White Blood Cell Count Is Associated With Macro- and Microvascular Complications in Chinese Patients With Type 2 Diabetes

The Prince of Wales Hospital is the teaching hospital of the Chinese University of Hong Kong and serves a population of >1.2 million. Each week, the Diabetes Clinic registers 15–20 new patients who were referred from the community and hospital clinics or discharged from the hospital and reviews 120–150 patients. Since 1995, as part of a continuous quality improvement program, all newly referred patients to the clinic underwent a comprehensive assessment of complications and risk factors based on the European DiabCare protocol (11). This included documentation of demographic data, use of health care resources, and previous medication use.

Clinical assessments included measurement of BMI, waist-to-hip ratio (WHR), blood pressure, visual acuity, fundoscopy through dilated pupils, and foot examination using monofilament and a graduated tuning fork. Mean arterial pressure was defined as the sum of the diastolic blood pressure and 33% of the pulse pressure. Peripheral vascular disease was defined by the absence of foot pulses on palpation, confirmed by Doppler ultrasound examination of the ankle-to-brachial ratio of <0.9. Fundi were examined by a physician or ophthalmologist. Retinopathy was defined by the presence of dot and blot hemorrhages, hard exudates, cotton wool spots, neovascularization, laser scars, and a history of vitrectomy. Sensory neuropathy was diagnosed if two of the following findings were present: reduced sensation to monofilament examination in any part of the sole with normal skin, a score of ≤6/8 (age ≤65 years) or ≤4/8 (age >65 years) using the graduated tuning fork, or typical symptoms of numbness or abnormal sensation over both lower limbs. Fasting plasma blood samples were taken for measurement of glucose, lipid levels (including total cholesterol, HDL cholesterol, triglycerides [TGs], and calculated LDL cholesterol), and renal and liver function. Samples for complete blood count and HbA_1c were collected in EDTA tubes. Timed urine collections were obtained on two occasions. Normal albuminuria was defined as 24-h urinary albumin excretion (UAE) <30 mg/day, microalbuminuria as UAE of 30–300 mg/day, and macroalbuminuria as UAE >300 mg/day. The albuminuric status was defined by concordant results from both urine samples (12).

To minimize the confounding effect of infection, only subjects with a WBC count within the normal range (3.5–12.5 × 10⁹/l) were included in the analysis. Patients presenting with symptoms suggestive of type 1 diabetes, defined as diabetic ketoacidosis, acute presentation with heavy ketonuria (3+), or continuous requirement of insulin within 1 year of diagnosis were excluded (13). At baseline, the presence of macrovascular complications was defined by a history of angina, myocardial infarction, stroke, or peripheral vascular disease. To enhance the specificity of microvascular complications of diabetes, both retinopathy and albuminuria were required to be present. Information on smoking habits was assessed by a standardized questionnaire. Patients’ smoking status was classified as never having smoked, former smoker (ceased smoking for at least 1 year), or current smoker. In this study, former and current smokers were analyzed as a group and compared with those who had never smoked.

Plasma glucose was measured by a hexokinase method (Hitachi 911 automated analyzer; Boehringer Mannheim, Mannheim, Germany). HbA_1c was measured using an automated ion-exchange chromatographic method (Bio-Rad Laboratory, Hercules, CA; reference range 5.1–6.4%). The inter- and intra-assay coefficient of variation (CV) for HbA_1c was ≤3.1% at values <6.5%. Total cholesterol, TGs, and HDL cholesterol were measured by enzymatic methods on the Hitachi 911 automated analyzer using reagent kits supplied by the manufacturer. LDL cholesterol was calculated by Friedewald’s equation for TG <4.5 mmol/l (14). The precision performance of these assays was within the manufacturer’s specifications. Urinary creatinine (Jaffe’s kinetic method) and albumin (immunoturbidimetry method) were also measured on the Hitachi 911 analyzer using reagent kits supplied by the manufacturer. The interassay precision CV was 12.0 and 2.3% for urinary albumin concentrations of 8.0 and 68.8 mg/l, respectively. The lowest detection limit was 3.0 mg/l. Plasma creatinine (Jaffe’s kinetic method) was measured on a Dimension AR system (Dade Behring, Deerfield, IL). A complete blood profile, including WBC count, was measured using an automated cell counter (GEN-S; Beckman Coulter, Miami, FL).

The analysis was performed using the SPSS (version 10.1) statistical package. Plasma TGs, plasma creatinine, and albuminuria were logarithmically transformed because of skewed distributions. All data are expressed as means ± SD or median (interquartile range), as appropriate. Student’s t test or ANOVA was used for between-group comparisons for continuous variables, and the χ² test was used for categorical variables. WBC count was grouped into quintiles, and the adjusted odds ratio (OR) for these quintiles was computed with the lowest category as the referent group. The logistic regression model was used to estimate the OR (95% CI) for diabetic complications. In the regression analysis, known conventional risk factors for cardiovascular diseases were included as covariates in the final model. P < 0.05 (two tailed) was considered to be significant.

In this cross-sectional analysis of 3,776 type 2 diabetic patients, 42.1% were men and 28.7% were smokers. The mean age was 59.0 ± 13.5 years, and the duration of diabetes was 7.4 ± 6.6 years. The mean WBC count was 7.2 ± 1.7 ×10⁹/l. Table 1 summarizes the demographic, anthropometric, and metabolic characteristics of the study sample according to the quintiles of WBC count. Subjects with higher WBC counts had longer disease duration, higher systolic blood pressure, diastolic blood pressure, BMI, WHR, HbA_1c, fasting plasma glucose, LDL cholesterol, TGs, plasma creatinine, and UAE, and lower HDL cholesterol (P < 0.001 for trend for all). Univariate analysis revealed positive associations of WBC count with BMI, waist circumference, blood pressure, glycemic control, LDL cholesterol, TGs, and UAE, and a negative association with HDL cholesterol (all P < 0.001) (Table 2).

Of the 3,776 subjects, 14.4% had macrovascular complications at baseline. The prevalence rates of albuminuria and retinopathy were 42.8 and 26.5%, respectively. The frequencies of macro- and microvascular complications increased in a dosage-dependent manner as the WBC count increased (P < 0.001 for trend for all) (Fig. 1).

Smokers had a higher risk of having macrovascular complications (OR 1.83 [95% CI 1.51–2.21]; P < 0.001) and microvascular complications (1.36 [1.13–1.63]; P = 0.001) compared with nonsmokers. Using nonsmokers in the lowest quintile of WBC count as the referent, the age-adjusted ORs for macrovascular diseases in the highest WBC quintile were 2.88 (1.92–4.31; P < 0.001) for nonsmokers and 3.61 (2.33–5.59; P < 0.001) for former or current smokers (Fig. 2A). For the coexistence of retinopathy and albuminuria, the age-adjusted ORs in the highest WBC quintile were 2.54 (1.78–3.61; P < 0.001) for nonsmokers and 2.56 (1.73–3.81; P = 0.003) for former or current smokers (Fig. 3A).

After adjustment for sex, age, smoking status, disease duration, BMI, WHR, mean arterial pressure, HbA_1c, HDL cholesterol, and LDL cholesterol, patients with higher WBC counts remained at increased risk for having macrovascular (Fig. 2B) or microvascular complications (Fig. 3B) (P < 0.001 for trend for both). When entered as a continuous variable in the multivariate-adjusted models for both macro- and microvascular diseases, the WBC count remained a significant factor of complications after controlling for the conventional risk factors. For macrovascular complications, a 1-unit (1 × 10⁹/l) increment of WBC count was associated with a 15.8% (95% CI 9.3–22.6; P < 0.001) increased risk. The correlation between WBC count and macrovascular complications remained significant after the inclusion of UAE as a covariate in the final model (1.14 [1.07–1.21]; P < 0.001). For microvascular complications, the risk was increased by 12.3% (5.8–19.1; P < 0.001) with each unit increment of WBC count.

In this large-scale cross-sectional study involving 3,776 Chinese patients with type 2 diabetes, WBC count, even within the normal range, was independently associated with both macro- and microvascular complications of diabetes in a dosage-related manner and after controlling for conventional risk factors, including smoking, blood pressure, lipids, albuminuria, and glucose control as well as obesity. This finding is in agreement with current evidence regarding the association of inflammatory markers, including WBC count, with the development of metabolic syndrome, coronary heart disease, and all-cause mortality (4–8,15–20).

In this study, the prevalence of macrovascular disease was 2.4-fold higher in subjects with WBC count >8.6 ×10⁹/l than in those with counts <5.6 ×10⁹/l. Despite the potential effects of smoking on WBC count (21,22) and the controversy regarding the effect of WBC count on cardiovascular diseases independent of smoking (5,23,24), we were able to demonstrate a dosage-dependent association between WBC count and macrovascular diseases in both smokers and nonsmokers.

Chinese patients with type 2 diabetes have an excessive burden of microvascular complications (25). In our study cohort, the prevalence rates of retinopathy and albuminuria were two and three times the rate of macrovascular complications, respectively. We even defined microvascular complications as the presence of both retinopathy and albuminuria to minimize false positive classification. The OR of microvascular complications increased in a stepwise fashion with progressive quintiles of WBC count. This finding is in agreement with that of the Insulin Resistance Atherosclerosis Study, which showed an association of C-reactive protein and fibrinogen with albuminuria in subjects with type 2 diabetes (26). A recent study has also demonstrated a relation between albuminuria and WBC count (10), as is seen in our present study. By contrast, the association between fibrinogen and albuminuria was less reproducible, although the former parameter was not measured in our present study (27–31). Unlike macrovascular complications, smoking status did not appear to have an additional effect on the prevalence of microvascular complications in our study.

The close association between the WBC count and both micro- and macrovascular complications raises the hypothesis that inflammation may be a common linking factor. In support of this notion, the inflammatory process is now recognized to be a major component of atherosclerosis (4). Mononuclear leukocytes are recruited to the site of endothelial injury and form foam cells in the plaque (32). Activation of neutrophil leads to changes in rheological properties and adherence to the endothelium, all of which lead to capillary plugging and tissue ischemia (33). Furthermore, various cytokines and growth factors, such as interleukins, tumor necrosis factor-α, and transforming growth factor-β1 (TGF-β1) are released from activated leukocytes (34,35) to cause endothelial dysfunction (36,37). In this connection, the leukocyte count has been shown to be an independent predictor of both endothelium-dependent and -independent vasodilation in type 2 diabetic patients (38). Furthermore, increased secretion of TGF-β1 by mononuclear cells has been demonstrated in patients with diabetic nephropathy (39). Elevated TGF-β1 levels in the glomeruli stimulate proliferation of mesangial and epithelial cells, leading to a matrix expansion typical of glomerulosclerosis (40,41). In addition, activated leukocytes can release superoxide radicals and proteases, all of which promote oxidative stress. The latter can then activate the transcription factor nuclear factor-κB in peripheral mononuclear blood cells. All these pathways can lead to diabetic nephropathy (42). Taken together, it is plausible that low-grade chronic inflammatory responses can interact with other risk factors, leading to widespread vascular damage, endothelial dysfunction, increased oxidative stress, and increased production of growth factors and cytokines to cause micro- and macrovascular complications in type 2 diabetic patients.

Despite the circumstantial evidence, it remains plausible that the elevated WBC count may be an effect of vascular complications (e.g., inflammation after micro-infarction). Nevertheless, the inflammatory response has been shown to be involved in the development of vascular occlusion and ischemic organ damage in other diseases. A notable example is sickle cell anemia, a common hereditary hemoglobinopathy caused by a single-point mutation of the β-globin chain of the hemoglobin. Elevated WBC counts have been shown to correlate with the incidence of stroke in children with sickle cell anemia (43) and may precipitate sickle crisis (44). Increased expression of adhesion molecules by leukocytes may be important in the pathogenesis of sickle cell complications. Patients with complications of sickle cell disease have a high expression of αMβ₂ integrin and l-selectin (45). More importantly, an improvement in symptoms after hydroxyurea therapy is accompanied by a marked drop in WBC counts without the rise in hemoglobin F (46) and reduced expression of adhesion molecules in leukocytes (45). In multivariate analysis, the reduction of total WBC count is a predictor of clinical response to hydroxyurea (47). Elevated WBC counts may therefore contribute to the development of vascular complications, although prospective studies are required to address this issue. Another limitation of the present study was the lack of measurement of C-reactive proteins, a specific marker of inflammation. Given the close association between C-reactive proteins and WBC count, we opted to measure the latter, which was more feasible in a cohort study involving large number of subjects.

In conclusion, elevated WBC count, although still within the normal range, is associated with both macro- and microvascular complications in Chinese patients with type 2 diabetes. Chronic inflammation may play a crucial role in the pathogenesis of retinopathy and cardiovascular and renal complications of diabetes. Further research is required to establish the causal relation of WBC and diabetic complications and the underlying mechanisms.

The study was supported by funding from the Hong Kong Foundation for Research and Development in Diabetes.

We thank our research nurses for their professionalism as well as all medical and nursing staff at the Diabetes and Endocrine Centre, the Prince of Wales Hospital, for their dedication.