ETA2023 Poster Presentations Thyroid Eye Disease (9 abstracts)
1University of Catania, Catania, Italy; 2University of Catania, Clinical and Experimental Medicine, Catania, Italy; 3University of Catania, Clinical and Experimental Medicine, Endocrinology Unit, Catania, Italy; 4Universita DI Catania, University of Catania, Endocrinologia, Garibaldi Nesima, Catania, Italy
Introduction: Oxidative stress (OX) plays a role in the pathogenesis of moderate to severe Graves ophthalmopathy (AMS-GO).
Aim: We evaluated the impact of clinical conditions related to OX on the outcome of parenteral glucocorticoids (PGLUC) therapy in AMS-GO.
Methods: We retrospectively evaluated patients with AMS-GO treated with PGLUC from January 2013 to May 2022. GO clinical evaluation was performed at baseline, at 6 (W6), at 12 (W12) and at 24 weeks (W24) after starting PGLUC. Patients were classified as Improved (I) or Not Improved (NI) by the EUGOGO overall clinical criteria. We then performed multiple univariate binomial logistic regression analyses including the following covariates as OX conditions: W6 outcome, total and calculated LDL cholesterol (LDLc), Body Mass Index (BMI), fasting glycemic values, history of hypertension (HoH), smoking, age and sex. Outcomes at W12 and W24 were set as dependent variables in different analyses. Some multivariate models were finally built with more representative variables by univariate analyses.
Results: 139 patients, 40 males and 99 females, median age of 47 (36-55) years, with AMS-GO received a median PGLUC cumulative dose of 49 (36-65) milligrams/kg body weight. 56/122 (45.9%) patients were classified as I at W6, 60/139 (43.2%) at W12. 86 patients completed the follow-up at W24. Among them, 33/86 (38.4%) were I. When compared to NI, early I at W6 showed an 8 and 7 times greater chance of being classified as I at W12 and at W24, respectively (OR 8.1, 95% CI 2.8-23.0, P < 0.001 and OR 7.0, 95% CI 2.5-19.7, P < 0.001), I vs NI). At univariate regression analyses, total and LDLc cholesterol, HoH and age were predictive variables towards both W12 and W24 outcomes (Table 1). The multivariate model aimed at predicting W12 outcome was significant, χ2(3)=15.094, P = 0.002. Among covariates, LDLc and HoH resulted significant (P = 0.05 and P = 0.04, respectively). The multivariate model aimed at predicting W24 outcome resulted significant, χ2(2)=9.458, P = 0.009. LDLc and HoH almost reached significance (P = 0.09 and P = 0.052, respectively).
W12 Outcome | W24 Outcome | |
W6 Outcome | P < 0.001 | P < 0.001 |
Total cholesterol | P = 0.037 | P = 0.12 |
LDLc | P = 0.015 | P = 0.042 |
BMI | P = 0.989 | P = 0.319 |
Fasting glycemia | P = 0.356 | P = 0.308 |
History of hypertension | P = 0.003 | P = 0.031 |
Smoking | P = 0.941 | P = 0.442 |
Age | P = 0.023 | P = 0.017 |
Sex | P = 0.781 | P = 0.046 |
Conclusions: An early I (at 6W) to PGLUC is predictive of the medium and long term clinical response. HoH, total and LDL cholesterol are predictors of low clinical response to PGLU at W12. Although our data suggest the existence of a relationship with the long term outcome, further studies are needed to deepen it.