Association of the Neutrophil to High-Density Lipoprotein Cholesterol Ratio and Platelet to High-Density Lipoprotein Cholesterol Ratio with Early Vascular Aging

Abstract

Purpose

This cross-sectional study aimed to investigate the association of complete blood count parameters and lipid metabolism‑related inflammatory indices with early vascular aging (EVA).

Methods

A total of 1114 participants were included with an overall EVA prevalence of 19.6%. Logistic regression was adopted to assess the independent associations of neutrophil to high-density lipoprotein cholesterol ratio (NHR) and platelet to high-density lipoprotein cholesterol ratio (PHR) with EVA after adjusting for traditional cardiovascular risk factors. Restricted cubic spline (RCS) models were applied to examine nonlinear relationships. Subgroup analyses were conducted to evaluate effect modification, and decision curve analysis (DCA) was used to assess the net benefit for risk stratification.

Results

In the fully adjusted model, both NHR (OR = 1.583, 95% CI: 1.078–2.327, P = 0.019) and PHR (OR = 2.050, 95% CI: 1.325–3.173, P = 0.001) remained independently associated with increased odds of EVA. The RCS analysis did not reveal significant nonlinear relationships for either biomarker (P for non-linearity > 0.05), suggesting that the associations were adequately captured by linear models. Subgroup analyses revealed that the positive association between NHR and EVA risk was more pronounced in men and smokers. For PHR, there was no significant interaction among different subgroups. DCA demonstrated that both NHR and PHR conferred modest net benefit gains (maximum 0.0127 and 0.0162, respectively) across clinically relevant threshold probabilities.

Conclusion

In this cross‑sectional population, both NHR and PHR were independently associated with EVA, with essentially linear dose‑response relationships. The association between NHR and EVA was particularly significant in men and smokers, showing population heterogeneity. Both indices provided a modest net benefit for identifying individuals with EVA, but the modest magnitude of this benefit suggests limited immediate clinical utility. Limitations include the cross‑sectional design and single‑center cohort, which limit causal inference and generalizability. Further prospective, multi‑center studies are needed to validate these findings.

Introduction

Early vascular aging (EVA) is a condition characterized by an accelerated rate of vascular aging in certain individuals, arising from intimal changes induced by genetic predisposition, environmental factors, or arterial injury.¹ EVA may exacerbate the onset and progression of cardiovascular disease (CVD).² Therefore, early detection and intervention of EVA are crucial for preventing or delaying the occurrence of CVD. Arterial stiffness is a sign of vascular aging, and the most accurate assessment method is to measure the pulse wave velocity (PWV). Carotid-femoral pulse wave velocity (cfPWV) is regarded as the gold standard for this evaluation.³ However, due to the complexity of the measurement procedure and patient discomfort, particularly during femoral assessment in the groin region, cfPWV is rarely implemented in clinical practice. Currently, the primary method for measuring vascular aging is brachial-ankle pulse wave velocity (baPWV), a non-invasive, easily operated technique. However, it is currently mainly applied in physical examination center in China and has not yet been widely implemented among outpatients and inpatients.

Inflammatory indices derived from complete blood count and lipid profiles are low‑cost and easily obtained. Specifically, the neutrophil to HDL cholesterol ratio (NHR) and platelet to HDL cholesterol ratio (PHR) reflect a dual burden of heightened inflammation and impaired lipid metabolism. Both neutrophils and platelets are key drivers of vascular inflammation and thrombosis, whereas HDL C exerts protective effects including anti-inflammatory and endothelium preserving functions. Recent studies have reported associations of NHR and PHR with various cardiovascular conditions, including coronary artery disease, stroke, and heart failure.^4–9 However, the existing data is scarce and there is no research focusing on EVA. Therefore, it is necessary to investigate whether these inflammatory indices are independently associated with EVA. This cross-sectional study aims to evaluate the association of NHR and PHR with EVA and to explore their potential utility for risk stratification.

Materials and Methods

Study Design

This was a cross-sectional study. All adults aged 18 years or older who underwent a routine health check-up between January 2020 and March 2021 at the Physical Examination Center of Changzhou First People’s Hospital in China were eligible for inclusion. Participants were excluded if they had any condition potentially affecting the blood system, such as active infection (n=15), hepatic (n=20), renal (n=7), or hemolytic disease (n=1), chronic inflammatory disorders (n=13), malignancy (n=18), heart disease (n=12), or a history of stroke (n=13). Following these exclusions, a total of 1114 residents were ultimately enrolled in the study (Supplementary Figure S1). This study followed the REporting of studies Conducted using Observational Routinely-collected health Data (RECORD) guidelines for studies using routinely collected health data.¹⁰ The RECORD checklist is provided as Supplementary Material 3. Ethical approval for this study was obtained from the Ethics Committee of the Research Department at Changzhou First People’s Hospital (No. [2021]KeDi 156). The study was conducted in accordance with the ethical principles outlined in the Declaration of Helsinki. All participants had provided written informed consent for their clinical data to be used for research purposes at the time of their routine health check-up, and the data were anonymized prior to analysis. As this was a retrospective analysis of routinely collected data with no additional interventions, the requirement for re-consent was waived by the ethics committee, and participants were not contacted again. Participation in the original health check-up was voluntary, and individuals could decline or withdraw from the check-up without consequence.

Measurement of EVA

In accordance with expert consensus,⁶ EVA was defined as a baPWV value exceeding the age‑specific 97.5th percentile of the general healthy population.¹¹ The baPWV measurements and the age‑specific reference percentiles – the latter derived from a database of 7881 healthy participants – were both obtained using a non‑invasive arteriosclerosis detection device (BP‑203 RPE III; Omron; Japan).¹² All measurements were performed by trained nurses between 7:00 and 9:00 a.m. using the BP-203 RPE III device (Omron Health Medical, China) in accordance with the manufacturer’s instructions. Participants were instructed to refrain from smoking, caffeine, and alcohol for at least three hours and from exercise for 30 minutes prior to examination. Following a five-minute seated rest in a temperature-controlled room (22–25°C), participants lay supine and remained still during the procedure. Cuffs were applied to both arms and ankles, with the lower edge positioned 2–3 cm above the elbow crease and 1–2 cm above the medial malleolus, respectively. Electrocardiogram electrodes were attached to both wrists, and a heart sound sensor was placed at the left sternal border. Given the strong correlation between left and right-side measurements, the average baPWV value was used for analysis.

Data Collection

Anthropometric measurements, including body weight, height, and resting heart rate, were systematically recorded. Body mass index (BMI) was calculated as weight in kilograms divided by the square of height in meters (kg/m²). Smoking status was categorized as ever smoker or non-smoker. Ever smokers were defined as those who had smoked at least 100 cigarettes during their lifetime.¹³ Drinking status was classified as current drinker or non-drinker, with current drinking defined as consuming at least 12 alcoholic beverages annually.¹⁴ Hypertension was diagnosed based on any of the following: an average systolic/diastolic blood pressure ≥140/90 mmHg from two independent readings, or current use of antihypertensive medication. Blood pressure was measured twice from the right arm at five-minute intervals, and the average value was calculated for analysis. Diabetes mellitus was defined by a fasting plasma glucose level ≥7.0 mmol/L, a random glucose level ≥11.1 mmol/L, or ongoing antidiabetic treatment. Hyperlipidemia was determined by self-reported physician diagnosis, current lipid-lowering therapy, or laboratory evidence of dyslipidemia (total cholesterol >6.2 mmol/L, low-density lipoprotein cholesterol >4.1 mmol/L, or triglycerides >2.3 mmol/L).

Blood samples were collected after an overnight fast. For complete blood count parameters (including white blood cell count, red blood cell count, hemoglobin, platelet count, and differential white blood cell counts), samples were analyzed within 4 hours of collection. For biochemical assays, plasma/serum was separated by centrifugation within 2 hours and stored at −80°C until analysis, after which levels of alanine aminotransferase (ALT), aspartate aminotransferase (AST), glutamyl transpeptidase (GGT), alkaline phosphatase (ALP), fasting venous blood glucose (FBG), urea nitrogen, creatinine, uric acid, total cholesterol (TC), triglycerides (TG), high-density lipoprotein cholesterol (HDL-C), and low-density lipoprotein cholesterol (LDL-C) were determined on a Roche autoanalyser, while homocysteine concentration was measured by high-performance fluorescence chromatography. All composite inflammatory indices were calculated as below (HDL-C unit: mmol/L):

1. Neutrophil-to-lymphocyte ratio (NLR): neutrophil count/lymphocyte count

2. Monocyte-to-lymphocyte ratio (MLR): monocyte count/lymphocyte count

3. Neutrophil-monocyte-to-lymphocyte ratio (NMLR): (neutrophil + monocyte count)/lymphocyte count

4. Neutrophil-to-HDL-C ratio (NHR): neutrophil count/HDL-C concentration

5. Monocyte-to-HDL-C ratio (MHR): monocyte count/HDL-C concentration

6. Lymphocyte-to-HDL-C ratio (LHR): lymphocyte count/HDL-C concentration

7. Platelet-to-HDL-C ratio (PHR): platelet count/HDL-C concentration

8. Systemic immune-inflammation index (SII): (platelet count × neutrophil count)/lymphocyte count

9. Systemic inflammation response index (SIRI): (monocyte count × neutrophil count)/lymphocyte count

Statistical Analysis

Normally distributed continuous variables were expressed as mean ± standard deviation ( ± s) and compared between two groups using independent samples t‑test; non‑normally distributed continuous variables were expressed as median (interquartile range) and compared using Wilcoxon rank‑sum test. Categorical variables were presented as counts (percentages) and compared using chi‑square test. Missing continuous variables were handled using multiple imputation with chained equations (20 imputations). No missing data were present for any categorical variable. The percentage of missing data for each variable was reported in Supplementary Table S1.

The least absolute shrinkage and selection operator (LASSO) logistic regression was employed to identify independent inflammatory indicators for the occurrence of EVA. The correlation between the inflammatory indicators and EVA in three different models was tested through a multivariate logistic regression model. Model 1 (unadjusted), Model 2 (adjusted for sex, age and BMI), and Model 3 (further adjusted for hypertension, diabetes mellitus, smoking history, alcohol intake, FBG and hyperlipidemia). E-values were calculated for Model 3 primary effect estimates to evaluate robustness against unmeasured residual confounding (diet, physical activity, long-term medication). Lacking external validation cohorts, 10-fold cross-validation with 1000 replications was implemented for internal validation to mitigate overfitting.

Subsequently, restricted cubic spline regression (RCS) analysis was used to explore any nonlinear relationship of the inflammatory indicators and EVA. Subgroup analysis was used to explore the potential differences among different subgroups. We employed decision curve analysis (DCA) to explore the clinical net benefit of inflammatory indicators in the context of EVA. Subsequently, the discriminatory power of these markers for predicting EVA in males and females separately was quantified by analyzing the area under the ROC curve. Data analyses were conducted using R Statistical Software (Version 4.3.0) and STATA 15.0 Statistical software, with a significance threshold at P < 0.05.

Results

Baseline Characteristics

After rigorous application of the inclusion and exclusion criteria, 1114 participants were included in the analysis, with an overall EVA prevalence of 19.6%. The median age of the study population was 50 years old, and 59.07% were male. Baseline demographic characteristics and laboratory parameters of the population were shown in Table 1. Compared to those without EVA, participants with EVA were more likely to be smokers and drinkers (all P < 0.05). Because EVA was defined using age‑specific baPWV percentiles (>97.5th percentile of healthy peers), the EVA group included younger individuals whose baPWV was elevated for their age, even if still within the absolute normal range for older adults. Therefore, the mean age in the EVA group was lower than that in the non‑EVA group (Table 1). This pattern reflects the age‑standardized definition of EVA and does not suggest a protective effect of younger age.

Table 1.

Baseline Characteristics

	Non-EVA (n= 896)	EVA (n= 218)	P value
Age, year	50.35±11.70	48.66±8.21	0.0015
Male n (%)	528 (58.93)	130 (59.63)	0.8500
BMI, kg/m2	24.64 (4.58)	25.05 (4.38)	0.1453
Smoking n (%)	55 (6.14)	30 (13.76)	0.0001
Drinking n (%)	20 (2.23)	14 (6.42)	0.0010
Hypertension n (%)	183 (20.42)	54 (24.77)	0.1600
Diabetes mellitus n (%)	53 (5.92)	11 (5.04)	0.6210
Hyperlipidemia n (%)	286 (31.92)	85 (38.99)	0.0470
Systolic pressure, mmHg	124 (25)	136 (25)	<0.0001
Diastolic pressure, mmHg	74 (15)	81 (15)	<0.0001
Heart rate, bpm	74 (15)	77 (15)	0.0005
WBC, 109/L	5.87 (1.96)	6.18 (2.26)	0.0178
RBC, 1012/L	4.80 (0.72)	4.91 (0.65)	0.0036
Platelet, 109/L	221 (74)	234 (74)	0.0010
Hemoglobin, g/L	147 (23)	148 (22)	0.2437
Neutrophil, 109/L	3.34 (1.40)	3.52 (1.42)	0.0084
Lymphocyte, 109/L	1.96 (0.73)	1.97 (0.75)	0.3366
Monocyte, 109/L	0.33 (0.15)	0.33 (0.14)	0.9437
ALT, U/L	20.4 (16.0)	22.0 (16.0)	0.3017
AST, U/L	22.6 (9.0)	21.8 (8.0)	0.0189
GGT, U/L	24.0 (25.0)	29.0 (25.0)	0.0360
ALP, U/L	78 (30)	79 (32)	0.8342
FBG, mmol/L	5.07 (0.83)	5.20 (1.13)	0.0005
Urea nitrogen, mmol/L	4.82 (1.46)	4.53 (1.52)	0.0098
Creatinine, umol/L	74 (19.5)	74 (18)	0.3370
Uric acid, umol/L	345.2 (121.5)	344.0 (122.1)	0.3263
TC, mmol/L	5.08 (1.25)	5.14 (1.12)	0.9533
TG, mmol/L	1.54 (1.17)	1.82 (1.23)	0.0001
HDL-C, mmol/L	1.20 (0.40)	1.14 (0.40)	0.0025
LDL-C, mmol/L	2.92 (1.04)	2.88 (1.00)	0.1658
NLR	1.71 (0.84)	1.78 (0.89)	0.1250
MLR	0.17 (0.07)	0.17 (0.06)	0.3679
NMLR	1.89 (0.89)	1.96 (0.93)	0.1713
NHR	2.74 (1.46)	2.96 (2.14)	0.0006
MHR	0.27 (0.15)	0.28 (0.19)	0.1012
LHR	1.58 (0.85)	1.72 (0.91)	0.0096
PHR	183.51 (90.55)	203.77 (102.96)	<0.0001
SII	380.85 (232.25)	425.55 (217.47)	0.0008
SIRI	0.56 (0.38)	0.59 (0.38)	0.2738

Notes: Continuous variables that satisfy the normal distribution were expressed as mean ± standard deviation (X±S), while those that do not conform to the normal distribution were expressed as median (interquartile range). Classified data were expressed in terms of the number of cases (percentage). EVA was defined using age‑specific percentiles; thus, the EVA group includes younger individuals with baPWV above the 97.5th percentile for their age, resulting in a lower mean age compared with the non‑EVA group.

Abbreviations: BMI, body mass index; WBC, white blood cell; RBC, red blood cell; ALT, alanine aminotransferase; AST, aspartate aminotransferase; GGT, glutamyl transpeptidase; ALP, alkaline phosphatase; FBG, fasting venous blood glucose; TC, total cholesterol; TG, triglycerides; HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol; NLR, neutrophil-to-lymphocyte ratio; MLR, monocyte-to-lymphocyte ratio; NMLR, neutrophil-monocyte to lymphocyte ratio; NHR, neutrophil (×10⁹/L) to high-density lipoprotein cholesterol (mmol/L) ratio; MHR, monocyte (×10⁹/L) to high-density lipoprotein cholesterol (mmol/L) ratio; LHR, lymphocyte (×10⁹/L) to high-density lipoprotein cholesterol (mmol/L) ratio; PHR, platelet (×10⁹/L) to high-density lipoprotein cholesterol (mmol/L) ratio; SII, systemic immune-inflammation index; SIRI, systemic inflammation response index.

In the comparison of the general information and chronic medical history of the two groups of participants, there were statistically significant differences between the two cohorts in terms of hyperlipidemia, systolic pressure, diastolic pressure and heart rate (P < 0.05). Comparison of laboratory parameters revealed statistically significant differences in WBC, RBC, platelet, neutrophil, AST, GGT, FBG, urea nitrogen, TG, and HDL-C (P < 0.05). Comparison of inflammatory markers revealed statistically significant differences in NHR, LHR, PHR, and SII between the two groups (all P < 0.05).

Association of NHR and PHR with EVA

The LASSO regression analysis was performed to identify significant inflammatory markers. The results demonstrated that two inflammatory markers, NHR and PHR, remained significantly and independently associated with EVA. Because NHR and PHR showed right-skewed distributions (Shapiro–Wilk test, both P < 0.001), they were log-transformed before being entered into the logistic regression models. After adjusting for traditional cardiovascular risk factors, the logistic regression model revealed a significant association between the NHR, PHR and EVA (both P<0.05) (Table 2). To assess the robustness of these associations against potential unmeasured confounding (eg, medication use, diet, and physical activity), we calculated the E‑values for the primary estimates derived from Model 3. The E‑values for NHR and PHR were 2.54 and 2.82, respectively, indicating that an unmeasured confounder would need to be associated with both exposure and outcome by a risk ratio of at least these magnitudes to completely explain away the observed effects - a threshold exceeding the effect sizes of most lifestyle factors, suggesting reasonable robustness. To assess the internal validity and generalizability of our fully adjusted model (Model 3), we performed 10‑fold cross‑validation with 1000 repetitions. The cross‑validated mean AUCs for Model 3 were 0.631 (95% CI: 0.573–0.690) for NHR and 0.635 (95% CI: 0.590–0.680) for PHR, suggesting modest discriminative ability for both biomarkers.

Table 2.

Logistic Regression Analyses Between the Inflammatory Indicators with EVA

Variable	Model 1		Model 2		Model 3
	OR (95% CI)	P-value	OR (95% CI)	P-value	OR (95% CI)	P-value
Log (NHR)	1.961 (1.370–2.807)	<0.001	1.900 (1.313–2.750)	0.001	1.583 (1.078–2.327)	0.019
Log (PHR)	2.401 (1.586–3.634)	<0.001	2.330 (1.534–3.538)	<0.001	2.050 (1.325–3.173)	0.001

Notes: Model 1: no covariates were adjusted; Model 2, gender, age and BMI were adjusted; Model 3, gender, age, BMI, hypertension, diabetes mellitus, smoking, drinking, FBG and hyperlipidemia were adjusted.

Abbreviations: NHR, neutrophil (×10⁹/L) to high-density lipoprotein cholesterol (mmol/L) ratio; PHR, platelet (×10⁹/L) to high-density lipoprotein cholesterol (mmol/L) ratio.

Based on Model 3, we employed the RCS model to assess the relationship between the NHR, PHR and EVA (Figure 1). The nonlinear components were not statistically significant for either NHR (P = 0.0648) or PHR (P = 0.4477), indicating that the associations were adequately described by linear models. Accordingly, no meaningful threshold effects or inflection points were identified.

Figure 1.

Association between the inflammatory indicators and EVA with the RCS function. (A) Association of Log NHR with EVA; (B) Association of Log PHR with EVA. The logistic regression was adjusted for gender, age, BMI, hypertension, diabetes mellitus, smoking, drinking, FBG and hyperlipidemia.

Subgroup Analysis

As shown in Figure 2, a stratified analysis by gender, BMI, hypertension, diabetes mellitus, smoking history, drinking history and hyperlipidemia further explored the relationship between inflammatory markers and EVA. For the NHR, subgroup analyses showed that the association between NHR and EVA was statistically significant in the following subgroups: male, hypertensive, smokers, hyperlipidemic, those with BMI ≥24, and those without diabetes or drinking history. Subgroup analyses revealed that the NHR–EVA association was significantly modified by gender (P for interaction = 0.018) and smoking status (P for interaction = 0.008), with stronger effects observed in males and smokers. No significant interactions were detected for other covariates (all P > 0.05). For the PHR, subgroup analyses indicated that the relationship still remained significant among the male groups, hypertension category, smoking category, those with a BMI of 24 or above, and those without a history of diabetes, drinking or hyperlipidemia. The association was not significantly modified by subgroup factors, with all interaction P values > 0.05.

Figure 2.

Association between the inflammatory indicators and EVA in various subgroups.

DCA and ROC Curve Analyses

Decision Curve Analysis revealed a moderate clinical decision value for both NHR and PHR in improving model utility (Figure 3). Specifically, the maximum net benefit increase was 0.0127 for NHR and 0.0162 for PHR. The full model showed greater net benefit than the base model across a wide threshold probability range (6.0–50.0% for NHR and 7.0–50.0% for PHR), suggesting a potential but modest incremental benefit. A receiver operating characteristics curve (ROC) analysis, which was quantified by the area under the curve (AUC), was used to assess the value of the novel indicator NHR/PHR for predicting EVA in different genders (Figure 4). In males, the AUCs for predicting EVA were 0.697 (95% CI: 0.646–0.749) for NHR and 0.698 (95% CI: 0.646–0.749) for PHR, both higher than those in females, which were 0.613 (95% CI: 0.552–0.674) and 0.616 (95% CI: 0.553–0.680), respectively. Calibration was assessed using the Hosmer‑Lemeshow test (NHR: χ² = 9.94, P = 0.269; PHR: χ² = 11.97, P = 0.152).

Figure 3.

Decision curve analysis displaying the ability of NHR and PHR to predict EVA.

Note: NHR and PHR were natural log-transformed before analysis to improve model fit.

Figure 4.

ROC curve displaying the ability of NHR and PHR to predict EVA according to gender.

Note: NHR and PHR were natural log-transformed before analysis to improve model fit.

Discussion

This study revealed that NHR and PHR were independently associated with the occurrence of EVA, with no evidence of nonlinearity in their dose-response relationships. Notably, individuals with EVA were younger in our cohort, which is consistent with the early vascular aging phenotype characterized by accelerated arterial stiffening before advanced age. The association between NHR and EVA was particularly pronounced in males and smokers, showing significant population heterogeneity. However, the clinical utility of NHR and PHR was modest, as suggested by the small net benefit gains in DCA (0.0127 for NHR and 0.0162 for PHR) and the modest discriminative performance (AUCs of approximately 0.70). Therefore, although the incremental value of NHR and PHR beyond traditional risk factors appears modest, they may still serve as supplementary biomarkers in selected subgroups (eg, males or smokers) where their association with EVA was more pronounced.

Compared with other phenotypes of vascular aging, the early vascular aging phenotype is characterized by an abnormal increase in arterial stiffness. EVA predicts cardiovascular events more accurately than traditional risk scores.¹⁵ The prevalence of EVA in this study (19.6%) was comparable to that reported in the Spanish general population (18%), though higher than the OPTIMO study (5.7%).^16,17 However, direct comparisons of prevalence across studies are strongly dependent on the diagnostic criteria used to define EVA, as well as differences in age, sex, ethnicity, and the use of age-specific percentiles for arterial stiffness parameters. Therefore, these estimates should be interpreted with caution.

Previous studies have shown a positive correlation between NHR and the severity of coronary artery disease.⁶ NHR demonstrated a sensitivity and specificity of up to 77.6% and 74.0%, respectively, for identifying severe coronary artery stenosis.¹⁸ It has also been evaluated as a predictive marker for metabolic syndrome.¹⁹ Ren et al found that NHR was independently associated with the risk of acute coronary syndrome in patients with type 2 diabetes.²⁰ Substantial evidence indicates that inflammatory and immune mechanisms, as well as dyslipidemia, are involved in the initiation and progression of atherosclerosis.²¹ Inflammatory cell infiltration has been observed at all stages of atherosclerotic disease. Neutrophils release various inflammatory mediators, such as cytokines and chemokines, which recruit additional immune cells to sites of inflammation and infection, potentially exacerbating the inflammatory response.⁷ HDL-C is well‑known for its anti‑inflammatory, antioxidant, and endothelial protective properties, playing a crucial role in cardiovascular health. Low HDL‑C levels may impair these protective functions, which according to existing literature could further promote atherosclerosis.^8,22 In the context of the present study, our findings suggest that PHR may reflect platelet activity and lipid metabolism status, and may also be associated with inflammation and oxidative stress. However, it is important to emphasize that our cross‑sectional design does not allow us to establish causal mechanisms. The observed associations are biologically plausible given that platelet activation plays a critical role in all steps of coronary artery disease.⁹ Activated platelets are involved in thrombus formation in response to atherosclerotic plaque rupture or endothelial erosion, and are believed to contribute to the development of atherothrombotic disease or adverse cardiovascular events.²³ A bivariate Mendelian randomization study showed that platelet count was positively associated with blood pressure, suggesting that platelets might exacerbate cardiovascular risk factors such as hypertension.²⁴ Based on these prior findings and our own results, we hypothesize that elevated PHR could indicate a prothrombotic and proinflammatory state, which might contribute to the development and progression of vascular changes such as EVA. However, direct evidence linking PHR to future cardiovascular events is still lacking and warrants further investigation. Nevertheless, this hypothesis requires testing in future mechanistic and prospective studies.

The results of this study showed that the EVA group had lower HDL-C levels and higher levels of neutrophil, PLT, NHR, and PHR than the non-EVA group (all P < 0.05), suggesting that lipid metabolism disorders, platelet activation, and inflammatory responses may jointly influence the progression of EVA. Even after adjusting for demographic, lifestyle, and clinical covariates, elevated NHR and PHR remained significantly associated with an increased risk of EVA. The linear dose‑response relationships of NHR and PHR with EVA suggest a progressive increase in risk without a threshold, simplifying clinical interpretation. Recent mediation analysis revealed that neutrophils mediate 13.9% of the association between cardiometabolic index and all-cause mortality,²⁵ and adding multiple inflammatory markers (including NLR, CRP, etc) to traditional risk models significantly improved the C-index from 0.702 by 0.0069,²⁶ reinforcing the incremental value of inflammation linked biomarkers. Compared with established inflammatory indices such as NLR and CRP, NHR and PHR offer a distinct advantage by integrating both inflammatory and lipid metabolic pathways—particularly relevant for atherosclerotic vascular aging, where HDL‑mediated anti‑inflammation and reverse cholesterol transport are central. However, the modest discriminative ability (AUC≈0.70) supports their role as supplementary rather than primary screening tools.^6,19 We acknowledge that no single biomarker can replace comprehensive multiparameter risk scores (eg, Framingham, SCORE).²⁷ Nevertheless, their low cost, wide availability, and lack of additional blood draw render them practical adjuncts for primary care screening, especially in resource‑limited settings where comprehensive tools are less feasible.

Subgroup analysis showed no significant interactions for PHR. However, the association between elevated NHR and increased EVA risk was stronger in males and smokers, suggesting that sex and smoking status modify this relationship. ROC analysis confirmed this pattern, with higher AUCs for both NHR (male 0.697 vs female 0.613) and PHR (male 0.698 vs female 0.616), indicating better discriminative performance in males. These differences may reflect hormonal influences (eg, estrogen-related anti-inflammatory effects) and differential inflammatory profiles, although the smaller proportion of female participants in our cohort warrants caution.²⁸ Although the AUCs of NHR and PHR were modest, they are comparable to those of established inflammatory markers.²⁶ Importantly, decision curve analysis demonstrated positive net benefits across clinically relevant thresholds, supporting the potential use of these inexpensive, readily available biomarkers as screening tools in primary care, particularly for high-risk male populations.

Clinically, identifying individuals with EVA remains a priority, which is of great significance for the prevention of cardiovascular disease. The convenience of calculating NHR and PHR from readily available laboratory tests, coupled with their potential mechanistic implications in reflecting inflammatory and lipid-related pathways, justifies further exploration of their clinical application. Further research is warranted to determine how NHR and PHR can be effectively integrated into clinical decision-making and whether they can guide personalized therapeutic interventions.

Limitations

This study has several limitations. First, the cross‑sectional design and single‑center Chinese cohort limited causal inference and generalizability. Second, although we adjusted for cardiovascular risk factors and performed E‑value analyses, residual confounding from unmeasured lifestyle factors could not be entirely excluded; internal validation via 10‑fold cross‑validation was performed, but external validation in independent cohorts remains necessary. Third, arterial stiffness was assessed using baPWV rather than the carotid‑femoral gold standard, which should be considered when interpreting our results. Future prospective multi‑center studies with diverse populations, detailed lifestyle data, and cfPWV measurements are warranted.

Conclusions

This cross‑sectional study identifies NHR and PHR as independent correlates of EVA, with the NHR‑EVA association significantly stronger in men and smokers. Despite modest AUCs (~0.70) and design limitations that preclude routine clinical use, their low cost and wide availability support their role as screening adjuncts in primary care. Prospective multi‑center studies with hard cardiovascular endpoints are warranted to validate their clinical utility.

Funding Statement

The authors declare that financial support was received for the research and publication of this article. This work is supported by the 2024 Changzhou Municipal Health Commission Science and Technology Project (QN202436), the 2024 General Project of Philosophy and Social Sciences Research in Colleges and Universities (2024SJYB0504) and the 2024 Research Project on Geriatric Health of Jiangsu Provincial Health Commission (LKM2024034).

Abbreviations

EVA, early vascular aging; NHR, neutrophil to high-density lipoprotein cholesterol ratio; PHR, platelet to high-density lipoprotein cholesterol ratio; RCS, restricted cubic spline; DCA, decision curve analysis; CVD, cardiovascular disease; PWV, pulse wave velocity; cfPWV, carotid-femoral pulse wave velocity; baPWV, brachial-ankle pulse wave velocity; BMI, body mass index; WBC, white blood count; RBC, red blood count; ALT, alanine aminotransferase; AST, aspartate aminotransferase; GGT, glutamyl transpeptidase; ALP, alkaline phosphatase; FBG, fasting venous blood glucose; TC, total cholesterol; TG, triglycerides; HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol; NLR, neutrophil-to-lymphocyte ratio; MLR, monocyte-to-lymphocyte ratio; NMLR, neutrophil-monocyte to lymphocyte ratio; MHR, monocyte to high-density lipoprotein cholesterol ratio; LHR, lymphocyte to high-density lipoprotein cholesterol ratio; SII, systemic immune-inflammation index; SIRI, systemic inflammation response index; LASSO regression, least absolute shrinkage and selection operator regression; ROC, receiver operating characteristics curve; AUC, area under the curve.

Data Sharing Statement

The data that support the findings of this study are available from the Changzhou No.1 People’s Hospital, but restrictions apply to the availability of these data, which were used under license for the current study, and so are not publicly available. Data are however available from the authors upon reasonable request and with permission of the Changzhou No.1 People’s Hospital.

Ethics Approval and Consent to Participate

This study was performed in line with the principles of the Declaration of Helsinki. Approval was granted by the Ethics Committee of the Third Affiliated Hospital of Soochow University (No. [2021]KeDi 156). All the research subjects signed the informed consent form.

Author Contributions

All authors meet the five authorship criteria of Dove Medical Press.

Ruoling Teng: Conceptualization, Writing – original draft, Funding acquisition, Investigation, Project administration, Visualization.

Huajie Han: Conceptualization, Writing – original draft, Investigation, Project administration.

Yi Ding: Conceptualization, Writing – review & editing, Supervision.

Yujiao Yang: Conceptualization, Writing – review & editing, Supervision.

Boyu Tan: Conceptualization, Writing – review & editing, Supervision.

Long Wang: Conceptualization, Writing – review & editing, Supervision.

Fenfen Liu (Corresponding author): Conceptualization, Writing – review & editing, Funding acquisition, Investigation, Project administration, Visualization.

All authors made significant intellectual contributions to the conception, design, execution, or analysis/interpretation of the work (Criterion 1); drafted or critically revised the manuscript (Criterion 2); approved the submission to this journal (Criterion 3); reviewed and agreed on all versions of the manuscript before submission and during revision (Criterion 4); and agree to take full responsibility and accountability for the integrity and accuracy of the content (Criterion 5). All authors gave final approval of the version to be published; have agreed on the journal to which the article has been submitted; and agree to be accountable for all aspects of the work.

Disclosure

The authors declare no competing interests.