Association between fasting plasma glucose and high-sensitivity C-reactive protein: gender differences in a Japanese community-dwelling population

C-reactive protein (CRP) is an acute phase reactant and a sensitive marker of inflammation. Several studies support the concept that high-sensitivity C-reactive protein (hsCRP), even when within the clinical normal range, is an important precursor of the metabolic syndrome (MetS) and type 2 diabetes [1‐3], and it may be an independent predictor that reflects early stage cardiovascular disease (CVD) [4‐6].

A recent review of 20 studies revealed that there was also a significant exponential association between glucose and CVD in nondiabetic participants that extended below the usual "diabetic threshold" [7], and fasting plasma glucose (FPG) is an important predictor of CVD after adjusting for potential confounders. Experimental studies have shown that hyperglycemia stimulates the release of the inflammatory cytokines tumor necrosis factor-alpha (TNF-α) and interleukin-6 (IL-6) from various cells such as monocytes [8]. Recent data have demonstrated that hsCRP is stimulated and produced in the liver by proinflammatory cytokines (e.g., TNF-α and IL-6) produced by visceral adiposity [9]. Thus, elevated FPG is associated with elevated concentrations of hsCRP [10‐12].

On the other hand, Nakanishi et al. reported that hsCRP levels are much higher in Westerners than in Japanese [13], and in women compared with men [3, 14]. In addition, gender differences have been reported to be consistent across all ethnic subgroups even after multivariable adjustment [14]. However, the question of whether modification by gender has an effect on the association between FPG and inflammation in Japanese has not been investigated in detail.

The aim of this study was to determine whether elevated FPG levels are associated with elevated hsCRP concentrations, and whether this association is independent of gender, body mass index (BMI) and other confounders of CVD. We examined cross-sectional data from Japanese community-dwelling participants.

In this cross-sectional, population-based study, we determined the relation between glucose levels and hsCRP by gender in healthy participants. This study showed that hsCRP levels increase continuously across the spectrum of FPG, starting from the lowest FPG in both men and women. When analyzing FPG and BMI simultaneously, it is apparent that increased hsCRP levels are mainly due to increased hsCRP concentrations according to the status of obesity. These findings suggest that the FPG-hsCRP relationship is stronger for subjects with obesity than for those without. In addition, increase in hsCRP levels was greater in women than men, and we demonstrated that there is a gender interaction between increased FPG and hsCRP after adjustment for age, BMI, smoking status, SBP, HDL-C, uric acid, FPG, and HMW adiponectin. To our knowledge, few studies indicate modification of the effect by gender in the association between FPG and inflammation in Japanese community-dwelling adults.

Several previous studies show that gender, FPG and obesity are independent confounding factors for increased hsCRP. In a total of 1,737 Korean subjects aged >60 years, multiple regression analysis using hsCRP as a dependent variable showed that BMI and white blood cell count as well as fasting insulin, post-load 2 h glucose, hematocrit and LDL-C were significant independent variables [16]. In 520 Chinese patients with type 2 diabetes, insulin resistance was associated with CRP levels independent of abdominal obesity [17]. Unek et al. [18] report that obese patients show a significant increase in hsCRP and soluble CD40 ligand levels compared with normal weight subjects, which might contribute to the known proinflammatory milieu found in these patients. In a total of 16,616 men and non-pregnant women aged ≥ 17 years, after adjustment for potential confounders, the adjusted OR of obese men and women for increased CRP was 2.13 (95% CI, 1.56-2.91) and 6.21 (95% CI, 4.94-7.81), respectively. In addition, BMI was associated with increased CRP levels in women only [19]. In a population-based cross-sectional study, fasting glucose remained significantly and independently related to CRP levels [10, 11], and hsCRP as well as the insulin resistance indices increased gradually even in the normal fasting glucose range, supporting the rationale for expanding the range of fasting hyperglycemia [12]. Moreover, an increase in CRP levels was greater in women than men in the obesity and higher fasting glucose category, and this association was more pronounced in women [11]. In our study, increased FPG levels were positively associated with increased hsCRP, independent of other confounders, especially in obese women and the association was also shown to be within the normal range in obese participants.

There are several possible mechanisms linking FPG, gender, obesity and hsCRP. This result shows that high FPG levels can increase levels of key proteins (e.g., hsCRP) that result in inflammation. In experimental studies, human monocytes show a glucose-dependent increase in TNF-α and IL-6 production [20‐22]. In female Turkish subjects, subjects with impaired glucose tolerance had higher fasting serum TNF-α levels than subjects with normal glucose tolerance (p < 0.01), and serum TNF-α and IL-6 concentrations were elevated during the glucose challenge (for each comparison, p < 0.01) [23]. It has also been demonstrated that IL-6 and high glucose synergistically stimulate matrix metalloproteinase-1 (MMP-1) expression in mononuclear phagocytes, and IL-6 and high glucose synergistically upregulate MMP-1 expression by U937 mononuclear phagocytes [24]. Obesity per se is also an important risk factor of the development of type 2 diabetes and is associated with low-grade systemic inflammation [25]. Furthermore, Fat accumulation in the liver or adipose tissues can induce inflammatory cytokines such as TNF-α, IL-6, and IL-8 [26]. These cytokines produced by adipocytes stimulate hepatic synthesis of CRP, which is an acute-phase protein, and influence insulin resistance as well as lipid and glucose metabolisms [3, 27‐29]. All these factors result in the production of CRP.

Other mechanisms may contribute to the altered glucose metabolism demonstrated. Adiponectin is an adipocyte-secreted cytokine, which has insulin sensitizing, anti-inflammatory, and anti-atherosclerotic properties [30, 31], and is decreased in persons that are obese and have diabetes [32]. Decreased adiponectin could affect blood glucose and systemic inflammation. However, in our study, an association between FPG and hsCRP remained significant after adjusting for confounding factors including BMI and HMW adiponectin.

Our finding showed that gender interaction in the association between MetS and hsCRP suggest that hormones may play an important role in the inflammatory mechanism [33, 34]. In our study, increased BMI was associated with increased hsCRP concentrations in both genders, however, the correlation coefficients between hsCRP and BMI were much higher in women (r = 0.355) than in men (r = 0.188). These results demonstrated that inflammatory response may be more enhanced through adipose tissue-secreted cytokine expression in women [11]. Thus, sex hormones may be responsible for the association of fasting glucose with hsCRP. In postmenopausal women, this gender-specific effect of fasting glucose on inflammation might help to explain an increase in the relative risk for MetS and CVD [35]

Several limitations should be considered in this study. First, our cross-sectional study design does not eliminate potential causal relationships between CRP and FPG. Second, FPG categories are based on a single assessment of blood, which may introduce a misclassification bias. Third, we could not eliminate the possible effect of medication for hypertension on the present findings. Therefore the demographics and referral source may limit generalizability.

In conclusion, the present study showed that FPG levels are strongly associated with hsCRP in the general population with obesity. The underlying mechanism behind this relationship is unclear, but seems to be independent of traditional cardiovascular risk factors such as age, BMI, smoking status, alcohol consumption, blood pressure, lipids, or uric acid. Thus, the FPG levels (especially, higher values of normal FPG, IFG, and women) in persons with obesity might provide an important marker for the assessment of atherogenic risk as well as a therapeutic target for the modification of atherogenic risks. For healthy community-dwelling persons, prospective population-based studies are needed to investigate the mechanisms underlying this association to determine whether intervention, such as effective lifestyle modifications and medication (e.g., antihypertensive, antilipidemic, and diabetic medication) that decrease BMI and FPG in adults, will decrease hsCRP.