Identification of Ocular and Systemic Predictors and Development of a Prediction Model in Ocular Myasthenia Gravis

Purpose: Predicting patients with ocular myasthenia gravis (OMG) achieving minimal manifestation status (MMS) or better is crucial for guiding treatment decisions. Prognostic models are becoming increasingly popular in the medical field due to their high accuracy. This study aimed to develop a clinical tool based on both ocular and systemic factors. Methods: We reviewed patients from our hospital from 2020 to 2023. Logistic and Least Absolute Shrinkage and Selection Operator (LASSO) regressions were applied to determine key predictors. A logistic nomogram model was constructed based on the selected factors. The model's performance was rigorously evaluated through internal validation for discrimination, calibration, and clinical utility. Results: Among 216 patients, 126 individuals were included in the study. A total of 61 patients (48.4%) achieved MMS or better status in the 1-year outcome. Five key variables were selected to conduct the prognostic model, including ocular deviation angle, duration, antibody, corticosteroid therapy, and thyroid ultrasound. The area under the curve (AUC) was 0.889 (95% confidence interval [CI] = 0.832-0.947), demonstrating a good performance. The calibration curves also showed excellent agreement between predicted and observed probabilities in both training and validation cohorts. Conclusions: This study provides a clinically applicable predictive model, which accurately identifies patients with OMG likely to achieve MMS or better status within 1 year. The model is predictive of individual patients with great performance, thereby increasing clinical confidence and utility. Translational Relevance: This study translates specific ocular examination findings into a practical prognostic tool. The developed model directly assists clinicians in identifying patients with OMG with a high probability of achieving remission within 1 year, thereby informing personalized management strategies.