A retrospective study of the relationship between low-density lipoprotein cholesterol and pregnancy outcomes after assisted reproductive technology treatments in women with diabetes mellitus and infertility.

Chun Yuan: Conceptualization, Data curation, Funding acquisition, Investigation, Writing – original draft, Writing – review & editing. Yuan Guo: Data curation. Jing Wang: Data curation, Methodology, Supervision. Jinghui Guo: Supervision. Jinyong Liu: Supervision.

This study was supported by the Chronic Disease Management Research Project of the National Health Commission Capacity Building and Continuing Education Center ( GWJJMB202510010079 ), Key Grant of Jiangsu Provincial Maternal and Child Health Research Project ( F202304 ) and Natural Science Foundation of Jiangsu Province ( BK20191076 ).

This study included 474 women with T1/2D who underwent IVF/ICSI treatment ( Figure 1 ). After PSM analysis, 416 women were divided into 2 groups based on ovarian stimulation protocols (GnRH antagonist group, n=262; early follicular phase long-acting GnRH agonist group, n=154). To compare pregnancy outcomes and other parameters, patients were divided into high- and low-LDL groups ( Figure 1 ). In the GnRH antagonist protocol group, there were no significant differences in baseline characteristics (age, AMH, BMI, glucose, ALT, AST) between the high- and low-LDL-C groups; however, total cholesterol and triglyceride levels were significantly higher in the high-LDL-C group ( Table 1 ). TABLE 1 Baseline/treatment characteristics of patients with diabetes and infertility in this study (GnRH antagonist protocol) TABLE 1 dummy alt text High LDL-C (>3mmol/L, n=142) Low LDL-C (<3mmol/L, n=120) X 2 /F P Age (year) 32.55±0.42 31.82±0.41 1.250 0.219 AMH (ng/ml) 6.18±0.43 6.59±0.43 1.027 0.520 BMI (Kg/M 2 ) 25.01±0.34 25.04±0.25 1.545 0.952 Glucose(mmol/L) 6.23±0.21 5.83±0.21 1.336 0.170 ALT (U/L) 26.87±1.71 24.40±1.93 1.080 0.339 AST (U/L) 23.84±0.92 22.19±1.00 1.013 0.227 Triglycerides (mmol/L) 1.82±0.09 1.43±0.09 1.162 0.005* Total cholesterol(mmol/L) 5.61±0.06 4.16±0.05 1.682 <0.0001*** HDL-C (mmol/L) 1.30±0.02 1.24±0.03 1.074 0.123 IVF/ICSI Treatment characteristics Gn dosage (IU) 2016±59.21 2003±72.01 1.259 0.891 Duration of stimulation (Days) 9.34±0.16 9.31±0.17 1.004 0.893 No of oocytes retrieved 9.57±0.50 8.93±0.48 1.003 0.361 Oocyte retrieval rate (%) 0.74±0.02 0.76±0.01 1.316 0.269 IVF Fertilization Rate (%) 0.71±0.02 0.70±0.02 1.153 0.819 IVF good-quality rate (%) 4.31±0.36 4.57±0.43 1.158 0.643 IVF Blastocyst Formation Rate (%) 0.58±0.03 0.57±0.04 1.304 0.803 IVF Good-quality blastocyst rate (%) 0.50±0.05 0.47±0.05 1.508 0.663 ICSI Fertilization Rate (%) 0.87±0.03 0.85±0.03 1.197 0.706 ICSI good-quality embryo rate (%) 0.67±0.05 0.77±0.04 1.874 0.114 ICSI Blastocyst Formation Rate (%) 0.41±0.11 0.31±0.05 3.525 0.334 ICSI Good-quality blastocyst rate (%) 0.30±0.06 0.43±0.07 1.689 0.220 Notes: BMI, body mass index; AMH anti-Müllerian hormone; ALT, Alanine aminotransferase; AST, Aspartate aminotransferase; GnRH, gonadotropin-releasing hormone; Gn, gonadotropin; IVF, in vitro fertilization; ICSI, intracytoplasmic sperm injection; * P <0.05, ⁎⁎ P <0.001, ⁎⁎⁎ P <0.0001, Bold values. Yuan. A retrospective study of the relationship between low-density lipoprotein cholesterol and pregnancy outcomes after assisted reproductive technology treatments in women with diabetes mellitus and infertility. Am J Obstet Gynecol 2026. Baseline/treatment characteristics of patients with diabetes and infertility in this study (GnRH antagonist protocol) Notes: BMI, body mass index; AMH anti-Müllerian hormone; ALT, Alanine aminotransferase; AST, Aspartate aminotransferase; GnRH, gonadotropin-releasing hormone; Gn, gonadotropin; IVF, in vitro fertilization; ICSI, intracytoplasmic sperm injection; * P <0.05, ⁎⁎ P <0.001, ⁎⁎⁎ P .05; TABLE 1 , TABLE 2 , Figure 2 ). However, the rate of early miscarriage in GnRH antagonist protocol-FET cycles had a significantly positive association with LDL-C level (OR=21.727, 95% CI, [3.90–121.04], P =.0004, Figure 3 , Supplemental Table 1 ). TABLE 2 Pregnant outcomes in different groups (GnRH antagonist protocol) TABLE 2 dummy alt text Fresh embryo transfer cycles (Cleavage-stage embryos) High LDL-C (n=46) Low LDL-C(n=33) X 2 P Implantation rate (%) 19 (33.3) 9(21.4) 1.690 0.260 Clinical pregnant rate (%) 19(41.3) 10(30.3) 3.569 0.059 Early miscarriage rate (%) 9(19.5) 4(12.1) 0.775 0.541 Live birth Rate (%) 10(21.7) 6(18.1) 0.151 0.782 Frozen embryo transfer cycles (all blastocysts) High LDL-C (n=42) Low LDL-C(n=31) X 2 P Implantation rate (%) 28(49.1%) 24(52.1%) 2.755 0.134 Clinical pregnant rate (%) 28(52.8%) 24(53.3%) 0.002 0.561 Early miscarriage rate (%) 9(17.3%) 2(4.4%) 4.121 0.041* Live birth Rate (%) 19(35.8%) 20(44.4%) 0.750 0.255 Cumulative clinical pregnant rate (%) 49(49.4) 37(46.2) 0.187 0.764 Cumulative live birth rate (%) 30(30.3%) 28(35.0%) 0.142 0.751 Notes: *P<0.05, **P<0.001, ***P<0.0001, Bold values. Yuan. A retrospective study of the relationship between low-density lipoprotein cholesterol and pregnancy outcomes after assisted reproductive technology treatments in women with diabetes mellitus and infertility. Am J Obstet Gynecol 2026. Figure 2 Cumulative live birth rates in different groups (GnRH antagonist protocol). Figure 2 dummy alt text Yuan. A retrospective study of the relationship between low-density lipoprotein cholesterol and pregnancy outcomes after assisted reproductive technology treatments in women with diabetes mellitus and infertility. Am J Obstet Gynecol 2026. Figure 3 Regression curve of the relationship of LDL-C -and early miscarriage rate. Figure 3 dummy alt text Yuan. A retrospective study of the relationship between low-density lipoprotein cholesterol and pregnancy outcomes after assisted reproductive technology treatments in women with diabetes mellitus and infertility. Am J Obstet Gynecol 2026. Pregnant outcomes in different groups (GnRH antagonist protocol) Notes: *P<0.05, **P<0.001, ***P<0.0001, Bold values. Cumulative live birth rates in different groups (GnRH antagonist protocol). Regression curve of the relationship of LDL-C -and early miscarriage rate. In the long-acting GnRH agonist protocol group, there were no significant differences in baseline characteristics (age, AMH, BMI, glucose) between the high- and low-LDL-C groups; however, total cholesterol, triglycerides, and AST/ALT levels were significantly higher in the high-LDL-C group ( Table 3 ). Moreover, aside from the ICSI fertilization rate, most ART parameters did not differ significantly between the 2 groups ( Table 3 ). It is noteworthy that, although the clinical pregnancy rate in the high-LDL-C group was considerably higher than that of the low-LDL-C group, there was no statistical difference in live birth or cumulative live birth rates between the 2 groups. ( Table 4 , Figure 4 ). TABLE 3 Baseline/treatment characteristics of patients with type 2 diabetes and infertility ( Early follicular phase long-acting GnRH agonist protocol ) TABLE 3 dummy alt text High LDL-C(n=85) Low LDL-C(n=69) F P Age (year) 33.98±0.44 32.83±0.55 1.249 0.102 AMH (ng/ml) 3.53±0.34 3.32±0.24 2.239 0.622 BMI (Kg/M 2 ) 23.92±0.31 23.55±0.35 1.075 0.438 Glucose(mmol/L) 5.93±0.25 5.92±0.30 1.219 0.170 ALT (U/L) 23.11±1.71 17.59±1.56 1.413 0.019* AST (U/L) 22.22±0.97 19.38±0.75 2.009 0.032* Triglycerides (mmol/L) 2.56±0.62 1.20±0.08 74.84 0.049* Total cholesterol(mmol/L) 5.56±0.07 4.20±0.07 1.382 <0.0001 ⁎⁎⁎ HDL (mmol/L) 1.32±0.03 1.28±0.04 1.525 0.369 Gn dosage (U) 2629±99.02 2646±133.9 1.538 0.914 Gn Days 10.95±0.30 11.30±0.28 1.307 0.405 No of oocytes 8.34±0.49 8.83±0.55 1.082 0.497 Oocyte retrieval rate (%) 0.74±0.02 0.78±0.02 1.691 0.123 IVF Fertilization Rate (%) 0.71±0.04 0.77±0.04 1.139 0.258 IVF good-quality rate (%) 0.72±0.05 0.67±0.04 0.745 0.458 IVF Blastocyst Formation Rate (%) 0.53±0.06 0.57±0.04 1.304 0.803 IVF Good-quality blastocyst rate (%) 0.50±0.05 0.47±0.05 1.508 0.663 ICSI Fertilization Rate (%) 0.82±0.02 0.70±0.05 2.106 0.039* ICSI good-quality embryo rate (%) 0.78±0.04 0.75±0.05 0.380 0.704 ICSI Blastocyst Formation Rate (%) 0.37±0.04 0.45±0.05 1.118 0.268 ICSI Good-quality blastocyst rate (%) 0.38±0.06 0.54±0.08 1.538 0.131 Notes: *P<0.05, **P<0.001, ***P<0.0001, Bold values. Yuan. A retrospective study of the relationship between low-density lipoprotein cholesterol and pregnancy outcomes after assisted reproductive technology treatments in women with diabetes mellitus and infertility. Am J Obstet Gynecol 2026. TABLE 4 Pregnant outcomes in different groups (Early follicular phase long-acting GnRH agonist protocol) TABLE 4 dummy alt text Fresh embryo transfer cycles (Cleavage-stage embryos) High LDL-C (n=39) Low LDL-C (n=40) X 2 P Implantation rate (%) 34(65.4%) 17(34.6%) 4.297 0.038* Clinical pregnant rate (%) 27(69.2%) 16(40%) 15.340 0.001** Early abortion rate (%) 7(17.9%) 1(2.5%) 5.178 0.029* Live birth Rate (%) 10(25.6%) 12(30%) 0.032 0.859 Frozen embryo transfer cycles (blastocysts) High LDL-C (n=27) Low LDL-C (n=18) Implantation rate (%) 12(44.4%) 7(38.9%) 0.134 0.715 Clinical pregnant rate (%) 12(46.1%) 7(38.8%) 0.224 0.636 Early abortion rate (%) 5(19.2%) 0 3.905 0.048* Live birth Rate (%) 7(26.9%) 6(33.3%) 0.205 0.651 Cumulative clinical pregnant rate (%) 49(59.7%) 24(40%) 5.414 0.027* Cumulative live birth rate (%) 23(28%) 17(28.3%) 0.001 0.970 Notes: *P<0.05, **P<0.001, ***P<0.0001, Bold values. Yuan. A retrospective study of the relationship between low-density lipoprotein cholesterol and pregnancy outcomes after assisted reproductive technology treatments in women with diabetes mellitus and infertility. Am J Obstet Gynecol 2026. Figure 4 Cumulative live birth rates in different groups (Early follicular phase long-acting GnRH agonist protocol). Figure 4 dummy alt text Yuan. A retrospective study of the relationship between low-density lipoprotein cholesterol and pregnancy outcomes after assisted reproductive technology treatments in women with diabetes mellitus and infertility. Am J Obstet Gynecol 2026. Baseline/treatment characteristics of patients with type 2 diabetes and infertility ( Early follicular phase long-acting GnRH agonist protocol ) Notes: *P<0.05, **P<0.001, ***P<0.0001, Bold values. Pregnant outcomes in different groups (Early follicular phase long-acting GnRH agonist protocol) Notes: *P<0.05, **P<0.001, ***P<0.0001, Bold values. Cumulative live birth rates in different groups (Early follicular phase long-acting GnRH agonist protocol).

Female patients aged 21 to 45 years who were diagnosed with either Type 1 or Type 2 diabetes at least 2 years prior, between January 1, 2018, and July 30, 2024. All patients underwent in vitro fertilization (IVF) or Intracytoplasmic Sperm Injection (ICSI) treatment at the Reproductive Medicine Clinical Center, First Affiliated Hospital of Nanjing Medical University. A flowchart of the study protocol is shown in Figure 1 . Figure 1 Flowchart for individuals included. AMH, anti-Müllerian hormone; BMI, body mass index; GnRH, gonadotropin-releasing hormone; LDL-C, low-density lipoprotein cholesterol; PSM, propensity score matching. Figure 1 dummy alt text Yuan. A retrospective study of the relationship between low-density lipoprotein cholesterol and pregnancy outcomes after assisted reproductive technology treatments in women with diabetes mellitus and infertility. Am J Obstet Gynecol 2026 . Flowchart for individuals included. AMH, anti-Müllerian hormone; BMI, body mass index; GnRH, gonadotropin-releasing hormone; LDL-C, low-density lipoprotein cholesterol; PSM, propensity score matching. The patients were divided into 2 groups: the high LDL-C group (≥3 mmol/L) and the low LDL-C group (<3 mmol/L). In this retrospective cohort study, we collected baseline characteristics and IVF/ICSI data from patients and analyzed pregnancy outcomes. According to the protocol of ovarian stimulation, the clinical characteristics in IVF/ICSI cycles and pregnancy outcomes were compared separately. We employed propensity score matching (PSM) using a 1:1 nearest-neighbor algorithm to create matched pairs of the high LDL-C group and the low LDL-C group across different IVF/ICSI protocols. Standardized mean differences <0.1 were considered indicative of a well-balanced matching process, ensuring comparability between the 2 groups. Categorical variables were presented as numbers and percentages and compared using the χ 2 test. Continuous variables were reported as the mean and standard deviation or as the median and interquartile range. The unpaired Student's t -test or the Mann–Whitney U test was used to compare continuous variables between 2 groups, depending on the data distribution. This study received approval from the ethics committee of the First Affiliated Hospital of Nanjing Medical University (2018-SR-063).

This study uses clinical data from ART treatments to examine the relationship between lipid levels and pregnancy outcomes in IVF/ICSI therapy. The findings indicate a significant link between abnormal lipid levels and a higher risk of early miscarriage in infertile women with diabetes mellitus after ART procedures. Reproductively aged women with diabetes mellitus are commonly associated with reproductive dysfunction, which is often neglected. According to a comprehensive population-based study in Sweden, women with diabetes had a significantly lower chance of fertilization compared to women without diabetes mellitus. Furthermore, women with diabetes might take longer to conceive and face a greater risk of negative pregnancy outcomes. 9 Moreover, the presence of obesity, polycystic ovary syndrome (PCOS), and abnormal ovulation could compound the issue of subfertility in women with diabetes. Therefore, for women with diabetes, a healthy pregnancy and live birth are especially important. According to many recent studies, obesity and abnormal glucose levels increase the risk of miscarriage, low embryo implantation rates, hypertensive disorders during pregnancy, preterm deliveries, and abnormal offspring weight. 10 Moreover, several lifestyle factors are associated with ART outcomes. In IVF/ICSI treatments, multiple studies have demonstrated a significant decrease in pregnancy rates related to smoking. 11 Moderate alcohol consumption, moderate caffeine intake, and bisphenol A (BPA) exposure all reduce IVF pregnancy rates, 12 , 13 while healthy Mediterranean diets and exercise were found to have inconsistent effects on IVF outcomes. 14 , 15 However, few clinical investigations have reported the relationship between lipid metabolism and ART treatments in women with diabetes. Dyslipidemia is typically characterized by elevated LDL-C and/or triglycerides, concurrent with decreased HDL-C. 16 Alterations in gestational lipid levels are linked to miscarriage, pregnancy-related complications, gestational diabetes, and fetal and neonatal morbidities. 17 However, few studies examine the connection between serum lipid disturbances and reproductive outcomes following ART procedures. In recent studies, dyslipidemia negatively impacts ART outcomes, regardless of PCOS and obesity. 18 However, there are few clinical reports focusing on how dyslipidemia affects ART outcomes in women with diabetes. In clinical labs, it is noteworthy that the normal ranges of blood lipids are too broad to overlook dyslipidemia in diabetic women before ART treatments. In general, preexisting chronic diseases are one of several pathological conditions associated with infertility that may require ART treatments. 18 Women with diabetes, hypertension, or other diseases should have a physical exam before IVF/ICSI treatments to prevent maternal and neonatal complications. 19 , 20 , 21 Our study analyzes pregnancy outcomes of various ovarian stimulation protocols in women with diabetes mellitus to prevent bias. In cycles of the GnRH antagonist protocol, most women are advised to undergo frozen embryo transfer to prevent ovarian hyperstimulation syndrome (OHSS). Women with poor or normal ovarian response are recommended to use the long-acting GnRH agonist protocol. 22 For women with endometriosis or adenomyosis, if a fresh embryo transfer is planned, a long-acting GnRH agonist protocol might be beneficial. 23 Our data showed that diabetic women with high LDL-C have a higher risk of miscarriage during ART treatments (cycles of fresh/frozen embryo transfer), indicating that dyslipidemia is a risk factor for pregnancy loss. The results are due to links between dyslipidemia and inflammation, poor oocyte/embryo quality, 24 and low blastocyst scores. 25 Other studies showed that LDL-C had the strongest negative correlation with pregnancy outcomes after ART treatments, such as lower pregnancy rates, lower live birth rates, and higher pregnancy loss rates. 26 On the contrary, we observed a higher rate of clinical pregnancy and implantation associated with increased LDL-C in fresh embryo transfer cycles of the long-acting GnRH agonist protocol. It remains to be determined whether embryo quality or endometrial environment is responsible for these results. This is the first study to explore how LDL-C is essential in physical examinations before ART treatments in infertile women with diabetes mellitus, providing significant evidence for the need for diabetic women with dyslipidemia to use blood lipid-lowering treatments before IVF/ICSI-ET. Importantly, we demonstrated that the standard clinical range of LDL-C (2.6–4.1 mmol/L) does not provide sufficient information to guide treatment decisions, as it fails to predict which patients are at high risk of pregnancy loss. In the analysis evaluating dyslipidemia and pregnancy outcomes together, the rate of cumulative clinical pregnancy in the high LDL-C group was temporarily increased. Although there was no significant difference in the cumulative live birth rate, our data showed that the GnRH antagonist protocol was more effective than the long-acting GnRH agonist protocol in women with diabetes and infertility, due to the shorter duration and the higher cumulative live birth rate. This study demonstrates that in infertile women with diabetes mellitus undergoing ART, elevated pretreatment LDL-C is a significant and modifiable marker associated with an increased risk of early miscarriage and a reduced live birth rate. The impact of LDL-C varies by ovarian stimulation protocol. Specifically, within FET cycles using the GnRH antagonist protocol, the risk of early miscarriage shows a direct positive correlation with LDL-C levels. Considering comprehensive pregnancy outcomes, the GnRH antagonist protocol appears more favorable than the long-acting GnRH agonist protocol for this patient population. This study has some limitations. Firstly, due to the wide variety of baseline characteristics among women with diabetes mellitus, the number of patients included was reduced after PSM analysis. Secondly, the decision to use fresh or frozen embryo transfer in this study was based on the ovarian stimulation protocol and the risk of OHSS, which results in an imbalance between fresh and frozen embryo transfer cycles. In conclusion, infertile women with diabetes mellitus and elevated lipid levels before ART treatment are at increased risk for early miscarriage and have a lower live birth rate. Women with diabetes mellitus should have their blood lipids tested and pretreatment before ART, just as they do with blood glucose levels. These findings underscore the importance of integrating preconception lipid management into the clinical pathway and adopting a protocol-specific risk assessment to optimize pregnancy outcomes.

Diabetes mellitus is a more complex systemic metabolic disorder, characterized by chronic hyperglycemia, dyslipidemia, and an increased risk of vascular and metabolic complications. 1 With increasing maternal age at childbirth and the rising prevalence of metabolic diseases among women of reproductive age, a growing proportion of women with diabetes are seeking assisted reproductive technology (ART) treatment. 2 In this metabolic context, women with diabetes often face higher clinical risk during ART, with evidence suggesting that type 2 diabetes is associated with impaired ovarian reserve and adverse pregnancy outcomes during ART compared with nondiabetic women. 2 However, existing studies in this population have predominantly focused on pregnancy complications and maternal–fetal outcomes, whereas evidence specifically addressing ART treatment processes and related reproductive outcomes remains limited. 2 , 3 In patients with diabetes, low-density lipoprotein cholesterol (LDL-C) is a central metabolic marker in cardiovascular risk management and has traditionally been used to guide prevention of cardiovascular disease. 4 Elevated LDL-C and dyslipidemia have also been associated with adverse reproductive and fertility outcomes, including lower pregnancy and live birth rates in couples undergoing fertility treatment. 5 , 6 , 7 , 8 Nevertheless, in the field of reproductive medicine, LDL-C has rarely been considered a key determinant of reproductive outcomes, and its potential role in ART treatment processes and associated reproductive outcomes has yet to be systematically evaluated. The present study aimed to investigate women with diabetes undergoing ART and to systematically assess the association between LDL-C levels and ART outcomes, with particular focus on pregnancy and live birth outcomes.

The authors report no conflict of interest.