Clinical results of anti-inflammatory therapy in Graves' ophthalmopathy and association with thyroidal autoantibodies.

OBJECTIVE: Graves' ophthalmopathy (GO) is clinically associated with autoimmune thyroid disease, and autoantibodies to thyroidal antigens, especially to the TSH-receptor (TRAb), might be involved in the disease process. While there is mounting evidence that TRAb are associated with GO at the onset of the disease, so far no studies have looked at the association between thyroidal autoantibodies and the clinical outcome of GO therapy. The aim of this retrospective study was to evaluate whether TSH binding inhibitory immunoglobulins (TBII) and thyroid stimulating antibodies (TSAb) are still associated with the clinical activity and severity of GO after the completion of anti-inflammatory therapy. In addition, we wanted to elucidate whether thyroid peroxidase (TPO) or thyroglobulin (TG) autoantibodies (TPOAb and TGAb) are in any way related to GO. DESIGN PATIENTS AND MEASUREMENTS: Clinical activity score (CAS) and the severity of GO (modified NOSPECS score) were assessed in 108 patients with GO after steroid therapy and, if indicated, orbital irradiation. Patients were grouped according to their clinical presentation and autoantibody levels (TBII, TSAb, TPOAb and TGAb) were measured. After therapy for hyperthyroidism, all patients were clinically euthyroid but showed clear heterogeneity for GO 4-12 months after the end of anti-inflammatory therapy. Fifty-two patients had inactive GO, 41 had moderately active and 15 still had very active (non-responsive) GO. Concerning severity, 27 patients had mild GO, 64 moderately severe and 17 severe GO.
RESULTS: TBII titres were still positive in 14 (93%) of 15 patients in the non-responsive group (CAS > 6) compared to 22 (42%) of 52 patients (P < 0.001) with post-therapeutic inactive GO (CAS </= 2). A similar result was seen between TBII levels and the NOSPECS score. The simultaneous presence of TBII and TSAb was associated with significantly higher activity and severity [odds ratios: 4.9 (activity); 9.0 (severity)] than the presence of TBII without measurable TSAb [odds ratios: 2.1 (activity), 2.0 (severity)] in comparison to absence of both antibodies. Only TBII and TSAb, but not TPOAb or TGAb medians, increased statistically significantly with CAS or NOSPECS scores. Both scores were positively associated with TBII (CAS: r = 0.31 P < 0.001; NOSPECS: r = 0.38, P < 0.0001) and, to a lesser degree, with TSAb (CAS: r = 0.27, P < 0.007, and NOSPECS: r = 0.29, P < 0.003). This association was independent of the treatment of hyperthyroidism, although highest levels of TBII were seen after radioiodine treatment. The NOSPECS score was negatively associated with TGAb (r =-0.27, P < 0.01) but not with TPOAb, while both showed no association with the CAS score.
CONCLUSIONS: We conclude that the persistence of TBII and TSAb levels in patients with therapy-resistant disease in comparison to patients with inactive disease supports the role of TRAb in the pathogenesis of GO. Furthermore, the fact that, even after anti-inflammatory therapy, TBII and TSAb levels and prevalence still correlate with the severity and activity of GO suggests not only a trigger but also a possible role in the maintenance of the autoimmune process in the orbits.