Case Study Two:
Case Report:
A 9-year-old girl with intractable absence seizures and a possible focal lesion
Douglas R. Nordli, Jr., M.D.
Director of Children's Epilepsy
The Lorna S. and James P. Langdon Chair of Pediatric Epilepsy
Children's Memorial Hospital, Chicago, IL
Ellen G. is a 9-year-old girl who initially presented with what appeared to be typical absence seizures. During the previous several months she had repeated stereotyped attacks characterized by staring and unresponsiveness, which lasted several seconds. She was amnestic for test items presented during the seizures. The episodes usually lacked any motor manifestations, although her father, after repeated interviews, believed that on a few occasions he may have seen some subtle deviation of her eyes and perhaps subtle tonic posturing of one arm.
Early development was normal and there were no risk factors for epilepsy. She was in a 4th-grade class with 32 children and had a history of mild learning disabilities going back to the first or second grade. Her parents complained of some difficulties with her current school performance.
General physical and detailed neurological examinations were normal. Hyperventilation for 1 minute and 45 seconds induced a typical absence seizure lasting approximately 10 seconds.
Routine EEGs, obtained with her awake and with hyperventilation, revealed highly stereotyped 3Hz spike-wave discharges with a generalized distribution. The electrographic findings were interpreted as being consistent with childhood absence epilepsy.
MRI revealed a small hyperintense signal on T2-weighted scans in the left globus pallidus. The clinical significance of this abnormality was uncertain, and repeat imaging six months later showed no change in the lesion.
She was treated with divalproex sodium with doses as high as 45 mg/kg/day. Her parents noted a change in her personality during this time, but it was not clear whether it was related to the medication. Attacks continued, so the medication was withdrawn. Ethosuximide was begun, but low doses caused stomach upset and her parents requested a change of medication.
Video EEG monitoring revealed a normal background with a well-developed and modulated 9 Hz alpha rhythm. The most conspicuous feature of the recording was the presence of numerous bursts of generalized spike-wave discharges with a repetition rate of 3Hz (Figure 1). These discharges were more prominent in the frontal derivations and sometimes had polyphasic components. The discharges lasted 5-10 seconds and were accompanied by unresponsiveness and amnesia for test items. There were no motor manifestations or automatisms associated with the discharges. Careful scrutiny of the discharges revealed other interesting features. By altering the "paper speed" of the display to 60 cm/sec and adding a channel comparing F3 to F4, one could appreciate a subtle, left-hemisphere lead-in to many of the discharges (Figure 2).
We concluded that these findings were consistent with a form of localization-related epilepsy with rapid secondary generalization, mimicking typical absence epilepsy. Based on these findings, gabapentin was begun, but seizures worsened. She was switched to acetazolamide, which resulted a remarkable reduction of the seizures. Alternative medications currently under consideration are lamotrigine and zonisamide.
Discussion
This patient presented with what appeared to be absence seizures, which proved refractory to classic absence medications. The question then becomes: is this a case of refractory absence epilepsy, or was the initial diagnosis incorrect? There are new agents available that may have some efficacy in refractory absence seizures. There are reports from Japan, for example, on refractory absence seizures with focal features that were treated with zonisamide with good results.
There are rare genetic causes of intractable absence that might be considered, such as glucose-transported protein defect and a condition involving hyperaminemia and hypoglycemia. Another possibility is that the abnormality in the left globus pallidus seen on the MRI may be related to a structural anomaly somewhere in the cortex that is the source of the seizures.
The lesson here is that, in patients who don't respond to appropriate therapy for their seizure type, one should consider alternative diagnoses. This patient was treated unsuccessfully with drugs for generalized seizures that work in the vast majority of patients with absence; this was a cue for consideration of a diagnosis other than absence.
Based on the rather subtle EEG findings, it was concluded that this patient has complex partial seizures with rapid secondary generalization. She remains on acetazolamide at only 10 mg/kg/day and is remarkably improved, with a marked reduction in seizure frequency. It should be noted, however, that this is a short-term follow up, only a couple of weeks after initiating therapy. Acetazolamide is notorious for having short-lived efficacy in some patients. Some patients have done well long term on acetazolamide, but often it remains effective for only a brief period. It is possible that Ellen will require some other agent long term, perhaps something in combination with divalproex sodium or an alternative AED like lamotrigine or zonisamide. Once she is clinically in remission, objective data will be obtained on seizure frequency.
A noteworthy feature of this case is the extremely rapid secondary generalization. It was so rapid, in fact, it could only be appreciated at the equivalent of a paper speed of 60 cm/sec and by using a technique with the EEG electrodes that originated in a report in the Journal of Clinical Neurophysiology. One of the frontal electrodes, for example F3, is plugged into one of the channels, for example Grid 1. F4 is then plugged into the other channel, for example Grid 2. The signal is then passed through the EEG machine. Since the EEG is basically a differential amplifier, one then examines the differences between the signals. If the signal were hitting the frontal regions precisely on time, completely synchronized and at the same amplitude, then the difference between the F3 and F4 signals would be zero; there would be a straight line. In other words, if the seizure were a pure, generalized discharge, it should produce a straight line under these conditions. If the amplitude is higher in one channel or the other, the line would be skewed slightly in the direction of the higher amplitude. But if the signal is starting in one place and secondarily generalizing, there would be a more complex tracing in the combined F3 to F4 channel. The latter is, in fact, what was seen on this patient's EEG: a wave form that looked more complex than the signal at either channel independently. This was very suggestive of a partial seizure with very rapid secondary generalization.
What about Ellen's prognosis? Because a focal lesion appears to be reponsible for the seizures, she may be a candidate for surgery. A repeat MRI was done; we looked very carefully at the region above the globus pallidus to see if any abnormality could be visualized in the adjacent cortex or overlying cortex, but unfortunately none was apparent. Nevertheless, this patient is a potential candidate for epilepsy surgery, particularly if the spells continue to be difficult to control with medications and if we can ultimately visualize the lesion on imaging.
|
