Abstracts

Rationale: Discontinuation of treatment is typically considered jointly by the treating physician and patient after prolonged seizure freedom. However, patients not uncommonly elect to self-discontinue treatment. Most studies evaluating treatment discontinuation have been in patients with chronic epilepsy and where the decision to withdraw treatment has been physician-directed. Few studies have focused on newly diagnosed patients, nor those who self-discontinued. We studied the frequency, risk factors, reasons and sequelae of self-discontinuation of treatment with newly diagnosed epilepsy.

Methods: Patients attending First Seizure Clinics of publicly funded hospitals in Western Australia between 1 May 1999 and 31 May 2016 were prospectively recruited. Patients with newly diagnosed epilepsy who commenced treatment with antiseizure medications (ASMs) were studied. Patients in this cohort who self-discontinued treatment were identified and compared to patients who remained on treatment. Patients whose treatment was discontinued by their neurologist were excluded. We also analysed seizure recurrence, seizure-related injury and treatment re-commencement after discontinuation.

Results: We studied 489 patients with newly diagnosed epilepsy (median age 41, range 14-88, 62% male) followed for a median duration of 3.0 years (interquartile range [IQR]: 1.2-6.0) after commencement of treatment. 78 (16.0%) patients self-discontinued ASMs after a median treatment duration of 1.4 years (IQR: 0.4-2.9), at a median time from last seizure of 11.8 months (IQR: 4.6-31.8). 51% patients were not seizure-free for 12 months preceding discontinuation.

The most common reasons for treatment discontinuation provided by the patients were adverse effects (32/78, 41%), perception that ASMs were no longer required (27/78, 35%) and planned or current pregnancy (9/78, 12%) (Table 1). Patients who self-discontinued treatment were less likely to have epileptogenic neuroimaging lesions compared to those who did not (29.5% vs. 49.9%, hazard ratio [HR]=0.53, 95% confidence interval [CI]: 0.29-0.97), had a slightly shorter interval since the last seizure (median 0.99 vs. 1.05 years, HR=0.77, 95% CI: 0.68-0.88), and were less likely to present initially with seizure clusters (11.5% vs 22.9%, HR=0.49, 95% CI: 0.24-0.99) (Table 2).

After discontinuation, 51 patients (65%) experienced seizure recurrence after a median of 6.1 months (IQR 1.2 months-1.9 years), of whom 43 (84%) restarted treatment during follow-up. 28% (22/78) patients experienced a seizure-related injury after discontinuation of treatment.

Conclusions: More than 1 in 7 adults with newly diagnosed epilepsy self-discontinued treatment, often within 18 months of commencing treatment and without attaining seizure freedom. This is likely an underestimate given the possibility of self-discontinuation beyond follow up. Self-discontinuation of treatment was frequently followed by seizure relapse, sometimes resulting in injuries. Reasons for discontinuation highlight areas for improved discussion with patients during the early course of their condition, including the chronicity of epilepsy, and pre-emptive management strategies for current or potential adverse effects.

Funding: Please list any funding that was received in support of this abstract.: N/A.