Association between serum uric acid to high-density lipoprotein cholesterol ratio and female infertility in women aged 18 to 45 years: A study based on NHANES 2013 to 2018.

In summary, this nationally representative cross-sectional study suggests that an elevated UHR is associated with a higher prevalence of infertility among reproductive-aged women. Further prospective and mechanistic studies are needed to confirm this association and explore underlying pathways.

Infertility, defined as the inability to conceive after at least 1 year of regular unprotected intercourse, affects approximately 10% to 15% of couples worldwide, posing significant public health concerns and psychological distress for affected individuals, particularly women. [ 1 ] The etiology of infertility is multifactorial, involving complex interactions among genetic, metabolic, hormonal, and environmental factors. [ 2 ] Identifying modifiable risk factors and biomarkers for early detection and intervention remains a critical goal in reproductive medicine. [ 3 ] Accumulating evidence indicates that systemic or local inflammation and immune responses are among the most critical factors underlying unexplained infertility. [ 4 ] Immune cells responsible for maintaining endocrine homeostasis, including macrophages, natural killer (NK) cells, dendritic cells, and T cells, as well as immunoregulatory molecules such as interleukin (IL)-6, IL-10, tumor necrosis factor-alpha (TNF-α), and transforming growth factor-beta 1 (TGF-β1), exhibit abnormal activity in women with infertility-associated disorders. [ 5 , 6 ] Previous studies have demonstrated that the counts of lymphocytes, proportions of CD4 + T lymphocytes, and NK cells are significantly elevated in women diagnosed with polycystic ovary syndrome (PCOS), and that CD4 + T cells and NK cells serve as independent risk factors for PCOS, [ 7 ] suggesting that peripheral inflammatory immune cells may represent promising predictive factors for infertility in these patients. Uric acid (UA), traditionally associated with gout and renal disease, has recently been recognized as a marker of oxidative stress, inflammation, and metabolic disturbances, which may impair female fertility. [ 8 – 11 ] Several studies have highlighted the role of elevated UA in conditions like polycystic ovary syndrome (PCOS) and insulin resistance, [ 12 , 13 ] which are known to affect reproductive outcomes. High-density lipoprotein cholesterol (HDL-C) plays a crucial role in modulating inflammation and oxidative stress, [ 14 , 15 ] with lower levels of HDL-C being linked to infertility. [ 16 , 17 ] The UA to HDL-C ratio (UHR), an emerging biomarker initially proposed to assess cardiometabolic risk and inflammation, integrates these 2 key metabolic parameters into a single indicator. [ 18 , 19 ] Recent studies suggest that elevated UHR may effectively reflect systemic inflammation, oxidative stress, and insulin resistance status, potentially impacting reproductive health adversely. [ 20 , 21 ] Despite growing interest in UHR, no studies to date have specifically examined the association between UHR and female infertility. Previous clinical research has demonstrated that UHR is associated with increased risk of metabolic syndrome, insulin resistance, hypertension, and cardiovascular diseases in various populations, sometimes outperforming UA or HDL-C alone as a predictor of subclinical inflammation and metabolic dysregulation. [ 20 – 22 ] These findings provide a biological rationale for exploring its potential relevance to female reproductive health. While both elevated UA and reduced HDL-C are individually associated with reproductive disorders, the combination of these 2 factors into a single metric, the UHR, may provide a more comprehensive reflection of the balance between oxidative stress and inflammation, as well as antioxidant protection. UHR captures not only the pro-inflammatory and oxidative burden (through elevated UA) but also the counteracting anti-inflammatory and antioxidant properties of HDL-C. This dual nature may make UHR a more informative biomarker for infertility risk, as it reflects both the metabolic and inflammatory disturbances that contribute to reproductive dysfunction. Utilizing data from the National Health and Nutrition Examination Survey (NHANES) 2013 to 2018, this study aims to explore the association between UHR and infertility among reproductive-aged women. The findings could provide new insights into potential metabolic biomarkers for infertility and highlight novel avenues for early screening and targeted interventions in reproductive healthcare. Given the cross-sectional nature of the NHANES dataset, this study aims to examine the association rather than infer causality.

Conceptualization: Heng Zhang. Data curation: Heng Zhang. Formal analysis: Heng Zhang. Funding acquisition: Heng Zhang. Investigation: Heng Zhang. Methodology: Heng Zhang. Project administration: Heng Zhang. Resources: Heng Zhang. Software: Heng Zhang. Supervision: Heng Zhang. Validation: Heng Zhang. Visualization: Heng Zhang. Writing – original draft: Heng Zhang. Writing – review & editing: Heng Zhang.

This cross-sectional analysis utilized data from the NHANES, conducted by the National Center for Health Statistics (NCHS). NHANES is a comprehensive survey designed to collect representative information on the health and nutritional status of the civilian, noninstitutionalized U.S. population, including demographics, socioeconomic status, dietary habits, and health-related conditions. To ensure sample diversity and representativeness, NHANES employs a complex, stratified, multistage probability sampling method to select participants nationwide. The NHANES study protocol was approved by the CDC NCHS Ethics Review Board, and all participants provided written informed consent. NHANES data are publicly accessible at https://www.cdc.gov/nchs/nhanes/ . This analysis specifically focused on NHANES data from the 2013 to 2018 cycles. Initially, 29,400 participants were included in the NHANES dataset. The exclusion criteria were as follows: Male participants (n = 14,452) were excluded because the study focused on female infertility. Individuals aged younger than 18 years (n = 5630) and those older than 45 years (n = 4995) were excluded to focus on women of reproductive age (18–45 years). Women with missing infertility data (n = 656) were excluded to ensure complete outcome information. Women who had undergone bilateral oophorectomy (n = 41) or hysterectomy (n = 100) were excluded, as these conditions would render them incapable of natural conception. Participants with missing data for key variables, including UA (n = 8), HDL-C (n = 199), or other important covariates (n = 872), were excluded to ensure the integrity and accuracy of the analysis. Missing data for laboratory measures (UA and HDL-C) and covariates were handled using listwise deletion, meaning that participants with missing data on any key variable were excluded from the corresponding analysis. This approach was used to ensure the integrity and consistency of the data included in the final analysis After applying these exclusion criteria, 2447 eligible female participants aged 18–45 years were included in the final analysis. The detailed sample selection process is illustrated in Figure 1 . Flowchart of participant election from NHANES 2013 to 2018. NHANES = National Health and Nutrition Examination Survey. The exposure variable, UA to HDL-cholesterol ratio (UHR), was calculated using fasting blood sample measurements from the NHANES 2013 to 2018 database, which provided data for serum UA and HDL-C. In NHANES, HDL-C levels were measured by direct immunoassay or precipitation methods. Serum UA concentrations were assessed using a timed endpoint enzymatic method. Specifically, UA was measured by the DxC800 automated chemistry analyzer, which quantifies UA concentrations by monitoring the absorbance change of the chromogenic product generated from the reaction between hydrogen peroxide – produced by the oxidation of UA by uricase – and 4-aminoantipyrine (4-AAP) catalyzed by 3,5-dichloro-2-hydroxybenzene sulfonate (DCHBS). The primary outcome – infertility – was assessed based on self-reported responses from the reproductive health questionnaire of the NHANES dataset. Women were classified as infertile if they responded affirmatively to either of the following questions: RHQ074, asking whether they had ever unsuccessfully tried to conceive for at least 12 months; or RHQ076, asking whether they had ever sought medical assistance due to difficulty conceiving. Participants who answered negatively to both questions were classified as fertile and served as the reference group. It should be acknowledged that this classification, which relies on self-report rather than clinical diagnosis, may encompass a broader spectrum of fertility difficulties, including those not formally evaluated or diagnosed. Women who answered negatively to both questions may include those who have never attempted to conceive, and thus may not truly be at risk for infertility. This potential misclassification is an inherent limitation of using self-reported questionnaire data. This definition aligns with previous epidemiological studies utilizing NHANES data to explore infertility prevalence and its associated risk factors. [ 23 , 24 ] To account for potential confounding, several relevant covariates were included in the analysis. Demographic characteristics encompassed age, body mass index (BMI), race/ethnicity, educational attainment, marital status, and poverty-income ratio (PIR). Behavioral and lifestyle variables included smoking status, alcohol intake, and physical activity levels. Alcohol consumption was categorized into 5 groups: lifetime abstainers (fewer than 12 alcoholic beverages consumed over a lifetime), former drinkers (≥12 drinks per year in the past but none in the previous 12 months), light drinkers (<2 drinks per day), moderate drinkers (≥2 drinks per day), and heavy drinkers (≥3 drinks per day). Smoking status was classified as never-smokers (<100 cigarettes during lifetime), former smokers (≥100 cigarettes lifetime but currently not smoking), and current smokers (≥100 cigarettes lifetime and actively smoking during the survey period). Physical activity levels were derived from self-reported leisure-time activity, classified according to intensity as vigorous activity (causing large increases in heart rate or breathing for at least 10 minutes) or moderate activity (causing small yet noticeable increases in heart rate or breathing for at least 10 minutes). Clinical comorbidities, including diabetes mellitus (DM) and hypertension, were also adjusted for. DM was categorized as present, borderline, or absent. Participants were considered diabetic if they had a self-reported diagnosis, were currently using insulin or oral hypoglycemic medication, or exhibited fasting plasma glucose ≥126 mg/dL or a 2-hour oral glucose tolerance test plasma glucose ≥200 mg/dL. Hypertension was defined by a self-reported medical diagnosis, current antihypertensive treatment, or measured systolic/diastolic blood pressure ≥140/90 mm Hg during examination at the mobile examination center (MEC). Several reproductive and gynecological variables known to affect fertility were also included as covariates, such as age at menarche, menstrual cycle regularity, use of hormone therapy, contraceptive use, and history of pelvic infection. Statistical analyses were performed using EmpowerStats software ( www.empowerstats.com ; X&Y Solutions Inc., Boston) and R software version 4.0.5 ( http://www.R-project.org ; The R Foundation for Statistical Computing). Given the complex, multistage sampling design of NHANES, all analyses incorporated the appropriate fasting subsample weights (WTSAF2YR) for each cycle, with combined weights adjusted for the pooled 6-year cycles following NCHS recommendations. We applied these weights in descriptive analyses, univariate tests, and multivariable logistic regression models using survey-weighted procedures. Participants were divided into 2 groups based on infertility status. Weighted Student’s t tests were used to compare continuous variables, and weighted chi-square tests were applied for categorical variables. Continuous variables were expressed as weighted means ± standard error, while categorical variables were presented as weighted percentages. To examine the relationship between UHR and female infertility, 3 multivariate logistic regression models were constructed. Model 1 was unadjusted; Model 2 was adjusted for sociodemographic variables, including age, race/ethnicity, educational level, marital status, and PIR; Model 3 was further adjusted for behavioral and clinical variables, including smoking status, alcohol use, physical activity, DM, hypertension, and reproductive health variables. Generalized additive models with smooth curve fitting were utilized to assess potential nonlinear associations between UHR and infertility risk. Statistical significance was determined using a 2-sided P -value threshold of <.05. Subgroup analyses were conducted to further explore whether the association between UHR and infertility differed across key demographic and clinical categories, including age, BMI, race/ethnicity, education level, PIR, marital status, smoking status, DM, hypertension, regular menstrual periods, and history of pelvic infection. Multivariate logistic regression models adjusted for corresponding covariates were used for analyses within each subgroup.

Table 1 summarizes characteristics of 2447 women (329 infertile, 2118 fertile). Infertile women were significantly older (34.92 ± 0.59 vs 31.60 ± 0.23 years, P  <.0001) and had higher BMI, serum UA, and log-transformed UHR (all P  <.01). They were also more likely to be ≥35 years old, obese (BMI ≥30 kg/m²), living with a partner, and had higher prevalence of hypertension, pelvic infections, and birth control pill use ( P  <.05). Other characteristics showed no significant differences between groups. Baseline characteristics of participants according to infertility status. BMI = body mass index, DM = dmellitus, HDL-C = high-density lipoprotein cholesterol, PIR = poverty-income ratio, UA = uric acid. Table 2 showed multivariate logistic regression analyses evaluating the association between log-transformed UHR and infertility risk. In Model 1 (unadjusted), higher log (UHR) was significantly associated with increased infertility risk. After adjustment for sociodemographic variables (Model 2), the association remained significant and became slightly stronger. The fully adjusted model (Model 3) showed a significant relationship (OR = 2.04, 95% CI: 1.28–3.27, P  = .01). Multivariable logistic regression analysis of the association between log (UHR) and female infertility. Model 1 no covariates were adjusted. Model 2 adjusts age, race/ethnicity, marital status, education level, and PIR. Model 3 further adjusts for clinical factors (alcohol intake, smoking status, physical activity, history of hypertension or DM, age of first menstrual, menstrual status, pelvic infection, ever use of female hormones, and birth control pills). BMI = body mass index, CI = confidence interval, DM = diabetes mellitus, HDL-C = high-density lipoprotein cholesterol, OR = odds ratio, PIR = poverty-income ratio, UA = uric acid, UHR = the uric acid to HDL-cholesterol ratio. When log (UHR) was analyzed by quartiles, a positive dose-response relationship was observed across all models. In the fully adjusted model, compared with the lowest quartile (Q1, reference), the odds ratios (ORs) for infertility risk increased progressively: Q2 (OR = 1.72, 95% CI: 1.09–2.72, P  = .02), Q3 (OR = 1.91, 95% CI: 1.17–3.12, P  = .01), and Q4 (OR = 1.98, 95% CI: 1.15–3.41, P  = .02), with a significant linear trend ( P for trend = .017). Figure 2 showed smooth curve fitting results illustrating the relationship between log-transformed UHR and infertility risk. In the unadjusted model (Fig. 2 A), infertility risk increased notably with rising log (UHR). After adjusting for sociodemographic, behavioral, clinical, and reproductive covariates (fully adjusted Model 3, Fig. 2 B), the positive association remained significant, confirming a stable dose-response relationship between increased UHR levels and higher infertility risk. Smooth curve fitting of the association between log (UHR) and the prevalence of infertility. (A) Unadjusted model; (B) fully adjusted model. UHR = the uric acid to HDL-cholesterol ratio. Figure 3 and Table 3 present subgroup analyses assessing potential effect modification by several key factors. In the analysis of UHR as a continuous variable (Fig. 3 ), stronger associations were observed among women younger than 35 years, those living with a partner, nonsmokers, women with regular menstrual periods, no history of DM, no hypertension, and those with or without pelvic infections (both significantly elevated risk; all P  <.05). Interaction tests showed no statistically significant differences between subgroups. Subgroup analysis of UHR quartiles and risk of infertility in reproductive-aged women. Model 1 no covariates were adjusted. Model 2 adjusts age, race/ethnicity, marital status, education level, and PIR. Model 3 further adjusts for clinical factors (alcohol intake, smoking status, physical activity, history of hypertension or DM, age of first menstrual, menstrual status, pelvic infection, ever use of female hormones, and birth control pills). BMI = body mass index, CI = confidence interval, DM = diabetes mellitus, HDL-C = high-density lipoprotein cholesterol, OR = odds ratio, PIR = poverty-income ratio, UA = uric acid, UHR = the uric acid to HDL-cholesterol ratio. Subgroup analyses of the association between log (UHR) and infertility risk across stratified variables. UHR = the uric acid to HDL-cholesterol ratio. Subgroup analyses (Table 3 ) revealed significant associations between higher UHR quartiles and infertility risk in specific groups. Women aged ≥35 years showed elevated risks in Q2 (OR = 2.27; 95% CI: 1.22–4.20), 3 (OR = 2.57; 95% CI: 1.18–5.59), and 4 (OR = 2.16; 95% CI: 1.06–4.39). Significant associations were also observed among those living with a partner (Q2–Q4 ORs ranging from 1.85–2.71), never-smokers (ORs = 2.17–3.12), women without DM (ORs = 1.79–2.17), women without hypertension (ORs = 1.83–2.27), those reporting regular menstrual periods (ORs = 1.94–2.31), and notably higher among those with a history of pelvic infection (ORs = 2.81–4.64). Subgroup analyses indicated a suggestive association between elevated UHR and infertility risk in overweight women (OR = 2.33, 95% CI: 0.44–5.34). However, the wide confidence interval, which is close to null, suggests considerable uncertainty in this result. Therefore, while the association appears to be positive, the findings should be interpreted with caution, and further studies are needed to validate these observations. All reported associations were statistically significant ( P  <.05). No significant interactions were observed across subgroups.

In this cross-sectional analysis based on NHANES 2013 to 2018 data, we observed that higher levels of the UHR were significantly associated with increased infertility risk among reproductive-aged women. This association remained robust after adjusting for multiple demographics, clinical, behavioral, and reproductive covariates. Furthermore, a clear dose-response relationship was identified, with infertility risk progressively increasing across higher quartiles of UHR. Subgroup analyses confirmed the consistency of these associations, particularly among women aged ≥35 years, those living with partners, never-smokers, and individuals with regular menstrual cycles or a history of pelvic infections. Chronic inflammation and immune dysregulation are increasingly recognized as central contributors to multiple etiologies of female infertility, including PCOS, endometriosis, pelvic inflammatory disease, and unexplained infertility. A delicate immunological balance between fetal tolerance and infection defense is essential for successful implantation and pregnancy maintenance. Disruptions in immune cell populations or function – such as altered levels of T cells, NK cells, and dendritic cells – have been implicated in adverse reproductive outcomes, including implantation failure, recurrent miscarriage, and preterm birth. [ 25 ] Despite the crucial role of inflammation, current diagnostic approaches for infertility often involve complex, costly, and invasive procedures that may cause physical or psychological discomfort. A recent study based on NHANES data suggested an inverse association between systemic inflammatory markers and female fertility, highlighting the clinical relevance of inflammation-related biomarkers in reproductive assessment. [ 26 ] Several prior studies have explored the relationship between UA and fertility outcomes, albeit with varying results. For example, elevated UA levels have been associated with conditions like PCOS and insulin resistance, both of which are linked to infertility. [ 12 , 13 ] In addition to inflammation, lipid metabolism also plays a critical role in female reproductive health. Dyslipidemia – particularly low levels of HDL-C – has been associated with hormonal imbalance, oxidative stress, and impaired ovarian function, and is frequently observed in women with PCOS and other metabolic disorders. [ 17 ] Abnormal lipid profiles may interfere with follicular development, endometrial receptivity, and oocyte quality, ultimately reducing fertility potential. [ 27 ] Together, these lines of evidence underscore the importance of both inflammatory and metabolic pathways in the pathophysiology of female infertility. Given the intertwined roles of inflammation and lipid metabolism in reproductive dysfunction, the UHR emerges as a promising integrated biomarker reflecting both pro-inflammatory and metabolic stress states. Unlike traditional single-parameter indicators, UHR captures the imbalance between elevated UA – a marker of oxidative stress and systemic inflammation – and reduced HDL-C, which is known for its anti-inflammatory and antioxidant properties. [ 28 , 29 ] Previous studies have linked UHR to CVD, insulin resistance, and metabolic syndrome, suggesting its broad utility in detecting subclinical inflammatory and metabolic disturbances. [ 22 , 30 , 31 ] However, to our knowledge, this is the first population-based study to demonstrate a significant association between elevated UHR levels and infertility risk in women, highlighting its potential clinical relevance in reproductive health screening and risk stratification. In addition to UHR, other inflammation-based ratios such as the neutrophil-to-lymphocyte ratio (NLR) and platelet-to-lymphocyte ratio (PLR) have also been investigated as potential markers of systemic inflammation and oxidative stress, [ 32 ] which may affect reproductive outcomes. Unlike NLR and PLR, which primarily reflect immune cell balance and innate immune activation, UHR integrates metabolic and inflammatory pathways by combining a marker of oxidative stress and metabolic dysfunction (serum UA) with a marker of anti-inflammatory and antioxidant defense (HDL-C). This dual nature may provide complementary information to purely hematologic ratios. While prior studies have linked elevated NLR and PLR with adverse fertility outcomes, direct comparisons with UHR in this context remain limited. Our study contributes to this area by highlighting that different composite indices may capture distinct aspects of systemic inflammation and metabolic stress. Future studies should consider including multiple ratios simultaneously to better understand their relative and combined predictive value for infertility risk. Both serum UA and HDL-C are individually relevant to oxidative stress, inflammation, and metabolic dysregulation, which are important pathways potentially influencing female fertility. The UHR combines these opposing effects into a single index that reflects the net balance between pro-inflammatory/oxidative burden and anti-inflammatory/antioxidant defenses. Prior studies have suggested that this ratio may have stronger predictive value for cardiometabolic risk and subclinical inflammation than either component alone. Unlike classic inflammatory markers such as CRP, UHR can be readily calculated from routine biochemistry panels, providing a cost-effective and widely accessible surrogate for integrated metabolic-inflammation risk. Nevertheless, our findings indicate that UHR may serve as an accessible and convenient biomarker worth further investigation for its potential role in reproductive health research. Prospective validation studies are needed to confirm this association, clarify its causal relevance, and explore whether monitoring UHR could help identify women who might benefit from targeted metabolic or lifestyle interventions to improve fertility outcomes. This study has several limitations that should be considered when interpreting the results. First, the cross-sectional design of NHANES precludes any inference of causality. As a result, we cannot establish the temporal relationship between elevated UHR and infertility. Infertility itself could potentially affect metabolic markers, such as UA and HDL-C, through mechanisms like stress, hormonal changes, or the use of fertility treatments, which may lead to reverse causation. Second, infertility status was assessed through self-reported questionnaire items in NHANES, which may introduce recall bias or misclassification. Specifically, classifying women who responded “no” to all infertility-related questions as non-infertile may result in non-differential misclassification, as it does not account for undiagnosed infertility or subfertility in women who may not recognize or report reproductive difficulties. Such misclassification could attenuate the observed associations. Third, although we adjusted for a broad range of potential confounders, residual confounding cannot be ruled out. Important biological mediators – such as systemic inflammatory markers (e.g., C-reactive protein, interleukins), oxidative stress biomarkers, or reproductive hormones (e.g., AMH, FSH) – were not included in the analysis due to data limitations. These factors may influence both UHR and fertility status and could further elucidate underlying mechanisms. Moreover, we could not adjust for possible lifestyle or medical interventions related to infertility treatment that might affect UA and HDL-C levels, and thus UHR. Such residual confounding may influence the observed associations. Prospective studies incorporating treatment details and repeated biomarker measurements are needed to address this potential bias. Detailed data on medications known to affect UA and HDL-C levels, such as urate-lowering therapies or statins, were not fully available or were missing for a substantial proportion of participants. As a result, residual confounding due to medication use may remain and should be addressed in future studies with more comprehensive medication tracking. Lastly, UHR may be affected by transient physiological conditions, dietary factors, hydration status, or medication use (e.g., urate-lowering or lipid-modifying drugs), which were not fully accounted for in our models. Despite these limitations, our study also has notable strengths. We used a large, nationally representative dataset (NHANES), which enhances the generalizability of our findings to U.S. women of reproductive age. The complex survey design and use of appropriate weighting improve the reliability of population-level estimates. Given the findings of this study, it is important to note that the results are hypothesis-generating. While we observed a potential association between elevated UHR and infertility, further studies, particularly those with longitudinal designs, are needed to validate these findings and clarify the directionality of the relationship. As such, UHR should not yet be considered a definitive biomarker for infertility. Any potential clinical applications, including its integration into routine reproductive health assessments, would require prospective validation to determine its true utility in practice.

Thanks to all NHANES participants and staff.