Anatomical hemispherectomy, removing a whole cerebral hemisphere, could stop seizures in catastrophic unilateral epilepsy, but years later some patients developed superficial cerebral haemosiderosis from repeated bleeding into the large cavity. Hemispherotomy keeps most of the hemisphere in place and disconnects it instead, aiming for similar seizure control with fewer late complications.
Stop removing the brain. Just cut the connections. That, in one line, is how one of the biggest operations in epilepsy surgery changed over the twentieth century.
In 1928, Walter Dandy removed an entire cerebral hemisphere. The patient survived. Over the following decades, anatomical hemispherectomy became an established operation for catastrophic epilepsy arising from one hemisphere.
The principle was direct: remove the diseased hemisphere, and the seizures may stop. In many patients, they did.
The late complications took time to become clear. Years after an apparently successful operation, some patients developed headache, drowsiness and progressive neurological decline.
Repeated small haemorrhages into the large postoperative cavity could lead to superficial cerebral haemosiderosis, with iron deposited on the surface of the remaining brain and damaging it. Hydrocephalus was also a recognised problem. In severe cases, these complications could be fatal long after the original surgery.
The field did not abandon the operation. It changed what the operation was.
Hemispherotomy aims to disconnect the diseased hemisphere functionally rather than remove it. The key pathways divided include:
Most of the hemisphere stays in place, with its blood supply, so there is no large empty cavity. Several techniques have been described, including peri-insular (lateral) and vertical parasagittal approaches.
In appropriately selected patients, modern series report long-term seizure freedom in roughly 70 to 80% of cases.
It remains a major decision. Patients may accept permanent deficits such as hemiparesis and homonymous hemianopia, although many already have significant deficits from the underlying disease. These risks must be discussed openly before surgery.
Given similar seizure-freedom rates and a lower risk of late complications, is there any remaining indication for anatomical hemispherectomy today?
Consider, for example, failed disconnection, or anatomy that makes a complete disconnection difficult.
Educational content for clinicians and trainees, reflecting the author's understanding at the date of review. It is not medical advice: if you are a patient, please discuss your care with your own clinical team.
