OBJECTIVE—Recent studies have shown that C-reactive protein (CRP) predicts future risk of diabetes in healthy Caucasians. We determined whether plasma CRP level was elevated in Chinese subjects with impaired glucose tolerance (IGT) and whether CRP level could be used to predict progression to type 2 diabetes or reversion to normal glucose tolerance (NGT) in these high-risk individuals.

RESEARCH DESIGN AND METHODS—A total of 228 subjects with IGT at baseline from the Hong Kong Cardiovascular Risk Factors Prevalence Study underwent repeat oral glucose tolerance testing after 2 years. Plasma high-sensitivity CRP was measured from their stored baseline samples and from 228 subjects with NGT matched for age and BMI by an immunoturbidimetric assay.

RESULTS—Subjects with IGT at baseline had higher plasma CRP levels than subjects with NGT: 1.18 mg/l (0.52–2.52) vs. 0.87 mg/l (0.37–1.84), median (interquartile range), P = 0.01. At 2 years, 117 subjects with IGT reverted to NGT, 84 remained in IGT, and 21 progressed to diabetes. Individuals who progressed to diabetes had the highest plasma CRP levels at baseline (P < 0.0001). Those with baseline CRP levels in the third and top quartile had a relative risk of remaining in IGT or progressing to diabetes of 2.87 (95% CI 1.06–7.82) and 2.76 (1.06–7.31), respectively, after adjusting for anthropometric measure and lifestyle factors.

CONCLUSIONS—CRP independently predicts the risk of remaining in IGT or progressing to diabetes in Chinese subjects with IGT. CRP might provide an adjunctive measure for identifying subjects with the highest risk of progression to diabetes who would derive the greatest benefits from preventive interventions.

Recent studies have suggested that low-grade chronic inflammation may be involved in the pathogenesis of type 2 diabetes. Inflammatory markers such as high white blood cell count, high fibrinogen level, and C-reactive protein (CRP) level have been related to the development of type 2 diabetes. A number of studies have shown that plasma CRP level predicts the development of type 2 diabetes in middle-aged Caucasian men and women and in elderly subjects (1–5). A positive association between markers of acute-phase response and features of the insulin resistance syndrome has also been reported (6–8). Plasma CRP concentration is increased in patients with impaired glucose tolerance (IGT) and in newly detected type 2 diabetic patients compared with subjects with normal glucose tolerance (NGT), and the increase in inflammatory parameters is parallel to the stages of glucose intolerance (9–11). Recent data from Japan have shown that other inflammatory markers may also be a useful predictor in Asian subjects, and a strong association between white blood cell count and features of the metabolic syndrome has been reported in Japanese middle-aged men (12,13). The objectives of the present study are, first, to determine whether plasma CRP level is elevated in Chinese subjects with IGT because Chinese individuals, even in those with IGT, are generally less obese than Caucasians (14), and second, to determine whether plasma CRP level can be used to predict those individuals with IGT who will progress to type 2 diabetes or revert to NGT.

The Hong Kong Cardiovascular Risk Factors Prevalence Study is a population-based study of diabetes and cardiovascular risk factors (14). During 1995–1996, 2,900 subjects aged 25–74 years were recruited and all subjects without a history of diabetes underwent a 75-g oral glucose tolerance test (OGTT). A total of 434 subjects with IGT were identified and were subsequently invited to participate in a prospective study on the natural history of IGT in our community (15). A total of 322 subjects returned for a repeat OGTT at 2 years, and World Health Organization (WHO) 1999 criteria were used to define IGT and diabetes (16). Only subjects with stored baseline plasma samples available for measurement of CRP (n = 228) were included in the present study. The baseline demographic characteristics of the 228 subjects with IGT did not differ from those subjects with IGT who were excluded from the analysis. Plasma CRP was also measured in baseline samples from 228 subjects with NGT who were matched for age and BMI to establish a normal range.

Fasting plasma glucose and 2-h glucose levels were measured by hexokinase method on a Hitachi 747 analyzer (Boehringer Mannheim, Mannheim, Germany). Insulin was measured by microparticle enzyme immunoassay (MEIA; Abbott Laboratories, Tokyo, Japan). Homeostasis model assessment index was calculated to estimate insulin resistance (HOMA-IR = fasting glucose × fasting insulin/22.5). Plasma total cholesterol and triglyceride levels were determined enzymatically on a Hitachi 717 analyzer (Boehringer Mannheim). HDL cholesterol was measured by the same method after precipitation of apolipoprotein B (apoB)-containing lipoproteins with PEG 6000. LDL cholesterol was calculated using the Friedewald equation. Apolipoprotein AI and apoB were measured by rate nephelometry using the Beckman Array System (Beckman Coulter Instruments, Galway, Ireland). Plasma high-sensitivity CRP was measured using a particle-enhanced immunoturbidimetric assay (Roche Diagnostics, Mannheim, Germany) using anti-CRP mouse monoclonal antibodies coupled to latex microparticles. The assay was standardized against the CRM 470 Reference Preparation for Proteins in Human (RPPHS), with a functional sensitivity of 0.1 mg/l. The intra-assay coefficients of variation were 1.83% at 0.60 mg/l, 2.13% at 6.42 mg/l, and 0.49% at 19.38 mg/l; and interassay coefficients of variation were 7.01% at 0.58 mg/l, 2.55% at 6.39 mg/l, and 2.64% at 18.78 mg/l.

Results are expressed as means and SDs or as medians and interquartile ranges if the distribution of the data was found to be skewed. Data that were not normally distributed were logarithmically transformed before analyses were made. Student’s t test and ANOVA were used to analyze continuous variables from two groups and multiple groups, respectively, and proportions were compared using χ2 test. Subjects with plasma CRP levels >15 mg/l (six from the IGT group and four from the NGT group), indicating clinically relevant inflammatory conditions, were excluded from the analysis; none of the six subjects with IGT developed diabetes at 2 years, but four subjects reverted to NGT and two remained IGT. Quartile limits of CRP were determined based on the distribution in subjects with NGT. Quartile-specific risk estimates were obtained in the group of subjects with IGT at baseline using binary logistic regression with glycemic status at 2 years (reversion to NGT versus IGT/incident type 2 diabetes) as the dependent variable. Multinomial logistic regression was performed to determine the predictive power of various biochemical and demographic parameters in determining the progression from IGT to diabetes or reversion to NGT by comparing those two groups against those who remained in IGT at 2 years. The sensitivity and specificity of using various cutoff values of CRP in the prediction of diabetes were examined using the receiver operating characteristic curve analysis.

The baseline characteristics of subjects with NGT and IGT are shown in Table 1. Despite having similar BMI, waist circumference, and waist-to-hip ratio, subjects with IGT still had significantly higher plasma levels of CRP than subjects with NGT. Fasting and 2-h insulin levels were higher in subjects with IGT, and HOMA-IR was also elevated, suggesting that many of these subjects were insulin resistant with hyperinsulinemia.

At 2 years, 117 subjects with IGT had reverted back to NGT, 84 remained in IGT, and 21 had progressed to type 2 diabetes. The baseline clinical and biochemical characteristics based on the category of glucose tolerance by 2 years are shown in Table 2. Those who reverted back to NGT had lower baseline BMI, waist circumference, and waist-to-hip ratio than those who remained in IGT or progressed to diabetes (Table 2). They also had the lowest fasting glucose, 2-h glucose, and fasting triglyceride levels as well as systolic blood pressure. There were highly significant differences in plasma CRP levels among the three groups; those who progressed to diabetes had the highest plasma levels.

Plasma CRP levels were divided into quartiles based on the distribution in those 224 subjects with NGT at baseline: quartile 1, median 0.16 mg/l (range 0.10–0.36); quartile 2, 0.58 mg/l (0.37–0.87); quartile 3, 1.29 mg/l (0.88–1.90); quartile 4, 3.59 mg/l (1.91–14.43). A total of 17% of the IGT subjects had plasma CRP values in quartile 1, 22% in quartile 2, 25% in quartile 3, and 36% in quartile 4. In subjects with IGT at baseline, the odds ratio for remaining in IGT or progressing to diabetes per mg/l for CRP and for log(CRP) are 1.19 (1.03–1.39) and 2.72 (1.51–4.89), respectively. The distribution of parameters that are potential covariates, such as anthropometric measures, lifestyle factors, and various biochemical measurements by CRP quartiles, are shown in Table 3. The relative risk according to baseline CRP quartiles before and after adjusting for a number of covariates is shown in Table 4. Subjects with IGT whose baseline CRP levels were in quartiles 3 and 4 had a relative risk of remaining in IGT or progressing to diabetes of 3.5 and 3.9, respectively. The CRP cutpoint of 0.88 mg/l for quartile 3 has a sensitivity of 100% and specificity of 43% in the prediction of diabetes, whereas the cutpoint of 1.91 mg/l for quartile 4 has a sensitivity of 71% and specificity of 69%.

Multinomial logistic regression was performed to determine the predictive power of baseline parameters in determining progression from IGT to diabetes or reversion to NGT using those who remained in IGT at 2 years as the reference group. Baseline parameters that showed significant differences by 2-year glycemic status were used; results are shown in Table 5. For progression from IGT to diabetes, 2-h glucose, log(CRP), and sex were significant predictors, whereas for reversion from IGT to NGT, 2-h glucose, BMI, and age were the main predictors.

Previous studies have shown that fasting and postload glucose concentrations and BMI measured at the time of IGT recognition were the most consistent and strongest predictors of the progression from IGT to type 2 diabetes (17,18). We have shown that in our cohort of IGT subjects, plasma CRP level was also a useful predictor. Those IGT subjects with plasma CRP levels in the top two quartiles had approximately three times the risk of either remaining in IGT or progressing to diabetes compared with those individuals with plasma CRP in the lowest quartile. Previous studies have shown a graded increase in diabetes risk according to CRP quartiles or quintiles (3,4), but we did not find a graded increase in risk between the third and fourth quartiles of CRP in the present study. This is probably because of our relatively small sample size and the limited number of events. However, as shown in Table 4, a much larger proportion of subjects in quartile 4 progressed to diabetes compared with quartile 3 (20 vs. 9%, respectively). The predictive power of CRP remained significant even after adjusting for anthropometric measures and lifestyle factors such as smoking, physical activity, and alcohol consumption. The strongest predictors of progression to diabetes in this study were 2-h glucose and plasma CRP levels. To identify individuals who would revert to NGT, 2-h glucose level and BMI were the most useful indicators. Most of the subjects who progressed from IGT to diabetes were in the top CRP quartile. They had increased waist-to-hip ratio, waist circumference, and plasma triglyceride level and fulfilled the criteria of the metabolic syndrome by WHO standards (16). This is consistent with previous studies showing associations between inflammatory markers and features of the metabolic syndrome and suggests that chronic subclinical inflammation may play a role in the pathogenesis of disorders related to insulin resistance (6–8).

The conversion rate from IGT to diabetes in our study was 9.5% over 2 years or 4.8% per year. The yearly incidence of diabetes in our subjects with IGT was similar to that reported by epidemiological data (1.5–7.2%) (17) and the Finnish Study (6%) (19), but lower than that reported in the Da Qing Study (15.7%) (20) and the STOP-Type 2 Diabetes Study (12.4%) (21). It is well recognized that individuals with IGT can later revert to nondiabetic status and the significance of transient IGT may differ in different populations (22). More than 50% of the IGT subjects in our study reverted to NGT at 2 years without intervention. This may be partly due to the variability in OGTT results accounted for by the phenomenon of regression to the mean (biological variability) (17), awareness in the population examined, or factors known to impair glucose tolerance. None of our subjects were taking medications that could affect glucose tolerance, and no specific medical advice was given after the first OGTT. There is a need to refine our ability to identify subjects with IGT who will progress to type 2 diabetes because it is now possible to prevent diabetes in these high-risk subjects by lifestyle modification and/or pharmacological interventions (19–21,23). The validity of using CRP to improve our power to predict diabetes in the present study is limited by the short follow-up period and the small number of subjects progressing to diabetes at 2 years and will need to be confirmed by further prospective studies in the Chinese population.

In summary, we have shown that Chinese subjects with IGT have elevated plasma CRP levels and that CRP independently predicts the risk for either remaining in IGT or progressing to type 2 diabetes over a 2-year period. Plasma CRP level might provide an adjunctive measure for identifying subjects at highest risk for progression to diabetes, who, therefore, would derive the greatest benefits from preventive interventions.

Table 1—

Baseline characteristics of subjects with NGT and IGT

NGT (n = 224)IGT (n = 222)P value
Men/women (%) 44.6/55.4 41.9/58.1 NS 
Age (years) 50.4 ± 11.9 50.5 ± 11.8 NS 
Smoker/exsmoker/nonsmoker (%) 10.3/7.1/82.6 11.7/9.9/78.4 NS 
BMI (kg/m2) 25.2 ± 3.4 25.7 ± 3.4 NS 
Waist (cm) 81.6 ± 8.8 82.8 ± 9.6 NS 
Waist-to-hip ratio 0.86 ± 0.008 0.87 ± 0.008 NS 
CRP (mg/l)* 0.87 (0.37–1.84) 1.18 (0.52–2.52) 0.011 
Fasting glucose (mmol/l) 5.09 ± 0.48 5.42 ± 0.58 <0.0001 
2-h glucose (mmol/l) 5.84 ± 1.10 8.92 ± 0.83 <0.0001 
Fasting insulin (pmol/l) 5.88 ± 3.41 7.41 ± 5.32 <0.0001 
2-h insulin (pmol/l) 58.05 ± 40.35 97.54 ± 62.09 <0.0001 
HOMA-IR 1.34 ± 0.82 1.83 ± 1.44 <0.0001 
Total cholesterol (mmol/l) 5.07 ± 0.96 5.34 ± 0.95 0.003 
Triglyceride (mmol/l)* 1.11 ± 0.57 1.45 ± 0.79 <0.0001 
LDL cholesterol (mmol/l) 3.28 ± 0.87 3.46 ± 0.88 0.03 
HDL cholesterol (mmol/l) 1.28 ± 0.31 1.22 ± 0.33 NS 
ApoA (g/l) 1.39 ± 0.26 1.38 ± 0.29 NS 
ApoB (g/l) 0.98 ± 0.29 1.05 ± 0.28 0.007 
Systolic blood pressure (mmHg) 123.9 ± 18.5 126.4 ± 22.0 NS 
Diastolic blood pressure (mmHg) 76.4 ± 9.8 78.3 ± 11.6 NS 
NGT (n = 224)IGT (n = 222)P value
Men/women (%) 44.6/55.4 41.9/58.1 NS 
Age (years) 50.4 ± 11.9 50.5 ± 11.8 NS 
Smoker/exsmoker/nonsmoker (%) 10.3/7.1/82.6 11.7/9.9/78.4 NS 
BMI (kg/m2) 25.2 ± 3.4 25.7 ± 3.4 NS 
Waist (cm) 81.6 ± 8.8 82.8 ± 9.6 NS 
Waist-to-hip ratio 0.86 ± 0.008 0.87 ± 0.008 NS 
CRP (mg/l)* 0.87 (0.37–1.84) 1.18 (0.52–2.52) 0.011 
Fasting glucose (mmol/l) 5.09 ± 0.48 5.42 ± 0.58 <0.0001 
2-h glucose (mmol/l) 5.84 ± 1.10 8.92 ± 0.83 <0.0001 
Fasting insulin (pmol/l) 5.88 ± 3.41 7.41 ± 5.32 <0.0001 
2-h insulin (pmol/l) 58.05 ± 40.35 97.54 ± 62.09 <0.0001 
HOMA-IR 1.34 ± 0.82 1.83 ± 1.44 <0.0001 
Total cholesterol (mmol/l) 5.07 ± 0.96 5.34 ± 0.95 0.003 
Triglyceride (mmol/l)* 1.11 ± 0.57 1.45 ± 0.79 <0.0001 
LDL cholesterol (mmol/l) 3.28 ± 0.87 3.46 ± 0.88 0.03 
HDL cholesterol (mmol/l) 1.28 ± 0.31 1.22 ± 0.33 NS 
ApoA (g/l) 1.39 ± 0.26 1.38 ± 0.29 NS 
ApoB (g/l) 0.98 ± 0.29 1.05 ± 0.28 0.007 
Systolic blood pressure (mmHg) 123.9 ± 18.5 126.4 ± 22.0 NS 
Diastolic blood pressure (mmHg) 76.4 ± 9.8 78.3 ± 11.6 NS 

Data are means ± SD. ApoA, apolipoprotein A.

*

Median (interquartile range).

View Large
Table 2—

Baseline clinical and biochemical characteristics of subjects with IGT by their category of glucose tolerance at 2 years

NGT (n = 117)IGT (n = 84)Diabetes (n = 21)P value (ANOVA)
Men/women (%)† 35/65 44/56 71/29 0.007 
Age (years) 48.2 ± 11.6 53.0 ± 11.6 52.8 ± 11.8 0.011 
BMI (kg/m2) 24.9 ± 3.8 26.4 ± 2.9 26.7 ± 2.5 0.003 
Waist (cm) 80.1 ± 10.2 85.5 ± 7.5 87.4 ± 9.2 <0.0001 
Waist-to-hip ratio 0.85 ± 0.008 0.89 ± 0.007 0.90 ± 0.007 <0.0001 
CRP (mg/I)* 0.82 (0.42–2.09) 1.39 (0.68–2.70) 2.39 (1.82–2.97) <0.0001 
Fasting glucose (mmol/l) 5.34 ± 0.55 5.47 ± 0.56 5.63 ± 0.78 0.05 
2-h glucose (mmol/l) 8.69 ± 0.72 9.04 ± 0.81 9.69 ± 0.98 <0.0001 
Fasting insulin (pmol/l) 7.74 ± 6.39 7.13 ± 3.71 6.71 ± 3.96 NS 
2-h insulin (pmol/l) 103.26 ± 66.65 92.38 ± 57.53 85.9 ± 51.50 NS 
HOMA-IR 1.88 ± 1.72 1.77 ± 1.04 1.72 ± 1.05 NS 
Total cholesterol (mmol/l) 5.29 ± 0.97 5.44 ± 0.92 5.25 ± 0.97 NS 
Triglyceride (mmol/l)* 1.12 (0.80–1.80) 1.38 (1.00–1.93) 1.50 (1.24–2.26) 0.014 
LDL cholesterol (mmol/l) 3.42 ± 0.86 3.55 ± 0.92 3.34 ± 0.92 NS 
HDL cholesterol (mmol/l) 1.26 ± 0.32 1.18 ± 0.32 1.13 ± 0.40 NS 
ApoA (g/l) 1.37 ± 0.30 1.39 ± 0.29 1.35 ± 0.15 NS 
ApoB (g/l) 1.00 ± 0.27 1.11 ± 0.28 1.09 ± 0.29 0.018 
Systolic blood pressure (mmHg) 122.6 ± 21.5 129.9 ± 23.1 134.2 ± 16.6 0.015 
Diastolic blood pressure (mmHg) 77.1 ± 12.1 79.4 ± 11.2 80.9 ± 10.1 NS 
Exercise-rarely or never (%)† 56.4 63.1 61.9 NS 
Alcohol-rarely or never (%)† 80.3 82.1 85.7 NS 
NGT (n = 117)IGT (n = 84)Diabetes (n = 21)P value (ANOVA)
Men/women (%)† 35/65 44/56 71/29 0.007 
Age (years) 48.2 ± 11.6 53.0 ± 11.6 52.8 ± 11.8 0.011 
BMI (kg/m2) 24.9 ± 3.8 26.4 ± 2.9 26.7 ± 2.5 0.003 
Waist (cm) 80.1 ± 10.2 85.5 ± 7.5 87.4 ± 9.2 <0.0001 
Waist-to-hip ratio 0.85 ± 0.008 0.89 ± 0.007 0.90 ± 0.007 <0.0001 
CRP (mg/I)* 0.82 (0.42–2.09) 1.39 (0.68–2.70) 2.39 (1.82–2.97) <0.0001 
Fasting glucose (mmol/l) 5.34 ± 0.55 5.47 ± 0.56 5.63 ± 0.78 0.05 
2-h glucose (mmol/l) 8.69 ± 0.72 9.04 ± 0.81 9.69 ± 0.98 <0.0001 
Fasting insulin (pmol/l) 7.74 ± 6.39 7.13 ± 3.71 6.71 ± 3.96 NS 
2-h insulin (pmol/l) 103.26 ± 66.65 92.38 ± 57.53 85.9 ± 51.50 NS 
HOMA-IR 1.88 ± 1.72 1.77 ± 1.04 1.72 ± 1.05 NS 
Total cholesterol (mmol/l) 5.29 ± 0.97 5.44 ± 0.92 5.25 ± 0.97 NS 
Triglyceride (mmol/l)* 1.12 (0.80–1.80) 1.38 (1.00–1.93) 1.50 (1.24–2.26) 0.014 
LDL cholesterol (mmol/l) 3.42 ± 0.86 3.55 ± 0.92 3.34 ± 0.92 NS 
HDL cholesterol (mmol/l) 1.26 ± 0.32 1.18 ± 0.32 1.13 ± 0.40 NS 
ApoA (g/l) 1.37 ± 0.30 1.39 ± 0.29 1.35 ± 0.15 NS 
ApoB (g/l) 1.00 ± 0.27 1.11 ± 0.28 1.09 ± 0.29 0.018 
Systolic blood pressure (mmHg) 122.6 ± 21.5 129.9 ± 23.1 134.2 ± 16.6 0.015 
Diastolic blood pressure (mmHg) 77.1 ± 12.1 79.4 ± 11.2 80.9 ± 10.1 NS 
Exercise-rarely or never (%)† 56.4 63.1 61.9 NS 
Alcohol-rarely or never (%)† 80.3 82.1 85.7 NS 

Data are means ± SD. ApoA, apolipoprotein A.

*

Median (interquartile range);

†

proportions compared by χ2 test.

View Large
Table 3—

Baseline anthropometric measures, lifestyle factors, and biochemical parameters of subjects with IGT by CRP quartiles

CRP quartile 1 (n = 37)CRP quartile 2 (n = 50)CRP quartile 3 (n = 55)CRP quartile 4 (n = 80)P value (ANOVA)
Men/women (%)† 43/57 44/56 42/58 40/60 NS 
Age (years) 45.5 ± 10.4 49.3 ± 11.2 52.1 ± 11.7 52.3 ± 12.3 0.02 
BMI (kg/m2) 23.9 ± 3.8 25.0 ± 2.8 25.5 ± 2.8 27.0 ± 3.5 <0.0001 
Waist circumference (cm) 78.2 ± 11.5 80.9 ± 7.1 83.0 ± 9.7 86.0 ± 8.8 <0.0001 
Smoker/exsmoker/nonsmoker (%)† 5/5/90 14/14/72 11/11/78 14/9/77 NS 
Exercise: rarely or never (%)† 46 54 66 65 NS 
Alcohol: rarely or never (%)† 87 80 82 79 NS 
Fasting glucose (mmol/l) 5.45 ± 0.62 5.26 ± 0.42 5.37 ± 0.54 5.52 ± 0.65 NS 
2-h glucose (mmol/l) 8.94 ± 0.75 8.73 ± 0.74 8.74 ± 0.79 9.15 ± 0.90 0.01 
Triglyceride (mmol/l)* 1.11 (0.68–1.80) 1.03 (0.89–1.80) 1.33 (0.92–1.80) 1.54 (1.10–2.00) NS 
CRP quartile 1 (n = 37)CRP quartile 2 (n = 50)CRP quartile 3 (n = 55)CRP quartile 4 (n = 80)P value (ANOVA)
Men/women (%)† 43/57 44/56 42/58 40/60 NS 
Age (years) 45.5 ± 10.4 49.3 ± 11.2 52.1 ± 11.7 52.3 ± 12.3 0.02 
BMI (kg/m2) 23.9 ± 3.8 25.0 ± 2.8 25.5 ± 2.8 27.0 ± 3.5 <0.0001 
Waist circumference (cm) 78.2 ± 11.5 80.9 ± 7.1 83.0 ± 9.7 86.0 ± 8.8 <0.0001 
Smoker/exsmoker/nonsmoker (%)† 5/5/90 14/14/72 11/11/78 14/9/77 NS 
Exercise: rarely or never (%)† 46 54 66 65 NS 
Alcohol: rarely or never (%)† 87 80 82 79 NS 
Fasting glucose (mmol/l) 5.45 ± 0.62 5.26 ± 0.42 5.37 ± 0.54 5.52 ± 0.65 NS 
2-h glucose (mmol/l) 8.94 ± 0.75 8.73 ± 0.74 8.74 ± 0.79 9.15 ± 0.90 0.01 
Triglyceride (mmol/l)* 1.11 (0.68–1.80) 1.03 (0.89–1.80) 1.33 (0.92–1.80) 1.54 (1.10–2.00) NS 

Data are means ± SD.

*

Median (interquartile range);

†

proportions compared by χ2 test.

View Large
Table 4—

Relative risk of remaining in IGT or progressing to diabetes at 2 years according to baseline plasma CRP levels

CRP quartile 1CRP quartile 2CRP quartile 3CRP quartile 4
Glycemic status at 2 years: NGT/IGT/diabetes (n) 27/10/0 33/17/0 24/26/5 33/31/16 
Crude analysis 1 1.39 (0.55–3.53) 3.49 (1.42–8.58) 3.85 (1.64–9.01) 
Adjusted for BMI 1 1.25 (0.48–3.23) 3.04 (1.21–7.60) 2.87 (1.18–6.97) 
Model A (adjusted for age, sex, BMI, smoking, physical activity, and alcohol) 1 1.38 (0.50–3.81) 2.87 (1.06–7.82) 2.76 (1.04–7.31) 
Model A + fasting glucose 1 1.51 (0.54–4.23) 3.06 (1.11–8.45) 2.84 (1.07–7.57) 
Model A + 2-h glucose 1 1.62 (0.57–4.59) 3.63 (1.28–10.28) 2.69 (0.99–7.32) 
Model A + 2-h glucose and waist circumference 1 1.65 (0.58–4.71) 3.62 (1.27–10.31) 2.68 (0.98–7.31) 
Model A + 2-h glucose, waist circumference, and log(triglyceride) 1 1.65 (0.58–4.69) 3.62 (1.27–10.30) 2.68 (0.98–7.31) 
CRP quartile 1CRP quartile 2CRP quartile 3CRP quartile 4
Glycemic status at 2 years: NGT/IGT/diabetes (n) 27/10/0 33/17/0 24/26/5 33/31/16 
Crude analysis 1 1.39 (0.55–3.53) 3.49 (1.42–8.58) 3.85 (1.64–9.01) 
Adjusted for BMI 1 1.25 (0.48–3.23) 3.04 (1.21–7.60) 2.87 (1.18–6.97) 
Model A (adjusted for age, sex, BMI, smoking, physical activity, and alcohol) 1 1.38 (0.50–3.81) 2.87 (1.06–7.82) 2.76 (1.04–7.31) 
Model A + fasting glucose 1 1.51 (0.54–4.23) 3.06 (1.11–8.45) 2.84 (1.07–7.57) 
Model A + 2-h glucose 1 1.62 (0.57–4.59) 3.63 (1.28–10.28) 2.69 (0.99–7.32) 
Model A + 2-h glucose and waist circumference 1 1.65 (0.58–4.71) 3.62 (1.27–10.31) 2.68 (0.98–7.31) 
Model A + 2-h glucose, waist circumference, and log(triglyceride) 1 1.65 (0.58–4.69) 3.62 (1.27–10.30) 2.68 (0.98–7.31) 

Data are odds ratio (95% CI).

View Large
Table 5—

Predictive power of baseline parameters in determining progression from IGT to diabetes or reversion to NGT at 2 years

Approximately 1-SD changeRegression coefficientSEP valueOdds ratio (95% CI)
Diabetes      
 Intercept  −10.03 4.55 0.027  
 Log(CRP) 0.5 log(mg/l) 0.93 0.38 0.014 2.53 (1.21–5.29) 
 Log(triglyceride) 0.3 log(mmol/l) −0.003 0.41 0.99 0.99 (0.45–2.24) 
 Fasting glucose 0.5 mmol/l 0.04 0.23 0.84 1.05 (0.67–1.63) 
 2-h glucose 1 mmol/l 0.88 0.35 0.010 2.42 (1.23–4.76) 
 Age 10 years −0.25 0.26 0.33 0.78 (0.47–1.29) 
 Sex  1.53 0.62 0.023 4.59 (1.38–15.32) 
 BMI 3 kg/m2 −0.11 0.29 0.71 0.90 (0.51–1.57) 
 Systolic blood pressure 20 mmHg 0.12 0.29 0.69 1.13 (0.64–1.99) 
NGT      
 Intercept  9.54 2.58 0  
 Log(CRP) 0.5 log(mg/l) −0.18 0.17 0.30 0.83 (0.59–1.17) 
 Log(triglyceride) 0.3 log(mmol/l) −0.05 0.25 0.84 0.95 (0.59–1.54) 
 Fasting glucose 0.5 mmol/l −0.5 0.15 0.76 0.96 (0.72–1.28) 
 2-h glucose 1 mmol/l −0.45 0.20 0.02 0.64 (0.43–0.94) 
 Age 10 years −0.30 0.15 0.04 0.74 (0.55–0.99) 
 Sex  −0.25 0.34 0.46 0.78 (0.40–1.51) 
 BMI 3 kg/m2 −0.36 0.16 0.03 0.70 (0.51–0.96) 
 Systolic blood pressure 20 mmHg 0.00 0.17 0.99 1.00 (0.72–1.39) 
Approximately 1-SD changeRegression coefficientSEP valueOdds ratio (95% CI)
Diabetes      
 Intercept  −10.03 4.55 0.027  
 Log(CRP) 0.5 log(mg/l) 0.93 0.38 0.014 2.53 (1.21–5.29) 
 Log(triglyceride) 0.3 log(mmol/l) −0.003 0.41 0.99 0.99 (0.45–2.24) 
 Fasting glucose 0.5 mmol/l 0.04 0.23 0.84 1.05 (0.67–1.63) 
 2-h glucose 1 mmol/l 0.88 0.35 0.010 2.42 (1.23–4.76) 
 Age 10 years −0.25 0.26 0.33 0.78 (0.47–1.29) 
 Sex  1.53 0.62 0.023 4.59 (1.38–15.32) 
 BMI 3 kg/m2 −0.11 0.29 0.71 0.90 (0.51–1.57) 
 Systolic blood pressure 20 mmHg 0.12 0.29 0.69 1.13 (0.64–1.99) 
NGT      
 Intercept  9.54 2.58 0  
 Log(CRP) 0.5 log(mg/l) −0.18 0.17 0.30 0.83 (0.59–1.17) 
 Log(triglyceride) 0.3 log(mmol/l) −0.05 0.25 0.84 0.95 (0.59–1.54) 
 Fasting glucose 0.5 mmol/l −0.5 0.15 0.76 0.96 (0.72–1.28) 
 2-h glucose 1 mmol/l −0.45 0.20 0.02 0.64 (0.43–0.94) 
 Age 10 years −0.30 0.15 0.04 0.74 (0.55–0.99) 
 Sex  −0.25 0.34 0.46 0.78 (0.40–1.51) 
 BMI 3 kg/m2 −0.36 0.16 0.03 0.70 (0.51–0.96) 
 Systolic blood pressure 20 mmHg 0.00 0.17 0.99 1.00 (0.72–1.39) 

The given odds ratio for continuous covariates is for an approximate 1-SD change.

View Large

A table elsewhere in this issue shows conventional and Système International (SI) units and conversion factors for many substances.

This study was funded by grants from the Hong Kong Research Grant Council (HKU7290/97M) and Health Services Research Committee (HSRC no. 631011).

We thank Grace Cheung and Tena Li for their assistance with data collection and entry.

1
Pradhan AD, Manson JE, Rifai N, Buring JE, Ridker PM: C-reactive protein, interleukin 6, and risk of developing type 2 diabetes mellitus.
JAMA
 
286
:
327
–334,
2001
2
Barzilay JI, Abraham L, Heckbert SR, Cushman M, Kuller LH, Resnick HE, Tracy RP: The relation of markers of inflammation to the development of glucose disorders in the elderly: the Cardiovascular Health Study.
Diabetes
 
50
:
2384
–2389,
2001
3
Freeman DJ, Norrie J, Caslake MJ, Gaw A, Ford I, Lowe GD, O’Reilly DS, Packard CJ, Sattar N: C-reactive protein is an independent predictor of risk for the development of diabetes in the West of Scotland Coronary Prevention Study.
Diabetes
 
51
:
1596
–1600,
2002
4
Festa A, D’Agostino R Jr, Tracy RP, Haffner SM: Elevated levels of acute-phase proteins and plasminogen activator inhibitor-1 predict the development of type 2 diabetes: the Insulin Resistance Atherosclerosis Study.
Diabetes
 
51
:
1131
–1137,
2002
5
Thorand B, Lowel H, Schneider A, Kolb H, Meisinger C, Frohlich M, Koenig W: C-reactive protein as a predictor for incident diabetes mellitus among middle-aged men: results from the MONICA Augsburg cohort study, 1984–1998.
Arch Intern Med
 
163
:
93
–99,
2003
6
Frohlich M, Imhof A, Berg G, Hutchinson WL, Pepys MB, Boeing H, Muche R, Brenner H, Koenig W: Association between C-reactive protein and features of the metabolic syndrome: a population-based study.
Diabetes Care
 
23
:
1835
–1839,
2000
7
Festa A, D’Agostino R Jr, Howard G, Mykkanen L, Tracy RP, Haffner SM: Chronic subclinical inflammation as part of the insulin resistance syndrome: the Insulin Resistance Atherosclerosis Study (IRAS).
Circulation
 
102
:
42
–47,
2000
8
Han TS, Sattar N, Williams K, Gonzalez-Villalpando C, Lean ME, Haffner SM: Prospective study of C-reactive protein in relation to the development of diabetes and metabolic syndrome in the Mexico City Diabetes Study.
Diabetes Care
 
25
:
2016
–2021,
2002
9
McMillan DE: Increased levels of acute-phase serum proteins in diabetes.
Metabolism
 
38
:
1042
–1046,
1989
10
Ford ES: Body mass index, diabetes, and C-reactive protein among U.S. adults.
Diabetes Care
 
22
:
1971
–1977,
1999
11
Temelkova-Kurktschiev T, Henkel E, Koehler C, Karrei K, Hanefeld M: Subclinical inflammation in newly detected type II diabetes and impaired glucose tolerance (Letter).
Diabetologia
 
45
:
151
,
2002
12
Nakanishi N, Suzuki K, Tatara K: White blood cell count and clustered features of metabolic syndrome in Japanese male office workers.
Occup Med (Lond)
 
52
:
213
–218,
2002
13
Nakanishi N, Sato M, Shirai K, Nakajima K, Murakami S, Takatorige T, Suzuki K, Tatara K: Associations between white blood cell count and features of the metabolic syndrome in Japanese male office workers.
Ind Health
 
40
:
273
–277,
2002
14
Janus ED, Watt NM, Lam KS, Cockram CS, Siu ST, Liu LJ, Lam TH: The prevalence of diabetes, association with cardiovascular risk factors and implications of diagnostic criteria (ADA 1997 and WHO 1998) in a 1996 community-based population study in Hong Kong Chinese.
Diabet Med
 
17
:
741
–745,
2000
15
Wat NM, Lam TH, Janus ED, Lam KS: Central obesity predicts the worsening of glycemia in southern Chinese.
Int J Obes Relat Metab Disord
 
25
:
1789
–1793,
2001
16
World Health Organization:
Definition, Diagnosis and Classification of Diabetes Mellitus and its Complications: Report of a WHO Consultation. Part 1: Diagnosis and Classification of Diabetes Mellitus
. Geneva, World Health Org.,
1999
17
Harris MI: Impaired glucose tolerance - prevalence and conversion to NIDDM.
Diabet Med
 
13 (Suppl. 2)
:
S9
–S11,
1996
18
Edelstein SL, Knowler WC, Bain RP, Andres R, Barrett-Connor EL, Dowse GK, Haffner SM, Pettitt DJ, Sorkin JD, Muller DC, Collins VR, Hamman RF: Predictors of progression from impaired glucose tolerance to NIDDM: an analysis of six prospective studies.
Diabetes
 
46
:
701
–710,
1997
19
Tuomilehto J, Lindstrom J, Eriksson JG, Valle TT, Hamalainen H, Ilanne-Parikka P, Keinanen-Kiukaanniemi S, Laakso M, Louheranta A, Rastas M, Salminen V, Uusitupa M: Prevention of type 2 diabetes mellitus by changes in lifestyle among subjects with impaired glucose tolerance.
N Engl J Med
 
344
:
1343
–1350,
2001
20
Pan XR, Li GW, Hu YH, Wang JX, Yang WY, An ZX, Hu ZX, Lin J, Xiao JZ, Cao HB, Liu PA, Jiang XG, Jiang YY, Wang JP, Zheng H, Zhang H, Bennett PH, Howard BV: Effects of diet and exercise in preventing NIDDM in people with impaired glucose tolerance: the Da Qing IGT and Diabetes Study.
Diabetes Care
 
20
:
537
–544,
1997
21
Chiasson JL, Josse RG, Gomis R, Hanefeld M, Karasik A, Laakso M: Acarbose for prevention of type 2 diabetes mellitus: the STOP-NIDDM randomised trial.
Lancet
 
359
:
2072
–2077,
2002
22
Motala AA, Omar MA, Gouws E: Transient impaired glucose tolerance in South African Indians does not carry a risk for progression to NIDDM.
Diabetes Care
 
20
:
1101
–1107,
1997
23
Knowler WC, Barrett-Connor E, Fowler SE, Hamman RF, Lachin JM, Walker EA, Nathan DM: Reduction in the incidence of type 2 diabetes with lifestyle intervention or metformin.
N Engl J Med
 
346
:
393
–403,
2002
DIABETES CARE
2003