Abstract
Lutein, zeaxanthin, and β-cryptoxanthin are polar oxygenated carotenoids found to be detectable in more than 95% of the population in the United States. Research has linked these carotenoids with lower coronary heart disease prevalence. This study investigates the association of serum lutein/zeaxanthin and β-cryptoxanthin with erectile dysfunction (ED) among middle-aged and older men in the United States. Serum lutein/zeaxanthin and β-cryptoxanthin were independent variables. The outcome variable was ED. Analyzed data from 1,302 men (≥40 years old) who participated in the National Health and Nutrition Examination Survey 2001–2002 cross-sectional study were included. After adjusting for all covariates, serum lutein/zeaxanthin negatively correlated with ED (odds ratio [OR]: 0.972, 95% confidence interval [CI]: [0.951, 0.994], p = .011). However, a U-shaped association between serum lutein/zeaxanthin and ED was detected in men with diabetes or prevalent cardiovascular disease. A U-shaped non-linear association was observed between β-cryptoxanthin levels and ED. These findings suggest that while both lutein/zeaxanthin and β-cryptoxanthin are recognized as essential antioxidants, maintaining lower serum lutein/zeaxanthin levels and appropriate serum β-cryptoxanthin levels may offer potential benefits for individuals with ED. Further investigations, particularly prospective studies, are warranted to determine the role of serum lutein/zeaxanthin and β-cryptoxanthin in the biological mechanism associated with ED.
Introduction
Erectile dysfunction (ED) is a disorder in prevalence with age, referring to people who cannot obtain or maintain a penile erection during sexual intercourse and occurs mainly in males older than 40 years (Burnett et al., 2018; Shamloul & Ghanem, 2013). Reports from the United States and Europe suggest that ED affects approximately 52% of men aged 40 to 70 and 49% of men aged 50 to 80 (Feldman et al., 1994). A meta-analysis of nearly 6 years conducted in China revealed a progressive increase in the incidence of ED with age, ranging from 20.86% to 93.72% (Wang et al., 2017). The concrete mechanism between natural aging and ED remains elusive. Cellular senescence, oxidative stress, and inflammation are potential consequences of aging that may contribute to progressive damage over time (Echeverri Tirado et al., 2016). Medical conditions, including diabetes, hypertension, and coronary heart disease, have been identified as risk factors for ED (DeLay et al., 2016; Ponholzer et al., 2005).
Antioxidant phytochemicals have been a significant focus of nutrition research to mitigate and manage chronic ailments such as cardiovascular diseases and diabetes (Zhang et al., 2015). Lutein, zeaxanthin, and β-cryptoxanthin, characterized as polar oxygenated carotenoids due to their possession of 1 to 6 oxygen atoms, are presently included in a group of 11 newly proposed vitamins that are believed to foster healthy aging (Ames, 2018; Castenmiller & West, 1998). These extensively studied carotenoids are commonly found in various verdant vegetables, including kale, spinach, and broccoli. The dietary intake of lutein/zeaxanthin among American adults has experienced an increase from 1.3 mg/day to 1.7 mg/day within the previous three decades (Li et al., 2021). In mitigating coronary heart disease, including lutein in one’s diet has proven advantageous (Leermakers et al., 2016). A strong correlation exists between cardiovascular diseases and ED, as both conditions share similar risk factors and pathophysiological mechanisms (Terentes-Printzios et al., 2022).
There is, however, little information in previous research about the association between serum lutein/zeaxanthin and β-cryptoxanthin concentration and the prevalence of ED. Therefore, we conducted a cross-sectional study with data from the National Health and Nutrition Examination Survey (NHANES) 2001–2002 to estimate the associations between serum polar oxygenated carotenoids and ED.
Method
Study Population
NHANES has been conducting biennial surveys since 1999 to gather health-related data from a representative sample of the non-institutionalized population in the United States. The data collection involves in-person interviews, physical examinations, questionnaires, and laboratory tests. Detailed information regarding the survey methods can be found on the official website (http://www.cdc.gov/nchs/nhanes/index.htm).
The study utilized NHANES 2001–2002 data, specifically focusing on a cohort of 2,126 men with valid data on ED. Initially, we excluded 57 men with a prior diagnosis of prostate cancer, followed by the exclusion of 99 men with missing serum lutein/zeaxanthin measurements and seven men with missing serum β-cryptoxanthin measurements. We further narrowed the study’s scope to include only older men 40 and older to minimize potential confounding effects of age. Four men undergoing dialysis within the past 12 months were also removed from the analysis. Consequently, the final sample consisted of 1,302 men. Figure 1 illustrates the selection process of participants through a flow chart. Prior to the commencement of the study, ethical approval was obtained from the National Center for Health Statistics (NCHS) Ethics Review Board (Protocol Number: #98-12), and each participant provided written informed consent.
A Flowchart Showing the Selection of Study Population
Assessment of ED
The dependent variable in this study was the presence of ED. The NHANES 2001–2002 utilized the KIQ400 questionnaire to gather personal interview data on ED. The questionnaire specifically asked participants to describe their ability to achieve and maintain an erection sufficient for satisfactory intercourse. Response options included “always or almost always able,” “usually able,” “sometimes able,” or “never able.” This straightforward question was deemed a convenient and validated tool for assessing ED (O’Donnell et al., 2005). In the current study, individuals who responded with “sometimes able” or “never able” were classified as having ED, whereas those who indicated “always or almost always able” or “usually able” were categorized as not having ED.
Assessment of Serum Lutein/Zeaxanthin and β-cryptoxanthin (µg/dL)
The independent variables examined in this study were serum levels of lutein/zeaxanthin and β-cryptoxanthin. We divided the serum combined lutein/zeaxanthin levels into three tertiles: tertile 3 (≥18.6 µg/dL), tertile 2 (<18.6 and ≥12.1 µg/dL), and tertile 1 (<12.1 µg/dL). The serum β-cryptoxanthin levels were also divided into three tertiles: tertile 3 (≥10.3 µg/dL), tertile 2 (<10.3 and ≥5.5 µg/dL), and tertile 1 (<5.5 µg/dL). The serum lutein/zeaxanthin and β-cryptoxanthin levels were measured using high-performance liquid chromatography with photodiode array detection. The NHANES website extensively accounts for quality assurance and quality control procedures.
Assessment of Covariates
The present study identified covariates by utilizing previous clinical practice and method descriptions. The confounding variables in this study encompass a range of continuous variables, including age, C-reactive protein (CRP), the estimated glomerular filtration rate (EGFR), and high-density lipoprotein cholesterol (HDL-C). Categorical variables were considered, specifically race, education level, physical activities, body mass index (BMI), poverty income ratio (PIR), history of hypertension, history of prevalent cardiovascular disease, history of diabetes, smoking behavior, and alcohol use. The classification of race/ethnicity included Mexican American, Other Hispanic, Non-Hispanic White, Non-Hispanic Black, or Other Race—Including Multi-Racial. The participant’s level of education was evaluated through interviews and categorized into three levels: more than high school, high school, and less than high school. Regarding the average level of physical activity, participants were grouped into four categories: (1) individuals who frequently sit and do not walk often; (2) individuals who frequently stand or walk; (3) individuals who frequently climb stairs or hills; and (4) individuals engaged in heavy work or carrying heavy loads. The classification of BMI included three categories: under/average weight (<25.0 kg/m2), overweight (25.0–30.0 kg/m2), and obesity (≥30.0 kg/m2). We categorized the PIR as less than 1.3, between 1.3 and 1.8, and greater than 1.8. Hypertension was determined through self-reported history or current use of prescribed medication. In cases where respondents did not respond to the questionnaire, we utilized examination data consisting of four blood pressure measurements. According to the 2017 guideline of the American Heart Association, individuals who had a systolic blood pressure of at least 130 mmHg and a diastolic blood pressure of at least 80 mmHg when measured in the right arm on two or more occasions were classified as having hypertension (Whelton et al., 2018). Individuals who had a systolic blood pressure of less than 130 mmHg and at least 90 mmHg and a diastolic blood pressure of less than 80 mmHg and at least 60 mmHg when measured in the right arm on at least one occasion were considered to have normal blood pressure. Individuals who self-reported a history of diabetes or exhibited a glycohemoglobin level exceeding 6.5% were classified as individuals with diabetes. The term “prevalent cardiovascular disease” encompasses individuals who self-reported a history of coronary heart disease, myocardial infarction, congestive heart failure, or stroke. We classified smoking behaviors into non-smoking, past smoking, and current smoking. In light of the participants’ responses to alcohol consumption, four distinct groups were formed: lifetime abstainers (individuals who have never consumed more than 12 drinks in their lifetime), past drinkers (individuals who have consumed more than 12 drinks in their lifetime but have not consumed alcohol in the past year), current moderate drinkers (individuals whose weekly alcohol consumption was calculated by multiplying the average number of drinks per day by seven, with a threshold of less than 14 drinks for males and no instances of consuming five or more drinks in a single day in the past 12 months), and current excessive drinkers (individuals who consume more than 14 drinks per week for males or have consumed five or more drinks in a single day within the past 12 months; Ma et al., 2023). The chronic kidney disease epidemiology collaboration (CKD-EPI) equation was used to calculate the EGFR (Levey et al., 2009). The measurement of CRP and HDL-C followed a standard NHANES laboratory procedure.
Statistical Analysis
We undertook all statistical analyses using statistical software based on R (http://www.R-project.org, The R Foundation), including EmpowerStats (http://www.empowerstats.com, X&Y Solutions, Inc, CA, USA) and STATA 17.0 (StataCorp, College Station, TX, USA). The complicated survey design was considered by applying a suitable examination weight. In addition, we defined statistical significance as p < .05 (two-sided).
Results
Table 1 presents the general characteristics of the participants. Based on their serum combined lutein/zeaxanthin levels, individuals were classified into three tertiles: tertile 3 (≥18.6 µg/dL), tertile 2 (12.1–18.6 µg/dL), and tertile 1 (<12.1 µg/dL). The study included 1,302 participants, with 74.87% (n = 851) classified as non-ED and 25.13% (n = 451) classified as ED. The average age of the participants was 54.35 years. The mean serum combined lutein/zeaxanthin level was 16.46 ± 8.58 µg/dL, while the mean β-cryptoxanthin level was 8.77 ± 6.61 µg/dL. Significant differences were observed in most outcomes when examining tertiles of serum lutein/zeaxanthin levels, except for age, EGFR, diabetes, prevalent cardiovascular disease, hypertension, and physical activity (p > .05). Supplementary Table 1 provides an overview of the general characteristics of participants stratified by ED.
General Characteristics of Participants (n = 1,302) Stratified by Lutein/Zeaxanthin (tertile 1–3, µg/dL) in NHANES 2001–2002
Note. Weighted mean ± SD for continuous variables: p value was calculated by weighted linear regression model. Unweighted participant numbers and weighted percent for categorical variables: p value was calculated by weighted chi-square test. NHANES = National Health and Nutrition Examination Survey; HDL-C = high-density lipoprotein cholesterol; CRP = C-reactive protein; EGFR = estimated glomerular filtration rate; BMI = body mass index; ED = erectile dysfunction; MET = metabolic equivalent.
Table 2 presents the effect sizes for the relationship between serum lutein/zeaxanthin (categorized into three groups) and ED. In the unadjusted model, individuals with intermediate and deficient levels of lutein/zeaxanthin had odds ratios (ORs) of 1.126 (95% confidence interval [CI]: [0.847, 1.499], p = .143) and 1.524 (95% CI: [1.152, 2.017], p = .007) for ED, respectively, compared with those with optimal levels. When adjusting for multiple variables, the deficient category remained statistically significant in all adjustment models: full adjustment Model 4 (OR: 1.786, 95% CI: [1.243, 2.565], p = .003), Model 2 (OR: 1.942, 95% CI: [1.392, 2.710], p < .001), and Model 3 (OR: 1.735, 95% CI: [1.220, 2.467], p = .002).
Association Between Serum Combined Lutein/Zeaxanthin and ED, NHANES 2001–2002.
Note. Tertile 3 (≥18.6 µg/dL), Tertile 2 (<18.6 and ≥12.1 µg/dL), Tertile 1 (<12.1 µg/dL).
Model 1: unadjusted. Model 2: adjusted for age and race. Model 3: adjusted for Model 2 covariates + smoking, alcohol consumption, BMI, physical activity, PIR, and education. Model 4: adjusted for Model 3 covariates + cardiovascular disease, diabetes, hypertension, HDL-C, CRP, and EGFR. ED = erectile dysfunction; NHANES = National Health and Nutrition Examination Survey; BMI = body mass index; PIR = poverty income ratio; HDL-C = high-density lipoprotein cholesterol; CRP = C-reactive protein; EGFR = estimated glomerular filtration rate.
The study also examined the correlation between serum lutein/zeaxanthin levels and ED using generalized additive models with weighted parameters and smooth curve fittings. In the fully adjusted smooth curve fit models, a nearly negative linear association between serum lutein/zeaxanthin levels and ED was observed (Figure 2).
(A) Association Between Serum Lutein/Zeaxanthin and ED. (B) Association Between Serum β-Cryptoxanthin and ED
In subgroup analyses stratified by race, we observed that the association between serum lutein/zeaxanthin and ED was no longer significant for Mexican Americans, other Hispanics, or individuals of other races. However, a stronger negative association was found for non-Hispanic Black individuals compared with non-Hispanic White individuals (Figure 3). In analyses stratified by age, it was consistently observed that serum lutein/zeaxanthin was inversely related to ED for men, both above and below 60 years. However, a U-shaped relationship was identified between serum lutein/zeaxanthin and ED for men with diabetes or prevalent cardiovascular disease.
Association Between Serum Lutein/Zeaxanthin and ED Stratified by Race, Age, Diabetes, and Prevalent Cardiovascular Disease
The study presented the effect sizes of the relationship between serum β-cryptoxanthin levels (categorized into three groups) and ED in Table 3. In the unadjusted model, participants with intermediate and deficient β-cryptoxanthin levels, compared with those with optimal levels, exhibited ORs of 1.290 (95% CI: [0.894, 1.861], p = .173) and 1.569 (95% CI: [1.093, 2.252], p = .015) for ED, respectively. After adjusting for multiple variables, the pattern remained consistent, with only the deficient category showing statistically significant ORs in the fully adjusted Model 4 (OR: 1.689, 95% CI: [1.048, 2.721], p = .031), Model 2 (OR: 2.179, 95% CI: [1.439, 3.298], p < .001), and Model 3 (OR: 1.765, 95% CI: [1.104, 2.820], p = .018). In the fully adjusted smooth curve fit models, a U-shaped non-linear relationship between serum β-cryptoxanthin and ED was observed (Figure 2). We conducted subgroup analyses to stratify the data by race, revealing that Mexican Americans exhibited a U-shaped relationship between serum β-cryptoxanthin and ED (Figure 4). At the same time, non-Hispanic Whites and non-Hispanic Blacks displayed a negatively linear relationship. When the data were stratified by age, diabetes, and prevalent cardiovascular disease, a consistently U-shaped association was observed.
Association Between Serum β-Cryptoxanthin and ED, NHANES 2001–2002.
Note. Tertile 3 (≥10.3 µg/dL), Tertile 2 (<10.3 and ≥5.5 µg/dL), Tertile 1 (<5.5 µg/dL). Model 1: unadjusted. Model 2: adjusted for age and race. Model 3: adjusted for Model 2 covariates + smoking, alcohol consumption, BMI, physical activity, PIR, and education. Model 4: adjusted for Model 3 covariates + cardiovascular disease, diabetes, hypertension, HDL-C, CRP, and EGFR. ED = erectile dysfunction; NHANES = National Health and Nutrition Examination Survey; BMI = body mass index; PIR = poverty income ratio; HDL-C = high-density lipoprotein cholesterol; CRP = C-reactive protein; EGFR = estimated glomerular filtration rate.
Association Between Serum β-Cryptoxanthin and ED Stratified by Race, Age, Diabetes, and Prevalent Cardiovascular Disease
We performed a sensitivity analysis by excluding one subject with serum lutein/zeaxanthin levels equal to or greater than 60 µg/L. The analysis revealed that the association between serum lutein/zeaxanthin and ED remained negatively linear. However, upon excluding four subjects with serum β-cryptoxanthin levels equal to or greater than 50 µg/L, we observed a threshold effect between serum β-cryptoxanthin and ED. Supplementary Figures 1 and 2 provide a visual representation of these findings.
Discussion
This study utilized representative samples from NHANES 2001–2002 to assess the correlations between serum lutein/zeaxanthin, β-cryptoxanthin, and ED in middle-aged and elderly males. The findings revealed inverse associations between serum lutein/zeaxanthin and ED and non-linear associations between β-cryptoxanthin and ED. A U-shaped relationship was also observed between serum lutein/zeaxanthin concentration and ED in individuals with diabetes and those with prevalent cardiovascular disease.
Previous research has offered limited insights into the impact of serum lutein, zeaxanthin, and β-cryptoxanthin on ED. However, a cross-sectional study published this year conducted by Fujita et al. examined 335 men residing in the community who experienced ED (Fujita et al., 2023). The findings of this study revealed a significant inverse relationship between blood concentrations of lutein, zeaxanthin, and β-cryptoxanthin and the severity of ED. Ojo et al. conducted an experimental study utilizing a rat model, wherein their findings indicated that the ethyl acetate fraction derived from the Spondias mombin plant exhibited a significant inhibitory effect on phosphodiesterase type 5 (PDE5) and resulted in increased concentrations of cyclic guanosine monophosphate (cGMP) within the corpus cavernosal tissues (Ojo et al., 2019). Notably, the primary constituents of the ethyl acetate fraction were lutein, zeaxanthin, and β-cryptoxanthin. Our study has identified a negative linear correlation between serum lutein/zeaxanthin and ED and a U-shaped non-linear correlation between serum β-cryptoxanthin and ED. To further analyze the data more comprehensively, we conducted subgroup analyses following the guidelines outlined in strengthening the reporting of observational studies in epidemiology (STROBE). Regarding the association between serum lutein/zeaxanthin and ED, the subgroup analysis revealed that this association was insignificant among Mexican Americans, other Hispanics, or individuals of other races. However, the inverse association was strongest among non-Hispanic Black individuals, followed by non-Hispanic White individuals. A U-shaped curve relationship was also observed between diabetes and individuals with pre-existing cardiovascular disease.
The inverse relationship between lutein/zeaxanthin and ED may be attributed to reduced risk of atherosclerosis and endothelial dysfunction. Endothelial dysfunction is an early indicator of subclinical atherosclerosis (Solomon et al., 2003). Numerous studies have demonstrated a negative association between dietary lutein intake and the likelihood of developing coronary heart disease and stroke, potentially through mechanisms involving the mitigation of atherosclerosis and inflammation (Calvo, 2005; Dwyer et al., 2001; Leermakers et al., 2016). Lidebjer et al. discovered that stable and acute coronary artery disease patients exhibited significantly reduced plasma concentrations of oxygenated carotenoids (Lidebjer et al., 2007). An animal study conducted by Pongkan et al. demonstrated that β-cryptoxanthin had a more pronounced impact on cardioprotection in cases of cardiac ischemia-reperfusion injury (Pongkan et al., 2017). To effectively manage ED, regulating serum lutein/zeaxanthin and β-cryptoxanthin levels is imperative. There is a dearth of research and guidelines about the optimal levels of serum lutein/zeaxanthin and β-cryptoxanthin for preventing ED. In the future, the results of this study may be used to guide clinical practice of ED.
In summary, the results of this study hold significant clinical implications for the treatment of ED, emphasizing the necessity of early intervention for individuals at high risk. However, it is essential to acknowledge certain limitations within the present study. First, the study’s cross-sectional design restricts the ability to establish causal relationships between serum lutein/zeaxanthin, β-cryptoxanthin, and ED. Second, including patients with various medical conditions that may potentially confound the effects of ED further complicates the interpretation of the findings.
Conclusion
This cross-sectional study conducted among middle-aged and older men in the United States has indicated a positive correlation between elevated serum lutein/zeaxanthin levels and decreased ED. However, a U-shaped non-linear association between serum β-cryptoxanthin levels and ED has been observed. These findings suggest that both lutein/zeaxanthin and β-cryptoxanthin have distinct roles in the preventive mechanism of ED. Maintaining lower levels of serum lutein/zeaxanthin and appropriate levels of serum β-cryptoxanthin may prove beneficial in managing ED.
Supplemental Material
sj-tif-1-jmh-10.1177_15579883231216905 – Supplemental material for Associations Between Serum Polar Oxygenated Carotenoids Level and Erectile Dysfunction in Men Older Than 40 Years
Supplemental material, sj-tif-1-jmh-10.1177_15579883231216905 for Associations Between Serum Polar Oxygenated Carotenoids Level and Erectile Dysfunction in Men Older Than 40 Years by Wen Wei, Xiangyun Xu, Qinghua Luo and Leihua Cao in American Journal of Men's Health
Supplemental Material
sj-tif-2-jmh-10.1177_15579883231216905 – Supplemental material for Associations Between Serum Polar Oxygenated Carotenoids Level and Erectile Dysfunction in Men Older Than 40 Years
Supplemental material, sj-tif-2-jmh-10.1177_15579883231216905 for Associations Between Serum Polar Oxygenated Carotenoids Level and Erectile Dysfunction in Men Older Than 40 Years by Wen Wei, Xiangyun Xu, Qinghua Luo and Leihua Cao in American Journal of Men's Health
Supplemental Material
sj-xlsx-3-jmh-10.1177_15579883231216905 – Supplemental material for Associations Between Serum Polar Oxygenated Carotenoids Level and Erectile Dysfunction in Men Older Than 40 Years
Supplemental material, sj-xlsx-3-jmh-10.1177_15579883231216905 for Associations Between Serum Polar Oxygenated Carotenoids Level and Erectile Dysfunction in Men Older Than 40 Years by Wen Wei, Xiangyun Xu, Qinghua Luo and Leihua Cao in American Journal of Men's Health
Footnotes
Acknowledgements
We acknowledge the staff and the participants of the NHANES study for their valuable contributions.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
Supplemental Material
Supplemental material for this article is available online.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
