The article by Bakshi et al. (I) provides excellent clinical and MRI evidence that occipital lobe seizures can be the major manifestation of the reversible posterior leukoencephalopathy syndrome (RPLS). The authors mention, however, that ictal EEG recordings were unavailable in their patients for definitive proof. Indeed, although chronic occipital epileptic EEG abnormalities have been reported in two patients as a complication of eclampsia (2), and lateralized complex partial status epilepticus was demonstrated in one patient treated with cyclosporine (3), the many reports of RPLS have not yet captured the expected reversible occipital epileptic abnormalities.A case of RPLS with occipital lobe seizures was seen at this hospital in a 50-year-old man on cyclosporine postcardiac transplantation. Three weeks after transplantation, the patient became acutely confused and then had a generalized seizure. No further convulsions occurred after institution of phenytoin therapy, though confusion and intermittent agitation continued. A CT scan done on the day of the generalized seizure showed bilateral occipital white matter hypodensities with extension to involve the occipital grey matter on the right side (Fig. 1A). Focal white matter hypodensities were also noted in the right external capsule and right corona radiata, as well as in the white matter of both cerebellar hemispheres. An EEG 1 day later (Fig. 1B) showed nearly continuous periodic lateralized epileptiform discharges (PLEDs) with a regional predominance over the right occipital lobe and, in addition, the occurrence of two electrographic seizures with ictal onset over the right occipital region, both spreading to involve the entire right posterior quadrant, with reflection over the left occipital region, prior to ictal offset. Each seizure lasted 2-3 min, during which time the patient became increasingly confused and agitated.The presumed etiology of the clinical presentation was cyclosporine-induced neurotoxicity plus or minus acute hypertension. Nitroprusside was required for 3 days after the onset of neurological symptoms to control acutely elevated blood pressure in the range of 180/100. The cyclosporine level was 495 kg/L 2 days before the generalized seizure and the daily dose had been reduced at that time. Cyclosporine was discontinued with the onset of neurological symptomatology and replaced by mycophenalate mofetil as an immunosuppressant. The patient made a gradual, complete recovery from his confusional state over the following 10 days and an MRI scan done 11 days after the generalized seizure was normal. An EEG performed 3 weeks after the first (shortly after reinstitution of low-dose cyclosporine) showed occasional interictal epileptiform potentials over the right occipital pole.Two months later, the patient was doing well and tolerating cyclosporine with no recurrence of seizures. An EEG showed only a minimal residual nonepileptiform disturbance over the right occipital region.Thus, as presumed by Bakshi et al., RPLS in the acute phase may be a...

A central nystagmogenic area exists in humans that appears to be homologous to the nucleus of the optic tract, a region described in nonhuman primates to play a role in the generation of optokinetic nystagmus.

(3-7), and that the vast majority are published within 4-5 years (5,6). It is likely similar in neurology and neurology subspecialty meetings. Although it may be acceptable for abstracts to be cited in "ordinary" journal articles, the appropriateness of including abstracts in a "practice advisory" and other types of authoritative guidelines could be questioned. REPLY To the Editor:The author raises a valid point, that information presented in abstract form may not have undergone the same type of peer review process as published articles. He goes on to question the appropriateness of using such data for "practice advisories and other authoritative guidelines." However, the process by which evidence-based practice advisories are created requires a complete evaluation of the available literature. A systematic review should describe the highest quality evidence available; this may include data in abstract form (1). Publication bias against certain kinds of evidence may exist (e.g., negative studies are less likely to be published).The data in question was described as class I11 because of its preliminary nature. No other data was available on these points, so abstract information was cited. The source of this data was explicitly stated. This type of data did not lead to a strong statement of effectiveness in the practice advisory. If further information-in the form of well-controlled trials-were to become available, the strength of recommendation would increase accordingly. To the Editor: Bautista et al. (1) clearly demonstrate that poor surgical outcome in patients with neocortical epilepsy is correlated with interictal epileptiform abnormalities outside the area of surgical resection. Our recent study of the predictive value of acute electrocorticography (ECoG) in 60 patients with frontal lobe neocortical epilepsy resulted