Diabetes Metab J > Volume 44(5); 2020 > Article
Cho, Lee, Shim, and Kim: Sex-, Age-, and Metabolic Disorder-Dependent Distributions of Selected Inflammatory Biomarkers among Community-Dwelling Adults

Abstract

Background

Inflammatory cytokines are increasingly utilized to detect high-risk individuals for cardiometabolic diseases. However, with large population and assay methodological heterogeneity, no clear reference currently exists.

Methods

Among participants of the Cardiovascular and Metabolic Diseases Etiology Research Center cohort, of community-dwelling adults aged 30 to 64 without overt cardiovascular diseases, we presented distributions of tumor necrosis factor (TNF)-α and -β, interleukin (IL)-1α, -1β, and 6, monocyte chemoattractant protein (MCP)-1 and -3 and high sensitivity C-reactive protein (hsCRP) with and without non-detectable (ND) measurements using multiplex enzyme-linked immunosorbent assay. Then, we compared each markers by sex, age, and prevalence of type 2 diabetes mellitus, hypertension, and dyslipidemia, using the Wilcoxon Rank-Sum Test.

Results

In general, there were inconsistencies in direction and magnitude of differences in distributions by sex, age, and prevalence of cardiometabolic disorders. Overall, the median and the 99th percentiles were higher in men than in women. Older participants had higher TNF-α, high sensitivity IL-6 (hsIL-6), MCP-1, hsCRP, TNF-β, and MCP-3 median, after excluding the NDs. Participants with type 2 diabetes mellitus had higher median for all assayed biomarkers, except for TNF-β, IL-1α, and MCP-3, in which the medians for both groups were 0.00 due to predominant NDs. Compared to normotensive group, participants with hypertension had higher TNF-α, hsIL-6, MCP-1, and hsCRP median. When stratifying by dyslipidemia prevalence, the comparison varied significantly depending on the treatment of NDs.

Conclusion

Our findings provide sex-, age-, and disease-specific reference values to improve risk prediction and diagnostic performance for inflammatory diseases in both population- and clinic-based settings.

INTRODUCTION

The utility of inflammatory biomarker assays for detecting individuals at high-risk for cardiovascular and metabolic diseases has become increasingly prevalent in clinical and research settings [1,2,3,4,5]. In adjunct to the conventional chemistry and imaging tests, these biomarkers aid in predicting or diagnosing cardiovascular and metabolic diseases in the context of an appropriate clinical presentation [6,7,8].
Considering that the major contribution of inflammatory mechanisms to cardiometabolic diseases has been repeatedly emphasized, many efforts are continuously being made to discover novel markers and assay methodologies. In particular, cytokines/chemokines are critical in homeostatic trafficking and positioning of immune cells in response to inflammation [9,10,11]; they trigger and modify intracellular signaling pathways, thereby orchestrating innate and adaptive immune responses [12].
Of many, interleukin (IL), monocyte chemotactic proteins (MCP) and tumor necrosis factors (TNF) are critical to macrophagic inhibition [13]. They activate growth factors, stimulate procoagulant activity, and suppress antithrombotic pathways in endothelial cells [6]. By measuring their expression in atherosclerotic lesions, previous studies have observed the association between the aforementioned biomarkers and cardiometabolic outcomes, such as type 2 diabetes mellitus (T2DM), myocardial infarction (MI) and more [14,15,16]. However, particularly with the advent of newer assay methodologies and large population heterogeneity, no clear consensus exists regarding the universal reference range. Moreover, little is known about the distribution of these biomarkers across healthy community population.
To address this knowledge gap, the objective of this study was to determine distributions of selected inflammatory biomarkers among middle-aged Koreans. Then, we compared their descriptive statistics by sex, age, and metabolic disorders, including T2DM, hypertension and dyslipidemia.

METHODS

Approvals and participant consent

As part of the ethical committee process, the Cardiovascular and Metabolic Diseases Etiology Research Center (CMERC) Study has been approved by the Institutional Review Boards of Severance Hospital, Yonsei University Health System, Seoul, Korea (4-2013-0661) and Ajou University Hospital, Suwon, Korea (AJIRB-BMR-SUR-13-272). Written informed consent has been obtained from all participants prior to the baseline survey. Participants were ensured that they can withdraw from the study at any time, regardless of its cause, without any repercussions.

Study population

The CMERC cohort is a multi-centered, prospective, observational study of community-dwelling population residing in Seoul and capital regions in Republic of Korea [17]. The participants are between the age of 30 to 64 years and without history overt cardiovascular diseases (CVDs), recent history of malignant cancer and autoimmune or chronic inflammatory diseases. In the present study, among 8,108 participants who have undergone baseline examination between 2013 and 2018, 4,058 participants (excluding two with missing variables) from Center 1 underwent high sensitivity C-reactive protein (hsCRP) measurements. Among them, a random subset of participants enrolled between 2013 to 2015 underwent additional measurements via multiplex enzyme-linked immunosorbent assay (ELISA) assay, including TNF-α, TNF-β, IL-1α, IL-1β, hsIL-1β, IL-6, hsIL-6, MCP-1, and MCP-3. The varying number of measurements for each biomarker was due to different number of non-detectable (ND) measurements, requiring higher sensitivity analyses.

Data collection

Overnight-fasting blood samples were obtained in the morning. All biomarker assays were performed in accordance to standard protocols at a single laboratory (Seoul Clinical Laboratory R&D Center, Seoul, Korea).

hsCRP

We collected bubble-free serum samples in plastic containers and allowed them to clot for at least 30 minutes before centrifugation for 10 minutes at 1,000×g. Then, we aliquot and stored the samples below −20℃. During the sample preparation, we allowed all regents to warm to room temperature, premixed beat bottle for 30 seconds, vortexed for 1 minute and centrifuged thoroughly prior to use. No samples were frozen nor thawed for multiple times.
The samples were measured by the immunoturbidimetric CRP-N Assay LA CRP-S Nittobo D-Type high sensitive assay (Nittobo, Tokyo, Japan). The samples were loaded and measured using the latex-enhanced nephelometry technique on an automated analyzer ADVIA 1800 Chemistry System (Siemens Medical Sol., Malvern, PA, USA). The assays were based on the principle of particle-enhanced immunological agglutination method with a commercial test kit (N assay LA, CRP-S, Nittobo). A human CRP calibrator N assay LA CRP-S multi-point was employed to delineate the calibration curve.
The lower and upper detection limits of the hsCRP assay were 0.02 and 40 mg/dL, respectively. The functional sensitivity of the hsCRP assay was 0.45 mg/dL. To calculate intra-assay variability, we repeated the pool serum assay 20 times; the absorbance coefficient variability was less than 5%. Measurement was done in duplicates and any duplicates that were not within a three assay standard deviation from one another were re-run.

Cytokines and chemokines

The sample collection and preparation methods are the same as the aforementioned hsCRP assay. We quantitatively determined the steady state level of the circulating inflammatory cytokines/chemokines of interest: TNF-α, TNF-β, IL-1α, IL-1β, IL-6, MCP-1, and MCP-3. Their serum concentrations were measured using a Milliplex MAP Human Cytokine/Chemokine Multiplex Bead-based kit (Millipore, Burlington, MA, USA), with a 38-plex (HMCP3-MAG, HIL1A-MAG, HCYIL1B-MAG, HCYIL6-MAG, HCYMCP1-MAG, HCYTNFA-MAG, HTNFB-MAG) Millipore Human Cytokine Panel Kits. Specifically, 25 µL of serum was incubated with fluorescently labeled capture antibody-coated beads in a 96-well filter bottomed plate on a plate shaker overnight at 4℃. After incubation, the sample bead mix was removed, and the plate was washed two times using a vacuum manifold. The beads were resuspended in sheath fluid for 5 minutes on the plate shaker. Distinctly colored bead sets of 500 5.6-Nm polystyrene microspheres or 80 6.45-µm magnetic microspheres were created, each of which was coated with a distinctive capture antibody. After an analyte from the sample was captured by the bead, a biotinylated detection antibody was introduced and incubated on a plate shaker at room temperature for 30 minutes. The reaction mixture was incubated with streptavidin-phycoerythrin (Streptavidin-PE) conjugate to complete the reaction on the surface of each microsphere.
The Luminex Bio-Plex 100 analyzer (MAGPIX) identified individual microsphere, and the results were quantified based on fluorescent reporter signals using Luminex xPONENT acquisition software, Milliplex Analyst 5.1. We analyzed the median fluorescence intensity (MFI) using a 5-parameter logistic or spline curve-fitting method to calculate cytokine/chemokines concentrations in each sample. The Luminex MAGPIX instrument was calibrated with the MAGPIX Calibration Kit (EMD Millipore Catalog #40-049), and the performance was verified with the MAGPIX Performance Verification Kit (EMD Millipore Catalog #40-050).
All assays were performed by the same operator according to the manufacturers' instructions. For quality assurance, each sample was run twice, and the mean derived for each sample was used as the index value. Additionally, we reconstituted two kit-supplied quality control (#1, 2) with 250 µL of deionized water to run on each plate in duplicate. After inverting the vial several times, we allowed the vial to sit for 5 to 10 minutes, and then transferred the controls to appropriately labeled polypropylene microfuge tubes. We confirmed the samples to fall within the expected range in accordance with the kit-specific protocols provided by Millipore. Less than 0.5% cross-reactivity and interference were observed.

High sensitivity cytokines

The sample collection and preparation methods are the same as the aforementioned cytokine/chemokine assay. Here, the samples underwent a 2-fold dilution, maintaining a 125 µL of sample and 125 µL of Calibrator Diluent RD6-40 ratio, respectively.
We quantitatively determined the steady state level of the circulating inflammatory cytokines of interest: hsIL-1 and hsIL-6. Their serum concentrations were modulated using Milliplex MAP Human High Sensitivity Cytokine/Chemokine Base kit A (Millipore, Billerica, Burlington, USA) on a 96-well filter bottomed plate. Analyte-specific antibodies were pre-coated onto color-coded magnetic microparticles. Microparticles, standards, and samples were pipetted into wells, and the immobilized antibodies captured the analytes of interest. After washing away unbound substances, corresponding biotinylated antibody cocktail was added to each well. After a thorough wash, Streptavidin-PE conjugate was added to each well. A final wash removed residual unbound conjugate, and the microparticles were resuspended in buffer. The Luminex Bio-Plex 100 analyzer (MAGPIX) identified individual microsphere, and the results were quantified based on fluorescent reporter signals within 90 minutes of the run. The results were analyzed using Luminex xPONENT acquisition software, Milliplex Analyst 5.1.
To calculate a 4-fold dilution for the remaining levels, we referred to the standard concentrations provided by the manufacturer. We averaged the duplicate reading for each standard and sample, and subtracted the average blank MFI using a 5-parameter spline curve-fitting method. Since the samples were diluted, the concentration read from the standard curve were multiplied by the corresponding dilution factor. The Luminex MAGPIX instrument was calibrated with the MAGPIX Calibration Kit (EMD Millipore Catalog #40-049), and its performance was verified with the MAGPIX Performance Verification Kit (EMD Millipore Catalog #40-050).
Similar to the aforementioned cytokine and chemokine assay, we employed the manufacturer-recommended quality control (#11). Likewise, less than 0.5% cross-reactivity and interference were observed.
Data for each kit was analyzed with strict adherence to the manufacturers' guidelines. The proportion of the samples that had both readings within the accepted recovery range (between 70% and 130%) was determined. The reproducibility of the different multiplex methods was evaluated by describing the limits of agreement between duplicates in range for each combination of method and analyte using the Bland-Altman test.

Type 2 diabetes mellitus

Fasting plasma glucose levels were measured using colorimetry method (ADVIA1800 Auto Analyzer; Siemens Medical Sol.), and glycosylated hemoglobin (HbA1c) measurements were obtained via high-performance liquid chromatography (Variant II Turbo Hemoglobin Testing System; Bio-Rad., Hercules, CA, USA). T2DM was defined based on the Korean Clinical Practice Guidelines for T2DM [18]: participants with fasting glucose level ≥126 mg/dL, HbA1c ≥6.5% or current use of oral glucose-lowering drugs or insulin injection were considered to have T2DM.

Hypertension

Blood pressure was measured using a single-arm automated oscillometric device (HEM-7080; Omron Health, Matsusaka, Japan). Information regarding hypertension treatment was obtained via self-report. Hypertension was defined according to the 2013 Korean Society of Hypertension guidelines for the management of hypertension [19], equivalent to the Eight Joint National Committee guidelines [20]: participants with a mean systolic blood pressure (SBP) ≥140 mm Hg, diastolic blood pressure (DBP) ≥90 mm Hg or currently using antihypertensive medications were considered to have hypertension.

Dyslipidemia

Total cholesterol (TC), triglycerides (TG), high-density lipoprotein cholesterol (HDL-C), and low-density lipoprotein cholesterol (LDL-C) levels were analyzed enzymatically with an ADVIA 1800 Auto Analyzer. Dyslipidemia was defined based on the Korean Dyslipidemia Diagnosis criteria [21], equivalent to Adult Treatment Panel III (ATP III) guidelines [22]. Hypercholesterolemia was defined as TC ≥240 mg/dL; hypertriglyceridemia was defined as TG ≥200 mg/dL; hypoalphalipoproteinemia was defined as HDL-C <40 mg/dL; hyper-LDL-cholesterolemia was defined as LDL-C ≥160 mg/dL. Having any one type of the aforementioned cholesterol abnormality or current intake of lipid-lowering was regarded as prevalent dyslipidemia.

Statistical analysis

We defined the minimum, median, 75th, 90th, 95th, 99th percentile, maximum and mean levels for each biomarker using a non-parametric procedure. We presented distribution statistics of each inflammatory marker with and without NDs below the mechanical detection threshold. We additionally stratified by sex, age, T2DM, hypertension and dyslipidemia prevalence, and compared using Wilcoxon rank-sum test. No detectable outliers were excluded in the analysis to preserve the natural integrity of the cohort characteristics. All statistical analyses were performed using SAS version 9.4 (SAS Institute Inc., Cary, NC, USA).

RESULTS

A total of 4,058 participants underwent hsCRP measurements between 2013 and 2018 (Fig. 1). Among them, 1,285 participants whom enrolled in the CMERC study between 2013 and 2015, also underwent TNF-α, IL-1α, IL-1β, IL-6, and MCP-1 and three measurements. Additionally, we assayed hsIL-6 and hsIL-1β from 819 participants and TNF-β from 466 participants. TNF-α, IL-1α, IL-1β, IL-6, and MCP-3 had varying number of NDs.
At baseline, 35.1% of the participants were male, with mean age of 51.3 years (Table 1). In terms of prevalence of cardiometabolic disorder, 372 participants had T2DM (9.2%), 1,070 participants had hypertension (26.4%) and 1,882 participants had dyslipidemia (46.4%). Lifestyle factors (i.e., current smoking and drinking status), anthropometric measurements (i.e. body mass index) and biomarkers related to CVD risk differed significantly by sex.
Table 2, Supplementary Tables 1, 2, 3, 4, and Supplementary Figs 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15 illustrate the distribution of each inflammatory biomarker, including and excluding NDs. Five biomarkers had varying proportion of NDs: TNF-β (70.0%), IL-1α (72.8%), IL-1β (8.3%), IL-6 (8.2%), and MCP-3 (60.8%). In general, the descriptive statistics with and without NDs were different when comparing their respective median and upper percentiles. The overall skewedness to right primarily resulted from upper outliers and zero coding of the NDs.
Table 3 illustrates the Spearman's rank correlation coefficient. Overall, the markers showed weak to moderate positive association among each other. The exceptions are as follows: MCP-1 and TNF-β (r=−0.13), IL-1α and hsIL-6 (r=−0.04), and MCP-1 and MCP-3 (r=−0.01) showed weak yet negative associations, which persisted even after excluding NDs.
For all biomarkers, the median and the 99th percentiles were higher in male than in female regardless of NDs (Tables 4 and 5). For example, the median and the 99th percentile IL-6 values were 1.38 and 345.97 pg/mL in male and 0.37 and 280.06 pg/mL in female, respectively (P<0.001). Such sex-difference extended in high sensitivity analysis, where the median and the 99th percentile hsIL-6 values were 0.83 and 14.06 pg/mL in men and 0.75 and 7.08 pg/mL in women, respectively. Other IL, TNF and MCP subtypes demonstrated similar trend.
When stratifying by the population median age of approximately 55 years, the older group had significantly higher hsIL-6, MCP-1 and hsCRP median (Tables 4 and 5). In contrast, the median value for hsIL-1β was higher in the younger group. Furthermore, the 99th percentile value for hsIL-6 was also markedly higher in the younger group, even after excluding the NDs.
The distributions of the examined markers differed by the presence of each cardiometabolic disorder. Participants with T2DM had higher median for all assayed biomarkers, except for TNF-β, IL-1α, and MCP-3, in which the medians for both groups were 0.00 due to predominant NDs (Table 4). However, when excluding the NDs, the medians for IL-1α and MCP-3 were lower in participants with T2DM than their counterpart (35.24 pg/mL vs. 103.37 pg/mL and 29.85 pg/mL vs. 31.34 pg/mL, respectively), but the differences were statistically insignificant (Table 5). The 99th percentile values for TNF-β, IL-1α, IL-1β, hsIL-1β, hsIL-6, and hsCRP were higher in participants with T2DM, when considering NDs into account.
Compared to the normotensive group, participants with hypertension had higher median TNF-α, hsIL-6, MCP-1, and hsCRP and 99th percentile TNF-β, MCP-3, and hsCRP values. These differences generally persisted even after excluding NDs.
Participants with elevated lipid levels had higher median TNF-α, IL-1β, hsIL-1β, hsIL-6, MCP-1, and hsCRP values than those within normal cholesterol range. However, when examining distributions without NDs, participants without dyslipidemia had higher median TNF-β, IL-1α, IL-6, and MCP-3 levels. In terms of the 99th percentile, participants with dyslipidemia had higher IL-1α, hsIL-1β, hsIL-6, and MCP-1 levels, and such difference remained after excluding NDs.

DISCUSSION

In this middle-aged community-dwelling Korean population, we presented the distributions of varying groups of inflammatory biomarkers, which were distinctive by sex, age and prevalence of cardiometabolic disorders. No explicit recommendations nor universal reference range currently exist for these cytokines/chemokines, owing to insufficient evidence from the general population. Without existing manufacturer-recommended 99th percentile nor referent range, we illustrated their distributions detected in relatively healthy population without any data refinement. These results have important clinical implications that are almost certainly relevant to the application of all modern multiplex assays to both epidemiologic and clinical studies.

hsCRP

Previous literature in this field has noted heterogeneous distributions and reference ranges derived from their own study population. In a Chinese population, the median hsCRP was 0.99 mg/L in the middle-aged group and 1.76 mg/L in the elderly group without gender difference [23]. In a Thai adult population, the hsCRP concentration ranged from 0.2 to 7.9 mg/L without significant sex- nor age-differences [24]. Despite the different numerical range, our hsCRP measurements uniformly demonstrated significant right-skewed distribution, with older and participants with any of the cardiometabolic disorders embodying higher median. hsCRP is known to increase with wide range of both acute and chronic infections, tissue necrosis, neoplasia, insulin resistance, obesity, smoking, and stress [25]. Considering that the prevalence of such proinflammatory conditions are more prevalent with older age, our results align with other population-based studies, where hsCRP was associated with positively graded risk for frailty [26], incident T2DM [2,27], elevated lipid profile [23], and non-alcoholic fatty liver disease [7].
Yet, descriptive assays from population- or clinic-settings of other multiplex assayed inflammatory biomarkers are currently scarce. Several studies have done genetic, experimental and analytic examinations between cytokines/chemokines and diseases with inflammation as a primary basis for pathogenesis. They may provide biological plausibility for some of our current findings.

IL-6

IL-6 is produced in various tissues, including activated leukocytes, adipocytes, and endothelial cells [28]. Our participants of older age or with T2DM, hypertension or dyslipidemia showed higher median IL-6 levels. Similarly, a cross-sectional data from the Newcastle 85+ Study identified positive association between basal IL-6 and frailty risk, enacting as a mediator in inflammatory processes whilst physiological changes accompanied with aging, such as decreased lean body mass, osteopenia, low-grade anemia, decreased serum albumin and cholesterol and increased prevalence of lymphoproliferative disorders [26]. In terms of high IL-6 levels among participants with cardiometabolic disorder, parallel results were also observed in a study, which implicated the pro-coagulant effect of IL-6 from its production in arterial endothelial and smooth muscle cells [29]. Findings from a population-based study also confirmed that elevated levels of IL-6 are associated with increased risk of future MI, reinforcing its role in cytokine-mediated inflammation during the early stages of atherogenesis [3].

IL-1α and IL-β

IL-1 is crucial for host-defense responses to infection and injury by triggering inflammation in a pathway initiated through myeloid differentiation primary response 88 (MYD88) activation and culminating in NF-KB-induced transcription of inflammatory genes [30,31]. Whereas IL-1β is secreted and active only upon cleavage of its precursor by caspase-1 on the inflammasome [11], IL-1α can be found constitutively inside cells under normal homeostasis, and is active in its precursor as well as calpain-processed mature form [32,33]. These different order of IL-1 subtype expression and activation may explain explain the low correlation and varying detection rate despite their similar cellular niche. Likewise, whereas no differences were detected by age group for median IL-1α values, IL-1β, and hsIL-1β levels were higher in the younger group. Such results contrast a previous study that tested age-related augmentation of the systemic and myocardial inflammatory responses, in which the older mice displayed greater myocardial mononuclear cell accumulation and IL-1β production than younger mice due to depressed cardiac function with older age [34]. This result translates to higher vulnerability to endotoxemic cardiac depression with aging [34]. We suspect that such inconsistency may arise from different study population structure and size, as similar studies have also observed varying magnitude of association between IL-6 and age by participant sex, age, and race [34,35]. Yet, our results align in terms of the implication of IL-1 in inflammatory diseases. The genetic assay showed the distribution of oncogenic mutations and single nucleotype polymorphisms (SNPs) in the predicted and experimental structures of protein complexes in the IL-1 pathway, supporting its contributions to cancer development and other inflammatory diseases [36]. Although our study excluded participants with recent history of malignant cancer, the hsIL-1β concentrations remained consistently high among those with metabolic disorders, implying that moderate differences in IL-1 subtypes may be present even in non-critical inflammatory diseases.

TNF-α and TNF-β

TNFs play crucial role in innate and adaptive immunity by inducing proatherogenic changes in lipid metabolism, thereby implicated in lymphoid follicle development, production of proinflammatory cytokines and facilitation of fibroblasts and synviocytes proliferation [37]. Despite TNF-β and TNF-α share many common biological activities, TNF-β is more likely to induce the secretion of IL-6, IL-8, and matrix metallopeptidase 3 (MMP-3) than TNF-α even at low levels, thereby supporting differential productions and detections of TNF subtypes in inflammatory and autoimmune diseases [38,39]. Moreover, TNFs are highly implicated in impaired glucose metabolism and elevated insulin resistance via inhibition of insulin-stimulated tyrosine kinase activity of the insulin receptor by downregulating adiponectin and upregulating leptin production [40,41]. These results align with our current findings, in which the both subtypes of TNF were higher among participants with T2DM.

MCP-1 and MCP-3

MCPs selectively recruit monocytes, neutrophils, and lymphocytes, thereby are found during progression of immune disorders, pulmonary diseases, cancer, and vascular diseases [42,43]. Recent experiments demonstrated a reduction in atherosclerotic lesion formation in MCP-1 deficient mice [44,45]. Conversely, macrophage-specific overexpression of MCP-1 resulted in the acceleration of vascular lesion size and infiltration of macrophages in atherosclerosis-prone mice [45]; such findings are in align with higher MCP-1 levels detected among participants with T2DM, hypertension and dyslipidemia in our study population. Another study indicated high feasibility of MCP-1 as a surrogate measure of biological age, as its circulating levels were higher in frail participants [46], in parallel with higher percentiles embodied by our older participants.
Compared with previous findings, inconsistencies may be explained by the following reasons. First, the upper percentiles were largely derived from small counts of very high values; considering that we did not exclude any upper outliers, those extreme values may distort the true distribution differences between participants with and without metabolic disorder(s). Secondly, the assay methodological discrepancies challenge the direct comparison across different studies. Different kits and software extrapolate antigen concentration with different sample preparation and trade-off of sensitivity, specificity and referent. Lastly, the inflammatory biomarkers are highly dependent on the conditions of extraction. Depending on the study design, number of measurements done at specific time point(s) in disease progression may lead to high heterogeneity. These differences may explain the unexpected negative correlation observed between TNF-β and MCP-1, likely due to varying sample size, number of detectable measurements and its range, compared with previous studies.
In short, every family and subtypes of cytokine/chemokines play differential role in disease-specific inflammatory processes. Their concentrations vary by location and timing of secretion, expression and proliferation, depending on the stages of disease progression, individual- and population heterogeneity and genetic susceptibility. These variations challenge the current multiplex assay to provide a consensus on single reference range for risk prediction and diagnosis.

Strengths and limitations

To the extent of our knowledge, this is the first study to assay and to present a wide array of inflammatory markers in community-dwelling Korean population. Our findings may serve as an addendum in reinforcing biological mechanisms of inflammation and a basis for more comprehensive depiction of inflammatory states for both clinic and research purposes. In addition, despite a considerable proportion of NDs, we were still able to illustrate a wide range of inflammatory states from considerably large-sized population, thereby able to portray realistic distributions found in population settings. Future studies warrant larger population- and clinic-based sample with repeated measurements to obtain more robust reference values for diagnostic purposes.
Several limitations warrant cautious interpretations of our findings. The primary concern involves the study nature and the design of the study. Considering the relatively good health of the participants, a large proportion of NDs were observed. Although some of these were overcome by high sensitivity assay, we were unable to distinguish differences within the ND range. Moreover, due to the cross-sectional nature of the study design, we relied on a single measurement, thereby unable to determine whether the assayed biomarkers reflect acute or persistent inflammatory state. Another limitation arises from the detection methodology used. Although the samples were measured via identical kit and technologies, day-to-day variation in sample collection were inevitable. However, we have minimized potential batch effect by assaying the samples at a single time point. Lastly, our study did not limit eligibility nor collect information regarding history of acute infectious diseases.
In sum, the distributions of inflammatory biomarkers varied by its classes and inclusion/exclusion of NDs. Importantly, the median and the 99th percentile for most biomarkers were different when stratifying participants by sex, age and with and without cardiometabolic disorders. These findings lend weight to provide sex- and age-specific reference values to improve risk predictive and diagnostic performance for inflammatory diseases. A more considered approach would be to tailor the references to individual baseline characteristics and comorbidities. Given that independent clinical feasibility of these biomarkers is not yet warranted, they should yet be referred in conjunction with routine examinations, individual's clinical presentation, and medical history.

ACKNOWLEDGMENTS

None.

NOTES

CONFLICTS OF INTEREST

No potential conflict of interest relevant to this article was reported.

AUTHOR CONTRIBUTIONS

Conception or design: S.M.J.C., H.C.K.

Acquisition, analysis, or interpretation of data: S.M.J.C., H.L., J.S.S., H.C.K.

Drafting the work or revising: S.M.J.C., H.L.

Final approval of the manuscript: S.M.J.C., H.L., J.S.S., H.C.K.

SUPPLEMENTARY MATERIALS

Supplementary materials related to this article can be found online at https://doi.org/10.4093/dmj.2019.0119.

Supplementary Table 1

Distribution of inflammatory biomarkers in male participants (n=1,426)
dmj-2019-0119-s001.pdf

Supplementary Table 2

Distribution of inflammatory biomarkers in female participants (n=2,634)
dmj-2019-0119-s002.pdf

Supplementary Table 3

Distribution of inflammatory biomarkers in participants age below 55 (n=2,097)
dmj-2019-0119-s003.pdf

Supplementary Table 4

Distribution of inflammatory biomarkers in participants age 55 or above (n=1,963)
dmj-2019-0119-s004.pdf

Supplementary Fig. 1

Distribution of tumor necrosis factor-α (TNF-α) in total population (n=1,285).
dmj-2019-0119-s005.pdf

Supplementary Fig. 2

Distribution of tumor necrosis factor-β (TNF-β) including non-detectable (ND) measurements in total population (n=466).
dmj-2019-0119-s006.pdf

Supplementary Fig. 3

Distribution of tumor necrosis factor-β (TNF-β) excluding non-detectable measurements in total population (n=140).
dmj-2019-0119-s007.pdf

Supplementary Fig. 4

Distribution of interleukin-1α (IL-1α) including non-detectable (ND) measurements in total population (n=1,285).
dmj-2019-0119-s008.pdf

Supplementary Fig. 5

Distribution of interleukin-1α (IL-1α) excluding non-detectable measurements in total population (n=349).
dmj-2019-0119-s009.pdf

Supplementary Fig. 6

Distribution of interleukin-1β (IL-1β) including non-detectable (ND) measurements in total population (n=1,285).
dmj-2019-0119-s010.pdf

Supplementary Fig. 7

Distribution of interleukin-1β (IL-1β) excluding non-detectable measurements in total population (n=1,179).
dmj-2019-0119-s011.pdf

Supplementary Fig. 8

Distribution of high sensitivity interleukin-1β (hsIL-1β) in total population (n=819).
dmj-2019-0119-s012.pdf

Supplementary Fig. 9

Distribution of interleukin-6 (IL-6) including non-detectable (ND) measurements in total population (n=1,285).
dmj-2019-0119-s013.pdf

Supplementary Fig. 10

Distribution of interleukin-6 (IL-6) excluding non-detectable measurements in total population (n=1,180).
dmj-2019-0119-s014.pdf

Supplementary Fig. 11

Distribution of high sensitivity interleukin-6 (hsIL-6) in total population (n=819).
dmj-2019-0119-s015.pdf

Supplementary Fig. 12

Distribution of monocyte chemoattractant protein-1 (MCP-1) in total population (n=1,285).
dmj-2019-0119-s016.pdf

Supplementary Fig. 13

Distribution of monocyte chemoattractant protein-3 (MCP-3) including non-detectable (ND) measurements in total population (n=1,285).
dmj-2019-0119-s017.pdf

Supplementary Fig. 14

Distribution of monocyte chemoattractant protein-3 (MCP-3) excluding non-detectable measurements in total population (n=504).
dmj-2019-0119-s018.pdf

Supplementary Fig. 15

Distribution of high sensitivity C-reactive protein (hsCRP) in total population (n=4,058).
dmj-2019-0119-s019.pdf

REFERENCES

1. Popa C, Netea MG, van Riel PL, van der Meer JW, Stalenhoef AF. The role of TNF-alpha in chronic inflammatory conditions, intermediary metabolism, and cardiovascular risk. J Lipid Res 2007;48:751-762.
crossref pmid
2. Pradhan AD, Manson JE, Rifai N, Buring JE, Ridker PM. C-reactive protein, interleukin 6, and risk of developing type 2 diabetes mellitus. JAMA 2001;286:327-334.
crossref pmid
3. Ridker PM, Rifai N, Stampfer MJ, Hennekens CH. Plasma concentration of interleukin-6 and the risk of future myocardial infarction among apparently healthy men. Circulation 2000;101:1767-1772.
crossref pmid
4. Kaptoge S, Seshasai SR, Gao P, Freitag DF, Butterworth AS, Borglykke A, Di Angelantonio E, Gudnason V, Rumley A, Lowe GD, Jørgensen T, Danesh J. Inflammatory cytokines and risk of coronary heart disease: new prospective study and updated meta-analysis. Eur Heart J 2014;35:578-589.
crossref pmid pdf
5. Schöttker B, Herder C, Rothenbacher D, Roden M, Kolb H, Müller H, Brenner H. Proinflammatory cytokines, adiponectin, and increased risk of primary cardiovascular events in diabetic patients with or without renal dysfunction: results from the ESTHER study. Diabetes Care 2013;36:1703-1711.
crossref pmid pmc
6. Koch W, Kastrati A, Bottiger C, Mehilli J, von Beckerath N, Schomig A. Interleukin-10 and tumor necrosis factor gene polymorphisms and risk of coronary artery disease and myocardial infarction. Atherosclerosis 2001;159:137-144.
crossref pmid
7. Lee J, Yoon K, Ryu S, Chang Y, Kim HR. High-normal levels of hs-CRP predict the development of non-alcoholic fatty liver in healthy men. PLoS One 2017;12:e0172666.
crossref pmid pmc
8. Ridker PM. C-reactive protein and the prediction of cardiovascular events among those at intermediate risk: moving an inflammatory hypothesis toward consensus. J Am Coll Cardiol 2007;49:2129-2138.
crossref pmid
9. Parameswaran N, Patial S. Tumor necrosis factor-α signaling in macrophages. Crit Rev Eukaryot Gene Expr 2010;20:87-103.
crossref pmid pmc pdf
10. Tousoulis D, Oikonomou E, Economou EK, Crea F, Kaski JC. Inflammatory cytokines in atherosclerosis: current therapeutic approaches. Eur Heart J 2016;37:1723-1732.
crossref pmid pdf
11. Jia T, Serbina NV, Brandl K, Zhong MX, Leiner IM, Charo IF, Pamer EG. Additive roles for MCP-1 and MCP-3 in CCR2-mediated recruitment of inflammatory monocytes during Listeria monocytogenes infection. J Immunol 2008;180:6846-6853.
crossref pmid pmc
12. Bogavac-Stanojevic N, Jelic-Ivanovic Z, Spasojevic-Kalimanovska V, Spasic S, Kalimanovska-Ostric D. Lipid and inflammatory markers for the prediction of coronary artery disease: a multi-marker approach. Clin Biochem 2007;40:1000-1006.
crossref pmid
13. Fatkhullina AR, Peshkova IO, Koltsova EK. The role of cytokines in the development of atherosclerosis. Biochemistry (Mosc) 2016;81:1358-1370.
crossref pmid pmc pdf
14. Ramji DP, Davies TS. Cytokines in atherosclerosis: key players in all stages of disease and promising therapeutic targets. Cytokine Growth Factor Rev 2015;26:673-685.
crossref pmid pmc
15. Ait-Oufella H, Taleb S, Mallat Z, Tedgui A. Recent advances on the role of cytokines in atherosclerosis. Arterioscler Thromb Vasc Biol 2011;31:969-979.
crossref pmid
16. Shim JS, Song BM, Lee JH, Lee SW, Park JH, Choi DP, Lee MH, Ha KH, Kim DJ, Park S, Lee WW, Youm Y, Shin EC, Kim HC. Cohort profile: the cardiovascular and metabolic diseases etiology research center cohort in Korea. Yonsei Med J 2019;60:804-810.
crossref pmid pmc
17. Ko SH, Kim SR, Kim DJ, Oh SJ, Lee HJ, Shim KH, Woo MH, Kim JY, Kim NH, Kim JT, Kim CH, Kim HJ, Jeong IK, Hong EK, Cho JH, Mok JO, Yoon KH. Committee of Clinical Practice Guidelines. Korean Diabetes Association. 2011 Clinical practice guidelines for type 2 diabetes in Korea. Diabetes Metab J 2011;35:431-436.
crossref pmid pmc
18. Shin J, Park JB, Kim KI, Kim JH, Yang DH, Pyun WB, Kim YG, Kim GH, Chae SC. Guideline Committee of the Korean Society of Hypertension. 2013 Korean Society of Hypertension guidelines for the management of hypertension: part I-epidemiology and diagnosis of hypertension. Clin Hypertens 2015;21:1.
crossref pmid pmc pdf
19. James PA, Oparil S, Carter BL, Cushman WC, Dennison-Himmelfarb C, Handler J, Lackland DT, LeFevre ML, MacKenzie TD, Ogedegbe O, Smith SC Jr, Svetkey LP, Taler SJ, Townsend RR, Wright JT Jr, Narva AS, Ortiz E. 2014 Evidence-based guideline for the management of high blood pressure in adults: report from the panel members appointed to the Eighth Joint National Committee (JNC 8). JAMA 2014;311:507-520.
crossref pmid
20. Committee for the Korean Guidelines for the Management of Dyslipidemia. 2015 Korean guidelines for the management of dyslipidemia: executive summary (English translation). Korean Circ J 2016;46:275-306.
crossref pmid pmc
21. Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults. Executive summary of the third report of the National Cholesterol Education Program (NCEP) expert panel on detection, evaluation, and treatment of high blood cholesterol in adults (Adult Treatment Panel III). JAMA 2001;285:2486-2497.
crossref pmid
22. Zhai Y, Shi XM, Fitzgerald SM, Qian HZ, Kraus VB, Sereny M, Hu P, Yin ZX, Xu JW, Zeng Y. High sensitivity C-reactive protein associated with different health predictors in middle-aged and oldest old Chinese. Biomed Environ Sci 2012;25:257-266.
pmid pmc
23. Charuruks N, Laohajinda B, Rujiwanitgun S, Chaiworaporn M. Reference value for C-reactive protein and its distribution pattern in Thai adults. Circ J 2005;69:339-344.
crossref pmid
24. Hirschfield GM, Pepys MB. C-reactive protein and cardiovascular disease: new insights from an old molecule. QJM 2003;96:793-807.
crossref pmid pdf
25. Collerton J, Martin-Ruiz C, Davies K, Hilkens CM, Isaacs J, Kolenda C, Parker C, Dunn M, Catt M, Jagger C, von Zglinicki T, Kirkwood TB. Frailty and the role of inflammation, immunosenescence and cellular ageing in the very old: cross-sectional findings from the Newcastle 85+ Study. Mech Ageing Dev 2012;133:456-466.
crossref pmid
26. Effoe VS, Correa A, Chen H, Lacy ME, Bertoni AG. High-sensitivity C-reactive protein is associated with incident type 2 diabetes among African Americans: the Jackson Heart Study. Diabetes Care 2015;38:1694-1700.
crossref pmid pmc
27. Aniszewska A, Chlodzinska N, Bartkowska K, Winnicka MM, Turlejski K, Djavadian RL. The expression of interleukin-6 and its receptor in various brain regions and their roles in exploratory behavior and stress responses. J Neuroimmunol 2015;284:1-9.
crossref pmid
28. Qi L, Rifai N, Hu FB. Interleukin-6 receptor gene variations, plasma interleukin-6 levels, and type 2 diabetes in U.S. Women. Diabetes 2007;56:3075-3081.
crossref pmid
29. Lopez-Castejon G, Brough D. Understanding the mechanism of IL-1β secretion. Cytokine Growth Factor Rev 2011;22:189-195.
crossref pmid pmc
30. Liu T, Zhang L, Joo D, Sun SC. NF-κB signaling in inflammation. Signal Transduct Target Ther 2017;2:17023.
crossref pmid pmc
31. Rider P, Carmi Y, Guttman O, Braiman A, Cohen I, Voronov E, White MR, Dinarello CA, Apte RN. IL-1α and IL-1β recruit different myeloid cells and promote different stages of sterile inflammation. J Immunol 2011;187:4835-4843.
crossref pmid
32. Di Paolo NC, Shayakhmetov DM. Interleukin 1α and the inflammatory process. Nat Immunol 2016;17:906-913.
crossref pmid pmc pdf
33. Slimani H, Zhai Y, Yousif NG, Ao L, Zeng Q, Fullerton DA, Meng X. Enhanced monocyte chemoattractant protein-1 production in aging mice exaggerates cardiac depression during endotoxemia. Crit Care 2014;18:527.
crossref pmid pmc pdf
34. Hubbard RE, O'Mahony MS, Savva GM, Calver BL, Woodhouse KW. Inflammation and frailty measures in older people. J Cell Mol Med 2009;13(9B):3103-3109.
crossref pmid pmc
35. Acuner Ozbabacan SE, Gursoy A, Nussinov R, Keskin O. The structural pathway of interleukin 1 (IL-1) initiated signaling reveals mechanisms of oncogenic mutations and SNPs in inflammation and cancer. PLoS Comput Biol 2014;10:e1003470.
crossref pmid pmc
36. Han L, Song JH, Yoon JH, Park YG, Lee SW, Choi YJ, Nam SW, Lee JY, Park WS. TNF-α and TNF-β polymorphisms are associated with susceptibility to osteoarthritis in a Korean population. Korean J Pathol 2012;46:30-37.
crossref pmid pmc pdf
37. Svenungsson E, Gunnarsson I, Fei GZ, Lundberg IE, Klareskog L, Frostegard J. Elevated triglycerides and low levels of high-density lipoprotein as markers of disease activity in association with up-regulation of the tumor necrosis factor alpha/tumor necrosis factor receptor system in systemic lupus erythematosus. Arthritis Rheum 2003;48:2533-2540.
crossref pmid
38. Aggarwal BB. Signalling pathways of the TNF superfamily: a double-edged sword. Nat Rev Immunol 2003;3:745-756.
crossref pmid pdf
39. Stephens JM, Pekala PH. Transcriptional repression of the GLUT4 and C/EBP genes in 3T3-L1 adipocytes by tumor necrosis factor-alpha. J Biol Chem 1991;266:21839-21845.
pmid
40. Bruun JM, Lihn AS, Verdich C, Pedersen SB, Toubro S, Astrup A, Richelsen B. Regulation of adiponectin by adipose tissue-derived cytokines: in vivo and in vitro investigations in humans. Am J Physiol Endocrinol Metab 2003;285:E527-E533.
crossref pmid
41. Deshmane SL, Kremlev S, Amini S, Sawaya BE. Monocyte chemoattractant protein-1 (MCP-1): an overview. J Interferon Cytokine Res 2009;29:313-326.
crossref pmid pmc
42. Tsou CL, Peters W, Si Y, Slaymaker S, Aslanian AM, Weisberg SP, Mack M, Charo IF. Critical roles for CCR2 and MCP-3 in monocyte mobilization from bone marrow and recruitment to inflammatory sites. J Clin Invest 2007;117:902-909.
crossref pmid pmc
43. Gu L, Okada Y, Clinton SK, Gerard C, Sukhova GK, Libby P, Rollins BJ. Absence of monocyte chemoattractant protein-1 reduces atherosclerosis in low density lipoprotein receptor-deficient mice. Mol Cell 1998;2:275-281.
crossref pmid
44. Boring L, Gosling J, Cleary M, Charo IF. Decreased lesion formation in CCR2-/- mice reveals a role for chemokines in the initiation of atherosclerosis. Nature 1998;394:894-897.
crossref pmid pdf
45. Aiello RJ, Bourassa PA, Lindsey S, Weng W, Natoli E, Rollins BJ, Milos PM. Monocyte chemoattractant protein-1 accelerates atherosclerosis in apolipoprotein E-deficient mice. Arterioscler Thromb Vasc Biol 1999;19:1518-1525.
crossref pmid
46. Fried LP, Tangen CM, Walston J, Newman AB, Hirsch C, Gottdiener J, Seeman T, Tracy R, Kop WJ, Burke G, McBurnie MA. Cardiovascular Health Study Collaborative Research Group. Frailty in older adults: evidence for a phenotype. J Gerontol A Biol Sci Med Sci 2001;56:M146-M156.
crossref pmid pdf
Fig. 1

Flow diagram of the study inclusion criteria. CMERC, Cardiovascular and Metabolic Disease Etiology Research Center; TNF, tumor necrosis factor; IL, interleukin; hsIL, high sensitivity interleukin; MCP, monocyte chemoattractant protein; hsCRP, high sensitivity C-reactive protein.

dmj-2019-0119f1.jpg
Table 1.
General characteristics of the study participants by sex (n=4,058)
Variable Total (n=4,058) Sex
P value
Male (n=1,426) Female (n=2,632)
Age, yr 51.3±9.4 50.2±10.2 51.9±8.8 <0.001
BMI, kg/m2 23.9±3.1 24.9±2.9 23.4±3.0 <0.001
Alcohol intake <0.001
 Non-drinker 905 (22.3) 118 (8.3) 787 (29.9)
 Previous drinker 184 (4.5) 86 (6.0) 98 (3.7)
 Current drinker 2,969 (73.2) 1,222 (85.7) 1,747 (66.4)
Smoking status <0.001
 Non-smoker 2,778 (68.5) 322 (22.6) 2,456 (93.3)
 Previous smoker 724 (17.8) 626 (43.9) 98 (3.7)
 Current smoker 556 (13.7) 478 (33.5) 78 (3.0) 0.045
History of CVD
 Yes 26 (0.6) 14 (1.0) 12 (0.5)
 No 4,032 (99.4) 1,412 (99.0) 2,620 (99.5)
Fasting glucose, mg/dL 91.8±19.3 96.1±23.9 89.5±15.8 <0.001
HbA1c, % 5.7±0.7 5.7±0.8 5.6±0.6 <0.001
Diabetes mellitus <0.001
 Yes 372 (9.2) 178 (12.5) 194 (7.4)
 No 3,686 (90.8) 1,248 (87.5) 2,438 (92.6)
SBP, mm Hg 118.6±14.9 125.0±13.6 115.2±14.4 <0.001
DBP, mm Hg 76.2±9.9 80.6±9.9 73.8±9.1 <0.001
Hypertension <0.001
Yes 1,070 (26.4) 509 (35.7) 561 (21.3)
No 2,988 (73.6) 917 (64.3) 2,071 (78.7)
Total cholesterol, mg/dL 198.4±35.4 195.3±35.3 200.1±35.3 <0.001
HDL-C, mg/dL 57.6±14.7 51.3±12.8 61.0±14.5 <0.001
LDL-C, mg/dL 115.3±32.1 112.8±33.3 116.6±31.4 <0.001
Triglyceride, mg/dL 127.9±87.8 155.9±115.9 112.7±62.8 <0.001
Dyslipidemia <0.001
 Yes 1,882 (46.4) 793 (55.6) 1,089 (41.4)
 No 2,176 (53.6) 633 (44.4) 1,543 (58.6)

Values are presented as mean±standard deviation or number (%). P value was derived from the independent t-test, the Wilcoxon rank-sum test, or chi-square test.

BMI, body mass index; CVD, cardiovascular disease; HbA1c, glycosylated hemoglobin; SBP, systolic blood pressure; DBP, diastolic blood pressure; HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol.

Table 2.
Distribution of inflammatory biomarkers in total participants (n=4,058)
Biomarker No. of measurement No. of ND measurement Mean SD Minimum 25th percentile Median 75th percentile 90th percentile 95th percentile 99th percentile Maximum
Including ND measurements
TNF-α, pg/mL 1,285 0 19.59 28.62 0.47 8.11 12.26 19.49 33.23 58.52 150.29 301.65
TNF-β, pg/mL 466 326 7.98 88.97 0.00 0.00 0.00 0.07 1.15 9.48 168.50 1,840.00
IL-1α, pg/mL 1,285 936 79.19 281.69 0.00 0.00 0.00 2.28 193.19 534.02 1,296.00 4,444.00
IL-1β, pg/mL 1,285 106 5.10 18.35 0.00 0.13 0.52 2.01 9.54 24.37 95.95 276.47
hsIL-1β, pg/mL 819 0 0.90 1.00 0.10 0.40 0.70 1.00 1.80 2.57 4.90 15.20
IL-6, pg/mL 1,285 105 23.50 59.64 0.00 0.09 0.61 11.46 79.77 145.27 300.15 505.69
hsIL-6, pg/mL 819 0 1.31 3.32 0.22 0.60 0.76 1.13 1.97 3.03 10.25 69.60
MCP-1, pg/mL 1,285 0 709.40 355.67 80.75 511.57 667.21 843.22 1,056 1,192 1,698 8,129.00
MCP-3, pg/mL 1,285 781 20.50 48.36 0.00 0.00 0.00 17.76 69.45 104.44 243.70 525.80
hsCRP, mg/L 4,058 0 1.44 3.87 0.01 0.33 0.58 1.19 2.66 4.89 16.29 88.49
Excluding ND measurements
TNF-β, pg/mL 140 - 26.55 161.20 0.01 0.13 0.33 3.16 26.71 106.57 381.59 1,840.00
IL-1α, pg/mL 349 - 291.57 480.27 0.02 15.39 91.43 375.53 853.18 1,151 2,233 4,444.00
IL-1β, pg/mL 1,179 - 5.55 19.09 0.01 0.22 0.62 2.25 10.52 28.00 97.68 276.47
IL-6, pg/L 1,180 - 25.59 61.81 0.01 0.14 0.87 15.11 86.56 150.38 301.61 505.69
MCP-3, pg/mL 504 - 52.26 65.62 0.01 10.19 31.16 70.68 123.65 193.87 273.65 525.80

ND, non-detectable; SD, standard deviation; TNF, tumor necrosis factor; IL, interleukin; hsIL, high sensitivity interleukin; MCP, monocyte chemoattractant protein; hsCRP, high sensitivity C-reactive protein.

Table 3.
Spearman’s correlation among inflammatory biomarkers (n=4,058)
Biomarker TNF-α TNF-β IL-1α IL-1β hsIL-1β IL-6 hsIL-6 MCP-1 MCP-3 hsCRP
TNF-α 1.00 0.03 0.30c 0.26c 0.21c 0.46c 0.14c 0.32c 0.28c 0.19c
TNF-β 0.43c 1.00 0.25c 0.26c NA 0.27c NA -0.13b 0.35c 0.03
IL-1α 0.27c 0.11 1.00 0.44c 0.01 0.48c -0.04 0.03 0.29c 0.01
IL-1β 0.30c 0.42c 0.31c 1.00 0.23c 0.42c 0.10b 0.12c 0.43c 0.04
hsIL-1β 0.21c NA -0.06 0.23c 1.00 0.10b 0.45c 0.17c 0.06 0.11b
IL-6 0.47c 0.39c 0.38c 0.42c 0.04 1.00 0.15c 0.12c 0.39c 0.15c
hsIL-6 0.14c NA -0.01 0.10b 0.45c 0.10b 1.00 0.21c 0.01 0.41c
MCP-1 0.32c 0.16 0.12a 0.15c 0.17c 0.11b 0.21c 1.00 -0.01 0.13c
MCP-3 0.35c 0.48c 0.30c 0.59c 0.04 0.33c -0.01 0.10a 1.00 0.05
hsCRP 0.19c 0.16 -0.01 0.03 0.11b 0.14c 0.41c 0.13c 0.06 1.00

The correlation coefficient was obtained from Spearman's rank correlation. The upper right coefficients are obtained from biomarkers including non-detectable (ND) measurements. The lower left coefficients are obtained from biomarkers excluding ND measurements.

TNF, tumor necrosis factor; IL, interleukin; hsIL, high sensitivity interleukin; MCP, monocyte chemoattractant protein; hsCRP, high sensitivity C-reactive protein; NA, not available.

a P<0.05,

b P<0.01,

c P<0.001.

Table 4.
Distribution of inflammatory biomarkers by sex, age, and prevalence of type 2 diabetes mellitus, hypertension and dyslipidemia, including NDs (n=4,058)
Biomarker Sex
Age, yr
T2DM
Hypertension
Dyslipidemia
Male Female P value 55> 55≤ P value No Yes P value No Yes P value No Yes P value
TNF-α, pg/mL
Number 458 827 - 797 488 - 1,180 105 - 996 289 - 729 556 -
Median 14.51 11.17 <0.001 12.24 12.32 0.889 12.15 14.04 0.027 12.23 12.52 0.428 12.12 12.57 0.101
75th percentile 23.42 17.03 <0.001 20.61 18.67 <0.001 19.36 20.59 0.766 19.66 19.15 0.050 18.72 20.31 0.969
90th percentile 39.99 28.85 0.011 35.84 29.44 <0.001 33.15 33.51 0.798 33.82 31.50 0.042 33.56 32.83 0.387
95th percentile 76.12 54.16 0.107 67.35 48.44 0.002 58.49 64.58 0.989 59.36 53.92 0.240 66.46 50.83 0.041
99th percentile 157.74 139.54 0.405 158.96 135.10 0.205 157.14 101.24 0.252 157.74 150.29 0.788 209.14 112.73 0.025
TNF-β, pg/mL
Number 127 339 - 246 220 - 430 36 - 353 113 - 274 192 -
Median 0.00 0.00 0.352 0.00 0.00 <0.001 0.00 0.00 0.149 0.00 0.00 0.566 0.00 0.00 0.115
75th percentile 0.00 0.07 0.015 0.20 0.00 <0.001 0.07 0.00 0.119 0.07 0.01 0.086 0.13 0.03 0.002
90th percentile 1.31 1.09 0.440 1.31 0.50 0.276 1.12 4.40 0.876 0.98 3.90 0.386 2.03 0.58 0.088
95th percentile 9.48 10.19 0.539 11.33 7.49 0.513 9.18 10.19 0.571 9.18 11.33 0.755 14.44 3.90 0.035
99th percentile 178.94 92.26 0.244 125.54 168.50 0.601 125.54 1,840.00 0.401 92.26 381.59 0.088 178.94 92.26 0.237
IL-1α, pg/mL
Number 458 827 - 797 488 - 1,180 105 - 996 289 - 729 556 -
Median 0.00 0.00 0.048 0.00 0.00 0.117 0.00 0.00 0.130 0.00 0.00 0.436 0.00 0.00 0.021
75th percentile 8.33 0.04 0.030 3.77 0.00 <0.001 2.67 0.00 0.130 2.75 0.03 0.013 6.58 0.00 0.003
90th percentile 249.77 186.25 0.492 196.83 159.73 0.003 206.70 46.69 0.015 197.47 123.82 0.155 280.33 82.64 <0.001
95th percentile 647.87 496.05 0.107 544.72 470.97 0.009 555.84 117.21 0.032 543.18 530.10 0.373 644.42 371.13 0.011
99th percentile 1,380.00 1,191.00 0.157 1,270.00 1.380.00 0.876 1,296.00 458.64 0.854 1,311.00 1,135.00 0.368 1,269.00 1,380.00 0.589
IL-1β, pg/mL
Number 458 827 - 797 488 - 1,180 105 - 996 289 - 729 556 -
Median 0.69 0.45 <0.001 0.66 0.40 <0.001 0.52 0.68 0.478 0.55 0.48 0.133 0.52 0.53 0.814
75th percentile 2.63 1.67 0.001 2.47 1.26 <0.001 2.03 1.57 0.122 2.03 1.83 0.016 2.12 1.85 0.034
90th percentile 10.96 9.09 0.382 10.64 6.42 <0.001 10.00 5.31 0.034 9.38 11.97 0.683 9.09 10.34 0.975
95th percentile 32.99 24.13 0.572 32.99 15.86 <0.001 24.43 9.27 0.678 24.49 24.17 0.373 24.49 24.37 0.591
99th percentile 143.32 81.31 0.405 101.67 71.66 0.068 86.97 125.57 0.437 97.68 71.66 0.368 97.68 81.03 0.567
hsIL-1β, pg/mL
Number 331 488 - 551 268 - 750 69 - 643 176 - 455 364 -
Median 0.70 0.60 <0.001 0.70 0.60 0.310 0.64 0.72 0.153 0.66 0.64 0.696 0.62 0.67 0.221
75th percentile 1.20 0.90 <0.001 1.00 1.00 <0.001 1.03 1.08 0.764 1.03 1.05 0.190 0.98 1.07 0.573
90th percentile 2.00 1.50 <0.001 1.80 1.80 0.003 1.72 2.49 0.208 1.74 1.90 0.591 1.60 1.94 0.265
95th percentile 3.30 1.98 0.002 2.24 2.61 0.322 2.32 2.72 0.292 2.33 2.56 0.636 2.10 2.56 0.547
99th percentile 6.50 3.90 0.007 4.70 5.32 0.816 4.89 6.47 0.840 5.32 3.77 0.072 4.89 6.28 0.908
IL-6, pg/mL
Number 458 827 - 797 488 - 1,180 105 - 996 289 - 729 556 -
Median 1.38 0.37 <0.001 0.72 0.49 0.051 0.61 0.62 0.913 0.61 0.61 0.561 0.74 0.52 0.233
75th percentile 14.37 10.04 0.187 15.79 7.78 <0.001 12.29 7.26 0.056 12.99 8.77 0.004 16.23 8.44 0.002
90th percentile 89.78 69.66 0.070 86.17 59.02 <0.001 83.60 57.19 0.236 80.44 79.77 0.309 87.49 66.73 0.052
95th percentile 168.45 137.77 0.276 149.24 137.79 0.035 147.98 92.42 0.396 147.28 137.34 0.373 148.87 137.34 0.197
99th percentile 345.97 286.06 0.157 336.88 244.24 0.017 300.15 226.26 0.854 329.69 298.67 0.368 300.15 292.51 0.567
hsIL-6, pg/mL
Number 331 488 - 551 268 - 750 69 - 643 176 - 455 364 -
Median 0.83 0.75 <0.001 0.75 0.81 <0.001 0.75 0.98 <0.001 0.75 0.94 <0.001 0.69 0.85 <0.001
75th percentile 1.30 1.01 <0.001 1.05 1.31 0.024 1.11 1.43 0.103 1.04 1.45 0.010 0.98 1.30 0.003
90th percentile 2.11 1.77 <0.001 1.82 2.11 0.418 1.94 4.10 0.567 1.88 2.24 0.727 1.85 2.13 0.266
95th percentile 3.42 2.34 0.005 3.03 2.91 0.032 2.82 8.32 0.078 2.82 3.70 0.672 2.54 3.42 0.210
99th percentile 14.06 7.08 0.007 14.06 8.13 0.004 8.13 69.60 0.012 13.62 10.25 0.778 7.08 14.06 0.059
MCP-1, pg/mL
Number 458 827 - 797 488 - 1,180 105 - 996 289 - 729 556 -
Median 753.01 626.03 <0.001 649.71 704.10 0.002 664.72 707.78 0.061 657.93 706.85 0.007 644.85 704.38 <0.001
75th percentile 927.38 801.87 <0.001 836.75 856.04 0.003 836.88 911.94 0.088 834.37 866.72 0.802 815.68 900.94 0.088
90th percentile 1,132.00 985.86 <0.001 1,053.00 1,059.00 0.027 1,045.00 1,132.00 0.500 1,040.00 1,067.00 0.797 1,008.00 1,083.00 0.143
95th percentile 1,256.00 1,132.00 0.004 1,200.00 1,176.00 0.035 1,182.00 1,313.00 0.188 1,182.00 1,244.00 0.598 1,162.00 1,221.00 0.334
99th percentile 1,843.00 1,530.00 0.157 1,735.00 1,642.00 0.205 1,727.00 1,548.00 0.252 1,735.00 1,679.00 0.368 1,679.00 1,727.00 0.987
MCP-3, pg/mL
Number 458 827 - 797 488 - 1,180 105 - 996 289 - 729 556 -
Median 0.00 0.00 0.089 0.00 0.00 0.050 0.00 0.00 0.135 0.00 0.00 0.544 0.00 0.00 0.036
75th percentile 16.02 19.33 0.671 19.60 14.93 <0.001 18.12 14.45 0.354 18.18 17.04 0.083 20.66 14.54 0.029
90th percentile 73.58 69.26 0.617 70.75 67.06 0.010 70.68 42.37 0.381 66.60 79.65 0.841 70.60 66.87 0.387
95th percentile 115.10 101.10 0.408 116.92 98.13 0.009 104.37 113.04 0.986 101.99 116.92 0.807 118.22 100.31 0.197
99th percentile 256.00 224.82 0.405 245.16 212.02 0.205 243.25 243.70 0.437 228.80 249.45 0.322 252.64 173.63 0.092
hsCRP, mg/L
Number 1,426 2,632 - 2,097 1,961 - 3,686 372 - 2,988 1,070 - 2,176 1,882 -
Median 0.65 0.54 <0.001 0.55 0.61 <0.001 0.56 0.87 <0.001 0.53 0.75 <0.001 0.50 0.69 <0.001
75th percentile 1.39 1.10 <0.001 1.16 1.23 0.241 1.13 2.03 <0.001 1.07 1.57 <0.001 1.02 1.41 <0.001
90th percentile 3.26 2.45 <0.001 2.58 2.68 0.154 2.51 3.96 <0.001 2.40 3.47 0.007 2.41 2.93 0.022
95th percentile 6.28 4.01 <0.001 4.68 5.35 0.154 4.56 7.75 <0.001 4.21 6.25 0.007 4.23 5.43 0.022
99th percentile 21.53 11.41 0.01 15.62 17.37 0.709 14.30 22.42 <0.001 16.29 16.34 0.946 16.57 16.29 0.996

The P value was derived from the two-sided Wilcoxon rank-sum test.

ND, non-detectable; T2DM, type 2 diabetes mellitus; TNF, tumor necrosis factor; IL, interleukin; hsIL, high sensitivity interleukin; MCP, monocyte chemoattractant protein; hsCRP, high sensitivity C-reactive protein.

Table 5.
Distribution of inflammatory biomarkers by sex, age, and prevalence of type 2 diabetes mellitus, hypertension and dyslipidemia, excluding NDs (n=1,180)
Biomarker Sex
Age, yr
T2DM
Hypertension
Dyslipidemia
Male Female P value 55> 55≤ P value No Yes P value No Yes P value No Yes P value
TNF-β, pg/mL
Number 33 107 - 95 45 - 134 6 - 110 30 - 89 51 -
Median 0.56 0.30 0.279 0.31 0.42 0.445 0.30 6.95 0.032 0.30 1.12 0.088 0.42 0.30 0.233
75th percentile 6.38 2.31 0.644 1.31 6.38 0.977 2.31 10.19 0.642 1.52 8.59 0.146 4.94 1.31 0.075
90th percentile 120.88 21.03 0.964 14.44 92.26 0.682 22.26 1,840.00 0.793 19.33 182.73 0.427 49.03 9.49 0.181
95th percentile 178.94 62.92 0.223 62.92 168.50 0.641 92.26 1,840.00 0.639 52.78 381.59 0.065 125.54 68.91 0.089
99th percentile 1,840 244.58 0.660 1,840 381.59 0.963 244.58 1,840.00 0.046 168.50 1,840.00 0.018 381.59 1,840 0.919
IL-1α, pg/mL
Number 140 209 - 229 120 - 326 23 - 275 74 - 213 136 -
Median 74.54 106.12 0.887 105.07 81.66 0.888 103.37 35.24 0.094 101.66 61.48 0.515 134.07 53.98 0.013
75th percentile 413.38 371.13 0.110 373.14 383.13 0.013 410.01 117.21 0.024 375.53 389.99 0.304 442.14 271.13 0.003
90th percentile 940.64 852.97 0.337 902.85 843.77 0.045 902.85 242.31 0.194 902.85 804.14 0.214 967.91 760.22 0.034
95th percentile 1,282.50 1,116 0.408 1,135.00 1,281.50 0.422 1,151.00 458.64 0.595 1,239.00 1,053.00 0.141 1,151 1,311 0.523
99th percentile 2,233 2,054 0.534 2,233 1,769 0.351 2,054 2,233 0.273 2,509 1,552 0.231 1,769 2,233 0.885
IL-1β, pg/mL
Number 433 746 - 797 488 - 1,082 97 - 918 261 - 671 508 -
Median 0.77 0.53 <0.001 0.75 0.48 <0.001 0.60 0.70 0.501 0.67 0.55 0.242 0.61 0.63 0.975
75th percentile 2.82 1.91 0.001 2.80 1.54 <0.001 2.28 1.67 0.123 2.28 2.09 0.022 2.41 2.04 0.077
90th percentile 12.88 10.00 0.201 12.27 8.72 <0.001 10.96 6.72 0.060 10.00 13.43 0.981 9.56 10.96 0.892
95th percentile 33.86 25.88 0.370 34.35 18.40 <0.001 27.54 48.41 0.853 27.54 28.00 0.643 27.54 28.00 0.720
99th percentile 143.32 86.97 0.280 101.67 71.66 0.106 95.95 156.03 0.368 97.68 71.66 0.445 101.67 81.03 0.364
IL-6, pg/mL
Number 435 745 - 739 441 - 1,084 96 - 922 258 - 674 506 -
Median 1.90 0.62 <0.001 1.01 0.77 0.157 0.87 0.90 0.835 0.86 1.03 0.754 1.16 0.75 0.395
75th percentile 19.46 13.17 0.291 20.41 10.08 <0.001 17.48 8.77 0.035 17.78 10.89 0.003 19.46 10.22 0.002
90th percentile 94.72 79.71 0.201 94.72 76.49 0.001 87.49 59.03 0.119 87.16 85.52 0.132 94.04 79.46 0.045
95th percentile 170.20 145.27 0.241 154.29 148.68 0.026 151.04 118.07 0.522 149.71 152.64 0.447 168.19 137.79 0.377
99th percentile 345.97 296.78 0.092 336.88 244.24 0.028 301.61 326.23 0.920 329.69 298.67 0.445 329.69 292.51 0.743
MCP-3, pg/mL
Number 160 344 - 332 172 - 471 33 - 399 105 - 305 199 -
Median 39.31 26.70 0.045 30.47 32.61 0.400 31.34 29.85 0.577 29.00 37.61 0.072 31.34 30.66 0.957
75th percentile 80.09 67.04 0.606 68.92 73.71 0.007 70.75 69.45 0.266 66.60 81.40 0.715 68.58 73.26 0.421
90th percentile 127.49 113.04 0.982 127.57 110.97 0.007 118.22 131.70 0.742 117.95 131.70 0.860 139.47 112.60 0.015
95th percentile 194.87 183.39 0.955 198.58 173.63 0.029 193.87 243.70 0.794 193.87 203.51 0.703 203.51 138.24 0.005
99th percentile 480.20 252.64 0.106 257.04 432.06 0.935 301.97 256 0.436 301.97 256 0.590 346 245.16 0.023

The P value was derived from the two-sided Wilcoxon rank-sum test.

ND, non-detectable; T2DM, type 2 diabetes mellitus; TNF, tumor necrosis factor; IL, interleukin; MCP, monocyte chemoattractant protein.



ABOUT
BROWSE ARTICLES
FOR CONTRIBUTORS
Editorial Office
101-2104, Lotte Castle President, 109 Mapo-daero, Mapo-gu, Seoul 04146, Korea​
Tel: +82-2-714-9064    Fax: +82-2-714-9084    E-mail: diabetes@kams.or.kr                

Copyright © 2020 by Korean Diabetes Association. All rights reserved.

Developed in M2community

Close layer