Multivariable prediction model of drug resistance in adult patients with generalized epilepsy from Colombia: A case–control study

Abstract Introduction Patients with drug-resistant epilepsy (DRE) account for most of the burden of epilepsy, and they have poor prognosis in seizure control, higher morbidity, and mortality. Objectives The objective of the study was to develop a prognostic model of drug resistance in adult patients with generalized epilepsy from Colombia. Methods In this case–control study of patients with generalized epilepsy, patients were separated into two groups: one group with DRE (cases) according to the new International League Against Epilepsy (ILAE) definition after a complete evaluation performed by an epileptologist and the other group without DRE (control). Variables were analyzed to identify statistical differences between groups and were then selected to construct a prognostic model from a logistic regression. Results One hundred thirty-three patients with generalized epilepsy were studied. Thirty-eight (28.5%) patients had DRE, and 95 (71.5%) did not have DRE. History of status epilepticus, abnormal findings from neurological examination, aura, any degree of cognitive impairment, epileptic seizures at any moment of the day, and any comorbidity were risk factors. The presence of seizures only in the waking state and idiopathic etiology were protective factors. A prognostic model was constructed with previously reported risk factors for DRE and other variables available in the population of this study. In the multivariable analysis, the history of status epilepticus (odds ratio (OR): 5.6, confidence interval (CI): 1.1–20.0, p = 0.031), abnormal findings from neurological examination (OR: 5.7, CI: 2.3–13.9, p = 0.000), and aura (OR: 6.1, CI: 1.8–20.8, p = 0.003) were strongly associated with DRE. Conclusions In adult patients with generalized epilepsy, aura, abnormal findings from neurological examination, and history of status epilepticus were predictive factors for DRE.