) and Q4 (> 2.25 mmol/L) with Q1 (≤ 1.64 mmol/L) resulted in adjusted OR of 3.12 (95% CI: 1.14–6.73) and 5.17 (95% CI 2.14–12.49), respectively. The RCS analysis further confirmed this finding that plasma LDL-C levels at 4 and 7 weeks of gestation may have a linear relationship with the risk of EPL (Fig. 5D and H). No significant association was found between plasma TG and HDL-C concentrations and the risk of EPL.
Supplementary Table 3. showed the subgroup analysis according to the number of previous pregnancy losses. For women with < 2 pregnancy losses, elevated TC (aOR = 2.66, 95% CI 1.28–5.51) and LDL-C (aOR = 3.03, 95% CI 1.29–7.09) levels at 4 weeks of gestation were linked to a higher EPL risk. The adjusted OR was 3.43 (95% CI 1.08–10.90) when comparing Q4 to Q1 of TC. Similarly, the raw OR was 4.27 (95% CI 1.13–16.12) when contrasting the Q4 of LDL-C with the Q1. Nevertheless, no significant relationship was observed between plasma TC and LDL-C levels measured at 7 weeks of pregnancy and EPL. Among women with ≥ 2 pregnancy losses, higher TC levels at 7 weeks showed a significant association with EPL, with Q3 (aOR: 4.37, 95% CI 1.34–14.32) and Q4 (aOR: 7.43, 95% CI 2.34–23.65) having notably higher aOR values compared to the lowest quartile. For LDL-C levels measured at 4 weeks, Q3 also exhibited a significant increase in adjusted OR compared to Q1. Similarly, for LDL-C levels measured at 7 weeks, adjusted ORs were significantly higher in Q4, Q3, and Q2 compared to the reference group.
Supplementary Table 4 and Table 5 shows sub-group analysis, examining the correlation between plasma lipid levels and EPL risk within different age and BMI categories. In both age groups (under and over 30), increased plasma TC and/or LDL-C levels measured at 4- and 7-weeks’ gestation were associated with a higher risk of EPL. However, significant correlations between EPL and TC or LDL-C were observed only in women with a BMI ≤ 24 kg/m2 and lipid measurements taken at 7 weeks, due to limited BMI data availability.
Discussion
In this study, patients experiencing EPL during the first trimester exhibited higher TC and LDL-C levels compared to those with successful pregnancies. Elevated TC and LDL-C levels between 4 and 9 weeks of pregnancy were linked to an increased risk of EPL. Subgroup analyses by previous pregnancy losses and age showed consistent results. The increased risk of EPL was particularly evident when TC levels exceeded 3.25–3.78 mmol/L and LDL-C levels surpassed 1.92–2.04 mmol/L.
Lipoproteins, composed of various cholesterol forms, TG, phospholipids, and apolipoproteins, are vital for lipid transport in the blood, energy use, lipid storage, steroid hormone production, and bile acid synthesis [14, 15]. Plasma LDL-C reflects the cholesterol content carried by LDL particles and provides an estimate of circulating LDL concentration [14]. Studies link LDL-C levels to atherosclerotic cardiovascular disease (ASCVD) risk, suggesting that reducing LDL-C can lower this risk [14]. However, understanding the relationship between cholesterol, LDL-C, and adverse pregnancy outcomes is limited. Studies from various countries have shown that high TC levels are linked to an increased risk of preterm birth [16–18]. A study suggested that second and third trimester pregnancy losses may share this underlying cause with preterm birth [19]. Additionally, women with elevated serum lipid levels undergoing vitro fertilization (IVF) or intracytoplasmic sperm injection (ICSI) face a higher risk of pregnancy loss between 12 to 28 weeks' gestation compared to controls [20]. High LDL-C and low HDL-C levels before conception are linked to pregnancy loss before 28 weeks [11]. Prior studies have examined how pre-pregnancy lipid levels affect pregnancy outcomes later in pregnancy.
Cohort studies have shown that early pregnancy dyslipidemia, marked by high TC, TG, LDL-C, and/or low HDL-C, is associated with a higher risk of preeclampsia, preterm birth, large-for-gestational-age infants, and gestational diabetes [7, 9, 21]. However, evidence on the effect of gestational cholesterol levels on EPL is sparse. One study reported no link between TC and LDL-C levels at 4 and 8 weeks of gestation and pregnancy loss before 14 weeks [19]. Conversely, our study found a positive association between TC and LDL-C levels measured between 4 and 9 weeks and EPL, but only when these levels surpassed specific thresholds within the normal range. Although LDL-C is a known risk factor for ASCVD, the ideal LDL-C level for safety and efficacy is unclear [22]. Our study indicated that TC levels over 3.78 mmol/L at 4 weeks and 3.25 mmol/L at 7 weeks of gestation, along with LDL-C levels over 2.04 mmol/L at 4 weeks and 1.92 mmol/L at 7 weeks, may raise the risk of EPL. This highlighted the importance of identifying the optimal maternal blood cholesterol range during pregnancy for proper embryo development.
While there is limited data on the threshold or optimal range of cholesterol linked to EPL, research indicates that in the U.S., both low (< 4.12 mmol/L) and high (≥ 6.76 mmol/L) TC levels are linked to preterm birth in white mothers [23]. In China, TC levels of ≥ 5.20 mmol/L are associated with lower live birth rates in infertile women undergoing IVF/ICSI [20]. An HDL-C level of ≤ 1.995 mmol/L is optimal for detecting pregnancy-induced hypertension, while an LDL-C level of ≥ 3.425 mmol/L is best for identifying macrosomia [24]. Pre-pregnancy or early pregnancy HDL-C levels negatively correlate with pregnancy loss risk [19, 25], and slightly elevated TG levels in the first eight weeks are linked to a higher risk of pregnancy loss before the 14th week [26]. Our study found that high TC levels at certain gestational weeks were linked to a higher risk of EPL. Unlike previous research, we observed that the relationship between HDL-C levels and EPL risk varied by gestational week, showing both positive and negative correlations.
Our study has several strengths. It is the first to explore the relationship between plasma lipids and EPL at various gestational weeks, yielding consistent results. Previous studies that aggregated lipid data from different weeks might not be reliable due to fluctuations in lipid levels during pregnancy. Additionally, all women with pregnancy loss had blood lipid tests conducted prior to the diagnosis of pregnancy loss, rather than afterward. However, the study also has limitations. Lipid profiles were not assessed before pregnancy or in the first 1–2 weeks of gestation. Despite adjusting for confounding factors, there may still be unidentified confounders. Being a single-center retrospective study, it is also susceptible to recall bias. Future multicenter prospective cohort studies are needed to validate these findings.
In conclusion, our study suggests that elevated TC and LDL-C levels in early pregnancy may heighten the risk of EPL. Therefore, monitoring and managing cholesterol levels during this critical period could potentially mitigate this risk.
Supplementary Information
Below is the link to the electronic supplementary material.
Acknowledgements
We express our profound gratitude to all participants, researchers and staff of The Second Hospital of Lanzhou University involved in the Study.
Author contributions
Wei Zhang: conceptualization, data curation, analysis and interpretation of data, software, project administration, writing—original draft, writing—review and Editing. Ruifang Wang: conceptualization, data curation, project administration. Xin Yang: data curation, project administration. Zhiyuan Cheng: conceptualization, analysis and interpretation of data, writing—review and editing. Fang Wang: conceptualization, methodology, resources, data curation, supervision, project administration, and funding acquisition. Prof. Fang Wang had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. The final version for submission was approved by all coauthors.
Funding
This work was funded by Science Foundation of Lanzhou University Second Hospital (Grant No.CY2018-MS12) and Medical Innovation and Development Project of Lanzhou University (Grant No. lzuyxcx-2022-137).
Data availability
All available data are collected and stored. The raw data involved in this study are not publicly available due to the protection of privacy of all enrolled participants.
Declarations
Conflict of interests
The authors declare that there is no potential conflict of interest with respect to the research, authorship, and/or publication in this paper.
Footnotes
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Contributor Information
Zhiyuan Cheng, Email: chengzy@sustech.edu.cn.
Fang Wang, Email: ery_fwang@lzu.edu.cn.
References
- 1.Bender Atik R, Christiansen OB, Elson J et al (2018) ESHRE guideline: recurrent pregnancy loss. Hum Reprod Open 2:hoy004. 10.1093/hropen/hoy004 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Practice Committee of the American Society for Reproductive Medicine (2012) Evaluation and treatment of recurrent pregnancy loss: a committee opinion. Fertil Steril 98(5):1103–1111. 10.1016/j.fertnstert.2012.06.048 [DOI] [PubMed] [Google Scholar]
- 3.Neill S (2023) Management of early pregnancy loss. JAMA 329(16):1399–1400. 10.1001/jama.2023.0933 [DOI] [PubMed] [Google Scholar]
- 4.Ammon Avalos L, Galindo C, Li DK (2012) A systematic review to calculate background miscarriage rates using life table analysis. Birth Defects Res A Clin Mol Teratol 94(6):417–423. 10.1002/bdra.23014 [DOI] [PubMed] [Google Scholar]
- 5.Dimitriadis E, Menkhorst E, Saito S et al (2020) Recurrent pregnancy loss. Nat Rev Dis Prim 6(1):98. 10.1038/s41572-020-00228-z [DOI] [PubMed] [Google Scholar]
- 6.Provinciatto H, Moreira MVB, Neves GR et al (2024) Levothyroxine for subclinical hypothyroidism during pregnancy: an updated systematic review and meta-analysis of randomized controlled trials. Arch Gynecol Obstet 309(6):2387–2393. 10.1007/s00404-024-07512-3 [DOI] [PubMed] [Google Scholar]
- 7.Vrijkotte TG, Krukziener N, Hutten BA et al (2012) Maternal lipid profile during early pregnancy and pregnancy complications and outcomes: the ABCD study. J Clin Endocrinol Metab 97(11):3917–3925. 10.1210/jc.2012-1295 [DOI] [PubMed] [Google Scholar]
- 8.Catov JM, Ness RB, Wellons MF et al (2010) Prepregnancy lipids related to preterm birth risk: the coronary artery risk development in young adults study. J Clin Endocrinol Metab 95(8):3711–3718. 10.1210/jc.2009-2028 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Enquobahrie DA, Williams MA, Butler CL et al (2004) Maternal plasma lipid concentrations in early pregnancy and risk of preeclampsia. Am J Hypertens 17(7):574–581. 10.1016/j.amjhyper.2004.03.666 [DOI] [PubMed] [Google Scholar]
- 10.Cai WY, Luo X, Ma HL et al (2022) Association between preconception serum lipid concentrations and treatment outcomes in women with PCOS who underwent ovulation induction. Reprod Biomed Online 45(4):805–814. 10.1016/j.rbmo.2022.04.013 [DOI] [PubMed] [Google Scholar]
- 11.Cai WY, Luo X, Chen E et al (2021) Serum lipid levels and treatment outcomes in women undergoing assisted reproduction: a retrospective cohort study. Front Endocrinol 12:633766. 10.3389/fendo.2021.633766 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Bottomley C, Bourne T (2009) Diagnosing miscarriage. Best Pract Res Clin Obstet Gynaecol 23(4):463–477. 10.1016/j.bpobgyn.2009.02.004 [DOI] [PubMed] [Google Scholar]
- 13.Jun-Ren ZH, Run-Lin GA, Shui-Ping ZH, Guo-Ping LU, Dong ZH, Jian-Jun LI (2018) 2016 Chinese guidelines for the management of dyslipidemia in adults. J Geriatr Cardiol 15(1):1–29. 10.11909/j.issn.1671-5411.2018.01.011 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Mach F, Baigent C, Catapano AL et al (2020) 2019 ESC/EAS Guidelines for the management of dyslipidaemias: lipid modification to reduce cardiovascular risk. Eur Heart J 41(1):111–188. 10.1093/eurheartj/ehz455 [DOI] [PubMed] [Google Scholar]
- 15.von Eckardstein A, Nordestgaard BG, Remaley AT et al (2023) High-density lipoprotein revisited: biological functions and clinical relevance. Eur Heart J 44(16):1394–1407. 10.1093/eurheartj/ehac605 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Alleman BW, Smith AR, Byers HM et al (2013) A proposed method to predict preterm birth using clinical data, standard maternal serum screening, and cholesterol. Am J Obstet Gynecol 208(6):472.e1–11. 10.1016/j.ajog.2013.03.005 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Hao Y, Pang Y, Yan H et al (2019) Association of maternal serum copper during early pregnancy with the risk of spontaneous preterm birth: a nested case-control study in China. Environ Int 122:237–243. 10.1016/j.envint.2018.11.009 [DOI] [PubMed] [Google Scholar]
- 18.Catov JM, Bodnar LM, Kip KE et al (2007) Early pregnancy lipid concentrations and spontaneous preterm birth. Am J Obstet Gynecol 197(6):610.e1–7. 10.1016/j.ajog.2007.04.024 [DOI] [PubMed] [Google Scholar]
- 19.Grantz KL, Elmi A, Pugh SJ et al (2020) Maternal serum lipid trajectories and association with pregnancy loss and length of gestation. Am J Perinatol 37(9):914–923. 10.1055/s-0039-1689000 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Yang T, Zhao J, Zhang Q et al (2021) Associations between dyslipidaemia and pregnancy outcomes in the first complete cycle of IVF/ICSI: a real-world analysis. Reprod Biomed Online 43(6):1095–1105. 10.1016/j.rbmo.2021.08.030 [DOI] [PubMed] [Google Scholar]
- 21.Wang C, Zhu W, Wei Y et al (2017) The associations between early pregnancy lipid profiles and pregnancy outcomes. J Perinatol 37(2):127–133. 10.1038/jp.2016.191 [DOI] [PubMed] [Google Scholar]
- 22.Liu H, Li J, Liu F et al (2023) Efficacy and safety of low levels of low-density lipoprotein cholesterol: trans-ancestry linear and non-linear Mendelian randomization analyses. Eur J Prev Cardiol. 10.1093/eurjpc/zwad111 [DOI] [PubMed] [Google Scholar]
- 23.Edison RJ, Berg K, Remaley A et al (2007) Adverse birth outcome among mothers with low serum cholesterol. Pediatrics 120(4):723–733. 10.1542/peds.2006-1939 [DOI] [PubMed] [Google Scholar]
- 24.Wu Q, Zhang L, Huang L et al (2019) Second-trimester maternal lipid profiles predict pregnancy complications in an age-dependent manner. Arch Gynecol Obstet 299(5):1253–1260. 10.1007/s00404-019-05094-z [DOI] [PubMed] [Google Scholar]
- 25.Cai WY, Luo X, Chen E et al (2021) Serum lipid levels and treatment outcomes in women undergoing assisted reproduction: a retrospective cohort study. Front Endocrinol. 10.3389/fendo.2021.633766 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Laughon SK, McLain AC, Sundaram R et al (2014) Maternal lipid change in relation to length of gestation: a prospective cohort study with preconception enrollment of women. Gynecol Obstet Invest 77(1):6–13. 10.1159/000355100 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
All available data are collected and stored. The raw data involved in this study are not publicly available due to the protection of privacy of all enrolled participants.