A nurse-led nomogram for predicting the risk of myasthenic crisis in patients with myasthenia gravis and bulbar weakness.

Background: The present study aimed to develop and validate a nurse-led clinical prediction model for assessing the risk of myasthenic crisis (MC) in myasthenia gravis (MG) patients with bulbar weakness.

Methods: A retrospective analysis was conducted on MG patients meeting the inclusion criteria from January 2022 to June 2024. The training group included 308 patients (January 2022-January 2024), and the validation group included 77 patients (February 2024-June 2024). The primary outcome was MC occurrence. A binary logistic regression model was constructed and presented as a nomogram. Model performance was evaluated using the area under the receiver operating characteristic curve (AUC) values, calibration curves, and decision curve analysis (DCA). Internal validation was performed by bootstrap resampling, while external validation was conducted using a validation cohort.

Results: The study included 385 MG patients. Logistic regression analysis identified Myasthenia Gravis Foundation of America classification, presence of thymoma, body position, partial pressure of carbon dioxide, oxygenation index, and oropharyngeal secretions as the independent predictors of MC. The nomogram showed good discrimination, with AUC values of 0.806 (sensitivity: 76.0%, specificity: 71.7%) in the training group and 0.832 (sensitivity: 60.0%, specificity: 95.2%) in the validation group. The model also exhibited good calibration, and the results of DCA confirmed the clinical utility of the model across a range of risk thresholds.

Conclusion: This nomogram can serve as an effective tool for nurses to identify MG patients at a high risk of developing MC. Because this study was a single-center retrospective study, future multicenter validation studies are required to further verify and expand the clinical applicability of this model.