Platelet-to-high-density lipoprotein cholesterol ratio as a potential mediator in the obesity-chronic kidney disease association among hypertensive patients Authors: Xun Huang, Weihua Liu, Qinghua Liu, Lin Huang Published: 2025 ABSTRACT AbstractThis study investigated the relationship between the platelet-to-high-density lipoprotein cholesterol ratio (PHR) and chronic kidney disease (CKD) in a hypertensive population, exploring its mediating role in the obesity–CKD association. Using data from the National Health and Nutrition Examination Survey (NHANES) 2009–2018 on 10,467 hypertensive adults, we observed a complex association. Overall, PHR was positively correlated with CKD (odds ratio [OR] = 1.08, 95% CI: 1.05–1.14). However, the relationship was nonlinear and U-shaped, with an inflection point at a PHR Z-score of −0.32 (PHR = 166). Below this point, the association was negative (OR = 0.78, 95% CI: 0.65–0.91), whereas above it, the association was strongly positive (OR = 1.24, 95% CI: 1.15–1.36), particularly among individuals with obesity (BMI ≥ 30 kg/m2). Furthermore, PHR significantly mediated the relationship between obesity and CKD, accounting for 21.46% of the effect from BMI and 18.01% from waist circumference. In conclusion, PHR exhibits a U-shaped relationship with CKD in hypertensive individuals, most prominently in those with obesity, and may serve as an accessible biomarker of obesity-related inflammation for CKD risk stratification.Keywords: Chronic kidney disease, platelet-to-high-density lipoprotein ratio, NHANES, obesity, hypertension Introduction Chronic kidney disease (CKD) is defined as a clinical syndrome characterized by persistent abnormalities in kidney function or structure. This condition represents an irreversible and progressive deterioration that significantly increases the risk of cardiovascular mortality [1]. CKD has emerged as a major global health challenge, with increasing prevalence imposing substantial economic burdens on healthcare systems worldwide [2]. Recent epidemiological data indicate that CKD affects over 850 million people globally, accounting for trillions of dollars in annual medical expenditures and lost productivity [3]. Effective prevention and management of CKD progression require targeted interventional strategies, particularly for high-risk populations such as individuals with hypertension. The rising global incidence of hypertension can be largely attributed to demographic shifts such as population aging and unhealthy lifestyle factors, including obesity. Current data show that hypertension affects nearly one-third of adults worldwide, approximately 1.39 billion individuals [4]. Among hypertensive patients, CKD prevalence is notably high, ranging from 31% to 51% [5]. Both hypertension and CKD development are substantially driven by inflammatory mechanisms [6]. Chronic inflammation is recognized as a crucial pathogenic mechanism in atherosclerosis development, highlighting the importance of identifying early inflammatory biomarkers for diagnostic and therapeutic purposes [7]. The platelet-to-high-density lipoprotein ratio (PHR) was first proposed by Jialal et al. as a biomarker for predicting metabolic syndrome [8]. Subsequent population-based studies have demonstrated that PHR serves as a reliable inflammatory biomarker with diagnostic and prognostic value in cardiovascular and metabolic diseases [9–11]. PHR serves as an integrative index of the inflammatory thrombotic milieu, capturing the equilibrium between platelet-mediated prothrombotic and proinflammatory drive and the opposing antiplatelet and anti-inflammatory effects of high-density lipoprotein cholesterol (HDL-C) [12]. HDL-C particles modulate platelet reactivity and vascular homeostasis via SR‑BI and S1P receptor-dependent signaling, in part by promoting endothelial nitric oxide and prostacyclin production, thereby exerting antiplatelet, anti‑inflammatory, and antithrombotic effects [13]. In CKD, HDL-C often loses its normal function. It becomes less effective at removing cholesterol from cells, undergoes harmful chemical changes through oxidation and carbamylation and shows reduced antioxidant enzyme activity. These changes weaken HDL-C’s protective effects and promote inflammation and blood clotting [14,15]. Accordingly, PHR captures the net shift toward a pro-inflammatory and pro-thrombotic balance when platelets are elevated and HDL-C-mediated defenses are weakened, a construct shown to be relevant across cardiometabolic conditions [15]. Compared with LDL-C, which mainly indexes pro-atherogenic burden rather than an endogenous anti-inflammatory buffer, HDL-C is better suited as the denominator to represent the counter-regulatory arm in this ratio [16]. Prior studies have only identified an association between PHR and CKD within the general population [17]. Hypertensive individuals represent a distinct subgroup with a specific disease mechanism. We therefore focused on hypertensive adults, a high-risk, clinically actionable population with a high baseline prevalence of CKD. Given the chronic low-grade inflammatory state, endothelial dysfunction, oxidative stress, and dyslipidemia that characterize hypertension [18,19]. In patients with high blood pressure, the balance between inflammation and blood clotting that PHR measures becomes more apparent. This makes it easier to identify meaningful biological patterns and reduces variability while limiting heterogeneity that would arise from mixing normotensive participants. Obesity is a major risk factor for CKD and renal failure [20], contributing to disease progression through multiple mechanisms, including impaired vascular endothelial function, activation of the renin-angiotensin-aldosterone system (RAAS), and establishment of chronic inflammation [21,22]. Given that PHR represents an inflammatory metabolic biomarker integrating platelet activation and lipid metabolism, both key components in obesity-related kidney injury, which may serve as an important mediator linking obesity to CKD development. Therefore, our research aimed to assess the association between PHR and CKD among hypertensive patients, with particular focus on quantifying how PHR might mediate the obesity and CKD relationship in this high-risk population. Methods Design of the study and participants This cross-sectional study used data from the National Health and Nutrition Examination Survey (NHANES) conducted between 2009 and 2018. NHANES is a nationally representative survey that assesses health and nutritional status using a complex multi-stage sampling design. The research protocol was approved by the National Center for Health Statistics (NCHS) Ethics Review Committee, and all participants provided informed consent. This study included 10,467 individuals with hypertension. The primary exclusion criteria were: participants with missing platelet count (PC) or HDL-C data, those with missing CKD data, and those without hypertension. Detailed information is shown in Figure 1. Flowchart of participant selection for the study. A total of 49,693 NHANES 2009–2018 participants were screened. After excluding those missing HDL-C or platelet count data (n = 13,321), CKD data (n = 4,809), and nonhypertensive individuals (n = 21,096), 10,467 hypertensive participants with complete data were included in the final analysis. Abbreviations: NHANES: National Health and Nutrition Examination Survey; HDL-C: high-density lipoprotein cholesterol; PC: platelet count; CKD: chronic kidney disease. Hypertension was defined by the following criteria: (1) physician diagnosis, (2) current use of antihypertensive medications, (3) average systolic blood pressure ≥140 mmHg and/or diastolic blood pressure ≥90 mmHg [23]. Definition of the PHR and CKD PHR was calculated as platelet count (10³/μL) divided by HDL-C (mmol/L) [8]. Given the non-normal distribution of PHR, we applied Z-score standardization to normalize the variable for subsequent statistical analyses. The Z-score transformation was performed using the formula: Z = (X – μ)/σ, where X represents the individual PHR value, μ is the population mean, and σ is the standard deviation. We defined outliers as |Z-score| > 2.58 (p < 0.01) and extreme outliers as |Z-score| > 3.29 (p < 0.001) [24,25]. Among the 10,467 participants, we identified 87 (0.83%) extreme outliers and 234 (2.24%) outliers in PHR distribution. CKD was defined as eGFR <60 mL/min/1.73m2 and/or urinary albumin-to-creatinine ratio (uACR) ≥30 mg/g [26]. The eGFR was calculated using the 2021 Chronic Kidney Disease Epidemiology Collaboration creatinine equation [27]. Due to the cross-sectional nature of this study, CKD was defined based on single measurements of kidney function parameters. Potential covariates This study evaluated several covariates including demographic characteristics (age, biologic sex, race, education, and marital status), anthropometric measures (BMI, waist circumference), socioeconomic status (poverty income ratio), lifestyle factors (smoking and alcohol use), comorbidities (cardiovascular disease and diabetes), and medication use (antihypertensive medications, lipid-lowering drugs, including no/atorvastatin/rosuvastatin/others, and glucose-lowering drugs including no/insulin/oral glucose-lowering drugs/both insulin and oral drugs). To assess obesity, we measured waist circumference (WC) and calculated the weight-adjusted waist index (WWI) by dividing WC (cm) by the square root of weight (√kg) [28]. Diabetes was diagnosed based on the following criteria: self-reported diagnosis, documented medical records, use of glucose-lowering medications or insulin therapy, fasting blood glucose ≥7.0 mmol/L, HbA1c ≥6.5%, or 2-h oral glucose tolerance test glucose ≥11.1 mmol/L [29]. CVD diagnosis was based on participants’ self-reported history of congestive heart failure, coronary heart disease, angina, myocardial infarction, or stroke [30]. Statistical analyses Continuous variables following a normal distribution were expressed as mean ± standard deviation (SD), while variables that follow a non-normal distribution were summarized using the median and interquartile range, denoted as P50 (P25, P75). For categorical variables, we documented the raw count alongside the percentage distribution (n, %). The participants in this study were stratified into four distinct groups according to the quartiles of the PHR, allowing for comparisons of baseline characteristics. The chi-square test was used to compare categorical variables across groups. For continuous variables, differences were assessed using one-way analysis of variance (ANOVA) for normally distributed variables and the Kruskal–Wallis test for non-normally distributed variables. In this study, we applied multivariate logistic regression to calculate odds ratios (OR) along with 95% confidence intervals (CI) to analyze the relationship between the PHR and CKD. We created three analytical models: Model 1 contained no adjustments; Model 2 incorporated adjustments for factors including age, sex, race, level of education, marital status, BMI, poverty income ratio, WC, and WWI; and Model 3 of the study further included adjustments for diabetes, smoking status, alcohol consumption, CVD, and the administration of antihypertensive, lipid-lowering and glucose-lowering medications. We employed generalized additive models integrated with penalized splines to establish smooth curves for analyzing the dose-response relationship between PHR and CKD. The Z-score transformation improved the precision of threshold detection by standardizing the distribution and reducing the influence of extreme values, thereby enhancing the robustness of the analysis. Upon identifying nonlinearity, a recursive algorithm was used to compute the inflection point. We then constructed two separate piecewise logistic regression models for the regions above and below the identified inflection point. Additionally, subgroup analyses were conducted to examine potential effect modifications by sex (male/female), age (<65/≥65), obesity (<30/≥30), diabetes (no/yes), smoking (no/yes), alcohol use (no/yes), lipid-lowering drugs (no/atorvastatin/rosuvastatin/others) and glucose-lowering drugs (no/yes/insulin/oral glucose-lowering drugs/both insulin and oral drugs). We conducted a mediation analysis to investigate whether PHR mediates the association between obesity indicators (BMI, WC, and WWI) and CKD, adjusted for age, sex, race, educational level, marital status, poverty income ratio, diabetes, smoking status, drinking status, CVD, and use of antihypertensive, lipid-lowering, and glucose-lowering drugs. We first verified the prerequisite pathways using linear regression (obesity indices vs. PHR) and logistic regression (obesity indices vs. CKD). Mediating effects were confirmed when indirect effects showed statistical significance based on bootstrap confidence intervals. After careful examination of extreme values, we retained all outliers in the primary analysis as they likely represented genuine physiological or pathological conditions rather than measurement errors. However, sensitivity analyses were conducted to verify the robustness of our findings by excluding outliers. Analysis 1: Excluding outliers (|Z-score| > 2.58, n = 234, 2.24%); analysis 2: Excluding extreme outliers (|Z-score| > 3.29, n = 87, 0.83%). The threshold effect analysis was repeated for each sensitivity analysis to verify the consistency of the identified inflection point and effect sizes. Receiver operating characteristic (ROC) analysis was performed to compare the diagnostic performance of uACR alone versus uACR combined with PHR for CKD diagnosis. Areas under the curve (AUC) were compared using DeLong’s test. Net reclassification improvement (NRI) and integrated discrimination improvement (IDI) were calculated using bootstrap methods (n = 1000) to assess the incremental predictive value of PHR. Two-tailed approaches were used for the significance tests, with a significance level fixed at 0.05. Statistical analysis was performed through R software version 4.2.2 (R Foundation) and Empower Stats platform (http://www.empowerstats.com, X&Y Solutions, Inc., Boston, MA, USA). Results Demographic characteristics of participants Table 1 presents the baseline characteristics of 10,467 hypertensive participants categorized into quartiles according to their PHR. On average, the participants were 58.70 ± 15.44 years old, with males representing 50.20%. As PHR levels increased from Q1 to Q4: increases were observed in the proportion of male participants, BMI, WC, WWI, smoking rates, uACR, eGFR, PC, diabetes prevalence, and the percentage of married individuals. In contrast, higher levels of PHR were linked to a greater proportion of Other Hispanic individuals, lower poverty income ratios, and reduced educational levels. Additionally, there was a noted decrease in the percentage of current drinkers, alongside lowered HDL-C levels, with fewer participants being prescribed lipid-lowering and more glucose-lowering medications. No significant differences were found in the utilization of antihypertensive medications among the various groups. Importantly, CVD and CKD prevalence exhibited a decline from Q1 to Q3, followed by an increase in Q4. Table 1.Baseline characteristics of the participants based on PHR quartiles.CharacteristicPHR QuartilesP-valueQ1Q2Q3Q4N2809248124902687 PHR113.12 (94.93, 126.26)158.49 (148.39, 168.07)201.57 (189.68, 214.79)275.00 (248.65, 318.78)<0.001Age, (Years)64.27 ± 14.0260.83 ± 14.9257.89 ± 15.2253.42 ± 15.80<0.001Male1279 (45.53%)1230 (49.58%)1293 (51.93%)1452 (54.04%)<0.001Race <0.001Non-Hispanic White, %1152 (41.01%)999 (40.27%)1001 (40.20%)1084 (40.34%) Non-Hispanic Black, %871 (31.01%)624 (25.15%)601 (24.14%)664 (23.06%) Mexican American, %770 (11.12%)1049 (15.14%)1259 (18.18%)623 (23.19%) Other Hispanic, %285 (10.15%)304 (12.25%)348 (13.98%)376 (13.99%) Other races, %287 (10.22%)298 (12.01%)280 (11.24%)319 (11.87%) BMI, kg/m228.43 ± 6.6030.57 ± 7.1631.87 ± 7.3533.34 ± 7.60<0.001WWI, (cm/√kg)11.29 ± 0.8111.39 ± 0.7811.46 ± 0.7611.49 ± 0.75<0.001Poverty income ratio2.11 (1.16, 4.07)2.04 (1.10, 4.00)1.92 (1.06, 3.73)1.76 (1.00-3.43)<0.001Education level, n (%) 0.057Less than college1375 (48.95%)1249 (50.34%)1302 (52.29%)1394 (51.88%) College or higher1434 (51.05%)1232 (49.66%)1188 (47.71%)1293 (48.12%) Marital status, n (%) <0.001Married/living with a partner1511 (53.79%)1469 (59.21%)1477 (59.32%)1581 (58.84%) Divorced/separated/widowed994 (35.39%)739 (29.79%)710 (28.51%)695 (25.87%) Never married304 (10.82%)273 (11.00%)303 (12.17%)411 (15.30%) Smokers, %1353 (48.17%)1133 (45.67%)1187 (47.67%)1369 (50.95%)<0.001Current drinking (%)1026 (36.53%)848 (34.18%)825 (33.13%)974 (36.25%)0.027HDL-C, (mmol/L)1.77 ± 0.471.39 ± 0.281.22 ± 0.241.01 ± 0.22<0.001PC, (1000 cells/uL)188.16 ± 46.11220.22 ± 43.35245.79 ± 48.79292.24 ± 69.56<0.001uACR, (mg/g)18.38 (4.40, 10.86)20.48 (4.19, 11.08)17.06 (4.26, 11.01)23.90 (5.60, 25.76)0.024eGFR, (mL/min/1.73 m²)84.08 ± 24.7287.05 ± 24.8990.50 ± 25.3093.38 ± 27.27<0.001Diabetes721 (25.67%)775 (31.24%)820 (32.93%)967 (35.99%)<0.001CVD570 (20.29%)471 (18.98%)426 (17.11%)489 (18.20%)0.024CKD815 (29.01%)637 (25.68%)576 (23.13%)715 (26.61%)<0.001Antihypertensive drugs2046 (72.84%)1859 (74.93%)1827 (73.37%)1951 (72.61%)0.232Lipoprotein-lowering drugs 0.002No1815 (64.20%)1534 (61.83%)1595 (64.42%)1815 (67.65%) Atorvastatin42 (1.49%)46 (1.85%)37 (1.49%)35 (1.30%) Rosuvastatin181 (6.40%)134 (5.40%)144 (5.82%)132 (4.92%) Others789 (27.91%)767 (30.91%)700 (28.27%)701 (26.13%) Glucose-lowering drugs <0.001No2367 (83.73%)1950 (78.60%)1918 (77.46%)2019 (75.25%) Insulin60 (2.12%)79 (3.18%)67 (2.71%)96 (3.58%) Oral glucose-lowering drugs331 (11.71%)378 (15.24%)385 (15.55%)427 (15.92%) Using both Insulin and oral drugs69 (2.44%)74 (2.98%)106 (4.28%)141 (5.26%) Abbreviations: PHR: Platelet-to-HDL-Cholesterol Ratio; BMI: body mass index; WWI: weight-adjusted waist index; HDL-C: high-density lipoprotein cholesterol; PC: Platelet count; uACR: urinary albumin-to-creatinine ratio; eGFR: estimated glomerular filtration rate; CVD: cardiovascular disease; CKD: chronic kidney disease. Baseline characteristics of the participants based on PHR quartiles. Abbreviations: PHR: Platelet-to-HDL-Cholesterol Ratio; BMI: body mass index; WWI: weight-adjusted waist index; HDL-C: high-density lipoprotein cholesterol; PC: Platelet count; uACR: urinary albumin-to-creatinine ratio; eGFR: estimated glomerular filtration rate; CVD: cardiovascular disease; CKD: chronic kidney disease. Associations of PHR and CKD Table 2 shows the multivariate logistic regression analysis results examining the association between PHR and CKD, adjusting for various confounding factors. Per SD increase in PHR, the OR for CKD was 1.08 (95% CI: 1.05–1.14). In the fully adjusted model with PHR categorized into quartiles, the ORs for CKD compared to Q1 were: Q2: 0.89 (95% CI: 0.87–1.04); Q3: 1.39 (95% CI: 1.19–1.61); and Q4: 1.18 (95% CI: 1.01–1.37). Table 2.ORs adjusted for multiple variables, 95% CIs, and P values of the PHR associated to CKD.CharacteristicsModel 1aModel 2bModel 3cOR (95% CI)P valueOR (95% CI)P valueOR (95% CI)P valueContinuous (PHRd)0.98 (0.94, 1.03)0.4311.12 (1.07, 1.18)<0.0011.08 (1.05, 1.14)<0.001Q11 (Referent) 1 (Referent) 1 (Referent) Q20.85 (0.75, 0.95)0.0070.94 (0.82, 1.08)0.3830.89 (0.87, 1.04)0.094Q30.74 (0.65, 0.83)<0.0010.92 (0.80, 1.05)0.2221.39 (1.19, 1.61)0.036Q40.89 (0.79, 1.00)0.0471.30 (1.13, 1.50)<0.0011.18 (1.01, 1.37)0.039P for trend0.010 0.002 0.098 Abbreviations: BMI: body mass index; WWI: weight-adjusted waist index; CVD: cardiovascular disease; CKD: chronic kidney disease. a Model 1 unadjusted. b Model 2: adjusted for age, sex, race, education, marital status, poverty-income ratio, BMI and WWI. c Model 3: additionally adjusting smoker, drinking status, diabetes, CVD, antihypertensive drugs, lipoprotein-lowering drugs, and glucose-lowering drugs. d PHR was analyzed using Z-score transformed. ORs adjusted for multiple variables, 95% CIs, and P values of the PHR associated to CKD. Abbreviations: BMI: body mass index; WWI: weight-adjusted waist index; CVD: cardiovascular disease; CKD: chronic kidney disease. Model 1 unadjusted. Model 2: adjusted for age, sex, race, education, marital status, poverty-income ratio, BMI and WWI. Model 3: additionally adjusting smoker, drinking status, diabetes, CVD, antihypertensive drugs, lipoprotein-lowering drugs, and glucose-lowering drugs. PHR was analyzed using Z-score transformed. The P value for trend was >0.05, indicating a potential non-linear relationship between PHR and CKD. Figure 2 visualizes this non-linear relationship, demonstrating a clear U-shaped curve. Table 3 presents the threshold effect analysis using a two-piecewise logistic regression model. The log-likelihood ratio test confirmed that the two-piecewise model was superior to the single-line model (p < 0.05). The PHR Z-score inflection point was −0.32 (PHR = 166). Below this threshold, PHR was negatively associated with CKD (OR: 0.78, 95% CI: 0.65–0.91). Above this threshold, PHR was positively associated with CKD (OR: 1.24, 95% CI: 1.15–1.36). Nonlinear association between PHR Z-score and adjusted OR for CKD. The solid line represents the adjusted OR for CKD across the range of PHR Z-scores, and the dotted lines indicate the 95% confidence intervals. Covariates were adjusted in the model. Abbreviations: PHR: platelet-to-HDL-C ratio; OR: odds ratio; CKD: chronic kidney disease. Table 3.Results from the two-piece logistic regression model.PHRCKD, OR (95% CI)a,bP-valueInflection point (K)−0.32 (166) 2.58 (Z-score = −0.31, 166.77), and excluding |Z-score| > 3.29 (Z-score = −0.32, 166), confirming the robustness of our findings. Subgroup analyses Subgroup analysis was performed to further investigate the association between PHR and CKD (Figure 3). The association was significantly stronger among obese participants (p = 0.002). No significant interactions were observed for gender, age, diabetes status, smoking habits, alcohol consumption, or the use of lipid-lowering or glucose-lowering drugs (all P for interaction > 0.05). Subgroup analysis of the association between PHR and CKD. Forest plot showing adjusted ORs with 95% CIs for CKD across subgroups defined by sex, age, BMI, diabetes, smoking, current drinking status, lipid-lowering drugs, and glucose-lowering drugs. All covariates from Model 3 were included as adjustment factors, except for the stratification variable. Abbreviations: PHR: platelet-to-HDL-C ratio; OR: odds ratio; CI: confidence interval; BMI: body mass index; CKD: chronic kidney disease. The mediating role of PHR The three obesity indices showed significant positive associations with both PHR levels and CKD across all models showed in Tables S1 and S2. In the fully adjusted model (Model 3), each 1-unit increase in obesity indices was associated with higher PHR (BMI: β = 0.04, 95% CI: 0.02–0.05; waist circumference: β = 0.06, 95% CI: 0.02–0.08; WWI: β = 0.28, 95% CI: 0.25–0.31; all p < 0.001) and increased CKD risk (BMI: OR = 1.11, 95% CI: 1.00–1.12, p = 0.015; waist circumference: OR = 1.11, 95% CI: 1.08–1.21, p = 0.001; WWI: OR = 1.24, 95% CI: 1.14–1.33, p < 0.001). These associations supported the conditions for mediation analysis. Mediation analysis revealed that PHR significantly mediated the relationships between obesity indicators and CKD risk (Figures 4 and 5). For BMI as shown in Figure 4(A), the indirect effect via PHR was 0.003 (95% CI: 0.0013–0.0056, p < 0.001), accounting for 21.46% of the total effect. For waist circumference, Figure 4(B), the indirect effect was 0.004 (95% CI: 0.0014–0.0064, p = 0.002), representing 18.01% of the total effect. For WWI, Figure 4(C), the indirect effect was 0.004 (95% CI: 0.0012–0.0063, p = 0.004), with a mediation proportion of 9.32%. BMI showed the highest proportion mediated by PHR, followed by waist circumference and WWI. All indirect effects were statistically significant, supporting PHR as an important inflammatory indicator linking obesity to CKD in hypertensive individuals. Mediation analysis of the association between PHR and CKD. Three mediation models examining whether PHR mediates the relationships between (A) BMI, (B) waist circumference, and (C) WWI and CKD. Each panel shows the indirect effect, direct effect, total effect, and proportion of mediation with 95% confidence intervals and P values. Arrows indicate the direction of associations. All models were adjusted for age, sex, race, education, marital status, poverty-income ratio, smoker, drinking status, diabetes, CVD, antihypertensive drugs, lipid-lowering drugs, and glucose-lowering drugs. Abbreviations: PHR: platelet-to-HDL-C ratio; CKD: chronic kidney disease; BMI: body mass index; WWI: weight-adjusted waist index; CI: confidence interval. Forest plot for mediation proportions for BMI, waist circumference, and WWI. BMI: body mass index; WWI: weight-adjusted waist index. Diagnostic performance of PHR for CKD ROC analysis comparing uACR alone versus uACR combined with PHR showed modest but statistically significant improvement (Figure S1). Model 1 (uACR alone) achieved an AUC of 0.89 (95% CI: 0.87–0.89), while Model 2 (uACR + PHR) demonstrated an AUC of 0.90 (95% CI: 0.89–0.91, p < 0.001). The incremental discrimination analysis revealed small improvements: NRI of 0.011 (95% CI: 0.006–0.016) and IDI of 0.012 (95% CI: 0.005–0.017) (Table S4). It is important that future longitudinal studies should evaluate both discrimination and calibration performance when developing PHR-based prediction models for incident CKD. These results reflect the ability to better distinguish individuals with versus without CKD at the time of assessment, rather than the prediction of future CKD development. Discussion This large cross-sectional study demonstrates a novel nonlinear association between PHR and CKD prevalence in hypertensive individuals. The findings indicate that both excessively high and low PHR levels are associated with a higher prevalence of CKD. Stratified analysis showed that the association between PHR and CKD was more pronounced in the obese population with hypertension. Furthermore, subsequent mediation analysis revealed a significant role of PHR as an inflammatory mediator in the association between obesity and CKD in the hypertensive population. PHR was first proposed by Jialal in 2021 and was considered a valid predictor of metabolic syndrome (MetS), and potentially a biomarker for atherosclerotic thrombosis risk [8]. Subsequent studies further confirmed the predictive value of PHR as an inflammatory marker for MetS in severely obese adults, with a P value of 0.022 and an AUC of 0.6054 (95% CI 0.5226, 0.6882) [31]. Lu et al. validated the existence of an L-shaped association between the PHR and liver fibrosis [32]. Furthermore, recent studies have shown that PHR has been validated as a predictor of type 2 diabetes in high-risk obese populations, with an AUC of 0.700 (p < 0.001) [33], and it has also expanded its relevance to other metabolic diseases and CVD [10,11,34,35]. Notably, PHR is closely associated with hyperuricemia and kidney stones [36,37]. However, the study by Li et al. indicated that the association between PHR and CKD was observed only in the general population [17]. Hypertension amplifies a thrombo-inflammatory balance captured by PHR and strengthens its link with CKD. This research was designed to explore the relationship between PHR and CKD further, ultimately revealing a U-shaped relationship in the hypertensive population. Persistent low-grade inflammation plays a significant role in the onset and progression of CKD, potentially altering and disrupting the regulation of renal microcirculation and perfusion distribution, ultimately leading to kidney damage, renal insufficiency, and the development of chronic renal failure [6,38]. Hypertension seems to be both a cause and a result of inflammation. Hypertension can trigger a systemic inflammatory response, and effectively reducing this inflammatory response can significantly lower the damage caused by hypertension to the kidneys and blood vessels [39]. Hypertension causes persistent damage to the kidneys through a complex inflammatory cascade. Initially, factors such as angiotensin II, elevated sodium levels, and sympathetic nervous system activation raise the levels of reactive oxygen species (ROS) in endothelial cells [40]. This pathway stimulates the activation of nuclear factor kappa B (NF-κB), initiating inflammatory transcription [41]. Consequently, endothelial cells express adhesion molecules (such as VCAM-1 and ICAM-1) and chemokines, which guide leukocytes, particularly monocytes, to migrate to the vascular wall [42]. These monocytes can differentiate into inflammatory macrophages or dendritic cells, activating the NLRP3 inflammasome and releasing inflammatory factors such as IL-1β and IL-1843 [43]. Simultaneously, the complement system is activated, producing C3a and C5a, which further enhance immune cell infiltration and amplify the inflammatory response [44]. Oxidative stress leads to lipid peroxidation, resulting in the formation of isoLGs that modify proteins and act as neoantigens, thereby activating T cells [45,46]. This persistent inflammation not only elevates blood pressure but also induces renal fibrosis, cell death, and permanent alterations in organ function, creating a vicious cycle [47]. Thus, this process is both a consequence of hypertension and a significant driving factor in its progression. Building on prior work, we identified a U-shaped, nonlinear association between PHR and CKD in hypertensive adults [17]. Distinct mechanisms likely operate at each extreme of the ratio, consistent with the ‘HDL-C paradox’ [48]. Among obese patients with very low PHR (due to very high HDL-C levels), these excessively elevated HDL-C concentrations may indicate dysfunctional HDL-C particles. These altered HDL-C particles lose their ability to remove cholesterol from cells and their anti-inflammatory and antioxidant properties. In some cases, they may even develop pro-inflammatory characteristics [49]. Genetic variation in individuals, such as specific mutations in cholesteryl ester transfer protein (CETP), may provide one explanation. Besides, the genes ABCA1, LIPC, and SCARB1 are associated with a heightened risk of CVD and increased levels of HDL-C [50–52]. One of the main functions of HDL-C in the body is to transport cholesterol, particularly during the lipolysis of triglyceride-rich lipoproteins (TGRL), where HDL-C can extract free cholesterol from these lipoproteins [53]. Feng et al. created a novel in vitro fluorescence method and discovered that the capacity to transfer free cholesterol was decreased by 45% among individuals with low HDL-C. However, it decreased by 20% in those with higher HDL-C, indicating that elevated HDL-C does not guarantee preserved function [54]. At high PHR values, elevated platelet counts together with reduced HDL-C reflect a pro-inflammatory and pro-thrombotic milieu characterized by platelet activation, release of inflammatory mediators, microvascular thrombosis, and diminished HDL-C-mediated anti-inflammatory defenses [55]. Together, these mechanisms provide biological plausibility for the observed U-shaped relationship between PHR and CKD in hypertensive adults, wherein both dysfunctional HDL-C at low PHR and inflammatory–thrombotic states at high PHR converge on increased CKD risk. Notably, subgroup analysis demonstrated that the association between PHR and CKD was significant among obese populations. Statistical mediation analysis showed that PHR is associated with the relationship between obesity and CKD, with statistical mediation proportions of 21.46% for BMI, 18.01% for waist, and 9.32% for WWI, respectively. This finding suggests that PHR may reflect inflammatory processes linking obesity-induced systemic inflammation to CKD. It is well-established that obesity constitutes a powerful and independent risk factor for CKD [56]. Adipose tissue, especially visceral adipose tissue, can release various pro-inflammatory cytokines, including leptin, adiponectin, tumor necrosis factor (TNF), monocyte chemoattractant protein-1 (MCP-1), transforming growth factor, and angiotensin II (ANG II). These inflammatory mediators create a chronic low-grade inflammatory state that is effectively captured by PHR through two complementary mechanisms: enhanced platelet activation and aggregation, leading to increased platelet counts, and reduced HDL-C levels and impaired HDL-C anti-inflammatory function [57]. Thus, elevated PHR may reflect the dual impact of obesity-induced inflammation - increased thrombotic risk through platelet activation and diminished protective effects of HDL-C. Increased levels of TNF-α and MCP-1 are commonly observed in obesity, where MCP-1 primarily functions to recruit macrophages to adipose tissue, contributing to adipose tissue inflammation and systemic inflammatory responses. The accumulation of perinephric fat (PF) can directly exert pressure on the vascular structures of the kidneys and the kidney tissue, increasing renal interstitial hydrostatic pressure. This condition can impair renal hemodynamics and potentially affect glomerular filtration efficiency [58]. Additionally, an overabundance of PF can cause kidney damage through local or systemic release of adipokines along with inflammatory mediators, and by stimulating the sympathetic nervous system and the RAAS [59]. Interestingly, our mediation analysis revealed that BMI demonstrated the highest mediation proportion through PHR (21.46%), suggesting that general obesity may exert stronger effects through systemic inflammatory pathways compared to central obesity measures. While WWI showed the strongest direct association with CKD, its lower mediation proportion through PHR (9.32%) indicates that central obesity may additionally operate through non-inflammatory mechanisms such as direct mechanical compression effects on renal structures. This provides support for our statistical mediation analysis, suggesting that PHR may reflect inflammatory processes that are statistically associated with both obesity and CKD prevalence. Our findings show that PHR reflects inflammation and blood clotting processes that increase CKD risk in obese patients with high blood pressure. This is especially true when PHR values are high, as platelets become more active while HDL-C protective effects are reduced. First, in obese hypertensive individuals with elevated PHR, guideline‑directed antiplatelet therapy may be considered when clinically indicated, with careful attention to bleeding risk in CKD. Second, optimization of HDL-C function—rather than pharmacologic elevation of HDL‑C should be prioritized via weight reduction, increased physical activity, smoking cessation, and management of triglyceride-rich lipoproteins. Agents with proven cardio and renal benefit and favorable metabolic and inflammatory effects may also improve the overall milieu that PHR indexes in obese hypertensive patients. Finally, serial PHR measurement could be explored as an accessible adjunct biomarker to monitor the thrombo-inflammatory balance in future longitudinal and interventional studies. Several important limitations should be acknowledged. First, the cross-sectional design of NHANES prevents us from establishing the temporal persistence (≥3 months) required for a definitive CKD diagnosis according to KDIGO guidelines. Our findings, therefore, represent associations with reduced kidney function at a single time point rather than confirmed chronic kidney disease. However, this approach is consistent with established epidemiological research using NHANES data and provides valuable insights into kidney function associations. Second, our mediation analysis showed statistical significance, but it cannot confirm true causal mediation pathways due to the lack of temporal ordering inherent in cross-sectional data. The observed mediation effects should be interpreted as statistical associations rather than causal relationships. Furthermore, PHR may function as both a marker and mediator of CKD, and the directionality of these relationships cannot be definitively established from our cross-sectional design. Longitudinal studies are needed to validate these findings and establish causal directions. Third, we restricted our analysis to hypertensive adults, which limits generalizability to normotensive populations. External validation in normotensive and general-population cohorts is needed to confirm the broader applicability of our PHR-CKD findings. Moreover, it should be noted that the observed effect sizes between PHR and CKD in our study were relatively small. Although the associations were statistically significant, the weak magnitude suggests that PHR contributes only moderately to CKD risk. Nevertheless, even small effect sizes at the population level should not be neglected. Fourth, our diagnostic performance analysis evaluated cross-sectional discrimination rather than longitudinal prediction of CKD development. Longitudinal validation is needed to determine whether PHR can effectively predict future CKD development and whether these discrimination improvements translate to clinically meaningful risk stratification. Conclusion In this research, we observed that both elevated and reduced PHR are associated with an increased odds of CKD, with this association being more pronounced in obese populations. Our mediation analysis revealed that PHR is statistically associated with the relationship between obesity and CKD, with mediation proportions of 21.46% for BMI, 18.01% for waist circumference, and 9.32% for WWI. These findings suggest that the association between obesity and CKD may partly involve inflammatory processes reflected by PHR, with BMI showing stronger associations through systemic inflammatory mechanisms compared to central obesity measures. Therefore, PHR may serve as a potential biomarker for identifying obesity-related CKD risk in hypertensive patients, particularly as it appears to capture the inflammatory component associated with obesity-related kidney dysfunction. Supplementary Material Supplemental_material.docx Figure_S1.tif Funding Statement This work was supported by the National Natural Science Foundation of China (Grant No. 82560082) and the Natural Science Foundation of Jiangxi Province (Grant No. 20212BAB206042). Acknowledgments We express our gratitude to all participants and NHANES workers for their generous contribution, which enabled us to conduct this study. Declarations This study fully complies with the ethical standards of the Declaration of Helsinki. The research was authorized by the National Center for Health Statistics Ethics Review Committee. All participants provided written informed consent. Disclosure statement No potential conflict of interest was reported by the author(s). Availability of data and materials The data that support the findings of this study are openly available in the National Center for Health Statistics at https://www.cdc.gov/nchs/nhanes/index.htm. References Associated Data Supplementary Materials Supplemental_material.docx Figure_S1.tif Data Availability Statement The data that support the findings of this study are openly available in the National Center for Health Statistics at https://www.cdc.gov/nchs/nhanes/index.htm.