SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES

Objective

To identify factors that regulate glucose metabolism in rheumatoid arthritis (RA).

Methods

We evaluated the homeostatic model assessment of insulin resistance (HOMA-IR) and beta cell function (HOMA-B) in 94 RA patients. We investigated the relationship between characteristics known to affect glucose metabolism in the general population (age, abdominal obesity [waist circumference], hypertension, antihypertensive therapy) as well as characteristics of RA (disease activity, glucocorticoid therapy) and insulin resistance and beta cell function.

Results

Patients with high-grade inflammation (high-sensitivity C-reactive protein value >1.92 mg/liter) (n = 81) were more insulin resistant than patients with low-grade inflammation (n = 13), whereas beta cell function was similar in both groups. Insulin resistance and beta cell function were similar in both groups after adjustment for waist circumference. All recorded characteristics except for age were associated with HOMA-IR or/and HOMA-B in univariate analyses. In mixed regression models, abdominal obesity and patient's assessment of disease activity (by visual analog scale) were predictors of insulin resistance. The Disease Activity Score assessed using 28-joint counts for swelling and tenderness, tender joint count, and patient's assessment of disease activity were associated with reduced beta cell function, and the cumulative dose of glucocorticoids was associated with enhanced beta cell function. The cumulative glucocorticoid dose in all study patients was a mean of only 536 mg (95% confidence interval 239–1,173). In patients with high-grade inflammation, age was further associated with impaired beta cell function, whereas use of angiotensin-converting enzyme inhibitors or angiotensin II type 1 receptor blockers was associated with enhanced beta cell function.

Conclusion

The modifiable factors of abdominal obesity, antihypertensive therapy, disease activity, and use of glucocorticoids appear to affect glucose metabolism in RA.

Individuals with rheumatoid arthritis (RA) experience a more than 2-fold increased risk of myocardial infarction (1). Both traditional and nontraditional risk factors contribute to atherogenesis in RA (2, 3). Among these risk factors is insulin resistance, which is reported to be associated with inflammation and glucocorticoid therapy in RA (2, 4–6). In addition, RA patients experience a diabetes prevalence of 10–20% (7–9), and diabetes predicts increased rates of cardiovascular events in this disease (8, 9).

In the general population, insulin resistance is an established risk factor for cardiovascular disease and type 2 diabetes mellitus (DM) (10) and constitutes a central mechanism in the metabolic syndrome, wherein it clusters particularly with dyslipidemia and impaired glucose metabolism (11). In the general population, the main determinant of insulin resistance is abdominal obesity (11). Other contributing factors include age and hypertension (12, 13). In our experience, 50% of RA patients have hypertension (4). Furthermore, among the antihypertensive agents, β-blockers and diuretics have been reported to impair insulin sensitivity and to have the propensity to provoke the onset of type 2 DM, whereas angiotensin-converting enzyme (ACE) inhibitors and angiotensin II type 1 receptor blockers (ARBs) have favorable effects on glucose metabolism (14–16).

In normal subjects, decreased insulin sensitivity results in increased insulin secretion by pancreatic beta cells (12, 17). In contrast, in individuals predisposed to type 2 DM, this ability to compensate for insulin resistance is impaired (12, 17). Although type 2 DM is often (but not always) preceded by insulin resistance, impaired insulin secretion plays a more pivotal role than insulin resistance in the progression from normal glucose tolerance to type 2 DM (17).

In the Insulin Resistance Atherosclerosis Study, age and abdominal fat content were found to be main determinants of the presence of impaired insulin secretion (12). Enlarged fat cells release excess nonesterified fatty acids and adipokines that, in the long term, impair beta cell function (17). This process has been referred to as lipotoxicity (17–19). To our knowledge, insulin secretion has not been studied in RA. High-grade inflammation and glucocorticoid therapy, both of which are relevant in the context of RA, have the potential to adversely affect beta cell function (20–25).

In this study, we determined the relative contribution of characteristics known to affect glucose metabolism in the general population (age, abdominal obesity, hypertension, and antihypertensive agents) as well as characteristics of RA (disease activity and glucocorticoid therapy) to impaired insulin sensitivity and beta cell function in RA.

PATIENTS AND METHODS

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES

Patients.

We evaluated 94 white patients who met the American College of Rheumatology (formerly, the American Rheumatism Association) 1987 criteria for RA (26). Their demographic characteristics are shown in Table 1. We excluded patients in whom impaired glucose metabolism was diagnosed prior to the onset of RA and patients taking lipid-lowering or glucose-lowering agents. Included were patients with impaired fasting glucose or type 2 DM that was treated with dietary intervention only. Patients with hypothyroid were also excluded since we have recently identified an independent association between hypothyroidism and markers of insulin sensitivity in RA (27). The study was approved by the Ethics Committee for Research on Human Subjects (Medical) of the University of the Witwatersrand.

Table 1. Characteristics in all patients and in patients with low-grade and high-grade inflammation*
 All patients (n = 94)Patients with low-grade inflammation (n = 13)Patients with high-grade inflammation (n = 81)PAdjusted P
  • *

    Except where indicated otherwise, values are the mean (95% confidence interval). P values compare patients with low-grade inflammation (high-sensitivity C-reactive protein [hsCRP] ≤1.92 mg/liter) and high-grade inflammation (hsCRP >1.92 mg/liter) by Student's t-test and Fisher's exact test. Adjusted P values were adjusted for waist circumference in logistic regression analyses. BMI = body mass index; BP = blood pressure; ACE = angiotensin-converting enzyme; ARB = angiotensin II type 1 receptor blocker; HDL = high-density lipoprotein; LDL = low-density lipoprotein; HOMA-IR = homeostatic model assessment of insulin resistance; HOMA-B = homeostatic model assessment of beta cell function; ESR = erythrocyte sedimentation rate; VAS = visual analog scale (0–100-mm); DAS28 = Disease Activity Score assessed using 28-joint counts for swelling and tenderness; DMARD = disease-modifying antirheumatic drug; GCs = glucocorticoids.

  • Logarithmically transformed. Values are the geometric mean (95% confidence interval).

Demographics     
 Age, years55 (53–57)53 (47–58)55 (53–58)0.4
 Women, no. (%)62 (87)11 (85)71 (88)0.5
Disease duration, years5.8 (4.4–7.8)6.5 (3.4–12.0)5.8 (4.2–7.9)0.7
Rheumatoid factor positive, no. (%)71 (76)7 (54)64 (79)0.1
Metabolic features     
 BMI, kg/m225.5 (24.8–26.6)23.4 (21.4–25.3)25.9 (24.7–27.0)0.03
 Waist circumference, cm88 (85–90)81 (75–86)89 (86–92)0.01
 Systolic BP, mm Hg131 (127–134)127 (120–133)131 (127–135)0.30.9
 Diastolic BP, mm Hg84 (82–86)82 (75–89)84 (82–86)0.50.9
 Hypertension, no. (%)48 (51)5 (38)43 (53)0.30.8
 Antihypertensive agent use, no. (%)31 (33)3 (23)28 (35)0.10.9
 Diuretic and/or β-blocker use, no. (%)20 (21)2 (15)18 (22)0.41.0
 ACE inhibitor or ARB use, no. (%)25 (27)3 (23)22 (27)0.50.7
 Triglycerides, mmoles/liter1.1 (1.0–1.2)0.9 (0.7–1.2)1.2 (1.1–1.3)0.070.3
 HDL cholesterol, mmoles/liter1.6 (1.5–1.7)1.8 (1.6–2.0)1.5 (1.4–1.6)0.030.3
 Ratio of triglycerides to HDL cholesterol0.8 (0.7–0.9)0.5 (0.4–0.7)0.8 (0.7–0.9)0.010.07
 Ratio of total cholesterol to HDL cholesterol3.7 (3.2–4.2)3.0 (2.5–3.4)3.8 (3.2–4.4)0.020.3
 LDL cholesterol, mmoles/liter2.9 (2.8–3.1)3.0 (2.6–3.4)2.9 (2.7–3.1)0.60.3
 Impaired fasting glucose, no. (%)4 (4)0 (0)4 (5)0.51.0
 Diabetes, no. (%)4 (4)0 (0)4 (5)0.51.0
 Impaired fasting glucose or diabetes, no. (%)8 (8)0 (0)8 (10)0.31.0
 HOMA-IR1.25 (1.08–1.46)0.84 (0.67–0.96)1.34 (1.13–1.58)0.0020.2
 HOMA-B157 (134–185)151 (101–228)158 (132–189)0.80.8
Disease activity     
 hsCRP, mg/liter8.6 (6.4–11.8)0.7 (0.5–1.0)12.9 (10.0–16.6)<0.0001
 ESR, mm/hour15 (12–19)5 (3–8)18 (14–23)0.002
 No. of swollen joints (28 assessed)3 (2–5)1 (1–3)4 (3–5)0.0002
 No. of tender joints (28 assessed)4 (3–5)1 (1–3)5 (3–6)0.0009
 Patient's assessment of disease activity (by VAS), mm48 (41–55)23 (10–36)52 (45–59)0.0007
 DAS284.3 (3.9–4.7)2.1 (1.4–2.8)4.7 (4.2–5.1)<0.0001
Treatment     
 DMARD use, no. (%)55 (59)9 (69)46 (57)0.3
 Cumulative dose of GCs, mg536 (239–1,173)850 (165–4,466)489 (203–1,201)0.6
 Ever use of GCs, no. (%)72 (77)12 (92)60 (74)0.1
 Current use of prednisone, no. (%)13 (14)0 (0)13 (16)0.1
 Current prednisone dosage, mg/day0.2 (0.1–0.4)0 (0–0)0.3 (0.1–0.5)0.4

Outcome characteristics.

The outcome characteristics were the homeostatic model assessment of insulin resistance (HOMA-IR) ([insulin (mU/liter) × glucose (mmoles/liter)] ÷ 22.5) and the homeostatic model assessment of beta cell function (HOMA-B) ([20 × insulin (mU/liter)] ÷ [glucose (mmoles/liter) − 3.5]) (28, 29). Laboratory tests were performed on blood samples obtained from patients who were fasting; samples were taken between 8:00 AM and 10:00 AM. Details on methods for the evaluation of circulating plasma glucose and serum insulin concentrations in the laboratory that performs these assessments for us (Abbott, Chicago, IL) have been reported elsewhere (7, 29).

Other recorded characteristics.

Using previously reported methods (7, 30, 31), we recorded metabolic features, disease activity, and features of disease-modifying antirheumatic drug (DMARD) and glucocorticoid therapy (see Table 1). Disease activity variables comprised the high-sensitivity C-reactive protein (hsCRP) level, the erythrocyte sedimentation rate (ESR), the swollen joint count (in 28 joints), the tender joint count (in 28 joints), and the patient's assessment of disease activity (measured on a 0–100-mm visual analog scale [VAS]) (32). In addition, we calculated the Disease Activity Score assessed using 28-joint counts for swelling and tenderness (DAS28) using the following formula (32): DAS28 score = (0.56 × equation image + (0.28 × equation image + (0.70 × ln ESR) + (0.014 × VAS score).

The hsCRP level was measured using an immunoturbidimetric assay performed with Olympus OSR6185 (Olympus Diagnostics, Lismeehan, County Clare, Ireland). Impaired fasting glucose and diabetes were diagnosed in accordance with the recent recommendations as made by the American Diabetes Association (33). Lipid levels were determined by automated enzymatic methods, with reagents supplied by Olympus Diagnostics for measurement of total cholesterol, high-density lipoprotein (HDL) cholesterol, and triglycerides, and with reagents supplied by Randox Laboratories (Crumlin, UK) for measurement of low-density lipoprotein (LDL) cholesterol. Information on previous use of glucocorticoids was obtained from review of the patients' medical records. The cumulative dose of glucocorticoids was calculated as the cumulative oral prednisone dose plus the cumulative pulse prednisone equivalent dose administered intraarticularly, intramuscularly, or/and intravenously.

Statistical analysis.

Results were expressed as the mean with 95% confidence intervals (95% CIs) or the number with the percentage. Individual characteristics showing an abnormal distribution were logarithmically transformed. For these characteristics, the geometric means and 95% CIs are given.

We recently reported cardiovascular risk factors, including hsCRP concentrations, in 80 subjects that formed the control group in an investigation of endothelial function in RA (31). These subjects had no known diseases, impaired fasting glucose, or diabetes and were not taking any medications. The mean hsCRP level was 1.58 mg/liter (95% CI 1.30–1.92) in this cohort. Accordingly, for the present study, we considered the presence of an hsCRP level >1.92 mg/liter to reflect high-grade inflammation. We compared the recorded variables in patients with low-grade inflammation with those in patients with high-grade inflammation using Student's t-test (continuous variables) or Fisher's exact test (dichotomous variables). Comparisons of metabolic syndrome features in patients with high-grade inflammation versus patients with low-grade inflammation were further made in logistic regression models, with adjustment for the potentially confounding variable of waist circumference.

We then assessed the associations of age, waist circumference, disease activity, glucocorticoid therapy, hypertension, and the use of antihypertensive agents with insulin resistance and beta cell function in univariate analysis, using simple linear regression analysis or Student's t-test as appropriate. All characteristics studied, except for age, were associated with HOMA-IR or/and HOMA-B. Age is a known risk factor for impaired glucose metabolism (12), and therefore, all characteristics were entered as independent variables in mixed regression models for HOMA-IR and HOMA-B. In univariate analyses, the cumulative glucocorticoid dose constituted the glucocorticoid therapy variable that was most strongly associated with HOMA-B and this variable was therefore used as the glucocorticoid characteristic in the mixed regression models. The disease activity variables were all strongly interrelated and were therefore entered in separate mixed regression models (models 1–6 as described below).

As an estimate of the effect of independent variables that were significantly associated with insulin resistance or/and beta cell function in the multivariable models, we calculated the percentage change in R2 (i.e., the difference between the R2 from models with and without the variable of interest divided by the R2 of the model containing the variable) (12).

RESULTS

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES

Recorded variables in all patients and in patients with low-grade and high-grade inflammation.

The descriptive statistics in all patients are shown in Table 1. Thirty-one patients (33%) were taking ≥1 antihypertensive agent. These comprised a diuretic in 19 patients (20%), a β-blocker in 3 (3%), an ACE inhibitor in 12 (13%), an ARB in 13 (14%), and a calcium-channel blocker in 8 (9%). No other antihypertensive agents were used, and no patient was taking an ACE inhibitor and an ARB simultaneously. Treatment with DMARDs alone was used in 55 patients (59%). This was because 40 of these patients (43%) were seen for the first time at our clinic at the time of the study. The different DMARDs used were methotrexate in 47 patients (50%), chloroquine in 23 (25%), minocycline in 12 (13%), leflunomide in 5 (5%), sulfasalazine in 5 (5%), and azathioprine in 5 (5%). None of the patients were being treated with biologic agents at the time of the study.

As compared with patients with low-grade inflammation (hsCRP level ≤1.92 mg/liter), patients with high-grade inflammation (hsCRP level >1.92 mg/liter) had a higher body mass index and waist circumference, lower HDL cholesterol level, higher ratios of triglycerides to HDL cholesterol and of total cholesterol to HDL cholesterol, and a higher frequency of impaired fasting glucose or diabetes (Table 1). In addition, patients with high-grade inflammation were more insulin resistant, whereas insulin secretion did not differ significantly between the groups. This confirms the presence of impaired beta cell function (12, 17, 34) in the context of RA patients with high-grade inflammation as compared with RA patients with low-grade inflammation. The hsCRP level and waist circumference were significantly interrelated (R = 0.324, P = 0.0005). After adjustment for waist circumference, the HOMA-IR and other metabolic syndrome features were no longer significantly different between patients with low-grade inflammation and patients with high-grade inflammation.

The DMARD and glucocorticoid therapy variables did not differ significantly between patients with low-grade inflammation and patients with high-grade inflammation.

Univariate analysis of insulin resistance and beta cell function.

Tables 2 and 3 show the results of univariate analyses of insulin resistance and beta cell function. HOMA-IR was associated positively with waist circumference, hsCRP level, DAS28, ESR, patient's assessment of disease activity, presence of hypertension, and use of diuretics and/or β-blockers and ACE inhibitors or ARBs. HOMA-B was inversely associated with the DAS28, swollen joint count, tender joint count, and patient's assessment of disease activity, but positively associated with the cumulative glucocorticoid dose.

Table 2. Correlations between patient characteristics and HOMA-IR and HOMA-B*
 HOMA-IRHOMA-B
RPRP
  • *

    HOMA-IR = homeostatic model assessment of insulin resistance; HOMA-B = homeostatic model assessment of beta cell function; hsCRP = high-sensitivity C-reactive protein; DAS28 = Disease Activity Score assessed using 28-joint counts for swelling and tenderness; ESR = erythrocyte sedimentation rate; GCs = glucocorticoids.

  • Logarithmically transformed.

Age0.0780.5−0.0350.7
Waist circumference0.668<0.00010.1920.06
hsCRP0.2980.0040.0340.74
DAS280.2080.04−0.2630.01
ESR0.2410.02−0.0530.6
No. of swollen joints (28 assessed)0.0910.4−0.2290.03
No. of tender joints (28 assessed)0.1240.2−0.3040.003
Patient's assessment of disease activity0.2500.02−0.2890.005
Cumulative dose of GCs0.0080.90.2750.007
Current prednisone dosage0.0990.30.0100.9
Table 3. HOMA-IR and HOMA-B values according to the presence of hypertension, the use of antihypertensive agents, and the use of glucocorticoids*
 No. of patientsHOMA-IRPHOMA-BP
  • *

    Values are the geometric mean (95% confidence interval). HOMA-IR = homeostatic model assessment of insulin resistance; HOMA-B = homeostatic model assessment of beta cell function; ACE = angiotensin-converting enzyme; ARB = angiotensin II type 1 receptor blocker; GCs = glucocorticoids.

Hypertension     
  Yes481.66 (1.34–2.05)0.0001183 (146–230)0.06
  No460.94 (0.78–0.89)134 (107–168)
Diuretic and/or β-blocker use     
  Yes201.14 (0.96–1.34)0.01164 (112–242)0.8
  No741.81 (1.32–2.47)155 (129–186)
ACE inhibitor or ARB use     
  Yes251.86 (1.33–2.59)0.005200 (138–289)0.1
  No691.09 (0.92–1.27)144 (121–172)
Ever use of GCs     
  Yes721.23 (1.03–1.47)0.6171 (142–207)0.05
  No221.33 (0.98–1.80)119 (88–161)
Current prednisone use     
  Yes131.38 (0.84–2.28)1.0170 (92–313)0.8
  No811.23 (1.05–1.45)155 (132–184)

The different disease activity variables were all strongly interrelated (R = 0.36–0.88, P < 0.0004) (data not shown).

Mixed regression models for insulin resistance.

In the mixed regression models for HOMA-IR, abdominal obesity was the main determinant of insulin resistance, accounting for 39–56% of the model R2 for HOMA-IR (Table 4). Patient's assessment of disease activity was also associated with insulin resistance, accounting for 5% of the model R2 for HOMA-IR.

Table 4. Mixed regression models for HOMA-IR in 94 rheumatoid arthritis patients*
 Model 1 (R2 = 50)Model 2 (R2 = 50)Model 3 (R2 = 50)Model 4 (R2 = 49)Model 5 (R2 = 50)Model 6 (R2 = 51)
β ± SEP%Δ R2β ± SEP%Δ R2β ± SEP%Δ R2β ± SEP%Δ R2β ± SEP%Δ R2β ± SEP%Δ R2
  • *

    HOMA-IR = homeostatic model assessment of insulin resistance; β = regression coefficient; %Δ = percentage change; ACE = angiotensin-converting enzyme; ARB = angiotensin II type 1 receptor blocker; GCs = glucocorticoids; hsCRP = high-sensitivity C-reactive protein; DAS28 = Disease Activity Score assessed using 28-joint counts for swelling and tenderness; ESR = erythrocyte sedimentation rate.

  • Logarithmically transformed.

Age−0.0042 ± 0.00260.1−0.0040 ± 0.00250.1−0.0041 ± 0.00260.1−0.0035 ± 0.00250.2−0.0037 ± 0.00250.2−0.0041 ± 0.00250.1
Waist circumference0.0138 ± 0.0021<0.0001520.0143 ± 0.0021<0.0001560.0141 ± 0.0021<0.0001470.0147 ± 0.0021<0.0001390.0147 ± 0.0021<0.0001400.0141 ± 0.0021<0.000147
Hypertension0.0980 ± 0.06210.10.0942 ± 0.06220.10.0999 ± 0.06210.10.0989 ± 0.06290.10.0911 ± 0.06260.20.0939 ± 0.06150.1
Diuretic and/or β-blocker use0.0036 ± 0.07161.00.0010 ± 0.07181.00.0156 ± 0.07160.80.0081 ± 0.07290.9−0.0053 ± 0.02760.9−0.0053 ± 0.07110.9
ACE inhibitor or ARB use0.0790 ± 0.07170.30.0788 ± 0.07170.30.0637 ± 0.07120.40.0673 ± 0.07230.90.0830 ± 0.07250.30.0868 ± 0.07120.2
Cumulative dose of GCs0.0070 ± 0.01510.60.0126 ± 0.01570.40.0067 ± 0.01510.70.0076 ± 0.01650.70.0139 ± 0.01620.40.0128 ± 0.01530.4
hsCRP0.0601 ± 0.04040.1               
DAS28   0.0183 ± 0.01230.1            
ESR      0.0719 ± 0.04980.2         
No. of swollen joints         0.0119 ± 0.05880.8      
No. of tender joints            0.0713 ± 0.05460.2   
Patient's assessment of disease activity               0.0156 ± 0.00770.0465

Mixed regression models for beta cell function.

In all patients, the mixed regression models for HOMA-B revealed that the cumulative glucocorticoid dose was associated with enhanced beta cell function, accounting for 11–55% of the model R2 for HOMA-B (Table 5). Three of the disease activity variables (DAS28, tender joint count, and patient's assessment of disease activity) were inversely associated with the HOMA-B. These variables accounted for 19–25% of the model R2 for HOMA-B.

Table 5. Mixed regression models for HOMA-B in 94 rheumatoid arthritis patients*
 Model 1 (R2 = 20)Model 2 (R2 = 23)Model 3 (R2 = 20)Model 4 (R2 = 21)Model 5 (R2 = 23)Model 6 (R2 = 25)
β ± SEP%Δ R2β ± SEP% Δ R2β ± SEP%Δ R2β ± SEP%Δ R2β ± SEP%Δ R2β ± SEP%Δ R2
  • *

    HOMA-B = homeostatic model assessment of beta cell function; β = regression coefficient; %Δ = percentage change; ACE = angiotensin-converting enzyme; ARB = angiotensin II type 1 receptor blocker; GCs = glucocorticoids; hsCRP = high-sensitivity C-reactive protein; DAS28 = Disease Activity Score assessed using 28-joint counts for swelling and tenderness; ESR = erythrocyte sedimentation rate.

  • Logarithmically transformed.

Age−0.0062 ± 0.00350.08−0.0057 ± 0.00340.09−0.0064 ± 0.00350.07−0.0064 ± 0.00340.06−0.0063 ± 0.00330.07−0.0057 ± 0.00330.09
Waist circumference0.0042 ± 0.00280.10.0046 ± 0.00280.10.0043 ± 0.00290.10.0042 ± 0.00280.10.0039 ± 0.00270.20.0049 ± 0.00270.08
Hypertension0.1030 ± 0.08460.20.1186 ± 0.08270.20.1079 ± 0.08450.20.1132 ± 0.08390.20.1271 ± 0.08300.10.1180 ± 0.08120.2
Diuretic and/or β-blocker use−0.1143 ± 0.09710.2−0.0941 ± 0.09550.3−0.1204 ± 0.09740.2−0.0973 ± 0.09730.3−0.0814 ± 0.09630.4−0.0892 ± 0.09440.4
ACE inhibitor or ARB use0.1454 ± 0.09750.10.1320 ± 0.09530.20.1593 ± 0.09690.10.1445 ± 0.09650.10.1190 ± 0.09620.20.1210 ± 0.09450.2
Cumulative dose of GCs0.0682 ± 0.02060.001550.0548 ± 0.02080.01260.0668 ± 0.02060.002500.0550 ± 0.02200.01270.0502 ± 0.02150.02110.0558 ± 0.02030.00726
hsCRP−0.0009 ± 0.00090.3               
DAS28   −0.0362 ± 0.01630.0319            
ESR      −0.0592 ± 0.06780.4         
No. of swollen joints         −0.1161 ± 0.07840.1      
No. of tender joints            −0.1606 ± 0.07240.0319   
Patient's assessment of disease activity               −0.0273 ± 0.01030.00925

Thirteen patients in this cohort (14%) had low-grade inflammation (hsCRP level ≤1.92 mg/liter) (Table 1). In the general population, low-grade inflammation is associated with insulin resistance but not with beta cell function (35). Therefore, we repeated the regression models for beta cell function after exclusion of patients with low-grade inflammation (Table 6). This did not materially alter the associations between the cumulative glucocorticoid dose as well as the disease activity variables and beta cell function. However, the model R2 increased from a range of 20–25 (Table 5) to a range of 26–33 (Table 6). Moreover, in patients with high-grade inflammation, age was associated with decreased beta cell function, accounting for 13–23% of the model R2 for HOMA-B, whereas the use of ACE inhibitors and/or ARBs was associated with enhanced beta cell function in 3 of the 6 regression models, accounting for 13–18% of the model R2 for HOMA-B in the respective models.

Table 6. Mixed regression models for HOMA-B in 81 rheumatoid arthritis patients with high-grade inflammation*
 Model 1 (R2 = 26)Model 2 (R2 = 30)Model 3 (R2 = 26)Model 4 (R2 = 30)Model 5 (R2 = 32)Model 6 (R2 = 33)
β ± SEP%Δ R2β ± SEP%Δ R2β ± SEP%Δ R2β ± SEP%Δ R2β ± SEP%Δ R2β ± SEP%Δ R2
  • *

    High-grade inflammation was defined as a high-sensitivity C-reactive protein (hsCRP) value >1.92 mg/liter. HOMA-B = homeostatic model assessment of beta cell function; β = regression coefficient; %Δ = percentage change; ACE = angiotensin-converting enzyme; ARB = angiotensin II type 1 receptor blocker; GCs = glucocorticoids; DAS28 = Disease Activity Score assessed using 28-joint counts for swelling and tenderness; ESR = erythrocyte sedimentation rate.

  • Logarithmically transformed.

Age−0.0090 ± 0.00360.0223−0.0076 ± 0.00350.0315−0.0086 ± 0.00360.0222−0.0084 ± 0.00350.0220−0.0081 ± 0.00340.0217−0.0072 ± 0.00340.0413
Waist circumference0.0037 ± 0.00300.20.0041 ± 0.00280.20.0040 ± 0.00290.20.0039 ± 0.00280.20.0035 ± 0.00280.20.0045 ± 0.00280.1
Hypertension0.1545 ± 0.08920.090.1604 ± 0.08640.070.1529 ± 0.08940.090.1643 ± 0.08730.060.1727 ± 0.08590.050.1609 ± 0.08530.06
Diuretic and/or β-blocker use−0.1294 ± 0.10050.2−0.1056 ± 0.09750.3−0.1298 ± 0.10050.2−0.1050 ± 0.09860.3−0.0879 ± 0.09740.4−0.0977 ± 0.09640.3
ACE inhibitor or ARB use0.2145 ± 0.10030.04180.1882 ± 0.09710.060.2136 ± 0.09990.04180.1964 ± 0.09780.048130.1696 ± 0.09720.090.1749 ± 0.09630.07
Cumulative dose of GCs0.0704 ± 0.02120.001420.0587 ± 0.02120.007230.0706 ± 0.02120.001430.0558 ± 0.02210.01210.0523 ± 0.02160.02170.0591 ± 0.02070.00623
hsCRP0.0188 ± 0.07330.8                
DAS28   −0.0397 ± 0.01770.0316            
ESR      −0.0239 ± 0.07670.8         
No. of swollen joints         −0.1587 ± 0.08330.06      
No. of tender joints            −0.1933 ± 0.07640.0119   
Patient's assessment of disease activity               −0.0287 ± 0.01080.0120

Although the hsCRP value was associated with waist circumference in this cohort (see above), after omission of waist circumference in the models shown in Tables 5 and 6, the association between the hsCRP value and HOMA-B remained not statistically significant (data not shown).

DISCUSSION

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES

In the present study, patients with high-grade inflammation had increased insulin resistance and reduced beta cell function as compared with patients with low-grade inflammation. Furthermore, patients with high-grade inflammation also had lower HDL cholesterol and higher ratios of triglycerides to HDL cholesterol and total cholesterol to HDL cholesterol, characteristics that form part of the metabolic/insulin resistance syndrome (10, 11, 36). Also, all 8 patients with impaired fasting glucose or diabetes had high-grade inflammation, although this association was not statistically significant. Impaired fasting glucose and diabetes are clinical hallmarks of impaired beta cell function (17). An unanticipated finding, however, was the higher body mass index and waist circumference in RA patients with high-grade inflammation, whereas in the general population, obesity contributes to both insulin resistance and reduced beta cell function (12). Adjustment for waist circumference in the logistic regression models indicated that abdominal obesity may explain differences in lipids, insulin resistance, and beta cell function between patients with high-grade inflammation and patients with low-grade inflammation.

In population studies, adiposity is closely correlated with circulating levels of cytokines and accounts for as much as 30% of the (low-grade) systemic inflammatory burden (2). The association of waist circumference with hsCRP concentrations in the present cohort therefore suggests that enlarged abdominal fat cell–derived adipokines, such as interleukin-6 (in addition to inflamed joint derived interleukin-6), also contribute to CRP concentrations in RA. We recently observed an increased waist circumference in RA patients as compared with healthy controls (P = 0.01), whereas the body mass index was similar in both groups (P = 0.1) (31). Westhovens et al (37) previously reported a decreased fat content in the legs, an increased abdominal fat content, and a generalized reduction in lean body mass in patients with RA. These characteristics form part of the so-called “rheumatoid cachexia” that affects more than 50% of RA patients and is related to cytokine production and use of glucocorticoids (37, 38). The contribution of abdominal obesity to CRP concentrations in RA may complicate the interpretation of this acute-phase response when used as a disease activity marker in these patients.

We found that insulin resistance was associated with waist circumference, hsCRP level, DAS28, ESR, patient's assessment of disease activity, presence of hypertension, and use of antihypertensive agents (a marker of hypertension in univariate analysis). Such associations between acute-phase responses and insulin resistance are well documented in RA (4, 5). In mixed regression models, waist circumference was the main contributor to insulin resistance, and among the RA characteristics, only the patient's assessment of disease activity made a significant, but small (5%), contribution to the model R2 of HOMA-IR. Previous studies of insulin resistance in RA did not report on the relative role of excess weight.

Obesity is the main determinant of insulin resistance in the general population (10–12). Our findings therefore suggest that the interventions that are increasingly recommended for the treatment of insulin resistance in the general population may be equally important in RA patients with insulin resistance (39–41). However, the present investigation was of a cross-sectional design, and longitudinal studies have clearly shown a marked improvement in insulin sensitivity upon suppression of disease activity with short-term glucocorticoid therapy and DMARDs in RA (5, 6). The relative importance of weight control and disease activity suppression in enhancing insulin sensitivity in RA deserves further study.

It is noteworthy that the patient's assessment of disease activity constituted the only disease activity variable that was independently associated with insulin resistance. As opposed to the other disease activity parameters that were recorded, the patient's assessment of disease activity is a self reported variable, and such characteristics are prone to be influenced by psychological status (42). This may be relevant in the present context, since depression is common in RA (43) and is related to insulin resistance (44).

We have previously reported an independent association between glucocorticoid use and insulin resistance in RA (7). This finding, together with other reported limitations of glucocorticoid therapy in RA (45–47), led to a radical reduction in glucocorticoid use in our clinics. Although the mean disease duration in this cohort (5 years) was shorter that that in our previous investigation (11 years) (7), the most relevant difference between the 2 cohorts was the extent of glucocorticoid therapy. The mean cumulative dose of glucocorticoids in the present study was <0.5 gm, as compared with 7.3 gm in our previous investigation (7). Short-term use of glucocorticoids in RA patients with insulin resistance and active disease was also reported to enhance insulin sensitivity (5). Taken together, the findings indicate that when glucocorticoids are used sparingly and tailored to disease activity, they may not adversely affect insulin sensitivity in RA. This may be most easily achieved by using low-dose intraarticular pulse glucocorticoid therapy (in combination with DMARD initiation or intensification), as was frequently done in the current cohort.

The DAS28, tender joint count, and patient's assessment of disease activity were independent negative predictors of beta cell function. Joint-derived cytokines induce the production of CRP and fibrinogen by the liver (2), whereas fibrinogen is the main determinant of the ESR (48), and certain cytokines adversely affect beta cell function (20–23). However, in contrast to the clinical disease activity variables, the CRP and ESR were not related to beta cell function in the present investigation. Mediators of altered beta cell function other than cytokines include adipocyte-derived nonesterified fatty acids and hormones such as adiponectin and leptin (17). Bokarewa et al (49) recently described increased circulating concentrations of leptin that were not related to the CRP concentrations in RA. The cross-sectional design of the present study precludes a determination of causality between RA disease activity and beta cell function. The interactions between adipocyte-derived substances, disease activity characteristics, and beta cell function in RA require further investigation.

Cumulative doses of glucocorticoids were strongly associated with enhanced beta cell function in the present cohort. In a longitudinal investigation, glucocorticoid therapy accounted for the increased incidence of diabetes in RA (8). These apparently contrasting findings may again relate to differences in glucocorticoid use. In the latter study (8), 33% of the patients were taking prednisone, as compared with 13% in the present investigation. In our 15 patients who were taking prednisone, the dosage was ≤4 mg/day in 8 of them. Although our results need to be confirmed longitudinally, when glucocorticoids are used sparingly in RA, they may improve beta cell function through their antiinflammatory effects, which outweigh their known adverse effects on glucose metabolism (7, 25).

In patients with high-grade inflammation, age was further associated with decreased beta cell function in the mixed regression models. Age was found to be the most important determinant of impaired beta cell function in a large population study (12). Also, the use of ACE inhibitors and/or ARBs was associated with enhanced beta cell function. Recent clinical trials have revealed that the use of these agents lowers the risk of type 2 DM in patients with hypertension (16). There is also evidence that blockade of the renin–angiotensin system may promote the differentiation of adipocytes, which may result in improved insulin sensitivity and decreased storage of excess calories in the pancreas, thereby reducing lipotoxicity and improving insulin secretion (50). Our results may further support the use of the latter agents as first-line therapy for hypertension in RA.

As mentioned above, a limitation of this study is its cross-sectional design, and our findings need to be confirmed in a longitudinal investigation. We evaluated only 94 patients. An analysis of a larger data set may reveal a greater number of significant associations between the independent variables assessed in the current study and insulin resistance and impaired beta cell function. Also, we used the HOMA-IR and HOMA-B (29) to evaluate insulin resistance and beta cell function, respectively. Alternative methods include dynamic tests, such as the hyperinsulinemic euglycemic clamp for insulin resistance and the hyperglycemic clamp for beta cell function. However, the latter techniques are not necessarily superior to the HOMA-IR and the HOMA-B (29). Rather, they give information about different aspects of insulin resistance and beta cell function. Each of these different investigations has its own limitations. Furthermore, there is a good correlation between estimates of insulin resistance derived from the HOMA-IR and from the euglycemic clamp, and between estimates of beta cell function derived from the HOMA-B and from the hyperglycemic clamp (29). Finally and as alluded to previously, in view of the sparing use of glucocorticoids, our results may not apply to RA patients treated with higher doses of the corticosteroids.

In conclusion, abdominal obesity and RA disease activity were determinants of insulin resistance in this cohort of patients with RA. Age and disease activity were associated with reduced beta cell function, whereas treatment of hypertension with ACE inhibitors and/or ARBs and the sparing use of glucocorticoids were associated with enhanced beta cell function. Since abdominal obesity, antihypertensive therapy, disease activity, and glucocorticoid use constitute factors that are modifiable, our findings may have implications for the prevention and treatment of cardiovascular disease and diabetes in RA.

REFERENCES

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES
  • 1
    Maradit-Kremers H, Crowson CS, Nicola PJ, Ballman KV, Roger VL, Jacobsen SJ, et al. Increased unrecognized coronary heart disease and sudden deaths in rheumatoid arthritis. Arthritis Rheum 2005; 52: 40211.
  • 2
    Sattar N, McCarey DW, Capell H, McInnes IB. Explaining how “high-grade” systemic inflammation accelerates vascular risk in rheumatoid arthritis. Circulation 2003; 108: 295763.
  • 3
    Del Rincon I, Freeman GL, Haas RW, O'Leary DH, Escalante A. Relative contribution of cardiovascular risk factors and rheumatoid arthritis clinical manifestations to atherosclerosis. Arthritis Rheum 2005; 52: 341323.
  • 4
    Dessein PH, Stanwix AE, Joffe BI. Cardiovascular risk in rheumatoid arthritis versus osteoarthritis: acute phase response related decreased insulin sensitivity and high-density lipoprotein cholesterol as well as clustering of metabolic syndrome features in rheumatoid arthritis. Arthritis Res 2002; 4: R5.
  • 5
    Svenson KL, Pollare T, Lithell H, Hallgren R. Impaired glucose handling in active rheumatoid arthritis: relationship to peripheral insulin resistance. Metabolism 1988; 37: 12530.
  • 6
    Dessein PH, Joffe BI, Stanwix AE. Effects of disease modifying agents and dietary intervention on insulin resistance and dyslipidemia in inflammatory arthritis: a pilot study. Arthritis Res 2002; 4: R12.
  • 7
    Dessein PH, Joffe BI, Stanwix AE, Christian BF, Veller M. Glucocorticoids and insulin sensitivity in rheumatoid arthritis. J Rheumatol 2004; 31: 86774.
  • 8
    Wolfe F, Michaud K. Corticosteroids increase the risk of diabetes mellitus in rheumatoid arthritis and contribute to the risk of myocardial infarction and heart failure [abstract]. Ann Rheum Dis 2004; 63 Suppl 1: 495.
  • 9
    Maradit-Kremers H, Nicola PJ, Crowson CS, Ballman KV, Gabriel SE. Cardiovascular death in rheumatoid arthritis: a population-based study. Arthritis Rheum 2005; 52: 72232.
  • 10
    Grundy SM. What is the contribution of obesity to the metabolic syndrome? Endocrinol Metab Clin North Am 2004; 33: 26782.
  • 11
    Grundy SM, Brewer HB Jr, Cleeman JI, Smith SC, Lenfant C. Definition of metabolic syndrome. Report of the National Heart, Lung, and Blood Institute/American Heart Association Conference on Scientific Issues Related to Definition. Circulation 2004; 109: 4338.
  • 12
    Wagenknecht LE, Langefeld CD, Scherzinger AL, Norris JM, Haffner SM, Saad MF, et al. Insulin sensitivity, insulin secretion, and abdominal fat. The Insulin Resistance Atherosclerosis Study (IRAS) Family Study. Diabetes 2003; 52: 24906.
  • 13
    Ferrannini E, Buzzigoli G, Bonadonna R, Giorico MA, Oleggini M, Graziadei L, et al. Insulin resistance in essential hypertension. N Engl J Med 1987; 317: 3507.
  • 14
    Grassi G, Seravalle G, Dell'Oro R, Trevano FQ, Bombelli M, Scopelliti F, et al. Comparative effects of candesartan and hydrochlorothiazide on blood pressure, insulin sensitivity, and sympathetic drive in obese hypertensive individuals: results of the CROSS study. J Hypertens 2003; 21: 17619.
  • 15
    Reneland R, Alvarez E, Andersson PE, Haenni A, Byberg L, Lithell H. Induction of insulin resistance by β-blockade but not ACE-inhibition: long-term treatment with atenolol or trandolapril. J Hum Hypertens 2000; 14: 17580.
  • 16
    Jandeleit-Dahm KA, Tikellis C, Reid CM, Johnston CI, Cooper ME. Why blockade of the renin-angiotensin system reduces the incidence of new-onset diabetes. J Hypertens 2005; 23: 46373.
  • 17
    Stumvoll M, Goldstein BJ, van Haeften TW. Type 2 diabetes: principles of pathogenesis and therapy. Lancet 2005; 365: 133346.
  • 18
    Matsui J, Terauchi Y, Kubota N, Takamoto I, Eto K, Yamashita T, et al. Pioglitazone reduces islet triglyceride content and restores impaired glucose-stimulated insulin secretion in heterozygous peroxisome proliferators-activated receptor-γ-deficient mice on a high-fat diet. Diabetes 2004; 53: 284454.
  • 19
    Assimacopoulos-Jeannet F. Fat storage in pancreas and in insulin-sensitive tissues in pathogenesis of type 2 diabetes. Int J Obes Relat Metab Disord 2004; 28 Suppl 4: S537.
  • 20
    Yang Z, Chen M, Ellett JD, Carter JD, Brayman KL, Nadler JL. Inflammatory blockade improves human pancreatic islet function and viability. Am J Transplant 2005; 5: 47583.
  • 21
    Chen M, Yang ZD, Smith KM, Carter JD, Nadler JL. Activation of 12-lipoxygenase in proinflammatory cytokine-mediated beta cell toxicity. Diabetologia 2005; 48: 48695.
  • 22
    Parkash J, Chaudhry MA, Rhoten WB. Tumor necrosis factor-α-induced changes in insulin-producing beta-cells. Anat Rec A Discov Mol Cell Evol Biol 2005; 286: 98293.
  • 23
    Groeneveld JA, Beishuizen A, Visser FC. Insulin: a wonder drug in the critically ill? Crit Care 2002; 6: 1025.
  • 24
    Delaunay F, Khan A, Cintra A, Davan LD, Ling ZC, Andersson A, et al. Pancreatic β-cells are important targets for the diabetogenic effects of glucocorticoids. J Clin Invest 1997; 100: 20948.
  • 25
    Lambillotte C, Gilon P, Henquin JC. Direct glucocorticoid inhibition of insulin secretion. An in vitro study of dexamethasone effects in mouse islets. J Clin Invest 1997; 99: 41423.
  • 26
    Arnett FC, Edworthy SM, Bloch DA, McShane DJ, Fries JF, Cooper NS, et al. The American Rheumatism Association 1987 revised criteria for the classification of rheumatoid arthritis. Arthritis Rheum 1988; 31: 31524.
  • 27
    Dessein PH, Joffe BI, Stanwix AE. Subclinical hypothyroidism is associated with insulin resistance in rheumatoid arthritis. Thyroid 2004; 14: 4436.
  • 28
    Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF, Turner RC. Homeostasis model assessment: insulin resistance and beta-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia 1985; 28: 4129.
  • 29
    Wallace TM, Levy JC, Matthews DR. Use and abuse of HOMA modeling. Diabetes Care 2004; 27: 148795.
  • 30
    Dessein PH, Joffe BI, Veller MG, Stevens BA, Tobias M, Reddi K, et al. Traditional and nontraditional cardiovascular risk factors are associated with atherosclerosis in rheumatoid arthritis. J Rheumatol 2005; 32: 43542.
  • 31
    Dessein PH, Joffe BI, Singh S. Biomarkers of endothelial dysfunction, cardiovascular risk factors and atherosclerosis in rheumatoid arthritis. Arthritis Res Ther 2005; 7: R63443.
  • 32
    Prevoo ML, van 't Hof MA, Kuper HH, van Leeuwen MA, van de Putte LB, van Riel PL. Modified disease activity scores that include twenty-eight–joint counts: development and validation in a prospective longitudinal study of patients with rheumatoid arthritis. Arthritis Rheum 1995; 38: 448.
  • 33
    American Diabetes Association. Diagnosis and classification of diabetes mellitus. Diabetes Care 2005; 28: 53742.
  • 34
    The Diabetes Prevention Program Research Group. Role of insulin secretion and sensitivity in the evolution of type 2 diabetes in the Diabetes Prevention Program. Effects of lifestyle intervention and metformin. Diabetes 2005; 54: 240414.
  • 35
    Festa A, Hanley AJ, Russell PT, D'Agostino RD, Haffner SM. Inflammation in the prediabetic state is related to increased insulin resistance rather than decreased insulin secretion. Circulation 2003; 108: 182230.
  • 36
    Jeppesen J, Facchini FS, Reaven GM. Individuals with high total cholesterol/HDL cholesterol ratios are insulin resistant. J Intern Med 1998; 243: 2938.
  • 37
    Westhovens R, Nijs J, Taelman V, Dequeker J. Body composition in rheumatoid arthritis. Br J Rheum 1997; 36: 4448.
  • 38
    Rall LC, Roubenoff R. Rheumatoid cachexia: metabolic abnormalities, mechanisms and interventions. Rheumatology (Oxford) 2004; 43: 121923.
  • 39
    Knowler WE, Barrett-Connor E, Fowler SE, Hamman RF, Lachin JM, Walker EA, et al. Reduction in the incidence of type 2 diabetes with lifestyle intervention or metformin. N Engl J Med 2002; 346: 393403.
  • 40
    Buchanan TA, Xiang AH, Peters RK, Kjos SL, Marroquin A, Goico J, et al. Preservation of pancreatic beta-cell function and prevention of type 2 diabetes by pharmacological treatment of insulin resistance in high-risk Hispanic women. Diabetes 2002; 51: 2796803.
  • 41
    Durbin RJ. Thiazolidinedione therapy in the prevention/delay of type 2 diabetes in patients with impaired glucose tolerance and insulin resistance. Diabetes Obes Metab 2004; 6: 2805.
  • 42
    Bardwell WA, Nicassio PM, Weisman MH, Gevirtz R, Bazzo D. Rheumatoid Arthritis Severity Scale: a brief, physician-completed scale not confounded by patient self-report of psychological functioning. Rheumatology (Oxford) 2002; 41: 3845.
  • 43
    Ang DC, Choi H, Kroenke K, Wolfe F. Comorbid depression is an independent risk factor for mortality in patients with rheumatoid arthritis. J Rheumatol 2005; 32: 10139.
  • 44
    Weber-Hamann B, Hentschel F, Kniest A, Deuschle M, Colla M, Lederbogen F, et al. Hypercortisolemic depression is associated with increased intra-abdominal fat. Psychosom Med 2002; 64: 2747.
  • 45
    Van Gestel AM, Laan RF, Haagsma CJ, van de Putte LB, van Riel PL. Oral steroids as bridge therapy in RA patients starting with parental gold: a randomized double-blind placebo-controlled trial. Br J Rheumatol 1995; 34: 34751.
  • 46
    Capell HA, Madhok R, Hunter JA, Porter D, Morrison E, Larkin J, et al. Lack of radiological and clinical benefit over two years of low dose prednisolone for rheumatoid arthritis: results of a randomized controlled trial. Ann Rheum Dis 2004; 63: 797803.
  • 47
    Wolfe F, Michaud K, Caplan L, Kamen D, Furst DE, Russell AS. Corticosteroid therapy is associated with serious side effects over a broad spectrum of outcomes: results from a large longitudinal study [abstract]. Arthritis Rheum 2004; 50 Suppl 9: S381.
  • 48
    Wolfe F. Comparative usefulness of C-reactive protein and erythrocyte sedimentation rate in patients with rheumatoid arthritis. J Rheumatol 1997; 24: 147785.
  • 49
    Bokarewa M, Bokarew D, Hultgren O, Tarkowski A. Leptin consumption in the inflamed joints of patients with rheumatoid arthritis. Ann Rheum Dis 2003; 62: 9526.
  • 50
    Sharma AM, Janke J, Gorzelniak K, Engeli S, Luft FC. Angiotensin blockade prevents type 2 diabetes by formation of fat cells. Hypertension 2002; 40: 60911.