Red cell distribution width (RDW) indicates the size variability of circulating erythrocytes and often reported as a part of the complete blood count for the differential diagnosis of anemia [
Ultrasonographic measurement of carotid intima-media thickness (C-IMT) is a relatively simple, noninvasive way to assess subclinical atherosclerosis in high-risk patients. CVD is the most common cause of death in people with type 2 diabetes, and C-IMT has been widely used to predict CVD risk and related outcomes in these people [
In this study, we analyzed the relationship between RDW and subclinical atherosclerosis measured by C-IMT and examined its potential role as a marker carotid atherosclerosis in Koreans with type 2 diabetes without CVD.
Four hundred sixty-nine people with type 2 diabetes at the Diabetes Center of Gangnam Severance Hospital, Korea, were enrolled in this cross-sectional study. The subjects were retrospectively recruited from Cohort Study for Clinical Research in Gangnam Severance Hospital. This study is an observational study designed to systemically collect clinical and biochemical information of people with impaired glucose metabolism in the Gangnam area in Seoul, Korea, and to establish a cohort to be followed for the incidence of diabetes among those at prediabetic phase and also diabetic complications. Previously diagnosed diabetes patients based on self-reported responses and newly diagnosed diabetes patients according to the American Diabetes Association criteria were all included. People with concurrent acute illnesses including clinically significant infectious diseases, chronic kidney or hepatic diseases, malignancy, and any systemic hematologic disorders that could affect red blood cells were excluded. Those with prior cardiovascular or cerebrovascular diseases were also excluded. Among the 577 type 2 diabetes patients enrolled in Cohort Study for Clinical Research in Gangnam Severance Hospital between 2013 and 2014, 61 subjects with a history of coronary artery disease or cerebrovascular accident, 22 subjects with chronic kidney disease or chronic hepatitis disease, 4 subjects with cancer, and 2 subjects with acute infection were excluded, and 469 subjects were analyzed. The institutional review board of Yonsei University College of Medicine approved this study protocol, and written informed consent was obtained from all subjects.
Body weight and height were measured in the morning, without clothing and shoes, and body mass index (BMI) was calculated by dividing the weight (kg) by the square of the height (m2). Systolic and diastolic blood pressures were measured by an experienced technician by placing the left arm at heart level after a five-minute rest using EASY X 800 (Jawon Medical Co. Ltd, Seoul, Korea). Current smoking was defined as having smoked cigarettes regularly over the previous 6 months.
Blood samples were taken from all subjects after an overnight fast. Standard methods were used for complete blood count and biochemistry. Fasting plasma glucose (FPG), total cholesterol (TC), high-density lipoprotein cholesterol (HDL-C), and triglycerides (TG) levels were determined using enzymatic methods with a Hitachi 7600-120 automated chemistry analyzer (Hitachi, Tokyo, Japan). Low-density lipoprotein cholesterol (LDL-C) was calculated according to the Friedewald formula. Hemoglobin A1c (HbA1c) was determined by high-performance liquid chromatography (Variant II, Bio-Rad, Hercules, CA, USA). RDW, hemoglobin, and white blood cell (WBC) count were measured as part of the automated complete blood count using an ADVIA 2120 (Siemens, Erlangen, Germany). Fasting serum insulin was determined by chemiluminescence (RIA kit, Daiichi, Japan), and insulin resistance was calculated using the homeostasis model assessment of insulin resistance (HOMA-IR) index, using following formula: HOMA-IR = fasting insulin (
C-IMT was measured by high-resolution B-mode ultrasonography on a single machine (Aloka, Tokyo, Japan) with a 7.5 MHz linear array transducer by the same investigator throughout the study. The probe scanned the far wall of the middle and distal common carotid artery by a lateral longitudinal projection. IMT was defined as the distance between the lumen-intima interface and media-adventitia interface. The measurement was made at 20 mm proximal to the origin of the right and left carotid bulb using the computer-assisted analyzing system (M’ATH, METRIS Co., Argenteuil, France) in conjunction with the ultrasound exam. The mean value of 99 computer-based points in this region was calculated to be the C-IMT [
Data were expressed as the mean ± S.D. One way analysis of variance (ANOVA) was used to compare various continuous variables among the groups. Correlation coefficients between C-IMT and various clinical factors were calculated with Pearson’s test. Triglycerides, insulin, and HOMA-IR were log-transformed for analysis since they showed skewed distributions. Enter method multiple linear regression analysis was used to assess whether the factors shown to be significantly associated with C-IMT remained significant after adjustment. The association of the RDW with carotid atherosclerosis was further explored by categorizing the RDW into tertiles and using the first tertile as the reference. Adjusted odds ratios (ORs) and corresponding 95% confidence intervals were estimated with the use of multivariate logistic regression analysis models. Statistical analyses were carried out using SPSS for Windows 20.0 (SPSS Inc., Chicago, IL, USA).
Clinical and biochemical characteristics of the study participants are summarized in Table
Clinical characteristics of subjects according to RDW tertiles.
I (10.9–12.1%) | II (12.2–12.6%) | III (12.7–16.0%) | ||
---|---|---|---|---|
158 | 157 | 154 | ||
Male (%) | 105 (66.5) | 88 (56.1) | 100 (64.9) | 0.22 |
Age (years) | 54.29 ± 10.57 | 56.88 ± 11.96 | 59.92 ± 12.44 | <0.01 |
Diabetes duration (years) | 6.78 ± 6.81 | 7.25 ± 7.57 | 7.61 ± 7.43 | <0.01 |
BMI (kg/m2) | 24.15 ± 2.77 | 24.58 ± 3.41 | 25.28 ± 3.57 | 0.01 |
SBP (mmHg) | 124.18 ± 13.31 | 127.03 ± 14.19 | 128.07 ± 14.54 | 0.04 |
DBP (mmHg) | 77.08 ± 9.06 | 77.66 ± 8.27 | 77.48 ± 8.85 | 0.37 |
FPG (mmol/L) | 9.09 ± 3.94 | 9.16 ± 3.73 | 9.57 ± 4.09 | 0.16 |
HbA1c (%) | 8.62 ± 2.24 | 8.63 ± 2.43 | 8.69 ± 2.23 | 0.39 |
HbA1c (mmol/mol) | 70.7 ± 1.0 | 70.8 ± 3.1 | 71.5 ± 0.9 | 0.41 |
TC (mmol/L) | 4.46 ± 1.06 | 4.37 ± 1.09 | 4.55 ± 1.11 | 0.33 |
TG (mmol/L) | 1.55 (1.15–2.19) | 1.36 (1.06–1.99) | 1.32 (1.01–1.72) | 0.07 |
LDL-C (mmol/L) | 2.55 ± 1.03 | 2.56 ± 0.93 | 2.57 ± 0.98 | 0.76 |
HDL-C (mmol/L) | 1.13 ± 0.33 | 1.14 ± 0.33 | 1.15 ± 0.31 | 0.85 |
Insulin (mIU/dL) | 6.0 (3.0–9.2) | 6.75 (2.5–9.8) | 7.6 (2.8–11.4) | 0.31 |
HOMA-IR | 2.50 (1.07–3.87) | 2.31 (1.03-4.25) | 2.32 (1.04–4.56) | 0.23 |
Hemoglobin (g/dL) | 14.28 ± 1.50 | 13.78 ± 1.61 | 13.55 ± 1.63 | <0.01 |
WBC (103/mL) | 6.76 ± 1.78 | 6.84 ± 1.88 | 6.93 ± 1.63 | 0.22 |
Antidiabetic medication (%) | ||||
Thiazolidinedione | 45 (28.5) | 46 (29.3) | 41 (26.6) | 0.45 |
Metformin | 134 (84.8) | 139 (82.5) | 128 (83.1) | 0.67 |
Sulfonylurea | 80 (50.6) | 76 (48.4) | 84 (54.5) | 0.51 |
DPP-IV inhibitor | 39 (24.7) | 37 (23.6) | 39 (25.3) | 0.20 |
Insulin | 21 (13.3) | 25 (15.9) | 22(14.3) | 0.37 |
Smoking (%) | 50 (31.6) | 53 (33.8) | 69 (44.8) | 0.02 |
Hypertension (%) | 75 (47.4) | 83 (52.9) | 91 (59.1) | <0.01 |
Statin use (%) | 92 (58.2) | 94 (59.9) | 99 (64.3) | 0.42 |
Data are represented as the mean ± SD, number (percentage), or median (interquartile range). RDW: red blood cell distribution width; BMI: body mass index; SBP: systolic blood pressure; DBP: diastolic blood pressure; FPG: fasting plasma glucose; HbA1c: hemoglobin A1c; TC: total cholesterol; TG: triglyceride; LDL-C: low-density lipoprotein cholesterol; HDL-C: high-density lipoprotein cholesterol; HOMA-IR: homeostasis model assessment of insulin resistance; WBC: white blood cell.
C-IMT values were 0.740 ± 0.120 mm, 0.772 ± 0.138 mm, and 0.795 ± 0.139 mm, in the order of first, second, and third RDW tertiles, respectively, and ANOVA showed significant differences in C-IMT values between tertiles 1 and 2 and tertiles 2 and 3 (Figure
Relationship between C-IMT and RDW tertiles in type 2 diabetes.
The C-IMT values significantly increased across RDW tertiles.
Correlation analyses revealed that C-IMT significantly correlated with age, male gender, BMI, SBP, DBP, insulin, HOMA-IR, smoking, and RDW. In multiple linear regression analysis, RDW was associated with C-IMT after control for covariates (Table
Correlations and multiple regression of risk factors associated with C-IMT.
Age | 0.497 | <0.01 | 0.228 | <0.001 |
Sex (M versus F) | 0.229 | <0.01 | 0.191 | <0.001 |
DM duration (yrs) | 0.216 | <0.01 | 0.087 | 0.108 |
BMI (kg/m2) | 0.089 | 0.047 | 0.039 | 0.459 |
SBP (mmHg) | 0.196 | 0.031 | 0.135 | 0.041 |
DBP (mmHg) | 0.107 | 0.036 | 0.057 | 0.323 |
FPG (mmol/L) | 0.013 | 0.590 | — | — |
HbA1c (%) | 0.027 | 0.565 | — | — |
TC (mmol/L) | 0.050 | 0.192 | — | — |
TG (mmol/L) | 0.105 | 0.074 | — | — |
LDL-C (mmol/L) | 0.082 | 0.322 | — | — |
HDL-C (mmol/L) | −0.062 | 0.190 | — | — |
Insulin (mIU/dL) | 0.092 | 0.061 | — | — |
HOMA-IR | 0.125 | 0.020 | 0.100 | 0.042 |
RDW (%) | 0.162 | <0.01 | 0.112 | 0.030 |
Smoking (smokers versus nonsmokers) | 0.173 | <0.01 | 0.131 | 0.015 |
C-IMT: carotid intima-media thickness; M: male; F: female; DM: diabetes mellitus; BMI: body mass index; SBP: systolic blood pressure; DBP: diastolic blood pressure; FPG: fasting plasma glucose; HbA1c: hemoglobin A1c; TC: total cholesterol; TG: triglyceride; LDL-C: low-density lipoprotein cholesterol; HDL-C: high-density lipoprotein cholesterol; HOMA-IR: homeostasis model assessment of insulin resistance; RDW: red blood cell distribution width. Continuous variables with skewed distributions (TG, insulin, and HOMA-IR) were log-transformed for analysis.
The association between RDW and carotid atherosclerosis was further explored by categorizing RDW into three groups and using the first group as the reference. Compared to subjects at the lowest tertile of RDW, those at the 2nd and 3rd tertile of RDW were at a significantly higher risk of having C-IMT ≥ 1.0 mm (OR 1.68 and 2.12, resp.), after adjusting for age and sex. Those at the highest tertile RDW were 1.95 times more likely to have C-IMT ≥ 1.0 mm after adjusting for various factors related to atherosclerosis, including blood pressure, smoking, BMI, use of statin, and lipid and glucose parameters (Table
Odds ratios and 95% confidence intervals for increased carotid IMT according to RDW.
RDW tertile | ||||
---|---|---|---|---|
I | II | III | ||
OR (model 1) | 1.00 | 1.89 (1.10–3.25) | 2.77 (1.63–4.70) | <0.01 |
OR (model 2) | 1.00 | 1.68 (1.03–2.80) | 2.12 (1.18–4.23) | <0.01 |
OR (model 3) | 1.00 | 1.28 (0.87–2.01) | 1.95 (1.08–3.52) | 0.03 |
RDW: red blood cell distribution width; OR: odds ratio. Carotid atherosclerosis was defined as C-IMT ≥ 1.0 mm. Model 1: unadjusted. Model 2: adjusted for age and sex. Model 3: adjusted for age, sex, DM duration, BMI, SBP, DBP, TG, HDL-C, LDL-C, HOMA-IR, and smoking.
In this study, we investigated the relationship between RDW and C-IMT as a marker of subclinical atherosclerosis in people with type 2 diabetes without a history of cardiovascular or cerebrovascular diseases. A positive association was identified between RDW and C-IMT, and it remained significant after adjusting for other cardiovascular risk factors. Participants with highest tertile RDW were over twice more likely to have C-IMT greater than or equal to 1.0 mm compared to those with the lowest tertile. To the best of our knowledge, this is the first study to demonstrate a correlation between RDW and C-IMT in people with type 2 diabetes.
Our results showed that elevated RDW is associated with age, prevalence of hypertension, BMI, SBP, and smoking in people with type 2 diabetes mellitus. These findings are consistent with those of previous studies [
Recently, several studies have demonstrated the association between RDW and increased cardiovascular events and mortality [
Although C-IMT measurement can be done without concerns for radiation exposure or radiocontrast dye-induced nephropathy compared with other diagnostic tools for atherosclerosis, such as cardiac computerized tomography or brain magnetic resonance angiography, it is expensive and many clinics are not equipped with the ultrasound. Therefore, along with other well-known risk factors for atherosclerosis, an increased RDW could be considered another marker for atherosclerosis in people with type 2 diabetes without CVD.
Although the exact pathophysiological mechanism has not been clarified, it may be linked to oxidative stress and inflammation. Chronic inflammation and oxidative stress play important roles in arterial atherosclerosis [
Our study had several limitations. First, owing to the cross-sectional design, we were unable to determine whether there was a causal relationship between RDW and C-IMT in people with type 2 diabetes. Second, our study population was relatively small. A prospective, larger population studies are needed to confirm the relationship, and they may also provide an RDW cutoff point for a clinically significant carotid atherosclerosis or a need for further evaluation. Also, carotid plaque volume has been shown to be a better predictor of cardiovascular risk compared with intima-media thickness [
We found a significant association between RDW and C-IMT in people with type 2 diabetes without a history of cardiovascular or cerebrovascular diseases. Increased RDW might be a risk factor for subclinical atherosclerosis in people with type 2 diabetes.
These study results have been presented as a poster at the Korean Association of Internal Medicine Spring Conference 2016.
The authors declare that there are no conflicts of interest regarding the publication of this article.
This study was supported by a grant from the Seoul R&BD Program, Republic of Korea (10526).