White blood cell count combined with LDL cholesterol as a valuable biomarker for coronary artery disease

Abstract

Objective 

Inflammation and dyslipidemia are important pathophysiological bases for the occurrence and development of coronary artery disease (CAD); however, combination of these two entities is rarely used to diagnose CAD and its severity. Our aim was to determine whether the combination of white blood cell count (WBCC) and LDL cholesterol (LDL-C) can serve as a biomarker for CAD.

Methods 

We enrolled 518 registered patients and measured serum WBCC and LDL-C on admission. The clinical data were collected, and the Gensini score was used to assess the severity of coronary atherosclerosis.

Results 

WBCC and LDL-C levels in the CAD group were higher than in the control group (P < 0.01). Spearman correlation analysis showed that WBCC combined with LDL-C was positively correlated with the Gensini score (r = 0.708, P < 0.01) and the number of coronary artery lesions (r = 0.721, P < 0.01). Receiver operating characteristic curve analysis revealed that WBCC combined with LDL-C had a higher predictive value for CAD, severe CAD, and three-vessel CAD [area under the curve (AUC) values were 0.909, 0.867, and 0.811, respectively] than WBCC (AUC values were 0.814, 0.753, 0.716, respectively) and LDL-C (AUC values were 0.779, 0.806, 0.715, respectively) alone (all P < 0.05).

Conclusion 

WBCC combined with LDL-C is correlated with the degree of coronary artery lesion. It had high sensitivity and specificity in the diagnosis of CAD, severe CAD, and three-vessel CAD.

Introduction

Atherosclerosis is a chronic systemic inflammatory disease ignited by the accumulation of lipids and inflammatory immune cells [¹]. It is the underlying pathological process of coronary artery disease (CAD). The early clinical manifestations of CAD are insidious, with serious complications and extremely poor prognosis [²]. Although the diagnostic strategy for CAD has made continuous progress in recent years, it still depends on coronary artery imaging examination in the clinic, which is expensive and technical with the risk of contrast agent allergy [³]. Considering circulating biomarkers to reflect CAD and its severity easily and efficiently will be a new trend.

Previous studies have shown that increased white blood cell count (WBCC) is associated with the development of coronary atherosclerosis and CAD events [⁴]. A large American clinical study followed white men without cardiovascular disease for 15 years and found that participants with increased WBCC were at a higher risk for developing CAD [⁵]. In addition, patients with severe coronary stenosis had significantly higher WBCCs than patients with normal coronary arteries [⁶]. WBCC is both a validated biomarker of the extent of the CAD inflammatory process and a key player in the development of atherosclerotic ischemic heart disease that subsequently manifests as myocardial infarction [⁷].

In addition to inflammation, dyslipidemia is another important pathophysiological basis for atherosclerotic events. LDL cholesterol (LDL-C) is involved in the occurrence and development of atherosclerosis, and elevated LDL-C is an independent risk factor for atherosclerotic cardiovascular disease [⁸].

Although there have been extensive studies on the relationship between inflammation, lipid metabolism, and coronary atherosclerosis, there are few reports on the diagnostic value of WBCC combined with LDL-C in CAD. WBCC combined with LDL-C may better reflect inflammatory factors and lipid metabolism disorders that develop over the course of CAD. In this study, we explored the correlation between WBCC combined with LDL-C and coronary artery lesions, as well as further evaluated its diagnostic value for CAD and its severity.

Methods

Study population

We enrolled consecutive patients 518 patients who were hospitalized in the Department of Cardiology of Haian Hospital Affiliated to Nantong University between January 2019 and December 2021. The inclusion criteria were patients over 18 years old who were admitted to the hospital due to chest pain and underwent coronary angiography. The exclusion criteria were heart failure (New York Heart Association class III–IV), acute myocarditis, cardiomyopathy, mental illness, pulmonary heart disease, pericardial effusion, aortic dissection, or heart valve disease; acute abdomen, cerebral hemorrhage, acute cerebral infarction, or coma; severe liver or kidney insufficiency, thyroid disease, infection, immune disease, connective tissue disease, or malignant tumor; and use of oral statins and other lipid-lowering drugs. The study has been approved by the Ethics Committee of Haian Hospital Affiliated to Nantong University (Approval No. HKL201837), and all patients signed informed consent forms.

Study groups

According to the coronary angiography guidelines of the American Heart Association and the American College of Cardiology (AHA/ACC) [⁹], the left and right coronary arteries were assessed via the Judkins method. At least four conventional projection positions were selected for the left coronary artery and at least two conventional projection positions were selected for the right coronary artery to fully visualize the segments of the left and right coronary arteries, other projection positions were added if necessary. Coronary artery stenosis ≥ 50% of any coronary artery trunk or branch diameter ≥ 1.5 mm was diagnosed as CAD, as prescribed by the AHA/ACC [⁹]. The angiographic results were interpreted by three senior deputy chief cardiologists and above and were divided into a CAD group or a normal control group according to the angiography results.

Gensini score

Gensini score was used to assess the severity of CAD. According to the degree of coronary artery stenosis, the score for no coronary stenosis was 0, <25% stenosis was 1, 25–49% stenosis was 2, 50–74% stenosis was 4, 75–89% stenosis was 8, 90–99% stenosis was 16, and 100% stenosis was 32. The determination coefficient was 5 in the left main coronary artery, 2.5, 1.5, and 1 in the proximal, middle, and distal segments of the anterior descending branch, respectively, 1.0 and 0.5 in the first and second diagonal branches, respectively, 2.5 and 1.0 in the proximal and distal segments of the circumflex branch, respectively, 1.0 in the blunt marginal branch and posterior descending branch, and 1.0 in the proximal, middle, and distal right coronary arteries. The coronary lesion score was the product of the coronary stenosis score and the corresponding lesion location coefficient, and the sum of each branch coronary lesion score was the Gensini score [¹⁰]. A higher Gensini score indicated more serious CAD. According to the Gensini score, the CAD group was divided into two subgroups: mild (1–30) and severe (>30) [¹¹].

Data collection

General data were collected on admission, including age, sex, history of smoking, history of hypertension and type 2 diabetes, and left ventricular ejection fraction (LVEF). Fasting venous blood was collected in the early morning after admission and was used to assess routine blood parameters, liver and kidney function, and blood lipids with an automatic biochemical instrument (Beckman, Brea, California, USA). WBCC, monocyte count, neutrophil count, lymphocyte count, high sensitivity C-reactive protein (hs-CRP), total cholesterol (TC), total triglyceride, LDL-C, HDL cholesterol (HDL-C), fasting blood glucose (FBG), creatinine, alanine aminotransferase (ALT), and aspartate aminotransferase (AST) were recorded.

Statistical analysis

SPSS 25.0 (IBM Corporation, Armonk, New York, USA) was used for statistical analysis. If the measurement data did not obey the normal distribution as the median (quartile spacing), the two-sample rank sum test was used for comparison between the two groups; if the measurement data obeyed the normal distribution as the mean ± SD, the independent samples t-test was used for comparison between two groups. Counting data are expressed as examples and percentages, and the chi-square test was used for comparison between groups. After binary classification, the continuous variables in the patient data were included in univariate logistic regression analysis. According to the standard of P < 0.2, the relevant influencing factors were selected for multivariate logistic binary regression analysis, and the independent risk factors of CAD and severe CAD were obtained. Spearman correlation analysis was used to evaluate the correlation between WBCC combined with LDL-C and coronary artery lesions. We used SPSS 25.0 to generate the receiver operating characteristic (ROC) curve of WBCC, LDL-C, and their combination to predict CAD, severe CAD, and three-vessel CAD and compared the area under the curve (AUC). P < 0.05 was considered statistically significant.

Results

Patient description

Of the 518 patients, 304 were male and 214 were female, aged between 31 and 95 years old, with an average age of 63.40 (55.00, 72.00) years. There were 265 patients with normal coronary arteries (131 males and 134 females) and 253 patients with CAD (173 males and 80 females). In the CAD group, there were 106 cases of one-vessel disease, 84 cases of two-vessel disease, and 63 cases of three-vessel disease. Compared with the control group, patients in the CAD group were older, had a higher frequency of hypertension, type 2 diabetes, and smoking, and biochemical analysis revealed higher serum WBCC, neutrophil count, lymphocyte count, monocyte count, creatinine, ALT, AST, FBG, total triglycerides, TC, LDL-C, hs-CRP, and Gensini score, while HDL-C level and LVEF were lower (P < 0.05, Table 1).

Table 1 - Comparison of clinical data between control group and coronary artery disease group

Variables	Control group (n = 265)	CAD group (n = 253)	Statistic	P value
Age	61.67 ± 12.30	65.50 ± 12.45	−3.530	<0.001
Sex, male (%)	131 (49.4)	173 (68.4)	19.161	<0.001
Hypertension, case (%)	77 (29.1)	169 (66.8)	73.933	<0.001
Smoking, case (%)	89 (33.6)	110 (43.5)	5.355	0.021
Diabetes, case (%)	23 (8.7)	74 (29.2)	35.983	<0.001
WBC/(L−1, ×109)	7.04 (6.63, 7.38)	10.13 (7.76, 13.07)	−12.350	<0.001
Neutrophils/(L−1, ×109)	5.76 (5.35, 6.11)	7.86 (6.01, 10.09)	−10.628	<0.001
Lymphocytes/(L−1, ×109	1.49 (1.34, 1.61)	1.64 (1.26, 1.96)	−4.172	<0.001
Monocyte/(L−1, ×109)	0.51 (0.44, 0.57)	0.69 (0.49, 0.95)	−9.316	<0.001
Creatinine/(μmol•L−1)	67.80 (58.05, 81.40)	80.85 (66.33, 99.65)	−5.887	<0.001
ALT/(U•L)	17.40 (12.65, 22.85)	29.80 (14.40, 50.87)	−6.771	<0.001
AST/(U•L)	19.30 (15.50, 26.73)	43.40 (19.18, 112.93)	−9.137	<0.001
FBG/(mmol•L−1)	6.34 (5.11, 8.12)	7.73 (5.91, 9.48)	−3.054	0.002
Triglycerides/(mmol•L−1)	4.17 (3.45, 4.89)	4.31 (3.61, 4.90)	−1.409	0.159
TC/(mmol•L−1)	1.35 (1.00, 1.94)	1.52 (1.14, 1.99)	−2.388	0.017
LDL-C/(mmol•L−1)	1.26 (1.04, 1.66)	2.45 (1.44, 3.24)	−10.998	<0.001
HDL-C/(mmol•L−1)	2.34 (1.48, 3.01)	1.18 (1.00, 1.73)	−9.283	<0.001
Hs-CRP/(g•L−1)	0.80 (0.10, 3.40)	3.10 (0.65, 11.99)	−5.774	<0.001
LVEF (%)	56.12 (49.69, 61.25)	49.00 (40.19, 57.98)	−6.745	<0.001
Gensini score	8.73 (5.66, 11.87)	43.20 (29.31, 60.41)	−18.037	<0.001

ALT, alanine aminotransferase; AST, aspartate aminotransferase; CAD, coronary artery disease; CHD, coronary heart disease; FBG, fasting blood glucose; HDL-C, HDL cholesterol; hs-CRP, high sensitivity C-reactive protein; LDL-C, LDL cholesterol; LVEF, left ventricular ejection fraction; TC, total cholesterol; WBC, white blood cell.

Risk factors for coronary artery disease

The factors with P < 0.2 between the control group and the CAD group were included in univariate logistic regression analysis. The results showed that sex, age, hypertension, smoking, type 2 diabetes, WBCC, neutrophil count, lymphocyte count, monocyte count, creatinine, ALT, AST, FBG, LDL-C, and hs-CRP were risk factors for CAD [odds ratio (OR) = 1.016–146.450, P < 0.05]. The factors of univariate logistic regression analysis were included in the multivariate logistic binary regression analysis to identify the factors that independently influenced CAD. Multivariate logistic regression analysis showed that LDL-C [OR = 4.018; 95% confidence interval (CI), 1.894–8.525; P < 0.01], ALT (OR = 1.065; 95% CI, 1.019–1.112; P < 0.01), and hypertension (OR = 7.178; 95% CI, 2.704–19.057; P < 0.01) were independent risk factors for CAD (Table 2). The binary logistic regression model provided the following estimation of the logit function:

Table 2 - Univariate and multivariate binary logistic regression analysis between coronary artery disease and different factors

Variables	Univariate			Multivariate
	OR	95% CI	P value	OR	95% CI	P value
Age	1.026	1.011–1.040	0.001
Sex, male (%)	2.212	1.546–3.164	<0.001
Hypertension, case (%)	4.912	3.384–7.130	<0.001	7.178	32.704–19.057	<0.001
Smoking, case (%)	1.521	1.065–2.172	0.021
Diabetes, case (%)	4.350	2.622–7.216	<0.001
WBC/(L−1, ×109)	1.939	1.694–2.218	<0.001
Neutrophils/(L−1, ×109)	1.847	1.620–2.106	<0.001
Lymphocytes/(L−1, ×109	2.192	1.407–3.416	0.001
Monocyte/(L−1, ×109)	146.45	47.99–446.83	<0.001
Creatinine/(μmol•L−1)	1.016	1.009–1.022	<0.001
ALT/(U•L)	1.044	1.032–1.057	<0.001	1.065	1.019–1.112	<0.001
AST/(U•L)	1.032	1.024–1.040	<0.001
FBG/(mmol•L−1)	1.677	1.496–1.881	<0.001
LDL-C/(mmol•L−1)	4.101	3.120–5.390	<0.001	4.018	1.894–8.525	<0.001
hs-CRP/(g•L−1)	1.047	1.026–1.069	<0.001

ALT, alanine aminotransferase; AST, aspartate aminotransferase; CI, confidence interval; FBG, fasting blood glucose; hs-CRP, high sensitivity C-reactive protein; LDL-C, LDL cholesterol; OR, odds ratio; WBC, white blood cell.

$$\text{Logit}\left(\text{p}\right)=-3.863+1.951\times \text{ hypertension }+0.947\times \text{ LDL-C }+0.069\times \text{ ALT}$$

The model discrimination test revealed that the area under the ROC curve was 0.829 (95% CI, 0.765–0.856; P < 0.01), which was >0.75, indicating that the prediction model has good discrimination.

Risk factors for severe coronary artery disease

The factors with P < 0.2 between mild and severe groups were included in univariate logistic regression analysis. The results showed that sex, age, hypertension, smoking, type 2 diabetes, WBCC, neutrophil count, lymphocyte count, monocyte count, creatinine, AST, ALT, FBG, LDL-C, and hs-CRP were risk factors for severe CAD (OR = 1.020–18.989, P < 0.05). The factors of univariate logistic regression analysis were included in the multivariate logistic binary regression analysis. The results showed that LDL-C (OR = 4.256; 95% CI, 2.958–6.123; P < 0.01), AST (OR = 1.007; 95% CI, 1.002–1.012; P < 0.05), and monocyte count (OR = 42.362; 95% CI, 14.985–119.756; P < 0.05) were independent risk factors for severe CAD. The binary logistic regression model provided the following estimation of the logit function:

$$\text{Logit}\left(\text{p}\right)=-6.594+1.445\times \text{ LDL-C }+0.007\times \text{ AST }+3.491\times \text{ monocyte count}$$

The model discrimination test revealed that the area under the ROC curve was 0.906 (95% CI, 0.881–0.931; P < 0.01), which was >0.75, indicating that the prediction model has good discrimination.

Univariate logistic regression analysis was used to compare the factors between coronary artery lesions with less than three branches and more than three branches (P < 0.2). The results showed that sex, age, hypertension, type 2 diabetes, WBCC, neutrophil count, lymphocyte count, monocyte count, AST, ALT, FBG, TC, total triglycerides, LDL-C, and hs-CRP were risk factors for three-vessel CAD (OR = 1.004–22.724, P < 0.05). The factors of univariate logistic regression analysis were included in the multivariate logistic binary regression analysis. The results showed that LDL-C (OR = 1.758; 95% CI, 1.067–2.896; P < 0.05), hs-CRP (OR = 1.034; 95% CI, 1.014–1.054; P < 0.01), hypertension (OR = 2.673; 95% CI, 1.096–6.520; P < 0.05), type 2 diabetes (OR = 3.077; 95% CI, 1.280–7.395; P < 0.05), and monocyte count (OR = 7.499; 95% CI, 1.192–47.158; P < 0.05) were independent risk factors for three-vessel CAD. The binary logistic regression model provided the following estimation of the logit function:

$$\begin{array}{l}\text{Logit}\left(\text{p}\right)=-9.620+0.655\times \text{ LDL-C}\\ +0.027\times \text{ hs-CRP}\\ +1.121\times \text{ hypertension }\\ +2.026\times \text{ type }2\text{ diabetes}\\ +3.545\times \text{ monocyte count}\end{array}$$

The model discrimination test revealed that the area under the ROC curve was 0.894 (95% CI, 0.858–0.929; P < 0.01), which was > 0.75, indicating that the prediction model has good discrimination.

Predictive value of white blood cell count, LDL-cholesterol, and their combination in coronary artery disease and its severity

The ROC curve was used to measure the predictive value of WBCC, LDL-C, and combined WBCC and LDL-C in predicting CAD and severe CAD. The binary logistic regression model provided the following estimation of the logit function:

$$\text{Logit}\left(\text{p}\right)=-8.188+0.613\times \text{ WBCC }+1.664\times \text{ LDL-C}$$

When the critical value of WBCC combined with LDL-C was 0.588, the sensitivity and specificity were 0.791 and 0.928, respectively. The AUCs of the three groups were 0.814 for WBCC, 0.779 for LDL-C, and 0.909 for combined WBCC and LDL-C (Fig. 1a). When the critical value of WBCC combined with LDL-C was 0.588, the predictive efficiency of severe CAD was the highest; the sensitivity was 0.839, and the specificity was 0.825. The AUCs of the three groups were 0.753 for WBCC, 0.806 for LDL-C, and 0.867 for combined WBCC and LDL-C (Fig. 1b).

Fig. 1: Predictive value of WBCC, LDL-C, and their combination in CAD and its severity. (a) ROC curve of WBCC, LDL-C, and their combination for predicting CAD. (b) ROC curve of WBCC, LDL-C, and their combination for predicting severe CAD. (c) ROC curve of WBCC, LDL-C, and their combination for predicting three-vessel CAD. AUC, area under the curve; CAD, coronary artery disease; LDL-C, LDL cholesterol; ROC, receiver operating characteristic; WBCC, white blood cell count.

Correlation between white blood cell count combined with LDL-cholesterol and Gensini score

Spearman correlation analysis was performed on WBCC combined with LDL-C and the Gensini score for 518 patients, and the two were positively correlated (r = 0.708, P < 0.01), indicating that higher values of WBCC combined with LDL-C correlated with more serious CAD.

Relationship between white blood cell count combined with LDL-cholesterol and the number of coronary vascular lesions

Spearman correlation analysis was performed on WBCC combined with LDL-C and the number of coronary vascular lesions in 518 patients, and the two were positively correlated (r = 0.721, P < 0.01), indicating that higher values of WBCC combined with LDL-C correlated with more numerous coronary vascular lesions. The ROC curve was used to measure the predictive value of WBCC, LDL-C, and their combination on the number of coronary vascular lesions. When the critical value of WBCC combined with LDL-C was 0.969, the sensitivity and specificity were 0.556 and 0.901, respectively. The AUCs of the three groups were 0.716 for WBCC, 0.715 for LDL-C, and 0.811 for combined WBCC and LDL-C (Fig. 1c).

Discussion

Inflammation and dyslipidemia are important factors for the occurrence and development of CAD [¹²]. This study combined blood lipids and inflammation to explore the diagnostic value of WBCC combined with LDL-C in CAD. By comparing the AUCs of WBCC, LDL-C, and WBCC combined with LDL-C in predicting CAD, severe CAD, and three-vessel CAD, the superiority of WBCC combined with LDL-C was observed. Spearman correlation analysis showed that higher values of WBCC combined with LDL-C were correlated with more severe CAD. In addition, the value of WBCC combined with LDL-C was positively correlated with the number of vascular lesions, and the value of WBCC combined with LDL-C in the group with three or more coronary artery involvement was higher than that in the other groups. The results of multivariate logistic binary regression analysis showed that LDL-C was an independent risk factor for CAD, severe CAD, and three-vessel CAD, which was consistent with previous reports [¹³]. Leukocytosis can affect CAD through multiple pathological mechanisms and has been reported to be an independent risk factor and prognostic indicator for cardiovascular outcomes [¹⁴]. Among blood lipid parameters, LDL-C has a significant correlation with CAD and is an independent risk factor for CAD [¹⁵]. Our study showed that the diagnostic performance of WBCC combined with LDL-C in CAD, severe CAD, and CAD with three or more vessels was superior to WBCC or LDL-C alone, which indicates that WBCC combined with LDL-C is better than single LDL-C.

WBCC is a widely used inflammatory marker in clinical practice in addition to granulocyte count, lymphocyte count, and monocyte count. The pathogenesis of CAD is complex and involves various types of inflammatory cells. After vascular endothelial injury, monocytes are recruited from peripheral blood to the vascular wall, differentiate into macrophages, phagocytose lipids, secrete metalloproteinases [^16–18], and secrete or induce neutrophils and mast cells to accumulate in plaques [^19–21]. Over time, the recruitment and accumulation of inflammatory cells increase the lipid and inflammatory cell content of plaques, and by mediating inflammation, causes proteolytic and oxidative damage to endothelial cells [²²], clogging the microcirculatory system, inducing a state of hypercoagulability, and promoting further ischemia and infarction [²³]. Multiple clinical studies have shown that high WBCC is associated with increased CAD-related morbidity and mortality and is an independent risk factor for atherosclerosis [^24–26]. Leukocyte count is widely used in clinical practice and is easy to obtain, it is a promising indicator in the diagnosis of CAD; however, the low specificity of leukocytes limits its application. Our study combined WBCC with LDL-C, which helped to increase its diagnostic specificity for CAD.

Numerous laboratory and clinical studies have shown that elevated LDL-C is the main risk factor for the occurrence of atherosclerosis [²⁷]. Furthermore, it is involved in the process of atherosclerosis, which can be confirmed at the following three levels: first, studies at the genetic level have found that LDL-C receptor gene mutation leads to the formation of acute atherosclerosis plaques; second, animal experiments have found that upregulation of LDL-C in non-CAD animals lead to the formation of CAD; and finally, epidemiological data in the population have shown that, as LDL-C increase, the prevalence of CAD increases [^28–30]. In addition, the clinical application of reducing LDL-C as a strategy can significantly improve the outcome and prognosis of CAD, which further supports the correlation between LDL-C and CAD [^27,31,32].

In summary, the literature indicates that blood lipid indexes and inflammatory indexes are related to the diagnosis of CAD; however, few studies have combined these indexes to diagnose CAD and assess the severity of CAD. In this study, inflammatory and lipid abnormalities involved in the pathophysiological mechanism of atherosclerosis were combined, and we found that the combination of WBCC and LDL-C had diagnostic value for CAD. Higher values of WBCC combined with LDL-C were associated with more coronary artery involvement and increased disease severity in patients with CAD. Because blood lipid levels and WBCC are relatively easy to obtain in medical institutions at all levels, their combination can indicate risk status or the need for further investigation in clinical practice, especially in hospitals that do not have the resources to perform coronary angiography. Identifying patients at high risk of CAD to modify drug treatment or transfer them to higher-level hospitals for further diagnosis and treatment is essential to reduce the morbidity and mortality of CAD.

This study has some limitations. First, the blood lipid and WBCC values collected in this study were all collected in a single preoperative blood collection, and no dynamic monitoring was performed. Second, while the Gensini score is the most used scoring system for evaluating coronary lesions, it cannot reflect special conditions, such as branch lesions, calcified lesions, and twisted lesions. Third, the population of this study was highly selected, and the sample size was relatively small; therefore, some clear risk factors for CAD, such as smoking, were not identified. Finally, this study only discussed the diagnostic value of WBCC combined with LDL-C for CAD, and further research is needed on its value in long-term prognosis and prediction of adverse events in CAD.

Conclusion

WBCC combined with LDL-C has predictive value for CAD, severe CAD, and three-vessel CAD. It can provide clinicians with a new perspective to identify patients with high-risk coronary artery lesions as early as possible and to select the best treatment for patients to reduce the risk of cardiovascular events.

Acknowledgements

H. D. was responsible for designing the experiments. Z. L., Y. Y., and S. G. performed the experiments. Y.L. and H.H. collected data. All authors contributed to manuscript preparation and revision. The study has been approved by the Ethics Committee of Haian Hospital Affiliated to Nantong University (Approval No. HKL201837), and all patients signed informed consent forms. The data are available upon request.

Conflicts of interest

There are no conflicts of interest.