Development of the LARyNGTH score to predict central neurogenic respiratory failure in lateral medullary infarction

Background Central neurogenic respiratory failure (CNRF) is a fatal complication associated with lateral medullary infarcts (LMI). Its detection may be critical for timely intervention and to reduce the risk of mortality. This study aimed to identify the clinical predictors of CNRF in patients with LMI. Methods This study included patients with LMI who were admitted to a university hospital between 2010 and 2022. CNRF was defined as unexpected intubations due to hypercapnia, cardiopulmonary resuscitation owing to respiratory causes, or clinical detection of CNRF. Clinical, neurological, and imaging data were also collected. Univariable and multivariable analyses were performed to identify the clinical predictors of respiratory complications. Results Among the 214 patients, 12 (5.6%) developed CNRF. Hypertension (91.7% vs. 55.9%, p = 0.033) and diabetes mellitus (58.3% vs. 26.2%, p = 0.04) were more common in the CNRF group. Patients with CNRF exhibited higher National Institutes of Health Stroke Scale (NIHSS) scores (6.0 [4.25–7.0] vs. 2.0 [1.0–4.0], p < 0.001) and lower modified Gugging Swallowing Screen (GUSS) test scores (9.0 [3.5–13.75] vs. 15 [13–15], p = 0.014). Imaging showed a higher rate of large rostral infarcts (83.3% vs. 9.4%, p < 0.001) and combined extramedullary lesions (50.0% vs. 13.4%, p < 0.001). Multivariable analysis identified the LARyNGTH score, which included the above variables. A score >5.5 predicted CNRF with an AUROC of 0.968 (95% CI: 0.940–0.996), a sensitivity of 91.7%, a specificity of 94.6%, and an accuracy of 94.4%. Conclusion In patients with LMI, a combined predictive model can predict CNRF and facilitate preemptive intervention.