Drug-resistant epilepsy: A retrospective observational study at a tertiary care hospital

Abstract

Objective: To determine the prevalence and predictors of drug-resistant epilepsy among patients with epilepsy. Methodology: A retrospective observational study was conducted at the Neurology Department of Liaquat National Hospital, Karachi, for the period from November, 2024 to October, 2025. Patients with epilepsy were enrolled, excluding neonates and those with psychiatric comorbidities or active substance abuse. Participants were recruited consecutively. Descriptive statistics summarized patient characteristics, while chi-square tests evaluated associations between variables such as drug adherence, epilepsy type, and antiepileptic drug (ASM) selection. Odds ratios (ORs) with 95% confidence intervals were calculated to quantify the strength of these associations. Results: A total of 382 participants were included (49.5% male, 50.5% female), with a mean age of 31.73 +/- 16.82 years and a mean seizure onset age of 25.3 years. The majority were urban residents with higher educational levels and 19.4% had refractory epilepsy. On multivariate logistic regression analysis, normal MRI brain findings (a OR = 0.29, 95% CI: 0.142–0.570; p < 0.001), normal EEG findings (a OR = 0.50, 95% CI: 0.256–0.974; p = 0.042) and adherence to the given treatment (a OR= 0.516, 95% CI: 0.234-1.138; p=0.101) have lower odds of having drug resistance. Lower level of education (a OR = 3.496, 95% CI: 1.687–7.245; p < 0.001), and a history of status epilepticus (a OR = 8.029, 95% CI: 4.186-15.400: p <0.001) have higher odds of refractory epilepsy. Conclusion: By identifying key predictors of drug-resistant epilepsy, our study contributes to a better understanding of factors associated with treatment resistance and may inform future interventions.