- Email: [email protected]
Bilateral loss of cortical somatosensory evoked potential at birth predicts cerebral palsy in term and near-term newborns
Early Human Development 86 (2010) 93–98
Contents lists available at ScienceDirect
Early Human Development j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / e a r l h u m d ev
Bilateral loss of cortical somatosensory evoked potential at birth predicts cerebral palsy in term and near-term newborns A. Suppiej a,⁎, A. Cappellari a, M. Franzoi a, A. Traverso a, M. Ermani b, V. Zanardo c a b c
Child Neurology and Clinical Neurophysiology Unit, Department of Paediatrics, University of Padova, Italy Bio statistical Unit, Department of Neurosciences, University of Padova, Italy Neonatal Intensive Care Unit, Department of Paediatrics, University of Padova, Italy
a r t i c l e
i n f o
Article history: Received 16 July 2009 Received in revised form 21 January 2010 Accepted 29 January 2010 Keywords: Encephalopathy Cerebral palsy SEP Neonate Near-term Term
a b s t r a c t Bilateral loss of cortical somatosensory evoked potential (SEP) is considered the single best indicator of adverse outcome in acute encephalopathy of adult patients and older children. This study determines whether the presence or absence of the neonatal cortical SEP can predict cerebral palsy at two years in survivors of neonatal encephalopathy scored according to Sarnat criteria. We also compare SEPs with visual evoked potentials (VEPs), the EEG and neonatal neurological status. Fifty-nine neonates admitted to the neonatal intensive care unit had SEP, VEP and EEG recordings analysed according to the presence (n = 37, 63%) or absence (n = 22, 37%) of neonatal encephalopathy (score ≥ 1). Cortical SEP was always present in the perinatal period in those surviving without major neurological disability, while it was bilaterally absent in all but one patient with a subsequent diagnosis of cerebral palsy. Multivariate analysis using the logistic regression model showed that bilateral loss of cortical SEP and Sarnat Score correctly classified the neurological outcome in all patients. Bilateral absence of cortical SEP indicates early identification of neonates at risk of cerebral palsy indicating that EPs have a clinical role in the workup of neonatal encephalopathy. © 2010 Elsevier Ireland Ltd. All rights reserved.
1. Introduction Evoked potentials (EPs) are electrical potential changes of sensory receptors, neural pathways and the brain following external stimuli, they reflect activity of the corresponding sensory pathways ascending to cerebral cortex following sensory input. Evoked potential studies can be performed at bedside in the neonatal intensive care unit (NICU); similarly to electroencephalography they can be regarded as non-invasive techniques. In adults and older children the clinical role of EPs in neurological prognosis of coma is well established. The single best indicator of adverse outcome (death or severe neurological impairment) is severe abnormality of median nerve somatosensory evoked potential (SEP) consisting in absence of the N20 cortical component. The results of a meta-analysis of 41 studies show that adult comatose patients with absent somatosensory evoked potential responses have less than 1% chance of awakening [1]. Bilateral loss of cortical SEP (BLC-SEP) in comatose children has a positive predictive value of 100% in post anoxic aetiology [2]. Failure to elicit the N20 cortical SEP component or its severe latency prolongation in infancy has been also shown to be predictive of developmental disturbance and especially motor ⁎ Corresponding author. Via Giustiniani n. 3, 35128 Padova, Italy. Tel.: +39 049 8213505; fax: +39 049 8213509. E-mail address: [email protected] (A. Suppiej). 0378-3782/$ – see front matter © 2010 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.earlhumdev.2010.01.024
handicap. Although SEPs do not give direct evidence about the state of the motor pathways, they have proved to be a valuable early indicator of severe motor impairment if they remain abnormal in follow-up examinations during the first 4 weeks of life [3]. Neonatal encephalopathy is a significant cause of infant mortality and morbidity and a risk factor for neurological sequelae in the survivors of neonates admitted to Neonatal Intensive Care Units (NICU) [4,5]. A wide spectrum of neurodevelopment problems and intellectual delay has been reported in survivors of neonatal encephalopathy, of which the most devastating and permanent outcome is cerebral palsy (CP). The definitive diagnosis of CP is usually made by the age of 2 years, by observation of the characteristic signs of abnormal control of movement or posture, muscle weakness, ataxia, and rigidity, frequently accompanied by sensory impairment, and cognitive limitation, earlier diagnosis of this condition is warranted in view of potential benefit of family counselling at discharge from the hospital and early rehabilitation programmes. A classification of neonatal signs and symptoms of neonatal encephalopathy was described by Sarnat and Sarnat [6] in which: Sarnat 1 is characterized by hyperalertness, uninhibited Moro and stretch reflexes and sympathetic effects, Sarnat 2 is marked by obtundation, hypotonia, strong distal flexion, and multifocal seizures and Sarnat 3 includes stupor, flaccidity and suppression of brain stem and autonomic functions. This grading system has helped to define the neonatal encephalopathy categories as mild, moderate or severe
94
A. Suppiej et al. / Early Human Development 86 (2010) 93–98
(stages 1, 2 or 3) and is clinically useful for prognosis. It is generally agreed that the combination of clinical neurological scores and EEG background activity allows accurate prognosis of a good neuromotor outcome in normal or mildly abnormal cases and a poor outcome in severe cases. However, defining the outcome may be difficult in cases with moderate abnormalities in both the early neurological examination and the EEG [6,7]. Furthermore in the clinical setting the reliability of neurological examination and the EEG may be reduced for coexisting medical complications. The prognostic role of BLC-SEP has not yet been evaluated in the neonatal period, although the prognostic role and technical feasibility of EPs in neonatal intensive care units have been studied [for a review see 8,9]. In the neonatal period EP recordings are hampered by the need for a dedicated methodology taking into account receptor, peripheral and central nervous system immaturity, specially trained technicians and an adequate understanding of the normal waveforms at different post conceptional ages [10]. Though the cortical SEP of the term neonate is very variable in form and latency between subjects and normal neurophysiology differs in near-term in comparison to term infants, a flat trace due to the loss of cortical response is easy to interpret in both cases and could be a robust indicator of abnormality [10,11]. The primary aim of the present study was to determine if the presence or absence of the cortical SEP response could predict the occurrence of cerebral palsy in term and near-term survivors of neonatal encephalopathy. Additionally, somatosensory evoked potentials were compared with visual evoked potentials (VEPs), the electroencephalogram (EEG) and the neurological status during the first days of life classified according to Sarnat and Sarnat [6]. 2. Materials and methods 2.1. Subjects All term and near-term neonates (34/0–36//7 gestational weeks) delivered at the Obstetric Clinic of Padua University (Italy) between January 2003 and December 2006, who required positive pressure resuscitation or were affected by respiratory failure at birth, and were consecutively admitted to third level NICU of the Paediatric University Hospital of Padua, were eligible for this neurophysiological study. Twelve out of 79 eligible patients were excluded from analysis because of failure to complete EEG, VEP or SEP recordings: 3 died in the first week of life, 5 were transferred to second level NICU, and the parents of 4 refused neurophysiological testing. Another 8 infants were lost at follow-up. Thus, 59 neonates were finally included in the study analysis. Subjects were subsequently classified in two groups according to presence or absence of neonatal encephalopathy (NE) defined and graded according to Sarnat and Sarnat criteria [6]. The ethical committee of the hospital approved the study. 2.2. Neonatal neurophysiological testing A “Galileo NT” system (EBNeuro/Florence Italy) was used for EEG and EP recordings. EPs were performed when neonates were clinically stable, whilst lying in their cribs; the skin temperature of the stimulated limb was monitored and kept constant throughout the recording session. The behavioural state of the neonate was monitored by behavioural scoring [9]. SEPs were recorded with the method of our laboratory [10] after the first week of life, because false positive results have been reported in the first week of life [12]. In brief, SEP responses were elicited by right and left median nerve stimulation with electrical pulses at motor threshold intensity, rate 0.5 Hz, duration 0.2 ms, time base 100 ms. The N13 spinal component was recorded at the 7th cervical vertebra with Fz reference while the cortical N20 component at the central
location contralateral to stimulation referred to the ipsilateral C3 or C4. Recordings were done in active sleep or awake state. Only the N20 component in the presence of a normal cervical potential N13 was considered for analysis and scored as follows: 0 bilaterally recorded N20 cortical component; 1 unilateral absence of N20 cortical component; 2 bilateral absence of N20 cortical component. VEPs were elicited by binocular white flashes, rate 0.5 Hz, filters 1–200 Hz, time base 1000 ms, recorded at Oz, and midway between Oz and left and right mastoids all referred to Fz; neonates were tested in the awake state, after the first week of life [12]. The large positive wave at around 200 ms (P200) was considered for analysis. Results were scored as follows: 0 presence of the P200 at Oz; 1 left/right occipital asymmetry N50%; 2 absence of response. EEGs were recorded after the first 8 h of life [13]. A 16 channel montage was used to record 12 channels of EEG using the 10–20 International System of electrode placement modified for neonates (F4, C4, T4, P4, O2, Fz, Cz, F3, C3, T3, P3, O1), plus four polygraphic channels (electrooculogram, sub-mentalis electromyogram, electrocardiogram and pneumogram).The EEG background activity of each patient was scored according to Pressler et al. [13]: 0 when normal for gestational age or mildly abnormal (mild asymmetries, mild voltage reduction); 1 when moderately abnormal (discontinuous activity with interburst intervals (IBI) of less than 10 s, other types of continuous activity, clear asymmetry or asynchrony); 2 when showing major abnormalities (IBI of 10–60 s, severe depression, no wake–sleep cycles); 3 when the EEG was inactive or background activity was less than 10 μV and IBI greater than 60 s. All EEG and EP recordings were interpreted by the same neurophysiologist (A.S.) and agreed independently by a second neurophysiologist (A.C.). 2.3. Follow-up Sequential follow-up examinations were performed for all infants up to a median age of twenty-eight months (range 18–36). Each evaluation included a medical history and a structured neurological examination [14] carried out by a child neurologist (M.F.) blind to the neurophysiological results. Hearing was tested with auditory evoked potentials [15] before leaving the NICU or during the first 3 months after leaving hospital. Results of the routine ophthalmologic followup, as well as occurrence of epilepsy were also recorded. The outcome was defined on the basis of neurological findings according to Amiel-Tison and Stewart [14] at the last follow-up, brain stem auditory evoked potentials assessment of hearing and the results of the routine opthalmological follow-up, as follows: 1 normal or impaired with no disability (minor abnormalities of tone and reflexes and high tone sensorineural hearing loss which did not require aiding) constituted the no cerebral palsy group (no-CP group); 2 impaired with disability (cerebral palsy, sensorineural hearing loss requiring aiding, registered as partially sighted or blind, epilepsy) constituted the cerebral palsy group (CP group). 2.4. Statistical analysis Normally distributed variables were analysed using the Student T test in the case of two independent groups or the Analysis of Variance
A. Suppiej et al. / Early Human Development 86 (2010) 93–98
in the case of more than two groups. Mann–Witney U and Kruskal Wallis test were used for ordinal variables. The chi square test was used for qualitative non-ordinal variables. Multivariate analysis was done using the logistic regression model to evaluate if the outcome as dependent variable could be improved combining dichotomised SEP and Sarnat score as independent variables. The significance was set at p b 0.05. All analyses used STATISTICA 6.0 for Windows (StatSoft, Tulsa, OK).
3. Results 3.1. Demographic and neonatal data The demographic and clinical characteristics of the 59 neonates studied are summarized in Table 1. Thirty seven out of 59 were affected by neonatal encephalopathy (NE-group) while 22 constituted the no-neonatal encephalopathy (no-NE) group. NE and no-NE groups were significantly different only with respect to 5′ Apgar score, pH, multi-organ failure, and the occurrence of neonatal seizures. In the NE-group, 21 neonates (57%) were classified as Sarnat 1, 10 (27%) Sarnat 2, and 6 (16%) Sarnat 3. A significant degree of linear correlation (rho = 0.33, p = 0.048) was found between severity of Sarnat score and time in hospital. Neonatal seizures were effectively treated with phenobarbital using a loading dose of 20 mg/kg and maintenance dose of 5 mg/kg per day, plasma levels were within therapeutic range in all patients. SEPs and VEPs were recorded at a median age of 13 days (range 7– 21 days), EEGs at a median age of 48 h (range 10–72 h). EEGs and VEPs were reliably recorded in all patients, SEPs in all but one whose data were excluded from analysis because of artefacts. Independent agreement between two neurophysiologists occurred in interpretation of all EP traces while disagreement occurred in the interpretation of two EEG traces, the final score being assigned after discussing the cases. Neonatal neurophysiological abnormalities are summarized in Table 2. All children in the no-NE-group had, in the neonatal period, a normal or mildly abnormal EEG (score 0), VEP and SEP cortical components were recorded in all of this group, 4 neonates showed left/right occipital asymmetry of the VEPs. Examples of individual SEP and VEP representative waveforms in one neonate without encephalopathy and in three neonates affected by encephalopathy Sarnat1, 2 and 3 are shown in Figs. 1 and 2, respectively.
95
Table 2 Neurophysiological abnormalities in the neonatal encephalopathy (NE) and noneonatal encephalopathy (no-NE) groups. NE-group Test
Type of Sarnat 1 Sarnat 2 Sarnat 3 Total N (%) abnormalities (N = 21) (N = 10) (N = 6) N (%) (N = 37) (N=22)
EEG 0 (normal 17 score and mild) 1 (moderate) 4 2 (major) 0 3 (severe) 0 Present 19 SEPa bilaterally Absent 1 unilaterally Absent 1 bilaterally VEP Present: no 18 asymmetry O1/O2 Present: yes 3 asymmetry O1/O2 Absent 0 a
No NEgroup
4
0
21 (57%)
3 2 1 6
0 1 5 0
7 (19%) 3 ( 8%) 6 (16%) 25 (68%)
0
1
2 ( 5%)
22 (100%) 0 0 0 22 (100%) 0
3
5
9 (24%)
0
6
2
26 (70%)
2
0
5 (13%)
4 (18%)
2
4
6 (16%)
0
18 (82%)
No data available for one patient.
3.2. Outcome None of the infants affected by neonatal encephalopathy was subsequently diagnosed as suffering from genetic or metabolic syndromes. Nine out of 37 children with neonatal encephalopathy were classified in the CP outcome group; the clinical details are shown in Table 3.Out of the 12 patients who had neonatal seizures three, all in the CP group, developed epilepsy in the first six months of life. The remaining 28/37 children constituted the no-CP outcome group; the routine ophthalmological evaluation did not disclose visual deficits in all, none developed epilepsy, one had mild sensorineural hearing loss which did not require aiding. Minor abnormalities of tone and reflexes were however observed in 5/28. At the clinical follow-up none of the children in the no-NE group showed motor deficits, none had hearing loss or brainstem dysfunction shown by auditory evoked potentials, none had visual deficits. 3.3. Outcome and neonatal data
Table 1 Demographic and perinatal clinical data in the neonatal encephalopathy (NE) and noneonatal encephalopathy (no NE) groups. No
Gender (male), n (%) Birth weight, grams (mean ± SD) Gestational age, weeks(mean ± SD) Apgar score 5′ ≤ 3, n (%) Type of delivery (vaginal), n (%) Dystocic vaginal delivery, n (%) Emergency caesarean, n (%) Resuscitation, n (%) pH b 7, n (%)* BE N 12, n (%)* MOF, n (%) Neonatal seizures, n (%) Time in hospital, days (mean ± SD)
Neonatal encephalopathy
Neonatal encephalopathy
p
(37 cases)
(22 cases)
19/37 (51%) 2915 ± 574 38.3 ± 2.33 6/36 (17%)
13/22 (59%) 3050 ± 604 38.18 ± 2.04 0/22 (0%)
ns ns ns 0.04
15/37 (41%) 6/15 (40%) 19/22 (86%) 34/37 (92%) 9/37 (24%) 19/37 (51%) 23/37 (62%) 12/37 (32%) 15 ± 11
10/22 (45%) 4/10 (40%) 9/12 (75%) 17/21 (81%) 0/18 (0%) 10/18 (56%) 3/21 (14%) 0/22 (0%) 10 ± 6
ns ns ns ns 0.022 ns 0.0004 0.002 ns
BE = base excess, * within the first hour of life, MOF = multi-organ failure (defined as at least one organ liver, kidney, heart involvement in adjunct to the brain).
BLC-SEP was observed in all but one cases affected by CP, the exception being a patient with Sarnat 1 encephalopathy and EEG score 0. Unilateral loss of cortical SEP was observed in two patients (one in the no-CP outcome group and one affected by hemiplegia). Cortical SEPs were bilaterally present in all children not developing cerebral palsy (no false negative results). All 21 children with neonatal encephalopathy Sarnat 1 were classified in the no-CP outcome group, while Sarnat 3 (6 patients) was always associated with cerebral palsy (Table 3, patients ID 2–7). The outcome of the 10 neonates with grade 2 encephalopathy however could not be predicted only on clinical basis (7 no-CP group versus 3 CP group—Table 3, patients ID 1, 8, 9). A multivariate analysis using the logistic regression model was performed to evaluate if the outcome as dependent variable could be improved combining dichotomised SEP (scores 1 and 2 vs score 0) and Sarnat score as independent variables. We found that this combination correctly classified the outcome in all patients (p b 0.0001). All patients with absent VEP were affected by CP and were partially sighted at follow-up. VEP asymmetry (score 1) was observed in children without CP. All NE patients with a score 0 or 1 of neonatal
96
A. Suppiej et al. / Early Human Development 86 (2010) 93–98
Fig. 1. Examples of SEP in one neonate without encephalopathy (A) and in three neonates affected by neonatal encephalopathy Sarnat 1 (B), 2 (C) and 3 (D).
EEG were in the no-CP group, while all those with a score of 2 or 3 were in the CP outcome group. The clinical outcome, EEG, SEP and VEP scores in patients with Sarnat 2 encephalopathy are summarized in Table 4, it shows that only patients with bilaterally absent SEPs and EEG score 2 or 3 were affected by CP. 4. Discussion This study found that cortical SEP was always present in neonates surviving without major neurological disability, while it was bilater-
ally absent in all but one patient demonstrating by two years of age definite neurological signs consistent with the diagnosis of cerebral palsy. These data extend to neonatal age previous findings of the role BLC-SEP in adults and children [1,2]. As in infants and older children [16,17], but differently from adults [1], the predictive value of a normal SEP appears in neonates to be superior to that of an absent SEP. The higher rate of false positive cases in some neonatal studies, as well as in the one patient in the present report as compared with adult studies suggests that in the neonatal period clinical and methodological factors may affect the reliability of the responses. Indeed it has been shown that the number of stimuli,
Fig. 2. Examples of VEP in one neonate without encephalopathy (A) and in three neonates affected by neonatal encephalopathy respectively Sarnat 1 (B), 2 (C) and 3 (D).
A. Suppiej et al. / Early Human Development 86 (2010) 93–98 Table 3 Details of the outcome of the 9 patients affected by cerebral palsy. ID Outcome 1. 2. 3. 4. 5. 6. 7. 8. 9.
Tetraparetic cerebral palsy, epilepsy. Hemiplegic cerebral palsy. Tetraparetic cerebral palsy, sensorineural hearing loss, partially sighted. Tetraparetic cerebral palsy, partially sighted. Tetraparetic cerebral palsy, partially sighted. Tetraparetic cerebral palsy, epilepsy, partially sighted. Tetraparetic cerebral palsy, sensorineural hearing loss, partially sighted. Tetraparetic cerebral palsy, epilepsy, sensorineural hearing loss, partially sighted. Diplegic cerebral palsy, partially sighted.
ID: identification number.
stimulation rate and filter settings have to be adjusted to take account of habituation and fatigue of the immature neonatal nervous system and the increased low-frequency content of the responses (for a review see [10]). The vigilance level may also affect the cortical components, but differences are not apparent during active sleep and the awake state [18]. Age related topographic changes need also to be taken into account, particularly before term. The electrode position at C3 (or C4) adopted in the present study is not the best location for the N20 recording in more mature infants, since it is near the isopotential area between the frontal P20 and the parietal N20, but this location was more suitable for near-term infants because of the movement of the N20 maximal amplitude from frontal to parietal sites related to the posterior displacement of the central sulcus consequent to the development of the frontal lobe, between 31 and 40 weeks of gestational age [19]. In future, more sophisticated applications of SEPs in neonates, should include both central and parietal (P3 or P4) locations. A limitation of the present study is the absence of a comparison group of healthy neonates from postnatal wards. However we think that this will not have affected the overall results since it is well known that the cortical SEP is always recorded in normal neonates under recording conditions comparable to those adopted in the present study [10]. Interestingly enough, the combination of SEP and Sarnat score correctly predicted the neurological outcome in all our patients. Neonatal encephalopathy was associated in our patients with low Apgar scores and major organ system injury, while other causes were ruled out, hence, our results apply to patients in whom encephalopathy was related to perinatal asphyxia. Indeed, assessing the proportion of neonatal encephalopathy that is due to asphyxia during birth is difficult because of problems in defining asphyxia and in attributing the cause of Table 4 Scores of severity of SEP, VEP, EEG and outcome in patients with Sarnat 2 neonatal encephalopaty. SEP: 0 = bilaterally recorded N20 cortical component; 2 = bilateral absence of N20 cortical component. VEP: 0 = presence of the P200 at Oz; 1 = left/right occipital asymmetry N 50%; 2 = absence of response. EEG: 0 = normal for gestational age or mildly abnormal (mild asymmetries, mild voltage reduction); 1 = discontinuous activity with inter-burst intervals (IBI) of less than 10 s, other types of continuous activity, clear asymmetry or asynchrony; 2 = IBI of 10–60 s, severe depression, no wake–sleep cycles; 3 = inactive EEG or background activity less than 10 μV and IBI greater than 60 s. Outcome: 1 = no cerebral palsy; 2 = cerebral palsy. ID
SEP
VEP
EEG
Outcome
10 11 12 13 1 14 15 8 16 9
0 0 0 0 2 0
0 0 1 1 0 0 0 2 0 2
1 0 1 1 2 0 0 3 0 2
1 1 1 1 2 1 1 2 1 2
a
a
2 0 2
Data excluded because of artefact.
97
neonatal neurological illness. Perinatal hypoxic ischemic brain injury has been reported in up to 90% of infants with neonatal encephalopathy [20–22], however only 21% of the neonates showing a neonatal neurological syndrome [23] satisfy the standard criteria for intrapartum asphyxia defined by the American College of Obstetricians and Gynaecologists and the American Academy of Pediatrics [24,25]. An early-onset encephalopathy in infants born ≥34 weeks' gestation with low Apgar scores is considered the best single criteria for occurrence of acute perinatal asphyxia potentially severe enough to cause brain damage [26]. Recent studies based on systematic use of magnetic resonance imaging (MRI) suggest that the brain areas preferentially involved following a prolonged period of mild to moderate hypotension lie between the territories of the circulation of the anterior, middle and posterior cerebral arteries, while damage which occurs as a result of a short period of profound hypoxic ischemia causes injury to the posterior putamen of the lentiform nuclei, the ventrolateral nuclei of the thalami, the hippocampus and the peri-rolandic cortex because of their high metabolic rate [27]. The functional status of the above anatomical locations of brain injury can be explored by means of evoked potentials (EPs). SEP abnormalities may point to dysfunction of sub-cortical pathways ascending to cortex including basal ganglia and internal capsule as well as to cortical areas located in the border zones between anterior and middle cerebral arteries. In agreement with previous finding [7,27], we found that the severity of the neurological syndrome graded with the Sarnat and Sarnat score [6] accurately predicted the outcome in severe and mild cases, but not in those of intermediate severity. In the present study VEPs were found not useful in neurological prognosis, when cortical response was present this finding did not guarantee a good outcome. The absence of cortical response was observed only in some patients, all developing cerebral palsy, in agreement with other authors showing that all neonates with abnormal VEPs throughout the first week of life, or with absent VEPs, either died or were left with neurological sequelae [28,29]. VEPs may have critical role in visual prognosis: all our patients with absent VEP developed visual deficit, in agreement with other authors' findings [30]. The use of neonatal electroencephalogram as a prognostic tool in neonatal medicine is widely recognised. In hypoxic–ischaemic encephalopathy a normal EEG has been associated with a favourable outcome while a severely abnormal background activity was invariably associated with a poor outcome. Prognosis remains uncertain in cases of moderately abnormal EEGs. In the attempt to improve the prognostic role of EEG different timing of the first recording, repeated recordings in the first hours or days of life and different scoring systems have been studied [31–38]. A higher predictive value was documented by Pressler et al. [13] when testing was done between 8 and 72 h of life. The score of EEG severity proposed by these authors and adopted in the present study and the temporal window of 10–72 h seem to accurately predict neurological examination at two years of age also in patients with moderate encephalopathy and may explain the excellent prognostic role of the EEG in prognostication of cerebral palsy. In agreement with other studies [39], the use of phenobarbital (maintained in the therapeutic range) in neonates with seizures did not affect the prognostic value of EEG recordings. From our knowledge the present study is the first that has analysed the predictive value of BLC-SEP and that of the neonatal neurological syndrome severity, VEP and the EEG in the same population. Our results support the role of BLC-SEP and the EEG in early identification of patients at risk of cerebral palsy following neonatal encephalopathy. The functional status of the central nervous system can be assessed in NICU admitted neonates at bedside, non invasively and at low cost using the EEG and SEPs. Although in the present study SEP did not pick up additional abnormalities in comparison with EEG results, in daily clinical practice these neurophysiological tools can support the clinical examination
98
A. Suppiej et al. / Early Human Development 86 (2010) 93–98
of patients affected by neonatal encephalopathy but, it is important to implement a cost effective strategy. Considering the different timings for prediction of outcome, SEP at 7 days of age and EEG 1–5 days of age, they appear complimentary investigations. The high reliability of BLC-SEP apply to patients tested after the first week of life, considerably later than the therapeutic window for neuroprotection, at present based mainly on hypothermia; thus for this purpose the neurophysiological technique of choice remains the EEG. By contrast SEPs may offer complementary data at a later stage. Indeed research suggest that the interpretation of severe EEG patterns such as electrocerebral silence and burst suppression may become complex in patients treated with intravenous anaesthetic agents with depressant effect on cortical metabolism or in the occurrence of metabolic disturbances; in fact in the absence of cortical damage the above EEG patterns are reversible in parallel with improvement of clinical status [9]. Furthermore we think that it may be easier to analyse SEP waveform for presence or absence of the N20 cortical component than to classify an EEG abnormality as being of intermediate severity. Considering that in the present study Sarnat score and dichotomised SEP correctly classified the outcome in all patients, a combination of clinical evaluation and SEP could be a valid alternative to undertaking repeated EEGs in selected cases. In conclusion the data of the present study support the role of BLC-SEP in early identification of neonates at risk of cerebral palsy and should be included in the clinical workup of term and near-term encephalopathy. Acknowledgements The authors thank the technicians Maria Tesone, Elena Bizzaro and Loretta Peppato.The project has been supported by a grant from the Italian Ministry of Health, Project No. 0AN/F. There are no known potential conflicts of interest. References [1] Robinson LR, Micklesen PJ, Tirschwell DL, Lew HL. Predictive value of somatosensory evoked potentials for awakening from coma. Crit Care Med 2003;31:960–7. [2] Wolrab G, Bolthauser E, Schmitt B. Neurological outcome in comatose children with bilateral loss of cortical somatosensory evoked potentials. Neuropediatrics 2001;32:271–4. [3] Görke W. Somatosensory evoked cortical potentials indicating impaired motor development in infancy. Dev Med Child Neurol 1986;28:633–41. [4] Finer NN, Robertson CM, Peters KL, Coward JH. Factors affecting outcome in hypoxic–ischemic encephalopathy in term infants. Am J Dis Child 1983;137:21–5. [5] Marlow N, Rose AS, Rands CE, Draper ES. Neuropsychological and educational problems at school age associated with neonatal encephalopathy. Arch Dis Child: Fetal and Neonatal Ed 2005;90:380–7. [6] Sarnat HB, Sarnat MS. Neonatal encephalopathy following fetal distress. A clinical and elecroencephalographic study. Arch Neurol 1976;33:696–705. [7] Van Handel M, Swaab H, de Vries LS, Jongmans MJ. Long term cognitive and behavioral consequences of neonatal encephalopathy following perinatal asphyxia: a review. Eur J Paediatr 2007;166:645–54. [8] Suppiej A. Evoked potentials in neonatal hypoxic–ischaemic encephalopathy: review of the literature. Ann Ist Super Sanita 2001;37:515–25. [9] Pressler R, Bady B, Binnie CD, Boylan GB, Connell JA, Lütschg J, et al. Neurophysiology of the neonatal period. In: Binnie C, Cooper R, Mauguière F, Osselton JW, Prior PF, Tedman BM, editors. Clinical neurophysiology, Volume 2. EEG, paediatric neurophysiology, special techniques and applications. Amsterdam: Elsevier; 2003. p. 450–506. [10] Suppiej A. General characteristics of evoked potentials. In: Pressler R, Binnie CD, Cooper R, Robinson R, editors. Neonatal and paediatric clinical neurophysiology. Edinburgh: Churchill Livingstone,Elsevier; 2007. p. 111–54. [11] Gibson NA, Graham M, Levene MI. Somatosensory evoked potentials and outcome in perinatal asphyxia. Arch Dis Child 1992;67:393–8.
[12] Taylor MJ, Murphy WJ, Whyte HE. Prognostic reliability of somatosensory and visual evoked potentials of asphyxiated term infants. Dev Med Child Neurol 1992;34:507–15. [13] Pressler R, Boylan G, Morton M, Binnie C, Rennie J. Early serial EEG in hypoxic ischaemicencephalopathy. Clin Neurophysiol 2001;112:31–7. [14] Amiel-Tison C, Stewart A. Follow up studies during the first five years of life: a pervasive assessment of neurological function. Arch Dis Child 1989;64:496–502. [15] Suppiej A, Rizzardi E, Zanardo V, Franzoi M, Ermani M, Orzan E. Reliability of hearing screening in high-risk neonates: comparative study of otoacoustic emission, automated and conventional auditory brainstem response. Clin Neurophysiol 2007;118:869–76. [16] Mandel R, Martinot A, Delepoulle F, Lamblin MD, Laureau E, Vallee L, et al. Prediction of outcome after hypoxic–ischemic encephalopathy: a prospective clinical and electrophysiologic study. J Pediatr 2002;141:45–50. [17] Abend NS, Licht DJ. Predicting outcome in children with hypoxic ischemic encephalopathy. Pediatric Crit Care Med 2007;8:1–7. [18] Hrbek A, Hrbkova M, Leonard H. Somatosensory, auditory and visual evoked responses in newborn infants during sleep and wakefulness. Elettroencephalogr Clin Neurophysiol 1969;26:597–603. [19] Karniski W, Wyble L, Lease L, Blair RC. The late somatosensory evoked potential in premature and term infants. II. Topography and latency development. Electroencephalogr Clin Neurophysiol 1992;84:44–54. [20] Badawi N, Kurinczuk JJ, Keogh JM, Alessandri LM, O'Sullivan F, Burton PR, et al. Intrapartum risk factors for newborn encephalopathy: the Western Australian case control study. BMJ 1998;317:1554–8. [21] Cowan F, Rutherford M, Groenendaal F, Eken P, Mercuri E, Bydder GM, et al. Origin and timing of brain lesion in term infants with neonatalencephalopathy. Lancet 2003;361:736–42. [22] Pierrat V, Haouari N, Liska A, Thomas D, Subtil D, Truffert P, et al. Prevalence, causes and outcomes at 2 years of age of newborn encephalopathy: a populationbased study. Arch Dis Child 2005;90:257–61 Fetal and Neonatal Ed. [23] Korst LM, Phelan JP, Wang YM, Martin GI, Ahn MO. Acute fetal asphyxia and permanent brain injury: a retrospective analysis of current indicators. Matern Fetal Med 1999;8:101–6. [24] MacLennan A. A template for defining a causal relation between acute intrapartum events and cerebral palsy: international consensus statement. BMJ 1999;319:1054–9. [25] Hankins GD, Speer M. Defining the pathogenesis and pathophysiology of neonatal encephalopathy and cerebral palsy. Obstet Gynecol 2003;102:628–36. [26] Hill A. Current concepts of hypoxic–ischemic cerebral injury in term newborn. Pediatr Neurol 1991;7:317–25. [27] Rennie JM, Hagmann CF, Robertson NJ. Outcome after intrapartum hypoxic ischaemia at term. Seminars in fetal and neo med 2007;12:398–407. [28] Whyte HE, Taylor MJ, Menzies R, Chin KC, MacMillan LJ. Prognostic utility of visual evoked potentials in term asphyxiated neonates. Pediatr Neurol 1986;2:220–3. [29] Muttitt SC, Taylor MJ, Kobayashi JS, MacMillan L, Whyte HE. Serial visual evoked potentials and outcome in term birth asphyxia. Pediatr Neurol 1991;7:86–90. [30] McCulloch DL, Taylor MJ, Whyte HE. Visual evoked potentials and visual prognosis following perinatal asphyxia. Arch Ophthalmol 1991;109:229–33. [31] Watanabe K, Miyazaki S, Hara K, Hakamada S. Behavioral state cycles, background EEGs and prognosis of newborns with perinatal hypoxia. Electroencephalogr Clin Neurophysiol 1980;49:618–25. [32] Pezzani C, Radvanyi-Bouvet MF, Relier JP, Monod N. Neonatal electroencephalography during the first twenty-four hours of life in full-term newborn infants. Neuropediatrics 1986;17:11–8. [33] Takeuchi T, Watanabe K. The EEG evolution and neurological prognosis of neonates with perinatal hypoxia. Brain Dev 1989;11:115–20. [34] D'Allest AM, Andre M, Radvanji-Bouvet MF. Apport de l'Electroencephalogramme pour le Diagnostic et le Pronostic de l'Asphyxie Perinatal du Nouveau-né a Term. Arch Pediatr 1996;3:254–6. [35] Biagioni E, Mercuri E, Rutherford M, Cowan F, Azzopardi D, Frisone MF, et al. Combined use of electroencephalogram and magnetic resonance imaging in fullterm neonates with acute encephalopathy. Pediatrics 2001;107:461–8. [36] Zeinstra E, Fock JM, Begeer JH, van Weerden TW, Maurits NM, Zweens MJ. The prognostic value of serial EEG recordings following acute neonatal asphyxia in full-term infants. Eur J Paediatr Neurol 2001;5:155–60. [37] Douglass LM, Wu JY, Rosman NP, Stafstrom CE. Burst suppression electroencephalogram pattern in the newborn: predicting the outcome. J Child Neurol 2002;17: 403–8. [38] Khan RL, Lahorgue Nunes M, Garcias da Silva LF, da Costa JC. Predictive value of sequential electroencephalogram (EEG) in neonates with seizures and its relation to neurological outcome. J Child Neurol 2008;23:144–50. [39] Staudt F, Scholl ML, Coen RW, Bickford RB. Phenobarbital therapy in neonatal seizures and the prognostic value of the EEG. Neuropediatrics 1982;13:24–33.
Contents lists available at ScienceDirect
Early Human Development j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / e a r l h u m d ev
Bilateral loss of cortical somatosensory evoked potential at birth predicts cerebral palsy in term and near-term newborns A. Suppiej a,⁎, A. Cappellari a, M. Franzoi a, A. Traverso a, M. Ermani b, V. Zanardo c a b c
Child Neurology and Clinical Neurophysiology Unit, Department of Paediatrics, University of Padova, Italy Bio statistical Unit, Department of Neurosciences, University of Padova, Italy Neonatal Intensive Care Unit, Department of Paediatrics, University of Padova, Italy
a r t i c l e
i n f o
Article history: Received 16 July 2009 Received in revised form 21 January 2010 Accepted 29 January 2010 Keywords: Encephalopathy Cerebral palsy SEP Neonate Near-term Term
a b s t r a c t Bilateral loss of cortical somatosensory evoked potential (SEP) is considered the single best indicator of adverse outcome in acute encephalopathy of adult patients and older children. This study determines whether the presence or absence of the neonatal cortical SEP can predict cerebral palsy at two years in survivors of neonatal encephalopathy scored according to Sarnat criteria. We also compare SEPs with visual evoked potentials (VEPs), the EEG and neonatal neurological status. Fifty-nine neonates admitted to the neonatal intensive care unit had SEP, VEP and EEG recordings analysed according to the presence (n = 37, 63%) or absence (n = 22, 37%) of neonatal encephalopathy (score ≥ 1). Cortical SEP was always present in the perinatal period in those surviving without major neurological disability, while it was bilaterally absent in all but one patient with a subsequent diagnosis of cerebral palsy. Multivariate analysis using the logistic regression model showed that bilateral loss of cortical SEP and Sarnat Score correctly classified the neurological outcome in all patients. Bilateral absence of cortical SEP indicates early identification of neonates at risk of cerebral palsy indicating that EPs have a clinical role in the workup of neonatal encephalopathy. © 2010 Elsevier Ireland Ltd. All rights reserved.
1. Introduction Evoked potentials (EPs) are electrical potential changes of sensory receptors, neural pathways and the brain following external stimuli, they reflect activity of the corresponding sensory pathways ascending to cerebral cortex following sensory input. Evoked potential studies can be performed at bedside in the neonatal intensive care unit (NICU); similarly to electroencephalography they can be regarded as non-invasive techniques. In adults and older children the clinical role of EPs in neurological prognosis of coma is well established. The single best indicator of adverse outcome (death or severe neurological impairment) is severe abnormality of median nerve somatosensory evoked potential (SEP) consisting in absence of the N20 cortical component. The results of a meta-analysis of 41 studies show that adult comatose patients with absent somatosensory evoked potential responses have less than 1% chance of awakening [1]. Bilateral loss of cortical SEP (BLC-SEP) in comatose children has a positive predictive value of 100% in post anoxic aetiology [2]. Failure to elicit the N20 cortical SEP component or its severe latency prolongation in infancy has been also shown to be predictive of developmental disturbance and especially motor ⁎ Corresponding author. Via Giustiniani n. 3, 35128 Padova, Italy. Tel.: +39 049 8213505; fax: +39 049 8213509. E-mail address: [email protected] (A. Suppiej). 0378-3782/$ – see front matter © 2010 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.earlhumdev.2010.01.024
handicap. Although SEPs do not give direct evidence about the state of the motor pathways, they have proved to be a valuable early indicator of severe motor impairment if they remain abnormal in follow-up examinations during the first 4 weeks of life [3]. Neonatal encephalopathy is a significant cause of infant mortality and morbidity and a risk factor for neurological sequelae in the survivors of neonates admitted to Neonatal Intensive Care Units (NICU) [4,5]. A wide spectrum of neurodevelopment problems and intellectual delay has been reported in survivors of neonatal encephalopathy, of which the most devastating and permanent outcome is cerebral palsy (CP). The definitive diagnosis of CP is usually made by the age of 2 years, by observation of the characteristic signs of abnormal control of movement or posture, muscle weakness, ataxia, and rigidity, frequently accompanied by sensory impairment, and cognitive limitation, earlier diagnosis of this condition is warranted in view of potential benefit of family counselling at discharge from the hospital and early rehabilitation programmes. A classification of neonatal signs and symptoms of neonatal encephalopathy was described by Sarnat and Sarnat [6] in which: Sarnat 1 is characterized by hyperalertness, uninhibited Moro and stretch reflexes and sympathetic effects, Sarnat 2 is marked by obtundation, hypotonia, strong distal flexion, and multifocal seizures and Sarnat 3 includes stupor, flaccidity and suppression of brain stem and autonomic functions. This grading system has helped to define the neonatal encephalopathy categories as mild, moderate or severe
94
A. Suppiej et al. / Early Human Development 86 (2010) 93–98
(stages 1, 2 or 3) and is clinically useful for prognosis. It is generally agreed that the combination of clinical neurological scores and EEG background activity allows accurate prognosis of a good neuromotor outcome in normal or mildly abnormal cases and a poor outcome in severe cases. However, defining the outcome may be difficult in cases with moderate abnormalities in both the early neurological examination and the EEG [6,7]. Furthermore in the clinical setting the reliability of neurological examination and the EEG may be reduced for coexisting medical complications. The prognostic role of BLC-SEP has not yet been evaluated in the neonatal period, although the prognostic role and technical feasibility of EPs in neonatal intensive care units have been studied [for a review see 8,9]. In the neonatal period EP recordings are hampered by the need for a dedicated methodology taking into account receptor, peripheral and central nervous system immaturity, specially trained technicians and an adequate understanding of the normal waveforms at different post conceptional ages [10]. Though the cortical SEP of the term neonate is very variable in form and latency between subjects and normal neurophysiology differs in near-term in comparison to term infants, a flat trace due to the loss of cortical response is easy to interpret in both cases and could be a robust indicator of abnormality [10,11]. The primary aim of the present study was to determine if the presence or absence of the cortical SEP response could predict the occurrence of cerebral palsy in term and near-term survivors of neonatal encephalopathy. Additionally, somatosensory evoked potentials were compared with visual evoked potentials (VEPs), the electroencephalogram (EEG) and the neurological status during the first days of life classified according to Sarnat and Sarnat [6]. 2. Materials and methods 2.1. Subjects All term and near-term neonates (34/0–36//7 gestational weeks) delivered at the Obstetric Clinic of Padua University (Italy) between January 2003 and December 2006, who required positive pressure resuscitation or were affected by respiratory failure at birth, and were consecutively admitted to third level NICU of the Paediatric University Hospital of Padua, were eligible for this neurophysiological study. Twelve out of 79 eligible patients were excluded from analysis because of failure to complete EEG, VEP or SEP recordings: 3 died in the first week of life, 5 were transferred to second level NICU, and the parents of 4 refused neurophysiological testing. Another 8 infants were lost at follow-up. Thus, 59 neonates were finally included in the study analysis. Subjects were subsequently classified in two groups according to presence or absence of neonatal encephalopathy (NE) defined and graded according to Sarnat and Sarnat criteria [6]. The ethical committee of the hospital approved the study. 2.2. Neonatal neurophysiological testing A “Galileo NT” system (EBNeuro/Florence Italy) was used for EEG and EP recordings. EPs were performed when neonates were clinically stable, whilst lying in their cribs; the skin temperature of the stimulated limb was monitored and kept constant throughout the recording session. The behavioural state of the neonate was monitored by behavioural scoring [9]. SEPs were recorded with the method of our laboratory [10] after the first week of life, because false positive results have been reported in the first week of life [12]. In brief, SEP responses were elicited by right and left median nerve stimulation with electrical pulses at motor threshold intensity, rate 0.5 Hz, duration 0.2 ms, time base 100 ms. The N13 spinal component was recorded at the 7th cervical vertebra with Fz reference while the cortical N20 component at the central
location contralateral to stimulation referred to the ipsilateral C3 or C4. Recordings were done in active sleep or awake state. Only the N20 component in the presence of a normal cervical potential N13 was considered for analysis and scored as follows: 0 bilaterally recorded N20 cortical component; 1 unilateral absence of N20 cortical component; 2 bilateral absence of N20 cortical component. VEPs were elicited by binocular white flashes, rate 0.5 Hz, filters 1–200 Hz, time base 1000 ms, recorded at Oz, and midway between Oz and left and right mastoids all referred to Fz; neonates were tested in the awake state, after the first week of life [12]. The large positive wave at around 200 ms (P200) was considered for analysis. Results were scored as follows: 0 presence of the P200 at Oz; 1 left/right occipital asymmetry N50%; 2 absence of response. EEGs were recorded after the first 8 h of life [13]. A 16 channel montage was used to record 12 channels of EEG using the 10–20 International System of electrode placement modified for neonates (F4, C4, T4, P4, O2, Fz, Cz, F3, C3, T3, P3, O1), plus four polygraphic channels (electrooculogram, sub-mentalis electromyogram, electrocardiogram and pneumogram).The EEG background activity of each patient was scored according to Pressler et al. [13]: 0 when normal for gestational age or mildly abnormal (mild asymmetries, mild voltage reduction); 1 when moderately abnormal (discontinuous activity with interburst intervals (IBI) of less than 10 s, other types of continuous activity, clear asymmetry or asynchrony); 2 when showing major abnormalities (IBI of 10–60 s, severe depression, no wake–sleep cycles); 3 when the EEG was inactive or background activity was less than 10 μV and IBI greater than 60 s. All EEG and EP recordings were interpreted by the same neurophysiologist (A.S.) and agreed independently by a second neurophysiologist (A.C.). 2.3. Follow-up Sequential follow-up examinations were performed for all infants up to a median age of twenty-eight months (range 18–36). Each evaluation included a medical history and a structured neurological examination [14] carried out by a child neurologist (M.F.) blind to the neurophysiological results. Hearing was tested with auditory evoked potentials [15] before leaving the NICU or during the first 3 months after leaving hospital. Results of the routine ophthalmologic followup, as well as occurrence of epilepsy were also recorded. The outcome was defined on the basis of neurological findings according to Amiel-Tison and Stewart [14] at the last follow-up, brain stem auditory evoked potentials assessment of hearing and the results of the routine opthalmological follow-up, as follows: 1 normal or impaired with no disability (minor abnormalities of tone and reflexes and high tone sensorineural hearing loss which did not require aiding) constituted the no cerebral palsy group (no-CP group); 2 impaired with disability (cerebral palsy, sensorineural hearing loss requiring aiding, registered as partially sighted or blind, epilepsy) constituted the cerebral palsy group (CP group). 2.4. Statistical analysis Normally distributed variables were analysed using the Student T test in the case of two independent groups or the Analysis of Variance
A. Suppiej et al. / Early Human Development 86 (2010) 93–98
in the case of more than two groups. Mann–Witney U and Kruskal Wallis test were used for ordinal variables. The chi square test was used for qualitative non-ordinal variables. Multivariate analysis was done using the logistic regression model to evaluate if the outcome as dependent variable could be improved combining dichotomised SEP and Sarnat score as independent variables. The significance was set at p b 0.05. All analyses used STATISTICA 6.0 for Windows (StatSoft, Tulsa, OK).
3. Results 3.1. Demographic and neonatal data The demographic and clinical characteristics of the 59 neonates studied are summarized in Table 1. Thirty seven out of 59 were affected by neonatal encephalopathy (NE-group) while 22 constituted the no-neonatal encephalopathy (no-NE) group. NE and no-NE groups were significantly different only with respect to 5′ Apgar score, pH, multi-organ failure, and the occurrence of neonatal seizures. In the NE-group, 21 neonates (57%) were classified as Sarnat 1, 10 (27%) Sarnat 2, and 6 (16%) Sarnat 3. A significant degree of linear correlation (rho = 0.33, p = 0.048) was found between severity of Sarnat score and time in hospital. Neonatal seizures were effectively treated with phenobarbital using a loading dose of 20 mg/kg and maintenance dose of 5 mg/kg per day, plasma levels were within therapeutic range in all patients. SEPs and VEPs were recorded at a median age of 13 days (range 7– 21 days), EEGs at a median age of 48 h (range 10–72 h). EEGs and VEPs were reliably recorded in all patients, SEPs in all but one whose data were excluded from analysis because of artefacts. Independent agreement between two neurophysiologists occurred in interpretation of all EP traces while disagreement occurred in the interpretation of two EEG traces, the final score being assigned after discussing the cases. Neonatal neurophysiological abnormalities are summarized in Table 2. All children in the no-NE-group had, in the neonatal period, a normal or mildly abnormal EEG (score 0), VEP and SEP cortical components were recorded in all of this group, 4 neonates showed left/right occipital asymmetry of the VEPs. Examples of individual SEP and VEP representative waveforms in one neonate without encephalopathy and in three neonates affected by encephalopathy Sarnat1, 2 and 3 are shown in Figs. 1 and 2, respectively.
95
Table 2 Neurophysiological abnormalities in the neonatal encephalopathy (NE) and noneonatal encephalopathy (no-NE) groups. NE-group Test
Type of Sarnat 1 Sarnat 2 Sarnat 3 Total N (%) abnormalities (N = 21) (N = 10) (N = 6) N (%) (N = 37) (N=22)
EEG 0 (normal 17 score and mild) 1 (moderate) 4 2 (major) 0 3 (severe) 0 Present 19 SEPa bilaterally Absent 1 unilaterally Absent 1 bilaterally VEP Present: no 18 asymmetry O1/O2 Present: yes 3 asymmetry O1/O2 Absent 0 a
No NEgroup
4
0
21 (57%)
3 2 1 6
0 1 5 0
7 (19%) 3 ( 8%) 6 (16%) 25 (68%)
0
1
2 ( 5%)
22 (100%) 0 0 0 22 (100%) 0
3
5
9 (24%)
0
6
2
26 (70%)
2
0
5 (13%)
4 (18%)
2
4
6 (16%)
0
18 (82%)
No data available for one patient.
3.2. Outcome None of the infants affected by neonatal encephalopathy was subsequently diagnosed as suffering from genetic or metabolic syndromes. Nine out of 37 children with neonatal encephalopathy were classified in the CP outcome group; the clinical details are shown in Table 3.Out of the 12 patients who had neonatal seizures three, all in the CP group, developed epilepsy in the first six months of life. The remaining 28/37 children constituted the no-CP outcome group; the routine ophthalmological evaluation did not disclose visual deficits in all, none developed epilepsy, one had mild sensorineural hearing loss which did not require aiding. Minor abnormalities of tone and reflexes were however observed in 5/28. At the clinical follow-up none of the children in the no-NE group showed motor deficits, none had hearing loss or brainstem dysfunction shown by auditory evoked potentials, none had visual deficits. 3.3. Outcome and neonatal data
Table 1 Demographic and perinatal clinical data in the neonatal encephalopathy (NE) and noneonatal encephalopathy (no NE) groups. No
Gender (male), n (%) Birth weight, grams (mean ± SD) Gestational age, weeks(mean ± SD) Apgar score 5′ ≤ 3, n (%) Type of delivery (vaginal), n (%) Dystocic vaginal delivery, n (%) Emergency caesarean, n (%) Resuscitation, n (%) pH b 7, n (%)* BE N 12, n (%)* MOF, n (%) Neonatal seizures, n (%) Time in hospital, days (mean ± SD)
Neonatal encephalopathy
Neonatal encephalopathy
p
(37 cases)
(22 cases)
19/37 (51%) 2915 ± 574 38.3 ± 2.33 6/36 (17%)
13/22 (59%) 3050 ± 604 38.18 ± 2.04 0/22 (0%)
ns ns ns 0.04
15/37 (41%) 6/15 (40%) 19/22 (86%) 34/37 (92%) 9/37 (24%) 19/37 (51%) 23/37 (62%) 12/37 (32%) 15 ± 11
10/22 (45%) 4/10 (40%) 9/12 (75%) 17/21 (81%) 0/18 (0%) 10/18 (56%) 3/21 (14%) 0/22 (0%) 10 ± 6
ns ns ns ns 0.022 ns 0.0004 0.002 ns
BE = base excess, * within the first hour of life, MOF = multi-organ failure (defined as at least one organ liver, kidney, heart involvement in adjunct to the brain).
BLC-SEP was observed in all but one cases affected by CP, the exception being a patient with Sarnat 1 encephalopathy and EEG score 0. Unilateral loss of cortical SEP was observed in two patients (one in the no-CP outcome group and one affected by hemiplegia). Cortical SEPs were bilaterally present in all children not developing cerebral palsy (no false negative results). All 21 children with neonatal encephalopathy Sarnat 1 were classified in the no-CP outcome group, while Sarnat 3 (6 patients) was always associated with cerebral palsy (Table 3, patients ID 2–7). The outcome of the 10 neonates with grade 2 encephalopathy however could not be predicted only on clinical basis (7 no-CP group versus 3 CP group—Table 3, patients ID 1, 8, 9). A multivariate analysis using the logistic regression model was performed to evaluate if the outcome as dependent variable could be improved combining dichotomised SEP (scores 1 and 2 vs score 0) and Sarnat score as independent variables. We found that this combination correctly classified the outcome in all patients (p b 0.0001). All patients with absent VEP were affected by CP and were partially sighted at follow-up. VEP asymmetry (score 1) was observed in children without CP. All NE patients with a score 0 or 1 of neonatal
96
A. Suppiej et al. / Early Human Development 86 (2010) 93–98
Fig. 1. Examples of SEP in one neonate without encephalopathy (A) and in three neonates affected by neonatal encephalopathy Sarnat 1 (B), 2 (C) and 3 (D).
EEG were in the no-CP group, while all those with a score of 2 or 3 were in the CP outcome group. The clinical outcome, EEG, SEP and VEP scores in patients with Sarnat 2 encephalopathy are summarized in Table 4, it shows that only patients with bilaterally absent SEPs and EEG score 2 or 3 were affected by CP. 4. Discussion This study found that cortical SEP was always present in neonates surviving without major neurological disability, while it was bilater-
ally absent in all but one patient demonstrating by two years of age definite neurological signs consistent with the diagnosis of cerebral palsy. These data extend to neonatal age previous findings of the role BLC-SEP in adults and children [1,2]. As in infants and older children [16,17], but differently from adults [1], the predictive value of a normal SEP appears in neonates to be superior to that of an absent SEP. The higher rate of false positive cases in some neonatal studies, as well as in the one patient in the present report as compared with adult studies suggests that in the neonatal period clinical and methodological factors may affect the reliability of the responses. Indeed it has been shown that the number of stimuli,
Fig. 2. Examples of VEP in one neonate without encephalopathy (A) and in three neonates affected by neonatal encephalopathy respectively Sarnat 1 (B), 2 (C) and 3 (D).
A. Suppiej et al. / Early Human Development 86 (2010) 93–98 Table 3 Details of the outcome of the 9 patients affected by cerebral palsy. ID Outcome 1. 2. 3. 4. 5. 6. 7. 8. 9.
Tetraparetic cerebral palsy, epilepsy. Hemiplegic cerebral palsy. Tetraparetic cerebral palsy, sensorineural hearing loss, partially sighted. Tetraparetic cerebral palsy, partially sighted. Tetraparetic cerebral palsy, partially sighted. Tetraparetic cerebral palsy, epilepsy, partially sighted. Tetraparetic cerebral palsy, sensorineural hearing loss, partially sighted. Tetraparetic cerebral palsy, epilepsy, sensorineural hearing loss, partially sighted. Diplegic cerebral palsy, partially sighted.
ID: identification number.
stimulation rate and filter settings have to be adjusted to take account of habituation and fatigue of the immature neonatal nervous system and the increased low-frequency content of the responses (for a review see [10]). The vigilance level may also affect the cortical components, but differences are not apparent during active sleep and the awake state [18]. Age related topographic changes need also to be taken into account, particularly before term. The electrode position at C3 (or C4) adopted in the present study is not the best location for the N20 recording in more mature infants, since it is near the isopotential area between the frontal P20 and the parietal N20, but this location was more suitable for near-term infants because of the movement of the N20 maximal amplitude from frontal to parietal sites related to the posterior displacement of the central sulcus consequent to the development of the frontal lobe, between 31 and 40 weeks of gestational age [19]. In future, more sophisticated applications of SEPs in neonates, should include both central and parietal (P3 or P4) locations. A limitation of the present study is the absence of a comparison group of healthy neonates from postnatal wards. However we think that this will not have affected the overall results since it is well known that the cortical SEP is always recorded in normal neonates under recording conditions comparable to those adopted in the present study [10]. Interestingly enough, the combination of SEP and Sarnat score correctly predicted the neurological outcome in all our patients. Neonatal encephalopathy was associated in our patients with low Apgar scores and major organ system injury, while other causes were ruled out, hence, our results apply to patients in whom encephalopathy was related to perinatal asphyxia. Indeed, assessing the proportion of neonatal encephalopathy that is due to asphyxia during birth is difficult because of problems in defining asphyxia and in attributing the cause of Table 4 Scores of severity of SEP, VEP, EEG and outcome in patients with Sarnat 2 neonatal encephalopaty. SEP: 0 = bilaterally recorded N20 cortical component; 2 = bilateral absence of N20 cortical component. VEP: 0 = presence of the P200 at Oz; 1 = left/right occipital asymmetry N 50%; 2 = absence of response. EEG: 0 = normal for gestational age or mildly abnormal (mild asymmetries, mild voltage reduction); 1 = discontinuous activity with inter-burst intervals (IBI) of less than 10 s, other types of continuous activity, clear asymmetry or asynchrony; 2 = IBI of 10–60 s, severe depression, no wake–sleep cycles; 3 = inactive EEG or background activity less than 10 μV and IBI greater than 60 s. Outcome: 1 = no cerebral palsy; 2 = cerebral palsy. ID
SEP
VEP
EEG
Outcome
10 11 12 13 1 14 15 8 16 9
0 0 0 0 2 0
0 0 1 1 0 0 0 2 0 2
1 0 1 1 2 0 0 3 0 2
1 1 1 1 2 1 1 2 1 2
a
a
2 0 2
Data excluded because of artefact.
97
neonatal neurological illness. Perinatal hypoxic ischemic brain injury has been reported in up to 90% of infants with neonatal encephalopathy [20–22], however only 21% of the neonates showing a neonatal neurological syndrome [23] satisfy the standard criteria for intrapartum asphyxia defined by the American College of Obstetricians and Gynaecologists and the American Academy of Pediatrics [24,25]. An early-onset encephalopathy in infants born ≥34 weeks' gestation with low Apgar scores is considered the best single criteria for occurrence of acute perinatal asphyxia potentially severe enough to cause brain damage [26]. Recent studies based on systematic use of magnetic resonance imaging (MRI) suggest that the brain areas preferentially involved following a prolonged period of mild to moderate hypotension lie between the territories of the circulation of the anterior, middle and posterior cerebral arteries, while damage which occurs as a result of a short period of profound hypoxic ischemia causes injury to the posterior putamen of the lentiform nuclei, the ventrolateral nuclei of the thalami, the hippocampus and the peri-rolandic cortex because of their high metabolic rate [27]. The functional status of the above anatomical locations of brain injury can be explored by means of evoked potentials (EPs). SEP abnormalities may point to dysfunction of sub-cortical pathways ascending to cortex including basal ganglia and internal capsule as well as to cortical areas located in the border zones between anterior and middle cerebral arteries. In agreement with previous finding [7,27], we found that the severity of the neurological syndrome graded with the Sarnat and Sarnat score [6] accurately predicted the outcome in severe and mild cases, but not in those of intermediate severity. In the present study VEPs were found not useful in neurological prognosis, when cortical response was present this finding did not guarantee a good outcome. The absence of cortical response was observed only in some patients, all developing cerebral palsy, in agreement with other authors showing that all neonates with abnormal VEPs throughout the first week of life, or with absent VEPs, either died or were left with neurological sequelae [28,29]. VEPs may have critical role in visual prognosis: all our patients with absent VEP developed visual deficit, in agreement with other authors' findings [30]. The use of neonatal electroencephalogram as a prognostic tool in neonatal medicine is widely recognised. In hypoxic–ischaemic encephalopathy a normal EEG has been associated with a favourable outcome while a severely abnormal background activity was invariably associated with a poor outcome. Prognosis remains uncertain in cases of moderately abnormal EEGs. In the attempt to improve the prognostic role of EEG different timing of the first recording, repeated recordings in the first hours or days of life and different scoring systems have been studied [31–38]. A higher predictive value was documented by Pressler et al. [13] when testing was done between 8 and 72 h of life. The score of EEG severity proposed by these authors and adopted in the present study and the temporal window of 10–72 h seem to accurately predict neurological examination at two years of age also in patients with moderate encephalopathy and may explain the excellent prognostic role of the EEG in prognostication of cerebral palsy. In agreement with other studies [39], the use of phenobarbital (maintained in the therapeutic range) in neonates with seizures did not affect the prognostic value of EEG recordings. From our knowledge the present study is the first that has analysed the predictive value of BLC-SEP and that of the neonatal neurological syndrome severity, VEP and the EEG in the same population. Our results support the role of BLC-SEP and the EEG in early identification of patients at risk of cerebral palsy following neonatal encephalopathy. The functional status of the central nervous system can be assessed in NICU admitted neonates at bedside, non invasively and at low cost using the EEG and SEPs. Although in the present study SEP did not pick up additional abnormalities in comparison with EEG results, in daily clinical practice these neurophysiological tools can support the clinical examination
98
A. Suppiej et al. / Early Human Development 86 (2010) 93–98
of patients affected by neonatal encephalopathy but, it is important to implement a cost effective strategy. Considering the different timings for prediction of outcome, SEP at 7 days of age and EEG 1–5 days of age, they appear complimentary investigations. The high reliability of BLC-SEP apply to patients tested after the first week of life, considerably later than the therapeutic window for neuroprotection, at present based mainly on hypothermia; thus for this purpose the neurophysiological technique of choice remains the EEG. By contrast SEPs may offer complementary data at a later stage. Indeed research suggest that the interpretation of severe EEG patterns such as electrocerebral silence and burst suppression may become complex in patients treated with intravenous anaesthetic agents with depressant effect on cortical metabolism or in the occurrence of metabolic disturbances; in fact in the absence of cortical damage the above EEG patterns are reversible in parallel with improvement of clinical status [9]. Furthermore we think that it may be easier to analyse SEP waveform for presence or absence of the N20 cortical component than to classify an EEG abnormality as being of intermediate severity. Considering that in the present study Sarnat score and dichotomised SEP correctly classified the outcome in all patients, a combination of clinical evaluation and SEP could be a valid alternative to undertaking repeated EEGs in selected cases. In conclusion the data of the present study support the role of BLC-SEP in early identification of neonates at risk of cerebral palsy and should be included in the clinical workup of term and near-term encephalopathy. Acknowledgements The authors thank the technicians Maria Tesone, Elena Bizzaro and Loretta Peppato.The project has been supported by a grant from the Italian Ministry of Health, Project No. 0AN/F. There are no known potential conflicts of interest. References [1] Robinson LR, Micklesen PJ, Tirschwell DL, Lew HL. Predictive value of somatosensory evoked potentials for awakening from coma. Crit Care Med 2003;31:960–7. [2] Wolrab G, Bolthauser E, Schmitt B. Neurological outcome in comatose children with bilateral loss of cortical somatosensory evoked potentials. Neuropediatrics 2001;32:271–4. [3] Görke W. Somatosensory evoked cortical potentials indicating impaired motor development in infancy. Dev Med Child Neurol 1986;28:633–41. [4] Finer NN, Robertson CM, Peters KL, Coward JH. Factors affecting outcome in hypoxic–ischemic encephalopathy in term infants. Am J Dis Child 1983;137:21–5. [5] Marlow N, Rose AS, Rands CE, Draper ES. Neuropsychological and educational problems at school age associated with neonatal encephalopathy. Arch Dis Child: Fetal and Neonatal Ed 2005;90:380–7. [6] Sarnat HB, Sarnat MS. Neonatal encephalopathy following fetal distress. A clinical and elecroencephalographic study. Arch Neurol 1976;33:696–705. [7] Van Handel M, Swaab H, de Vries LS, Jongmans MJ. Long term cognitive and behavioral consequences of neonatal encephalopathy following perinatal asphyxia: a review. Eur J Paediatr 2007;166:645–54. [8] Suppiej A. Evoked potentials in neonatal hypoxic–ischaemic encephalopathy: review of the literature. Ann Ist Super Sanita 2001;37:515–25. [9] Pressler R, Bady B, Binnie CD, Boylan GB, Connell JA, Lütschg J, et al. Neurophysiology of the neonatal period. In: Binnie C, Cooper R, Mauguière F, Osselton JW, Prior PF, Tedman BM, editors. Clinical neurophysiology, Volume 2. EEG, paediatric neurophysiology, special techniques and applications. Amsterdam: Elsevier; 2003. p. 450–506. [10] Suppiej A. General characteristics of evoked potentials. In: Pressler R, Binnie CD, Cooper R, Robinson R, editors. Neonatal and paediatric clinical neurophysiology. Edinburgh: Churchill Livingstone,Elsevier; 2007. p. 111–54. [11] Gibson NA, Graham M, Levene MI. Somatosensory evoked potentials and outcome in perinatal asphyxia. Arch Dis Child 1992;67:393–8.
[12] Taylor MJ, Murphy WJ, Whyte HE. Prognostic reliability of somatosensory and visual evoked potentials of asphyxiated term infants. Dev Med Child Neurol 1992;34:507–15. [13] Pressler R, Boylan G, Morton M, Binnie C, Rennie J. Early serial EEG in hypoxic ischaemicencephalopathy. Clin Neurophysiol 2001;112:31–7. [14] Amiel-Tison C, Stewart A. Follow up studies during the first five years of life: a pervasive assessment of neurological function. Arch Dis Child 1989;64:496–502. [15] Suppiej A, Rizzardi E, Zanardo V, Franzoi M, Ermani M, Orzan E. Reliability of hearing screening in high-risk neonates: comparative study of otoacoustic emission, automated and conventional auditory brainstem response. Clin Neurophysiol 2007;118:869–76. [16] Mandel R, Martinot A, Delepoulle F, Lamblin MD, Laureau E, Vallee L, et al. Prediction of outcome after hypoxic–ischemic encephalopathy: a prospective clinical and electrophysiologic study. J Pediatr 2002;141:45–50. [17] Abend NS, Licht DJ. Predicting outcome in children with hypoxic ischemic encephalopathy. Pediatric Crit Care Med 2007;8:1–7. [18] Hrbek A, Hrbkova M, Leonard H. Somatosensory, auditory and visual evoked responses in newborn infants during sleep and wakefulness. Elettroencephalogr Clin Neurophysiol 1969;26:597–603. [19] Karniski W, Wyble L, Lease L, Blair RC. The late somatosensory evoked potential in premature and term infants. II. Topography and latency development. Electroencephalogr Clin Neurophysiol 1992;84:44–54. [20] Badawi N, Kurinczuk JJ, Keogh JM, Alessandri LM, O'Sullivan F, Burton PR, et al. Intrapartum risk factors for newborn encephalopathy: the Western Australian case control study. BMJ 1998;317:1554–8. [21] Cowan F, Rutherford M, Groenendaal F, Eken P, Mercuri E, Bydder GM, et al. Origin and timing of brain lesion in term infants with neonatalencephalopathy. Lancet 2003;361:736–42. [22] Pierrat V, Haouari N, Liska A, Thomas D, Subtil D, Truffert P, et al. Prevalence, causes and outcomes at 2 years of age of newborn encephalopathy: a populationbased study. Arch Dis Child 2005;90:257–61 Fetal and Neonatal Ed. [23] Korst LM, Phelan JP, Wang YM, Martin GI, Ahn MO. Acute fetal asphyxia and permanent brain injury: a retrospective analysis of current indicators. Matern Fetal Med 1999;8:101–6. [24] MacLennan A. A template for defining a causal relation between acute intrapartum events and cerebral palsy: international consensus statement. BMJ 1999;319:1054–9. [25] Hankins GD, Speer M. Defining the pathogenesis and pathophysiology of neonatal encephalopathy and cerebral palsy. Obstet Gynecol 2003;102:628–36. [26] Hill A. Current concepts of hypoxic–ischemic cerebral injury in term newborn. Pediatr Neurol 1991;7:317–25. [27] Rennie JM, Hagmann CF, Robertson NJ. Outcome after intrapartum hypoxic ischaemia at term. Seminars in fetal and neo med 2007;12:398–407. [28] Whyte HE, Taylor MJ, Menzies R, Chin KC, MacMillan LJ. Prognostic utility of visual evoked potentials in term asphyxiated neonates. Pediatr Neurol 1986;2:220–3. [29] Muttitt SC, Taylor MJ, Kobayashi JS, MacMillan L, Whyte HE. Serial visual evoked potentials and outcome in term birth asphyxia. Pediatr Neurol 1991;7:86–90. [30] McCulloch DL, Taylor MJ, Whyte HE. Visual evoked potentials and visual prognosis following perinatal asphyxia. Arch Ophthalmol 1991;109:229–33. [31] Watanabe K, Miyazaki S, Hara K, Hakamada S. Behavioral state cycles, background EEGs and prognosis of newborns with perinatal hypoxia. Electroencephalogr Clin Neurophysiol 1980;49:618–25. [32] Pezzani C, Radvanyi-Bouvet MF, Relier JP, Monod N. Neonatal electroencephalography during the first twenty-four hours of life in full-term newborn infants. Neuropediatrics 1986;17:11–8. [33] Takeuchi T, Watanabe K. The EEG evolution and neurological prognosis of neonates with perinatal hypoxia. Brain Dev 1989;11:115–20. [34] D'Allest AM, Andre M, Radvanji-Bouvet MF. Apport de l'Electroencephalogramme pour le Diagnostic et le Pronostic de l'Asphyxie Perinatal du Nouveau-né a Term. Arch Pediatr 1996;3:254–6. [35] Biagioni E, Mercuri E, Rutherford M, Cowan F, Azzopardi D, Frisone MF, et al. Combined use of electroencephalogram and magnetic resonance imaging in fullterm neonates with acute encephalopathy. Pediatrics 2001;107:461–8. [36] Zeinstra E, Fock JM, Begeer JH, van Weerden TW, Maurits NM, Zweens MJ. The prognostic value of serial EEG recordings following acute neonatal asphyxia in full-term infants. Eur J Paediatr Neurol 2001;5:155–60. [37] Douglass LM, Wu JY, Rosman NP, Stafstrom CE. Burst suppression electroencephalogram pattern in the newborn: predicting the outcome. J Child Neurol 2002;17: 403–8. [38] Khan RL, Lahorgue Nunes M, Garcias da Silva LF, da Costa JC. Predictive value of sequential electroencephalogram (EEG) in neonates with seizures and its relation to neurological outcome. J Child Neurol 2008;23:144–50. [39] Staudt F, Scholl ML, Coen RW, Bickford RB. Phenobarbital therapy in neonatal seizures and the prognostic value of the EEG. Neuropediatrics 1982;13:24–33.