Nomogram for predicting pregnancy-related relapse of myasthenia gravis

Abstract

Background Myasthenia gravis (MG) is an autoimmune disease mediated by autoantibodies primarily affecting the neuromuscular junction. This study aims to identify risk factors for pregnancy-related MG relapse and develop a predictive model to improve clinical outcomes. Methods We enrolled 113 MG female patients with a pregnancy history during follow-up at Huashan Hospital affiliated with Fudan University, between January 2015 and October 2021. The study analyzed relapse rates and risk factors during pregnancy and postpartum using multivariate logistic regression. A nomogram was constructed to predict relapse probability, with model performance evaluated by discrimination and calibration metrics. Results Of the 113 patients, 52 (46.02%) experienced 115 relapses, including 52 (45.22%) occurring during the first trimester of pregnancy, 11 (9.56%) during the second trimester of pregnancy, and 52 relapses (45.22%) during the three months after delivery/abortion. Significant factors associated with pregnancy-relate relapse, included age at delivery/abortion (OR 0.21, 95% CI 0.06–0.65), MG stable duration (OR 0.24, 95% CI 0.09–0.63), thymic hyperplasia (OR 3.45, 95% CI 1.35–9.3), pre-pregnancy thymectomy (OR 0.08, 95% CI 0.01–0.36), and inadequate treatment during pregnancy (OR 4.44, 95% CI 1.35–17.76). The Nomogram model demonstrated robust predictive performance. Conclusion The first trimester of pregnancy and three months following delivery or abortion are high-risk periods for MG relapse. Younger ages, shorter MG stable duration before pregnancy, thymic hyperplasia, and inadequate treatments during pregnancy increase relapse risk.