Abstract / Summary
Abstract Background Diabetic nephropathy (DN) is a major microvascular complication among diabetic patients, with its high prevalence and poor prognosis presenting significant challenges to global public health. The neutrophil percentage-to-albumin ratio (NPAR), a novel biomarker that integrates inflammatory status and nutritional metabolism, has not been clearly linked to DN risk. This study aims to systematically examine the association between NPAR and DN using a cross-sectional design, with a focus on identifying potential nonlinear relationships, thereby providing evidence for DN risk assessment and clinical decision-making. Methods A total of 8,336 diabetic patients from the National Health and Nutrition Examination Survey (NHANES,1999–2018) were included in the analysis. Weighted logistic regression models were used to evaluate the association between NPAR and the risk of DN. Restricted cubic spline (RCS) modeling was employed to assess dose-response relationships and detect potential nonlinear patterns, with the inflection point identified using a stepwise recursive method. A two-stage regression model was applied to evaluate the differential effects on either side of the inflection point. Subgroup analyses were conducted to explore potential effect modifications, while multiple imputation was performed to address missing data and confirm the robustness of the findings. Results Among the 8,336 participants, 1,927 (20.14%) were diagnosed with DN. After adjusting for confounding variables, weighted logistic regression revealed a significant positive association between NPAR and DN risk (OR = 1.39, 95% CI: 1.24–1.55). Quartile analysis indicated that individuals in the highest quartile (Q4) had significantly higher odds of DN compared to the reference group (OR = 2.24, 95% CI: 1.85–2.70). RCS analysis confirmed a significant nonlinear association (overall P < 0.001, nonlinear P < 0.001), characterized by a J-shaped curve, with the inflection point identified at 14.48. The two-stage regression model showed that when NPAR was below 14.48, no significant association was observed (OR = 1.04, 95% CI: 0.99–1.09). However, when NPAR was equal to or greater than 14.48, each unit increase in NPAR was significantly associated with a higher risk of DN (OR = 1.21, 95% CI: 1.17–1.26). Subgroup analysis revealed a significant interaction between education level and NPAR’s effect on DN risk (P interaction < 0.05), and sensitivity analysis confirmed the stability of the results. Conclusion This study provides the first systematic evidence of a nonlinear association between NPAR and DN risk, characterized by a J-shaped curve with an inflection point at 14.48. When NPAR is equal to or exceeds this threshold, an increase in NPAR is significantly associated with higher DN risk, whereas no significant association is observed below this level. These findings offer a quantitative basis for early DN risk stratification—suggesting that maintaining NPAR below 14.48 may help reduce risk, while individuals with NPAR above this threshold should undergo enhanced screening and evaluation. These results support the clinical application of NPAR as a biomarker for individualized intervention strategies.