Nonconvulsive seizures are common in critically ill children.

BACKGROUND: Retrospective studies have reported the occurrence of nonconvulsive seizures in critically ill children. We aimed to prospectively determine the incidence and risk factors of nonconvulsive seizures in critically ill children using predetermined EEG monitoring indications and EEG interpretation terminology. METHODS: Critically ill children (non-neonates) with acute encephalopathy underwent continuous EEG monitoring if they met institutional clinical practice criteria. Study enrollment and data collection were prospective. Logistic regression analysis was utilized to identify risk factors for seizure occurrence. RESULTS: One hundred children were evaluated. Electrographic seizures occurred in 46 and electrographic status epilepticus occurred in 19. Seizures were exclusively nonconvulsive in 32. The only clinical risk factor for seizure occurrence was younger age (p=0.03). Of patients with seizures, only 52% had seizures detected in the first hour of monitoring, while 87% were detected within 24 hours. CONCLUSIONS: Seizures were common in critically ill children with acute encephalopathy. Most were nonconvulsive. Clinical features had little predictive value for seizure occurrence. Further study is needed to confirm these data in independent high-risk populations, to clarify which children are at highest risk for seizures so limited monitoring resources can be allocated optimally, and to determine whether seizure detection and management improves outcome.

Electroencephalographic monitoring during hypothermia after pediatric cardiac arrest.

BACKGROUND: Hypoxic ischemic brain injury secondary to pediatric cardiac arrest (CA) may result in acute symptomatic seizures. A high proportion of seizures may be nonconvulsive, so accurate diagnosis requires continuous EEG monitoring. We aimed to determine the safety and feasibility of long-term EEG monitoring, to describe electroencephalographic background and seizure characteristics, and to identify background features predictive of seizures in children undergoing therapeutic hypothermia (TH) after CA. METHODS: Nineteen children underwent TH after CA. Continuous EEG monitoring was performed during hypothermia (24 hours), rewarming (12-24 hours), and then an additional 24 hours of normothermia. The tolerability of these prolonged studies and the EEG background classification and seizure characteristics were described in a standardized manner. RESULTS: No complications of EEG monitoring were reported or observed. Electrographic seizures occurred in 47% (9/19), and 32% (6/19) developed status epilepticus. Seizures were nonconvulsive in 67% (6/9) and electrographically generalized in 78% (7/9). Seizures commenced during the late hypothermic or rewarming periods (8/9). Factors predictive of electrographic seizures were burst suppression or excessively discontinuous EEG background patterns, interictal epileptiform discharges, or an absence of the expected pharmacologically induced beta activity. Background features evolved over time. Patients with slowing and attenuation tended to improve, whereas those with burst suppression tended to worsen. CONCLUSIONS: EEG monitoring in children undergoing therapeutic hypothermia after cardiac arrest is safe and feasible. Electrographic seizures and status epilepticus are common in this setting but are often not detectable by clinical observation alone. The EEG background often evolves over time, with milder abnormalities improving and more severe abnormalities worsening.

Limitations of single-channel EEG on the forehead for neonatal seizure detection.

OBJECTIVES: In amplitude-integrated EEG, lead placement across the forehead is convenient, but this location has unknown effects on neonatal seizure (NS) detection sensitivity. This study describes the limits of NS detection by a single forehead EEG channel. STUDY DESIGN: Records were taken from a digital library of conventional EEGs (CEEGs) with NS, previously characterized at a bicentral channel, C(3) --> C(4). We analyzed electrographic characteristics in a single forehead channel, Fp(3) --> Fp(4). RESULT: A total of 330 seizures from 125 CEEGs were included. With Fp(3) --> Fp(4), at least one NS was detected in 66% of records vs 90% using C(3) --> C(4) (P<0.0001). Of 330 seizures, 46% appeared in Fp(3) --> Fp(4) vs 73% in C(3) --> C(4) (P<0.0001). Seizures appeared briefer in Fp(3) --> Fp(4) than C(3) --> C(4) (P<0.006) and CEEG (P<0.0001). CONCLUSION: NSs are significantly more difficult to detect with a single forehead channel than bicentrally or on CEEG. In Fp(3) --> Fp(4), a third of records with seizures were missed and over half of seizures were undetected.

Dystrophic neuritic processes in epileptic cortex.

Cortical dysplasia is a frequent finding in cortical resections from children with refractory epilepsy. Diagnostic criteria and a classification scheme for cortical dysplasia has been proposed, though the relationship between specific cortical dysplasia features and their causal relationship with epilepsy is poorly understood. We reviewed 28 surgical resections from children and identified a common and easily recognized feature of cortical dysplasia: maloriented, misshapen and occasionally coarse neurofilament stained process forming a dystrophic neuritic background. The dystrophic neuritic background was associated with other features of cortical dysplasia in all 28 patients with cortical dysplasia, 26 with refractory epilepsy and 2 patients with other neurologic diagnoses. In seven children with refractory epilepsy due to other pathologic diagnosis such as vascular or glial lesions, the dystrophic neuritic background was only found in one patient with a ganglioglioma and other features suggestive of an associated cortical dysplasia. Our data indicate that a dystrophic neuritic background is a common and relatively specific neuropathologic finding in cortical dysplasia.

Dysplasia: a common finding in intractable pediatric temporal lobe epilepsy.

BACKGROUND: Risk factors for temporal lobe epilepsy (TLE) include history of CNS infection, family history of epilepsy, and history of febrile convulsions (FC). Pre-existing cortical dysplasia (CD) may also predispose to refractory TLE, independent of other risk factors for epilepsy. METHODS: The authors reviewed the neuropathologic features of surgical tissue from temporal lobectomies of 33 pediatric patients with refractory TLE, with and without a history of epilepsy risk factors. RESULTS: CD was found in 64% (21/33) of all patients with refractory TLE, including 73% (11/15) patients with a history of FC, 66% (2/3) patients with CNS infections, and 83% (5/6) patients with a family history of epilepsy. Disrupted cortical lamination, dystrophic and maloriented neurons, and balloon cells characterized the CD found in the temporal neocortex. CONCLUSION: CD was seen in 21 of 33 surgical specimens from children with refractory TLE, including those with and without other epilepsy risk factors.

Incidental intracranial hemorrhage after uncomplicated birth: MRI before and after neonatal heart surgery.

We investigated the prevalence of intracranial hemorrhage (ICH) before and after neonatal heart surgery. We carried out pre- and postoperative MRI looking for brain lesions in 24 full-term newborns with known congenital heart disease. They underwent heart surgery with cardiopulmonary bypass (CPB), usually with deep hypothermic circulatory arrest (DHCA). The first MRI was 1-22 days after birth. There were 21 children born after uncomplicated vaginal delivery and three delivered by cesarean section (CS). ICH was seen in 13 (62%) of the vaginal delivery group but in none of the CS group. We saw subdural bleeding along the inferior surface of the tentorium in 11 (52%) and supratentorially in six (29%) of the 21 children with ICH. Small hemorrhages were present in the choroid plexus in seven (33%), in the parenchyma in one (5%) and in the occipital horn in one (5%). There were 26 foci of bleeding in these 21 patients (1.2 per patient). None was judged by formal neurologic examination to be symptomatic from the hemorrhage. Follow-up MRI after cardiac surgery was obtained in 23 children, showing 37 foci of ICH (1.6 per patient), but all appeared asymptomatic. Postoperatively, ICH had increased in 10 children (43%), was unchanged in seven (30%) and was less extensive in six (26%).

Allopurinol neurocardiac protection trial in infants undergoing heart surgery using deep hypothermic circulatory arrest.

OBJECTIVE: This pharmacologic protection trial was conducted to test the hypothesis that allopurinol, a scavenger and inhibitor of oxygen free radical production, could reduce death, seizures, coma, and cardiac events in infants who underwent heart surgery using deep hypothermic circulatory arrest (DHCA). DESIGN: This was a single center, randomized, placebo-controlled, blinded trial of allopurinol in infant heart surgery using DHCA. Enrolled infants were stratified as having hypoplastic left heart syndrome (HLHS) and all other forms of congenital heart disease (non-HLHS). Drug was administered before, during, and after surgery. Adverse events and the clinical efficacy endpoints death, seizures, coma, and cardiac events were monitored until infants were discharged from the intensive care unit or 6 weeks, whichever came first. RESULTS: Between July 1992 and September 1997, 350 infants were enrolled and 348 subsequently randomized. A total of 318 infants (131 HLHS and 187 non-HLHS) underwent heart surgery using DHCA. There was a nonsignificant treatment effect for the primary efficacy endpoint analysis (death, seizures, and coma), which was consistent over the 2 strata. The addition of cardiac events to the primary endpoint resulted in a lack of consistency of treatment effect over strata, with the allopurinol treatment group experiencing fewer events (38% vs 60%) in the entire HLHS stratum, compared with the non-HLHS stratum (30% vs 27%). In HLHS surgical survivors, 40 of 47 (85%) allopurinol-treated infants did not experience any endpoint event, compared with 27 of 49 (55%) controls. There were fewer seizures-only and cardiac-only events in the allopurinol versus placebo groups. Allopurinol did not reduce efficacy endpoint events in non-HLHS infants. Treated and control infants did not differ in adverse events. CONCLUSIONS: Allopurinol provided significant neurocardiac protection in higher-risk HLHS infants who underwent cardiac surgery using DHCA. No benefits were demonstrated in lower risk, non-HLHS infants, and no significant adverse events were associated with allopurinol treatment.congenital heart defects, hypoplastic left heart syndrome, induced hypothermia, ischemia-reperfusion injury, neuroprotective agents, allopurinol, xanthine oxidase, free radicals, seizures, coma.

Impact of prenatal diagnosis on survival and early neurologic morbidity in neonates with the hypoplastic left heart syndrome.

BACKGROUND: Prenatal echocardiography can identify the fetus that has complex congenital heart disease and may improve early management and surgical outcome. Prenatal diagnosis may be particularly beneficial to patients who have hypoplastic left heart syndrome (HLHS) and who are at risk for hypoxic-ischemic insult at presentation. OBJECTIVES: We sought to determine whether prenatal diagnosis reduces neurologic morbidity and operative mortality in patients who undergo palliative surgery for the HLHS. METHODS: Data from all patients who had HLHS, except for those with lethal genetic anomalies, and who were admitted to our institution between July 1992 and September 1997 were analyzed to assess the impact of prenatal diagnosis on preoperative management, neurologic morbidity, and surgical mortality. The primary outcome measures were hospital mortality and the incidence of adverse neurologic events (seizure or coma). RESULTS: There were 216 patients who had HLHS and were referred for surgical palliation, 79 (36.6%) of whom had been diagnosed prenatally. All patients who had been diagnosed prenatally were delivered in an advanced nursery and were started on prostaglandin E(1) on the first day of life. Patients whose HLHS was diagnosed postnatally were begun on prostaglandin E(1) later in life (median = day 2 [range = 1-28 days]). There were 4 preoperative deaths and 53 operative or postoperative deaths. Overall hospital mortality was 26.4% and did not differ between patients whose HLHS had been diagnosed prenatally and those whose HLHS had been diagnosed postnatally. With the use of multivariable analysis, prenatal diagnosis was associated with fewer adverse perioperative neurologic events in the patients whose HLHS had been diagnosed prenatally than in those whose HLHS had been diagnosed postnatally (odds ratio = 0.46). CONCLUSIONS: These data suggest that prenatal diagnosis has a favorable impact on treatment of patients who have HLHS and are undergoing staged palliation and reduces early neurologic morbidity. Prenatal diagnosis was not associated with reduced hospital mortality. It is possible that prenatal diagnosis may improve long-term neurologic outcome.

Horner syndrome associated with implantation of a vagus nerve stimulator.

PURPOSE: To report a case of Horner syndrome that occurred after implantation of a vagus nerve stimulator. METHODS: Case report. RESULTS: A 6-year-old female with cerebral dysgenesis and intractable partial seizures presented with Horner syndrome after vagus nerve stimulator implantation. CONCLUSION: Horner syndrome can occur as a result of the vagus nerve stimulator implant procedure and should be included as one of its possible surgical complications.

Neonatal seizures: early-onset seizure syndromes and their consequences for development.

The determination of the developmental consequences of seizure syndromes in the neonate is based upon a number of factors which include: understanding of the clinical and electroencephalographic (EEG) features of neonatal seizures; current theories of the mechanisms by which neonatal seizures are generated; a current classification of neonatal seizures; potential etiologic and risk factors for seizures; and therapies. In addition, different seizure types, mechanisms of generation and etiologies of cerebral dysfunction may vary with conceptional age of the infant. There are a few distinct neonatal epileptic syndromes, which are rare, have been well described: benign neonatal convulsions; benign neonatal familial convulsions; early myoclonic encephalopathy and early infantile epileptic encephalopathy. The prognosis for the first two is relatively good while the outcome for the other two with encephalopathy is catastrophic. However, the majority of neonatal seizures occur as acute, reactive events in association with a wide range of etiologic factors. These etiologic factors, as well as those of the more traditionally defined syndromes, are the main determinants of eventual developmental outcome of neonates who experience seizures. Although experimental data suggests that some epileptic seizures eventually may have physiological, histological, metabolic, or behavioral consequences, there is yet direct evidence in humans to suggest that the occurrence of seizures themselves in the neonate is the main determinant of long-term outcome.

CHOP Infant Coma Scale ("Infant Face Scale"): a novel coma scale for children less than two years of age.

The Glasgow Coma Scale (GCS) is the most frequently used tool worldwide for assessing the severity of neurologic injury after brain trauma, although applying this scale to infants and younger children can be problematic. The CHOP Infant Coma Scale, or Infant Face Scale (IFS), is a novel scale for children under 2 years of age which differs from other pediatric coma scales in the following ways: (1) it relies on objective behavioral observations; (2) it assesses cortical as well as brainstem function; (3) it parallels the GCS in scoring but is based on infant-appropriate behaviors; and (4) it can be applied to intubated patients. We report the results of a prospective study designed to compare interrater reliability between the IFS and GCS in children less than 2 years of age. Seventy-five hospitalized children less than 2 years of age were assessed simultaneously by a pair of observers, representing a spectrum of health care professionals, who scored the children using both the IFS and GCS. Interrater reliability for each pair of observers for each scale was assessed using the kappa statistic. A second series of 10 infants in the intensive care unit with specific diagnoses of acute traumatic or hypoxic/ischemic brain injury were similarly assessed. In the 75 hospitalized infants with a variety of diagnoses, interrater reliability for the GCS was in the "almost perfect," "slight," and "fair" range for the eye-opening, motor, and verbal subtests, respectively. In contrast, the IFS showed interrater reliability in the "almost perfect," "substantial," and "almost perfect" ranges for the three subtests. When applied to infants in an intensive care unit with acute traumatic brain injury or hypoxia/ischemia, the GCS interrater reliability scores were in the "fair" range, while the IFS scores were in the "almost perfect" range. The IFS demonstrates improved interrater reliability in direct comparison to the GCS, particularly in the "verbal/face" component where most pediatric coma scales are deficient. The IFS may prove to be a simple and practical bedside index of brain injury severity in children less than two years of age.

Tonsillar pain mimicking glossopharyngeal neuralgia as a complication of vagus nerve stimulation: case report.

An adolescent girl presented with severe, lancinating tonsillar pain exacerbated by swallowing 6 weeks after initiation of left vagus nerve stimulation for intractable epilepsy. Her symptoms mimicked those seen in glossopharyngeal neuralgia and were relieved by temporary cessation of stimulation. Gradual reinstitution of therapy with alteration in stimulus parameters resulted in improved seizure control as well as cessation of pain symptoms. Direct stimulation of the vagus nerve may result in vagoglossopharyngeal neuralgia, which, in this case, was amenable to stimulus modification.

Neurodevelopmental outcome and lifestyle assessment in school-aged and adolescent children with hypoplastic left heart syndrome.

OBJECTIVES: The purposes of this study are to describe the quality of life and cognitive function in school-aged children who have undergone staged palliation for hypoplastic left heart syndrome (HLHS), and to identify factors that are predictive of neurodevelopmental outcome in this population. METHODS: School-aged survivors with HLHS who had undergone palliative surgery at our institution were identified and mailed a questionnaire to assess subjectively quality of life, school performance, and incidence of medical complications. A subgroup of local patients underwent standardized testing of cognitive function and neurologic examination. These patients were compared with the larger (remote) group of questionnaire respondents to determine whether results may be generalizable to the entire HLHS population. Potential predictors of neurologic and cognitive outcome were tested for their association with test scores using multivariate regression analysis. RESULTS: Questionnaire results were obtained from 115 of 138 eligible children (83%; mean age: 9.0 +/- 2.0 years). Standardized testing was performed in 28 of 34 (82%) eligible local patients (mean age: 8.6 +/- 2.1 years). The majority of parents or guardians described their child's health as good (34%) or excellent (45%) and their academic performance as average (42%) or above average (42%). One third of the children, however, were receiving some form of special education. Chronic medication usage was common (64%); the incidence of medical complications was comparable to that previously reported in children with Fontan physiology. Cognitive testing of the local group demonstrated a median full scale IQ of 86 (range: 50-116). Mental retardation (IQ: <70) was noted in 18% of patients. In multivariate analysis, only the occurrence of preoperative seizures predicted lower full scale IQ. CONCLUSIONS: Although the majority of school-aged children with HLHS had IQ scores within the normal range, mean performance for this historical cohort of survivors was lower than that in the general population.

Preoperative risk-of-death prediction model in heart surgery with deep hypothermic circulatory arrest in the neonate.

OBJECTIVE: Our goal was to generate a preoperative risk-of-death prediction model in selected neonates with congenital heart disease undergoing surgery with deep hypothermic circulatory arrest. METHODS: We completed a single-center, prospective, randomized, double-blind, placebo- controlled neuroprotection trial in selected neonates with congenital heart disease requiring operations for which deep hypothermic circulatory arrest was used. An extensive database was generated that included preoperative, intraoperative, and postoperative variables. Variables (delivery, maternal, and infant related) were evaluated to produce a preoperative risk-of-death prediction model by means of logistic regression. An operative risk-of-death prediction model including duration of deep hypothermic circulatory arrest was also generated. RESULTS: Between July 1992 and September 1997, 350 (74%) of 473 eligible infants were enrolled with 318 undergoing deep hypothermic circulatory arrest. The mortality was 52 of 318 (16.4%), unaffected by investigational drug. The resulting preoperative risk model contained 4 variables: (1) cardiac anatomy (two-ventricle vs single ventricle surgery, with/without arch obstruction), (2) 1-minute Apgar score (5), (3) presence of genetic syndrome, and (4) age at hospital admission for surgery (5 days). Mortality for two-ventricle repair was 3.2% (4/130). Mortality for single ventricle palliation was 25.5% (48/188) and was significantly influenced by Apgar score, genetic diagnosis, and admission age. The preoperative model had a prediction accuracy of 80%. The operative risk model included duration of deep hypothermic circulatory arrest, which significantly (P =.03) increased risk of death, with a prediction accuracy of 82%. CONCLUSIONS: In this selected population, postoperative mortality risk is significantly affected by preoperative conditions. Identification of infants with varying mortality risks may affect family counseling, therapeutic intervention, and risk stratification for future study designs.

Successful use of alternate waste nitrogen agents and hemodialysis in a patient with hyperammonemic coma after heart-lung transplantation.

BACKGROUND: Lethal hyperammonemic coma has been reported in 2 adults after lung transplantation. It was associated with a massive elevation of brain glutamine levels, while plasma glutamine levels were normal or only slightly elevated. In liver tissue, glutamine synthetase activity was markedly reduced, and the histologic findings resembled those of Reye syndrome. The adequacy of therapy commonly used for inherited disorders of the urea cycle has not been adequately evaluated in patients with this form of secondary hyperammonemia. OBJECTIVE: To determine whether hemodialysis, in conjunction with intravenous sodium phenylacetate, sodium benzoate, and arginine hydrochloride therapy, would be efficacious in a patient with hyperammonemic coma after solid-organ transplantation. DESIGN: Case report. SETTING: A children's hospital. PATIENT: A 41-year-old woman with congenital heart disease developed a hyperammonemic coma with brain edema 19 days after undergoing a combined heart and lung transplantation. METHODS: Ammonium was measured in plasma. Amino acids were quantitated in plasma and cerebrospinal fluid by column chromatography. The effectiveness of therapy was assessed by measuring plasma ammonium levels and intracranial pressure and performing sequential neurological examinations. RESULTS: The patient had the anomalous combination of increased cerebrospinal fluid and decreased plasma glutamine levels. To our knowledge, she is the first patient with this complication after solid-organ transplantation to survive after combined therapy with sodium phenylacetate, sodium benzoate, arginine hydrochloride, and hemodialysis. Complications of the acute coma included focal motor seizures, which were controlled with carbamazepine, and difficulty with short-term memory. CONCLUSIONS: The aggressive use of hemodialysis in conjunction with intravenous sodium phenylacetate, sodium benzoate, and arginine hydrochloride therapy may allow survival in patients after solid-organ transplantation. An acute acquired derangement in extra-central nervous system glutamine metabolism may play a role in the production of hyperammonemia in this illness that resembles Reye syndrome, and, as in other hyperammonemic disorders, the duration and degree of elevation of brain glutamine levels may be the important determining factors in responsiveness to therapy.

Purine metabolism and inhibition of xanthine oxidase in severely hypoxic neonates going onto extracorporeal membrane oxygenation.

The effect of allopurinol to inhibit purine metabolism via the xanthine oxidase pathway in neonates with severe, progressive hypoxemia during rescue and reperfusion with extracorporeal membrane oxygenation (ECMO) was examined. Twenty-five term infants meeting ECMO criteria were randomized in a double-blinded, placebo-controlled trial. Fourteen did not receive allopurinol, whereas 11 were treated with 10 mg/kg after meeting criteria and before cannulation, in addition to a 20-mg/kg priming dose to the ECMO circuit. Infant plasma samples before cannulation, and at 15, 30, 60, and 90 min, and 3, 6, 9, and 12 h on bypass were analyzed (HPLC) for allopurinol, oxypurinol, hypoxanthine, xanthine, and uric acid concentrations. Urine samples were similarly evaluated for purine excretion. Hypoxanthine concentrations in isolated blood-primed ECMO circuits were separately measured. Hypoxanthine, xanthine, and uric acid levels were similar in both groups before ECMO. Hypoxanthine was higher in allopurinol-treated infants during the time of bypass studied (p = 0.022). Xanthine was also elevated (p < 0.001), and uric acid was decreased (p = 0.005) in infants receiving allopurinol. Similarly, urinary elimination of xanthine increased (p < 0.001), and of uric acid decreased (p = 0.04) in treated infants. No allopurinol toxicity was observed. Hypoxanthine concentrations were significantly higher in isolated ECMO circuits and increased over time during bypass (p < 0.001). This study demonstrates that allopurinol given before cannulation for and during ECMO significantly inhibits purine degradation and uric acid production, and may reduce the production of oxygen free radicals during reoxygenation and reperfusion of hypoxic neonates recovered on bypass.

Using age-appropriate prolactin levels to diagnose children with seizures in the emergency department.

OBJECTIVE: To assess the utility of serum and cerebrospinal fluid (CSF) prolactin levels for identifying children who have experienced seizures. METHODS: A prospective cohort study was performed in a pediatric ED at an urban children's hospital. A convenience sample of children underwent blood and CSF analyses in the ED over a 2-year period. RESULTS: Thirty-five children (aged 3 months-15 years) with generalized tonic-clonic seizures and 48 ill control patients were studied. Both groups included febrile and afebrile patients. The patient characteristics in the seizure and control groups were similar with respect to age, fever, current medications, and blood, urine, and CSF cultures. When serum prolactin levels were assigned age-adjusted dichotomous values of "elevated" or "normal," the rates of elevation between the seizure and control patients were different (p < 0.001). The positive and negative predictive values of these age-adjusted levels were 68% (95% CI 47-85%) and 76% (95% CI 61-87%), respectively. The mean CSF prolactin levels of the seizure and control groups were not significantly different. In addition, there was no single threshold CSF prolactin level that could delineate seizure patients from control patients. CONCLUSIONS: Age-adjusted serum prolactin levels are useful only as an adjunct in the prospective evaluation of the individual pediatric patient for epileptic seizure activity. CSF prolactin levels are not useful in the diagnosis of generalized seizures in children in the acute care setting.

Sensitivity of a seizure activity detection computer in childhood video/electroencephalographic monitoring.

PURPOSE: Few data describe the sensitivity of a "Seizure Activity Detection Computer" (SzAC) in childhood video/EEG (VEEG) monitoring, especially for very young children. We examined the accuracy of SzAC in childhood VEEG monitoring in different pediatric age groups. METHODS: We visually analyzed VEEG monitoring samples for randomly designated subsets of 56 patients with childhood epilepsy, reviewing 335 visually detected electrographic seizures to analyze the sensitivity of SzAC for each age group as well as the electrographic characteristics affecting the sensitivity of automated computer-based seizure detection. RESULTS: SzAC was positive in 227 of 335 (67.8%) visually identified electrographic seizures in the entire study group. The SzAC sensitivity for infants (age 2 months to 2 years) was 59.7% (43 of 72 seizures); for young children (age 3-10 years), sensitivity 56.5% (91 of 161 seizures). In adolescents (age 11-18 years), SzAC was positive in 93 of 102 (91.2%) seizures--i.e., in a significantly greater number of seizures than in younger age groups. SzAC was significantly less sensitive in detecting electrographic seizures characterized as being of short duration or of low voltage. CONCLUSIONS: The overall sensitivity of SzAC in detecting childhood electrographic seizures was 67.8%. The sensitivity was significantly less in younger age groups as compared with that in children aged >11 years. Electrographic seizures of relatively short duration or low voltage were often missed by SzAC.