NATIONAL HARBOR, MD—Serial EEG has been confirmed as an accurate biomarker for risk of epilepsy in infants with tuberous sclerosis complex (TSC). In a prospective observational study involving five centers, the positive predictive value of epileptiform discharges for subsequent epilepsy was 100%, according to data presented at the 2015 Child Neurology Society Annual Meeting.
“The average interval between the time that EEG demonstrated epileptiform activity and the onset of the first seizure is between two to three months,” reported E. Martina Bebin, MD, a pediatric neurologist at the University of Alabama, Birmingham. She suggested that this interval is important because it may allow clinicians to pursue strategies designed to prevent or at least delay seizure onset.
E. Martina Bebin, MD
While EEG is currently recommended at the time of TSC diagnosis, these data support the ability of serial EEG to screen for impending epilepsy, Dr. Bebin said. She noted that a recommendation for serial EEG had been issued in Europe through a consensus statement. The work by Dr. Bebin and colleagues, which was characterized as the first multicenter prospective study to establish EEG as a biomarker for epilepsy in TSC, also was published online ahead of print September 25 in Pediatric Neurology.
Heightened Risk
Studies have shown that as much as 80% of patients with TSC eventually develop epilepsy, and onset within the first six months of life is common. In a retrospective series conducted in Europe cited by Dr. Bebin, the majority of patients diagnosed with TSC early in life developed seizures before age 6 months. Many seizures were observed after a period of latent epileptogenesis. Other data linking EEG abnormalities to increased risk of overt seizures led to the multicenter trial to evaluate whether EEG could serve as a biomarker for epilepsy risk.
The investigators enrolled seizure-free children younger than 7 months old with a clinical or genetic diagnosis of TSC. Children previously treated with vigabatrin or inhibitors of the mammalian target of rapamycin were not eligible. Prematurity after at least 32 weeks’ gestation was not an exclusion criterion if there were no complications involving major organs. In addition to the University of Alabama, participating centers included Boston Children’s Hospital, Cincinnati Children’s Hospital, University of California at Los Angeles, and the University of Texas Medical School at Houston.
After enrollment, physical and neurologic evaluations were accompanied by a baseline one-hour EEG. The intent was to monitor brain activity during periods of wakefulness and sleep. Subsequent patient evaluations and EEGs were performed every six weeks for the first six months, every three months for another six months, and then every six months thereafter. The researchers considered more frequent monitoring over the first six months, which is the most common period of epilepsy onset, but the time commitment for families was a concern.
Forty children were enrolled. Dr. Bebin presented interim data on 28 children who had been followed past 12 months of age. Of the remaining children, six children were younger than 12 months, three had not yet had their EEG data fully analyzed, and three were treated with antiepileptic therapies on the basis of seizure activity observed on EEG and therefore excluded from analysis.
Initial Seizures
Clinical seizures developed in 19 children (67.8%) after enrollment. The average age at the time of first seizure was 6.7 months, with a range of 2 months to 20 months. Epileptic spasms were the most common type of initial seizures observed, accounting for 52.6%. In most other cases, patients had focal seizures either alone (26.3%) or in conjunction with epileptic spasms (15.8%).
EEG abnormalities, primarily in the form of epileptiform discharges, preceded the onset of the first seizure in 14 (73.7%) of the 19 infants. These EEG abnormalities were first observed between 1.2 and 9.0 months of age. The median interval between the epileptiform discharges and the first clinical seizure was 1.9 months and the average was 2.8 months. In the remaining five infants (26.3%) of the 19 who developed seizures, no EEG abnormalities were detected. Again, this group also largely developed focal seizures and epileptic spasms, but the only generalized tonic-clonic seizure occurred in this group.
Nine infants remained seizure-free off therapy, and all nine had normal EEGs over the period of evaluation.
As no infant demonstrated abnormal EEG without subsequently developing epilepsy, the positive predictive value was 100%. The negative predictive value, based on the five infants who developed epilepsy without first demonstrating an EEG abnormality, was 64%. Dr. Bebin pointed out that it is possible that abnormal EEG activity was missed in at least some of these infants due to the interval between EEG evaluations. In these patients, seizures developed at between two weeks and two months after the last EEG.