Background Oxidative stress plays a role in disease activity and premature atherosclerosis seen in patients with systemic lupus erythematosus (SLE). Advanced glycation end products (AGEs) are prevalent in the Western diet. The accumulation of serum AGEs disrupts protein function, and interaction with its receptor induces production of reactive oxygen species and activation of vascular endothelial cells, leading to increased oxidative stress. Our hypothesis is that serum AGE levels, indicators of dietary habits, correlate with inflammation and potentially impact autoimmunity.

Materials and methods We evaluated clinical data and serum samples from 80 Gullah African American participants enrolled prospectively into the SLE in Gullah Health (SLEIGH) study. Of the 80 participants, 50 were patients with SLE and 30 were unaffected controls (15 related controls and 15 unrelated controls). Serum samples were assessed with the AGE Competitive ELISA kit from Cell Bio Labs. All samples were normalised to total protein concentration.

The cohort consisted of 50 patients (25 with no history of cardiovascular disease, and 25 with history of cardiovascular disease or renal disease) and 30 controls (15 related controls, 15 unrelated healthy controls). Student’s T-test was used to compare AGE levels among SLE patients to controls. We used logistic regression models to examine predictors of autoantibody status and a nested case-control design to compare AGE levels between different disease subsets of patients.

Results Overall there were no significant differences in mean AGE levels between SLE patients (2.8 mcg/mL ± 1.8), related controls (5.0 mcg/mL ± 3.1) or unrelated controls (1.2 mcg/mL ± 0.6). Obese patients (BMI ≥ 30) had significantly higher AGE levels than non-obese patients (p = 0.03), though there was no difference among controls. Smoking history was associated with higher AGE levels (p = 0.03). Although on average higher, AGE levels were not significantly associated with diabetes, hyperlipidemia, or stroke history. There was no difference in mean AGE levels with presence of hypertension or current corticosteroid use. Regression models demonstrated no significant influence of AGE level on patient or control status (OR 0.93, p = NS), including when adjusted for gender, age (in years) and BMI. Interestingly, among controls, ANA positivity significantly correlated with higher AGE levels (p = 0.01), when adjusted for age (years).

Conclusions Although there was not a difference in AGE levels between SLE patients and controls, the AGE levels were higher with ANA positivity among controls. This finding suggests that serum AGE levels may play a role as a modifiable risk factor for autoimmunity and further study is warranted.