Diminished ovarian reserve is associated with metabolic disturbances and hyperhomocysteinemia in women with infertility

Abstract Background To distinguish the metabolic profile between women with diminished ovarian reserve (DOR) and those with normal ovarian reserve (NOR). Methods In this retrospective study, we enrolled 524 women under the age of 40 who were experiencing infertility: 261 in the DOR group and 263 in the NOR group. Physical characteristics and metabolic parameters were compared between these two groups. Results Women with DOR exhibited a higher propensity for elevated parameters including body mass index (BMI), systolic blood pressure (SBP), diastolic blood pressure (DBP), as well as heightened serum levels of homocysteine (Hcy), triglycerides (TG), low-density lipoprotein (LDL), and triglyceride-glucose (TyG) index, while concurrently experiencing reduced serum levels of high-density lipoprotein (HDL) (P < 0.05). Furthermore, the incidence rates of TG ≥ 1.7 mmol/L, hyperhomocysteinemia (HHcy), BMI ≥ 25 kg/m2, SBP/DBP ≥ 130/85 mmHg, and metabolic syndrome (MS) were significantly elevated within the DOR group as compared to the NOR group (P < 0.05). Conclusion The prevalence of metabolic disturbances and HHcy were notably elevated in women with infertility and DOR compared to those with NOR. PLAIN LANGUAGE SUMMARY This study focused on the metabolic condition of women who had difficulty getting pregnant and had a decreased ovarian reserve. The findings indicated that these women had a higher likelihood of glucose and lipid metabolic disorders and elevated serum homocysteine levels compared to those with a normal ovarian reserve. These metabolic issues and elevated serum homocysteine levels were associated with an increased risk of cardiovascular disease.


Introduction
Diminished ovarian reserve (DOR) is a condition characterised by a decrease in the quantity of oocytes, as evidenced clinically by reduced serum anti-M€ ullerian hormone (AMH) levels, a diminished antral follicle count (AFC), elevated basal follicle-stimulating hormone (bFSH) levels, and declined fertility (Mutlu andErdem 2012, ASRM 2020).Various studies have reported the prevalence of DOR to range from 10% to 26% among women of reproductive age, with an increasing incidence observed within the assisted reproductive technology (ART) population (Levi et al. 2001, Devine et al. 2015).
It is widely acknowledged that postmenopausal women and those with premature ovarian insufficiency (POI) have a higher propensity for weight gain and hypertension.They often exhibit an unfavourable lipid and/or glucose metabolic profile, increasing their vulnerability to cardiovascular disease (CVD) due to hormonal fluctuations (Ates et al. 2014, Newson 2018, Honigberg et al. 2019, El Khoudary 2020, Huang et al. 2021).Oestrogen deficiency is a critical factor contributing to these metabolic disturbances (Christ et al. 2018).DOR represents an early stage of ovarian ageing, occurring prior to menopause or POI.Most women with DOR continue to experience regular menstrual cycles and ovulations (ASRM 2020), suggesting that oestrogen production may not decrease significantly.As ovarian ageing is a continuous process, and it is known that atherosclerotic lesions can develop before menopause (McMahan et al. 2005), concerns have arisen regarding the metabolic profiles and the CVD risk among women with DOR.While a few studies have garnered widespread attention by revealing metabolic issues in women with DOR (Bleil et al. 2013, Tehrani et al. 2014, Verit et al. 2014, Verit et al. 2016), the precise CVD risk in this population remains elusive, with some other reports yielding contradictory results, both in human research (Yang et al. 2022) and animal model (Appt et al. 2012).
On the other hand, oxidative stress is an important mechanism of ovarian ageing (L.Wang et al. 2021, Yan et al. 2022), with numerous factors associated with oxidative stress, including hyperhomocysteinemia (HHcy) (Perna et al. 2003, Herrmann andHerrmann 2022).HHcy is also recognised as an independent risk factor for CVD (Ganguly andAlam 2015, Djuric et al. 2018), with even mild or moderate increases in blood homocysteine (Hcy) levels linked to an elevated cardiovascular risk.However, there is a lack of data regarding Hcy levels in patients with DOR.
In the present study, our objective was to investigate the metabolic profile, encompassing serum lipids, fasting blood glucose (FG), and Hcy levels, between women with DOR and the NOR controls.

Study design and patients
In this retrospective study, a total of 524 women with infertility who underwent in vitro fertilisation (IVF), intracytoplasmic sperm injection (ICSI), or intra-uterine insemination (IUI) cycles from January 2018 to August 2023 were recruited from the Centre for Reproductive Medicine of Shaoxing Women and Children's Hospital in Shaoxing, China.Institutional review board approval and written consent were obtained from all patients before data collection.The inclusion criteria for women with DOR at our centre were as follows: (1) AMH � 1.1 ng/ml, and/or (2) total antral follicle count (AFC) � 7, and/or (3)10 � bFSH� 25 IU/l.Exclusion criteria included: (1) POI, (2) a history of ovarian surgery within the past year, (3) previously diagnosed endocrine system diseases such as diabetes mellitus and thyroid disease, (4) a history of malignancy, (5) severe systematic or major organ disorders, (6) receiving hormonal therapy within 6 months of their initial visit or taking medications that could interfere with metabolic conditions.This study was conducted in accordance with the standard approved by the Ethics Committee of Shaoxing Women and Children's Hospital.
Sample size calculations were based on two independent proportion hypothesis tests with preliminary analysis, which resulted in at least 135 patients in each group.
In total, 2966 patients underwent IVF/ICSI/IUI cycles from January 2018 to August 2023 at our centre and 335 women were diagnosed with DOR.Seventy-four cases were excluded according to the exclusion criteria.Finally, 261 eligible cases were recruited for the DOR group.To enhance data comparability and control potential selection bias, 263 women from a pool of 1859 eligible cases with regular menstrual cycles and normal ovarian reserve during the same period were randomly selected for the control group.The flow chart of the study is presented in Figure 1.

Procedures
All the patients underwent a comprehensive standardised screening panel conducted by trained medical personnel before IVF/ICSI/IUI, including age, duration of infertility, smoking habits, gravidity, parity, age at menarche, menstrual cycle situation, height, weight, waist circumference, blood pressure (BP), blood test, and ultrasound scans.Waist circumference was measured as the smallest circumference at the level of the umbilicus.BPs were measured twice after a 20-30 minutes rest in a seated position, and the lower measurements were recorded.BMI was calculated as weight (kg)/height (m)2 .All data were recorded in our centre's assisted reproductive system software and retrospectively collected by professionally trained medical personnel.
All of the patients followed an unrestricted diet, and serum endocrine and metabolic parameters were measured during the first visit before IVF/ICSI/IUI.Blood samples were taken between 8:00 and 11:00 after a 12-hour overnight fast, and sex hormone levels were measured during the early follicular phase.Blood samples were centrifuged within 2 hours of collection and assessed on the same day.AFC was counted during the early follicular phase by a trained doctor using transvaginal ultrasound.
HHcy was defined as plasma Hcy levels � 10.0 lmol/L and metabolic parameters were stratified according to Chinese guidelines (China Joint Society 2016, Chinese Diabetes Society 2018, Bone Health and Nutrition Branch of Chinese Nutrition Society 2020).The triglyceride-glucose (TyG) index was calculated as follows: TyG index ¼ ln [Fasting triglyceride (mg/dl) � fasting glucose (mg/dl)/2], as previously reported (Tao et al. 2022).

Statistical analysis
All data are presented as mean ± SD for continuous variables unless for categorical data.Statistical analyses were carried out using the Statistical Package for the Social Sciences version (SPSS) 25.0 (SPSS Inc., Chicago, Illinois, USA).The normality of distributions of all continuous variables was assessed using the Kolmogorov-Smirnov test.Independent-sample ttests were used for normal distribution data, while the twosample Wilcoxon test was used for non-normal distributed data.Pearson's chi-squared test and Fisher's exact test were used to analyse differences in categorical data and to calculate odds ratios (OR).Binary logistic regression was applied to further adjust for potential confounding factors, mainly age, infertility duration, and BMI, in categorical data, and to obtain adjusted ORs.Additionally, multivariate logistic regression analysis was used to analyse the risk factors for metabolic disturbances and HHcy, variables including age, duration of infertility, smoking habits, gravidity, parity, AFC, AMH, bFSH, bLH, the FSH/LH ratio, TT, and bE2, were taken into consider.A P-value <0.05 was considered statistically significant.

Results
The demographic, clinical, and hormonal characteristics of the patients in the DOR and NOR groups are summarised in Table 1.In the DOR group, we observed significantly higher values for age, duration of infertility, BMI, SBP, DBP, bFSH, FSH/LH ratio, and basal E 2 (bE 2 ), as well as significantly lower levels of basic LH (bLH), AMH, and AFC compared to the NOR group (P < 0.05).No statistically significant differences were found between the two groups for PRL, TT and TSH (P > 0.05).
Table 2 presents the metabolic parameters, showing significantly higher levels of serum Hcy, TG, LDL, and the TyG index, as well as significantly lower levels of serum HDL in DOR group compared to the NOR group (P < 0.05).No statistical differences were observed between the two groups for FG, TC, Apo-A, and Apo B (P > 0.05).
Table 4 reveals the risk factors for metabolic disturbances and HHcy.For TG � 1.7 mmol/L, age and high BMI were

Discussion
In the present study, we observed elevated levels of serum Hcy, TG, and LDL, along with decreased levels of serum HDL in women with infertility and DOR compared to the control group.Additionally, the DOR group exhibited a tendency towards higher BMI, SBP, and DBP.Furthermore, we noted a significantly higher prevalence of metabolic disturbances in the DOR group, including elevated rates of TG � 1.7 mmol/L, HHcy, BMI � 25 kg/m 2 , SBP/DBP � 130/85 mmHg, and MS.
Emerging evidence suggests that women with infertility and DOR may face an increased risk of CVD, primarily attributed to a higher prevalence of obesity, hypertension, and disturbances in glucose and lipidmetabolism.Obesity, recognised as a pro-inflammatory state, is associated with an elevated risk of cardiovascular disorders (Battineni et al. 2021).Inflammation triggered by obesity contributes to a low-grade inflammatory condition, resulting in various metabolic deregulations, including heightened insulin resistance and endothelial dysfunction, which can precipitate CVD (Battineni et al. 2021).Our finding revealed that the BMI of the DOR group and prevalence of BMI � 25 kg/m 2 were significantly higher compared to the control group.This observation aligns with the report by Bleil, et al. (2013).Logistic regression analysis identified ageing, low AMH, low bLH, and a high FSH/LH ratio as risk factors.Notably, low AMH, low bLH and a high FSH/LH ratio are the indicators of DOR (Noci et al. 1998, Barroso et al. 2001, Fanchin et al. 2003).Thus, alterations in hormone levels may contribute to obesity in individuals with DOR.
Hyperlipemia, a well-established risk factor for atherosclerosis, is one of the most perilous contributors to CVD (Karr 2017).Hypertriglyceridaemia is a fairly common lipid metabolic disorder and often associated with an increased risk of CVD (Chait 2022).Genetic cause and environmental factors, such as ageing, obesity, high-fat diet and alcohol consumption, are key contributors to hypertriglyceridaemia (Simha 2020).Our study observed significantly higher serum TG levels and greater prevalence of TG � 1.7 mmol/L in the DOR group.While ageing and BMI were the primary risk factors for TG � 1.7 mmol/L, the difference remains statistically significant even after adjusting for age and BMI.Additionally, serum LDL and HDL levels differed significantly between the two groups, with the DOR group displaying higher LDL and lower HDL levels.These lipid metabolic disturbances were also observed in previous studies by Verit et al. (2014Verit et al. ( , 2016)).
Studies have suggested that insulin resistance may also play a significant role in vascular endothelial dysfunction and CVD (Cui et al. 2022).In our study, we observed a dramatic elevation in the TyG index, which is another proven and reliable tool for assessing insulin resistance, besides the homeostasis model assessment ratio (HOMA-IR) (Tao et al. 2022), in the DOR group.This result was also consistent with those of Verit et al. (2014Verit et al. ( , 2016)).
Hypertension, among the risk factors for CVD, boasts the strongest evidence of causation (Fuchs and Whelton 2020).Our study revealed significantly higher SBP and DBP levels, along with a greater prevalence of SBP/DBP � 130/85 mmHg in the DOR group.These findings corroborate data presented by Bleil et al. (2013).Furthermore, logistic regression analysis demonstrated that higher BMI was the primary risk factor, suggesting that the elevated prevalence of obesity in the DOR group might contribute to the higher blood pressure.
Metabolic syndrome is a chronic non-infective syndrome characterised by a series of vascular risk factors, including insulin resistance, hypertension, abdominal obesity, impaired glucose metabolism, and dyslipidaemia (Silveira Rossi et al. 2022).Prior research has shown dramatically lower ovarian reserve in patients with MS (Balkan et al. 2014) compared to normal controls.Our study also found a significantly higher prevalence of MS in the DOR group (OR ¼ 3.131,95%CI: 1.817-5.394,P < 0.001).Logistic regression analysis revealed that low AMH and low bLH were independent risk factors, suggesting that hormonal changes in individuals with DOR may contribute to the development of MS.
Additionally, we observed higher serum Hcy levels and a greater prevalence of HHcy in women with DOR.Hcy, an intermediate product in the normal biosynthesis of the amino-acids methionine and cysteine, is another well-established risk factor for CVD and stroke (Perna et al. 2003, Herrmann andHerrmann 2022).Currently, little is known about the mechanisms linking Hcy metabolism and ovarian function.A previous study reported increased serum Hcy concentrations in Chinese premature ovarian failure patients, which correlated with serum oestradiol/FSH concentrations (Hou et al. 2016).Another report suggested age as the primary factor contributing to high Hcy levels in DOR (H.Wang et al. 2022).we identified high BMI and a longer duration of infertility as the independent risk factors.Given the high prevalence of obesity in individuals with DOR and recent findings indicating a correlation between obesity and HHcy (Xiang et al. 2021, Yoo et al. 2022), it is plausible that elevated BMI contributes to higher Hcy levels in individuals with DOR.Additionally, daily dietary habits and folic acid supplementation can influence Hcy levels (Kaye et al. 2020).Clinically, we have observed that patients with long duration of infertility were more likely to discontinue folic acid supplementation, resulting in higher Hcy levels.
Our study has several limitations.First, it is a retrospective study with a limited sample size, preventing us from establishing definitive causal relationships.Moreover, other potential covariates, such as dietary patterns and physical activity, were not included in the analysis.
To reiterate, a growing body of research underscores that DOR is not solely indicative of ovarian ageing; it is also accompanied by a range of metabolic issues.Our study contributes valuable supplementary evidence to this body of research, particularly in elucidating the relationship between DOR and elevated homocysteine levels (HHcy).Nevertheless, robust, large-scale, well-designed studies are warranted to further corroborate our findings.

Table 1 .
Demographic, clinical and endocrine characteristics of the groups.Data are presented as mean ± SD, t-test.Bold values represent the significance of P values.

Table 2 .
Metabolic parameters of the groups.

Table 3 .
Comparison of metabolic abnormalities between the two groups.

Table 4 .
Logistic repression for risk factors of metabolic disturbances and HHcy.