Epilepsia, 32( 1):69-76. 1991
Raven Press, Ltd., New York 0 International League Against Epilepsy
Postnatal Epilepsy After EEG-Confirmed Neonatal Seizures Robert Ryan Clancy and Agustin Legido The Division of Neurology, The Children’s Hospital of Philadelphia, and the Departments of Neurology and Pediatrics, University of Pennsylvania School of Medicine, Philadelphia, Pennsylvania, U . S . A .
Summary: We examined infants whose neonatal seizures were confirmed by randomly recorded ictal EEG tracings to determine the types and frequency of postnatal epilepsy (PNE)-unprovoked, recurring postnatal seizures. Perinatal and postnatal clinical and EEG variables were also examined for their relevance to PNE. Forty infants with EEG-documented neonatal seizures of diverse etiologies were studied. The 27 survivors were followed for a mean of 31 months. PNE developed in 56% (15 of 27) of the cohort. The first seizure appeared at a meancorrected age of 12.7 months and occurred despite ongoing antiepileptic medication in 60% (9 of 15) of the group. Seizures were classified as infantile spasms or minor motor (7 patients), complex partial (4 patients), or generalized tonic-clonic (4 patients). Perinatal variables that significantly correlated with PNE included the presence of coma but not the age at seizure onset, the estimated gestational age, or Apgar scores. PNE occurred in 68% (13 of 19) of patients with moderately or markedly abnormal EEG backgrounds but in only 25% (2 of 8) without (p = 0.035). There was a strong trend for PNE to develop in
patients with >I0 electrographic seizures per hour but in only 45% (9 of 20) of infants with fewer seizures (p = 0.058). Several postnatal variables were significantly related to PNE-the presence of cerebral palsy (CP), mental retardation (MR), CP with MR, and follow-up EEGs. PNE occurred in only 27% (3 of 11) of patients without spikes or sharp waves on postnatal EEGs performed at age 3 months but in 100% ( 3 of 3 ) of patients with spikes or sharp waves (p = 0.022). We conclude that in infants whose clinical neonatal seizures are confirmed by ictal EEG tracings obtained randomly, PNE (often in the form of serious seizure syndromes such as infantile spasms) occurs far more frequently than previously reported and often despite ongoing treatment. Infants comatose during the neonatal period who had serious background EEG abnormalities and who later developed CP/MR and displayed spikeskharp waves on postnatal follow-up EEGs were at greatest risk to develop PNE. Key Words: Neonatal seizures-Epilepsy-ElectroencephalographyCerebral palsy-Mental retardation.
Neonatal seizures have long been recognized as an important clinical sign of serious encephalopathy in newborn infants (Legids et al., 1988; Clancy, 1989; Volpe, 1989). In this clinical setting, they usually represent “acute seizures,” i.e., transient seizures reflecting the presence of acute encephalopathies. Previous studies have shown that clinically diagnosed neonatal seizures forecast a high risk of death or permanent neurologic morbidity (McInerny and Schubert, 1969; Rose and Lombroso, 1970; Keen and Lee, 1973; Dennis, 1978; Holden et
al., 1982; Mellits et al., 1982; Bergman et al., 1983; Clancy, 1983). It is natural to examine the risk of a chronic seizure disorder (postnatal epilepsy) after neonatal seizures. Indeed, the relationship between neonatal seizures and the chronic condition “epilepsy” has ill-defined boundaries (Camfield and Camfield, 1987). The recognition of neonatal seizures has traditionally rested primarily or solely on clinical grounds; i.e., the observing clinician diagnoses neonatal seizures based on the appearance of distinctive abnormal “attacks” of stereotyped movements, postures, or behavior. Recent studies have shown the inherent difficulty in clinical recognition of seizures (Mizrahi and Kellaway, 1987). Errors of overdiagnosis and underdiagnosis are common. Innocent paroxysmal clinical events may erroneously be called “seizures” and many abnormal clinical attacks are not founded on a specific epileptic
Received October 1989; revision accepted February 1990. This study was presented at the Child Neurology Society Meeting in San Diego, California, 1987 and an abstract published in Ann Neurol 1987;22:407. Address correspondence and reprint requests to Dr. R. R. Clancy at Division of Neurology, The Children’s Hospital of Philadelphia, 34th St. and Civic Center Blvd., Philadelphia, PA 19104, U.S.A.
69
R . R . CLANCY AND A . LEGIDO
70
mechanism (Clancy, 1983, 1989). Conversely, many definite electrographic seizures do not incite distinctive clinical signs visible as “seizures” to the observing physician. Such “occult” or subclinical electrographic seizures are very common (Clancy et al., 1988). As a result of the inherent uncertainties in clinical recognition and quantification of neonatal seizures, previous studies of epilepsy after neonatal seizures that lack EEG confirmation raise important concerns regarding the study population (Scher et al., 1989). We examined the incidence of postnatal epilepsy in neonates whose seizures were definitely confirmed by randomly recorded ictal EEG tracings. Furthermore, several clinical, EEG, and quantitative measures of seizure abundance were examined for their relevance to the later appearance of postnatal epilepsy. PATIENTS AND METHODS Newborn infants admitted to the Children’s Hospital of Philadelphia Intensive Care Unit underwent randomly timed EEG examinations for clinically observed or suspected seizures. Only infants whose EEGs displayed unequivocal electrographic seizures comprised the study population. We examined serially all study infants neurologically during the neonatal period while they were acutely ill. Neurologic evaluation focused on their mental status (coma, lethargy, or normal), activity (absent movements, inactive, or normal) and tone (flaccid, hypotonic, or normal). Global scores were assigned reflecting their composite neurologic picture: normal, mildly abnormal, moderately abnormal, or severely abnormal. Severely abnormal neonates were comatose, flaccid, and inactive. Normal neonates showed preserved alertness, activity, and tone. Mildly and moderately abnormal infants had intermediate levels of neurological dysfunction. The children were also examined to determine their estimated gestational age (EGA), birth weight, 1- and 5-min Apgar scores, legal age at the time of seizure onset, and etiologic diagnosis. The diagnosis of neonatal asphyxia was based on the presence of abnormal Apgar scores, need for immediate resuscitation in the delivery room, abnormal arterial blood gas values, and an acute hypoxic-ischemic encephalopathy with depressed level of consciousness, hypotonia, and inactivity. Each patient was treated in a routine clinical fashion and antiepileptic drugs (AEDs) were administered as indicated. EEG examinations Routine EEG examinations were performed at the patient’s bedside with a 21-channel Grass elecEpilepsia, Vol. 32, N o . I , 1991
troencephalograph. Surface electrodes were applied with electrode paste according to the International 10-20 system modified for neonates (Clancy, in press). Tracings were obtained with a paper speed of 15 mm/s and a typical recording sensitivity of 7 pV/mm. Sixteen channels recorded EEG data and five recorded polygraphic parameters, including extraocular movements, EKG, nasal air flow, and/or thoracic respiratory movements. The EEG background was examined interictally and classified as normal or mildly, moderately, or severely abnormal according to criteria outlined in Table 1. Electrographic seizures were detected by visual analysis of the recording and were considered to exist if they lasted a minimum of 10 s and exhibited a typical evolution in space and time of abnormal repetitive stereotyped waveforms (Clancy , 1989) (Fig. 1). Each EEG was analyzed in detail to determine the following quantitative measures of seizure activity: (a) the number of seizures per hour, (b) the mean seizure duration, (c) the longest recorded seizure, and (d) the percentage of each EEG tracing during which any channel showed electrographic seizure activity (Fig. 2). Follow-up We followed all survivors and performed serial clinical neurologic examinations and Denver Developmental Screening Tests. Formal developmental test results were also available for many of the older survivors. The existence of MR, definite developmental delay (DQ 25 FV) Excessive discontinuity for state Isoelectric Marked, generalized low voltage (lOh was associated with the presence of PNE in 86% of cases, whereas only 45% of patients with seizure frequency 37 Weight (kg) 5 Age at seizure onset s1 Day > I Day Neonatal clinical neurologic evaluation Normal-Moderately abnormal Severely abnormal
Yes
NO
% (n)
% (n)
p Value
50 (3/6) 57 (12/21)
50 (3/6) 43 (9/21)
NS
80 (4/5) 50 (1 1/12)
20 (1/5) 50 (1 1/22)
NS
60 (3/5) 53 (8/15)
40 (215) 47 (7/15)
NS
75 (9/12) 40 (6/15)
25 (3/12) 60 (9/15)
NS
42 (8/19) 87 (7/8)
58 (11/19) 13 (1/8)
0.02
PNE, postnatal epilepsy; NS, not significant. Not pertinent for some patients.
NEONATAL SEIZURES A N D EPILEPSY TABLE 5. Neonatal EEG parameters and PNE in the whole group PNE EEG parameters Interictal EEG background Normal or mildly abnormal Moderate-severely abnormal Seizuredh s10 >I0 Mean seizure duration (min) s2 >2 Longest seizure duration (min) GI0 > 10 Percentage of EEG with seizure activity s10
> 10
Yes % (n)
No % (n)
25 (2/8)
75 (6/8)
68 (13/19)
32 (6/19)
45 (9/20) 86 (6/7)
55 (1 1/20) 14 (1/7)
0.058
65 (1 1/17) 40 (4/10)
35 (6/17) 60 (6110)
NS
54 (13/24) 67 (2/3)
46 (11/24) 33 ( I n )
NS
50 (5/10) 59 (10/17)
50 (5/10) 41 (7/17)
NS
p Value
0.035
Abbreviations as in Table 4.
ence of seizure etiology on PNE, we repeated an identical analysis of the clinical and EEG variables that may significantly relate to PNE on a subset of the entire group-the 10 survivors whose seizures were caused by perinatal asphyxia. Seventy percent of these survivors developed PNE. The only clinical parameter that achieved significance was a severely abnormal neurologic evaluation, associated with PNE in 100% (5 of 5 ) of patients (p = 0.002). In this patient subgroup, no EEG variables correlated significantly with PNE; however, 78% of asphyxiated infants with moderately to severely abnormal EEG background developed PNE. DISCUSSION The term “epilepsy” generally connotes a chronic neurologic condition characterized by reTABLE 6. Relationship between PNE, MR, and CP PNE Clinical parameters
Yes % (n)
% (n)
p Value
100 (15/15) 0 (oils)
25 (3/12) 75 (9112)
0.002
86 (13/15) 14 (2/15)
25 (3/12) 75 (9/12)
0.001
86 (13/15) 14 (2/15)
25 (3/12) 75 (9/12)
0.001
No
MR Yes
No CP Yes No MR + CP Yes No
PNE, postnatal epilepsy; MR, mental retardation; CP, cerebral palsy.
73
peated attacks of altered brain function (seizures) owing to electrographic ictal activity unprovoked by acute systemic disease. It is rare for the usual classified types of epilepsy to appear during the neonatal period (Camfield and Camfield, 1987). Nevertheless, there are a few examples of “neonatal epilepsy.” Newborn infants with cerebral dysgenesis and neurocutaneous disorders such as incontinenti pigmenti or organoid nevus syndrome (Clancy et al., 1985~)may have unprovoked epileptic seizures. Similarly, in “benign neonatal seizures,” (Plouin, 1985), familial neonatal seizures (Zonana et al., 1984) and, according to French authors “fifth-day fits” (Navalet et al., 1981), spontaneous clinicalEEG seizures may arise unexpectedly, sometimes followed by more typical forms of epilepsy in later life. These are uncommon causes of neonatal seizures, however. Indeed, in this study only 8% (3 of 40) of infants had an undetermined neonatal seizure cause. In contemporary neonatal medicine, most neonatal seizures represent “acute seizures,” immediately provoked by transient, unwanted system or neurologic disturbances such as hypoxia-ischemia (Clancy et al., 1988a), cranial trauma, or stroke (Clancy et al., 1985~).In these circumstances, the CNS is compelled to show temporary signs of dysfunction such as seizures and there is no inherent “lowered seizure threshold” as is implied in chronic epilepsy. Once the inciting disturbance has resolved, the cause of the seizures has passed, and the clinical EEG seizures often stop. How long should AEDs be continued in this context? Although few data suggest that PB or phenytoin (PHT) might be harmful to human newborns, it is most prudent to discontinue any medication that is no longer clearly indicated. The risk of PNE after neonatal seizures must be examined, and the relationship between these transient acute neonatal seizures and the chronic condition epilepsy must be better understood. One major obstacle to a clear understanding of neonatal seizures and epilepsy is the inherent difficulty in distinguishing between clinical neonatal seizures triggered by definite ictal EEG activity and similar-appearing clinical attacks with no simultaneous EEG seizure. Mizrahi and Kellaway (1987) suggest that many types of clinical neonatal seizures are frequently not founded on a specific epileptic mechanism and may actually represent postures, movements, or behaviors released from brainstem centers owing to transient interruption of descending inhibition. To overcome this critical difficulty, we chose our patient population from infants with clinical seizures whose EEGs had unequivocal ictal activity. Scher et al. (1989) also emphasized the critical value of confirming clinical seiEpilepsia, Vol. 32, N o . I , 1991
R . R . CLANCY AND A . LEGIDO
74
zures with EEG. Although this guarantees that all patients had epileptic-based neonatal seizures, we recognize that it also biases our study population to include sicker infants whose EEG seizures were sufficiently frequent to be captured by randomly recorded EEG tracings. We also recognize that the brief duration of the routine EEG recordings is a shortcoming of the quantitative analysis of seizure abundance. Our population with EEG-confirmed neonatal seizures had a 56% rate of PNE. This is much higher than previous reports, which generally cite risk
Raven Press, Ltd., New York 0 International League Against Epilepsy
Postnatal Epilepsy After EEG-Confirmed Neonatal Seizures Robert Ryan Clancy and Agustin Legido The Division of Neurology, The Children’s Hospital of Philadelphia, and the Departments of Neurology and Pediatrics, University of Pennsylvania School of Medicine, Philadelphia, Pennsylvania, U . S . A .
Summary: We examined infants whose neonatal seizures were confirmed by randomly recorded ictal EEG tracings to determine the types and frequency of postnatal epilepsy (PNE)-unprovoked, recurring postnatal seizures. Perinatal and postnatal clinical and EEG variables were also examined for their relevance to PNE. Forty infants with EEG-documented neonatal seizures of diverse etiologies were studied. The 27 survivors were followed for a mean of 31 months. PNE developed in 56% (15 of 27) of the cohort. The first seizure appeared at a meancorrected age of 12.7 months and occurred despite ongoing antiepileptic medication in 60% (9 of 15) of the group. Seizures were classified as infantile spasms or minor motor (7 patients), complex partial (4 patients), or generalized tonic-clonic (4 patients). Perinatal variables that significantly correlated with PNE included the presence of coma but not the age at seizure onset, the estimated gestational age, or Apgar scores. PNE occurred in 68% (13 of 19) of patients with moderately or markedly abnormal EEG backgrounds but in only 25% (2 of 8) without (p = 0.035). There was a strong trend for PNE to develop in
patients with >I0 electrographic seizures per hour but in only 45% (9 of 20) of infants with fewer seizures (p = 0.058). Several postnatal variables were significantly related to PNE-the presence of cerebral palsy (CP), mental retardation (MR), CP with MR, and follow-up EEGs. PNE occurred in only 27% (3 of 11) of patients without spikes or sharp waves on postnatal EEGs performed at age 3 months but in 100% ( 3 of 3 ) of patients with spikes or sharp waves (p = 0.022). We conclude that in infants whose clinical neonatal seizures are confirmed by ictal EEG tracings obtained randomly, PNE (often in the form of serious seizure syndromes such as infantile spasms) occurs far more frequently than previously reported and often despite ongoing treatment. Infants comatose during the neonatal period who had serious background EEG abnormalities and who later developed CP/MR and displayed spikeskharp waves on postnatal follow-up EEGs were at greatest risk to develop PNE. Key Words: Neonatal seizures-Epilepsy-ElectroencephalographyCerebral palsy-Mental retardation.
Neonatal seizures have long been recognized as an important clinical sign of serious encephalopathy in newborn infants (Legids et al., 1988; Clancy, 1989; Volpe, 1989). In this clinical setting, they usually represent “acute seizures,” i.e., transient seizures reflecting the presence of acute encephalopathies. Previous studies have shown that clinically diagnosed neonatal seizures forecast a high risk of death or permanent neurologic morbidity (McInerny and Schubert, 1969; Rose and Lombroso, 1970; Keen and Lee, 1973; Dennis, 1978; Holden et
al., 1982; Mellits et al., 1982; Bergman et al., 1983; Clancy, 1983). It is natural to examine the risk of a chronic seizure disorder (postnatal epilepsy) after neonatal seizures. Indeed, the relationship between neonatal seizures and the chronic condition “epilepsy” has ill-defined boundaries (Camfield and Camfield, 1987). The recognition of neonatal seizures has traditionally rested primarily or solely on clinical grounds; i.e., the observing clinician diagnoses neonatal seizures based on the appearance of distinctive abnormal “attacks” of stereotyped movements, postures, or behavior. Recent studies have shown the inherent difficulty in clinical recognition of seizures (Mizrahi and Kellaway, 1987). Errors of overdiagnosis and underdiagnosis are common. Innocent paroxysmal clinical events may erroneously be called “seizures” and many abnormal clinical attacks are not founded on a specific epileptic
Received October 1989; revision accepted February 1990. This study was presented at the Child Neurology Society Meeting in San Diego, California, 1987 and an abstract published in Ann Neurol 1987;22:407. Address correspondence and reprint requests to Dr. R. R. Clancy at Division of Neurology, The Children’s Hospital of Philadelphia, 34th St. and Civic Center Blvd., Philadelphia, PA 19104, U.S.A.
69
R . R . CLANCY AND A . LEGIDO
70
mechanism (Clancy, 1983, 1989). Conversely, many definite electrographic seizures do not incite distinctive clinical signs visible as “seizures” to the observing physician. Such “occult” or subclinical electrographic seizures are very common (Clancy et al., 1988). As a result of the inherent uncertainties in clinical recognition and quantification of neonatal seizures, previous studies of epilepsy after neonatal seizures that lack EEG confirmation raise important concerns regarding the study population (Scher et al., 1989). We examined the incidence of postnatal epilepsy in neonates whose seizures were definitely confirmed by randomly recorded ictal EEG tracings. Furthermore, several clinical, EEG, and quantitative measures of seizure abundance were examined for their relevance to the later appearance of postnatal epilepsy. PATIENTS AND METHODS Newborn infants admitted to the Children’s Hospital of Philadelphia Intensive Care Unit underwent randomly timed EEG examinations for clinically observed or suspected seizures. Only infants whose EEGs displayed unequivocal electrographic seizures comprised the study population. We examined serially all study infants neurologically during the neonatal period while they were acutely ill. Neurologic evaluation focused on their mental status (coma, lethargy, or normal), activity (absent movements, inactive, or normal) and tone (flaccid, hypotonic, or normal). Global scores were assigned reflecting their composite neurologic picture: normal, mildly abnormal, moderately abnormal, or severely abnormal. Severely abnormal neonates were comatose, flaccid, and inactive. Normal neonates showed preserved alertness, activity, and tone. Mildly and moderately abnormal infants had intermediate levels of neurological dysfunction. The children were also examined to determine their estimated gestational age (EGA), birth weight, 1- and 5-min Apgar scores, legal age at the time of seizure onset, and etiologic diagnosis. The diagnosis of neonatal asphyxia was based on the presence of abnormal Apgar scores, need for immediate resuscitation in the delivery room, abnormal arterial blood gas values, and an acute hypoxic-ischemic encephalopathy with depressed level of consciousness, hypotonia, and inactivity. Each patient was treated in a routine clinical fashion and antiepileptic drugs (AEDs) were administered as indicated. EEG examinations Routine EEG examinations were performed at the patient’s bedside with a 21-channel Grass elecEpilepsia, Vol. 32, N o . I , 1991
troencephalograph. Surface electrodes were applied with electrode paste according to the International 10-20 system modified for neonates (Clancy, in press). Tracings were obtained with a paper speed of 15 mm/s and a typical recording sensitivity of 7 pV/mm. Sixteen channels recorded EEG data and five recorded polygraphic parameters, including extraocular movements, EKG, nasal air flow, and/or thoracic respiratory movements. The EEG background was examined interictally and classified as normal or mildly, moderately, or severely abnormal according to criteria outlined in Table 1. Electrographic seizures were detected by visual analysis of the recording and were considered to exist if they lasted a minimum of 10 s and exhibited a typical evolution in space and time of abnormal repetitive stereotyped waveforms (Clancy , 1989) (Fig. 1). Each EEG was analyzed in detail to determine the following quantitative measures of seizure activity: (a) the number of seizures per hour, (b) the mean seizure duration, (c) the longest recorded seizure, and (d) the percentage of each EEG tracing during which any channel showed electrographic seizure activity (Fig. 2). Follow-up We followed all survivors and performed serial clinical neurologic examinations and Denver Developmental Screening Tests. Formal developmental test results were also available for many of the older survivors. The existence of MR, definite developmental delay (DQ 25 FV) Excessive discontinuity for state Isoelectric Marked, generalized low voltage (lOh was associated with the presence of PNE in 86% of cases, whereas only 45% of patients with seizure frequency 37 Weight (kg) 5 Age at seizure onset s1 Day > I Day Neonatal clinical neurologic evaluation Normal-Moderately abnormal Severely abnormal
Yes
NO
% (n)
% (n)
p Value
50 (3/6) 57 (12/21)
50 (3/6) 43 (9/21)
NS
80 (4/5) 50 (1 1/12)
20 (1/5) 50 (1 1/22)
NS
60 (3/5) 53 (8/15)
40 (215) 47 (7/15)
NS
75 (9/12) 40 (6/15)
25 (3/12) 60 (9/15)
NS
42 (8/19) 87 (7/8)
58 (11/19) 13 (1/8)
0.02
PNE, postnatal epilepsy; NS, not significant. Not pertinent for some patients.
NEONATAL SEIZURES A N D EPILEPSY TABLE 5. Neonatal EEG parameters and PNE in the whole group PNE EEG parameters Interictal EEG background Normal or mildly abnormal Moderate-severely abnormal Seizuredh s10 >I0 Mean seizure duration (min) s2 >2 Longest seizure duration (min) GI0 > 10 Percentage of EEG with seizure activity s10
> 10
Yes % (n)
No % (n)
25 (2/8)
75 (6/8)
68 (13/19)
32 (6/19)
45 (9/20) 86 (6/7)
55 (1 1/20) 14 (1/7)
0.058
65 (1 1/17) 40 (4/10)
35 (6/17) 60 (6110)
NS
54 (13/24) 67 (2/3)
46 (11/24) 33 ( I n )
NS
50 (5/10) 59 (10/17)
50 (5/10) 41 (7/17)
NS
p Value
0.035
Abbreviations as in Table 4.
ence of seizure etiology on PNE, we repeated an identical analysis of the clinical and EEG variables that may significantly relate to PNE on a subset of the entire group-the 10 survivors whose seizures were caused by perinatal asphyxia. Seventy percent of these survivors developed PNE. The only clinical parameter that achieved significance was a severely abnormal neurologic evaluation, associated with PNE in 100% (5 of 5 ) of patients (p = 0.002). In this patient subgroup, no EEG variables correlated significantly with PNE; however, 78% of asphyxiated infants with moderately to severely abnormal EEG background developed PNE. DISCUSSION The term “epilepsy” generally connotes a chronic neurologic condition characterized by reTABLE 6. Relationship between PNE, MR, and CP PNE Clinical parameters
Yes % (n)
% (n)
p Value
100 (15/15) 0 (oils)
25 (3/12) 75 (9112)
0.002
86 (13/15) 14 (2/15)
25 (3/12) 75 (9/12)
0.001
86 (13/15) 14 (2/15)
25 (3/12) 75 (9/12)
0.001
No
MR Yes
No CP Yes No MR + CP Yes No
PNE, postnatal epilepsy; MR, mental retardation; CP, cerebral palsy.
73
peated attacks of altered brain function (seizures) owing to electrographic ictal activity unprovoked by acute systemic disease. It is rare for the usual classified types of epilepsy to appear during the neonatal period (Camfield and Camfield, 1987). Nevertheless, there are a few examples of “neonatal epilepsy.” Newborn infants with cerebral dysgenesis and neurocutaneous disorders such as incontinenti pigmenti or organoid nevus syndrome (Clancy et al., 1985~)may have unprovoked epileptic seizures. Similarly, in “benign neonatal seizures,” (Plouin, 1985), familial neonatal seizures (Zonana et al., 1984) and, according to French authors “fifth-day fits” (Navalet et al., 1981), spontaneous clinicalEEG seizures may arise unexpectedly, sometimes followed by more typical forms of epilepsy in later life. These are uncommon causes of neonatal seizures, however. Indeed, in this study only 8% (3 of 40) of infants had an undetermined neonatal seizure cause. In contemporary neonatal medicine, most neonatal seizures represent “acute seizures,” immediately provoked by transient, unwanted system or neurologic disturbances such as hypoxia-ischemia (Clancy et al., 1988a), cranial trauma, or stroke (Clancy et al., 1985~).In these circumstances, the CNS is compelled to show temporary signs of dysfunction such as seizures and there is no inherent “lowered seizure threshold” as is implied in chronic epilepsy. Once the inciting disturbance has resolved, the cause of the seizures has passed, and the clinical EEG seizures often stop. How long should AEDs be continued in this context? Although few data suggest that PB or phenytoin (PHT) might be harmful to human newborns, it is most prudent to discontinue any medication that is no longer clearly indicated. The risk of PNE after neonatal seizures must be examined, and the relationship between these transient acute neonatal seizures and the chronic condition epilepsy must be better understood. One major obstacle to a clear understanding of neonatal seizures and epilepsy is the inherent difficulty in distinguishing between clinical neonatal seizures triggered by definite ictal EEG activity and similar-appearing clinical attacks with no simultaneous EEG seizure. Mizrahi and Kellaway (1987) suggest that many types of clinical neonatal seizures are frequently not founded on a specific epileptic mechanism and may actually represent postures, movements, or behaviors released from brainstem centers owing to transient interruption of descending inhibition. To overcome this critical difficulty, we chose our patient population from infants with clinical seizures whose EEGs had unequivocal ictal activity. Scher et al. (1989) also emphasized the critical value of confirming clinical seiEpilepsia, Vol. 32, N o . I , 1991
R . R . CLANCY AND A . LEGIDO
74
zures with EEG. Although this guarantees that all patients had epileptic-based neonatal seizures, we recognize that it also biases our study population to include sicker infants whose EEG seizures were sufficiently frequent to be captured by randomly recorded EEG tracings. We also recognize that the brief duration of the routine EEG recordings is a shortcoming of the quantitative analysis of seizure abundance. Our population with EEG-confirmed neonatal seizures had a 56% rate of PNE. This is much higher than previous reports, which generally cite risk
