Clinico-etiological and EEG profile of neonatal seizures
http://www.100md.com
《美国医学杂志》
Department of Pediatrics, Maulana Azad Medical College and Associated Lok Nayak Hospital, New Delhi, India
Objective. To evaluate clinical, etiological and EEG profile of neonatal seizures. Methods. In all the neonates enrolled in the study baseline information recorded was sex, gestational age, weight, ponderal index etc. Clinical profile of seizure episode included age at onset of seizure, type and duration of seizure, associated autonomic changes, medication given, response time to medication and possible cause. Relevant maternal history including antenatal and peripartum history was recorded. Relevant history and examination of newborn were noted. Essential investigations done in all subjects included blood glucose, serum calcium, serum sodium and arterial pH. USG cranium and EEG were done at earliest in all the subjects wherever feasible. Additional investigations were done as guided by history, physical examination and essential investigations. Results. Ninety babies with clinical seizures were enrolled into the study over one year period with an overall incidence of 1.17% (0.69% in term babies and 6.14% in preterm babies). Abnormal EEG's were found in one third cases out of 60 EEG's done in 90 babies. 26.7% of babies with perinatal asphyxia had abnormal EEG's (8/30). While 60% of babies with HIE II had abnormal discharges, background activity was suppressed in 66.66% EEG's in babies with HIE III. Conclusions. Overall incidence of neonatal seizures was 11.7/1000 live births, majority being preterm very low birth weight babies before 5 days of life. Perinatal asphyxia was responsible in 44.44% babies followed by metabolic abnormalities (23.33%). EEG abnormalities were present in 33.34% babies.
Keywords: Neonatal seizures; Electroencephalography
Clinical seizures are defined as paroxysmal alteration in neurological function i.e. behavioral, motor and/or autonomic function. Newborn period is the time when the incidence of seizures is the highest, yet their clinical recognition is difficult, therefore true incidence of neonatal seizures is difficult to determine. Seizures are often the first sign of neurological dysfunction in newborn but their clinical expression at this age is quite variable, poorly organized and often subtle. Volpe[1] has classified seizures into five clinical types, viz. subtle, multifocal clonic, focal clonic, generalized tonic and myoclonic. Neonatal seizures are difficult to investigate and consequently determination of etiology and initiation of therapy may be delayed which results in poor neurological outcome. Neonatal seizures can be due to various causes like hypoxic-ischemic encephalopathy, intracranial hemorrhage, meningitis, hypoglycemia, hypocalcemia, congenital malformation, etc. Though EEG provides a useful non-invasive test to diagnose neonatal seizures and evaluate degree of perinatal damage to brain and long term prognosis, yet its interpretation is influenced by variations in normal maturation process of brain. It is well recognized that not all seizures can be picked by surface recorded EEG and many clinically silent electrographic seizures have been reported. Therefore present study was conducted to evaluate the correlation of EEG with clinical and etiological profile of neonatal seizures.
Materials and Methods
The study was conducted in the tertiary level Neonatal Unit of a teaching hospital over one year period. All the babies born at Obstetric Unit of the hospital and admitted with clinically identified seizures before 28 days of age were enrolled into the study. Ninety babies developed clinical seizures over study period. In all neonates enrolled into the study, the information outlined below was collected and recorded.
(A) Baseline Characteristics: These included sex, gestational age, intrauterine growth status, maternal education and socio-economic status (Kuppuswami scale).[2] Weight, head circumference and length were recorded at birth by standard technique and Ponderal index was calculated. Babies were categorized into appropriate for gestational age (AGA) and small for gestational age (SGA)[symmetrical and asymmetrical] categories according to percentile curves developed by Manmohan et al.[3]
(B) Clinical Profile of Seizures : Each seizure episode reported by mother and subsequently observed by resident doctor on duty was recorded and relevant information was gathered: age at onset of seizure in day(s), type of seizure, duration of seizure in minutes, associated autonomic changes (HR, RR, O 2 saturation), medication required to control seizures, response time to medication, and possible cause.
(C) Determination of Etiology : Relevant clinical information was collected and recorded. Antenatal history of infections including TORCH, drug intake, pregnancy induced hypertension (PIH), gestational diabetes, etc. was taken. Peripartum history of PIH, antepartum hemorrhage, chorioamnionitis, fetal distress (fetal heart rate abnormalities or meconium stained liquor) was recorded. Mode of presentation and delivery was noted. Resuscitation details if done were also recorded. All babies were examined daily, including standard neurological examination till they were seizure free for 3 days or till discharge whichever was later.
(D) Investigations: Essential Investigations done in all the subjects included blood glucose, total serum calcium, ionized serum calcium, serum sodium, and arterial pH. USG of cranium and EEG were done at earliest preferably within 48 hours of onset of seizure or when the neonate became stable, whichever was earlier. Electro-encephalography was obtained on SLE microscibe EEG machine having 21 channels manufactured by SLE, UK. The techniques used were as per guidelines of the American EEG society for recording of EEG in the neonates.[4] The montages used were monopolar, bipolar and referential. Triclophos was used in dose of 50 mg/Kg body weight for inducing sleep where needed. Babies were also subjected to photic stimulation to provoke any abnormalities in brain activity. Each EEG recording was made for 30 minutes with different montages. Additional investigations were done as guided by history, physical examination and essential investigations. They included sepsis screen, blood culture, cerebrospinal fluid (CSF) examination, hematocrit, serum bilirubin levels, serum urea and creatinine, "TORCH" serology, VDRL, serum magnesium, serum parathyroid hormone, metabolic profile for inborn errors of metabolism. CT scan and MRI were done whenever indicated. All babies were treated using standard treatment protocol.
Statistical Analysis: The data were entered into computer using Epi info 6 software. Normally distributed data were analysed using mean (SD) while non guassian distributed data were analysed using median(range).
Results
During the study period, a total of 7728 babies were born live. Clinical seizures were identified in 90 babies during hospital stay, with an incidence of 11.65/1000 live births. The incidence in preterm babies was 6.14% in contrast to only 0.69% in term babies. Early onset seizures (< 48 hours) were observed in 2.84% preterm neonates and 0.47% term neonates. Seizures developed in 10.14% babies in very low birth weight, 1.51% babies in low birth weight (1500-2499 grams) and 0.59% babies in normal birth weight (2500-3999 grams) groups. Out of 90 babies developing clinical seizures, 47 were small for gestational age (twelve being symmetrical).
A total of 232 seizure episodes occurred in ninety babies with an average of 2.58 seizure episodes per baby (range 1 to 7 episodes) in first 28 days of life. The median age of onset of first seizure episode was 41.5 hours (in term babies 38 hours, in preterm babies 55 hours), with 75% of seizure episodes occurring before 115 hours of age. Fifty two babies (57.8%) developed first seizure within first 48 hours of life, out of which 20 babies had seizure in less than 12 hours of life. Autonomic changes (heart rate, respiratory rate or oxygen saturation of blood) were present in 108 seizure episodes (46.55%). Out of 232 seizure episodes, multifocal clonic seizures contributed a major proportion with 42.24%, followed by generalized tonic seizures (21.55%). Among mixed seizure episodes, multifocal clonic with subtle seizures were present in 21 seizure episodes followed by generalized tonic with subtle seizures in 18 episodes.
Perinatal asphyxia was the most common cause of neonatal seizures in both preterm (39.02%) and term (48.98%) babies. Thirty three babies had abnormal neurological state between seizures (hypoxic ischemic encephalopathy) and thus were diagnosed to have seizures because of birth asphyxia [Table - 1]. The median (range) age of onset of seizures in HIE stages I, II and III were 11 hours (1-47 hours), 12 hours (1-58 hours) and 23 hours (4-437 hours) respectively. Out of 108 seizure episodes in babies with perinatal asphyxia, 11 and 18 were associated with hypoglycemia and hypocalcemia while rest of 79 episodes were due to HIE per se [Table - 2]. Multifocal clonic and generalized tonic seizures were the most common type of seizures associated with perinatal asphyxia per se . 0 EEG was done in 30 babies out of 40 babies with perinatal asphyxia [Table - 1][Table - 2]. Five babies had abnormal EEG record out of eleven EEG's done in 13 babies in HIE stage II. Background abnormality was present in two babies, one having low voltage record and other very low voltage record. In another three babies multifocal discharges were present, one having spikes and two having sharp waves [Table - 3]. Brain edema was detected in 3 babies with HIE stage II on USG cranium.
Out of ninety babies, 21 babies (23.34%) had seizures associated with hypoglycemia . Ten babies were low birth weight while other eleven babies had associated morbidity like birth asphyxia, meningitis etc. Multifocal clonic seizures were the most common type of seizure episodes associated with hypoglycemia. In 89 babies in whom total serum calcium level was estimated, 22 babies (24.72%) had seizures associated with low total serum calcium level (preterm baby n=8, serum calcium < 7.5 mg/dL; term baby n=14, serum calcium < 8mg/dL). Nine babies were low birth weight while other thirteen babies had associated morbidities similar to hypoglycemic babies. Hypocalcemia alone or associated with other morbidities had multifocal clonic seizure type in majority of seizure episodes (31/56). Abnormal discharges were noted in two and three babies with hypoglycemia and hypocalcemia respectively out of nine and eight EEG's done. One baby with hypoglycemia had background abnormality also [Table - 3]. One baby had hypernatremia (serum sodium 185 meq/L) associated with high content of breast milk sodium (59.6 meq/L). Eighteen seizure episodes were noticed in 7 babies with meningitis , out of which seven were multifocal clonic seizures.
In 13 babies cause of seizure could not be ascertained [Table - 1]. Nine babies died, out of which 7 babies within 3 days of onset of first seizure episode and hence could not undergo essential (USG cranium and EEG) and relevant additional investigations. In rest of the 6 babies, EEG abnormality was present in 3 cases [Table - 3]. Multifocal discharges were present in all the 3 babies along with background abnormalities in 2 cases (one with low voltage record and other asymmetrical background activity). USG cranium failed to detect any abnormality responsible for seizure episode.
Discussion
In most of the clinical settings including neonatal intensive care units, seizures in the newborn are identified by clinical observation. However, some of the clinically identified motor and behavioral phenomena characterized as seizures do not have simultaneous EEG correlate, thus overestimating neonatal seizures. Conversely many electrographic seizures are not accompanied by clinically observed alternations in neonatal motor or behavioral function, thus underestimating neonatal seizures. The incidence of neonatal seizures as reported by various studies range from 0.1-0.5% in term neonates and 10-22.7% in preterm neonates.[5],[6],[7] The present study has found an overall incidence of clinical seizures in 1.17% babies with 0.69% in term babies and 6.14% in preterm babies. Early onset seizures were observed in 0.47% term neonates in the present study though the reported incidence in literature varies from 0.87/1000 to 2.8/1000 live births.[8] Low birth weight is an important risk factor, as clinical seizures in them often indicate severe brain injury and are associated with serious morbidity and a high mortality risk. The risk of neonatal seizures varies inversely with birth weight. As reported by Lanska et al[9] also, in the present study the incidence of seizures in very low birth weight babies was 10.14% and 0.59% in normal birth weight babies. Seizure types in newborn infants vary considerably from those observed in older infants, and the types in premature infants differ from those in term infants. Unlike older infants, neonates rarely have well organized generalized tonic clonic seizures. Premature infants have even less well organized spells than full term babies. In the present study multifocal clonic seizures (42.24%) were the most common seizure type followed by generalized tonic (21.55%), subtle (8.19%), focal clonic (6.47%) and myoclonic (0.86%). However, Mizrahi and Kellaway[10] and Scher et al[11] have reported subtle seizures as the most common type of neonatal seizures in their studies.
Perinatal asphyxia is the most frequent cause of neonatal seizures. In various studies reported incidence is 15-53%.[12],[13] Similarly in present study perinatal asphyxia was the most common cause of neonatal seizures with 44.4% babies having seizures. Though characteristic time of onset of seizure in HIE is 8-36 hours,[5] the median age of onset of seizures in HIE I, II and III was 11 hours, 12 hours and 23 hours respectively in the present study. In babies having seizures because of perinatal asphyxia per se (after excluding seizure episodes associated with hypocalcemia and hypoglycemia), multifocal clonic seizures were the most common type in the present study as opposed to myoclonic seizures present in 64.7% of cases reported by Mizrahi and Kellaway et al[10] The exact contribution of hypoglycemia and hypocalcemia as a cause of seizure in babies with perinatal asphyxia is not certain. In the present study, out of 108 seizure episodes in babies with perinatal asphyxia, 11 episodes and 18 episodes were associated with hypoglycemia and hypocalcemia respectively. EEG evidence of diffuse cerebral injury like burst suppression, multifocal sharp waves, dysmaturity, attenuation of background and electrical silence is seen in HIE and meningitis[5] but exact incidence of these findings has not been well established. In HIE stage I, EEG is usually normal, in stage II EEG shows discontinuous pattern and stage III isoelectric or markedly discontinuous EEG parttern.[14] In present study, out of 30 EEG's done in babies with HIE, 8 (26.67%) were abnormal. Out of these, abnormalities of background (low voltage record and very low voltage record in one case each) were seen in 2 cases each of HIE II and III. Multifocal discharges (2 having sharp waves and spikes each) were seen in 3 cases of HIE II and 1 case of HIE I. Earlier workers usually did not correlate specific abnormalities recorded in standard EEG to the stage of HIE but just noted whether it was normal or abnormal. Rose and Lombroso[15] and Mizrahi and Kellaway[10] reported abnormalities in standard EEG in HIE in 70% and 46.3% cases respectively.
Hypoglycemia is most frequent in low birth weight babies, most of them being small for gestational age and in infants of diabetic mothers. Incidence of hypoglycemia as reported in literature varies from 2% to 26.6%.[13],[16] The incidence of seizures due to hypoglycemia in the present study was 11.11%. Seizures due to hypoglycemia typically occur on 2nd postnatal day.[17] Similarly in the present study the median age of onset of seizures was 63.5 hours. Multifocal clonic seizures were the most common seizure type, being present in 36.36% seizure episodes. In neonatal hypoglycemic seizures, EEG produces multifocal discharges. However there is no systematic study available showing incidence of EEG changes in hypoglycemia. Most of these observations are based on individual case reports. In the present study, out of 9 EEG's done in babies with hypoglycemia, 3 (33.3%) were abnormal, one having marked suppression of background activity and 2 having discharges (1 each with multifocal and focal discharges).Abnormal record in standard EEG in hypoglycemia in neonatal seizures has been reported in 42.9% cases by Rose and Lombroso[15] and in 4.9% cases by Mizrahi and kellaway.[10] The incidence of hypocalcemic seizures ranges from 1.1% to 22%.[18],[19] Early onset hypocalcemia occurs in first 2-3 days of life.[16] 10% of babies in the present study at the median age of 45 hours developed hypocalcemic seizures, with multifocal clonic seizures in 55.36% cases. The electroclinical seizures of primary hypocalcemia with typical clonic movements show spike discharges in normal full term neonates.[14] In the present study, EEG's were done in 7 babies with hypocalcemia. Out of these 3 (42.86%) were abnormal having epileptiform discharges. Two of these were multifocal and one was focal. Abnormal record in standard EEG in hypocalcemia has been reported in 34.4% by Keen and Lee[20] and in 27.8% by Rose and Lombroso.[15]
EEG's were done in 2 babies with meningitis, out of which 1(50%) was abnormal with marked suppression of background activity. Abnormal standard EEG's in meningitis with or without other sepsis (not separated by authors) has also been reported in 33.3% cases by Rose and Lombroso[15] and 17% by Mizrahi and Kellaway.[10] The reported incidence of undetermined etiology of clinical seizures varies from 2.4% to 5.3%.[10],[19] In the present study, in 14.44% babies, no definite cause of clinical seizures could be ascertained. Possible reasons could be non-availability of in-house USG cranium and EEG facilities, short period survival between first seizure episode and death etc.
References
1.Volpes J. Neonatal seizures. N Engl J Med 1973; 289 : 164-168.
2.Kuppuswami Scale for Scoio-economic status in urban area (1994). Source: Kuppuswami B. Manual of Socio-economic status Scale (URBAN) Manasayan, 32, Netaji Subhash Nagar, Delhi - 110006.
3.Manmohan, Shivprasad SR, Chellani HK, Kapni V. Intrauterine Growth curves in North Indian babies: weight, length, head circumference and ponderal index. Indian Pediatr 1990; 27 : 43-51.
4.American EEG Society. Guidelines in EEG, 1-7. Report of the American EEG Society Ad Hoc Committee on Techniques . 1985.
5.Rennie JM, Bylan GB. Neonatal seizures. In David TJ, ed. Recent Advances in Paediatrics 18. Churchill Livingstone, Edinburgh 2002; pp. 19-32.
6.Laroia N. Current Controversies in Diagnosis and Management of Neonatal Seizures. Indian Pediatr 2000; 37 : 367-372. [PUBMED] [FULLTEXT]
7.Bernes SM, Kaplan AM. Evolution of neonatal seizures. Ped Clin North Am 1994; 41(5) : 1069-1104.
8.Rennie JM. Neonatal seizures. Eur J Pediatrics 1997; 156 : 83-87. [PUBMED] [FULLTEXT]
9.Lanska MJ, Lanska DJ, Baumann RJ, Kryscio RJ. A populations based study of neonatal seizures in Fayette county, Kentucky. Neurology 1995; 45 : 724-732. [PUBMED]
10.Mizrahi EM, Kellaway P. Characterization and Classification of neonatal seizures. Neurology 1987; 37 : 1837-1844. [PUBMED]
11.Scher MS, Aso K, Beggarly ME, Hamid MY, Steppe DA, Painter MJ. Electrographic Seizures in Preterm and Full-term Neonates: Clinical Correlates, Associated Brain Lesions and Risk for Neurologic Sequelae. Pediatrics 1993; 91(1) : 128-134.
12.Levene MI, Trounce JQ. Causes of Neonatal Convulsions. Arch Dis Child 1986; 61 : 78-79. [PUBMED]
13.Goldberg HJ. Neonatal convulsions - a ten year review. Arch Dis Child 1983; 57 : 633-635.
14.Rajadhyaksha S, Shah KN, Jadhav VM. Electroencephalogram in a Neonate. Jour Neonatol 2000; 14(2) : 35-39.
15.Rose AL, Lombroso CT. Neonatal seizure status: A Study of Clinical, Pathological and Electroencephalographic features in 137 Full Term Babies with a Long Term Follow Up. Pediatrics 1970; 45 : 404-425. [PUBMED]
16.Naveen B, Vanitha AJ, Vijai Kumar HV, Gururaj S, Satish, Rajashekhar. Etiological Profile of Neonatal Seizures in a Referral Centre. Abstract of scientific paper presented at 20th Annual convention of NNF. at Mumbai, Nov 2000.
17.Volpe JJ. Neonatal seizures. In: volpe JJ. Neurology of Newborn 4th Ed. W.B. Saunders, Philadelphia 2001;pp178-214.
18.Bergman I, Painter MJ, Hirsch RP et al. Outcome in Neonates with Convulsions treated in ICU. Ann Neurol 1983;14:642-647.
19.Westerlaine CGN and Vert P (eds.). Analysis of neonatal seizures In: Westerline CGN and Vert P. Neonatal seizures. Raven Press, New York 1990;pp 1-9.
20.Keen JH , Lee D. Sequelae of Neonatal Convulsions: A Study of 112 Infants. Arch Dis Child 1973; 48 : 542-546. [PUBMED](Kumar Ajay, Gupta Ashish, Talukdar Bibek)
Objective. To evaluate clinical, etiological and EEG profile of neonatal seizures. Methods. In all the neonates enrolled in the study baseline information recorded was sex, gestational age, weight, ponderal index etc. Clinical profile of seizure episode included age at onset of seizure, type and duration of seizure, associated autonomic changes, medication given, response time to medication and possible cause. Relevant maternal history including antenatal and peripartum history was recorded. Relevant history and examination of newborn were noted. Essential investigations done in all subjects included blood glucose, serum calcium, serum sodium and arterial pH. USG cranium and EEG were done at earliest in all the subjects wherever feasible. Additional investigations were done as guided by history, physical examination and essential investigations. Results. Ninety babies with clinical seizures were enrolled into the study over one year period with an overall incidence of 1.17% (0.69% in term babies and 6.14% in preterm babies). Abnormal EEG's were found in one third cases out of 60 EEG's done in 90 babies. 26.7% of babies with perinatal asphyxia had abnormal EEG's (8/30). While 60% of babies with HIE II had abnormal discharges, background activity was suppressed in 66.66% EEG's in babies with HIE III. Conclusions. Overall incidence of neonatal seizures was 11.7/1000 live births, majority being preterm very low birth weight babies before 5 days of life. Perinatal asphyxia was responsible in 44.44% babies followed by metabolic abnormalities (23.33%). EEG abnormalities were present in 33.34% babies.
Keywords: Neonatal seizures; Electroencephalography
Clinical seizures are defined as paroxysmal alteration in neurological function i.e. behavioral, motor and/or autonomic function. Newborn period is the time when the incidence of seizures is the highest, yet their clinical recognition is difficult, therefore true incidence of neonatal seizures is difficult to determine. Seizures are often the first sign of neurological dysfunction in newborn but their clinical expression at this age is quite variable, poorly organized and often subtle. Volpe[1] has classified seizures into five clinical types, viz. subtle, multifocal clonic, focal clonic, generalized tonic and myoclonic. Neonatal seizures are difficult to investigate and consequently determination of etiology and initiation of therapy may be delayed which results in poor neurological outcome. Neonatal seizures can be due to various causes like hypoxic-ischemic encephalopathy, intracranial hemorrhage, meningitis, hypoglycemia, hypocalcemia, congenital malformation, etc. Though EEG provides a useful non-invasive test to diagnose neonatal seizures and evaluate degree of perinatal damage to brain and long term prognosis, yet its interpretation is influenced by variations in normal maturation process of brain. It is well recognized that not all seizures can be picked by surface recorded EEG and many clinically silent electrographic seizures have been reported. Therefore present study was conducted to evaluate the correlation of EEG with clinical and etiological profile of neonatal seizures.
Materials and Methods
The study was conducted in the tertiary level Neonatal Unit of a teaching hospital over one year period. All the babies born at Obstetric Unit of the hospital and admitted with clinically identified seizures before 28 days of age were enrolled into the study. Ninety babies developed clinical seizures over study period. In all neonates enrolled into the study, the information outlined below was collected and recorded.
(A) Baseline Characteristics: These included sex, gestational age, intrauterine growth status, maternal education and socio-economic status (Kuppuswami scale).[2] Weight, head circumference and length were recorded at birth by standard technique and Ponderal index was calculated. Babies were categorized into appropriate for gestational age (AGA) and small for gestational age (SGA)[symmetrical and asymmetrical] categories according to percentile curves developed by Manmohan et al.[3]
(B) Clinical Profile of Seizures : Each seizure episode reported by mother and subsequently observed by resident doctor on duty was recorded and relevant information was gathered: age at onset of seizure in day(s), type of seizure, duration of seizure in minutes, associated autonomic changes (HR, RR, O 2 saturation), medication required to control seizures, response time to medication, and possible cause.
(C) Determination of Etiology : Relevant clinical information was collected and recorded. Antenatal history of infections including TORCH, drug intake, pregnancy induced hypertension (PIH), gestational diabetes, etc. was taken. Peripartum history of PIH, antepartum hemorrhage, chorioamnionitis, fetal distress (fetal heart rate abnormalities or meconium stained liquor) was recorded. Mode of presentation and delivery was noted. Resuscitation details if done were also recorded. All babies were examined daily, including standard neurological examination till they were seizure free for 3 days or till discharge whichever was later.
(D) Investigations: Essential Investigations done in all the subjects included blood glucose, total serum calcium, ionized serum calcium, serum sodium, and arterial pH. USG of cranium and EEG were done at earliest preferably within 48 hours of onset of seizure or when the neonate became stable, whichever was earlier. Electro-encephalography was obtained on SLE microscibe EEG machine having 21 channels manufactured by SLE, UK. The techniques used were as per guidelines of the American EEG society for recording of EEG in the neonates.[4] The montages used were monopolar, bipolar and referential. Triclophos was used in dose of 50 mg/Kg body weight for inducing sleep where needed. Babies were also subjected to photic stimulation to provoke any abnormalities in brain activity. Each EEG recording was made for 30 minutes with different montages. Additional investigations were done as guided by history, physical examination and essential investigations. They included sepsis screen, blood culture, cerebrospinal fluid (CSF) examination, hematocrit, serum bilirubin levels, serum urea and creatinine, "TORCH" serology, VDRL, serum magnesium, serum parathyroid hormone, metabolic profile for inborn errors of metabolism. CT scan and MRI were done whenever indicated. All babies were treated using standard treatment protocol.
Statistical Analysis: The data were entered into computer using Epi info 6 software. Normally distributed data were analysed using mean (SD) while non guassian distributed data were analysed using median(range).
Results
During the study period, a total of 7728 babies were born live. Clinical seizures were identified in 90 babies during hospital stay, with an incidence of 11.65/1000 live births. The incidence in preterm babies was 6.14% in contrast to only 0.69% in term babies. Early onset seizures (< 48 hours) were observed in 2.84% preterm neonates and 0.47% term neonates. Seizures developed in 10.14% babies in very low birth weight, 1.51% babies in low birth weight (1500-2499 grams) and 0.59% babies in normal birth weight (2500-3999 grams) groups. Out of 90 babies developing clinical seizures, 47 were small for gestational age (twelve being symmetrical).
A total of 232 seizure episodes occurred in ninety babies with an average of 2.58 seizure episodes per baby (range 1 to 7 episodes) in first 28 days of life. The median age of onset of first seizure episode was 41.5 hours (in term babies 38 hours, in preterm babies 55 hours), with 75% of seizure episodes occurring before 115 hours of age. Fifty two babies (57.8%) developed first seizure within first 48 hours of life, out of which 20 babies had seizure in less than 12 hours of life. Autonomic changes (heart rate, respiratory rate or oxygen saturation of blood) were present in 108 seizure episodes (46.55%). Out of 232 seizure episodes, multifocal clonic seizures contributed a major proportion with 42.24%, followed by generalized tonic seizures (21.55%). Among mixed seizure episodes, multifocal clonic with subtle seizures were present in 21 seizure episodes followed by generalized tonic with subtle seizures in 18 episodes.
Perinatal asphyxia was the most common cause of neonatal seizures in both preterm (39.02%) and term (48.98%) babies. Thirty three babies had abnormal neurological state between seizures (hypoxic ischemic encephalopathy) and thus were diagnosed to have seizures because of birth asphyxia [Table - 1]. The median (range) age of onset of seizures in HIE stages I, II and III were 11 hours (1-47 hours), 12 hours (1-58 hours) and 23 hours (4-437 hours) respectively. Out of 108 seizure episodes in babies with perinatal asphyxia, 11 and 18 were associated with hypoglycemia and hypocalcemia while rest of 79 episodes were due to HIE per se [Table - 2]. Multifocal clonic and generalized tonic seizures were the most common type of seizures associated with perinatal asphyxia per se . 0 EEG was done in 30 babies out of 40 babies with perinatal asphyxia [Table - 1][Table - 2]. Five babies had abnormal EEG record out of eleven EEG's done in 13 babies in HIE stage II. Background abnormality was present in two babies, one having low voltage record and other very low voltage record. In another three babies multifocal discharges were present, one having spikes and two having sharp waves [Table - 3]. Brain edema was detected in 3 babies with HIE stage II on USG cranium.
Out of ninety babies, 21 babies (23.34%) had seizures associated with hypoglycemia . Ten babies were low birth weight while other eleven babies had associated morbidity like birth asphyxia, meningitis etc. Multifocal clonic seizures were the most common type of seizure episodes associated with hypoglycemia. In 89 babies in whom total serum calcium level was estimated, 22 babies (24.72%) had seizures associated with low total serum calcium level (preterm baby n=8, serum calcium < 7.5 mg/dL; term baby n=14, serum calcium < 8mg/dL). Nine babies were low birth weight while other thirteen babies had associated morbidities similar to hypoglycemic babies. Hypocalcemia alone or associated with other morbidities had multifocal clonic seizure type in majority of seizure episodes (31/56). Abnormal discharges were noted in two and three babies with hypoglycemia and hypocalcemia respectively out of nine and eight EEG's done. One baby with hypoglycemia had background abnormality also [Table - 3]. One baby had hypernatremia (serum sodium 185 meq/L) associated with high content of breast milk sodium (59.6 meq/L). Eighteen seizure episodes were noticed in 7 babies with meningitis , out of which seven were multifocal clonic seizures.
In 13 babies cause of seizure could not be ascertained [Table - 1]. Nine babies died, out of which 7 babies within 3 days of onset of first seizure episode and hence could not undergo essential (USG cranium and EEG) and relevant additional investigations. In rest of the 6 babies, EEG abnormality was present in 3 cases [Table - 3]. Multifocal discharges were present in all the 3 babies along with background abnormalities in 2 cases (one with low voltage record and other asymmetrical background activity). USG cranium failed to detect any abnormality responsible for seizure episode.
Discussion
In most of the clinical settings including neonatal intensive care units, seizures in the newborn are identified by clinical observation. However, some of the clinically identified motor and behavioral phenomena characterized as seizures do not have simultaneous EEG correlate, thus overestimating neonatal seizures. Conversely many electrographic seizures are not accompanied by clinically observed alternations in neonatal motor or behavioral function, thus underestimating neonatal seizures. The incidence of neonatal seizures as reported by various studies range from 0.1-0.5% in term neonates and 10-22.7% in preterm neonates.[5],[6],[7] The present study has found an overall incidence of clinical seizures in 1.17% babies with 0.69% in term babies and 6.14% in preterm babies. Early onset seizures were observed in 0.47% term neonates in the present study though the reported incidence in literature varies from 0.87/1000 to 2.8/1000 live births.[8] Low birth weight is an important risk factor, as clinical seizures in them often indicate severe brain injury and are associated with serious morbidity and a high mortality risk. The risk of neonatal seizures varies inversely with birth weight. As reported by Lanska et al[9] also, in the present study the incidence of seizures in very low birth weight babies was 10.14% and 0.59% in normal birth weight babies. Seizure types in newborn infants vary considerably from those observed in older infants, and the types in premature infants differ from those in term infants. Unlike older infants, neonates rarely have well organized generalized tonic clonic seizures. Premature infants have even less well organized spells than full term babies. In the present study multifocal clonic seizures (42.24%) were the most common seizure type followed by generalized tonic (21.55%), subtle (8.19%), focal clonic (6.47%) and myoclonic (0.86%). However, Mizrahi and Kellaway[10] and Scher et al[11] have reported subtle seizures as the most common type of neonatal seizures in their studies.
Perinatal asphyxia is the most frequent cause of neonatal seizures. In various studies reported incidence is 15-53%.[12],[13] Similarly in present study perinatal asphyxia was the most common cause of neonatal seizures with 44.4% babies having seizures. Though characteristic time of onset of seizure in HIE is 8-36 hours,[5] the median age of onset of seizures in HIE I, II and III was 11 hours, 12 hours and 23 hours respectively in the present study. In babies having seizures because of perinatal asphyxia per se (after excluding seizure episodes associated with hypocalcemia and hypoglycemia), multifocal clonic seizures were the most common type in the present study as opposed to myoclonic seizures present in 64.7% of cases reported by Mizrahi and Kellaway et al[10] The exact contribution of hypoglycemia and hypocalcemia as a cause of seizure in babies with perinatal asphyxia is not certain. In the present study, out of 108 seizure episodes in babies with perinatal asphyxia, 11 episodes and 18 episodes were associated with hypoglycemia and hypocalcemia respectively. EEG evidence of diffuse cerebral injury like burst suppression, multifocal sharp waves, dysmaturity, attenuation of background and electrical silence is seen in HIE and meningitis[5] but exact incidence of these findings has not been well established. In HIE stage I, EEG is usually normal, in stage II EEG shows discontinuous pattern and stage III isoelectric or markedly discontinuous EEG parttern.[14] In present study, out of 30 EEG's done in babies with HIE, 8 (26.67%) were abnormal. Out of these, abnormalities of background (low voltage record and very low voltage record in one case each) were seen in 2 cases each of HIE II and III. Multifocal discharges (2 having sharp waves and spikes each) were seen in 3 cases of HIE II and 1 case of HIE I. Earlier workers usually did not correlate specific abnormalities recorded in standard EEG to the stage of HIE but just noted whether it was normal or abnormal. Rose and Lombroso[15] and Mizrahi and Kellaway[10] reported abnormalities in standard EEG in HIE in 70% and 46.3% cases respectively.
Hypoglycemia is most frequent in low birth weight babies, most of them being small for gestational age and in infants of diabetic mothers. Incidence of hypoglycemia as reported in literature varies from 2% to 26.6%.[13],[16] The incidence of seizures due to hypoglycemia in the present study was 11.11%. Seizures due to hypoglycemia typically occur on 2nd postnatal day.[17] Similarly in the present study the median age of onset of seizures was 63.5 hours. Multifocal clonic seizures were the most common seizure type, being present in 36.36% seizure episodes. In neonatal hypoglycemic seizures, EEG produces multifocal discharges. However there is no systematic study available showing incidence of EEG changes in hypoglycemia. Most of these observations are based on individual case reports. In the present study, out of 9 EEG's done in babies with hypoglycemia, 3 (33.3%) were abnormal, one having marked suppression of background activity and 2 having discharges (1 each with multifocal and focal discharges).Abnormal record in standard EEG in hypoglycemia in neonatal seizures has been reported in 42.9% cases by Rose and Lombroso[15] and in 4.9% cases by Mizrahi and kellaway.[10] The incidence of hypocalcemic seizures ranges from 1.1% to 22%.[18],[19] Early onset hypocalcemia occurs in first 2-3 days of life.[16] 10% of babies in the present study at the median age of 45 hours developed hypocalcemic seizures, with multifocal clonic seizures in 55.36% cases. The electroclinical seizures of primary hypocalcemia with typical clonic movements show spike discharges in normal full term neonates.[14] In the present study, EEG's were done in 7 babies with hypocalcemia. Out of these 3 (42.86%) were abnormal having epileptiform discharges. Two of these were multifocal and one was focal. Abnormal record in standard EEG in hypocalcemia has been reported in 34.4% by Keen and Lee[20] and in 27.8% by Rose and Lombroso.[15]
EEG's were done in 2 babies with meningitis, out of which 1(50%) was abnormal with marked suppression of background activity. Abnormal standard EEG's in meningitis with or without other sepsis (not separated by authors) has also been reported in 33.3% cases by Rose and Lombroso[15] and 17% by Mizrahi and Kellaway.[10] The reported incidence of undetermined etiology of clinical seizures varies from 2.4% to 5.3%.[10],[19] In the present study, in 14.44% babies, no definite cause of clinical seizures could be ascertained. Possible reasons could be non-availability of in-house USG cranium and EEG facilities, short period survival between first seizure episode and death etc.
References
1.Volpes J. Neonatal seizures. N Engl J Med 1973; 289 : 164-168.
2.Kuppuswami Scale for Scoio-economic status in urban area (1994). Source: Kuppuswami B. Manual of Socio-economic status Scale (URBAN) Manasayan, 32, Netaji Subhash Nagar, Delhi - 110006.
3.Manmohan, Shivprasad SR, Chellani HK, Kapni V. Intrauterine Growth curves in North Indian babies: weight, length, head circumference and ponderal index. Indian Pediatr 1990; 27 : 43-51.
4.American EEG Society. Guidelines in EEG, 1-7. Report of the American EEG Society Ad Hoc Committee on Techniques . 1985.
5.Rennie JM, Bylan GB. Neonatal seizures. In David TJ, ed. Recent Advances in Paediatrics 18. Churchill Livingstone, Edinburgh 2002; pp. 19-32.
6.Laroia N. Current Controversies in Diagnosis and Management of Neonatal Seizures. Indian Pediatr 2000; 37 : 367-372. [PUBMED] [FULLTEXT]
7.Bernes SM, Kaplan AM. Evolution of neonatal seizures. Ped Clin North Am 1994; 41(5) : 1069-1104.
8.Rennie JM. Neonatal seizures. Eur J Pediatrics 1997; 156 : 83-87. [PUBMED] [FULLTEXT]
9.Lanska MJ, Lanska DJ, Baumann RJ, Kryscio RJ. A populations based study of neonatal seizures in Fayette county, Kentucky. Neurology 1995; 45 : 724-732. [PUBMED]
10.Mizrahi EM, Kellaway P. Characterization and Classification of neonatal seizures. Neurology 1987; 37 : 1837-1844. [PUBMED]
11.Scher MS, Aso K, Beggarly ME, Hamid MY, Steppe DA, Painter MJ. Electrographic Seizures in Preterm and Full-term Neonates: Clinical Correlates, Associated Brain Lesions and Risk for Neurologic Sequelae. Pediatrics 1993; 91(1) : 128-134.
12.Levene MI, Trounce JQ. Causes of Neonatal Convulsions. Arch Dis Child 1986; 61 : 78-79. [PUBMED]
13.Goldberg HJ. Neonatal convulsions - a ten year review. Arch Dis Child 1983; 57 : 633-635.
14.Rajadhyaksha S, Shah KN, Jadhav VM. Electroencephalogram in a Neonate. Jour Neonatol 2000; 14(2) : 35-39.
15.Rose AL, Lombroso CT. Neonatal seizure status: A Study of Clinical, Pathological and Electroencephalographic features in 137 Full Term Babies with a Long Term Follow Up. Pediatrics 1970; 45 : 404-425. [PUBMED]
16.Naveen B, Vanitha AJ, Vijai Kumar HV, Gururaj S, Satish, Rajashekhar. Etiological Profile of Neonatal Seizures in a Referral Centre. Abstract of scientific paper presented at 20th Annual convention of NNF. at Mumbai, Nov 2000.
17.Volpe JJ. Neonatal seizures. In: volpe JJ. Neurology of Newborn 4th Ed. W.B. Saunders, Philadelphia 2001;pp178-214.
18.Bergman I, Painter MJ, Hirsch RP et al. Outcome in Neonates with Convulsions treated in ICU. Ann Neurol 1983;14:642-647.
19.Westerlaine CGN and Vert P (eds.). Analysis of neonatal seizures In: Westerline CGN and Vert P. Neonatal seizures. Raven Press, New York 1990;pp 1-9.
20.Keen JH , Lee D. Sequelae of Neonatal Convulsions: A Study of 112 Infants. Arch Dis Child 1973; 48 : 542-546. [PUBMED](Kumar Ajay, Gupta Ashish, Talukdar Bibek)