Abstract & Commentary
Sources: Jonas R, et al. Cerebral hemispherectomy: Hospital course, seizure, developmental, language, and motor outcomes. Neurology. 2004;62:1712-1721; Duchowney M. Hemispherectomy for epilepsy: When is one half better than two? [Editorial] Neurology. 2004;62:1664-1665; Bien CG, et al. An open study of tacrolimus therapy in Rasmussen encephalitis. Neurology. 2004;62:2106-2109.
Jonas and colleagues have collected the largest series of cases of unilateral hemispherectomy (UH) and assessed outcomes for the parameters listed in the title of their manuscript. In 21 of 115 patients (18%), the indication for surgery was Rasmussen encephalitis (RE). Other indications included hemimegalencephaly and hemispheric cortical dysplasia. Among these disorders, patients with hemimegalencephaly had the worst post-operative outcome. Regardless of underlying pathology, shorter seizure duration, better seizure control, and higher preoperative developmental quotients (DQ) predicted better postoperative DQ.
In the realm of non-surgical treatment of RE, Bien and colleagues administered tacrolimus to 7 patients with histopathologically proven RE. They compared seizure outcome, neurologic function (NIH Stroke Scale Motor Item score), and progression of cerebral hemiatrophy by MRI planimetry in these patients versus 12 historical controls (RE patients not exposed to tacrolimus). Tacrolimus-treated patients displayed slower progression of hemiparesis and hemiatrophy than controls. Unfortunately, seizure outcome was no different between tacrolimus patients and controls.
RE is a catastrophic neurological disorder of early childhood. The initial presentation is commonly focal seizures, often evolving into epilepsia partialis continua, a form of focal motor status epilepticus. Eventually, RE evolves to include progressive neurological decline, with hemiparesis, and cognitive dysfunction and concordant contralateral cerebral hemiatrophy on neuroimaging. What makes RE unique is that the ultimate decision to treat with UH requires that the disease has progressed to the point that the clinical diagnosis is certain and that the neurologic deficit includes hemiplegia. After all, only the particularly brave (or cocksure) neurologist and neurosurgeon will recommend UH for "probable" RE, or undertake a procedure that will convert a child who has mild or moderate hemiparesis into one with complete hemiplegia.
The current studies address surgical and non-surgical treatments for RE that may help to settle the issue of when to treat. Jonas et al, along with accompanying editorial by Duchowney, indicate that UH is a relatively safe and effective procedure for RE and can render almost 60% of operative patients seizure-free. Of the RE patients we have encountered, seizure-freedom is the most important factor in affording these patients any neurologically meaningful quality of life. The UH study does indicate that diagnosis and referral for surgery is being made in increasingly younger patients, and, although the data are not specifically analyzed as far as age at surgery relative to outcome, the fact that seizure duration was inversely correlated with post-operative developmental quotient suggests that the trend towards earlier referral is a favorable one. Moreover, functional recovery from UH will also be enhanced if performed at a young enough age to take advantage of developmental plasticity.
While treatment with tacrolimus does not significantly alter seizure frequency, as acknowledged by Bien and colleagues, this limited study is very intriguing. Alert is encouraged that new treatment options (ie, a well-established suppressor of T-cell activation) are again being explored that build upon the initial insights regarding the pathophysiology of RE. This particularly involves antibodies directed towards the R3 sub-unit of the glutamate receptor in RE patients (Science. 1994;265:648-651) and cytotoxic T-cells (Ann Neurol. 2002;51:311-318). Prior studies of immunomodulatory therapies (eg, plasmapheresis, intravenous immunlobulin, corticosteroids) founded on the basis that either outcome data were equivocal or the therapy was impractical given the chronic progressive nature of RE vs the short duration of benefit. Perhaps currently approved chemotherapeutic agents might work where other non-surgical treatments have failed. It is certainly worth a randomized controlled trial (even with such a low-incidence disorder) to prove the point.
Certainly, the tacrolimus study is flawed. The most obvious flaw is the study design employing historical controls. Given that this was an open study, and thus more prone to investigator bias, it is all the more disappointing that tacrolimus, unlike UH, led to no improvement in seizure outcome. Since RE patients usually have medically intractable seizures, functional debilitation on an ictal basis is expected to continue regardless of "superior outcome" for neurological function in tacrolimus-treated patients. As has been emphasized many times by Neurology Alert, the call goes forth for earlier diagnosis and treatment of various epileptic conditions. In the case of RE, further exploration of diagnosis and treatment is bound to provide insight into both epileptogenesis and immune-mediated neurological disease. — Andy Dean
Dr. Dean, Assistant Professor of Neurology and Neuroscience, Director of the Epilepsy Monitoring Unit, Department of Neurology, New York Presbyterian Hospital, Cornell Campus, is Assistant Editor of Neurology Alert.