Our study showed that measurements of RFNL were significantly related with plasma fT3 and fT4, TRAbs and antiTPO levels in patients with GO but not TSH levels. Baseline intraocular pressure and macular thickness were also correlated with serum baseline TSH levels. This study suggested fT3, fT4 and TRAbs levels were the main significant factor for the evaluation of intraocular structure especially optic nerve. RFNL, macular thickness and intraocular pressure measurements were similar before and after the treatment and this result could be related with the baseline mild orbitopathy of all patients. By providing those objective measures of retinal damage, early and effective antithyroid therapy is necessary for all patients with GO.
GO is an uncommon, disfiguring and disabling autoimmune condition and more than 90% of cases occur in patients presenting with hyperthyroidism due to Graves’ Disease (14). Thyroid eye diseases and Graves’ hyperthyroidism have been commonly seen in female patients who are 30–50 years old. In addition, some studies (15, 16) reported severity of the disease tends to be worse in men and in patients over the 50 years old. Our study showed that mean age values were similar. Although mean serum fT3, fT4 values and CAS were more increased in male patients than the female patients but we did not find any significance between the two genders. This may be related with the small sample size of our study.
Graves orbitopathy is related with involvement of the orbital fibroblast as a result of the autoimmune processes that collectively induce proliferation excess adipogenesis and overproduction of the extracellular matrix (17, 18). Extracellular matrix including glycosaminoglycans is able to absorb up to 1000 times its weight in water. This results in an increase in the orbital volume and therefore an increase in intraorbital pressure (19). This is evaluated clinically as the disease activity by using clinical activity score. In general, optic nerve involvement is expected in patients with severe thyroid eye disease (20). Tehrani et al. (21) showed that macular and peripapillary vessel density were significantly lower in patients with active GO compared to nonactive noncompressive which reported CAS as equal to or less than 3. They also reported mean intraocular pressure for active GO and nonactive noncompressive GO were similar (18.7 ± 5.4 vs 16.4 ± 3.3). In their cohort, TRAbs was present in 90% of patients and anti-thyroid peroxidase was positive in 12.7% patients. Dave et al. (22) showed that peripapillary and macular vascularity indexes were lower in patients with active GO. They also presented that RFNL thickness was increased in active GO versus the inactive eyes and healthy controls. Sayın et al. (23) demonstrated that intraocular pressure (14.3 ± 2.7 vs 12.9 ± 1.9, p = 0.00) was higher in patients with GO (NOSPECS score; 2.4 ± 0.9) than the healthy controls. But they did not show any significant difference of RFNL (98.3 ± 29.4 vs 101.7 ± 30.7) among the patients with GO and healthy controls. In the above mentioned study, we did not know baseline and follow-up hormone levels or autoantibodies between the patient groups. In our study we did not find any correlation of RFNL, intraocular pressure and macular thickness measurements between pre and after antithyroid therapy. Different from these results, we found a robust correlation between serum fT3, fT4, TRAbs, anti TPO levels and RFNL measurements. Before the ATDs intraocular pressure and macular thickness were negatively correlated with TSH levels. Although mean baseline intraocular pressure (14.8 ± 2.6 vs 14.6 ± 2.5) and macular thickness (270.53 ± 42.42 vs 270.07 ± 42.07) measurements were similar during the following period, we found serum TSH levels were a significant factor that affected macular thickness and intraocular pressure. We believe that this can be related with maintaining long term TSH suppression even normalized fT3, fT4 levels.
Optic nerve pathology in thyroid ophthalmopathy is compressive neuropathy. Some studies (24–27) reported RFNL thickness was decreased in patients with GO versus the inactive eyes and healthy controls. The reduction in the RFNL may be hypothesized to be due to hypoxia and resultant ischemia caused by edema in the orbital soft tissue. But some of the other studies (28–30) reported the opposite argument related with RFNL thickness. They suggested that increase in RFNL thickness may be related with initial edema caused by optic nerve compression. But it has been known that RFNL measurements vary according to age, ethnicity and duration of the diseases. To find more accurate results of these measurements, they should be evaluated in the same patient population during the treatment or matched with similar age healthy controls. During the follow-up time, we observed similar thickness of RFNL in patients with GO. This can be explained as all of the patients having lower initial CAS. But we found a negative correlation between initial fT4, fT3 and TRAbs titer and RFNL thickness. Although patients with GO had similar measurements of RFNL; initial fT3, fT4 have been independent risk factors that influence the thickness of RFNL.
It has been known that TSH receptor antibodies are a major risk factor for the occurrence of GO. TSH receptor activation enhances the differentiation of orbital preadipocytes into adipocytes, favoring the expansion orbital adipose tissue (31–33). A multicenter prospective study (34) from 2018 proposed a predictive value of developing GO with baseline ocular inflammation, smoking, duration of thyroid disfunction and especially the TRAbs titer as the four key risk factors. Even though patients received euthyroid status during the follow-up, ophthalmopathy could progresses in patients with high titer of TRAbs. We found a significant negative correlation between serum TRAbs and before-after RFNL thickness. This may be related with long-term positivity of TRAbs titers .
There were several limitations in our study. This was an observational prospective study with a small sample size. Long-term follow-up studies need to evaluate the changes in the intraocular structure. In addition, all measurements including intraocular pressure, RFNL and macular thickness of patients could be compared with the age and sex of healthy controls.
The strength of our study is that we detailed evaluated intraocular structure before and after antithyroid therapy. All patients had lower CAS at baseline namely nonactive GO, but we found serum TRAbs, fT3 and fT4 had a significant factor for the RFNL. Up to date there have not been any studies which showed the correlation between macular thickness, RFNL and serum fT3, FT4 and TRAbs titers in patients with GO.
In conclusion, serum fT3, fT4 and TSH levels were independent and important parameters for the retinal nerve thickness in patients even with non-active orbitopathy. We postulate that these results could reflect early retinal effects by the thyroid hormone and autoantibody levels should be evaluated in all patients with GO.