Dear Editor, We read with interest the cross-sectional analysis by Wu et al of NHANES 2001–2004 data, in which a low hemoglobin-to-red cell distribution width ratio (HRR) emerged as the strongest inflammatory predictor of self-reported erectile dysfunction (ED) in US men1. Their observation that HRR performed best (AUC ≈ 0.68) is intriguing and deserves follow-up. While the work adds to the growing literature linking systemic inflammation to ED, several methodological points deserve clarification before HRR could be recommended as a “screening tool” in andrological practice. This commentary was prepared in accordance with the TITAN Guidelines 2025 for the declaration and use of artificial intelligence in scientific manuscripts2. ED definition ED was assessed using a structured questionnaire, measured using only one question from the Massachusetts Male Aging Study. Subsequent evidence has shown that the self-reported question used in this study had a correlation of only 0.71–0.78 with the International Index of Erectile Function (an internationally used standard), potentially leading to minor misclassification of results3,4. While this bias is usually non-differential, it may lead to outcome contamination that could attenuate or inflate the observed odds ratios. Finding a solution to this issue (e.g., combining questionnaires with testosterone levels for assessment) would make the conclusions of this study more convincing5. 2. Testosterone adjustment Serum testosterone levels were not considered in the data analysis of this study, but previous studies have shown that this indicator was present in the NHANES 2001–2004 cycle5,6. Low testosterone is both a cause of ED and a determinant of hemoglobin concentration7. Consequently, low HRR may partially reflect unrecorded hypogonadism rather than simple systemic inflammation. Re-examining this association by adjusting for testosterone levels in future studies will help clarify the independent contribution of HRR. 3. Survival bias in an elderly-heavy sample Mean age of ED-positive men was approximately 66 years, and about 70 % had a smoking history in this study. Men with conditions such as occult hemoglobinopathies, chronic kidney disease, or cancer (which may lower HRR) may have already passed away before the NHANES examination. A cross-sectional design cannot distinguish whether a low HRR is a true predictor of ED or is simply a surrogate for unmeasured systemic disease that simultaneously reduces HRR and increases ED risk. 4. Drug interference This study did not control for the effects of drug usage, although the variable is actually included in the NHANES database8. For example, centrally acting antihypertensives, among prescription medications, are independently associated with ED and, via hemodilution, lower hemoglobin values9. If these drugs were initiated before the blood draw and the ED interview, they would act as a classical collider: simultaneously reducing HRR and increasing ED risk, thereby inflating the observed odds ratio. In addition, over-the-counter iron, vitamin B12, or folate preparations, among other medications, could increase hemoglobin and reduce red blood cell distribution width, thereby increasing HRR. If men with mild anemia and ED self-medicated before the blood draw, their HRR would be artificially elevated, potentially weakening the association of the study’s findings (i.e., making high HRR appear even more strongly linked to high ED risk). Once again, we sincerely thank the authors for their creative contribution to the field. By transforming routine clinical laboratory data into clinically relevant hypotheses, they have opened an exciting new avenue for the early recognition of ED. We look forward to seeing these initial findings refined in larger, prospective cohorts and, ultimately, tested in real-world screening programs. Such work will undoubtedly help translate inflammation-based biomarkers into practical tools that benefit ED patients worldwide.
Sun et al. (Mon,) studied this question.