Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Open Access

Peer-reviewed

Research Article

Associations between leukocyte count and lipid-related indices: Effect of age and confounding by habits of smoking and alcohol drinking

Abstract

Leukocyte count in peripheral blood is an acute-phase reactant and is associated with the risk of atherosclerotic diseases. Blood lipid profile, a major risk factor of cardiovascular disease, is known to be associated with leukocyte count, but it remains to be determined how this association is affected by other factors including lifestyle and age. The subjects were 11261 Japanese middle-aged men (30~65 years old) who had received health checkup examinations. The relationships of leukocyte count with lipid-related indices (ratio of LDL cholesterol to HDL cholesterol [LDL-C/HDL-C], ratio of triglycerides to HDL cholesterol [TG/HDL-C] and cardiometabolic index [CMI]) were investigated. Leukocyte count, LDL-C/HDL-C, TG/HDL-C and CMI were significantly higher in smokers than in nonsmokers, while leukocyte count and LDL-C/HDL-C were significantly lower in regular drinkers than in nondrinkers. Both in overall subjects and subjects without habits of smoking and drinking, LDL-C/HDL-C, TG/HDL-C and CMI were significantly higher in the 2nd and 3rd tertiles for leukocyte count than in the 1st tertile and tended to be higher with an increase of the tertile. Odds ratios for high TG/HDL-C and high CMI of the 3rd vs. 1st tertiles for leukocyte count tended to be lower with an increase of age, and odds ratios for high TG/HDL-C and high CMI of the interaction term, consisting of age (60~65 vs. 30~39 years) and tertile (3rd vs. 1st tertiles for leukocyte count), were significantly lower than the reference level. In conclusion, leukocyte count is associated with lipid-related indices, and the associations are independent of smoking and alcohol drinking and tend to be weaker with an increase of age in Japanese middle-aged men.

Citation: Wakabayashi I (2023) Associations between leukocyte count and lipid-related indices: Effect of age and confounding by habits of smoking and alcohol drinking. PLoS ONE 18(1): e0281185. https://doi.org/10.1371/journal.pone.0281185

Editor: Anindita Dutta, University of Chicago, UNITED STATES

Received: August 26, 2022; Accepted: January 17, 2023; Published: January 31, 2023

Copyright: © 2023 Ichiro Wakabayashi. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Data Availability: All data are within the paper and its Supporting information files.

Funding: This work was supported by the Grant-in-Aid for Scientific Research (No. 21H03386) from the Japan Society for the Promotion of Science. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Competing interests: The author has declared that no competing interests exist.

Introduction

Leukocyte count in peripheral blood, a marker of inflammation, is known to be a predictor of cardiovascular events [1]. Blood lipid profile, a major risk factor of cardiovascular disease, is also related to leukocyte count: leukocyte count showed positive associations with total cholesterol, LDL cholesterol and triglycerides and a negative association with HDL cholesterol [2, 3]. These associations are mediated by the chronic inflammatory process of arterial walls in progression of atherosclerosis as the common pathogenesis of cardiovascular disease. In addition, hypercholesterolemia induces proliferation of hematopoietic stem cell progenitors in mice [4], though there is little evidence currently that demonstrates a similar process in humans and thus further studies are needed to confirm this hypothesis.

Lifestyles including diet, nutrition, physical activity and habits of smoking and alcohol drinking are modifiable factors influencing cardiovascular risk. Leukocyte count has been shown to be associated with smoking and alcohol drinking: Cigarette smoking causes an elevation of leukocyte count, and smoking cessation leads to recovery of leukocyte count [5–7]. Leukocyte count was shown to be inversely associated with alcohol consumption [6, 8]. Thus, smoking and drinking are thought to show opposite effects on leukocyte count.

Lipid-related indices, consisting of two or more variables including blood lipids, have been proposed as a discriminator of cardiometabolic risk. The ratio of LDL cholesterol to HDL cholesterol (LDL-C/HDL-C ratio) is a classical atherogenic index that predicts eventual coronary heart disease [9]. The ratio of triglycerides to HDL cholesterol (TG/HDL-C ratio) has been proposed as a better predictor than LDL-C/HDL-C ratio for cardiovascular disease [10, 11] and has been shown to reflect small dense LDL [12], a potent risk factor of cardiovascular diseases that is independent of LDL cholesterol levels [13]. More recently, cardiometabolic index (CMI), which is calculated as the product of TG/HDL-C ratio and the ratio of waist circumference to height (waist-to-height ratio), has been proposed as a discriminator of diabetes [14] and has been shown to be associated with atherosclerosis in patients with peripheral arterial disease [15] and to discriminate the risk of incidental coronary heart disease [16]. TG/HDL-C ratio has been reported to be correlated positively with leukocyte count in overweight and obese adolescents [17]. However, there has been little information on the relationships of leukocyte count with lipid-related indices in adults. Moreover, it remains to be determined whether and how these relationships are confounded by smoking and alcohol drinking.

The aim of this study was therefore to clarify whether and how smoking and alcohol drinking affect the relationships between leukocyte count and lipid-related indices. The relationship between leukocyte count and metabolic syndrome, a cluster of accumulated cardiovascular risks, including obesity, dyslipidemia, hypertension and diabetes [18, 19], was also investigated with adjustment for smoking and alcohol drinking. In addition, the above relationships between leukocyte count and cardiometabolic risk were investigated in different age groups of middle-aged men.

Methods

Subjects

The subjects were 11261 middle-aged men (30 ~ 65 years) who had undergone annual health checkup examinations at their workplaces in Yamagata Prefecture in Japan. The participants were working at various kinds of companies including construction, manufacturing, information and communications, transport, wholesale and retail trade, eating and drinking places, accommodations, and services. The participants were recruited from April 2008 to March 2009. Individuals with histories of infectious diseases, autoimmune diseases and malignant diseases that are known to influence leukocyte count in peripheral blood were excluded from the subjects of this study. The design and protocol of this study were approved by the Ethics Committee of Yamagata University School of Medicine (No. 112 in the period from April 2005 to March 2006) and the Hyogo College of Medicine Ethics Committee (No. 3003 in 2020).

Present histories of lifestyles, including smoking, alcohol drinking, regular exercise, illness and therapy for illness, were surveyed by using self-written questionnaires. Cigarette smokers were defined as those who had had a habit of smoking for 6 months or longer and had it for the past month or longer. The subjects were divided by average daily cigarette consumption into three groups of nonsmokers, light smokers (20 or less cigarettes per day) and heavy smokers (21 or more cigarettes per day). The subjects were also divided by frequency of weekly alcohol drinking into three groups of nondrinkers, occasional drinkers and regular drinkers, who answered as “never”, “sometimes” and “every day”, respectively, to the question, “How frequently do you drink alcohol?”. Regarding a habit of regular exercise, subjects who answered in the questionnaire that they perform exercise almost every day for 30 min or longer per day were defined as those with a habit of regular exercise. The subjects were divided into two groups of those with and those without a habit of regular exercise.

Measurements

Height and body weight were measured with light clothes, and body mass index (BMI) was calculated as weight in kilograms divided by the square of height in meters. Waist circumference was measured at the navel level as defined by the Japanese Committee for the Diagnostic Criteria of Metabolic Syndrome [18]. Blood pressure was measured using a mercury sphygmomanometer after each subject had rested quietly in a sitting position with Korotkoff phase V being used to define diastolic pressure. Mean arterial pressure was defined as diastolic pressure plus one third of the difference between systolic pressure and diastolic pressure.

After overnight fasting, blood was collected from each subject in the morning. A part of the blood was immediately transferred to a 2-ml glass tube containing 3.8 mg EDTA-2K, and leukocyte count was measured by flow cytometry (with a red laser at 633 nm) using an automatic hematology analyzer (Sysmex XE-2100, Sysmex Corp., Kobe, Japan). The rest of the blood was used to obtain serum by centrifugation. Concentrations of triglycerides, HDL cholesterol and LDL cholesterol in serum were enzymatically measured with commercial kits by using an automatic analyzer (Hitachi Model 7350, Hitachi Corp., Tokyo, Japan). Hemoglobin A1c was measured by the latex cohesion method using a commercial kit (Determiner HbA1c, Kyowa Medex, Tokyo, Japan). Hemoglobin A1c values were calibrated by using a formula proposed by the Japan Diabetes Society as described before [20]. CMI was defined as the product of waist circumference (cm)-to-height (cm) ratio and triglycerides (mg/dl)-to-HDL cholesterol (mg/dl) ratio [14].

According to the criteria by the International Diabetes Federation [19] with a slight modification, metabolic syndrome was defined as the presence of two or more risk factors in addition to abdominal obesity diagnosed as high waist-to-height ratio. The risk factors included in the criteria of metabolic syndrome are abdominal obesity (high waist-to-height ratio), hypertension, dyslipidemia (high triglycerides and/or low HDL cholesterol) and diabetes evaluated by hemoglobin A1c level and a history of medication therapy for diabetes. The criterion for each of the above risk factors was defined as follows: abdominal obesity, waist-to-height ratio ≥ 0.5 [21]; hypertension, systolic blood pressure ≥ 140 mmHg and/or diastolic blood pressure ≥ 90 mmHg [22]; high triglycerides, triglycerides ≥ 150 mg/dl; low HDL cholesterol, HDL cholesterol < 40 mg/dl [19, 23]; diabetes, hemoglobin A1c ≥ 6.5% [24]. The cut-off values used for high LDL-C/HDL-C ratio, high TG/HDL-C ratio and high CMI were 3.5, 2.967 and 0.8, respectively [14, 25]. Subjects who were receiving medication therapy for hypertension and diabetes were included in the groups of subjects with hypertension and diabetes, respectively.

Statistical analysis

Leukocyte counts of the subjects were arranged in ascending order. Then the subjects were divided into three tertile groups of approximately equal sizes. The subjects were also divided into four age groups of subjects at ages of 30~39, 40~49, 50~59 and 60~65 years. Continuous variables, which are summarized as means with standard deviations or medians with interquartile ranges, were compared among the groups of smoking, alcohol drinking, or tertile for leukocyte count by using analysis of variance (ANOVA) followed by the Scheffé’s F test as a post-hoc test in the univariable analysis and by using analysis of covariance (ANCOVA) followed by Student’s t-test with Bonferroni’s multiplicity correction in the multivariable analysis. Since levels of triglycerides, TG/HDL-C ratio and CMI did not display normal distributions, they were compared non-parametrically by using the Kruskal-Wallis test followed by the Steel-Dwass test as a post-hoc test in the univariable analysis and were compared by using ANCOVA after logarithmic transformation with a base of 10 in the multivariable analysis. Pearson’s chi-square test was used to compare categorical variables, which are summarized as frequencies and percentages. Dichotomous variables, including high LDL-C/HDL-C ratio, high TG/HDL-C ratio, high CMI and metabolic syndrome, were compared by using logistic regression analysis. In univariable logistic regression analysis, crude odds ratios of each group pair were compared by using the Breslow-Day test. Odds ratios of an interaction term consisting of the tertile group of leukocyte count and the age group were also estimated in multivariable logistic regression analysis. In ANCOVA and multivariable logistic regression analysis, adjustments were performed for age, BMI, habits of smoking, alcohol drinking and regular exercise, and a history of medication therapy for dyslipidemia as appropriate. All probability (p) values are two-sided, and p values less than 0.05 were defined as significant. The statistical analyses were performed using a computer software program (IBM SPSS Statistics for Windows, Version 25.0. Armonk, NY: IBM Corp).

Results

Characteristics of overall subjects and the tertile groups of subjects classified by leukocyte count

Table 1 shows profiles of overall subjects and subjects in the tertile groups of leukocyte count. The percentage of smokers, BMI, waist-to-height ratio, systolic blood pressure, diastolic blood pressure, triglycerides, LDL cholesterol, LDL-C/HDL-C ratio, TG/HDL-C ratio, CMI, hemoglobin A1c and the percentage of subjects with diabetes tended to be higher with an increase of the tertile for leukocyte count. HDL cholesterol tended to be lower with an increase of the tertile for leukocyte count. Age and the percentage of drinkers were slightly but significantly younger and lower, respectively, in the 3rd tertile group of leukocyte count than in the 1st tertile group.

thumbnail
Download:
Table 1. Characteristics of overall subjects and subjects in each tertile group of leukocyte count.

https://doi.org/10.1371/journal.pone.0281185.t001

Correlations of leukocyte count with cardiometabolic risk factors

S1 Table shows Spearman’s rank correlation coefficients between cardiovascular risk factors. Leukocyte count was significantly correlated with waist-to-height ratio, mean arterial pressure, triglycerides, HDL cholesterol, LDL cholesterol, LDL-C/HDL-C ratio, TG/HDL-C ratio, CMI and hemoglobin A1c. The correlation coefficients of leukocyte count with the lipid-related indices, LDL-C/HDL-C ratio (r = 0.247, p < 0.01), TG/HDL-C ratio (r = 0.256, p < 0.01) and CMI (r = 0.258, p < 0.01), tended to be higher than those with other cardiovascular risk factors. Mean arterial pressure and hemoglobin A1c showed stronger correlations with waist-to-height ratio than those with other cardiovascular risk factors.

Comparison of mean levels of leukocyte count and lipid-related indices in the groups classified by a habit of smoking

Both in univariable analysis and multivariable analysis, mean levels of leukocyte count, TG/HDL-C ratio and CMI were significantly higher in light and heavy smokers than in nonsmokers and were significantly higher in heavy smokers than in light smokers (Table 2). Mean levels of LDL-C/HDL-C ratio were also significantly higher in light and heavy smokers than in nonsmokers and tended to be higher with an increase of cigarette consumption (Table 2).

thumbnail
Download:
Table 2. Comparison of mean levels of leukocyte count and lipid-related indices in non-, light and heavy smokers in overall subjects.

https://doi.org/10.1371/journal.pone.0281185.t002

Comparison of mean levels of leukocyte count and lipid-related indices in the groups classified by a habit of alcohol drinking

Both in univariable analysis and multivariable analysis, mean levels of leukocyte count and LDL-C/HDL-C ratio were significantly lower in occasional and regular drinkers than in nondrinkers and tended to be lower with an increase in frequency of drinking (Table 3). In univariable analysis, TG/HDL-C ratio and CMI were significantly lower in occasional and regular drinkers than in nondrinkers and tended to be lower with an increase in frequency of drinking, while there were no significant differences in TG/HDL-C ratio and CMI of non-, occasional and regular drinkers in multivariable analysis (Table 3).

thumbnail
Download:
Table 3. Comparison of mean levels of leukocyte count and lipid-related indices in non-, occasional and regular drinkers in overall subjects.

https://doi.org/10.1371/journal.pone.0281185.t003

Comparison of mean levels of lipid-related indices in the tertile groups of leukocyte count in overall subjects and subjects without habits of smoking and alcohol drinking

As shown in Table 4, both in overall subjects and subjects without habits of smoking and alcohol drinking, mean levels of LDL-C/HDL-C ratio, TG/HDL-C ratio and CMI were significantly higher in the 2nd and 3rd tertile groups of leukocyte count than in the 1st tertile group and tended to be higher with an increase of the tertile. The above findings were observed both in univariable analysis and multivariable analysis.

thumbnail
Download:
Table 4. Comparison of mean levels of each lipid-related index in the tertile groups of leukocyte count in overall subjects and subjects without habits of smoking and alcohol drinking.

https://doi.org/10.1371/journal.pone.0281185.t004

Comparison of prevalences of high lipid-related indices and metabolic syndrome in overall subjects and subjects without habits of smoking and alcohol drinking

As shown in Table 5, both in overall subjects and subjects without habits of smoking and alcohol drinking, prevalences of high LDL-C/HDL-C ratio, high TG/HDL-C ratio, high CMI and metabolic syndrome were significantly higher in the 2nd and 3rd tertile groups of leukocyte count than in the 1st tertile group and tended to be higher with an increase of the tertile.

thumbnail
Download:
Table 5. Comparison of prevalences of each cardiometabolic risk factor in the tertile groups of leukocyte count in overall subjects and subjects without habits of smoking and alcohol drinking.

https://doi.org/10.1371/journal.pone.0281185.t005

Odds ratios of the 2nd and 3rd tertile groups of leukocyte count vs. the 1st tertile group for high lipid-related indices and metabolic syndrome in overall subjects and subjects without habits of smoking and alcohol drinking

As shown in Table 6, both in overall subjects and subjects without habits of smoking and alcohol drinking, odds ratios of the 2nd and 3rd tertile groups vs. the 1st tertile group of leukocyte count for high LDL-C/HDL-C ratio, high TG/HDL-C ratio, high CMI and metabolic syndrome were significantly high when compared with the reference level of the 1st tertile group. The above odds ratios tended to be higher in the subject group of nonsmokers and nondrinkers than in the overall subjects.

thumbnail
Download:
Table 6. Odds ratios for each cardiometabolic risk factor in the 2nd and 3rd tertile groups vs. the 1st tertile group of leukocyte count.

https://doi.org/10.1371/journal.pone.0281185.t006

Relationships between age and cardiovascular risk factors

Results of comparison of cardiovascular risk factors in the age groups (30 ~ 39, 40 ~ 49, 50 ~ 59 and 60 ~ 65 years) are shown in Table 7. Leukocyte count was significantly lower in the oldest group than in the youngest group and tended to be lower with an increase of age. Waist-to-height ratio, mean arterial pressure, hemoglobin A1c and the percentage of subjects with metabolic syndrome were significantly lower in the youngest group than in the other three groups and tended to be higher with an increase of age. Triglycerides, LDL cholesterol, TG/HDL-C ratio, CMI and the percentages of subjects with high TG/HDL-C ratio and high CMI were higher in the 40 ~ 49 and 50 ~ 59 years groups than in the other two groups, thus showing inverted U-shaped relationships with age. HDL cholesterol was slightly but significantly higher in the 50 ~ 59 and 60 ~ 65 years groups than in the youngest group and tended to be higher with an increase of age.

thumbnail
Download:
Table 7. Comparison of cardiometabolic risk factors in the different age groups of overall subjects.

https://doi.org/10.1371/journal.pone.0281185.t007

Correlations between leukocyte count and lipid-related indices in the different age groups

In all of the age groups, leukocyte count was significantly correlated with the three lipid-related indices (S2 Table). The correlation coefficients with log-transformed TG/HDL-C ratio and CMI were lower in the oldest group than in the youngest group (TG/HDL-C ratio, 0.195 vs. 0.253 [p = 0.084]; CMI, 0.191 vs. 0.263 [p < 0.05]).

Odds ratios of the 2nd and 3rd tertile groups vs. the 1st tertile group of leukocyte count for high lipid-related indices and metabolic syndrome in the different age groups

As shown in Table 8, odds ratios of the 2nd and 3rd tertile groups vs. the 1st tertile group of leukocyte count for high lipid-related indices and metabolic syndrome were significantly high when compared with the reference level in all of the age groups except for the odds ratio of the 2nd vs. 1st tertile groups for high LDL-C/HDL-C ratio in the group of 60~65 years. The odds ratios tended to be lower in the oldest group than in the youngest group. Odds ratios of the interaction term consisting of age group (each age group vs. the youngest group) and leukocyte group (the 3rd vs. 1st tertile groups) for high TG/HDL ratio, high CMI and metabolic syndrome were significantly low in the oldest group when compared with the reference level.

thumbnail
Download:
Table 8. Odds ratios for each cardiometabolic risk factor of the 2nd and 3rd tertile groups vs. the 1st tertile group of leukocyte count in different age groups of overall subjects.

https://doi.org/10.1371/journal.pone.0281185.t008

Discussion

Both in overall subjects and subjects not having habits of smoking and alcohol drinking, leukocyte count was demonstrated to be positively associated with three lipid-related indices, including HDL-C/LDL-C ratio, TG/HDL-C ratio and CMI, and metabolic syndrome in ANCOVA and logistic regression analysis. Therefore, leukocyte count is associated with lipid-related indices and metabolic syndrome independently of habits of smoking and alcohol drinking. Leukocyte count and lipid-related indices were significantly higher in smokers than in nonsmokers and tended to be higher with an increase of cigarette consumption. Thus, smoking is a confounder for the relationship between leukocyte count and lipid-related indices. Leukocyte count and LDL-C/HDL-C ratio were significantly lower in drinkers than in nondrinkers and tended to be lower with an increase in frequency of drinking, suggesting that drinking is also a confounder of the association between leukocyte count and LDL-C/HDL-C ratio. Nonetheless, the associations of leukocyte count with the lipid-related indices and metabolic syndrome were independent of smoking and alcohol drinking as shown in the analyses of nonsmokers and nondrinkers alone and the multivariable analyses with adjustment for smoking and drinking. The three lipid-related indices evaluated in this study consist of HDL cholesterol, LDL cholesterol, triglycerides, and waist-to-height ratio, which were associated with leukocyte count (Table 1). In addition to these variables, blood pressure and hemoglobin A1c, which were also associated with leukocyte count (Table 1), are included in the criteria of metabolic syndrome. Therefore, it is reasonable that leucocyte count was associated with the lipid-related indices and metabolic syndrome. Interestingly, each odds ratio for high lipid-related indices of the 2nd or 3rd tertiles vs. the 1st tertile for leukocyte count in nonsmokers and nondrinkers tended to be higher than the corresponding odds ratio in overall subjects (Table 6). This suggests that the associations between leukocyte count and lipid-related indices after excluding effects of smoking and alcohol drinking are stronger than the associations before excluding them. To the best of my knowledge, this is the first study showing the relationships of leukocyte count with lipid-related indices and metabolic syndrome in relation to habits of smoking and drinking.

Leukocyte count, lipid-related indices and metabolic syndrome were also influenced by age: Leukocyte count was significantly lower in the oldest group of 60 ~ 65 years than in the youngest group of 30 ~ 39 years and tended to be lower with an increase of age; LDL-C/HDL-C ratio was significantly higher in the group of 40 ~ 49 years than in the group of 30 ~ 39 years; TG/HDL-C ratio and CMI were significantly higher in the groups of 40 ~ 49 and 50 ~ 59 years than in the group of 30 ~ 39 years; the prevalence of metabolic syndrome tended to be higher with an increase of age (Table 7). Thus, age is also a confounder for the relationships of leukocyte count with the lipid-related indices and metabolic syndrome. Therefore, age was always included in the explanatory variables in the multivariable analyses. However, as shown in S2 Table and Table 8, leukocyte count was associated with LDL-C/HDL-C ratio, TG/HDL-C ratio, CMI and metabolic syndrome in all of the age groups. Thus, these associations are independent of age. Interestingly, the correlation coefficients of leukocyte count with log-transformed TG/HDL-C ratio and log-transformed CMI tended to be lower with an increase of age, and the odds ratios of the interaction term consisting of age (vs. the youngest group) and leukocyte tertile (vs. the 1st tertile) for high TG/HDL-C, high CMI and metabolic syndrome were significantly lower in the group of 60 ~ 65 years compared with the reference level (30 ~ 39 years), suggesting weaker associations of leukocyte count with high TG/HDL-C, high CMI and metabolic syndrome in the oldest group than in the youngest group (Table 8). Moreover, the odds ratios for high TG/HDL-C ratio, high CMI and metabolic syndrome of the 2nd and 3rd tertiles for leukocyte count vs. the 1st tertile tended to be lower with an increase of age. Therefore, the associations of leukocyte count with high TG/HDL-C ratio, high CMI and metabolic syndrome are suggested to be weaker with age in middle-aged men. Although the reason for this age-dependent difference is unknown, one possible reason is higher variability of leukocyte count levels due to higher frequency of latent inflammatory diseases in older men. However, standard deviations of leukocyte count were not higher in older groups than in younger groups (Table 7).

Since atherosclerosis includes an inflammatory process [26], it is reasonable that higher leukocyte count is associated with higher risk of cardiovascular disease and its risk factors including visceral obesity and dyslipidemia, which are components of lipid-related indices. In this study, the association between leukocyte count and cardiometabolic risk remained significant after adjustment for various confounding factors including age, BMI, smoking, alcohol drinking and regular exercise. This finding supports the hypothesis that a high leukocyte count in healthy individuals reflects an inflammatory process in the pathogenesis of atherosclerosis. A recent study in which a longitudinal analysis was performed using a database of 9058 participants showed a robust positive association between triglycerides and leukocyte count in peripheral blood [27]. The authors of that report suggested a direct involvement of triglycerides in leukogenesis. The risk of high lipid-related indices in subjects without habits of smoking and drinking were three- to four-fold higher in the 3rd tertile group of leukocyte count than in the 1st tertile group in multivariable logistic regression analysis (Table 6). Thus, leukocyte count, which is a simple measure and is usually included in general health checkup examinations, is thought to be a useful variable to evaluate cardiovascular risk, although leukocyte count is influenced by infection and various inflammatory diseases. Because healthy individuals with a high leukocyte count in repeated health checkup examinations are suspected to have considerable atherosclerotic progression, they should be recommended to receive further examinations of atherosclerosis including ultrasonography of carotid arteries and aortic pulse wave velocity. The findings of stronger associations of leukocyte count with lipid-related indices and metabolic syndrome in younger subjects suggest that leukocyte count is more useful for discriminating cardiometabolic risk in younger middle-aged men than in older men.

Strengths

This study clearly demonstrated that leukocyte count in peripheral blood shows associations with cardiometabolic risk evaluated by prevalence of metabolic syndrome and high levels of lipid-related indices. These associations were independent of age and habits of smoking and alcohol drinking, although they confounded the relationships between leukocyte count and cardiometabolic risk. Thus, leukocyte count is a simple useful marker for discriminating cardiometabolic risk. It was also found in this study that the associations between leukocyte count and cardiometabolic risk declined with an increase of age (30 ~ 65 years). This suggests that leukocyte count is more useful as a discriminator of cardiometabolic risk in younger men.

Limitations

The subjects were middle-aged Japanese men. An age-dependent significant decline in the strength of the associations of leukocyte count with lipid-related indices and metabolic syndrome was observed in the group of 60 ~ 65 years, which is the oldest category in general workplaces. Thus, the above associations need to be investigated in further older subjects. Both the profiles of blood lipids and leukocyte count have been reported to differ by gender [28, 29]. It is known that there are ethnic differences in blood lipid levels and the risk of cardiovascular diseases [30]. Therefore, further studies using data for female subjects and subjects with other ethnicities are also needed to confirm the findings of the present study. The subjects were workers who had undergone annual health checkups at various companies, and thus there is a selection bias because they were generally healthier than those not receiving annual health checkups. Drinking, smoking and regular exercise were self-reported and thus there are possibilities of biases caused by underreporting or overreporting of them. Regarding the definition of smokers, subjects who had a habit of smoking for the past month or longer were categorized as smokers. However, there is a possibility that effects of cigarette smoking on leukocyte count and cardiometabolic risk persisted in subjects who recently quit smoking. The participants were recruited from April 2008 to March 2009, and at that time, individuals with a habit of vaping were rare in Japan. In a large population-based follow-up study in the UK, a positive association between leukocyte count and cardiovascular events mainly depended on the increase in neutrophils [31]. In a population-based cross-sectional study in the US, lymphocyte and basophil counts showed relatively strong positive and inverse correlations with triglycerides and HDL cholesterol, respectively [32]. However, information on leukocyte subtypes was not available in the present study. Since this study is cross-sectional in its design, further prospective studies are needed to discuss causal relationships of leukocyte count with lipid-related indices and metabolic syndrome.

Conclusion

There are clear associations of lipid-related indices and metabolic syndrome with total leukocyte count in middle-aged Japanese men, and the associations are independent of alcohol consumption and cigarette smoking. Leukocyte count in peripheral blood, a simple measure usually included in general health checkup examinations, is a more useful discriminator of cardiometabolic risk in young middle-aged men than in older men.

Supporting information

S1 Table. Correlations between each pair of leukocyte count and cardiometabolic risk factors in overall subjects.

Shown are Spearman’s rank correlation coefficients between each pair of leukocyte count and cardiometabolic risk factors in overall subjects. WHtR, waist-to-height ratio; MAP, mean arterial pressure. Asterisks denote significant correlations (**, p < 0.01).

https://doi.org/10.1371/journal.pone.0281185.s001

(DOCX)

S2 Table. Correlations between leukocyte count and each lipid-related index in different age groups of overall subjects.

Shown are Pearson’s correlation coefficients between leukocyte count and each lipid-related index (LDL-C/HDL-C ratio, TG/HDL-C ratio and CMI). Symbols indicate significant (**, p < 0.01) correlation coefficients and significant (†, p < 0.05) or marginally significant (#, p = 0.084) differences from the corresponding correlation coefficients in the youngest (30~39 years) group.

https://doi.org/10.1371/journal.pone.0281185.s002

(DOCX)

S1 Dataset.

https://doi.org/10.1371/journal.pone.0281185.s003

(XLSX)

References

  1. 1. Madjid M, Awan I, Willerson JT, Casscells SW. Leukocyte count and coronary heart disease: implications for risk assessment. J Am Coll Cardiol. 2004; 44(10): 1945–56. pmid:15542275
  2. 2. Liu Y, Kong X, Wang W, Fan F, Zhang Y, Zhao M, et al. Association of peripheral differential leukocyte counts with dyslipidemia risk in Chinese patients with hypertension: insight from the China Stroke Primary Prevention Trial. J Lipid Res. 2017; 58(1): 256–66. pmid:27879312
  3. 3. Lai YC, Woollard KJ, McClelland RL, Allison MA, Rye KA, Ong KL, et al. The association of plasma lipids with white blood cell counts: Results from the Multi-Ethnic Study of Atherosclerosis. J Clin Lipidol. 2019; 13(5): 812–20. pmid:31402296
  4. 4. Soehnlein O, Swirski FK. Hypercholesterolemia links hematopoiesis with atherosclerosis. Trends Endocrinol Metab. 2013; 24(3): 129–36. pmid:23228326
  5. 5. Petitti DB, Kipp H. The leukocyte count: associations with intensity of smoking and persistence of effect after quitting. Am J Epidemiol. 1986; 123(1): 89–95. pmid:3940445
  6. 6. Schwartz J, Weiss ST. Host and environmental factors influencing the peripheral blood leukocyte count. Am J Epidemiol. 1991; 134(12): 1402–9. pmid:1776614
  7. 7. Higuchi T, Omata F, Tsuchihashi K, Higashioka K, Koyamada R, Okada S. Current cigarette smoking is a reversible cause of elevated white blood cell count: Cross-sectional and longitudinal studies. Prev Med Rep. 2016; 4: 417–22. pmid:27583199
  8. 8. Nakanishi N, Suzuki K, Tatara K. Association between lifestyle and white blood cell count: a study of Japanese male office workers. Occup Med (Lond). 2003; 53(2): 135–7. pmid:12637599
  9. 9. Kannel WB. Lipids, diabetes, and coronary heart disease: insights from the Framingham Study. Am Heart J. 1985; 110(5): 1100–7. pmid:4061265
  10. 10. Gaziano JM, Hennekens CH, O’Donnell CJ, Breslow JL, Buring JE. Fasting triglycerides, high-density lipoprotein, and risk of myocardial infarction. Circulation. 1997; 96(8): 2520–5. pmid:9355888
  11. 11. Jeppesen J, Hein HO, Suadicani P, Gyntelberg F. Relation of high TG-low HDL cholesterol and LDL cholesterol to the incidence of ischemic heart disease. An 8-year follow-up in the Copenhagen Male Study. Arterioscler Thromb Vasc Biol. 1997; 17(6): 1114–20. pmid:9194762
  12. 12. Dobiásová M, Frohlich J. The plasma parameter log (TG/HDL-C) as an atherogenic index: correlation with lipoprotein particle size and esterification rate in apoB-lipoprotein-depleted plasma (FER(HDL)). Clin Biochem. 2001; 34(7): 583–8. pmid:11738396
  13. 13. Koba S, Hirano T, Kondo T, Shibata M, Suzuki H, Murakami M, et al. Significance of small dense low-density lipoproteins and other risk factors in patients with various types of coronary heart disease. Am Heart J. 2002; 144(6): 1026–35. pmid:12486427
  14. 14. Wakabayashi I, Daimon T. The "cardiometabolic index" as a new marker determined by adiposity and blood lipids for discrimination of diabetes mellitus. Clin Chim Acta. 2015; 438: 274–8. pmid:25199852
  15. 15. Wakabayashi I, Sotoda Y, Hirooka S, Orita H. Association between cardiometabolic index and atherosclerotic progression in patients with peripheral arterial disease. Clin Chim Acta. 2015; 446: 231–6. pmid:25920694
  16. 16. Higashiyama A, Wakabayashi I, Okamura T, Kokubo Y, Watanabe M, Takegami M, et al. The risk of fasting triglycerides and its related indices for ischemic cardiovascular diseases in Japanese community dwellers: the Suita Study. J Atheroscler Thromb. 2021; 28(12): 1275–88. pmid:34053965
  17. 17. Rumińska M, Witkowska-Sędek E, Artemniak-Wojtowicz D, Krajewska M, Majcher A, Sobol M, et al. Changes in leukocyte profile and C-reactive protein concentration in overweight and obese adolescents after reduction of body weight. Cent Eur J Immunol. 2019; 44(3): 307–15. pmid:31871419
  18. 18. Anonymous (2005) Metabolic syndrome-definition and diagnostic criteria in Japan. Nihon Naika Gakkai Zasshi. 94: 794–809 (in Japanese).
  19. 19. Alberti KG, Eckel RH, Grundy SM, Zimmet PZ, Cleeman JI, Donato KA, et al.; International Diabetes Federation Task Force on Epidemiology and Prevention; National Heart, Lung, and Blood Institute; American Heart Association; World Heart Federation; International Atherosclerosis Society; International Association for the Study of Obesity. Harmonizing the metabolic syndrome: a joint interim statement of the International Diabetes Federation Task Force on Epidemiology and Prevention; National Heart, Lung, and Blood Institute; American Heart Association; World Heart Federation; International Atherosclerosis Society; and International Association for the Study of Obesity. Circulation. 2009; 120(16): 1640–5. pmid:19805654
  20. 20. Wakabayashi I. Associations between polycythemia and cardiometabolic risk factors in middle-aged men. Clin Chim Acta. 2022; 531: 248–53. pmid:35421399
  21. 21. Hsieh SD, Muto T. Metabolic syndrome in Japanese men and women with special reference to the anthropometric criteria for the assessment of obesity: Proposal to use the waist-to-height ratio. Prev Med. 2006; 42(2): 135–9. pmid:16376978
  22. 22. Flack JM, Adekola B. Blood pressure and the new ACC/AHA hypertension guidelines. Trends Cardiovasc Med. 2020; 30(3): 160–4. pmid:31521481
  23. 23. Teramoto T, Sasaki J, Ishibashi S, Birou S, Daida H, Dohi S, et al; Japan Atherosclerosis Society. Executive summary of the Japan Atherosclerosis Society (JAS) guidelines for the diagnosis and prevention of atherosclerotic cardiovascular diseases in Japan -2012 version. J Atheroscler Thromb. 2013; 20(6): 517–23. pmid:23665881
  24. 24. Anonymous, American Diabetes Association. Diagnosis and classification of diabetes mellitus. Diabetes Care. 2010; 33 Suppl 1(Suppl 1): S62–9. pmid:20042775
  25. 25. Wakabayashi I, Daimon T. Comparison of discrimination for cardio-metabolic risk by different cut-off values of the ratio of triglycerides to HDL cholesterol. Lipids Health Dis. 2019; 18(1): 156. pmid:31351479
  26. 26. Libby P. Inflammation in atherosclerosis. Arterioscler Thromb Vasc Biol. 2012; 32(9): 2045–51. pmid:22895665
  27. 27. Tucker B, Sawant S, McDonald H, Rye KA, Patel S, Ong KL, et al. The association of serum lipid and lipoprotein levels with total and differential leukocyte counts: Results of a cross-sectional and longitudinal analysis of the UK Biobank. Atherosclerosis. 2021; 319: 1–9. pmid:33453490
  28. 28. Swiger KJ, Martin SS, Blaha MJ, Toth PP, Nasir K, Michos ED, et al. Narrowing sex differences in lipoprotein cholesterol subclasses following mid-life: the very large database of lipids (VLDL-10B). J Am Heart Assoc. 2014; 3(2): e000851. pmid:24755154
  29. 29. Chen Y, Zhang Y, Zhao G, Chen C, Yang P, Ye S, Tan X. Difference in Leukocyte composition between women before and after menopausal age, and distinct sexual dimorphism. PLoS One. 2016; 11(9): e0162953. pmid:27657912
  30. 30. Gazzola K, Reeskamp L, van den Born BJ. Ethnicity, lipids and cardiovascular disease. Curr Opin Lipidol. 2017; 28(3): 225–30. pmid:28338521
  31. 31. Groot HE, van Blokland IV, Lipsic E, Karper JC, van der Harst P. Leukocyte profiles across the cardiovascular disease continuum: A population-based cohort study. J Mol Cell Cardiol. 2020; 138: 158–64. pmid:31812706
  32. 32. Andersen CJ, Vance TM. Gender dictates the relationship between serum lipids and leukocyte counts in the National Health and Nutrition Examination Survey 1999–2004. J Clin Med. 2019; 8(3): 365. pmid:30875952