Association of different insulin resistance surrogates with erectile dysfunction： a cross-sectional study

doi:10.21203/rs.3.rs-5282845/v1

Purpose： Metabolic syndrome and obesity are more common among men who have erectile dysfunction (ED), and there is a strong association between insulin resistance (IR) and the development of these diseases. However, no research has studied an attachment between different IR surrogates and ED. As a result, we scrutinized their potential relationship Among individuals in general.

Materials and Methods： The NHANES 2001-2004 dataset provided the basis for our analysis. We utilizedmultivariable logistic regression to investigate the link between ED and various IR surrogates (including the triglyceride glucose [TyG] index, homeostasis model assessment of insulin resistance [HOMA-IR] index, and triglyceride glucose-body mass [TyG-BMI]index). Additionally, we utilized smoothed curve fitting and threshold effect examination to assess curvilinear correlations and validate our findings. Subgroup analyses were conducted to identify sensitive populations, while the Receiver Operating Characteristic curve (ROC) study assessed the forecasting ability involving different IR surrogates for IR and ED.

Results: When controlling potential covariates, both the TyG index and the HOMA-IR index demonstrated no significant association with ED. Conversely, the TyG-BMI index/10 exhibited a favorable correlation with ED (odds ratio [OR] = 1.03, 95% confidence interval [CI]: 1.01, 1.06; P < 0.05). Smooth curve fitting and threshold effect analysis revealed a curvilinear connection between the TyG-BMI index with ED, indicating a remarkable inflection point at 320.78. Subgroup analysis illustrated the interactive effects of age, marital status, and cardiovascular diseases (CVDs) on the connection that exists among the TyG-BMI index and ED (P for interaction > 0.05). Nonetheless, different IR surrogates showed no difference in predictive ability for ED: HOMA-IR (0.559, 95% CI: 0.53, 0.59) compared to TyG (0.604, 95% CI: 0.57, 0.63) and TyG-BMI (0.558, 95% CI: 0.53, 0.59).

Conclusions： This research demonstrates a positive relationship between highlevels of the TyG-BMI index and ED among adults in the United States.

Metabolic syndrome

Insulin resistance

Erectile dysfunction

NHANES

Cross-sectional studies

Erectile dysfunction (ED) is characterized by the failure to obtain or maintain an erection adequate for Satisfying sexual behavior [1]. As men age, a growing percentage of them suffer from ED, a frequent condition[2]. Before the age of 40, 1–10% of persons develop ED, and the frequency of ED rises rapidly to 52% after 40 years[3]. As the prevalence of ED increases, so does its socioeconomic burden, recent research estimates that the cost of preventing and treating ED has exceeded $15 billion[4]. ED has emerged as a significant issue impacting human development.

One of the key components of the metabolic syndrome is insulin resistance (IR), which is characterized by decreased insulin sensitivity in stimulating glucose uptake[5]. The gold standard for determining insulin resistance is the hyperinsulinemic-normal glucose clamp test., but its sophistication and invasiveness limit the application of this approach in clinical research. [6]. An established alternative is the homeostatic model assessment of insulin resistance (HOMA-IR)[7]. Nevertheless, routine measurement of circulating insulin in primary care settings is uncommon, leading to the development of simpler alternatives for IR evaluation, like the triglyceride glucose (TyG) index and triglyceride glucose-body mass (TyG-BMI) index [8–11].

ED may stem from organic reasons or psychogenic issues, with both often contributing to its onset in most patients [12]. Numerous risk factors and disorders are intricately intertwined with ED, including obesity, diabetes, smoking, drinking, high blood pressure, sedentary lifestyle, dyslipidemia, hypogonadism, and IR [2, 13, 14]. Recent analyses have identified IR as an independent predictor of ED among patients with gout [15]. Higher prevalence of ED associated with the TyG index [16]. Anyway, to date, no studies have assessed the causal connection between ED and these IR surrogates.

Therefore, in order to shed new light on the management of male sexual health, this study used data from the National Health and Nutrition Examination Survey (NHANES) to look into any potential correlation between ED and several IR surrogates.

2.1 Study population

The National Center for Health Statistics (NCHS) of the Centers for Disease Control and Prevention conducts a thorough cross-sectional survey every two years, from which the analytic data were taken. This survey assesses the nutritional and health status of Americans through questionnaires and physical examinations [17]. Each participant in the NHANES study signed the informed consent form, approved by the NCHS Research Ethics Review Board.

Our study specifically utilized data from the 2001 to 2004 dataset, as information on ED was exclusively available during this timeframe. Among the 21161 participants, due to the ED questionnaire being administered solely to adult males aged 20 and older, we excluded 10860 female participants, and 5347 participants younger than 20. Furthermore, the participants with incomplete ED assessment(n=838), a former diagnosis of prostate cancer(n=120), missing BMI(n=101), triglyceride(n=2167), fasting glucose, and serum insulin(n=8) data were excluded. As shown in Figure 1, we ultimately included 1720 participants in the final analysis.

2.2 Measurement of IR surrogates

Blood samples for triglycerides, low-density lipoprotein (LDL), insulin, and fasting glucose were taken in the morning after fasting. The TyG index was generated as Ln[fasting triglycerides (mg/dL) × fasting glucose (mg/dL)/2][8]. The TyG-BMI index was generated as Ln[fasting triglycerides (mg/dL) × fasting glucose (mg/dL)/2] × BMI (kg/m2)[10]. The HOMA-IR index was generated as fasting glucose (mmol/L) × fasting insulin (µU/mL)/22.5[18]. IR was characterized by a threshold of HOMA-IR index ≥ 2.2. [19].

2.3 Assessment of ED

ED was evaluated as an outcome variable using a single question adapted from the Massachusetts Male Aging Study [20]. Participants were asked, “How would you describe your ability to achieve and maintain an erection adequate for Satisfying sexual behavior?” Options for responses included “almost always or nearly always,” “usually,” “sometimes,” and “never.” In the present study, ED was defined as participants who responded “never able” or “sometimes able” to keep the erection going.

2.4 Covariates of interest

The list of covariates encompassed age, race, marital status, education level, the family poverty income ratio (PIR), smoking status, drinking status, hypertension, diabetes, physical activity, CVDs, and LDL. Marital status was classified as Living alone or Married/living with a partner. Educational attainment was grouped as less than high school, high school, and more than high school. Family PIR served as a proxy for household income and was divided into three categories: "< 1.3" "≥ 1.3 and < 3.5" and " ≥3.5"[21]. Smoking status was determined by questionnaire (having smoked a minimum of 100 tobacco products throughout their lifetime; Positive replies classified respondents as smokers). Drinking status was assessed using a questionnaire (Consumption of at least 12 units of alcoholic beverages in the past year; Positive replies identified individuals as drinkers). Hypertension, diabetes, the level of physical activity, and coronary heart diseases were also determined through questionnaire responses. CVDs meant participants previously diagnosed with heart attack, angina, or heart failure.

2.5 Statistical analysis

Based on the HOMA-IR index, participants in this study were divided into two groups: the "IR group" and the "Non-IR group." Whereas continuous variables were displayed as means and standard deviations, categorical data were represented as numbers (n) and percentages (%). Logistic regression models were employed to examine the association between ED and different IR surrogates, presenting results as OR with corresponding 95% CI. Following STROBE guidelines, three regression models were established: Model 1 without any covariate adjustments, Model 2 adjusting for race and age, and Model 3 adjusting for race, age, education level, marital status, family PIR, hypertension, diabetes, smoking status, drinking status, physical activity, CVDs, and LDL. On the basis of these analyses, we assessed discrepancies further in ED risk across different TyG-BMI quartiles. Additionally, in Model 3, smooth curve fitting was employed to investigate potential curvilinear associations with ED and the TyG-BMI index. Threshold effects analysis was also conducted to identify associations and inflection points between the ED and TyG-BMI index.

Furthermore, we conducted interaction and stratified analysis dependent on race, age., education level, marital status, family PIR, smoking, drinking, hypertension, diabetes, moderate activity, and CVDs.

Finally, we contrasted the predictive performance of these three IR surrogates for IR and ED using ROC curves and corresponding areas under the curve (AUC).

The statistical analyses were carried out using R studio, release 3.4.3, and EmpowerStats, release 2.0. (http://www.empowerstats.com, X&Y Solutions, Inc.). P-value <0.05 was regarded as significant.

3.1 Baseline characteristics of study participants

The study included 1720 participants, categorized into two groups: the "IR group (HOMA-IR ≥ 2.2)" and the "Non-IR group (HOMA-IR < 2.2)". We found that males with ED were significantly elevated by 5.71% within the IR group in contrast to those without IR(p=0.008). The majority of the individuals in the IR group were older, more probable to be obese, alcoholics, and to have raised levels of triglycerides, insulin, LDL, fasting glucose, total cholesterol, and TyG-BMI index when compared to the non-IR group. In addition, their risk of CVDs, diabetes, and hypertension was higher than that of the non-IR group. (all p < 0.001). Table 1 presents the specific baseline characteristics of both groups.

3.2 The linkage between different IR surrogates and ED

Table 2 displays the ORs and 95% CIs for each of the three logistic regression models' connections between ED and different IR surrogates. In Models 1 and 2, the TyG index, TyG-BMI index/10, and the HOMA-IR index showed significant positive correlations with ED (all P < 0.05). Nevertheless, after adjusting for all covariates in Model 3, we discovered no significant relationships between the TyG index or HOMA-IR index and ED (P > 0.05), while the TyG-BMI index/10 still maintained a positive relationship with ED (OR = 1.03, 95% CI: 1.01, 1.06; P < 0.05). This suggests that a 3% rise in the number of ED patients is linked to every 10-unit increase in the TyG-BMI index. On the basis of these findings, we subsequently evaluated the disparities in ED risk among different TyG-BMI quartile groups. In Model 3, participants in the fourth quartile (Q4) experienced a 23% greater risk of ED compared to those in the first quartile (Q1) (OR = 1.23, 95% CI: 0.82-1.86, p = 0.32), suggesting a noteworthy correlation between elevated TyG-BMI index and heightened ED risk.

3.3 Smooth curve fitting and threshold effect analysis between ED and TyG-BMI index

we employed both smoothed curve fitting and threshold effect analysis to further elucidate the connection between the ED and TyG-BMI index. According to our research, there is a positive relationship between the TyG-BMI index and ED, as seen in Figure 2, indicating a non-linear relationship with an inflection point at 320.78. Below this threshold, there was no significant correlation between ED and the TyG-BMI index (p = 0.9285). However, once TyG-BMI exceeded 320.78, the prevalence of ED increased proportionally, with an OR of 1.01 (95% CI: 1.00-1.02, p = 0.0020), as shown in Table 3.

3.4 Subgroup analysis and Interaction test

We performed a subgroup analysis to evaluate the impact of the TyG-BMI index on ED. According to Table 4, subgroup analyses revealed a statistically significant association between the TyG-BMI index and ED in several groups: those younger than 50 years (OR = 1.01; 95% CI: 1.00, 1.01), Non-Hispanic White individuals (OR = 1.00; 95% CI: 1.00, 1.01), individuals who were married or living with a partner (OR = 1.01; 95% CI: 1.00, 1.01), those with more than a high school education (OR = 1.01; 95% CI: 1.00, 1.01), those with a Family PIR ≥ 3.5 (OR = 1.01; 95% CI: 1.00, 1.01), non-smokers (OR = 1.01; 95% CI: 1.00, 1.01), individuals with hypertension (OR = 1.00; 95% CI: 1.00, 1.01), and those without CVDs (OR = 1.00; 95% CI: 1.00, 1.01). Interaction tests indicated that age, marital status, and CVDs played interactive roles in the association between ED and the TyG-BMI index.

3.5 Comparing different IR surrogates for IR and ED prediction

Table 5 and Figure 3 display the results of the ROC curve. The AUCs for predicting IR (characterized as HOMA-IR ≥ 2.2) were 0.816 (95% CI: 0.80, 0.84) for the TyG-BMI index and 0.763 (95% CI: 0.74, 0.79) for the TyG index, indicating that in terms of IR prediction, the TyG-BMI index was significantly superior to the TyG index (P < 0.001). Regarding ED, the AUCs were 0.558 (95% CI: 0.53, 0.59) for the TyG-BMI index, 0.604 (95% CI: 0.57, 0.63) for the TyG index, and 0.559 (95% CI: 0.53, 0.59) for the HOMA-IR index. There were no statistically significant differences in the AUCs among these different IR surrogates (P > 0.05).

In this cross-sectional investigation, we investigated the association between different IR surrogates and ED in the wider community. According to our research, there is a positive link between male ED in the US and elevated TyG-BMI levels.

Smooth curve fitting and threshold effect analysis further supported this relationship, highlighting an increase in ED prevalence with higher TyG-BMI levels, particularly at the inflection point (320.78). Subgroup analyses revealed significant associations between the TyG-BMI index and ED in specific groups: Non-Hispanic White individuals, those younger than 50 years, married or cohabiting individuals, those with more than a high school education, those with Family PIR ≥ 3.5, non-smokers, individuals with hypertension, and those without CVDs or moderate activity. Interaction tests suggested that age, marital status, and CVDs influenced the connection among the TyG-BMI index and ED. However, ROC curve analysis indicated no significant differences among different IR surrogates in predicting ED.

Obesity is frequently linked with glucose intolerance and IR[22]. These factors are collectively known as metabolic syndrome [23], which is thought to adversely affect male sexual function [24]. In a comprehensive prospective study, Fung et al. investigated seven traditional CVDs risk factors in male participants and assessed them for ED after a 25-year follow-up. They found that BMI, mean age, triglycerides, and cholesterol were all markedly linked to an elevated risk of ED.[25]. In another prospective study, 17 males without ED and 78 eugonadal individuals with ED were compared. Those with ED exhibited higher BMI, waist circumference (WC), and HOMA-IR levels, suggesting that greater degrees of IR and obesity are associated with poorer erectile function[26]. Bacon et al. carried out a 14-year prospective research to find out possible causes of risk for ED and discovered obesity was positively correlated with ED.[27]. Morelli et al. discovered that physical activity helped restore some of the hallmarks of rabbits on a high-fat diet that developed metabolic syndrome. For instance, vigorous activity has the potential to enhance erectile function and elevate the concentration of luteinizing hormone and testosterone[28]. Ji et al. examined 80 individuals with gout and 70 controls who were healthy. They demonstrated that HOMA-IR was an independent predictor for ED by Multivariate logistic regression analysis[15].

The negative effects of IR on erectile function are caused by a number of different mechanisms. All components of metabolic syndrome are commonly found in obese populations [29, 30]. Abdominal obesity promotes IR, leading to hyperinsulinemia and hyperglycemia [31, 32]. The metabolic syndrome has been identified in the literature not only as being highly prevalent in men with ED[33]but also identified as a distinctive risk factor for ED[34, 35]. Metabolic syndrome can cause ED through various mechanisms: a) atherosclerotic disease affects penile vascular tissue, leading to ED; b) endothelial dysfunction decreases vascular nitric oxide levels, impairing vasodilation; c) hyperglycemia results in glycation of penile cavernosal tissue, impairing collagen turnover and contributing to ED, while also changing the ratios of interleukin-10 and tumor necrosis factor-α, upregulating E-selectin, and inducing endothelial dysfunction. [36–38].

The hyperinsulinemic-euglycemic clamp (HIEC) represents the gold standard for in vivo measurement of metabolic IR[39, 40]. This technique quantitatively evaluates insulin's impact on systemic glucose uptake by administering a precise insulin dose while maintaining normoglycemia through adjustable glucose infusion rates based on regular arterialized glucose monitoring and negative feedback[39, 40]. There is interest in employing clinically accessible fasting measures of glucose homeostasis, such as the homeostatic model evaluation[18], TyG index[8], and TyG-BMI index[41], as alternative techniques to validate the IR diagnosis because of the complexity and cost of HIEC. In the present study, the TyG-BMI index demonstrated a superior association with ED compared to the other two surrogates, a finding consistent across other IR-related conditions. WANG et al. similarly found in their study using NHANES data that the TyG-BMI index outperformed other IR surrogates in forecasting hyperuricemia among non-diabetic patients[42]. The TyG-BMI index's enhanced predictive capability likely stems from its inclusion of BMI, an obesity index, alongside measures of faulty fatty acid metabolism and aberrant glucose metabolism, thus augmenting its diagnostic potential compared to the HOMA-IR index or TyG index.

The present study has several advantages. We utilized a large dataset from NHANES, adjusting for relevant covariates in multivariable logistic regression analyses to investigate the association between TyG-BMI and ED. Additionally, we employed smoothing curve fitting and threshold effect analyses to examine any potential non-linear relationships between the TyG-BMI index and ED. Subgroup analyses were also conducted to assess the consistency of these effects. According to what we understand, this study is the first to look into the link between different IR surrogates and ED within the broader populace. But there are certain shortcomings in our study. It is cross-sectional in nature, limiting our ability to establish causality. Furthermore, ED history was self-reported by survey participants, potentially underestimating the true prevalence of ED in the population.

Among adult males in the United States who are nationally represented, the TyG index and HOMA-IR index did not show any connection with ED after adjustments for covariates, whereas the TyG-BMI index demonstrated a consistent and robust positive association with ED, influenced by factors such as age, marital status, and cardiovascular diseases. These findings offer fresh perspectives on ED management and prevention. Nonetheless, the effectiveness of different IR surrogates in predicting ED was not different. Future cohort studies involving larger populations are necessary to confirm and expand upon these relationships.

Acknowledgements

Not applicable.

Author Contributions

ML: Writing – original draft, Validation, Methodology, Data curation. XL: Writing – original draft, Methodology. YX: Writing – original draft, data analysis. ZC: Writing – original draft. PH: Writing – review & editing, Project administration. YZ: Writing – review & editing, Supervision, Conceptualization.

Funding

The author(s) declare financial support was received for the research, authorship, and/or publication of this article. This project was supported by National Key Research and Development Program of China (2022YFA0806303), National Natural Science Foundation of China (82101739), Regional Joint Fund for Basic and Applied Basic Research of Guangdong Province (2021A1515111091), and President Foundation of Nanfang Hospital, Southern Medical University (2023A045, 2021C016).

Availability of data and material

The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found below: National Health and Nutrition Examination Survey (https://www.cdc.gov/nchs/nhanes).

Ethics approval and consent to participate

The studies involving humans were approved by National Center for Health Statistics Ethics Review Board. The studies were conducted in accordance with the local legislation and institutional requirements. The participants provided their written informed consent to participate in this study

Clinical trial number

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Yafi, F.A., et al., Erectile dysfunction. Nat Rev Dis Primers, 2016. 2: p. 16003.DOI:https://dx.doi.org/10.1038/nrdp.2016.3
Shamloul, R. and H. Ghanem, Erectile dysfunction. Lancet, 2013. 381(9861): p. 153-65.DOI:https://dx.doi.org/10.1016/s0140-6736(12)60520-0
Wang, W., et al., The association between heavy metal exposure and erectile dysfunction in the United States. Asian J Androl, 2023. 25(2): p. 271-276.DOI:https://dx.doi.org/10.4103/aja202237
Ayta, I.A., J.B. McKinlay, and R.J. Krane, The likely worldwide increase in erectile dysfunction between 1995 and 2025 and some possible policy consequences. BJU Int, 1999. 84(1): p. 50-6.DOI:https://dx.doi.org/10.1046/j.1464-410x.1999.00142.x
Lee, S.H., S.Y. Park, and C.S. Choi, Insulin Resistance: From Mechanisms to Therapeutic Strategies. Diabetes Metab J, 2022. 46(1): p. 15-37.DOI:https://dx.doi.org/10.4093/dmj.2021.0280
Matsuda, M. and R.A. DeFronzo, Insulin sensitivity indices obtained from oral glucose tolerance testing: comparison with the euglycemic insulin clamp. Diabetes Care, 1999. 22(9): p. 1462-70.DOI:https://dx.doi.org/10.2337/diacare.22.9.1462
Tang, Q., et al., Optimal cut-off values for the homeostasis model assessment of insulin resistance (HOMA-IR) and pre-diabetes screening: Developments in research and prospects for the future. Drug Discov Ther, 2015. 9(6): p. 380-5.DOI:https://dx.doi.org/10.5582/ddt.2015.01207
Ramdas Nayak, V.K., et al., Triglyceride Glucose (TyG) Index: A surrogate biomarker of insulin resistance. J Pak Med Assoc, 2022. 72(5): p. 986-988.DOI:https://dx.doi.org/10.47391/jpma.22-63
Nabipoorashrafi, S.A., et al., The accuracy of triglyceride-glucose (TyG) index for the screening of metabolic syndrome in adults: A systematic review and meta-analysis. Nutr Metab Cardiovasc Dis, 2022. 32(12): p. 2677-2688.DOI:https://dx.doi.org/10.1016/j.numecd.2022.07.024
Li, X., et al., Predictive Effect of Triglyceride Glucose-Related Parameters, Obesity Indices, and Lipid Ratios for Diabetes in a Chinese Population: A Prospective Cohort Study. Front Endocrinol (Lausanne), 2022. 13: p. 862919.DOI:https://dx.doi.org/10.3389/fendo.2022.862919
Er, L.K., et al., Triglyceride Glucose-Body Mass Index Is a Simple and Clinically Useful Surrogate Marker for Insulin Resistance in Nondiabetic Individuals. PLoS One, 2016. 11(3): p. e0149731.DOI:https://dx.doi.org/10.1371/journal.pone.0149731
Muneer, A., et al., Erectile dysfunction. Bmj, 2014. 348: p. g129.DOI:https://dx.doi.org/10.1136/bmj.g129
Dong, J.Y., Y.H. Zhang, and L.Q. Qin, Erectile dysfunction and risk of cardiovascular disease: meta-analysis of prospective cohort studies. J Am Coll Cardiol, 2011. 58(13): p. 1378-85.DOI:https://dx.doi.org/10.1016/j.jacc.2011.06.024
Rosen, R.C., et al., Epidemiology of erectile dysfunction: the role of medical comorbidities and lifestyle factors. Urol Clin North Am, 2005. 32(4): p. 403-17, v.DOI:https://dx.doi.org/10.1016/j.ucl.2005.08.004
Kim, J.H., et al., Insulin resistance is an independent predictor of erectile dysfunction in patients with gout. Korean J Intern Med, 2019. 34(1): p. 202-209.DOI:https://dx.doi.org/10.3904/kjim.2016.350
Li, L., et al., A higher TyG index is related with a higher prevalence of erectile dysfunction in males between the ages 20-70 in the United States, according to a cross-sectional research. Front Endocrinol (Lausanne), 2022. 13: p. 988257.DOI:https://dx.doi.org/10.3389/fendo.2022.988257
The National Health and Nutrition Examination Survey (NHANES), Database.; Available from: https://www.cdc.gov/nchs/nhanes/about_nhanes.htm.
Matthews, D.R., et al., Homeostasis model assessment: insulin resistance and beta-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia, 1985. 28(7): p. 412-9.DOI:https://dx.doi.org/10.1007/bf00280883
Greenwood, E.A., et al., Insulin resistance is associated with depression risk in polycystic ovary syndrome. Fertil Steril, 2018. 110(1): p. 27-34.DOI:https://dx.doi.org/10.1016/j.fertnstert.2018.03.009
Derby, C.A., et al., Measurement of erectile dysfunction in population-based studies: the use of a single question self-assessment in the Massachusetts Male Aging Study. Int J Impot Res, 2000. 12(4): p. 197-204.DOI:https://dx.doi.org/10.1038/sj.ijir.3900542
Xia, W., et al., Association between different insulin resistance surrogates and infertility in reproductive-aged females. BMC Public Health, 2023. 23(1): p. 1985.DOI:https://dx.doi.org/10.1186/s12889-023-16813-2
Cardozo, E., M.E. Pavone, and J.E. Hirshfeld-Cytron, Metabolic syndrome and oocyte quality. Trends Endocrinol Metab, 2011. 22(3): p. 103-9.DOI:https://dx.doi.org/10.1016/j.tem.2010.12.002
Eckel, R.H., S.M. Grundy, and P.Z. Zimmet, The metabolic syndrome. Lancet, 2005. 365(9468): p. 1415-28.DOI:https://dx.doi.org/10.1016/s0140-6736(05)66378-7
Schulster, M.L., S.E. Liang, and B.B. Najari, Metabolic syndrome and sexual dysfunction. Curr Opin Urol, 2017. 27(5): p. 435-440.DOI:https://dx.doi.org/10.1097/mou.0000000000000426
Fung, M.M., R. Bettencourt, and E. Barrett-Connor, Heart disease risk factors predict erectile dysfunction 25 years later: the Rancho Bernardo Study. J Am Coll Cardiol, 2004. 43(8): p. 1405-11.DOI:https://dx.doi.org/10.1016/j.jacc.2003.11.041
Knoblovits, P., et al., Erectile dysfunction, obesity, insulin resistance, and their relationship with testosterone levels in eugonadal patients in an andrology clinic setting. J Androl, 2010. 31(3): p. 263-70.DOI:https://dx.doi.org/10.2164/jandrol.109.007757
Bacon, C.G., et al., A prospective study of risk factors for erectile dysfunction. J Urol, 2006. 176(1): p. 217-21.DOI:https://dx.doi.org/10.1016/s0022-5347(06)00589-1
Morelli, A., et al., Physical activity counteracts metabolic syndrome-induced hypogonadotropic hypogonadism and erectile dysfunction in the rabbit. Am J Physiol Endocrinol Metab, 2019. 316(3): p. E519-e535.DOI:https://dx.doi.org/10.1152/ajpendo.00377.2018
Litwin, M. and Z. Kułaga, Obesity, metabolic syndrome, and primary hypertension. Pediatr Nephrol, 2021. 36(4): p. 825-837.DOI:https://dx.doi.org/10.1007/s00467-020-04579-3
Engin, A., The Definition and Prevalence of Obesity and Metabolic Syndrome. Adv Exp Med Biol, 2017. 960: p. 1-17.DOI:https://dx.doi.org/10.1007/978-3-319-48382-5_1
Vesa, C.M., D.C. Zaha, and S.G. Bungău, Molecular Mechanisms of Metabolic Syndrome. Int J Mol Sci, 2024. 25(10).DOI:https://dx.doi.org/10.3390/ijms25105452
McCracken, E., M. Monaghan, and S. Sreenivasan, Pathophysiology of the metabolic syndrome. Clin Dermatol, 2018. 36(1): p. 14-20.DOI:https://dx.doi.org/10.1016/j.clindermatol.2017.09.004
Bansal, T.C., et al., Incidence of metabolic syndrome and insulin resistance in a population with organic erectile dysfunction. J Sex Med, 2005. 2(1): p. 96-103.DOI:https://dx.doi.org/10.1111/j.1743-6109.2005.20120.x
Heidler, S., et al., Is the metabolic syndrome an independent risk factor for erectile dysfunction? J Urol, 2007. 177(2): p. 651-4.DOI:https://dx.doi.org/10.1016/j.juro.2006.09.043
Bal, K., et al., Prevalence of metabolic syndrome and its association with erectile dysfunction among urologic patients: metabolic backgrounds of erectile dysfunction. Urology, 2007. 69(2): p. 356-60.DOI:https://dx.doi.org/10.1016/j.urology.2006.09.057
Sanjay, S., et al., Metabolic syndrome: An independent risk factor for erectile dysfunction. Indian J Endocrinol Metab, 2015. 19(2): p. 277-82.DOI:https://dx.doi.org/10.4103/2230-8210.149322
Trussell, J.C. and R.S. Legro, Erectile dysfunction: does insulin resistance play a part? Fertil Steril, 2007. 88(4): p. 771-8.DOI:https://dx.doi.org/10.1016/j.fertnstert.2007.01.116
Fonseca, V. and A. Jawa, Endothelial and erectile dysfunction, diabetes mellitus, and the metabolic syndrome: common pathways and treatments? Am J Cardiol, 2005. 96(12b): p. 13m-18m.DOI:https://dx.doi.org/10.1016/j.amjcard.2005.07.005
Bergman, R.N., D.T. Finegood, and M. Ader, Assessment of insulin sensitivity in vivo. Endocr Rev, 1985. 6(1): p. 45-86.DOI:https://dx.doi.org/10.1210/edrv-6-1-45
DeFronzo, R.A., J.D. Tobin, and R. Andres, Glucose clamp technique: a method for quantifying insulin secretion and resistance. Am J Physiol, 1979. 237(3): p. E214-23.DOI:https://dx.doi.org/10.1152/ajpendo.1979.237.3.E214
Dang, K., et al., The association between triglyceride-glucose index and its combination with obesity indicators and cardiovascular disease: NHANES 2003-2018. Cardiovasc Diabetol, 2024. 23(1): p. 8.DOI:https://dx.doi.org/10.1186/s12933-023-02115-9
Wang, H., et al., Comparison of different insulin resistance surrogates to predict hyperuricemia among U.S. non-diabetic adults. Front Endocrinol (Lausanne), 2022. 13: p. 1028167.DOI:https://dx.doi.org/10.3389/fendo.2022.1028167

Table1 Baseline characteristics of participants according to different IR groups based on the HOMA-IR index

Characteristic	Non-IR	IR	P-value
Number	853	867
Age (years)	47.43 ± 19.10	50.37 ± 17.47	<0.001
BMI (kg/m²)	25.49 ± 3.84	30.31 ± 5.90	<0.001
Race			0.002
Mexican American	165 (19.34%)	209 (24.11%)
Other Hispanic	31 (3.63%)	20 (2.31%)
Non-Hispanic White	452 (52.99%)	484 (55.82%)
Non-Hispanic Black	178 (20.87%)	128 (14.76%)
Other Race	27 (3.17%)	26 (3.00%)
Marital status			<0.001
Living alone	308 (36.11%)	216 (24.91%)
Married or living with a partner	545 (63.89%)	651 (75.09%)
Education level			0.332
Less than high school	246 (28.84%)	227 (26.18%)
High school	199 (23.33%)	224 (25.84%)
More than high school	408 (47.83%)	416 (47.98%)
Family PIR			0.204
<1.3	191 (22.39%)	175 (20.18%)
≥1.3 and <3.5	375 (43.96%)	366 (42.21%)
≥3.5	287 (33.65%)	326 (37.60%)
Smoking status			0.344
No	339 (39.74%)	364 (41.98%)
Yes	514 (60.26%)	503 (58.02%)
Drinking status			0.001
No	117 (13.72%)	170 (19.61%)
Yes	736 (86.28%)	697 (80.39%)
Hypertension			<0.001
No	643 (75.38%)	554 (63.90%)
Yes	210 (24.62%)	313 (36.10%)
Diabetes			<0.001
No	817 (95.78%)	722 (83.28%)
Yes	27 (3.17%)	132 (15.22%)
Pre-diabetes	9 (1.06%)	13 (1.50%)
CVDs			<0.001
No	788 (92.38%)	753 (86.85%)
Yes	65 (7.62%)	114 (13.15%)
Moderate Activity			0.624
No	427 (50.06%)	417 (48.10%)
Yes	411 (48.18%)	437 (50.40%)
Unable to activity	15 (1.76%)	13 (1.50%)
LDL (mg/dL)	116.94 ± 34.07	123.02 ± 36.27	<0.001
Triglyceride (mg/dL)	113.22 ± 61.30	160.63 ± 72.09	<0.001
Total cholesterol (mg/dL)	191.93 ± 38.24	199.15 ± 39.29	<0.001
Fasting blood glucose (mg/dL)	96.36 ± 16.17	115.59 ± 37.20	<0.001
Insulin (uU/mL)	5.64 ± 2.14	18.27 ± 14.03	<0.001
TyG index	8.47 ± 0.52	9.00 ± 0.54	<0.001
TyG-BMI index	216.44 ± 38.23	272.85 ± 55.47	<0.001
Erectile dysfunction			0.008
No	643 (75.38%)	604 (69.67%)
Yes	210 (24.62%)	263 (30.33%)

Values are presented as mean ± standard deviation or n (%)

Abbreviations: BMI, body mass index; Family PIR, ratio of family income to poverty; CVDs, cardiovascular diseases; LDL, low-density lipoprotein; HOMA-IR, homeostasis model assessment of insulin resistance; TyG, triglyceride glucose index; TyG-BMI, triglyceride glucose-body mass

Table 2 Multivariate logistic regression analysis for the association between different IR surrogates and ED.

	Model 1 OR (95%CI) P-value	Model 2 OR (95%CI) P-value	Model 3 OR (95%CI) P-value
HOMA-IR index	1.07 (1.04, 1.09) <0.0001	1.06 (1.02, 1.09) 0.0005	1.02 (0.99, 1.06) 0.1433
TyG index	2.01 (1.67, 2.42) <0.0001	1.44 (1.15, 1.80) 0.0016	1.24 (0.96, 1.59) 0.0959
TyG-BMI index/10	1.04 (1.02, 1.06) <0.0001	1.04 (1.02, 1.07) 0.0003	1.03 (1.01, 1.06) 0.0190
TyG-BMI index quartile
Q1	1.0	1.0	1.0
Q2	1.44 (1.05, 1.97) 0.0223	0.92 (0.62, 1.36) 0.6732	0.97 (0.65, 1.46) 0.8908
Q3	1.46 (1.07, 1.99) 0.0183	1.01 (0.69, 1.50) 0.9468	1.11 (0.74, 1.68) 0.6140
Q4	1.74 (1.28, 2.36) 0.0004	1.37 (0.94, 2.01) 0.0998	1.23 (0.82, 1.86) 0.3196

OR, odds ratio; CI, confidence interval; Q1-Q4, quartile 1-quartile 4.

Model 1 was adjusted for no covariates;

Model 2 was adjusted for age, race;

Model 3 was adjusted for age, race, marital status, education level, the family PIR, smoking status, drinking status, hypertension, diabetes, moderate activity, cardiovascular diseases, low-density lipoprotein;

Table 3 Threshold effect analysis of TyG-BMI index on ED

TyG-BMI	OR (95% CI), P value
Model 1
A straight-line effect	1.00 (1.00, 1.01) 0.0190
Model 2
Fold points (K)	320.78
< K-segment effect 1	1.00 (1.00, 1.00) 0.9285
>K-segment Effect 2	1.01 (1.00, 1.02) 0.0020
Effect size difference of 2 versus 1	1.01 (1.00, 1.02) 0.0169
Equation predicted values at break points	-0.70 (-0.93, -0.48)
Log likelihood ratio tests	0.016

Adjusted for age, race, marital status, education level, the family PIR, smoking status, drinking status, hypertension, diabetes, moderate activity, cardiovascular diseases, and low-density lipoprotein.

Table 4 Subgroup analysis between TyG-BMI index and ED

Variables	OR (95%CI)	P value	P for interaction
Age			0.0037
<50	1.01 (1.00, 1.01)	0.0011
≥50	1.00 (1.00, 1.00)	0.5232
Race			0.0895
Mexican American	1.00 (0.99, 1.01)	0.913
Other Hispanic	1.00 (0.97, 1.02)	0.7067
Non-Hispanic White	1.00 (1.00, 1.01)	0.0188
Non-Hispanic Black	1.00 (1.00, 1.01)	0.2152
Other Race	1.08 (0.99, 1.17)	0.0871
Marital status			0.0188
Living alone	1.00 (0.99, 1.00)	0.4788
Married or living with a partner	1.01 (1.00, 1.01)	0.0011
Education level			0.1757
Less than high school	1.00 (1.00, 1.01)	0.7723
High school	1.00 (1.00, 1.01)	0.4937
More than high school	1.01 (1.00, 1.01)	0.003
Family PIR			0.0808
<1.3	1.00 (1.00, 1.01)	0.3116
≥1.3 and <3.5	1.00 (1.00, 1.01)	0.7267
≥3.5	1.01 (1.00, 1.01)	0.0013
Smoking status			0.0647
No	1.01 (1.00, 1.01)	0.0034
Yes	1.00 (1.00, 1.00)	0.3044
Drinking status			0.2807
No	1.01 (1.00, 1.01)	0.0311
Yes	1.00 (1.00, 1.01)	0.0831
Hypertension			0.5974
No	1.00 (1.00, 1.01)	0.2204
Yes	1.00 (1.00, 1.01)	0.032
Diabetes			0.079
No	1.00 (1.00, 1.00)	0.129
Yes	1.01 (1.00, 1.01)	0.0783
Pre-diabetes	1.03 (1.00, 1.06)	0.0607
CVDs			0.0259
No	1.00 (1.00, 1.01)	0.0035
Yes	1.00 (0.99, 1.00)	0.2172
Moderate Activity			0.1189
No	1.00 (1.00, 1.01)	0.0118
Yes	1.00 (1.00, 1.01)	0.1425
Unable to activity	0.99 (0.98, 1.01)	0.261

age, marital status, and CVDs played an interactive role in the association between TyG-BMI index and ED. All subgroups were adjusted for age, race, marital status, education level, family PIR, smoking status, drinking status, hypertension, diabetes, moderate activity, cardiovascular diseases, and low-density lipoprotein, except the stratification factor itself.

Table 5 Comparison of ROC curves for different surrogates to predict IR and ED

	Cutoff (Sensitivity, Specificity)	AUC (95% CI)	P-value
IR			<0.001
TyG index	8.665(0.742,0.668)	0.763(0.74,0.79)
TyG-BMI index	246.315(0.687,0.797)	0.816(0.80,0.84)
ED			>0.05
TyG index	8.912(0.497,0.661)	0.604(0.57,0.63)
HOMA-IR index	4.407(0.260,0.845)	0.559(0.53,0.59)
TyG-BMI index	206.735(0.806,0.279)	0.558(0.53,0.59)

AUC, area under the curve; CI, confidence interval

Z-test was used to compare tatistically significant differences between AUCs

No competing interests reported.

Association of different insulin resistance surrogates with erectile dysfunction： a cross-sectional study

Status:

Version 1

Abstract

Figures

1 Introduction

2 Materials and Methods

2.1 Study population

2.2 Measurement of IR surrogates

2.3 Assessment of ED

2.4 Covariates of interest

2.5 Statistical analysis

3 Results

3.1 Baseline characteristics of study participants

3.2 The linkage between different IR surrogates and ED

3.3 Smooth curve fitting and threshold effect analysis between ED and TyG-BMI index

3.4 Subgroup analysis and Interaction test

3.5 Comparing different IR surrogates for IR and ED prediction

4 Discussion

5 Conclusions

Declarations

References

Tables

Additional Declarations

Status:

Version 1