Cluster analysis of a large dataset of patients with juvenile myoclonic epilepsy: Predicting response to treatment. (February 2023)

• Record Type: Journal Article
• Authors: Asadi-Pooya, Ali A.; Farazdaghi, Mohsen
• Is Part Of: Seizure. Volume 105(2023)
• Page Start: 10
• Page End: 13
• Publication Date: 2023-02
• Subjects: Epilepsy -- Juvenile myoclonic -- Outcome -- Prognosis -- Seizure; Epilepsy -- Periodicals; Epilepsy -- Periodicals; Seizures -- Periodicals; Épilepsie -- Périodiques; Electronic journals; Electronic journals; 616.853
• DOI: 10.1016/j.seizure.2023.01.006
• Languages: English
• ISSNs: 1059-1311

Abstract

Highlights: Cluster analysis showed that there were two distinct clusters of homogeneous subgroups of patients with JME. Response to treatment was different between the clusters (as a trend) ( p = 0.076). It is ideal to apply this technique of statistical analysis on a very large dataset of patients in a multicenter study. Abstract: Objectives: The purpose of the current study was to apply Two-step cluster analysis on a large dataset of patients with juvenile myoclonic epilepsy (JME). We hypothesized that there are distinct subgroups of patients with similar clinical characteristics. We also hypothesized that the seizure outcome is different between these clusters. Methods: This was a retrospective study of a prospectively developed database. All patients with a diagnosis of JME were studied at the epilepsy center at Shiraz University of Medical Sciences, Shiraz, Iran, from 2008 until 2022. The Two-Step cluster analysis (Schwarz's Bayesian Criterion) was applied to the whole dataset. In the next step, the seizure outcome was compared between the clusters of patients. Results: Two hundred and ninety-five patients were included. Two-Step cluster analysis showed that there were two distinct clusters of homogeneous subgroups of patients with JME, presenting with more or less similar clinical characteristics, with a fair (0.4) silhouette measure of cohesion and separation. One hundred and eighty-one patients had a follow up duration of 12 months or longer at our center. Response to treatment at 12 months of follow-up was different between the clusters (as a trend): 43 patients (39.1%) from cluster 1 and 18 people (25.4%) from cluster 2 were free of all seizure types ( p = 0.076). Conclusion: The Two-Step cluster analysis identified two distinct clusters of patients with JME. Individuals with JME, who also have absence seizures, are less likely to enjoy a seizure free state with ASMs.