Development and evaluation of thirteen inflammation-related markers for erectile dysfunction in U.S. adults

Erectile dysfunction (ED) has increasingly become a widespread condition with both physical and psychological implications worldwide. This study aimed to examine the association between thirteen inflammation-related markers-HRR, IBI, SIRI, SII, NLR, PLR, AISI, PPN, CALLY, NPAR, MLR, LMR, and UHR-and the risk of ED, as well as to compare their predictive effectiveness for ED screening.

The 2001-2004 cycle of the National Health and Nutrition Examination Survey (NHANES) in the United States provided the data for this investigation. Logistic regression analyses and multivariable adjustments were conducted to examine the associations between thirteen indicators and ED. Receiver operating characteristic (ROC) curve analyses were performed to evaluate the screening capabilities of these indicators. Additionally, the correlations between the best predictive markers and ED were examined using smoothed curve fitting, while subgroup analyses explored differences in risk across various populations and the interactions of demographic and lifestyle factors.

1,150 people (29.25%) out of the 3,931 people that made up the final analysis were found to have ED. Multivariable logistic regression analysis identified HRR, SIRI, NLR, SII, PLR, NPAR, and MLR as independent risk factors for ED. Individuals in the top quartile of HRR (Q4) had considerably lower rates of ED than participants in the bottom quartile (Q1). HRR had an area under the ROC curve (AUC) of 0.6792. DeLong's test was conducted to compare the performance of different ROC curves, revealing that the AUC for HRR was significantly higher than that of the other metrics (P < 0.05). Smoothed curve fitting demonstrated a linear and negative correlation between HRR and ED (LLR > 0.05), particularly among smokers, alcohol users, and individuals with diabetes. However, interaction tests and subgroup analyses showed that the significant associations observed across different variable subgroups were independent of each other.

HRR, SIRI, NLR, SII, PLR, NPAR, and MLR are fundamental indicators for identifying ED. Among these, HRR demonstrated superior predictive ability for ED.