- Email: [email protected]
Prognostic utility of visual evoked potentials in term asphyxiated neonates
Prognostic Utility of Visual Evoked Potenfi s in Term Asphyxiated Neonates Hilary E. Whyte, MB*, MargotJ. Taylor, PhD ~, RosemaryMenzies, MB*, Kwei C. Chin, MB*, and LynnJ. MacMillan, RN~
The prognosis for term infants with birth asphyxia is variable and does not correlate well with the acute clinical state. Diagnostic tools such as electroencephalogmphy or imaging techniques have not been satisfactory predictors of outcome. Visual evoked potentials (VEPs) have changed acutely in birth asphyxia and may provide information for long-term prognosis. We studied VEPs serially in 25 term infants with documented birth asphyxia. The VEPs were dassified into three categories: normal, mildly abnormal, or severely abnormal, and then compared m each infant's acute clinical state and outcome at age six months. Eight of the nine infants with normal or mildly abnormal VEPs were normal when examined subsequently. All the patients with severely abnormal VEPs died, or suffered severe neurologic sequelae. The VEPs demonstrated good correlation with neurodevelopmental outcomes in infants with birth asphyxia and may be useful prognostically.
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. Introduction Birth asphyxia continues to be a major cause of mortality and morbidity among infants, being the single most common cause of cerebral palsy in later childhood [1]. Data on term asphyxiated infants reveal mortality rates of approximately 10-20% with morbidity rates ranging between 20-45 % [2-5]. Difficulty in determining the prognosis of each newborn results from an inability to determine adequately the severity of the insult by assessment of clinical status, imaging techniques, or by electroencephalographic (EEG) monitoring. Some studies have suggested that visual evoked potentials (VEPs) provide insight into acute alterations in central nervous system (CNS) function during and following asphyxia and may provide information for long-term prognosis [6-9]. We studied VEPs in term infants suffering from birth asphyxia to investigate the
From the Divisions of *Neonatology and ~qeurology; The Hospital for Sick Children; Toronto, Ontario. Received March 19, 1986; accepted May 14, 1986.
220
PEDIATRIC NEUROLOGY
Vol. 2 No. 4
possible correlations between VEPs in the acute phase, and subsequent neurodevelopmental outcome. Methods Twenty-five term infants were studied who had been born _ 37 weeks gestation and admitted to the neonatal intensive care unit of the Hospital for Sick Children in Toronto with a diagnosis of birth asphyxia. All infants were _<24-hours-old on admission. Birth asphyxia was classified by an Apgar score of -< 3 at 10 minutes, or by two of the following: (1) An Apgar score of -< 6 at 5 minutes; (2) Evidence of intrauterine asphyxia as manifested by a scalp pH -< 7, by late decelerations in fetal monitoring, or by meconium staining of the liquor; or, (3) Clinical evidence of asphyxia, such as altered tone, altered consciousness, or seizures. Neurologic examinations were performed daily until discharge. The clinical abnormality of each infant was determined on admission as either mild, moderate, or severe, depending on the degree of encephalopathy. Mild encephalopathy included infants with altered conscious states and altered tone. Moderate encepbalopattSy included infants with the characteristics of mild encephalopathy but who also had seizures. Severe encephalopathy included comatose infants with absent spontaneous respirations. Infants who developed seizures were treated initially with phenobarbital. Phenytoin and paraldehyde were used if seizures proved intractable. The majority of infants had cranial ultrasound studies and EEGs during the first week, with cranial computed tomography (CT) performed between the seventh and tenth day of life. VEPs were recorded from a single active electrode at Oz referenced to Fz. Gold cup electrodes were attached with paste and tape; impedance was below 5K ohms. A band pass of 1-100 Hz was employed, a gain of 10K, a sweep of 1 sec, and automatic artifact reject. Lightemitting diode (LED) goggles were held lightly over the infant's eyes; stimuli were presented at 0.5/sec. At least two averages of 64 trials each were collected with binocular stimulation. VEPs were performed daily for the first three days, again at the end of the first week, and weekly thereafter until death or discharge. Infants with absent VEPs also had electroretinograms (ERG) recorded. The VEPs were compared to our established normative values for neonates [10]. VEPs were classified into 6 categories: Grade 1-Normal; Grade 2Unusual wave form; Grade 3-Delayed latencies; Grade 4-Missing components; Grade 5-Delayed latencies plus missing components; Grade 6-Absent. During subsequent examinations, however, VEPs were classified into 3 categories: normal, mildly abnormal (i.e., a transiently abnormal or absent waveform of less than one week duration), or severely abnormal (i.e., prolonged absence or abnormality of waveform). Infants were seen subsequently at the ages of 6 weeks, 3 months, and 6 months. During each examination repeat
Communications should be addressed to: Dr. Whyte; The Hospital for Sick Children; 555 University Avenue; Toronto, Ontario; Canada M5G 1X8. Received March 19, 1986; accepted May 14, 1986.
BBR 39140 P2oo
+
6#V
3m
0
i
i
200
400
i
i
600
800
/
1000
ms
Figure 1. Normal VEPs at age 1 day and 3 months in an infant with mild asphyxia and normal outcome. VEPs were performed, as well as complete neurodevelopmental assessment. At age 3 months, a repeat EEG was performed, and at age 6 months, an ophthalmology examination was conducted.
Three infants had VEPs classified as mildly abnormal, recovering by the end of the first week. The CT scan and EEG were normal in one, mildly abnormal in the second, and severely abnormal in the third. Only one ultrasound was performed and was unremarkable. Two of these infants were normal at subsequent examination; previously, one of them had had a Grade 2 VEP abnormality, and the other had had a Grade 4 VEP, both of which resolved. The one infant who was globally delayed at age 6 months had absent VEPs (Grade 6) for 3 days with resolution at the end of the first week (Fig 3). The remaining 6 infants had absent VEPs (Grade 6); 12 patients improved to Grade 4 VEP by 6 weeks of age. All of these infants had abnormal EEGs ranging from burst-suppression to generalized depression, abnormal ultrasound examinations, and severe abnormalities on CTs. These infants were globally delayed at examination at age 6 months: 4 were microcephalic, 4 had seizures, 5 had spastic quadriplegia, 1 had diplegia.
SevereAsphyxia Nine infants were severely asphyxiated: 5 males and 4 females. EEGs were performed in 6 patients: 3 were BOG
41/40
Results Four infants were classified as mildly asphyxiated; 12 developed seizures with altered neurologic states and were classified as moderately asphyxiated, and 9 were severely asphyxiated.
ld
1
61~V
Mild Asphyxia The four infants with mild asphyxia all were male. EEGs performed in 2 patients were unremarkable; ultrasounds obtained in 3 patients a l s o were unremarkable. The VEPs were normal consistently (Grade 1) in 2 infants (Fig 1). Two other infants demonstrated abnormalities on the first day; one patient was categorized as Grade 2, and the other as Grade 5 (Fig 2), both of whom had reverted to normal by the second and fifth days of life, respectively. All four infants were normal at age 6 months; their EEGs and eye examinations were unremarkable.
P2oo
-~~~"8d 6,,v
Moderate Asphyxia There were 12 infants, 9 males and 3 females, in the moderate asphyxia category. Two of them had normal VEP, ultrasound, and EEG results; however, one patient demonstrated a mild hypoxic-ischemic encephalopathy on CT. Both infants were normal at subsequent examination at 6 months of age. A third with a Grade 1 VEP on day 1 had protracted seizures associated with a deterioration of his EEG, ultrasound, and mild to moderate changes in his CT scan. The VEP results deteriorated to Grade 4 during the first week and remained abnormal at ages 6 weeks and 3 months. The infant was microcephalic with spastic diplegia during the latter examination.
6#V 25m
200
400
600
800
1000
ms
Ftgure 2. Abnormal VEP (Grade 5) on day 1, with recovery o f normal VEP on day 5 and on day 8, and normal maturation o f the PrEP at 2.5 months o f age. The infant had mild asphyxia and a normal outcome. Whyte et al: Asphyxiated Neonates
221
BBB 41/40
~'V
2d VEPs
5d
N3oo P2oo
6d
,i
5 + 61~V
f
0
"
~
r
T
200
400
"
T
600
T
1
800
+
1
1000
ms
Figure 3. Infant w#b moderate asphyxia and global delay at age 6 months had absent VEPs with intact ERGs on day 2, but recovery of normal VEPs by day 6.
isoelectric, 1 demonstrated burst-suppression; 2 were depressed with seizure activity. Ultrasound examinations also were performed in 6 infants; 3 were unremarkable, and the other 3 demonstrated increased echogenicity. The CTs performed in 3 infants demonstrated moderate hypoxioischemic encephalopathy in 1, and severe hypoxic-ischemic encephalopathy in 2. All of these infants, however, had grossly abnormal VEPs; 5 were Grade 6, (Fig 4) and the remaining 4 deteriorated from Grade 4 or 5 to Grade 6 by day 3. Each of these infants died between day 1 and 3 months of age. In the 3 patients with isoelectric EEGs, treatment was withdrawn. The ERGs were normal in all the infants studied with the exception of 2 patients with severe asphyxia, on whom no ERGs could be recorded. Both of them had Grade 6 VEPs and died within 24 hours of birth. Fifteen infants received antiepileptic drugs. There was no obvious effect on the VEPs, apart from a transient latency shift following rapid bolus injection of phenobarbital in two patients.
clinical state as determined by Apgar scores has been found to be poorly correlated with outcome; even in the most extreme conditions, half of the survivors were free of major handicap [11]. Similarly, biochemical abnormalities occurring in infants with birth asphyxia have been poor predictors of outcome [12]. CT is the most useful investigative tool, having a 77% valid prediction rate when performed between the first and second week of life [13]; however, this rate is still not optimal. Although the occurrence of seizures within 24 hours of birth [14,15] and the appearance of certain changes in EEG patterns [16] have been used to indicate poor prognosis, other studies have demonstrated that these factors have a very poor correlation with outcome [17]. Since the neurologic outcome of infants with birth asphyxia has been compared previously to EEGs and CTs, and most recently to ultrasonography [18,19], these tests were examined in some infants. Because they were not performed in all of our patients, clear comparisons of their relative merit as outcome predictors cannot be made. From our results, however, there was a trend towards normal development in those who had normal or mildly abnormal CTs or EEGs; those with moderate or severe abnormalities on CTs or EEGs were severely delayed neurodevelopmentally or died. These findings were not corroborated by ultrasound examinations which were normal when performed prior to death in 3 infants. In an attempt to obtain a better measure of neurologic status, several authors have investigated VEPs and outcome in asphyxiated neonates [6,8]. Woods et al. [9,20,21] developed an animal model of
BBB 40f40
ld VEP
,.~o~..
ERG
2d
VEP 6,V
0
200
400
600
800
1010o
ms
Discussion In term infants with birth asphyxia, the initial 222
PEDIATRIC NEUROLOGY
Vol. 2 No. 4
Ftgure 4. Absent VEPs with normal ERGs in a severely asphyxiated infant who dted on day 3,
the effects of asphyxia on the VEPs. Using fetal lambs, they demonstrated that after only 2 minutes of asphyxia the VEPs were markedly abnormal, and after 4 minutes the VEPs were absent. After 8 minutes of asphyxia the lambs were resuscitated, but the VEPs remained abnormal for at least an hour. Post-asphyxia, those lambs with little evidence of sequelae had normal VEPs; those that did poorly had VEPs that remained atypical. VEPs thus provided information on brain recovery following asphyxia that was not available through other physiological measures. Hrbek et al. [6] examined visual and somatosensory evoked potentials and found a high abnormality rate in 57 asphyxiated infants. A scoring system was developed enabling quantitative evaluation of VEPs. "The evoked response risk score correlated well with the degree of asphyxia and a permanently high risk score on repeated testing was a serious prognostic sign." However, subsequent examinations were conducted only over a 3week period; no correlations could be drawn between VEPs and long-term prognosis. Gambi et al. [7] studied the VEPs of infants with respiratory distress and found them to be abnormal, demonstrating loss of the maturational characteristics of the waveform following anoxia. Those with the most severe respiratory distress still had abnormal VEPs when tested at age 9 months. Hakamada et al. [8] studied the evolution of VEPs in infants with a variety of perinatal disorders. All infants with abnormal VEPs for two weeks or more were abnormal at follow-up and the most severely affected infants had abnormal responses. There were, however, false negatives in 20.6% and false positives in 30% of the patients. To summarize our data, the four infants with normal VEPs were neurodevelopmentally normal on subsequent examinations. Of the five infants with mild VEP abnormalities, four were normal at subsequent examinations. Only one infant was neurodevelopmentally delayed at 3 months of age. Sixteen infants had severe abnormalities in their VEPs. Nine of these patients died; the other 7 had global delay: 1 at 3 months of age, and the other 6 patients at 6 months of age. Our results reveal a much higher correlation than has been reported previously. With one exception, all patients with normal or only transiently abnormal waveforms were normal during subsequent examinations. The one exception with a transient abnormality of VEP had an absent response for three days with rapid return to normal waveform. With further study it may become apparent that this absent response should have been classified with the severely abnormal VEPs, all of whom did poorly. Death or severe handicap could be predicted if abnormal waveforms persisted. VEPs in this study provided a good prediction of neurologic outcome. We found no false negatives and only one false positive result. Possible reasons for the improved predictive power of the VEPs include serial VEP studies starting on day 1 of life; investigating only well-defined asphyxi-
ated neonates; and, long-term follow-up. We recommend repeat VEP testing as a useful adjunct to the clinical assessment of term infants with birth asphyxia because it seems to provide better prognostic information than other diagnostic tests currently available. References [1] VolpeJJ. Neurology of the newborn. Philadelphia: Saunders, 1981. [2] BmwnJK, Purvis RJ, ForfarJO, et al. Neurological aspects of perinatal asphyxia. Dev Med Child Neurol 1974~16: 567. [3] Samat HB, Sarnat MS. Neonatal encephalopathy following fetal distress. Arch Neurol 1976;33:696-705. [4] DeSouza SW, Richards B. Neurological sequelae in newborn babies after perinatal asphyxia. Arch Dis Child 1978;53: 564-9. [5] Finer N, Richards R, Peters K. Hypoxic-ishcemic encephalopathy in term neonates: Perinatal factors and outcome. J Pediatr 1981 ;98:112-7. [6] Hrbek A, Karlberg P, Kjellmer I, et al. Clinical application of evoked electroencephalographic responses in newborn infants. I: Perinatal asphyxia. Dev Med Child Neurol 1977; 19: 34-44. [7] Gambi D, Rossini PM, Albertini G, Sollazzo D, Torrioli MG, Polidori GC. Follow-up of visual evoked potential in full-term and pre-term control newborns and in subjects who suffered from perinatal respiratory distress. Electroencephalogr Clin Neurophysiol 1980;48:509-16. [8] Hakamada S, Watanabe K, Hara K, Miyazaki S. The evolution of visual and auditory evoked potentials in infants with perinatal disorder. Brain Dev 1981; 3: 339-44. [9] Woods JR Jr, Coppes V, Brooks DE, et al. Measurement of visual evoked potential in the asphyctic fetus and during neonatal survival. AmJ Obstet Gynecol 1982; 143:944-51. [10l Taylor MJ, Menzies R, Whyte H. VEPs in normal full-term and premature neonates: Longitudinal vs cross sectional data. Electroencephalogr Clin Neurophysiol, In press. [11] Nelson K, EllenbergJ. Apgar scores as predictors of chronic neurologic disability. Pediatrics 1981 ;68: 36. [12] MyersRE. Experimental models of perinatal brain damage: Relevance to human pathology. In: Gluck L, ed. Intrauterine asphyxia and the developing fetal brain. Chicago: Year Book, 1977. [13] Fitzhardinge PM, Flodmark O, Fitz CR, Ashby S. The prognostic value of computed tomography as an adjunct to assessment of the term infant with post asphyxia] encephalopathy. J Pediatr 1981 ;99:777-81. [14] Rose AL, Lombroso CT. Neonatal seizure states: A study of clinical, pathological, and electroencephalographic features in 137 full-term babies with a long-term follow-up. Pediatrics 1970; 45:404-25. [15] Amiel-Tison C. Neurologic disorders in neonates associated with abnormalities of pregnancy and birth. Curr Probl Pediatr 1973;3:1. [16] Sokol RJ, Rosen MG, Chik L. Fetal electroencephalography. In: Beard RW, Nathaniels PW, eds. Fetal physiology and medicine. London: WB Saunders, 1976;476 -91. [171 Tortes F, Blaw ME. Longitudinal EEG--Clinical correlations in children from birth to 4 years of age. Pediatrics 1968;41:945. [18] Babcock DS, Ball W. Postasphyxial encephalopathy in fullterm infants: Ultrasound diagnosis. Radiology 1983; 148:417 -23. [19] Siegel M.J, Shackelford GD, Perlman JM, Fulling KH. Hypoxic-ischemic encephalopathy in term infants: Diagnosis and prognosis evaluated by ultrasound. Radiology 1984; 152: 395-9. [20] Woods JR Jr, Coppes V, Brooks DE, et al. Birth asphyxia I. Measurement of visual evoked potential (VEP) in the healthy foetus-newborn lamb. Pediatr Res 1981; 15:1429-32. [21] Woods JR Jr, Parisi V, Coppes V, Brooks DE. Maturational sequence of the visual system: Serial measurements of visual evoked potential and electroretinograrn in the healthy neonatal lamb. Am J Obstet Gynecol 1983;145:738-43. Whyte et al: Asphyxiated Neonates
223
The prognosis for term infants with birth asphyxia is variable and does not correlate well with the acute clinical state. Diagnostic tools such as electroencephalogmphy or imaging techniques have not been satisfactory predictors of outcome. Visual evoked potentials (VEPs) have changed acutely in birth asphyxia and may provide information for long-term prognosis. We studied VEPs serially in 25 term infants with documented birth asphyxia. The VEPs were dassified into three categories: normal, mildly abnormal, or severely abnormal, and then compared m each infant's acute clinical state and outcome at age six months. Eight of the nine infants with normal or mildly abnormal VEPs were normal when examined subsequently. All the patients with severely abnormal VEPs died, or suffered severe neurologic sequelae. The VEPs demonstrated good correlation with neurodevelopmental outcomes in infants with birth asphyxia and may be useful prognostically.
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. Introduction Birth asphyxia continues to be a major cause of mortality and morbidity among infants, being the single most common cause of cerebral palsy in later childhood [1]. Data on term asphyxiated infants reveal mortality rates of approximately 10-20% with morbidity rates ranging between 20-45 % [2-5]. Difficulty in determining the prognosis of each newborn results from an inability to determine adequately the severity of the insult by assessment of clinical status, imaging techniques, or by electroencephalographic (EEG) monitoring. Some studies have suggested that visual evoked potentials (VEPs) provide insight into acute alterations in central nervous system (CNS) function during and following asphyxia and may provide information for long-term prognosis [6-9]. We studied VEPs in term infants suffering from birth asphyxia to investigate the
From the Divisions of *Neonatology and ~qeurology; The Hospital for Sick Children; Toronto, Ontario. Received March 19, 1986; accepted May 14, 1986.
220
PEDIATRIC NEUROLOGY
Vol. 2 No. 4
possible correlations between VEPs in the acute phase, and subsequent neurodevelopmental outcome. Methods Twenty-five term infants were studied who had been born _ 37 weeks gestation and admitted to the neonatal intensive care unit of the Hospital for Sick Children in Toronto with a diagnosis of birth asphyxia. All infants were _<24-hours-old on admission. Birth asphyxia was classified by an Apgar score of -< 3 at 10 minutes, or by two of the following: (1) An Apgar score of -< 6 at 5 minutes; (2) Evidence of intrauterine asphyxia as manifested by a scalp pH -< 7, by late decelerations in fetal monitoring, or by meconium staining of the liquor; or, (3) Clinical evidence of asphyxia, such as altered tone, altered consciousness, or seizures. Neurologic examinations were performed daily until discharge. The clinical abnormality of each infant was determined on admission as either mild, moderate, or severe, depending on the degree of encephalopathy. Mild encephalopathy included infants with altered conscious states and altered tone. Moderate encepbalopattSy included infants with the characteristics of mild encephalopathy but who also had seizures. Severe encephalopathy included comatose infants with absent spontaneous respirations. Infants who developed seizures were treated initially with phenobarbital. Phenytoin and paraldehyde were used if seizures proved intractable. The majority of infants had cranial ultrasound studies and EEGs during the first week, with cranial computed tomography (CT) performed between the seventh and tenth day of life. VEPs were recorded from a single active electrode at Oz referenced to Fz. Gold cup electrodes were attached with paste and tape; impedance was below 5K ohms. A band pass of 1-100 Hz was employed, a gain of 10K, a sweep of 1 sec, and automatic artifact reject. Lightemitting diode (LED) goggles were held lightly over the infant's eyes; stimuli were presented at 0.5/sec. At least two averages of 64 trials each were collected with binocular stimulation. VEPs were performed daily for the first three days, again at the end of the first week, and weekly thereafter until death or discharge. Infants with absent VEPs also had electroretinograms (ERG) recorded. The VEPs were compared to our established normative values for neonates [10]. VEPs were classified into 6 categories: Grade 1-Normal; Grade 2Unusual wave form; Grade 3-Delayed latencies; Grade 4-Missing components; Grade 5-Delayed latencies plus missing components; Grade 6-Absent. During subsequent examinations, however, VEPs were classified into 3 categories: normal, mildly abnormal (i.e., a transiently abnormal or absent waveform of less than one week duration), or severely abnormal (i.e., prolonged absence or abnormality of waveform). Infants were seen subsequently at the ages of 6 weeks, 3 months, and 6 months. During each examination repeat
Communications should be addressed to: Dr. Whyte; The Hospital for Sick Children; 555 University Avenue; Toronto, Ontario; Canada M5G 1X8. Received March 19, 1986; accepted May 14, 1986.
BBR 39140 P2oo
+
6#V
3m
0
i
i
200
400
i
i
600
800
/
1000
ms
Figure 1. Normal VEPs at age 1 day and 3 months in an infant with mild asphyxia and normal outcome. VEPs were performed, as well as complete neurodevelopmental assessment. At age 3 months, a repeat EEG was performed, and at age 6 months, an ophthalmology examination was conducted.
Three infants had VEPs classified as mildly abnormal, recovering by the end of the first week. The CT scan and EEG were normal in one, mildly abnormal in the second, and severely abnormal in the third. Only one ultrasound was performed and was unremarkable. Two of these infants were normal at subsequent examination; previously, one of them had had a Grade 2 VEP abnormality, and the other had had a Grade 4 VEP, both of which resolved. The one infant who was globally delayed at age 6 months had absent VEPs (Grade 6) for 3 days with resolution at the end of the first week (Fig 3). The remaining 6 infants had absent VEPs (Grade 6); 12 patients improved to Grade 4 VEP by 6 weeks of age. All of these infants had abnormal EEGs ranging from burst-suppression to generalized depression, abnormal ultrasound examinations, and severe abnormalities on CTs. These infants were globally delayed at examination at age 6 months: 4 were microcephalic, 4 had seizures, 5 had spastic quadriplegia, 1 had diplegia.
SevereAsphyxia Nine infants were severely asphyxiated: 5 males and 4 females. EEGs were performed in 6 patients: 3 were BOG
41/40
Results Four infants were classified as mildly asphyxiated; 12 developed seizures with altered neurologic states and were classified as moderately asphyxiated, and 9 were severely asphyxiated.
ld
1
61~V
Mild Asphyxia The four infants with mild asphyxia all were male. EEGs performed in 2 patients were unremarkable; ultrasounds obtained in 3 patients a l s o were unremarkable. The VEPs were normal consistently (Grade 1) in 2 infants (Fig 1). Two other infants demonstrated abnormalities on the first day; one patient was categorized as Grade 2, and the other as Grade 5 (Fig 2), both of whom had reverted to normal by the second and fifth days of life, respectively. All four infants were normal at age 6 months; their EEGs and eye examinations were unremarkable.
P2oo
-~~~"8d 6,,v
Moderate Asphyxia There were 12 infants, 9 males and 3 females, in the moderate asphyxia category. Two of them had normal VEP, ultrasound, and EEG results; however, one patient demonstrated a mild hypoxic-ischemic encephalopathy on CT. Both infants were normal at subsequent examination at 6 months of age. A third with a Grade 1 VEP on day 1 had protracted seizures associated with a deterioration of his EEG, ultrasound, and mild to moderate changes in his CT scan. The VEP results deteriorated to Grade 4 during the first week and remained abnormal at ages 6 weeks and 3 months. The infant was microcephalic with spastic diplegia during the latter examination.
6#V 25m
200
400
600
800
1000
ms
Ftgure 2. Abnormal VEP (Grade 5) on day 1, with recovery o f normal VEP on day 5 and on day 8, and normal maturation o f the PrEP at 2.5 months o f age. The infant had mild asphyxia and a normal outcome. Whyte et al: Asphyxiated Neonates
221
BBB 41/40
~'V
2d VEPs
5d
N3oo P2oo
6d
,i
5 + 61~V
f
0
"
~
r
T
200
400
"
T
600
T
1
800
+
1
1000
ms
Figure 3. Infant w#b moderate asphyxia and global delay at age 6 months had absent VEPs with intact ERGs on day 2, but recovery of normal VEPs by day 6.
isoelectric, 1 demonstrated burst-suppression; 2 were depressed with seizure activity. Ultrasound examinations also were performed in 6 infants; 3 were unremarkable, and the other 3 demonstrated increased echogenicity. The CTs performed in 3 infants demonstrated moderate hypoxioischemic encephalopathy in 1, and severe hypoxic-ischemic encephalopathy in 2. All of these infants, however, had grossly abnormal VEPs; 5 were Grade 6, (Fig 4) and the remaining 4 deteriorated from Grade 4 or 5 to Grade 6 by day 3. Each of these infants died between day 1 and 3 months of age. In the 3 patients with isoelectric EEGs, treatment was withdrawn. The ERGs were normal in all the infants studied with the exception of 2 patients with severe asphyxia, on whom no ERGs could be recorded. Both of them had Grade 6 VEPs and died within 24 hours of birth. Fifteen infants received antiepileptic drugs. There was no obvious effect on the VEPs, apart from a transient latency shift following rapid bolus injection of phenobarbital in two patients.
clinical state as determined by Apgar scores has been found to be poorly correlated with outcome; even in the most extreme conditions, half of the survivors were free of major handicap [11]. Similarly, biochemical abnormalities occurring in infants with birth asphyxia have been poor predictors of outcome [12]. CT is the most useful investigative tool, having a 77% valid prediction rate when performed between the first and second week of life [13]; however, this rate is still not optimal. Although the occurrence of seizures within 24 hours of birth [14,15] and the appearance of certain changes in EEG patterns [16] have been used to indicate poor prognosis, other studies have demonstrated that these factors have a very poor correlation with outcome [17]. Since the neurologic outcome of infants with birth asphyxia has been compared previously to EEGs and CTs, and most recently to ultrasonography [18,19], these tests were examined in some infants. Because they were not performed in all of our patients, clear comparisons of their relative merit as outcome predictors cannot be made. From our results, however, there was a trend towards normal development in those who had normal or mildly abnormal CTs or EEGs; those with moderate or severe abnormalities on CTs or EEGs were severely delayed neurodevelopmentally or died. These findings were not corroborated by ultrasound examinations which were normal when performed prior to death in 3 infants. In an attempt to obtain a better measure of neurologic status, several authors have investigated VEPs and outcome in asphyxiated neonates [6,8]. Woods et al. [9,20,21] developed an animal model of
BBB 40f40
ld VEP
,.~o~..
ERG
2d
VEP 6,V
0
200
400
600
800
1010o
ms
Discussion In term infants with birth asphyxia, the initial 222
PEDIATRIC NEUROLOGY
Vol. 2 No. 4
Ftgure 4. Absent VEPs with normal ERGs in a severely asphyxiated infant who dted on day 3,
the effects of asphyxia on the VEPs. Using fetal lambs, they demonstrated that after only 2 minutes of asphyxia the VEPs were markedly abnormal, and after 4 minutes the VEPs were absent. After 8 minutes of asphyxia the lambs were resuscitated, but the VEPs remained abnormal for at least an hour. Post-asphyxia, those lambs with little evidence of sequelae had normal VEPs; those that did poorly had VEPs that remained atypical. VEPs thus provided information on brain recovery following asphyxia that was not available through other physiological measures. Hrbek et al. [6] examined visual and somatosensory evoked potentials and found a high abnormality rate in 57 asphyxiated infants. A scoring system was developed enabling quantitative evaluation of VEPs. "The evoked response risk score correlated well with the degree of asphyxia and a permanently high risk score on repeated testing was a serious prognostic sign." However, subsequent examinations were conducted only over a 3week period; no correlations could be drawn between VEPs and long-term prognosis. Gambi et al. [7] studied the VEPs of infants with respiratory distress and found them to be abnormal, demonstrating loss of the maturational characteristics of the waveform following anoxia. Those with the most severe respiratory distress still had abnormal VEPs when tested at age 9 months. Hakamada et al. [8] studied the evolution of VEPs in infants with a variety of perinatal disorders. All infants with abnormal VEPs for two weeks or more were abnormal at follow-up and the most severely affected infants had abnormal responses. There were, however, false negatives in 20.6% and false positives in 30% of the patients. To summarize our data, the four infants with normal VEPs were neurodevelopmentally normal on subsequent examinations. Of the five infants with mild VEP abnormalities, four were normal at subsequent examinations. Only one infant was neurodevelopmentally delayed at 3 months of age. Sixteen infants had severe abnormalities in their VEPs. Nine of these patients died; the other 7 had global delay: 1 at 3 months of age, and the other 6 patients at 6 months of age. Our results reveal a much higher correlation than has been reported previously. With one exception, all patients with normal or only transiently abnormal waveforms were normal during subsequent examinations. The one exception with a transient abnormality of VEP had an absent response for three days with rapid return to normal waveform. With further study it may become apparent that this absent response should have been classified with the severely abnormal VEPs, all of whom did poorly. Death or severe handicap could be predicted if abnormal waveforms persisted. VEPs in this study provided a good prediction of neurologic outcome. We found no false negatives and only one false positive result. Possible reasons for the improved predictive power of the VEPs include serial VEP studies starting on day 1 of life; investigating only well-defined asphyxi-
ated neonates; and, long-term follow-up. We recommend repeat VEP testing as a useful adjunct to the clinical assessment of term infants with birth asphyxia because it seems to provide better prognostic information than other diagnostic tests currently available. References [1] VolpeJJ. Neurology of the newborn. Philadelphia: Saunders, 1981. [2] BmwnJK, Purvis RJ, ForfarJO, et al. Neurological aspects of perinatal asphyxia. Dev Med Child Neurol 1974~16: 567. [3] Samat HB, Sarnat MS. Neonatal encephalopathy following fetal distress. Arch Neurol 1976;33:696-705. [4] DeSouza SW, Richards B. Neurological sequelae in newborn babies after perinatal asphyxia. Arch Dis Child 1978;53: 564-9. [5] Finer N, Richards R, Peters K. Hypoxic-ishcemic encephalopathy in term neonates: Perinatal factors and outcome. J Pediatr 1981 ;98:112-7. [6] Hrbek A, Karlberg P, Kjellmer I, et al. Clinical application of evoked electroencephalographic responses in newborn infants. I: Perinatal asphyxia. Dev Med Child Neurol 1977; 19: 34-44. [7] Gambi D, Rossini PM, Albertini G, Sollazzo D, Torrioli MG, Polidori GC. Follow-up of visual evoked potential in full-term and pre-term control newborns and in subjects who suffered from perinatal respiratory distress. Electroencephalogr Clin Neurophysiol 1980;48:509-16. [8] Hakamada S, Watanabe K, Hara K, Miyazaki S. The evolution of visual and auditory evoked potentials in infants with perinatal disorder. Brain Dev 1981; 3: 339-44. [9] Woods JR Jr, Coppes V, Brooks DE, et al. Measurement of visual evoked potential in the asphyctic fetus and during neonatal survival. AmJ Obstet Gynecol 1982; 143:944-51. [10l Taylor MJ, Menzies R, Whyte H. VEPs in normal full-term and premature neonates: Longitudinal vs cross sectional data. Electroencephalogr Clin Neurophysiol, In press. [11] Nelson K, EllenbergJ. Apgar scores as predictors of chronic neurologic disability. Pediatrics 1981 ;68: 36. [12] MyersRE. Experimental models of perinatal brain damage: Relevance to human pathology. In: Gluck L, ed. Intrauterine asphyxia and the developing fetal brain. Chicago: Year Book, 1977. [13] Fitzhardinge PM, Flodmark O, Fitz CR, Ashby S. The prognostic value of computed tomography as an adjunct to assessment of the term infant with post asphyxia] encephalopathy. J Pediatr 1981 ;99:777-81. [14] Rose AL, Lombroso CT. Neonatal seizure states: A study of clinical, pathological, and electroencephalographic features in 137 full-term babies with a long-term follow-up. Pediatrics 1970; 45:404-25. [15] Amiel-Tison C. Neurologic disorders in neonates associated with abnormalities of pregnancy and birth. Curr Probl Pediatr 1973;3:1. [16] Sokol RJ, Rosen MG, Chik L. Fetal electroencephalography. In: Beard RW, Nathaniels PW, eds. Fetal physiology and medicine. London: WB Saunders, 1976;476 -91. [171 Tortes F, Blaw ME. Longitudinal EEG--Clinical correlations in children from birth to 4 years of age. Pediatrics 1968;41:945. [18] Babcock DS, Ball W. Postasphyxial encephalopathy in fullterm infants: Ultrasound diagnosis. Radiology 1983; 148:417 -23. [19] Siegel M.J, Shackelford GD, Perlman JM, Fulling KH. Hypoxic-ischemic encephalopathy in term infants: Diagnosis and prognosis evaluated by ultrasound. Radiology 1984; 152: 395-9. [20] Woods JR Jr, Coppes V, Brooks DE, et al. Birth asphyxia I. Measurement of visual evoked potential (VEP) in the healthy foetus-newborn lamb. Pediatr Res 1981; 15:1429-32. [21] Woods JR Jr, Parisi V, Coppes V, Brooks DE. Maturational sequence of the visual system: Serial measurements of visual evoked potential and electroretinograrn in the healthy neonatal lamb. Am J Obstet Gynecol 1983;145:738-43. Whyte et al: Asphyxiated Neonates
223