Introduction

Cardiovascular diseases (CVD) remained the leading cause of death globally in 2021, responsible for nearly one-third of all fatalities1. The 2023 American Heart Association (AHA) report shows that global CVD mortality rose by 18.7% in 2020 compared to 20102. Given its profound impact on public health and the global economy, early identification and prediction of CVD are essential.

Although conventional risk factors including diabetes, hypertension, and hyperlipidemia are strong predictors of cardiovascular risk, they may not fully account for the variability in CVD risks3,4,5. Recent studies suggest that inflammation and nutritional status are key in atherosclerosis and CVD progression6,7,8.

The neutrophil percentage to albumin ratio (NPAR) is an emerging biomarker that integrates two critical factors: neutrophil percentage, an indicator of systemic inflammation, and albumin, a marker of nutritional status. Neutrophils are a key component of the immune response, and their involvement in inflammation has been linked to coronary artery disease9. Low albumin levels, on the other hand, are associated with poor cardiovascular outcomes and overall mortality10,11. Combining these two markers, NPAR provides a more comprehensive assessment of inflammation and nutrition, integral to cardiovascular health.

Previous studies have linked NPAR with adverse outcomes such as COPD, cancer, and acute kidney injury12,13,14. In addition, NPAR has been shown to outperform either neutrophil percentage or albumin alone in predicting mortality in patients with cardiogenic shock15. However, there is limited research on the association between NPAR and CVD in the overall adults of the United States.

This research aims to evaluate the relationship between NPAR and CVD prevalence using information from the National Health and Nutrition Examination Survey (NHANES). By exploring NPAR’s potential as a predictive biomarker, this research seeks to provide insights into more effective strategies for early CVD prevention.

Methods

Study population

For intricate sample weight computations, we adhered to NHANES rules to guarantee reliable and representative statistical analysis.

Our study analyzed information from 45,462 NHANES individuals from 2011 to 2020. We excluded 19,182 individuals under 20 years of age, 5 individuals with inadequate cardiovascular disease questionnaire data, and 50 individuals with missing education or smoking status. In total, 26,225 individuals were included in the final analysis.

Figure 1 presents a detailed flow chart illustrating the participant selection process for the study.

Fig. 1
figure 1

The process flow diagram for the systematic selection method. NPAR neutrophil-to-albumin ratio, CVD cardiovascular disease, CI confidence interval, OR odds ratio.

Full size image

Definition of the NPAR and CVD

The neutrophil percentage was obtained using the NHANES CBC Profile. The NHANES Standard Biochemistry Profile was used to calculate the albumin value. The dye bromcresol purple (BCP) is used in the albumin concentration measurement technique. The dye changes color at 600 nm when it binds to albumin preferentially in the pH range of 5.2–6.8. 700 nm is the secondary wavelength. This is an albumin-specific, two-point endpoint reaction.

NPAR is calculated by the formula: Neutrophil percentage (%) × 100/Albumin (g/dL).

The survey questionnaire responses from the participants in this study were used for defining CVD. If a participant reported having angina, myocardial infarction (MI), congestive heart failure (CHF), coronary heart disease (CHD), or stroke, a diagnosis of CVD was taken into consideration. As a result, any positive response to these particular requirements was taken to mean that CVD was present16,17.

Covariates

This study examined a range of variables and clinical conditions. Three levels of education were distinguished: underhigh school, secondary school or its equivalent, and colleges or higher. BMI was classified as ordinary (< 25 kg/m2), overweight (25–29.9 kg/m2), and obese (≥ 30 kg/m2). The family income-to-poverty ratio (PIR) was divided into three groups: poor, intermediate, and wealthy18. eGFR was calculated using creatinine-based equations19. Excessive alcohol consumption was defined as ≥ 4 drinks/day for men and ≥ 3 drinks/day for women. Moderate drinking was 3 drinks/day for men and 2 drinks/day for women, while lower intake was classified as minimal (< 3 drinks/day for men, < 2 drinks/day for women)20. The criterion for determining smoking status was a history of smoking more than 100 cigarettes in a year21. Total cholesterol > 200 mg/dL, triglycerides ≥ 150 mg/dL, high-density lipoprotein (HDL) ≤ 40 mg/dL in males and < 50 mg/dL in females, or low-density lipoprotein (LDL) ≥ 130 mg/dL were considered hyperlipidemia22,23. Hypertension was defined as an average systolic blood pressure (SBP) ≥ 140 mmHg, diastolic blood pressure (DBP) ≥ 90 mmHg, use of antihypertensive medication, or a documented history of hypertension. Diabetes was diagnosed if fasting plasma glucose was ≥ 126 mg/dL, glycated hemoglobin (HbA1c) was ≥ 6.5%, or if the participant used insulin or hypoglycemic medications.

Statistical analysis

Constant variables were presented as means with 95% confidence intervals (CIs), and classified variables as proportions with 95% CIs. For the study of baseline characteristics, individuals were separated into NPAR quartiles. Weighted generalized linear models were used to evaluate the association between NPAR and CVD. Model 1 was unadjusted, while Model 2 adjusted for demographic factors such as sex, age, and ethnicity. Model 3 further adjusted for sex, age, ethnicity, BMI, PIR, education, habit of smoking, drinking habits, hypertension, hyperlipidemia, and diabetes. Model 4 was adjusted for sex, age, ethnicity, BMI, PIR, education, habit of smoking, drinking habits, hypertension, LDL-C, and diabetes. Restricted cubic splines were used to analyze nonlinear associations, threshold effects investigation of inflection points and the association of NPAR and CVD. To assess the robustness of the NPAR-CVD association, weighted subgroup analyses were conducted, with unweighted logistic regression used for validation. Model discrimination was evaluated using receiver operating characteristic (ROC) curves, and area under curves (AUCs) comparisons were made with the DeLong test. All statistical analyses were conducted in R Studio (v4.2.2) and Empower Stats (v4.1), with statistical significance set at a two-sided P-value < 0.05.

Results

Characteristics of individuals

All of the 26,225 participants in this research, 51.59% were female and 48.41% were male, with an average age of 49.9 ± 17.7 years. The median NPAR was 1372.7, and the prevalence of CVD was 11.4%. Individuals were grouped into quartiles depending on NPAR. A noteworthy pattern revealed rising CVD prevalence throughout NPAR quartiles. (P < 0.001). Important differences in Demographic, lifestyle, and clinical factors were noted among the quartiles (P < 0.05) (Table 1).

Table 1 Characteristics of study participants by quartiles of NPAR.
Full size table

Association of the NPAR with CVD

Table 2 demonstrates a strong positive association between CVD and NPAR. In every model, the results were statistically significant (P < 0.05), this association holds consistent whether NPAR is viewed as a constant variable or as a classification variable. The CVD prevalence was consistently greater in the fourth NPAR quartile than in the first in all models. The CVD prevalence in the fourth quartile of the fully adjusted Model 3 was 46% greater than that in the first quartile [OR 1.46, 95 %CI (1.16, 1.83), P = 0.002]. Even after replacing hyperlipidemia with LDL-C in model 4, cardiovascular disease prevalence remained 40% higher in the fourth quartile of NPAR compared to the first quartile [1.40 (1.12, 1.75), P = 0.005]. Additionally, every model’s trend test has statistical significance (P < 0.05).

Table 2 The association between weighted NPAR and CVD.
Full size table

Furthermore, a nonlinear association between NPAR and CVD prevalence is depicted in Fig. 2.

Fig. 2
figure 2

The nonlinear association between CVD and NPAR. NPAR neutrophil-to-albumin ratio, CVD cardiovascular disease, CI confidence interval, OR odds ratio.

Full size image

Subgroup investigation of the association between the NPAR and CVD

A subgroup analysis based on clinical, lifestyle, and demographic variables was planned to evaluate the association between NPAR and CVD.

Interaction analyses revealed statistically significant differences in NPAR across subgroups defined by PIR and alcohol consumption (P < 0.05). In contrast, other factors did not significantly modify the positive association between NPAR and CVD (P > 0.05) (Table 3).

Table 3 Subgroup analysis of the association between weighted NPAR and CVD.
Full size table

Sensitivity analysis

The positive association between NPAR and the prevalence of CVD in all models is demonstrated by the sensitivity analysis employing unweighted logistic models in Table 4 (P < 0.05). The outcomes in Table 2 agree with these findings.

Table 4 The association between NPAR and CVD in sensitivity analysis using unweighted logistic regression analysis.
Full size table

Log-likelihood ratios indicated no significant improvement in model fit (Table S1).

Incremental value of the NPAR in predicting CVD

Figure 3 evaluates the added predictive value of albumin, neutrophil percentage, and NPAR to a baseline model with conventional CVD risk factors (age, sex, hypertension, diabetes, hyperlipidemia). Adding NPAR increased the AUC from 0.825 to 0.830, though there was no statistically significant difference. (P = 0.349)

Fig. 3
figure 3

Comparison of ROC curve areas for cardiovascular disease prediction models. Baseline models include sex, age, hypertension, diabetes, and hyperlipidemia. ROC receiver operating characteristic curve, AUC area under curve.

Full size image

Discussion

To our knowledge, this is the first research to investigate the association between cardiovascular disease prevalence and the novel biomarker NPAR. In this study, which used data from 26,225 participants in the NHANE database from 2011 to 2020, high NPAR was associated with a significant increase in cardiovascular prevalence even as a classification and constant variable (P < 0.05) after basic clinical, lifestyle, and demographic variables were taken into account. The stability of our results was further verified by the sensitivity analysis results, which were identical to the weighted results (P < 0.05).

Neutrophils are involved in every stage of the atherosclerotic process, they promote platelet adhesion, activate macrophages and endothelial cells, increase monocyte recruitment, and start the plaque formation process9,24. Furthermore, risk factors like obesity and diabetes raise the risk of cardiovascular disease by boosting the inflammatory response25,26. Research indicates that hypoalbuminemia is a separate risk factor for CVD. Albumin can sustain plasma colloid osmotic pressure and has anti-inflammatory, antioxidant, and anticoagulant properties10,27. NPAR is a new biomarker that combines albumin and neutrophil percentage. It is derived from routine blood and biochemical tests and is easy to use and repeat. Compared to albumin and neutrophil percentage alone, NPAR has a synergistic amplification effect that enhances the ability to evaluate a patient’s risk for cardiovascular disease. According to a study, NPAR is a separate risk factor for stroke-associated pneumonia and spontaneous intracerebral hemorrhage28. A retrospective cohort analysis revealed that among patients undergoing peritoneal dialysis, the risk of cardiovascular death was 1.57 times higher and the possibility of death of all causes was 1.51 times higher in the highest NPAR quartile than in the lowest quartile29. A follow-up study of 1,141 atrial fibrillation patients older than 80 years found a positive relationship between NPAR and the rates of cardiovascular and all-cause death after 28 days30.

Nonetheless, little study has examined the association between NPAR and CVD among the broader populace. Our results demonstrate that the association between cardiovascular disease prevalence and increasing NPAR is stable across various subgroup analyses, indicating that these factors do not influence the observed association (P > 0.05). In contrast, results from the PIR and drinking subgroups suggest that both factors may affect the NPAR-CVD association. A low-income poverty ratio was significantly associated with higher CVD prevalence, while the middle-income group also showed increased prevalence. No significant association was found for the high-income group, indicating that low-income individuals may be more vulnerable to adverse socioeconomic factors such as limited healthcare access and unhealthy lifestyles, increasing their CVD risk31. Thus, future interventions should target reducing the socioeconomic impact of CVD, particularly in low-income populations. Additionally, moderate alcohol consumption was significantly associated with increased CVD prevalence, while light and heavy drinking showed no significant link. This may be due to the cardiovascular effects of exceeding moderate alcohol intake, such as acute hypertension, arrhythmias, and chronic inflammation32. Additionally, drinking patterns, frequency, and unadjusted confounders could influence these outcomes. Light drinking may not reach the threshold for harm, and the sample size for heavy drinkers may have been insufficient to detect an effect. Future studies should explore the long-term impact of different drinking behaviors on CVD risks. And, threshold effect analysis indicated that, although a significant association was observed below the point, the log-likelihood ratio test (P > 0.05) suggests that the nonlinear effect between NPAR and CVD may be limited to specific subgroups or NPAR levels. Future studies should explore the underlying mechanisms between varying NPAR levels and CVD.

Among the many advantages of our study is that it is the first to investigate the association between NPAR and CVD in the general public. Second, our model fully accounts for any confounders. Third, we utilized information from the NHANES database, which includes information from 26,255 participants, strengthening the robustness of our findings. Nevertheless, there are several restrictions in our research. First, as a cross-sectional study, it is difficult to determine the cause relationships between NPAR and CVD, and prospective studies are needed to validate these findings. Second, NPAR changes could not be tracked over time, and relying on a single hematologic measurement may introduce errors. Additionally, CVD diagnoses were based on self-report questionnaires, which may lead to bias and limit the accuracy of disease classification. Furthermore, we lacked key systemic inflammation markers (e.g., CRP, ESR) and cardiovascular biomarkers (e.g., troponin, natriuretic peptides), which may affect the precision of the NPAR-CVD association. Future research should include more comprehensive diagnostic tools and biomarkers to confirm these results.

Conclusion

In conclusion, our study confirms the positive association between NPAR and CVD prevalence in the general public. Our research emphasizes the significance of the association between NPAR and CVD prevalence and points out its potential as a new biomarker for assessing CVD risk.