Association of Geriatric Nutritional Risk Index with erectile dysfunction in the elderly population

Introduction

Erectile dysfunction (ED), clinically defined by the inability to achieve or maintain a penile erection sufficient for satisfactory sexual performance, represents a significant global health concern, particularly among males over 40 years of age (1). Longitudinal data from the Massachusetts Male Aging Study revealed that 52% of men aged 40–70 years experience some degree of ED (2). Epidemiological evidence underscores that ED is substantially underdiagnosed and is increasingly recognized as a multisystem disorder rooted in vascular endothelial dysfunction. Translational research now firmly establishes ED as an early preclinical biomarker of generalized vascular impairment, revealing robust epidemiological associations with cardiovascular diseases (CVDs) (3). The pathophysiological convergence may originate from shared mechanisms, including endothelial nitric oxide synthase uncoupling, oxidative stress-mediated vascular senescence, and chronic low-grade inflammation—processes that promote accelerated vascular aging and often precede overt clinical cardiovascular manifestations by 2 to 5 years (4). Consequently, there is a growing imperative to develop effective preventive strategies for ED, underscoring the need to identify reliable biomarkers capable of detecting preclinical risk.

Malnutrition represents a significant global public health challenge, adversely affecting physiological homeostasis and clinical outcomes across a spectrum of diseases (5). Older adults are particularly susceptible to nutritional deficits, with epidemiological data indicating 32% prevalence of protein-energy imbalance in this population (6). While the Nutritional Risk Index (NRI) has been widely utilized for malnutrition screening, its applicability in elderly populations is constrained by its dependence on historical body weight data—a metric often unreliable due to age-related alterations in body composition and recall inaccuracy (7). To address these limitations, the Geriatric Nutrition Risk Index (GNRI) was introduced as an innovative tool integrating three essential components: anthropometric data (height and sex-adjusted ideal body weight) and serum albumin concentration (8). Serum albumin serves as a dual-purpose biomarker, reflecting both nutritional status and systemic inflammation, with its levels inversely correlated with acute and chronic inflammatory states as well as malignancies (9). Accumulating evidence has established the prognostic value of GNRI in various clinical contexts, including chronic respiratory diseases (10), osteoporosis (11), prostate cancer (12), depression (13), lower urinary tract symptoms (14), and CVD (15). Nevertheless, the relationship between GNRI-quantified nutritional status and ED prevalence remains insufficiently explored.

The National Health and Nutrition Examination Survey (NHANES) is a nationally representative cross-sectional program designed to assess the health and nutritional status of the non-institutionalized U.S. population. Renowned for its comprehensive and multilevel data collection, NHANES provides detailed information on sociodemographic characteristics, dietary intake, physical examinations, and laboratory biomarkers, enabling robust epidemiological investigations across diverse geographic and ethnic groups. In the present study, we utilized NHANES data to examine whether lower GNRI values are associated with an increased risk of ED among older men. Our findings indicate that GNRI may serve as a practical and objective biomarker for early risk stratification and preventive strategies against ED in geriatric populations. We present this article in accordance with the STROBE reporting checklist (available at https://tau.amegroups.com/article/view/10.21037/tau-2025-485/rc).

Methods

Study population

The NHANES, conducted by the Centers for Disease Control and Prevention (CDC), is a biennial survey that employs a complex, multi-stage, stratified, probability-based sampling design to generate nationally representative estimates of health and nutritional status in the United States. As a comprehensive cross-sectional initiative, NHANES systematically collects extensive biomedical, nutritional, and health-related data from a carefully selected sample of non-institutionalized U.S. civilians. Data acquisition is performed by trained personnel through standardized household interviews, detailed physical examinations, and laboratory assessments in Mobile Examination Centers (MECs), encompassing demographic variables, biochemical markers, anthropometric measurements, and detailed documentation of health behaviors. Detailed protocols and demographic data are publicly available via the official NHANES portal (http://www.cdc.gov/nchs/nhanes). The survey protocol was reviewed and approved by the National Center for Health Statistics (NCHS) Institutional Review Board, and all procedures strictly adhered to the U.S. Department of Health and Human Services regulations for the protection of human research subjects. Written informed consent was obtained from all participants prior to their inclusion in the study. This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments.

We utilized data from the NHANES cycles 2001–2004, as these were the only survey periods that included standardized assessments for ED. From an initial cohort of 21,161 participants enrolled during this period, we applied the following sequential exclusion criteria: (I) female participants (n=10,860); (II) individuals under 40 years of age (n=6,633); (III) those with missing ED diagnostic records (n=459), and (IV) subjects with incomplete Geriatric Nutritional Risk Index (GNRI) measurements (n=660). The final analytical cohort consisted of 2,549 male participants, who were further stratified into ED cases (n=1,005) and non-ED controls (n=1,544). A detailed flow diagram illustrating participant eligibility and the inclusion/exclusion process is provided in Figure 1.

Figure 1 The flowchart for participant screening in this study. ED, erectile dysfunction; GNRI, Geriatric Nutritional Risk Index; NHANES, National Health and Nutrition Examination Survey.

Assessment of ED

ED ascertainment employed a survey question adapted from the Massachusetts Male Aging Study (MMAS). Participants responded to the standardized query: “How would you characterize your capacity to attain and sustain erections adequate for pleasurable partnered sexual activity?” Response options included: (I) always or almost always able; (II) usually able; (III) sometimes able; (IV) never able. Based on established MMAS diagnostic criteria (16-20), respondents selecting options 3–4 were defined as ED cases, whereas those endorsing options 1–2 were classified as non-ED controls (16-20).

Definition of GNRI

The GNRI, an evidence-based risk stratification tool, was originally designed to quantify nutritional risk in elderly populations across clinical environments, particularly those requiring surgical interventions. This instrument utilized three clinically accessible variables: serum albumin concentration (g/L), anthropometric height (m), and body mass (kg), calculated through the following formula: GNRI = (1.489 × albumin) + [41.7 × (body weight / ideal body weight)] (10-12). The ideal body weight is calculated from 22 times the height (in meters) squared (10-12). The weight/ideal weight ratio is set to 1 when the present body weight is greater than the ideal body weight (10-12).

Covariates

The following covariates were collected based on previous studies (16-23): demographic parameters [age, racial/ethnic identity, socioeconomic status, and body mass index (BMI)], behavioral factors (ethanol intake, tobacco usage, and leisure-time physical activity), and clinical comorbidities (cardiovascular disorders, diabetes, hypertension, and dyslipidemia). Ethanol consumption was quantified via daily gram intake, stratified as non (0 g/d), light (<27.9 g/d), and heavy (≥28 g/d) drinkers (18-20). Tobacco usage was classified as nonsmokers (lifetime cigarette consumption <100 U), former smokers (cessation ≥6 months preceding survey) and current smokers (active use within the prior 30 days) (18-20). Physical activity quantification derived from self-reported metabolic equivalent expenditure in moderate-to-vigorous activities. CVD ascertainment required either physician-diagnosed angina or coronary artery disease. Diabetes mellitus classification incorporated either a documented diagnosis or fasting glucose ≥126 mg/dL (7.0 mmol/L). Individuals with systolic/diastolic pressures ≥140/90 mmHg or antihypertensive pharmacotherapy were defined as having hypertension (16,17,23). High cholesterol was diagnosed when total cholesterol was ≥240 mg/dL (6.2 mmol/L), or lipid-modifying medication use (16,17,23).

Statistical analysis

Descriptive statistics for the study population were presented as mean ± standard deviation (SD) for continuous variables and as counts (percentages) for categorical variables. Continuous variables with substantial missing values were converted into categorical formats, with missingness handled as a separate dummy variable group labeled “missing”. To examine the dose-response relationship between GNRI levels and ED, we constructed multivariable logistic regression models. GNRI was categorized into quartiles (Q1–Q4), with the lowest quartile (Q1) serving as the reference group. Three hierarchical models were fitted: (I) a crude, unadjusted model; (II) Model 1, adjusted for age and racial/ethnic background; and (III) Model 2, which additionally incorporated adjustments for the ratio of family income to poverty (PIR), educational attainment, marital status, BMI, alcohol consumption, smoking status, physical activity level, and history of CVD, diabetes, hypertension, and hypercholesterolemia. Moreover, subgroup analyses were performed to evaluate effect modification in the GNRI levels and ED association across clinically relevant subgroups: CVD and diabetes. Statistical significance thresholds were established at α=0.05 (two-sided). Analytical procedures employed R v4.4.1 (R Foundation for Statistical Computing) adhering to reproducible research standards.

Results

Table 1 delineates sociodemographic and clinical disparities between ED cases and non-ED controls. The ED cohort exhibited characteristics including older age (68.5±11.2 vs. 54.4±11.4 years, P<0.001), non-Hispanic White ethnicity (61.8% vs. 57.4%, P=0.049), lower annual household income (PIR less than 1.5: 28.6% vs. 22.9%, P<0.001), lower educational attainment (less than high school: 38.6% vs. 22.9%, P<0.001), higher BMI (BMI ≥30 kg/m², 31.3% vs. 29.4%, P=0.03), non-drinker (79.6% vs. 72.7%, P<0.001), former smoker (52.1% vs. 35.7%, P<0.001), physical inactivity, and a higher prevalence of CVD (23.4% vs. 9.4%, P<0.001), diabetes (24.6% vs. 8.4%, P<0.001), hypertension (62.5% vs. 40.7%, P<0.001), and hypercholesterolemia (49.1% vs. 44.4%, P=0.02; Table 1). Moreover, ED cases demonstrated significant nutritional compromise, characterized by lower serum albumin concentrations (4.2±0.3 vs. 4.3±0.3 g/L, P<0.001) and diminished GNRI scores (104.0±4.7 vs. 106.0±4.3, P<0.001), with concurrent lower weight (85.4±18.7 vs. 87.0±16.8 kg, P=0.04) and height (173±7 vs. 175±8 cm, P<0.001; Table 1).

Table 2 displayed the dose-response relationship between the GNRI and ED risk. In the crude model, the GNRI was associated with a lower risk of ED [odds ratio (OR), 0.90; 95% confidence interval (CI): 0.88–0.93]. After multivariable adjustment for socio-demographics, comorbidities, and lifestyle factors (Model 2), this inverse correlation persisted (OR, 0.95, 95% CI: 0.93–0.98). When GNRI was stratified into four quartiles, the OR (95% CI) decreased from the lowest to the highest quartile of GNRI, reaching 0.67 (0.53–0.99), 0.54 (0.33–0.72), and 0.31 (0.21–0.45) in the crude model, respectively (Table 2). The dose-response trend remained significant in the adjusted model 1 (Q4 vs. Q1: OR, 0.48; 95% CI: 0.31–0.75; P for trend =0.007) and model 2 (Q4 vs. Q1: OR, 0.62; 95% CI: 0.39–0.98; P for trend =0.04). These findings suggest a protective effect of GNRI against the development of ED. Additionally, we observed the non-linear relationship between GNRI and the prevalence of ED utilizing smooth curve fitting (P for non-linear <0.01, Figure 2).

Figure 2 Smooth curve fitting showing the relationship between GNRI and erectile dysfunction. CI, confidence interval; GNRI, Geriatric Nutritional Risk Index; OR, odds ratio.

We further performed subgroup analyses to assess the relationship between the GNRI and ED among various subpopulations. Following full adjustment, non-CVD cohorts demonstrated significant inverse associations (Q4 vs. Q1: OR, 0.54, 95% CI: 0.30–0.95; P for interaction <0.05). Additionally, individuals without diabetes (Q4 vs. Q1: OR, 0.66; 95% CI: 0.40–0.99; P for interaction <0.05) exhibited a negative correlation with the prevalence of ED compared with those in the lowest quartile (Figure 3).

Figure 3 Subgroup analysis of GNRI and erectile dysfunction. CI, confidence interval; CVD, cardiovascular disease; GNRI, Geriatric Nutritional Risk Index; OR, odds ratio.

Discussion

This is the first study to evaluate the association between the GNRI and ED in the elderly population. Analysis of this population-based cohort revealed distinct nutritional deficits in ED subjects, manifesting as reduced serum albumin levels (4.2±0.3 vs. 4.3±0.3 g/L, P<0.001) and lower GNRI values (104.0±4.7 vs. 106.0±4.3, P<0.001). Multivariate regression models adjusted for metabolic confounders demonstrated that the GNRI was associated with a lower risk of ED (OR, 0.90; 95% CI: 0.88–0.93). When categorized into quartiles, the highest GNRI group exhibited 38% lower ED risk compared to the lowest quartile (Q4 vs. Q1: OR, 0.62; 95% CI: 0.39–0.98; P for trend =0.04). Subgroup analyses further indicated this protective association was particularly evident in participants without CVD (OR, 0.54, 95% CI: 0.30–0.95) or diabetes mellitus (OR, 0.66; 95% CI: 0.40–0.99).

Malnutrition in geriatric populations, though frequently underrecognized, has emerged as a critical determinant of age-related comorbidities. Emerging evidence substantiates the clinical relevance of malnutrition screening in chronic disease management. Originally developed as a composite nutritional metric to predict nutrition-associated complications, the GNRI has been validated across multiple disease domains (9). This index combines anthropometric parameters (height-adjusted body mass) with serum albumin levels, enabling multidimensional evaluation of nutritional status. The incorporation of serum albumin—a biomarker reflecting both nutritional status and systemic inflammation—along with physical measurements, allows for an objective quantification of nutritional risk in aging adults. Previous studies revealed an inverse association between GNRI and CVD (15). Participants with the highest GNRI group exhibited 63% lower CVD risk compared to the lowest quartile (OR, 0.37; 95% CI: 0.21–0.65, P=0.005) (15). CVD and ED exhibit overlapping epidemiological profiles, manifesting shared modifiable risk factors such as sedentary behavior, low physical activity levels, elevated blood pressure, tobacco use, glucose metabolism disorders, adiposity, and dyslipidemia (24). These conditions further converge on shared pathological pathways, particularly through impaired endothelial nitric oxide bioavailability and systemic inflammatory activation (24). Therefore, it is biologically plausible that improved nutritional status may confer beneficial effects on erectile function. Our results revealed an inverse correlation between the GNRI and ED risk. This might be explained by the following reasons. Firstly, nutritional deficiencies may induce critical micronutrient depletion, particularly vitamin B and vitamin D insufficiency, which disrupts endothelium homeostasis and muscle contraction (17,25). Secondly, impaired nutritional status further promotes ED progression via inflammatory cascades. Antioxidant depletion in malnourished states exacerbates oxidative stress, triggering inflammatory activation (26). This cascade compromises vascular endothelial integrity. Concurrently, inflammatory mediators induce vasomotor dysfunction through impaired nitric oxide bioavailability and heightened vascular tone (26). Additionally, it has been reported that the GNRI was positively related to serum total testosterone level (27), which also plays a critical role in erectile function.

Interestingly, Stratified analyses revealed a protective association between GNRI and ED prevalence in males without CVD and diabetes. This phenomenon may be attributable to exacerbated oxidative damage pathways in aging populations and individuals with chronic metabolic disorders (28-30). Elevated redox imbalance potentially attenuates GNRI’ protective efficacy against endothelial dysfunction, thereby facilitating ED development in these specific subgroups.

This investigation presents notable merits. Firstly, it pioneers in establishing the epidemiological linkage between GNRI and ED prevalence. Secondly, stratified analyses were implemented to discern subgroups deriving maximal benefit from nutrition improvement. Thirdly, the stratification of GNRI into quartiles revealed a dose-dependent protective gradient against ED. Furthermore, utilizing NHANES datasets ensured robust statistical power and enhanced external validity through nationally representative sampling. However, several limitations should also be noted. Firstly, residual confounders, including genetic polymorphisms and nuanced lifestyle parameters, remain unaccounted. Secondly, the observational study design precludes causal inference, with persistent confounding risks. In addition, it is uncertain whether a lower GNRI is a determinant of ED risks or if ED itself exacerbates malnutrition. Furthermore, the interpretation of our subgroup analyses is constrained by limited statistical power within certain strata, particularly evident in smaller subgroups such as participants with comorbid CVD or diabetes. Finally, the generalizability of these findings is restricted to U.S. demographics owing to their derivation from the epidemiologically representative NHANES database.

Conclusions

Our study revealed an inverse correlation between the GNRI values and ED in geriatric populations without CVD and diabetes by utilizing a nationally representative database. Multivariable regression models identified GNRI as both an independent predictor and clinically applicable screening tool for ED risk stratification in aging male.

Acknowledgments

We sincerely appreciate the meticulous language editing and polishing of our manuscript by Dr. Yan Xie from Michigan State University.

Footnote

Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://tau.amegroups.com/article/view/10.21037/tau-2025-485/rc

Peer Review File: Available at https://tau.amegroups.com/article/view/10.21037/tau-2025-485/prf

Funding: This work was supported by the Technological Innovation Talent Program of Loudi City Science and Technology Bureau [No. (2023) 35].

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://tau.amegroups.com/article/view/10.21037/tau-2025-485/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Salonia A, Capogrosso P, Boeri L, et al. European Association of Urology Guidelines on Male Sexual and Reproductive Health: 2025 Update on Male Hypogonadism, Erectile Dysfunction, Premature Ejaculation, and Peyronie's Disease. Eur Urol 2025;88:76-102. [Crossref] [PubMed]
2. Saffati G, Seyan Z, Rendon DO, et al. Erectile dysfunction as a holistic indicator of well-being: a narrative review. Sex Med Rev 2025;13:11-9. [Crossref] [PubMed]
3. Bertini A, Pozzi E, Fallara G, et al. The atherosclerotic cardiovascular disease risk score is a reliable tool to identify patients with arteriogenic erectile dysfunction. Andrology 2023;11:1451-9. [Crossref] [PubMed]
4. Corona G, Cucinotta D, Di Lorenzo G, et al. The Italian Society of Andrology and Sexual Medicine (SIAMS), along with ten other Italian Scientific Societies, guidelines on the diagnosis and management of erectile dysfunction. J Endocrinol Invest 2023;46:1241-74. [Crossref] [PubMed]
5. Dent E, Wright ORL, Woo J, et al. Malnutrition in older adults. Lancet 2023;401:951-66. [Crossref] [PubMed]
6. Tomasiewicz A, Polański J, Tański W. Advancing the Understanding of Malnutrition in the Elderly Population: Current Insights and Future Directions. Nutrients 2024;16:2502. [Crossref] [PubMed]
7. Mizerska A, Durlik M, Kędzierska-Kapuza K. Nutritional Risk of Candidates for Simultaneous Pancreatic-Kidney Transplantation-A Narrative Review. Nutrients 2023;15:4179. [Crossref] [PubMed]
8. Cheng K, Zhang J, Ye LY, et al. Geriatric nutrition risk index in the prediction of all-cause and cardiovascular mortality in older adults with hyperlipidemia: NHANES 1999-2018. BMC Geriatr 2024;24:634. [Crossref] [PubMed]
9. Bouillanne O, Morineau G, Dupont C, et al. Geriatric Nutritional Risk Index: a new index for evaluating at-risk elderly medical patients. Am J Clin Nutr 2005;82:777-83. [Crossref] [PubMed]
10. Chen Z, Wen Y, Li W, et al. Geriatric nutritional risk index as a predictor of mortality in women with chronic inflammatory airway disease: evidence from NHANES 1999-2018. Front Nutr 2025;12:1547952. [Crossref] [PubMed]
11. Huang W, Xiao Y, Wang H, et al. Association of geriatric nutritional risk index with the risk of osteoporosis in the elderly population in the NHANES. Front Endocrinol (Lausanne) 2022;13:965487. [Crossref] [PubMed]
12. Miao S, Bao C, Zhang Y, et al. Associations of the Geriatric Nutritional Risk Index with high risk for prostate cancer: A cross-sectional study. Nutrition 2023;115:112164. [Crossref] [PubMed]
13. Li Z, Zhang L, Yang Q, et al. Association between geriatric nutritional risk index and depression prevalence in the elderly population in NHANES. BMC Public Health 2024;24:469. [Crossref] [PubMed]
14. Zheng T, Sun H, Tang Y, et al. Association of lower urinary tract symptoms and geriatric nutritional risk index in men: a cross-sectional study based on NHANES. Front Med (Lausanne) 2024;11:1356921. [Crossref] [PubMed]
15. Li C, Sun Y, Wu H, et al. Nonlinear Association Between Geriatric Nutritional Risk Index and Cardiovascular Disease in the Elderly Based on the NHANES Database. Int Heart J 2025;66:13-20. [Crossref] [PubMed]
16. Wang W, Chen J, Peng L, et al. Food Insecurity May be an Independent Risk Factor Associated With Erectile Dysfunction in the United States: Analysis of the National Health and Nutrition Examination Survey Data. Sex Med 2022;10:100549. [Crossref] [PubMed]
17. Wang W, Luo Q, Xiang L, et al. Dietary intakes of vitamin B(6), folate, vitamin B(12) and erectile dysfunction: a national population-based study. Transl Androl Urol 2024;13:1395-404. [Crossref] [PubMed]
18. Wang W, Ma Y, Chen J, et al. The Association Between 2, 4-Dichlorophenoxyacetic Acid and Erectile Dysfunction. Front Public Health 2022;10:910251. [Crossref] [PubMed]
19. Wang W, Xiang LY, Ma YC, et al. The association between heavy metal exposure and erectile dysfunction in the United States. Asian J Androl 2023;25:271-6. [Crossref] [PubMed]
20. Yang Z, Wang W, Lin L, et al. The association between urinary organophosphate insecticide metabolites and erectile dysfunction in the United States. Int J Impot Res 2024;36:226-31. [Crossref] [PubMed]
21. Chen G, Wang W, Wei X, et al. Single-cell transcriptomic analysis reveals that the APP-CD74 axis promotes immunosuppression and progression of testicular tumors. J Pathol 2024;264:250-69. [Crossref] [PubMed]
22. Luo Z, Wang W, Xiang L, et al. Association between the Systemic Immune-Inflammation Index and Prostate Cancer. Nutr Cancer 2023;75:1918-25. [Crossref] [PubMed]
23. Wang W, Lu X, Shi Y, et al. Association between food insecurity and kidney stones in the United States: Analysis of the National Health and Nutrition Examination Survey 2007-2014. Front Public Health 2022;10:1015425. [Crossref] [PubMed]
24. Miner M, Parish SJ, Billups KL, et al. Erectile Dysfunction and Subclinical Cardiovascular Disease. Sex Med Rev 2019;7:455-63. [Crossref] [PubMed]
25. Latic N, Erben RG. Vitamin D and Cardiovascular Disease, with Emphasis on Hypertension, Atherosclerosis, and Heart Failure. Int J Mol Sci 2020;21:6483. [Crossref] [PubMed]
26. Mirali Z, Majdizadeh G, Baghbani M, et al. Relationship Between Malnutrition Inflammation Score and Dietary Fat Quality Indices with Cardiovascular Diseases in Hemodialysis Patients. Iran J Kidney Dis 2023;17:156-67.
27. Li T, Jiang N, Liang X, et al. Association of geriatric nutritional risk index with total testosterone in elderly adults in the US: evidence from NHANES 2011-2016. Front Endocrinol (Lausanne) 2024;15:1457956. [Crossref] [PubMed]
28. Kaltsas A, Zikopoulos A, Dimitriadis F, et al. Oxidative Stress and Erectile Dysfunction: Pathophysiology, Impacts, and Potential Treatments. Curr Issues Mol Biol 2024;46:8807-34. [Crossref] [PubMed]
29. Hirsch GE, Heck TG. Inflammation, oxidative stress and altered heat shock response in type 2 diabetes: the basis for new pharmacological and non-pharmacological interventions. Arch Physiol Biochem 2022;128:411-25. [Crossref] [PubMed]
30. Hajam YA, Rani R, Ganie SY, et al. Oxidative Stress in Human Pathology and Aging: Molecular Mechanisms and Perspectives. Cells 2022;11:552. [Crossref] [PubMed]