After Disconnective Hemispherectomy, Epileptiform Discharges Are Common on the Operated Side in Seizure-Free Patients

Most pediatric patients who are seizure-free after successful disconnective hemispherectomy continue to have abnormalities evident by electroencephalography (EEG) in the disconnected hemisphere. Nevertheless, many of these patients can be weaned off antiseizure medications (ASMs) with no recurrence of seizures.

So reported researchers from Cleveland Clinic’s Charles Shor Epilepsy Center in a Dec. 5 platform presentation at the 2021 annual meeting of the American Epilepsy Society. The findings are from a series of 50 consecutive pediatric patients who became seizure-free following hemispherectomy.

“We want to raise awareness that young patients who are seizure-free after disconnective hemispherectomy may have EEG findings of epileptiform activity — and even subclinical seizures — in the disconnected hemisphere, but that this need not consign these patients to a lifetime of taking antiseizure medications,” says the study’s senior author, Ahsan N.V. Moosa, MD, a Cleveland Clinic pediatric epileptologist.

EEG findings and disconnective hemispherectomy outcomes

Disconnective hemispherectomy is an important option for selected patients with drug-resistant epilepsy. Cleveland Clinic’s experience from 170 patients who underwent the surgery between 1997 and 2009 resulted in about two-thirds being seizure-free after hemispheric surgery (Neurology. 2013;80:253-260).

However, a wide spectrum of abnormalities in the disconnected hemisphere may be seen on EEG after the operation. Although EEG abnormalities usually help determine whether a patient should be weaned off ASMs following surgery, the nature of these abnormalities and how they correlate with clinical outcomes have not previously been well characterized in the setting of disconnective hemispherectomy.

Profile of patients in the series

The current study analyzed 50 consecutive pediatric patients from Cleveland Clinic’s Epilepsy Surgery Outcome Database who were seizure-free at last follow-up after hemispherectomy. Mean age at surgery was 7.5 years (median, 6 years; range, 8 months to 21 years), and patients were evenly divided by sex (52% male/48% female).

Right hemispherectomy was performed in 23 patients (46%). Most underwent disconnective hemispherectomy (n = 47; 94%), with the rest having a modified anatomical hemispherectomy. Mean follow-up was 3.8 years (median, 3 years).

The most common causes of epilepsy were gliosis and encephalomalacia of various causes (including perinatal stroke), found in 35 patients (70%). Other causes were malformation of cortical development (n = 7), Rasmussen encephalitis (n = 6) and Sturge-Weber syndrome (n = 2). Thirty patients (60%) had a single seizure type; the rest had more than one.

Preoperative EEGs had features of epileptic encephalopathy in 12 patients (24%), consisting of hypsarrhythmia (n = 5), slow spike-and-wave complexes (n = 4) and continuous spike-waves during sleep (n = 3).

Postoperative EEG findings

Postoperatively, EEGs of all patients exhibited slowing and attenuation of physiological rhythms in the operated hemisphere. Other EEG findings were as follows:

• Epileptiform discharges in the operated hemisphere, 40 patients (80%)
• Epileptiform discharges in the unoperated hemisphere, 7 patients (14%)
• No epileptiform discharges on either side, 9 patients (18%)
• Bilateral independent epileptiform discharges, 6 patients (12%)
• Lateralized periodic discharges in the operated hemisphere, 2 patients (4%)
• Electrographic seizures on the side of surgery without spread to the other hemisphere and without clinical signs, 4 patients (8%)

Clinical outcomes

All patients were clinically seizure-free by study design. Thirty patients (60%) were weaned off ASMs without seizure recurrence.

Regarding the unoperated hemisphere:

• 6 of 7 patients (86%) who postoperatively had epileptiform discharges remained on ASMs.
• 14 of the 43 patients (33%) without discharges remained on ASMs.
• This difference was significant (P = 0.01).

Regarding the disconnected hemisphere:

• 19 of 34 patients (56%) with epileptiform discharges only in this hemisphere were successfully weaned off ASMs.
• 8 of 9 patients (89%) with no discharges were successfully weaned off ASMs.
• This difference was not statistically significant (P = 0.12).

Attempted weaning is worthwhile despite EEG abnormalities

Dr. Moosa highlights several key takeaways from the analysis of this series:

• Most patients who are seizure-free after disconnective hemispherectomy continue to show epileptiform discharges in the operated hemisphere, and some may even have subclinical EEG seizures.
• The disconnected hemisphere remains viable due to intact blood supply and may continue to show abnormal electrical activity, including seizures, but it is unable to generate clinical seizures because of the disconnection from the rest of the brain. “The epileptic activity in the ‘disconnected,’ ‘isolated’ brain tissue has no clinical relevance,” Dr. Moosa notes.
• The presence of subclinical EEG abnormalities on the operated side should not preclude attempts to taper ASMs in seizure-free patients.
• However, in a patient with epileptiform discharges on the nonoperated side, a decision to wean should be individualized based on risks versus benefits.