Bernard Rosenblatt

Automatic seizure detection in the newborn: methods and initial evaluation

Seizures are most common in the newborn period, but at that age seizures can be very difficult to identify by clinical observation. Therefore the EEG plays an even greater role in newborns than in older children and adults. The electrographic features of seizures and EEG background in the newborn are, however, very different to those found in adults. We present a set of methods for the automatic detection of seizures in the newborn. The methods are aimed at detecting a wide range of patterns, including rhythmic paroxysmal discharges at a wide range of frequencies, as well as repetitive spike patterns, even when they are not very rhythmic. The methods were developed using EEGs obtained from 55 newborns, recorded at 3 hospitals that used differing monitoring protocols. A total of 281 h of recordings containing 679 seizures were analyzed. An initial evaluation indicated that 71% of the seizures and 78% of seizure clusters (group of seizures separated by less than 90 s) were detected, with a false detection rate of 1.7/h. The methods were developed so that they can be implemented to operate in real time.

Automatic EEG analysis during long-term monitoring in the ICU.

To assist in the reviewing of prolonged EEGs, we have developed an automatic EEG analysis method that can be used to compress the prolonged EEG into two pages. The proposed approach of Automatic Analysis of Segmented-EEG (AAS-EEG) consists of 4 basic steps: (1) segmentation; (2) feature extraction; (3) classification; and (4) presentation. The idea is to break down the EEG into stationary segments and extract features that can be used to classify the segments into groups of like patterns. The final step involves the presentation of the processed data in a compressed form. This is done by providing the EEGer with a representative sample from each group of EEG patterns and a compressed time profile of the complete EEG. To verify the above approach, 41 6 h EEG records were assessed for normality via the AAS-EEG and conventional EEG approaches. The difference between the overall assessment via compressed and conventional EEG was within one abnormality level 100% of the time, and within one-half level for 73.6% of the records. We demonstrated the feasibility and reliability of automatically segmenting and clustering the EEG, thus allowing the reduction of a 6 h tracing to a few representative segments and their time sequence. This should facilitate review of long recordings during monitoring in the ICU.

The Determination of Sensory Deficits in Children With Hemiplegic Cerebral Palsy

Therapeutic intervention for children with cerebral palsy begins with accurate appraisal of abilities and disabilities. Currently, treatment focuses on the identified motor deficits, with any underlying sensory deficits often being overlooked. Sensory input is an essential component of motor function and motor control. Therefore, the objective of this study was to determine the presence and extent of sensory deficits in school-aged hemiplegic children using a formal clinical sensory battery, as well as somatosensory evoked potentials. A cross-sectional comparative study was performed comparing sensory function in hemiplegic children and healthy controls. Nine hemiplegic children and 41 healthy controls between 4 and 19 years of age were assessed. Hemiplegic children were included if they had a minimum level of receptive language function of 30 to 33 months and expressive language ability of 24 to 27 months, and no severe limitation of joint range of the hand. Significant bilateral sensory deficits (88.8%) were ascertained in hemiplegic children (P < .05), when compared to the performance of the healthy controls. Stereognosis and proprioception were the chief modalities affected bilaterally. The extent of sensory loss did not mirror the severity of motor deficit. Conversely, findings on somatosensory evoked potentials were closely related to motor function. Thus, a clinical sensory evaluation should be an integral part of the assessment of children with cerebral palsy. The likelihood of sensory impairment in one or more modalities on the hemiplegic or nonhemiplegic side is underappreciated and needs to be identified by rehabilitation specialists to maximize the functional potential of these children. (J Child Neurol 1995; 10:300-309).

Developmental and Functional Outcomes at School Entry in Children with Congenital Heart Defects

OBJECTIVE To describe developmental and functional outcomes of children with congenital heart defects (CHDs) at school entry after open heart surgery. STUDY DESIGN Infants with CHDs who underwent surgical repair in infancy were recruited and assessed prospectively for developmental progress. At 5 years of age (64.2 +/- 11.3 months), 94 subjects were evaluated in a blind fashion by using a variety of standardized measures. RESULTS Mean IQ scores were in the low average range (90-94). Receptive language was in the average range (103.6 +/- 14.4). Behavioral difficulties were common (27.1%), with internalizing problems being more frequent. Functional limitations in socialization (93.0 +/- 17.1), daily living skills (94.6 +/- 16.4), communication (90.0 +/- 14.1), and adaptive behavior (92.1 +/- 15.8) were noted in 11% to 17% of children. With the Functional Independence Measure for Children, 20% to 22% of subjects were more dependent than their peers in self-care and social cognition, although few (4.5%) had mobility restrictions. Predictors of developmental and functional limitations included: abnormal postoperative neurologic examination, microcephaly, deep hypothermic circulatory arrest time, palliation, acyanotic heart lesion, age at surgery, and maternal education. CONCLUSIONS After infant open-heart surgery, children with CHDs may exhibit a range of developmental difficulties at school entry that enhances risk for learning challenges and decreased social participation.

Health and well-being of children with congenital cardiac malformations, and their families, following open-heart surgery

Infants who survive open-heart surgery are at risk for developmental disability, which may impact on the well-being not only of the child, but also the family. The objective of our prospective study, therefore, was to determine the long-term health-related quality of life of children with congenital cardiac malformations following open-heart surgery, and to describe the persisting level of stress in their families. To this end, 49 parents completed the Child Health Questionnaire, the Parenting Stress Index, and the Child Behaviour Checklist as part of a developmental follow-up protocol when their child was 5 years of age. Mean scores on the Child Health Questionnaire were in the normal range, with physical well-being equal to 53.5, psychosocial well-being 50.9, with only 6.4 percent and 8.5 percent of subjects, respectively, falling within the suboptimal range of less than 40. The distribution of scores on the Parenting Stress Index, however, were more variable, with over one-quarter of parents indicating a high level of stress, with almost one-fifth having low levels of stress, and just over half scoring in the normal range, with the group mean being 52.6 plus or minus 32.3. An abnormal neurologic examination before surgery was associated with lower physical health (beta equal to -5.5, p equal to 0.02, r2 equal to 0.18), whereas lower arterial saturations of oxygen, less than 85 percent preoperatively, was associated with lower psychosocial health (beta equal to -6.6, p equal to 0.01, and r(2) equal to 0.14). The internalizing and externalizing behaviours of the child were significantly correlated with psychosocial well being, with r ranging from -0.32 to -0.52, and p less than 0.05. Parental stress also correlated with psychosocial health (r equal to -0.48 and p equal to 0.0009). Overall, the perception by the parents of the health-related quality of life of their child is favourable 5 years following open-heart surgery during infancy. Many parents, nonetheless, continue to feel either stressed or defensive about their child, particularly if their child exhibits behavioural difficulties. Our findings suggest that strategies need to be considered to enhance family well-being in the planning and delivery of health services to this population at high risk.

Evaluation of an automatic seizure detection method for the newborn EEG.

In another publication, we described a set of methods for automatic detection of EEG seizures in the newborn. We describe here the evaluation of these methods using a completely new set of data, which were not used in developing the method. This testing data set consisted of recording from 54 patients, lasting an average of 4.4 h. Recordings had 8-16 channels and were obtained, in approximately equal numbers, from 3 institutions in Canada, the USA and Australia. Recording conditions varied from short recordings fully attended by a technologist to overnight recordings largely unattended. The average seizure detection rate was 69% (77%, 53%, 84% in the 3 institutions). False detections occurred at the average rate of 2.3/h (4.1, 1.0, 2.7 in the 3 institutions), with fluctuations that reflected largely the technical quality and level of supervision of the recordings. The results are similar to those obtained in the commonly used method of epilepsy monitoring in adults and allow us to envisage clinical application.

Reliability of parental recall of developmental milestones

The developmental history is a key element of the pediatric evaluation. Precise determination of a childs milestones depends on the parents recollection of developmental events. As part of a prospective study examining neonatal predictors of outcome, healthy and high-risk newborns were assessed by a neurologist at 1, 3, and 5 years of age. The consistency of a parents recall of two milestones (i.e., first steps and first meaningful word) was examined at 3 and 5 years. The answers provided by the parent(s) at the first visit (mean/S.D.: 12.67/1.19 months) provided the comparison standard. Twenty-five healthy and 50 high-risk infants had acquired 1 or both developmental skills by the first visit. Of 75, 63 were reexamined at 3 and/or 5 years by the same neurologist. For first steps, excellent correlations were obtained at 3 years (r = 0.74) and modest correlations were determined at 5 years (r = 0.41). The mean difference in age reported between 1 and 3 years was 0.36 +/- 1.81 months, and 0.33 +/- 2.46 months between 1 and 5 years. For first word, correlations were poor at 3 and 5 years (r = 0.27 and r = -0.11, respectively), with mean discrepancies in recall of 2.43 +/- 4.37 months (3-year visit) and 2.74 +/- 5.56 months (5-year visit). Twenty percent (n = 13) of parents reported discrepancies of 6 or more months (mean: 9.4; range: 6-25 months). Furthermore, this error in reporting of first word was later than the correct age in all patients.(ABSTRACT TRUNCATED AT 250 WORDS)

PROGNOSTIC SIGNIFICANCE OF THE AUDITORY BRAINSTEM EVOKED RESPONSE HIGH‐RISK NEONATES

The prognostic significance of the auditory brainstem evoked response (ABR) was examined in this prospective study of neonates at risk for neurodevelopmental sequelae. ABR testing was performed in the neonatal period (37 to 45 weeks conceptional age) and at two and/or six months corrected age. 34 high‐risk newborns and 14 controls were followed to one year of age, when they received neurological and developmental assessments. Increased I to III and I to V interwave latencies predicted gross motor delay at one year, with a positive predictive power of 83 per cent and a specificity of 94.7 per cent. Increased brainstem conduction, dispersal of waves III to V and V/I amplitude ratio abnormalities predicted abnormal neurological findings at one year, with positive predictive values of 100, 100 and 80 per cent, respectively. A standard clinical examination of the newborn, performed on the at‐risk and control infants at 40 weeks conceptional age, was not found to be strongly predictive of neurodevelopmental deficits at one year.

Prognostic significance of multimodality evoked response testing in high-risk newborns ☆

Exposure to hypoxic-ischemic events in fetal or neonatal life may lead to permanent brain damage and subsequent neurodevelopmental deficits. Clinical and diagnostic tools have been somewhat helpful in identifying an at-risk group, particularly those patients sustaining significant neurologic sequelae. In this prospective study, the prognostic significance of multimodality evoked responses in high-risk newborns was examined. A group of 44 high-risk newborns, as well as 14 healthy newborns, were tested during the newborn period with auditory brainstem responses and somatosensory evoked responses; these tests were repeated at 2 and 6 months corrected age. A neonatal neurologic examination, the Einstein Neonatal Neurobehavioral Assessment Scale, was also conducted. At 1 year corrected age, both groups were assessed in a blind fashion by a pediatric neurologist and a psychologist to determine neurodevelopmental outcome. Results indicated that somatosensory evoked response abnormalities in particular predict an abnormal neurologic status at 1 year of age. Abnormalities that persisted or worsened correlated with severe neurologic impairment, whereas an abnormal somatosensory evoked response that improved or normalized in infancy was associated with mild to moderate neurologic sequelae. Increased brainstem conduction in the auditory brainstem responses was also associated with neurologic sequelae. Normal findings from auditory brainstem responses and somatosensory evoked responses predicted normal developmental scores in all areas, as well as a normal neurologic outcome at 1 year with negative predictive powers ranging from 85-100%. Evoked response testing appears to be an important adjunct to the neurologic investigation of high-risk newborns.

Lesionectomy of MRI detected lesions in children with epilepsy.

The results of complete excision of cerebral lesions detected by MRI in 18 children presenting with epilepsy were analyzed. There were 14 boys and 4 girls with a mean age of 9.2 years. The average age of onset of seizures was 6.8 years. The mean time from onset of seizures to surgery was 2.3 years. Often, CT scans suggested that the lesions were indolent. MRI was better in differentiating neoplastic from developmental lesions. Angiography was non-contributory in this series. Interictal EEGs showed epileptiform activity correlating with imaging studies in 54% of children. The lesion was completely surgically excised in all patients. This was confirmed by intra-operative ultrasound and postoperative imaging. Electrocorticography was performed prior to and after the resection, but residual spiking did not lead to further resection. The average postoperative follow-up was 5.7 years. Five patients had low grade astrocytomas, 4 had gangliogliomas, 1 a mixed astrocytoma-oligodendroglioma, 3 had cortical dysplasia, 2 infantile desmoplastic gangliogliomas, 2 hamartomata, and 1 cavernous angioma. Sixteen patients have been seizure-free since surgery. Only 2 have partial seizures. Thus, all patients benefited from the resection, with respect to seizure control. In those with temporal lobe lesions, improvement in IQ was seen postoperatively. Early consideration of surgery in patients with epilepsy and lesions demonstrated by MRI is suggested.