Abstract
Inflammation plays a key role in the pathogenesis of cardiovascular diseases via the development of atherosclerosis. Here, we evaluated the impact of serum C‐reactive protein (CRP) and the white blood cell (WBC) count on the risk of hypertension in middle‐aged Japanese men at a work site. We evaluated a total of 2991 Japanese male workers without hypertension who ranged in age from 18 to 64 years (mean age 40.4 ± 0.2 years) at a worksite in 2010. The hazard ratio (HR) for incident hypertension was estimated according to quartile levels of serum high‐sensitivity CRP (hs‐CRP) or WBC count. These men were followed up for 5 years from 2010 to 2015. During the follow‐up period, 579 (19.4%) subjects developed hypertension. In a multivariable analysis, the risk of incident hypertension was significantly increased with higher hs‐CRP levels: HR 1.00 (reference) for the lowest quartile, 1.39 (1.04‐1.85) for the 2nd quartile, 1.46 (1.08‐1.98) for the 3rd quartile, and 1.57 (1.17‐2.11) for the highest quartile. In contrast, the WBC count was not associated with a greater risk of incident hypertension after multivariable adjustment. These findings suggest that higher levels of serum hs‐CRP, but not the WBC count, are associated with the future incidence of hypertension in middle‐aged Japanese men.
Keywords: hs‐CRP, incident hypertension, Japanese, worksite population
1. INTRODUCTION
Chronic inflammation plays an important role in the progression and development of cardiovascular diseases.1 The commonly used marker C‐reactive protein (CRP) is an acute‐phase protein produced in response to infection, inflammation, and tissue injury. The white blood cell (WBC) count is also used to check inflammation in clinical practice. However, increasing evidence has shown that the CRP level and WBC count may contribute to an increased risk of cardiovascular diseases.2, 3, 4, 5 In an animal model, an injection of human CRP increased the myocardial infarct size in rats subjected to coronary artery ligation, and the administration of a specific small‐molecule inhibitor of CRP abrogated the increase in infarct size and cardiac dysfunction.4 In humans, an infusion of recombinant human CRP caused inflammation and coagulation.3 Other studies have shown that both higher levels of CRP or WBC count contribute to the development of diabetes, stroke, and cardiovascular diseases.2, 6, 7, 8
Hypertension is one of the most prevalent and important risk factors for cardiovascular diseases. Although several studies have reported that a higher CRP level or WBC count is an independent risk factor for incident hypertension,9, 10 the CRP level and the WBC count have been shown to be increased in individuals with smoking habits, as well as those with obesity, insulin resistance, or metabolic syndrome, all of which are traditional risk factors for cardiovascular diseases.2, 14, 15 Thus, inconsistent findings remain regarding whether a higher CRP or WBC count is an independent risk factor for incident hypertension and cardiovascular diseases or simply markers of accumulation of comorbid risk factors.21, 22
Adipose tissue was previously thought to be simply a storage site for accumulated fat, but adipose tissue is now recognized as an endocrine organ and an important source that produces a number of endocrine molecules (adipokines) including adiponectin, leptin, tumor necrosis factor‐alpha (TNF‐α), and interleukin (IL)‐6.25, 26 Adipose tissue from obese individuals releases large amounts of proinflammatory proteins such as IL‐6 that increase the levels of CRP, mainly in the liver.26 In obesity, white blood cells infiltrate adipose tissue and contribute to the immune system.18 Abdominal fat plays a key role in the pathogenesis of both obesity and metabolic syndrome,25, 26 and obesity and metabolic syndrome may thus contribute to increases in the serum CRP level and WBC count.18, 27
We conducted the present study in a working population to elucidate the effects of CRP and the WBC count on incident hypertension in young and middle‐aged workers, who are a particularly important population in whom cardiovascular events should be prevented. We also discuss the issue of whether obesity and the other cardiovascular risk factors affect the relationship between incident hypertension and the CRP level and WBC count.
2. METHODS
2.1. Study population
The study group consisted of 2991 male employees of a bus and railway company in Japan who ranged in age from 18 to 64 years (mean age 40.4 ± 0.2 years) in 2010. All were without hypertension. Females were excluded from the study, because they constituted only a small portion of this population. The subjects were followed up by routine annual physical examinations during the years 2010‐2015. The study protocol was approved by the Ethics Committee of Kyushu University.
2.2. Data collection
At the baseline examination in 2010, the blood pressure value in the right arm of the seated subject was measured. When the systolic and diastolic blood pressures were more than 140 and 90 mm Hg, respectively, the lower value was recorded after a repeated measurement. Waist circumference was measured at the umbilical level with the subject standing. Serum aspartate transaminase (AST), alanine transaminase (ALT), gamma‐glutamyl transpeptidase (γ‐GTP), low‐density lipoprotein (LDL) cholesterol, high‐density lipoprotein (HDL) cholesterol, triglycerides, serum creatinine, serum uric acid, glucose, hemoglobin A1c (HbA1c), WBC count, and high‐sensitivity CRP (hs‐CRP) were also measured. The value for HbA1c (%) was estimated as an National Glycohemoglobin Standardization Program (NGSP) equivalent value (%) calculated by the formula HbA1c (%) = HbA1c (%, Japan Diabetes Society [JDS]) + 0.4%, considering the relational expression of HbA1c (%, JDS) measured by the previous Japanese standard substance and measurement method and HbA1c (%, NGSP).28
The estimated glomerular filtration rate (eGFR) was calculated using the following formula, which is an equation modified for Japanese men: eGFR (mL/min/1.73 m2) = 194 × age−0.287 × serum creatinine−1.094.29 The body mass index (BMI) was defined as weight (in kg) divided by height (in m2). Information on medical history, habitual alcohol intake, and current smoking was obtained by interview or questionnaires.
2.3. Definitions of hypertension, diabetes mellitus, and dyslipidemia
Hypertension was defined as systolic blood pressure ≥140 mm Hg, diastolic blood pressure ≥90 mm Hg, or the use of antihypertensive drugs. Levels of blood pressure were classified into three groups according to the Japanese Society of Hypertension Guidelines for the Management of Hypertension (JSH2014) as follows: optimal blood pressure (SBP <120 mm Hg and DBP <80 mm Hg); normal blood pressure (SBP 120‐129 mm Hg or DBP 80‐84 mm Hg); or high‐normal blood pressure (SBP 130‐139 mm Hg or DBP 85‐89 mm Hg).30 Diabetes mellitus was defined as treatment with antidiabetic drugs or an HbA1c value ≥6.5%.31
2.4. Data analysis
The linear trends in the mean values and frequencies of covariates across quartiles of hs‐CRP and WBC counts were determined using a linear regression or logistic regression model. The serum levels of hs‐CRP were logarithmically transformed, because their distribution was skewed. Kaplan‐Meier curves were used to estimate the cumulative incidence of hypertension according to the quartiles of hs‐CRP values and WBC counts. We estimated the age‐adjusted or multivariable‐adjusted hazard ratio (HR) and 95% confidence intervals (95% CI) for the development of hypertension according to the levels of hs‐CRP and WBC count by using a Cox's proportional hazard model. The heterogeneity in the relationships between subgroups was tested by adding an interaction term to the statistical model. All data analyses were carried out using SAS software for Windows, ver. 9.4 (SAS Institute, Cary, NC) for statistical computing. A two‐tailed P‐value <0.05 was considered significant.
3. RESULTS
The baseline characteristics of the subjects according to quartile groups of hs‐CRP and WBC count are summarized in Tables 1 and 2, respectively. The subjects with higher hs‐CRP levels had greater mean values of BMI, waist circumference, systolic and diastolic blood pressures, LDL cholesterol, triglyceride, AST, ALT, γ‐GTP, glucose, HbA1c, hemoglobin, WBC count, smaller mean values of HDL cholesterol and eGFR, and a higher frequency of current smoking and prescription of lipid‐lowering or antidiabetic drugs (Table 1). The prescription rate of uric acid‐lowering drugs was not associated with the hs‐CRP quartiles (Table 1).
Table 1.
Baseline characteristics of subjects according to quartile of high‐sensitivity CRP level
| Variables | Baseline hs‐CRP (mg/dL) | P for trend | |||
|---|---|---|---|---|---|
| First quartile | Second quartile | Third quartile | Fourth quartile | ||
| ≤0.02 | 0.03‐0.05 | 0.06‐0.10 | ≥0.11 | ||
| n = 574 | n = 920 | n = 677 | n = 820 | ||
| Age (y) | 38 (0.4) | 42 (0.3) | 42 (0.4) | 42 (0.3) | <0.001 |
| BMI (kg/m2) | 21.8 (0.11) | 22.9 (0.10) | 23.8 (0.11) | 24.6 (0.13) | <0.001 |
| Waist circumference (cm) | 79 (0.3) | 82 (0.3) | 85 (0.3) | 87 (0.3) | <0.001 |
| Systolic blood pressure (mm Hg) | 115 (0.5) | 117 (0.4) | 117 (0.4) | 118 (0.4) | <0.001 |
| Diastolic blood pressure (mm Hg) | 73 (0.4) | 75 (0.3) | 75 (0.3) | 76 (0.3) | <0.001 |
| LDL cholesterol (mg/dL) | 113 (1.2) | 120 (1.0) | 122 (1.2) | 125 (1.1) | <0.001 |
| HDL cholesterol (mg/dL) | 62 (0.6) | 58 (0.4) | 55 (0.5) | 52 (0.4) | <0.001 |
| Triglyceride (mg/dL) | 117 (3.4) | 135 (2.9) | 170 (5.6) | 173 (4.7) | <0.001 |
| AST (IU/L) | 21 (0.3) | 22 (0.3) | 23 (0.3) | 25 (0.4) | <0.001 |
| ALT (IU/L) | 22 (0.5) | 27 (0.8) | 29 (0.7) | 33 (0.8) | <0.001 |
| γ‐GTP (IU/L) | 33 (1.2) | 39 (1.2) | 45 (1.6) | 53 (3.2) | <0.001 |
| eGFR (mL/min/1.73 m2) | 91 (0.6) | 88 (0.5) | 88 (0.6) | 88 (0.5) | 0.01 |
| Glucose (mg/dL) | 95 (0.7) | 95 (0.6) | 96 (0.8) | 100 (0.9) | <0.001 |
| Hemoglobin A1c (%) | 5.5 (0.01) | 5.6 (0.02) | 5.7 (0.02) | 5.8 (0.03) | <0.001 |
| Hemoglobin (g/dL) | 14.9 (0.04) | 15.0 (0.03) | 15.1 (0.04) | 15.1 (0.04) | <0.001 |
| White blood cell (1000 cells/µL) | 6.1 (0.06) | 6.5 (0.06) | 7.0 (0.06) | 7.6 (0.08) | <0.001 |
| Habitual alcohol intake (%) | 66.0 | 61.3 | 64.4 | 57.2 | 0.005 |
| Current smoking (%) | 39.9 | 46.9 | 51.6 | 52.8 | <0.001 |
| Lipid‐lowering drugs (%) | 2.3 | 2.7 | 4.7 | 4.0 | 0.02 |
| Antidiabetic drugs (%) | 0.7 | 0.8 | 1.2 | 2.1 | 0.009 |
| Uric acid‐lowering drugs (%) | 0.9 | 1.0 | 1.6 | 1.8 | 0.06 |
ALT, alanine transaminase; AST, aspartate transaminase; BMI, body mass index; eGFR, estimated glomerular filtration rate; HDL, high‐density lipoprotein; hs‐CRP, high‐sensitivity C‐reactive protein; LDL, low‐density lipoprotein; γ‐GTP, γ‐glutamyl transpeptidase.
Values are given as mean (SE) or frequency (%).
Table 2.
Baseline characteristics of subjects according to quartile of serum white blood cell count
| Variables | Baseline white blood cell (1000 cells/µL) | P for trend | |||
|---|---|---|---|---|---|
| First quartile | Second quartile | Third quartile | Fourth quartile | ||
| ≤5.4 | 5.5‐6.5 | 6.6‐7.8 | ≥7.9 | ||
| n = 714 | n = 767 | n = 760 | n = 750 | ||
| Age (y) | 40 (0.4) | 41 (0.4) | 42 (0.4) | 42 (0.3) | <0.001 |
| BMI (kg/m2) | 22.3 (0.10) | 23.1 (0.11) | 23.7 (0.11) | 24.2 (0.13) | <0.001 |
| Waist circumference (cm) | 81 (0.3) | 83 (0.3) | 84 (0.3) | 86 (0.4) | <0.001 |
| Systolic blood pressure (mm Hg) | 115 (0.4) | 116 (0.4) | 118 (0.4) | 119 (0.4) | <0.001 |
| Diastolic blood pressure (mm Hg) | 73 (0.3) | 75 (0.3) | 75 (0.3) | 75 (0.3) | <0.001 |
| LDL cholesterol (mg/dL) | 112 (1.1) | 119 (1.1) | 123 (1.1) | 127 (1.2) | <0.001 |
| HDL cholesterol (mg/dL) | 61 (0.6) | 57 (0.5) | 55 (0.5) | 53 (0.5) | <0.001 |
| Triglyceride (mg/dL) | 115 (3.4) | 141 (3.6) | 163 (4.5) | 177 (5.1) | <0.001 |
| AST (IU/L) | 22 (0.4) | 22 (0.3) | 23 (0.4) | 23 (0.3) | 0.15 |
| ALT (IU/L) | 25 (1.0) | 27 (0.6) | 30 (0.8) | 32 (0.8) | <0.001 |
| γ‐GTP (IU/L) | 36 (1.2) | 42 (1.6) | 47 (3.2) | 47 (1.6) | <0.001 |
| eGFR (mL/min/1.73 m2) | 88 (0.6) | 88 (0.6) | 88 (0.6) | 89 (0.5) | 0.08 |
| Glucose (mg/dL) | 93 (0.6) | 97 (0.7) | 97 (0.7) | 98 (0.8) | <0.001 |
| Hemoglobin A1c (%) | 5.5 (0.01) | 5.7 (0.02) | 5.7 (0.02) | 5.8 (0.02) | <0.001 |
| Hemoglobin (g/dL) | 14.8 (0.04) | 15.1 (0.04) | 15.1 (0.04) | 15.3 (0.04) | <0.001 |
| hs‐CRP (mg/dL) | 0.09 (0.008) | 0.14 (0.015) | 0.14 (0.015) | 0.24 (0.025) | <0.001 |
| Habitual alcohol intake (%) | 67.5 | 61.3 | 61.3 | 56.9 | <0.001 |
| Current smoking (%) | 28.9 | 52.9 | 52.9 | 71.6 | <0.001 |
| Lipid‐lowering drugs (%) | 2.2 | 3.1 | 3.2 | 4.5 | 0.04 |
| Antidiabetic drugs (%) | 0.7 | 1.2 | 1.2 | 1.6 | 0.16 |
| Uric acid‐lowering drugs (%) | 0.8 | 3.2 | 1.6 | 1.2 | 0.63 |
ALT, alanine transaminase; AST, aspartate transaminase; BMI, body mass index; eGFR, estimated glomerular filtration rate; HDL, high‐density lipoprotein; hs‐CRP, high‐sensitivity C‐reactive protein; LDL, low‐density lipoprotein; γ‐GTP, γ‐glutamyl transpeptidase.
Values are given as mean (SE), number or frequency (%).
A similar tendency was observed in the subjects with higher WBC counts, but the mean values of AST and eGFR and the prescription rates of antidiabetic drugs were not associated with the WBC quartiles (Table 2).
During the 5‐year follow‐up, 579 (19.4%) of the men developed hypertension. Figure 1 shows the Kaplan‐Meier curve for the cumulative rate of incident hypertension during the follow‐up period by quartile for hs‐CRP and WBC. Subjects in the highest quartile of hs‐CRP or WBC had the highest cumulative rate of incident hypertension. The age‐adjusted HR for the development of hypertension increased with higher levels of hs‐CRP (P for trend <0.001; Table 3) or WBC count (P for trend <0.001; Table 3). This association of hs‐CRP with incident hypertension remained unchanged after adjusting for potential confounding factors, namely, age, BMI, systolic blood pressure, HDL cholesterol, LDL cholesterol, triglyceride, eGFR, HbA1c, hemoglobin, use of lipid‐lowering, antidiabetic or uric acid‐lowering drugs, alcohol intake habit, and current smoking, where the waist circumference was not used due to high collinearity with BMI (r = 0.89; Table 3). In contrast, the association of WBC count with incident hypertension was not observed after the multivariable adjustment (Table 3). Sensitivity analyses using waist circumference instead of BMI for multivariable adjustment showed no obvious difference to the effects of hs‐CRP or WBC on the development of hypertension (Table S1).
Figure 1.
Kaplan‐Meier curves for incident hypertension by high‐sensitivity C‐reactive protein (hs‐CRP) and white blood cell (WBC) count quartiles
Table 3.
Age‐adjusted and multivariable‐adjusted hazard ratios for the incidence of hypertension according to quartiles of serum hs‐CRP and white blood cell
| No. of cases/at risk | Age‐adjusted | P for trend | Multivariable‐adjusted | P for trend | |||||
|---|---|---|---|---|---|---|---|---|---|
| HR | 95% CI | P value | HR | 95% CI | P value | ||||
| hs‐CRP (mg/dL) | |||||||||
| ≤0.02 | 67/574 | 1.00 | (reference) | <0.001 | 1.00 | (reference) | 0.03 | ||
| 0.03‐0.05 | 175/920 | 1.48 | (1.11‐1.96) | 0.007 | 1.39 | (1.04‐1.85) | 0.03 | ||
| 0.06‐0.10 | 141/677 | 1.69 | (1.20‐2.38) | <0.001 | 1.46 | (1.08‐1.98) | 0.01 | ||
| ≥0.11 | 196/820 | 1.93 | (1.39‐2.67) | <0.001 | 1.57 | (1.17‐2.11) | 0.003 | ||
| WBC (1000 cells/µL) | |||||||||
| ≤5.4 | 105/714 | 1.00 | (reference) | <0.001 | 1.00 | (reference) | 0.63 | ||
| 5.5‐6.5 | 132/767 | 1.14 | (0.88‐1.47) | 0.33 | 0.97 | (0.75‐1.26) | 0.81 | ||
| 6.6‐7.8 | 152/760 | 1.30 | (1.01‐1.66) | 0.04 | 1.01 | (0.78‐1.31) | 0.96 | ||
| ≥7.9 | 190/750 | 1.59 | (1.25‐2.02) | <0.001 | 1.12 | (0.86‐1.47) | 0.39 | ||
HR, hazard ratio; hs‐CRP, high‐sensitivity C‐reactive protein; WBC, white blood cell.
Multivariable adjustment was made for age, body mass index, systolic blood pressure, low‐density lipoprotein cholesterol, high‐density lipoprotein cholesterol, triglyceride, estimated glomerular filtration rate, hemoglobin A1c, hemoglobin, antidiabetic therapy, lipid‐lowering therapy, uric acid‐lowering therapy, alcohol intake, and current smoking.
We next conducted a subgroup analysis stratified by potential confounders to assess the consistency of the association between hs‐CRP and the incident hypertension by adding an interaction term to the statistical model (Table 4). We estimated the multivariable‐adjusted HRs and 95% CIs for the development of hypertension by comparing the highest with the lowest quartiles of hs‐CPR according to the levels of other risk factors as follows: pairs of groups by age, BMI, waist circumference, alcohol intake habit, current smoking, the presence of diabetes mellitus, the levels of triglycerides, LDL cholesterol, and HDL cholesterol, WBC count, and eGFR or three groups by blood pressure levels (Table 4). We found a significant interaction for the incidence of hypertension between hs‐CRP and the age groups (P = 0.049). The association was stronger in the younger (≤39 years old) subjects. No significant interactions were detected between hs‐CRP and BMI, waist circumference, alcohol intake habit, current smoking, diabetes, triglyceride, HDL cholesterol, LDL cholesterol, WBC count, eGFR, or blood pressure levels. No obvious difference was observed to the results of interaction, when we analyzed the data by using waist circumference instead of BMI for multivariable adjustment in the results of interactions (Table S2).
Table 4.
Multivariable‐adjusted hazard ratios and 95% CIs (highest vs lowest quartiles) for incidence of hypertension in all subjects according to the levels of risk factors
| Parameters | No. of cases/No. at risk | Multivariable‐adjusted hazard ratio (95% CI) | P for interaction | |
|---|---|---|---|---|
| Highest quartile (hs‐CRP ≥0.11) | Lowest quartile (hs‐CRP ≤0.02) | |||
| Age (y) | ||||
| ≤39 | 59/342 | 20/324 | 1.77 (1.02‐3.09) | 0.049 |
| ≥40 | 137/478 | 47/250 | 1.49 (1.05‐2.11) | |
| BMI (kg/m2) | ||||
| <25.0 | 50/296 | 42/407 | 1.72 (1.12‐2.64) | 0.92 |
| ≥25.0 | 146/524 | 25/167 | 1.72 (1.11‐2.66) | |
| Waist circumference (cm) | ||||
| <90.0 | 107/541 | 58/529 | 1.54 (1.11‐2.16) | 0.74 |
| ≥90.0 | 89/279 | 9/45 | 2.20 (1.09‐4.47) | |
| Alcohol intake | ||||
| Without | 68/351 | 10/195 | 1.92 (0.96‐3.85) | 0.78 |
| With | 128/469 | 57/379 | 1.50 (1.07‐2.09) | |
| Current smoking | ||||
| Without | 84/387 | 33/345 | 1.83 (1.20‐2.81) | 0.31 |
| With | 112/433 | 34/229 | 1.38 (0.91‐2.09) | |
| Diabetes mellitus | ||||
| Without | 172/740 | 66/564 | 1.55 (1.15‐2.10) | 0.44 |
| With | 24/80 | 1/10 | 2.02 (0.25‐16.55) | |
| Triglyceride (mg/dL) | ||||
| ≤149 | 91/447 | 53/453 | 1.63（1.14‐2.33） | 0.55 |
| ≥150 | 105/373 | 14/121 | 1.53 (0.85‐2.75) | |
| HDL cholesterol (mg/dL) | ||||
| <40 | 26/110 | 2/13 | 1.25（0.28‐5.64） | 0.24 |
| ≥40 | 170/710 | 65/561 | 1.56（1.15‐2.11） | |
| LDL cholesterol (mg/dL) | ||||
| <140 | 128/571 | 53/470 | 1.57（1.12‐2.21） | 0.56 |
| ≥140 | 68/249 | 14/104 | 1.83（0.97‐3.45） | |
| White blood cell count (1000 cell/µL) | ||||
| ≤6.5 | 55/285 | 38/382 | 1.71 (1.10‐2.65) | 0.79 |
| ≥6.6 | 141/535 | 29/192 | 1.56 (1.02‐2.37) | |
| eGFR (mL/min/1.73 m2) | ||||
| <88 | 116/430 | 35/253 | 1.73 (1.16‐2.58) | 0.92 |
| ≥88 | 80/390 | 32/321 | 1.31 (0.84‐2.06) | |
| Blood pressure level | ||||
| Optimal | 20/288 | 5/263 | 2.74 (1.00‐7.52) | 0.25 |
| Normal | 65/291 | 28/204 | 1.20 (0.75‐1.94) | |
| High normal | 111/241 | 34/107 | 1.58 (1.04‐2.38) | |
BMI, body mass index; eGFR, estimated glomerular filtration ratio; HDL, high‐density lipoprotein; LDL, low‐density lipoprotein; TG, triglyceride.
4. DISCUSSION
Our findings showed that higher levels of hs‐CRP were associated with incident hypertension in a population of middle‐aged Japanese men at a work site. In contrast, the subjects’ WBC count levels were not associated with the development of hypertension after the adjustment for potential confounding factors. In the stratified analysis, the association of hs‐CRP with incident hypertension was stronger in the younger subjects. The presence of obesity and diabetes, current smoking, alcohol intake habit, the levels of triglyceride, HDL cholesterol, and LDL cholesterol, the WBC count, the eGFR, and blood pressure levels did not alter this association between hs‐CRP and incident hypertension. These findings suggest that hs‐CRP is associated with incident hypertension independently of traditional risk factors for cardiovascular diseases.
One of the strengths of our study is that it was the first to evaluate the effects of both CRP and the WBC count on incident hypertension, although several studies showed the influence of inflammatory markers separately with either CRP or the WBC count on the development of hypertension.9, 10, 11, 13, 32 In addition, many previous studies have shown that increase in the CRP levels and WBC count is associated with cardiovascular disease,1, 2, 3 but little is known whether these inflammatory markers are related to incident hypertension especially younger individuals. There are several controversial prior findings regarding the relationship between higher levels of CRP and incident hypertension. Higher levels of CRP were likely to be associated with incident hypertension in studies of middle‐aged and older adults.9, 11, 12 In contrast, in the cross‐sectional British Women's Heart and Health Study investigating women aged 60‐79 years and using a Mendelian randomization approach, elevated levels of CRP were not associated with elevated blood pressure.23 The CARDINA study examined individuals ranging in age from 18 to 30 years; an association of CRP with incident hypertension was not observed after adjusting for BMI.21 It is thus possible that the relationship between CRP and incident hypertension varies in accord with the underlying heterogeneity in the background and the risk factors of the studied subjects. In line with these findings, here, we investigated the association of CRP with incident hypertension focusing on young and middle‐aged subjects (mean age, approx. 40 years), and our analyses revealed that a higher level of CRP was associated with incident hypertension.
Concerning the effects of the WBC count on incident hypertension, a few studies showed that a higher WBC count was associated with the development of hypertension,13, 33 whereas other study24 and our present investigation did not show such an association. The precise reasons for the differences between our result and those of previous studies are not yet known, but heterogeneity in the study design, study participants (including age, sex, and ethnic differences), and the distribution of WBC count values might be involved. In the National Health and Nutrition Examination Survey (NHANES)‐I Epidemiologic Follow‐up Study, a significant relationship between the WBC count and incident hypertension was observed only among older Caucasian men (≥65 years),13 suggesting the involvement of heterogeneity of study subjects in the hypertension/WBC count relationship. In addition, it was reported that the gradient of relative risk of the development of hypertension was 1.4 from the lowest quartile to the highest quartile of WBC count, whereas that of BMI, one of the established risk factors for hypertension, was 3.3.24, 34 Thus, the relationship between the WBC count and incident hypertension might not be strong compared to the confirmed risk factors for hypertension. It has also been suggested that adjustment by a multivariable analysis that includes the subjects’ baseline blood pressure values reduces the association between the WBC count and the subsequent development of hypertension.24 This is consistent with our present findings.
Both the CRP level and the WBC count are commonly measured as markers of systemic inflammation in clinical practice, and both contribute to atherogenesis and plaque disruption.35, 36 In a meta‐analysis of the associations between inflammatory markers and coronary heart disease by Danish et al, the odds ratios of the CRP level and the WBC count were 1.7 (95% CI, 1.4‐2.1) and 1.4 (95% CI 1.3‐1.5), respectively,37 showing that the CRP level and the WBC count have comparable effects on coronary heart disease. However, little is known about the effects of these markers of inflammation at the early stages of atherosclerosis and hypertension. The exact mechanisms underlying the effects of CRP and the WBC count in the development of hypertension must be clarified.
Another strength of our study is that the hazard ratio for the development of hypertension was significantly increased even for the very low levels of CRP observed in the second quartile of CRP (0.03‐0.05 mg/dL) compared to the lowest (≤0.02 mg/dL). Even a slight increase in CRP contributed to blood pressure elevation in our study population. In addition, there might be racial difference in the levels of CRP. The mean value of CRP in our subjects (0.13 mg/dL) is lower than those of Western populations. A statement for health care professionals from the US Centers for Disease Control and Prevention and the American Heart Association describes the CRP levels of <0.1, 0.1‐0.3, and >0.3 mg/dL as low, average, and high relative risk categories, respectively.1 The high‐risk category has an approx. twofold increase in the relative risk of cardiovascular diseases compared to the low‐risk category.1 The Multiethnic Cohort Study (MEC) conducted in the US showed that CRP levels after adjustment for BMI were lower in Japanese Americans (0.11 mg/dL) compared to Caucasian (0.21 mg/dL) or African Americans (0.32 mg/dL).38 The mean CRP of 0.13 mg/dL in the present study is compatible to that of Japanese Americans in the MEC study.
Epidemiological studies showed that higher levels of CRP were associated with obesity, insulin resistance, and diabetes,15, 16 where increased abdominal fat and the resultant insulin resistance may increase the levels of CRP by releasing vasoactive pathophysiological molecules from adipocytes, which cause inflammation in systematic tissues.25, 26 However, our present findings revealed that the effects of CRP on incident hypertension are observed independently of obesity, diabetes, and the levels of HDL cholesterol and triglycerides. In addition, when we evaluated the effects of CRP on incident hypertension in the subgroups divided by with and without other risk factors such as obesity, diabetes mellitus, or dyslipidemia, the effects of CRP on the risk of incident hypertension did not differ between the subjects with and without these risk factors. Of note, when we analyzed the data using waist circumference instead of BMI for multivariable adjustment, no obvious difference was observed compared to that using BMI (Tables S1 and S2). Thus, our results might be independent of abdominal fat although previous studies reported that the incorporating measurement of waist circumference as a measure of abdominal obesity in addition to BMI is important to evaluate the risk of hypertension.39, 40 These findings support the idea that higher CRP levels may contribute to the development of hypertension independently of insulin resistance and traditional risk factors for cardiovascular diseases such as obesity, diabetes, and dyslipidemia. Measurement of CRP may provide additional information for the detection of higher risk of subjects irrespective of traditional risk factors.
In contrast, the subgroup analysis regarding the age groups revealed that significant heterogeneity was observed in the effects of CRP on incident hypertension (P = 0.049), and the relationship between CRP and incident hypertension was stronger in the subjects aged ≤39 years. These results suggest that the higher levels of hs‐CRP among the younger subjects may be associated with a greater risk of incident hypertension. It may be important to evaluate the CRP levels especially in younger subjects irrespective of the presence of traditional risk factors for cardiovascular disease in order to predict and prevent the incidence of hypertension in the future.
Several potential limitations in our study regarding internal and external validities should be discussed. First, a potential limitation on the generalizability of our results should be noted. The present study was conducted focusing on the young and middle‐aged men. We also excluded women from the study, because of their small number; our finding thus cannot be applied to women and older people. Further studies might be required in women and older subjects including other races. Second, we lacked information on drug use, dietary habits, and physical activity. It has shown that some of the lipid‐lowering, antidiabetic, uric acid‐lowering drugs such as statins and metformins might affect the levels of CRP.41, 42 However, the prescription rates of antidiabetic, lipid‐lowering, or uric acid‐lowering drugs were low among our subjects. In addition, the sensitivity analysis excluding the subjects who were prescribed these drugs did not make any obvious difference to the findings (data not shown). Third, we used the lower values of blood pressure at routine annual physical examinations, which might underestimate the incidence of hypertension. Fourth, because this study was an observational investigation, we cannot rule out residual confounding factors despite our careful covariate adjustment.
In conclusion, the levels of CRP were associated with the future incidence of hypertension in a worksite population of middle‐aged Japanese men. The association of the levels of CRP and incident hypertension was stronger in the younger subjects and was observed irrespective of the presence of obesity, diabetes, metabolic syndrome, dyslipidemia, current smoking, and alcohol intake habit, suggesting that the effect of CRP on incident hypertension was independent of metabolic disorders. The measurement of CRP in middle‐aged Japanese males might be effective to predict the future development of hypertension.
CONFLICT OF INTEREST
TO received honoraria from Sanwa Kagaku Kenkyusho. TK received honoraria from Daiichi Sankyo and research funding from Daiichi Sankyo, Takeda Pharmaceutical, Astellas Pharma, Chugai Pharmaceutical, MSD, Boehringer Ingelheim, EA Pharma, Sanofi Aventis, Pfizer, Kissei Pharmaceutical, Kyowa Hakko Kirin, Asahi Kasei Medical, Otsuka Pharmaceutical, Torii Pharmaceutical, and Bayer. The other authors report no conflicts.
Supporting information
Kansui Y, Matsumura K, Morinaga Y, et al. C‐reactive protein and incident hypertension in a worksite population of Japanese men. J Clin Hypertens. 2019;21:524–532. 10.1111/jch.13510
Funding information
This study was supported in part by the Private University Research Branding Project.
REFERENCES
- 1. Pearson TA, Mensah GA, Alexander RW, et al. Markers of inflammation and cardiovascular disease: application to clinical and public health practice: a statement for healthcare professionals from the Centers for Disease Control and Prevention and the American Heart Association. Circulation. 2003;107:499‐511. [DOI] [PubMed] [Google Scholar]
- 2. Ridker PM, Cushman M, Stampfer MJ, Tracy RP, Hennekens CH. Inflammation, aspirin, and the risk of cardiovascular disease in apparently healthy men. N Engl J Med. 1997;336:973‐979. [DOI] [PubMed] [Google Scholar]
- 3. Bisoendial RJ, Kastelein J, Levels J, et al. Activation of inflammation and coagulation after infusion of C‐reactive protein in humans. Circ Res. 2005;96:714‐716. [DOI] [PubMed] [Google Scholar]
- 4. Pepys MB, Hirschfield GM, Tennent GA, et al. Targeting C‐reactive protein for the treatment of cardiovascular disease. Nature. 2006;440:1217‐1221. [DOI] [PubMed] [Google Scholar]
- 5. Kannel WB, Anderson K, Wilson PW. White blood cell count and cardiovascular disease. Insights from the Framingham Study. JAMA. 1992;267:1253‐1256. [PubMed] [Google Scholar]
- 6. Doi Y, Kiyohara Y, Kubo M, et al. C‐reactive protein is a predictor of the development of diabetes in a general Japanese population: the Hisayama Study. Diabetes Care. 2005;28:2497‐2500. [DOI] [PubMed] [Google Scholar]
- 7. Grau AJ, Boddy AW, Dukovic DA, et al. Leukocyte count as an independent predictor of recurrent ischemic events. Stroke. 2004;35:1147‐1152. [DOI] [PubMed] [Google Scholar]
- 8. Twig G, Afek A, Shamiss A, et al. White blood cells count and incidence of type 2 diabetes in young men. Diabetes Care. 2013;36:276‐282. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Sesso HD, Buring JE, Rifai N, Blake GJ, Gaziano JM, Ridker PM. C‐reactive protein and the risk of developing hypertension. JAMA. 2003;290:2945‐2951. [DOI] [PubMed] [Google Scholar]
- 10. Chuang SY, Hsu PF, Chang HY, Bai CH, Yeh WT, Pan HW. C‐reactive protein predicts systolic blood pressure and pulse pressure but not diastolic blood pressure: the cardiovascular disease risk factors two‐township study. Am J Hypertens. 2013;26:657‐664. [DOI] [PubMed] [Google Scholar]
- 11. Niskanen L, Laaksonen DE, Nyyssönen K, et al. Inflammation, abdominal obesity, and smoking as predictors of hypertension. Hypertension. 2004;44:859‐865. [DOI] [PubMed] [Google Scholar]
- 12. Nakanishi N, Sato M, Shirai K, Suzuki K, Tatara K. White blood cell count as a risk factor for hypertension: a study of Japanese male office workers. J Hypertens. 2002;20:851‐857. [DOI] [PubMed] [Google Scholar]
- 13. Gillum RF, Mussolino ME. White blood cell count and hypertension incidence. The NHANES I epidemiologic follow‐up study. J Clin Epidemiol. 1994;47:911‐919. [DOI] [PubMed] [Google Scholar]
- 14. Tatsukawa Y, Hsu W‐L, Yamada M, et al. White blood cell count, especially neutrophil count, as a predictor of hypertension in a Japanese population. Hypertens Res. 2008;31:1391‐1397. [DOI] [PubMed] [Google Scholar]
- 15. Ford ES. Body mass index, diabetes, and C‐reactive protein among U.S. adults. Diabetes Care. 1999;22:1971‐1977. [DOI] [PubMed] [Google Scholar]
- 16. Festa A, D'Agostino R Jr, Howard G, Mykkänen L, Tracy RP, Haffner SM. Chronic subclinical inflammation as part of the insulin resistance syndrome: the Insulin Resistance Atherosclerosis Study (IRAS). Circulation. 2000;102:42‐47. [DOI] [PubMed] [Google Scholar]
- 17. Pradhan AD, Cook NR, Buring JE, Manson JE, Ridker PM. C‐reactive protein is independently associated with fasting insulin in nondiabetic women. Arterioscler Thromb Vasc Biol. 2003;23:650‐655. [DOI] [PubMed] [Google Scholar]
- 18. Pecht T, Gutman‐Tirosh A, Bashan N, Rudich A. Peripheral blood leucocyte subclasses as potential biomarkers of adipose tissue inflammation and obesity subphenotypes in humans. Obes Rev. 2014;15:322‐337. [DOI] [PubMed] [Google Scholar]
- 19. Kianoush S, Bittencourt MS, Lotufo PA, et al. Association between smoking and serum GlycA and high‐sensitivity C‐reactive protein levels: the Multi‐Ethnic Study of Atherosclerosis (MESA) and Brazilian Longitudinal Study of Adult Health (ELSA‐Brasil). J Am Heart Assoc. 2017;6:e006545. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Fisch IR, Freedman SH. Smoking, oral contraceptives, and obesity. Effects on white blood cell count. JAMA. 1975;234:500‐506. [PubMed] [Google Scholar]
- 21. Lakoski SG, Herrington DM, Siscovick DM, Hulley SB. C‐reactive protein concentration and incident hypertension in young adults: the CARDIA study. Arch Intern Med. 2006;166:345‐349. [DOI] [PubMed] [Google Scholar]
- 22. Krishnan E. Interaction of inflammation, hyperuricemia, and the prevalence of hypertension among adults free of metabolic syndrome: NHANES 2009–2010. J Am Heart Assoc. 2014;3:e000157. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Smith GD, Lawlor DA, Harbord R, et al. Association of C‐reactive protein with blood pressure and hypertension: life course confounding and mendelian randomization tests of causality. Arterioscler Thromb Vasc Biol. 2005;25:1051‐1056. [DOI] [PubMed] [Google Scholar]
- 24. Friedman GD, Selby JV, Quesenberry CP Jr. The leukocyte count: a predictor of hypertension. J Clin Epidemiol. 1990;43:907‐911. [DOI] [PubMed] [Google Scholar]
- 25. Weisberg SP, McCann D, Desai M, Rosenbaum M, Leibel RL, Ferrante AW Jr. Obesity is associated with macrophage accumulation in adipose tissue. J Clin Invest. 2003;112:1796‐1808. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Collaboration E, Kaptoge S, Di Angelantonio E, et al. C‐reactive protein concentration and risk of coronary heart disease, stroke, and mortality: an individual participant meta‐analysis. Lancet. 2010;375:132‐140. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Saijo Y, Kiyota N, Kawasaki Y, et al. Relationship between C‐reactive protein and visceral adipose tissue in healthy Japanese subjects. Diabetes Obes Metab. 2004;6:249‐258. [DOI] [PubMed] [Google Scholar]
- 28. Committee of the Japan Diabetes Society on the Diagnostic Criteria of Diabetes Mellitus , Seino Y, Nanjo K, et al. Report of the committee on the classification and diagnostic criteria of diabetes mellitus. J Diabetes Investig. 2010;1:212‐228. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Matsuo S, Imai E, Horio M, et al. Collaborators developing the Japanese equation for estimated GFR. Revised equations for estimated GFR from serum creatinine in Japan. Am J Kidney Dis. 2009;53:982‐992. [DOI] [PubMed] [Google Scholar]
- 30. Shimamoto K, Ando K, Fujita T, et al. Japanese Society of Hypertension Committee for Guidelines for the Management of Hypertension. The Japanese Society of Hypertension Guidelines for the Management of Hypertension (JSH 2014). Hypertens Res. 2014;37:253‐390. [DOI] [PubMed] [Google Scholar]
- 31. International Expert Committee . International Expert Committee report on the role of the A1C assay in the diagnosis of diabetes. Diabetes Care. 2009;32:1327‐1334. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Cheung B, Ong Kl, Tso A, et al. C‐reactive protein as a predictor of hypertension in the Hong Kong Cardiovascular Risk Factor Prevalence Study (CRISPS) cohort. J Hum Hypertens. 2012;26:108‐116. [DOI] [PubMed] [Google Scholar]
- 33. Shankar A, Klein BE, Klein R. Relationship between white blood cell count and incident hypertension. Am J Hypertens. 2004;17:233‐239. [DOI] [PubMed] [Google Scholar]
- 34. Friedman GD, Selby JV, Quesenberry CP Jr, Armstrong MA, Klatsky AL. Precursors of essential hypertension: body weight, alcohol and salt use, and parental history of hypertension. Prev Med. 1988;17:387‐402. [DOI] [PubMed] [Google Scholar]
- 35. Yousuf O, Mohanty BD, Martin SS, et al. High‐sensitivity C‐reactive protein and cardiovascular disease: a resolute belief or an elusive link? J Am Coll Cardiol. 2013;62:397‐408. [DOI] [PubMed] [Google Scholar]
- 36. Madjid M, Awan I, Willerson JT, Casscells SW. Leukocyte count and coronary heart disease: implications for risk assessment. J Am Coll Cardiol. 2004;44:1945‐1956. [DOI] [PubMed] [Google Scholar]
- 37. Danesh J, Collins R, Appleby P, Peto R. Association of fibrinogen, C‐reactive protein, albumin, or leukocyte count with coronary heart disease: meta‐analyses of prospective studies. JAMA. 1998;279:1477‐1482. [DOI] [PubMed] [Google Scholar]
- 38. Morimoto Y, Conroy Sm, Ollberding Nj, et al. Ethnic differences in serum adipokine and C‐reactive protein levels: the multiethnic cohort. Int J Obes (Lond). 2014;38:1416‐1422. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39. Zhang M, Zhao Y, Wang G, et al. Body mass index and waist circumference combined predicts obesity‐related hypertension better than either alone in a rural Chinese population. Sci Rep. 2016;6:31935. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40. Janssen I, Katzmarzyk PT, Ross R. Body mass index, waist circumference, and health risk: evidence in support of current National Institutes of Health guidelines. Arch Intern Med. 2002;162:2074‐2079. [DOI] [PubMed] [Google Scholar]
- 41. Tsirpanlis G. The pattern of inflammation and a potential new clinical meaning and usefulness of C‐reactive protein in end‐stage renal failure patients. Kidney Blood Press Res. 2005;28:55‐61. [DOI] [PubMed] [Google Scholar]
- 42. Morin‐Papunen L, Rautio K, Ruokonen A, Hedberg P, Puukka M, Tapanainen JS. Metformin reduces serum C‐reactive protein levels in women with polycystic ovary syndrome. J Clin Endocrinol Metab. 2003;88:4649‐4654. [DOI] [PubMed] [Google Scholar]
- 43. Sezai A, Obata K, Abe K, Kanno S, Sekino H. Cross‐over trial of febuxostat and topiroxostat for hyperuricemia with cardiovascular disease (TROFEO Trial). Circ J. 2017;81:1707‐1712. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials