Outcome of Absence Epilepsy With Onset at 8-11 Years of Age: Watershed Ages When Syndromes Overlap

Abstract

Introduction: Absence seizures occur in various epilepsy syndromes, including childhood and juvenile absence epilepsy and juvenile myoclonic epilepsy. When children present with absence seizures at ages when syndromes overlap, initial syndrome designation is not always possible, making early prognostication challenging. For these children, the study objective is to determine clinical and initial electroencephalograph (EEG) findings to predict the development of generalized tonic-clonic seizures, which is a factor that affects outcome. Methods: Children with new-onset absence seizures between 8 and 11 years of age with at least 5 years of follow-up data were studied through the review of medical records and initial EEG tracings. Results: Ninety-eight patients were included in the study. The median age of absence seizure onset was 9 years (interquartile range [IQR] = 8.00, 10.00) and follow-up was 15 years (IQR = 13.00, 18.00). Forty-six percent developed generalized tonic-clonic seizures and 20% developed myoclonic seizures. On multiple regression analysis, a history of myoclonic seizures, anxiety, as well as bifrontal slowing and mild background slowing on initial EEG (P < .05) were associated with generalized tonic-clonic seizures. Although not statistically significant, a shorter duration of shortest EEG burst on baseline EEG was also associated with generalized tonic-clonic seizures. Conclusion: On initial EEG, bifrontal and background slowing and myoclonic seizures and anxiety are associated with developing generalized tonic-clonic seizures, which is of prognostic significance when early syndrome designation is difficult.

Absence seizures are characterized by brief impairments in awareness that arise and abate suddenly without post-ictal changes. ¹ Absence seizures most commonly occur in childhood, with an incidence of 6 to 8 per 100 000 in children up to 15 years of age. ² They appear in different epilepsy syndromes, predominantly childhood absence epilepsy and juvenile absence epilepsy, and less commonly juvenile myoclonic epilepsy. In these different syndromes, the timing when absence and generalized tonic-clonic seizures arise vary, but overlap. In childhood absence epilepsy, absence seizures occur from 2 to 8 years, with a peak of 5 years.^3,4 Patients often have good outcome, although generalized tonic-clonic seizures may occur in adolescence.^5,6 In juvenile absence epilepsy, absence seizures occur from 7 to 16 years, with a peak of 10-12 years of age. The absence seizures are usually less frequent than childhood absence epilepsy and 80% of patients will have generalized tonic-clonic seizures, peaking at 17 years, and infrequent myoclonic seizures. ⁷ In juvenile myoclonic epilepsy, approximately a third have absence seizures with onset between 5 and 16 years of age. Generalized tonic-clonic seizures and the hallmark myoclonic seizures occur 1-6 years later. ⁸ Juvenile absence epilepsy and juvenile myoclonic epilepsy patients usually require long-term ASM. Therefore, identification of a syndrome is of utmost importance for prognostication.

The EEGs of childhood absence epilepsy and juvenile absence epilepsy can also be difficult to distinguish at presentation, although nuances exist. In childhood absence epilepsy, the EEG is characterized by bilateral, synchronous, symmetrical 3-Hz (2.7-4 Hz) generalized spike wave discharges on a normal background. In juvenile absence epilepsy, the frequency of spike-wave discharges may be faster (3.5-4 Hz). ³ Occipital intermittent rhythmic delta activity is reported in 32% of children with childhood absence epilepsy, but can occur in children with other forms of genetic generalized epilepsy. ⁹ The incidence of photoparoxysmal response in childhood absence epilepsy and juvenile absence epilepsy vary among studies, with some reporting it to be present less, ¹⁰ similar, ¹¹ or more in juvenile absence epilepsy than childhood absence epilepsy. ¹² The EEG of juvenile myoclonic epilepsy consists of generalized 3.5- to 6-Hz polyspike wave discharges, ¹³ with a third demonstrating photoparoxysmal response.^10,14 However, these findings might not be apparent at onset of the absence seizures.

Therefore, there is considerable electro-clinical overlap between the syndromes and the cutoff ages of syndrome onset and offset remains controversial. Stratification into a particular syndrome is not always possible at the time of diagnosis and accurate classification may require additional evidence (including about the outcomes) accumulated over time to determine the syndrome. Specific counseling and determination of the risk of generalized tonic-clonic seizures has clinical relevance, as syndromes with associated with generalized tonic-clonic seizures often require long-term antiseizure medications. In addition, determining the risk of generalized tonic-clonic seizures is important, as individuals with generalized tonic-clonic seizures may face distinct challenges than those with absence seizures alone, including risk of seizure-related injury and of sudden unexpected death in epilepsy. The study objective is to determine clinical and initial EEG findings to predict the development of generalized tonic-clonic seizures in children presenting with absence seizures in an age range where absence epilepsy syndromes overlap.

Methods

Patient Inclusion

At BC Children's Hospital, since 1995, EEG and clinical data obtained by a technologist at each visit are entered into a database. In this retrospective study, the database was queried from 2000 to 2015 for patients 8-11 years of age with a first-time EEG that demonstrated new-onset absence seizures (280 were found). This age group was selected, as various epilepsy syndromes with absence seizures can overlap at this age. Patients were included in the study if there was no prior history of generalized tonic-clonic seizures, normal neurologic examination, no other brain pathology, and at least 5 years of clinical follow-up data available in the medical records. Included patients were not on any antiseizure medications at the time of the initial EEG. For all study patients, the diagnosis of absence seizures was confirmed by review of medical records and EEGs. Patients with eyelid myoclonia with absences and patients with an absence phenotype, but with focal epilepsy, were excluded, as they carry a different prognosis. For all included patients, the initial EEG at the time of diagnosis had to be available for investigators to review. On EEG, patients with moderate to severe background slowing or in a comatose state were excluded.

Clinical Features

Detailed chart reviews of the entire medical record were performed on the patients identified from the database. The diagnosis of absence epilepsy was made according to established criteria.^3,4 Clinical data, including age of seizure onset, sex, antiseizure medications tried, seizure frequency, development, school performance, and formally diagnosed epilepsy comorbidities, such as attention-deficit hyperactivity disorder (ADHD), autism spectrum disorder, and anxiety, were documented. “Difficulties in school performance” was defined as patients who required additional school assistance or a modified program to meet their learning potential. Standardized neuropsychological evaluations were not available for most patients, and chart reviews for comorbidities were predominantly based on reports from parents and clinicians. Specific features of seizure semiology, including the history of generalized tonic-clonic seizure and myoclonic seizures, were documented.

EEG Analysis

In this study, the first EEG recording performed at the time of the early diagnostic assessment, when patients were not yet on antiseizure medications, was evaluated. EEGs were recorded for 25-45 minutes with Biologic or Natus machines, using the international 10-20 system with a 256-Hz sampling rate and 0.5- to 70-Hz filters. The presence of mild background slowing, focal spikes, photoparoxysmal response, occipital intermittent rhythmic delta activity, polyspike wave discharges, and bifrontal slowing were documented. These findings are routinely coded in the EEG database. The primary investigators (AD, JC) reviewed all the initial EEG recordings to confirm the EEG findings coded in the database. In addition, the number and duration of seizures were manually counted, and the shortest and longest burst intervals were documented by reviewing the EEGs.

A normal EEG background was defined as evidence of a posterior dominant rhythm between 8 and 13 Hz during relaxed wakefulness. Polyspike wave discharges were defined as a sequence of multiple spikes that may be followed by a slow wave. They typically occurred in the form of high-amplitude rhythmic bursts with synchronized and generalized distribution. Photoparoxysmal response was defined as reproducible epileptiform discharges in response to photic stimulation that may or may not be self-sustaining. Occipital intermittent rhythmic delta activity was characterized by symmetrical or asymmetrical bursts of rhythmic, sinusoidal, 3-Hz, delta activity over the occipital region. ⁹ Bifrontal slowing was defined as paroxysmal slowing in the bilateral midfrontal regions (Fz, F3, F4) in the awake state, predominantly monophasic single slow waves of 0.25- to 0.5-second duration (Figure 1). This did not occur during hyperventilation. This is distinct from hypnagogic hypersynchrony (HH): a normal variant in drowsiness characterized by intermittent or sustained rhythmic slowing maximal in the central region ranging from 3- to 5-Hz activity.

Figure 1.

Examples of bifrontal slowing. Recording sensitivity settings were all 30 µV/mm. The distance between the bold vertical lines represents 1 second.

All bursts of GSW lasting 3 seconds or longer, with or without clinical signs, were considered seizures. Seizure onset was defined as the time of the first spike of a well-formed GSW complex, and seizure offset was defined as the end of the slow wave after the last spike. GSW bursts lasting less than 3 seconds were not considered seizures, unless a clinical correlate was noted.

Outcome Measures and Statistical Analysis

The primary outcome measure of the study was the development of generalized tonic-clonic seizures. All patients had at least 5 years of follow-up data. Patients were divided into two groups, based on the presence or absence of generalized tonic-clonic seizures, and clinical and EEG features were compared. Statistical analysis was performed using R statistical software (R Core team 2022, Vienna, Austria.). Data were summarized through descriptive statistics including medians and interquartile ranges (IQRs) for continuous variables and frequencies and percentages for categorical variables. Group comparisons for patients with and without generalized tonic-clonic seizures were performed with a t test for continuous variables and a chi-square test for categorical variables (Table 1). A univariate logistic regression model for the dichotomized outcomes of interest was fit to each clinically relevant risk factor, and odds ratios, 95% confidence intervals, and P values were reported. A multiple logistic regression model, including all risk factors, listed in Table 2, was also fit and the adjusted ORs, were reported to assess the robustness of the univariate results. The risk factors or electro-clinical variables for this model were selected based on the univariate analysis and the literature review (Table 2).

Table 1.

Clinical and EEG Features of Patients With and Without GTCs.^a

Characteristics	GTC present (n = 45)	GTC absent (n = 53)	Significant differences
Clinical features
Sex
Female	25 (55.6)	31 (58.5)
Male	22 (41.5)	20 (44.4)
Age of onset of absence seizures (y)	9.00 (8.00, 10.00)	9.00 (8.00, 10.00)
Age of onset of GTC (y)	10.00 (9.00, 12.00)	N/A
History of myoclonic seizures at diagnosis	16 (36.4)	4 (7.5)	<.001
Family history of epilepsy	12 (26.7)	15 (28.3)
Age at last follow-up, y	17 (2.14)	14 (2.74)	<.001
Epilepsy comorbidities at presentation
Developmental delay	6 (13.3)	6 (11.3)
ASD	4 (8.9)	3 (5.7)
School performance difficulties	5 (11.1)	8 (15.1)
ADHD	6 (13.3)	6 (11.3)
Mood disorder	6 (13.3)	6 (11.3)
Anxiety disorder	13 (28.9)	4 (7.5)	<.012
Initial EEG findings
Polyspike wave discharges	3 (6.7)	5 (9.4)
OIRDA	6 (13.3)	7 (13.2)
Focal epileptiform discharges	7 (15.6)	7 (13.2)
Photoparoxysmal response	7 (15.6)	5 (9.4)
Mild background slowing	8 (17.8)	2 (3.8)	<.040
Bifrontal slowing	12 (26.7)	1 (1.9)	<.001
Shortest seizure burst (s)	4.00 (3.00, 5.75)	5.00 (3.88, 7.50)
Longest seizure burst (s)	8.50 (6.00, 14.75)	10.50 (5.88, 19.00)
Mean seizure duration (s)	5.85 (4.35, 9.25)	7.00 (5.05, 11.35)

Abbreviations: ADHD, attention-deficit hyperactivity disorder; ASD, autism spectrum disorder; ASM, antiseizure medication; GTC, generalized tonic-clonic seizure; ORIDA, occipital intermittent rhythmic delta activity.

Numbers and percentages are provided for categorical variables. Medians and interquartile ranges are provided for continuous variables and frequencies.

Table 2.

Multiple Regression Analysis: Clinical and EEG Features Associated With Developing GTCs.^a

Electro-clinical Feature	Unadjusted OR (95% CI); P value	Adjusted OR (95% CI); P value
Intercept	NA	0.6 (0.18, 1.9); .401
Anxiety disorder	5 (1.6, 19); .0091	4.6 (1.3, 20); .027
Myoclonic jerks	7 (2.3, 26); .00135	4.1 (1.1, 18); .045
Bifrontal slowing	19 (3.5, 350); .00575	17 (2.1, 390); .0245
Mild background slowing	5.5 (1.3, 38); .0372	6.5 (1.2, 52); .0423
Duration of shortest burst (s)	0.89 (0.77, 1); .0795	0.87 (0.72, 1); .105
OIRDA	1 (0.3, 3.3); .985	1.6 (0.38, 6.7); .509
Photoparoxysmal response	1.8 (0.52, 6.4); .361	1.8 (0.4, 8.5); .436
Polyspike wave discharges	0.69 (0.13, 3); .62	1.2 (0.14, 7.1); .86
Focal epileptiform discharges	1.2 (0.38, 3.8); .741	0.56 (0.094, 2.7); .496

Abbreviations: GTC, generalized tonic-clonic seizure; NA, not applicable; OIRDA, occipital intermittent rhythmic delta activity.

^a The selection of electro-clinical features for the multiple logistic regression model were based on the univariate analysis and review of the literature. Features highlighted in gray are statistically significant.

Results

Ninety-eight patients with new-onset absence seizures between the ages of 8 and 11 years with five-year follow-up data met inclusion criteria (Figure 2). More than half of the patients (57.1%) were female. The median (IQR) age in years of onset of absence seizures was 9 (8.00, 10.00) and of last follow-up was 15.00 (13.00, 18.00). Twenty-seven (28.6%) had a family history of epilepsy. Approximately half (45 patients; 45.9%), experienced generalized tonic-clonic seizures and 20 (20.4%) had a history of myoclonic seizures. Classifying patients into epilepsy syndromes were often challenging for the primary clinicians at diagnosis. For example, by last follow-up, 9 patients were classified with absence epilepsy without specified syndrome, and 12 patients were initially diagnosed with childhood absence epilepsy, but the diagnosis was later changed to juvenile absence epilepsy. Five patients had juvenile myoclonic epilepsy, although that diagnosis was not made at initial presentation. Clinical and EEG results are summarized in Table 1.

Figure 2.

Flow diagram of rationale for patients included and excluded from the study. Each box denotes number of study patients.

Generalized Tonic-Clonic Seizures

In this cohort, 45 patients (45.9%) experienced generalized tonic-clonic seizures. There was no difference in median age of seizure onset, sex, or family history of seizures between patients with and without generalized tonic-clonic seizures. Regardless of developing generalized tonic-clonic seizures, the median age (IQR) of onset of absence seizures was 10 (9, 12). The majority (62.2%) developed their first generalized tonic-clonic seizure within 1 year from the onset of absence seizures.

Comorbidities

Regarding epilepsy comorbidities, 7 (7.1%) had autism spectrum disorder, 13 (13.3%) had school performance difficulties, 12 (12.2%) had ADHD, 12 (12.2%) had a mood disorder, and 17 (17.3%) had an anxiety disorder. In patients with or without generalized tonic-clonic seizures, there was no difference in the incidence of developmental delay, autism spectrum disorder, school performance difficulties, ADHD, or mood disorders. Notably, patients with generalized tonic-clonic seizures had a significantly higher incidence of anxiety disorders (P < .006).

EEG

Polyspike wave discharges were present in 8 (8.2%), occipital intermittent rhythmic delta activity in 13 (13.3%), focal spikes in 14 (14.3%), photosensitivity in 12 (12.3%), mild background slowing in 10 (10.2%), and bifrontal slowing in 13 (13.3%). There was no difference between patients with or without generalized tonic-clonic seizures in regard to the presence of polyspike wave discharges, focal spikes, photoparoxysmal, or occipital intermittent rhythmic delta activity. On multiple logistic regression analysis, a history of myoclonic jerks (adjusted OR 4.1 [95% CI 1.1, 18]; P < .045), bifrontal slowing (17 [2.1, 390]; .0245), abnormal background (6.5 [1.2, 52]; .0423), and anxiety (4.6 [1.3, 20]; .027) were associated with developing generalized tonic-clonic seizures.

On the initial EEG, there was no evidence of a difference in the number of seizures, seizure length, shortest burst, and longest burst between patients with and without generalized tonic-clonic seizures. However, there was a trend that a shorter duration of the shortest burst was associated with generalized tonic-clonic seizures (adjusted OR 0.87 [95% CI 0.72, 1]; P < .105). For patients with generalized tonic-clonic seizures, the mean shortest burst was 4.84 seconds, whereas it was a mean of 6.77 seconds without generalized tonic-clonic seizures.

Discussion

Absence seizures occur in various epilepsy syndromes, including childhood and juvenile absence epilepsy and juvenile myoclonic epilepsy. When children present with absence seizures in the watershed ages of various epilepsy syndromes, initial syndrome designation is not always possible, making early prognostication challenging. In this study, we determined that at the time of presentation of absence seizures, bifrontal and background slowing on EEG, myoclonic seizures, and anxiety are associated with developing generalized tonic-clonic seizures, which is of prognostic significance when early syndrome designation is difficult.

Significance of Generalized Tonic-Clonic Seizures

Approximately half of our cohort developed generalized tonic-clonic seizures. It is well established in the literature that the development of generalized tonic-clonic seizures has prognostic significance. For example, a meta-analysis of 23 study cohorts of patients with absence epilepsy (not stratified by syndrome) found that 78% of patients with absence seizures alone became seizure free, but only 35% of patients with absence seizures and generalized tonic-clonic seizures did so. ¹⁵ Within specific absence epilepsy syndromes, generalized tonic-clonic seizures have also been associated with less favorable prognosis and less likelihood of remission. A juvenile absence epilepsy study from Israel found that the outcome of patients with generalized tonic-clonic seizures was poorer than without generalized tonic-clonic seizures; seizure freedom was observed in 37.5% and 55.5%, respectively. ¹⁶ In a childhood absence epilepsy study, the development of generalized tonic-clonic seizures or myoclonic seizures during antiseizure medication treatment was ominous, predicting both lack of remission of childhood absence epilepsy and progression to juvenile myoclonic epilepsy. ⁵ Of note, in our cohort, myoclonic seizures were also associated with developing generalized tonic-clonic seizures. In one study, in addition to generalized tonic-clonic seizures, female gender and daily absence seizures at onset had a worse prognosis. ¹⁷ In our study, no difference was noted in relation to gender. Initial absence seizure frequency was difficult to ascertain in our retrospective study, although there was no statistically significant difference in the number of absence seizures on the initial EEG in those who did and did not develop generalized tonic-clonic seizures.

In our study, we included patients 8-11 years of age, when syndromes overlap. Others have also observed difficulties of syndrome designation at diagnosis in children presenting in intermediate age groups. In an absence epilepsy study, Trinka described 35 patients (22%) who could not be classified as childhood absence epilepsy or juvenile absence epilepsy (“overlap group”). ¹⁸ Almost half of the “overlap group” went on to develop generalized tonic-clonic seizures and 17% evolved to juvenile myoclonic epilepsy. The development of generalized tonic-clonic seizures was important in predicting seizure remission. Similarly in our study, nearly half, 46%, of our patients developed generalized tonic-clonic seizures. However, in contrast, only 5% developed juvenile myoclonic epilepsy, which may be related to the fact we had a shorter follow-up duration of 5 years, as opposed to their 25.8 years. Our study cohort differs from the above-mentioned overlapping group, as we included all patients presenting with absence seizures in the 8- to 11-year age group, as opposed to defining syndromes by strict cutoff ages and seizure frequency.

Comorbidities

We observed that a history of anxiety, at the time of the initial EEG, was significantly associated with a history of generalized tonic-clonic seizures. Ascertainment of the time of onset of anxiety disorders, whether prior to or around the time of onset of absence seizures, was beyond the scope of the study. In our cohort, all patients had anxiety prior to the onset of generalized tonic-clonic seizures. Other studies have observed that the presence of generalized tonic-clonic seizures, in itself, may lead to increased anxiety, including anticipatory anxiety, seizure phobia, and epileptic social phobia. ¹⁹ Furthermore, compared to other seizure types, generalized tonic-clonic seizures may be disproportionately related to increased morbidity. Sheikh et al observed that their cohort of patients with generalized tonic-clonic seizures had significantly higher levels of depression and anxiety and a lower quality of life compared with those who did not have generalized tonic-clonic seizures, even if the seizure frequency was less than those without generalized tonic-clonic seizures. ²⁰ It would be of interest to know if these psychiatric comorbidities preceded the onset of generalized tonic-clonic seizures, as observed in our study.

EEG

On the initial EEG, the presence of bifrontal slowing and mild background slowing were significantly associated with the development of generalized tonic-clonic seizures. Although not statistically significant, a shorter duration of shortest EEG burst on baseline EEG was also associated with generalized tonic-clonic seizures. EEG findings have previously been suggested as potential discriminators for outcome of absence epilepsy. Unlike our study, a small study comparing childhood absence epilepsy and juvenile absence epilepsy noted a tendency for EEG polyspike wave discharges to correlate with failure to achieve seizure control on antiseizure medication. ²¹ We noted a significant association with background slowing and developing generalized tonic-clonic seizures. Background slowing may reflect subtle underlying cerebral pathology and has predicted progression to generalized tonic-clonic seizures and lack of remission of typical absence epilepsy in other studies as well.^5,22,23

Focal epileptiform discharges were observed in 14% of our cohort and not associated with generalized tonic-clonic seizures. Conversely, Olsson found that focal abnormalities did predict development of generalized tonic-clonic seizures. ² Another adult study of patients with genetic generalized epilepsies, including absence epilepsy, observed that the ratio of seizure freedom was lower and the psychiatric disorders were significantly higher in patients with EEG focal abnormalities. ²⁴ ORIDA is considered a good prognostic factor in absence epilepsy, occurring predominantly in younger seizure patients.^2,25 In our study, patients were older (median age 9 years) and we only looked at the initial EEG, which may be why we observed occipital intermittent rhythmic delta activity in only 13% of our cohort and not as a prognostic factor. A large prospective study found that a shorter shortest burst duration was an early predictor of generalized tonic-clonic seizures in childhood absence epilepsy, possibly superimposing the characteristics seen in patients with juvenile absence epilepsy. Although not statistically significant in our study, conceivably related to our smaller sample size, we also noticed a trend that a shorter duration of shortest EEG burst on baseline EEG was associated with generalized tonic-clonic seizures.

In our study, bifrontal slowing on initial EEG was associated with generalized tonic-clonic seizures. The reason for this finding is not fully understood. In absence epilepsy, arousal signals trigger thalamic activation and focal or bilateral synchronous spikes in a regionally or multiregionally electrically hyperexcitable cortex. ²⁶ EEG recordings in patients with absence seizures are known to show a frontal predisposition,^27,28 with a frontal amplitude maximum of epileptiform discharges and seizure onset with unilateral frontal spikes in a third of patients with shifting lateralization.^27,29 Altered resting-state functional connectivity in the orbitofrontal cortex in childhood absence epilepsy is described. ³⁰ Combined MEG/EEG recordings demonstrated initial involvement of frontal areas, often combined with activity in the peri-insular and subcortical/thalamic areas during bilateral spikes discharges. ³¹ EEG-fMRI studies have demonstrated variable BOLD signal changes in the frontal cortex in patients with absence epilepsy. ³²

Differences in the activity of seizure networks in patients with absence seizures with and without generalized tonic-clonic seizures may contribute to why we noticed more bifrontal slowing in those with generalized tonic-clonic seizures. An EEG-fMRI absence study identified 2 groups of patients: those with and without a positive cortical BOLD signal change in the dorsolateral prefrontal cortex. Patients with this finding experienced seizures other than absence seizures and had a higher incidence of ongoing seizures than those without this finding, suggesting it could be a predictive marker for absence seizure outcome and severity. ³³

Limitations of our study include retrospective data collection, where the collection of clinical data may involve methodological difficulties, as there was no standard data collection form. In our study, patients had at least 5 years of follow-up. However, longer follow-up into adulthood would be optimal as remission rates vary with follow-up. Variations among studies regarding predictive EEG features and outcome could be related to the fact that patient populations of these studies are heterogeneous, with possible genetic differences, and that the influence of potential confounding factors such as age, arousal state, and selection bias with different definitions of the syndromes in the investigated populations. A strength of the study is that we have systematically coded EEG findings, including polyspike wave discharges and bifrontal slowing in our laboratory for 27 years in a consistent manner, which is not widespread practice. Multicenter prospective studies with larger cohorts, advanced neuroimaging techniques, and standardized neuropsychological and genetic testing are needed to further delineate this age group.

Conclusion

Childhood absence epilepsy, juvenile absence epilepsy, and juvenile myoclonic epilepsy may be regarded as age-dependent variants of a biological continuum. In patients experiencing absence seizures in age groups where syndromes overlap, initial syndrome designation is not always possible. In this group, the development of generalized tonic-clonic seizures is an important prognostic factor. Approximately half of this cohort developed generalized tonic-clonic seizure. A history of myoclonic seizures, anxiety, and EEG findings of intermittent bifrontal delta and mild background slowing on initial EEG was associated with generalized tonic-clonic seizures. We also noted a trend of a shorter duration of the shortest EEG burst on baseline EEG also being associated with generalized tonic-clonic seizures. Therefore, in this cohort, certain electro-clinical findings may be useful to prognosticate and counsel families. In addition, regular screening for epilepsy comorbidities is of utmost importance, including anxiety in patients with generalized tonic-clonic seizures.