Background Insulin resistance (IR) may contribute to an increase in cardiovascular risk in general population as well as in Systemic Lupus Erythematosus (SLE) patients.

Objectives The aim of this study was to examine the association between IR and disease activity, disease characteristics, drug exposure and subclinical atherosclerosis in patients with SLE.

Methods Cross-sectional study that encompassed 87 SLE patients and 82 sex-matched controls. IR by homeostatic model assessment (HOMA2), insulin, C-peptide serum levels and lipid profile were assessed in both groups. Activity (SLEDAI), severity (Katz) and damage (SLICC) scores, carotid intima-media thickness (cIMT) and carotid plaques (ultrasound) were assessed in SLE patients. A multivariable regression analysis, adjusted for IR related factors, was performed to evaluate the differences between groups in IR indexes and, in SLE patients, the interrelation between IR and disease activity/characteristics as well as subclinical atherosclerosis.

Results Median disease duration was 16 years (IQR 9–21). Body mass index, abdominal circumference, hypertension or dyslipidemia did not differ between groups. According to the SLEDAI score, 40% of patients were in no activity, while 32, 21, 18 and 1% were in mild, moderate, high and very high activity respectively.

HOMA-IR-C-peptide (beta coefficiente 0.53, [95% CI 0.25–0.82], p=0.00) was increased in SLE patients when compared to controls, as well as HOMA %B C peptide levels (beta coef. 35, 95% CI 18–52, p=0.00). Similarly, insulin sensitivity estimated through HOMA-S% was inferior in SLE patients (-beta coef. -37, 95% CI -63–11, p=0.01).This diferences remained significant even after adjustment for IR related factors.

SLICC damage index was clearly asociated with IR indexes; higher index values were related to higher HOMA-IR-C-peptide (beta coef. 37, [95% CI 16–57], p=0.00) and lower HOMA-S% levels (beta coef. 30%, [95% CI -47–14], p=0.00). Katz severity index showed correlation with HOMA-IR-C-peptide (beta coef.-5, [95% CI -11–0], p=0.04). These associations remained significant after adjustment for age, gender, smoking, hypertension and dyslipidemia, and, in relation with the SLICC index, also after adjustment for prednisone intake. SLEDAI activity index was not related to IR indexes.

The use of prednisone was positively associated with HOMA-IR both when considered binary (beta coef 47, [95% CI 31–63], p=0.00) and continous (beta coef 2 [95% CI 0–5] per mg, p=0.03). Hydroxycloroquine (or other drugs) use was not associated with IR indexes, neither were disease duration, antiDNA titers and complement serum levels.

Carotid plaques were found in 20% of the SLE patients. The presence of carotid plaques was correlated with a higher HOMA-IR-C-peptide (OR 3.15 [95% IC 1.17–8.51], p=0.02), and a higher cIMT value was associated with a lower HOMA-IR-S%>C-peptide (beta coef. 0.98 [95% CI 0.96–0.99], p=0.03). Nevertheless, after adjustment for cardiovascular risk factors this relation was lost.

Conclusions Activity and damage indexes in SLE patients are independently related to the development of IR. IR is associated with subclinical carotid atherosclerosis in SLE patients on the univariate analysis.

Disclosure of Interest None declared