Type 1 diabetes often occurs in association with other autoimmune diseases (thyroid disease [1–3], Addison’s disease [2], celiac disease [1,3], autoimmune gastropathy [4], and others), with autoimmune thyroiditis beingthe most common (1–3). To determine whether these autoimmune diseases are also more frequent in patients with latent autoimmune diabetes of adults (LADA), we compared the prevalence of anti–21 α-hydroxylase (21OHAb), anti–thyroid peroxidase (TPOAb), anti–parietal gastric cell antibodies, and rheumatoid factor in a group of patients with LADA versus patients with type 2 diabetes.
Fifty-four patients with LADA (>35 years of age at diagnosis who did not require insulin for at least 1 year after diagnosis and with positive GAD antibody [GADA]) (5) were included in the study (group A). Group B consisted of 54 individuals with type 2 diabetes (negative GADA) with the same characteristics (31 men and 23 women in the two groups, mean age 42 ± 5.8 and 43.2 ± 4.9 years, respectively). The interval between the diagnosis of diabetes and antibody measurement was <2 years (14 ± 3 and 13 ± 5 months, respectively). GADAs were determined by radioimmunoassay, with a level ≤1 unit/ml considered negative. TPOAbs were measured by a chemiluminescent assay, with a negative value <15 units/ml. Anti–parietal gastric cell antibodies were assayed by indirect immunofluorescence and 21OHAb by radioimmunoassay-iodine, with a level ≤1 unit/ml considered negative. Rheumatoid factor was measured by nefelometry (reference value <35 units/ml). Differences between the two groups were tested by Fisher’s exact test. A P value <0.05 was considered statistically significant.
The frequency of 21OHAb did not differ between the two groups (5.5% in group A vs. 0% in group B, P = 0.24). None of the three 21OHAb-positive patients presented clinical or laboratory evidence of primary adrenal insufficiency (determination of potassium, ACTH, cortisol, aldosterone, and plasma renin activity). There was also no difference in anti–parietal gastric cell antibodies (9.2 vs. 3.7%, P = 0.43), and none of the patients presented cobalamin deficiency (vitamin B12 <200 pg/ml). Rheumatoid factor was similar in the groups (7.4% in LADA vs. 1.8% in type 2 diabetes, P = 0.21), and all patients were clinically asymptomatic (rheumatoid arthritis). The prevalence of TPOAb was significantly higher in group A (24 vs. 5.5%, P = 0.006). In addition, a difference in the concentration of TPOAb was also detected (82 vs. 28 IU/ml, P = 0.003). Primary hypothyroidism (thyroid-stimulating hormone >5 IU/ml) was observed in 25% of the patients with TPOAb (3 of 13 patients in group A and 1 of 3 patients in group B).
Increased frequency of autoimmune disease–associated antibodies is observed in patients with type 1 diabetes (1–4). Similarly, we demonstrated that patients with LADA also show a higher prevalence of TPOAb when compared with type 2 diabetic individuals. The lack of a significant difference in 21OHAb, anti–parietal gastric cell antibodies, and rheumatoid factor might be attributed to the size of the sample. In agreement with our results, other studies have also reported a higher prevalence of autoantibodies in patients with LADA than in type 2 diabetic individuals, namely TPOAb (6–11), anti-gliadin and -endomysial antibodies (6), 21OHAb (7,9), and anti–parietal cell antibodies (11).
We conclude that patients with LADA show a higher frequency of autoimmune thyroiditis and should be routinely investigated for this condition.
