Peter Kellaway

Characterization and classification of neonatal seizures

To characterize and classify neonatal seizures, we studied 349 neonates, using a portable, cribside EEG/ polygraphic/video monitoring system. We recorded 415 clinical seizures in 71 infants; 11 other infants had electrical seizure activity without clinical accompaniments. Each seizure was analyzed in terms of its clinical character and its relationship to the presence of EEG seizure activity. Focal clonic seizures, some forms of myoclonic seizures, and focal tonic seizures were consistently associated with electrical seizure activity. Most “subtle” seizures, all generalized tonic seizures, and some forms of myoclonic seizures were either not associated with EEG seizure activity or had an inconsistent relationship with such activity. Seizures that were consistently and coherently related to focal EEG seizure activity had a high correlation with focal brain lesions and a favorable short-term outcome. Seizures with no relationship or an inconsistent relationship to EEG seizure activity were correlated with diffuse processes such as hypoxic-ischemic encephalopathy and a poor short-term outcome. The clinical and background EEG features of infants whose seizures were not accompanied by EEG seizure activity suggest that these seizures may not be epileptic in character.

Sleep and Epilepsy

Summary: Epileptic mechanisms in the brain are subject to long‐duration, time‐ordered neuromodulatory processes controlled by endogenous oscillators which are responsible for appropriately phased modulation of various normal physiological processes, including the 24‐h sleep/wakefulness cycle and the ultradian 100‐min cycle of rapid eye movement/non‐rapid eye movement sleep. Both focal and generalized types of epileptiform activity in humans are subject to biorhythmic modulation, and the various modulation patterns observed are in accord with a model which explains these patterns as a consequence of the interaction of two endogenous modulatory processes: one with a period of about 24 h, the other with a period of about 100 min. Differences in the phase angle between the two cyclic processes, determined by time of sleep onset, explain the various modulatory patterns observed. The mechanisms involved in the genesis and elaboration of electrical epileptiform activity in animal models are examined in relation to known processes involved in the physiology of sleep, and compared with data derived from long‐term studies of the time distribution of epileptic events in humans. In infantile spasms, clinical seizure activity and the ictal and interictal EEG patterns in relationship to the phases of the sleep cycle, the significant defects in the quality and quantity of sleep in this disorder, and the changes that take place in all of these when seizures are abolished by effective treatment, suggest that pontine mechanisms responsible for the sleep cycle may be involved in the elaboration of infantile spasms and hypsarrhythmia.

Hypsarrhythmia: Variations on the Theme

Summary: Prolonged monitoring studies of patients with infantile spasms have shown that hypsarrhythmia is a highly variable and dynamic electroencephalographic pattern. Variations of the prototypic pattern (modified hypsarrhythmia) include hypsarrhythmia with increased in‐terhemispheric synchronization, asymmetrical hypsarrhythmia, hypsarrhythmia with a consistent focus of abnormal discharge, hypsarrhythmia with episodes of attenuation, and hypsarrhythmia comprising primarily high‐voltage slow activity with little sharp‐wave or spike activity. Marked changes in the hypsarrhythmic pattern usually occur during sleep, chiefly during rapid eye movement sleep, when there is a marked reduction in, or total disappearance of, the hypsarrhythmic pattern. Relative normalization of the hypsarrhythmic pattern can also be seen immediately on arousal and during clusters of infantile spasms. Thus, the specific EEG features seen in a given patient depend on multiple factors, including the duration of the EEG recording, the clinical state of the patient, and the presence of various structural abnormalities of the brain.

Double-blind study of ACTH vs prednisone therapy in infantile spasms.

Twenty-four patients with infantile spasms were entered in a double-blind, placebo-controlled, crossover study to compare the therapeutic effectiveness of ACTH (20 to 30 units/day) with that of prednisone (2 mg/kg/day). Response to therapy was determined by utilizing a comprehensive monitoring system and was defined as a complete cessation of spasms and disappearance of the hypsarrhythmic EEG pattern. A major difference between the effectiveness of ACTH and that of prednisone in stopping the spasms and improving the EEG pattern was not demonstrated. Nine patients responded to ACTH (five initial drug, four crossover), and seven patients responded to prednisone (four initial drug, three crossover). Twelve responded within two weeks of initiation of therapy, and four within six weeks. Therapy was tapered and discontinued immediately after a response was obtained. Five patients had a relapse; four responded rapidly to a second course of therapy. Of the eight patients who failed to respond to hormonal therapy, seven were given clonazepam with no improvement.

Prospective study of outcome of infants with infantile spasms treated during controlled studies of ACTH and prednisone

We report the long-term outcome of 64 infants with infantile spasms, followed prospectively, using controlled treatment schedules and objective techniques (24-hour EEG and video monitoring) to determine response. Average age at follow-up was 50 months. Of the 64 infants, three (5%) died; of the others, 41 (67%) had developmental retardation of 50% or more or an IQ of 50 or less. Eight patients (13%) composed our cryptogenic study group and were so classified on the basis of normal CT scan, normal development prior to onset of infantile spasms, and undetermined cause. These patients had the better outcome; 38% had normal development or were only mildly retarded. Both the responders and nonresponders in our symptomatic group had a poor outcome; only 5% had normal development or mild impairment. Outcome was not significantly influenced by short versus long treatment lag or by response to therapy. Other types of seizures occurred in 34 patients (53%). In summary, the overall prognosis for long-term outcome in these 64 patients with infantile spasms was poor.

Sleep characteristics in infantile spasms

Sleep staging was performed on 32 patients with infantile spasms. All patients demonstrated significantly less total sleep time and lower percentage of REM time than the expected normal values for age. Seventeen were treated with ACTH or prednisone. There was no increase in total sleep time in those who received hormone therapy, whether or not it was successful. However, the percentage of REM time increased significantly in patients who responded to therapy; this increase occurred concurrently with clinical and electroencephalographic improvement. There was no significant change in REM-sleep values in patients who did not respond.

Anterior Temporal Lobectomy and Medically Refractory Temporal Lobe Epilepsy of Childhood

Summary: The evaluation and outcome of 22 patients who had onset of complex partial seizures (CPS) of temporal lobe origin in childhood and subsequently underwent anterior temporal lobectomy are described. All patients showed improved seizure control; 81.8% had a reduction ≥95% in seizure frequency. However, many patients had difficulty adjusting to a seizure‐free life. Psychosocial, behavioral, and educational problems occurred more frequently in patients whose surgery was delayed until adult life. We conclude that attempts should be made early in the course of CPS of childhood to determine whether the seizures are truly intractable to medical management so that surgical intervention can be expedited.

Electrical activity of the isolated cerebral hemisphere and isolated thalamus

Abstract A surgical technique has been devised which permits complete neural isolation of the cortex of an entire cerebral hemisphere while preserving the blood supply and drainage. The status of the arterial circulation and venous drainage of the isolated hemisphere is evaluated at the termination of each experiment by comparison of the intact and isolated sides following aortic injection of a particulate dye. Such preparations show continuous or semicontinuous electrical activity consisting of high voltage aperiodic 0.5–4 cycle/sec oscillations and a superimposed lower voltage 25–70 cycle/sec rhythm. This activity may be interrupted by brief isoelectric episodes which may occur in all regions of the cortex simultaneously. The fast rhythm is essentially similar to that seen in the intact brain, and appears to constitute an autochthonic activity of the superficial neural feltwork of the cortex sustained in the absence of afferent neural influences. The high voltage aperiodic slow activity appears to be a consequence of the surgical procedure, but is probably related to neural isolation rather than to circulatory or mechanical insult. The isoelectric episodes are directly related to the physiological condition of the isolated hemisphere, and any factor which tends to compromise this state increases their incidence and duration. The isolated hemisphere is totally inactive only in the presence of significant hypoxia or ischemia. In contrast, the surgically isolated thalamus exhibits continuous rhythmic 6–9 cycle/sec activity of sinusoidal waveform. Thus, the aperiodic slow activity of isolated cortex is not a necessary consequence of deafferentation per se but reflects some characteristic peculiar to the cortex, perhaps a dependence of certain elements of its neural population upon subcortical influences for the development of an organized rhythmic behavior.

The Electroencephalographic Features of Benign Centrotemporal (Rolandic) Epilepsy of Childhood

There is no experimental model of benign rolandic epilepsy, and the neurophysiologic and molecular mechanisms involved in the generation and elaboration of this type of epilepsy and its unique electroencephalographic features have yet to be determined. These EEG features are constituted by ionic currents generated at a cellular level. They are cortical in origin, and the cortical neurons responsible for their generation must be arranged and interconnected in such a manner as to permit their synchronous or virtually synchronous activation. This capacity for synchronization is inherent in the cellular arrangement and connectivity of the cortical network itself, but also may reflect imposed synaptic influences of thalamic origin. The dominant EEG feature of benign rolandic epilepsy is a focal surface negative spike (Fig. 1) with specific biophysical characteristics. The average duration is -74 ms (leading some authors to insist that it should be referred to as a sharp wave rather than a spike). The average sharpness or degree of curvature of the peak is 0.022 pV/ms/8 ms, indicating a relatively blunt character. The average amplitude is about 160 FV (I), but some individual spikes may exceed 300 pV. The high amplitude, the relatively prolonged duration, and the bluntness of the spike suggest that it is generated

“Contingent Negative Variation” in Rhesus Monkeys: An EEG Sign of a Specific Mental Process

Utilizing long time-constant EEG recording techniques, magnetic tape storage, and electronic averaging, the phenomenon designated the “contingent negative variation” or “expectancy wave” was studied in 3 Rhesus monkeys. Three basic conditioning paradigms were used: escape conditioning with a warning cue, discrimination task with aversive reinforcement following SD, and discrimination with appetitive reinforcement. Trials were presented in blocks of 12 and averaged on line with an Enhancetron electronic averager. In each case, during acquisition trials, a surface negative slow potential gradually developed only following the evoked response to SD and preceding the meaningful stimulus. No shift followed the unreinforced stimulus. During extinction trials, the shift gradually diminished. During discrimination reversal, a negative shift developed following the previously unreinforced stimulus, while the shift following what had been SD gradually diminished. Results demonstrate the appearance of the “contingent negative variation” in these monkeys and support the contention that this shift is a result of cerebral electrical activity during conation.