The Incidence of Acute and Remote Seizures

in Children with Intraventricular Hemorrhage

Jonathan B. Strober, MD
Robert S. Bienkowski, PhD
Joseph Maytal, MD

Summary: Seizures are a well-known complication of intraventricular hemorrhage (IVH) in

premature infants; however, the rate at which they occur is not known. The authors decided,
therefore, to investigate both the incidence of acute and remote seizures in infants with IVH and
the association with the grade of hemorrhage. One hundred and three infants with IVH were
identified and their records were reviewed for acute seizures, remote seizures, and associated
morbidity and mortality. The average gestational age of these infants was 29 weeks (range, 23-40
weeks). Of the 103 infants, 32 (31%) developed grade 4 IVH; 19 (18%), grade 3 IVH; and 52 (50%),
grades 1 and 2 1VH. Seventeen (17%) patients had acute seizures during their first month of life.
Six of the 61 patients (10%) who survived the neonatal period and for whom follow-up data were
available had, remote seizures. Infants with grade 4 IVH had significantly more acute seizures than
infants with grades 1 and 2. In this cohort, only infants with grades 3 and 4 IVH developed remote
seizures. Furthermore, among infants with grade 4 IVH acute seizures were a significant risk factor
for development of remote seizures. The use of long-term antiepileptic drug therapy in neonates
with a history of acute seizures is not established. These results suggest that antiepileptic drug
therapy beyond the neonatal period should be reserved for infants with grade 4 IVH with history
of acute seizures. Clin Pediatr. 1997;36:643-648

Introduction among premature infants.'-3 De- sual impairments.' Neonatal

pending on the extent of the he- seizures present a high risk for
I ntraventricular hemorrhage morrhage, infants can develop neonatal death and adverse neu-
(IVH) is an important cause seizures, cerebral palsy, mental re- rodevelopmental sequelae for
of morbidity and mortality tardation, hearing deficits, and vi- both preterm and full-term in-
fants.4,5 The information about
the incidence of acute seizures in
Division of Pediatric Neurology and Pediatric Research Center, Schneider Children's Hospital, neonates with IVH is scant,246
Long Island Jewiish Medical Center, The Long Island Campus for the Albert Einstein College and only little information is avail-
of Medicine, New Hyde Park, NY. able about the incidence of re-
Presented in part at the Child Neurology Society Meeting, San Francisco, California,
mote seizures in this population.
October 1994. In one study, 13% of infants with
IVH developed remote seizures;2
Reprint requests and correspondence to: Joseph Maytal, MD, Division of Pediatric Neurology, in another study, 63% of full-term
Schneider Children's Hospital, Long Island Jewish Medical Center, New Hyde Park, New York infants with IVH had remote
11040. seizures, mostly occurring after 1
© 1997 Westminster Publications, Inc., 708 Glen Cove Avenue, Glen Head, NY 11545, U.S.A. year of age.6

Downloaded from cpj.sagepub.com at UNIV OF PENNSYLVANIA on June 29, 2015

NOVEMBER 1997 CLINICAL PEDIATRICS 643
 Strober, Bienkowski, Maytal

Prophylactic administration of the ventricular system in associa- the incidence of acute and re-
antiepileptic drugs to infants with tion with ventriculomegaly. mote seizures and to calculate the
histories of neonatal seizures is Seizures were diagnosed by risk of remote seizures in grade 4
controversial, and no guidelines the health caregiver (i.e., doctor IVH children with history of acute
have been established regarding or nurse) witnessing the event neonatal seizures.
the role of long-term antiepileptic and the consulting pediatric neu-
drug therapy for these infants.7-9 rologist concurring with the diag-
Many authors suggest continuing nosis. Seizures were defined clini-
antiepileptic drug therapy be- cally as paroxysmal alteration in
Results
yond the neonatal period for in- neurologic function and were
fants with histories of perinatal in- classified as clonic, tonic, my- Characteristics ofStudy Cohort
sult and abnormal findings from oclonic,and subtle.12 Acute One hundred and three in-
neurologic examinations.10-12 seizures were defined as seizures fants (45 females, 58 males) with
However, there are concerns occurring during the first month all grades of IVH were identified;
about potential long-term delete- of life. Remote seizures were de- 18 (17%) of these infants experi-
rious effects of phenobarbital and fined as seizures occurring after enced acute and remote seizures.
other antiepileptic drugs.12 the first month of life. Other eti- Of the total 103 infants who de-
In this study we investigated ologies for acute seizures, e.g., hy- veloped IVH, 32 (31%) had grade
both the incidence of acute and poglycemia, were excluded be- 4. 19 (18%) had grade 3, 21
remote seizures in premature in- fore the seizures were ascribed to (20%) had grade 2, and 31 (30%)
fants who suffered from IVH IVH. had grade 1 IVH (Table 1). The
while in the neonatal intensive Follow-up information for gestational ages of these infants
care unit and the association be- those infants who survived the ranged from 23 to 40 weeks, with
tween the grade of hemorrhage neonatal period was obtained by an average of 29 weeks for grade
and the seizure outcome in these reviewing their records in the Di- 1, 28 weeks for grade 2, and 27
children. visions of Neurology and of Devel- weeks for grades 3 and 4 IVH. Of
opmental Pediatrics and by tele- the 13 infants who expired during
phoning the pediatricians of the neonatal period, one had
Materials and Methods record on the discharge sum- grade 3 IVH, and 12 had grade 4.
maries. Information was obtained Follow-up information was avail-
Patients were identified by about seizures after the neonatal able for 61 of the 90 infants who
searching the neonatal database period and the treatment, as well survived the neonatal period.
for infants with grades 1-4 IVH as the developmental and physi- This group included 18 of 20
who were discharged from the cal status of the children. Devel- (90%) surviving infants who had
Schneider Children's Hospital be- opmental delay was categorized as grade 4 IVH. The mean follow-up
tweenJanuary 1, 1992, and Octo- mild, moderate, or severe accord- time was 2 years and 7 months
ber 15, 1993. Data regarding mor- ing to the criteria developed by (range 6 months to 4 years, 3
bidity and mortality, radiologic Saigal et al.'4 Mild disability is de- months).
studies, and the occurrence of fined as mild diplegic cerebral Developmental delay was re-
acute and remote seizures and palsy and independent ambula- ported in 30 of the 61 surviving
their treatment were obtained tion; moderate disability is de- infants for whom follow-up data
from medical records. All patients fined as moderate cerebral palsy, were available. Eight infants had
with 1VH had ultrasonograms to inability to walk independently, grade 1 (seven mild delay and one
establish the diagnosis. Severity of and a need for some additional moderate delay), one infant had
IVH was classified as follows:'3 caregiving; severe disability is de- grade 2 (moderate), five infants
grade 1, hemorrhage limited and fined as severe cerebral palsy, in- had grade 3 (two mild, two mod-
around the choroid plexus; grade ability to walk or severe develop- erate, and one severe), and 16 in-
2, hemorrhage into the ventricle mental retardation, or severe fants had grade 4 IVH (nine mild,
without hydrocephalus; grade 3, special-sense disorder requiring five moderate, and two severe).
hemorrhage into the ventricle significant additional caregiving. All patients who developed re-
with development of hydro- Fisher's Exact Test was used to mote seizures were developmen-
cephalus; grade 4, hemorrhage determine the association be- tally delayed, four severely, two
into the brain parenchyma and tween the grade of the IVH and moderately, and one mildly.

644
644 CLINICAL PEI)IATRICS
PEDIATRICS
Downloaded from cpj.sagepub.com at UNIV OF PENNSYLVANIA on June 29, 2015
NOVEMBER
NOVEMBER 1997
 Acute and Remote Seizures in Premature Infants with IVH

Acute Seizures Other seizure types exhibited by Of the 18 patients with grade 4
Acute seizures were recorded infants with grade 4 IVH included IVH whom we followed up, four
in 17 infants (17%; 95% CI for the generalized clonic, myoclonic, suffered both acute and remote
incidence: 10 to 25%). Twelve in- and subtle seizures. One infant seizures, one suffered acute
fants had grade 4 IVH, two had with grade 2 had a focal tonic seizures but no remote seizures,
grade 3, and three had grade 2. seizure. All infants with acute and one suffered remote seizures
No infants with grade 1 IVH had seizures were treated with pheno- but no acute seizures. Thus pa-
acute seizures (Table 1). Because barbital and one also required tients who had acute seizures dur-
of the relatively small sample size phenytoin. ing the neonatal period had a ten-
and the association of grades 1 fold greater risk for developing
and 2 IVH with good prognosis,' Remote Seizures chronic seizures than patients
infants with these grades were Remote seizures occurred in who did not have acute seizures
combined into a single group for six (10%) of the 61 infants for (P=0.008).
statistical analysis. Infants with whom follow-up information was
grade 4 IVH had a significantly available. Five of these infants had
higher proportion of acute grade 4 IVH and one had grade 3 Discussion
seizures than infants with grades 1 (Table 1). One infant with grade 4
and 2 (P=0.0007, Fisher's Exact IVH without an acute seizure later In this study we found a high
Test). The relative risk of having developed remote seizures. All in- incidence (17%) of acute seizures
acute seizures in infants with fants were treated with either phe- in neonates who suffered IVH.
grade 4 IVH versus infants with nobarbital alone or in combina- The data in the literature regard-
grades 1 and 2 was 6.5 (95% CI: 20 tion with carbamazepine or ing this issue are scant and the
to 21.3). However, infants with phenytoin. None of the 32 infants sample sizes of the few studies
grade 3 IVH did not have a signif- with grades 1 and 2 IVH had that investigated it are small. A
icantly increased proportion of seizures after the neonatal period; previous study of the incidence of
acute seizures (P=0.6). however, five of 18 infants with electrographically confirmed
The majority of seizures were grade 4 P/H did. The difference seizures in preterm infants by
characterized as generalized in proportions is significant Scher et a14 found that 4% had
tonic, which occurred in seven in- (P-0.0036). The proportion of in- acute seizures. The number of
fants with grade 4 IVH, one infant fants with grade 3 IVH who devel- seizures in neonates with LVH was
with grade 3, and two infants with oped remote seizures did not dif- not reported, but 28 of 62 (45%)
grade 2. Focal clonic seizures oc- fer significantly from the preterm infants with seizures had
curred in one infant with grade 4 proportion of infants with grades IVH. Aso et a12 documented
PVH and one infant with grade 3. 1 and 2 IVH (P=0.6) seizures in 13% (4/32) of neo-

NOVEMBER 1997
NOVEMBER
Downloaded from cpj.sagepub.com at UNIV OF PENNSYLVANIA on June 29, 2015 CLINICAL PEDIATRJ.S
CLINICAL PEDIATRICS
645
645
 I
Strober, Rienkowski, Maytal

nates with IVH who required EEG mal hemorrhage may be a poten- tal period. Similar conclusions
recordings. A study by Sugimoto tially important epileptogenic were reached for older children
et a16 on full-term infants with factor. and adults who had acute seizures
IVH reported that four of eight The development of remote secondary to CNS insults such as
(50%) had acute seizures. Funato seizures was infrequent in our trauma, infection, or hemor-
et al15 performed a continuous IVH cohort (six of 61, 10%, 95% rhage.'7-'9 Furthermore, the risk
monitoring of the germinal layer CI: 4 to 20%); however, the rate of seizure recurrence following a
by linear scanning ultrasound and among the infants with grade 4 remote seizure is increased if the
reported that one of four (25%) IVH was much higher (28%). All patient has a history of an acute
immature infants with IVH devel- but one infant with remote seizure following CNS insult.20
oped acute seizures. The data seizures had a history of acute Although our overall follow-
from these studies are compared seizures. Comparison with other up rate was limited, it was high
in Figure 1. In the figure note that studies cannot be made directly (90%) for infants with grade 4
the 95% CI ranges of the studies because of different recruitment IVH, who are usually more im-
are very broad because the sam- criteria. In reviewing the out- paired and have high long-term
ples are small. Also note that the comes of very-low-birth-weight in- morbidity.' In infants with grades
confidence intervals for the four fants (less than 1,500 grams), 1 and 2 IVH, the follow-up rate

studies overlap. Combining the Cooke'6 found only a 1.8% inci- was 62% (32/52). The fact that
data for the four separate studies dence of remote seizures in a none of these 32 patients had re-
yields an estimated incidence for 3-year follow-up period, and mote seizures (Table 1) indicates
acute seizure in the range of Scher et a14 reported six of 49 that the maximal rate of remote
15-25%. (12%) preterm infants studied seizures in this group is about
Infants with IVH may have with acute seizures later devel- 10% (the 95% confidence for the
acute seizures from a variety of oped remote seizures. ratio of 0/32 is 0 to 0.09). Of the
causes.12 The high incidence of Our data indicate that infants 20 patients who were lost to fol-
acute seizures that we found in in- with grade 4 IVH and acute low-up in this group, one might
fants with grade 4 IVH as com- seizures are at increased risk of expect two, at most, to have re-
pared with the rest of the IVH co- developing remote seizures as mote seizures. Longer follow-up
hort, and as compared with some compared with infants with grade will be necessary to assess the true
historical data on premature in- 4 IVH who did not experience long-term incidence of seizure re-
fants,16 indicates that parenchy- acute seizures during the neona- currence. The incidence of re-

Funato et a115 p_

Sugimoto et a16 Figure 1. A comparison of

the incidence of acute
seizures in infants with IVH
Aso et a12 in four studies and with a
group that combines data
is from all these studies. The
Strober et al i -
horizontal lines indicate the
95% confidence intervals for
Combined the incidence ratios.

I
* I
* I
. I.

0 20 40 60 80 100
Incidence of Acute Seizure in IVH (% I

646
646 CLINICAL PEDIATRICS
NOVEMBER
NOVEMBER 1997
1997
Downloaded from cpj.sagepub.com at UNIV OF PENNSYLVANIA on June 29, 2015
 Acute and Remote Seizures in Premature Infants with IVH

mote seizures may increase over fants with grade 4 IVH with a his- tal period. Arch Dis Child. 1995;72:F97-
time. tory of acute neonatal seizures, F1IO.
Our study lacks electroen- because these patients are at a 12. Volpe JJ. Neurology of the Newborn.
cephalographic verification of higher risk for developing remote Philadelphia: WB Saunders Co;
acute seizures, and as a result we seizures. For this group of pa- 1995:172-207.
may have overestimated or under- tients we recommend antiepilep- 13. Papile LA, BursteinJ, Burstein R, Kof-
estimated the incidence of acute tic drug therapy at least for the fler H. Incidence and evolution of
subependymal and intraventricular
seizures in the neonatal period. first year of life before tapering
hemorrhage: a study of infants with
Nevertheless, in virtually all medications is considered.
birth weight less than 1,500 grams.
neonatal intensive care units, JPediatr. 1978;92:529-534.
seizures are diagnosed on clinical 14. Saigal S, Rosenbaum P, Hattersley B,
grounds and not by an EEG. An REFERENCES
Milner R. Decreased disability rate
EEG is usually obtained interic- 1. Volpe JJ. Neurology of the Newborn.
among 3-year-old survivors weighing
tally and therefore has lesser clin- Philadelphia: WB Saunders Co;
1995:403-463. 501 to 1,000 grams at birth and born
ical significance.21 Our clinical to residents of a geographically de-
2. Aso K, Scher M, Barmada MA. Cere-
data showing increased risk of bral infarcts and seizures in the
fined region From 1981 to 1984 com-
seizures in infants with grade 4 neonate.JChildNeurol. 1990;5:224-228.
pared with 1977 to 1980. J Pediatr.
IVH can be used to guide the clin- 3. Hill A, Volpe 1J. Seizures, hypoxic-is-
1989;1 14:839-846.
ician in these situations. These chemic brain injury, and intraventric- 15. Funato M, Tamai H, Kodaka R, et al.
data also support Volpe's recom- ular hemorrhage in the newborn. The moment of intraventricular hem-
mendation for empirical treat- Ann Neurol. 1981;10:109-121. orrhage. Brain Dev. 1988;10:325-327.
ment approach of neonatal 4. Scher MS, Aso K, Beggarly ME, et al. 16. Cooke RW. Annual audit of three year
seizures. 12 Electrographic seizures in preterm outcome in very low birth weight in-
The question of whether to and full-term neonates: clinical corre- fants. Arch Dis Child. 1993;69:295-298.
initiate long-term antiepileptic lates, associated brain lesions, and risk 17. AnnegersJF, Hauser WA, Beghai E, et
drug therapy in infants who expe- for neurologic sequelae. Pediatrics. al. The risk of unprovoked seizures af-
rienced IVH and acute seizures 1993;91:128-134. ter encephalitis and meningitis. Neu-

during the neonatal period is of 5. Mizrahi EM, Kellaway P. Characteriza- rology. 1986;38:1407-1410.
tion and classification of neonatal 18. AnnegersJF, GrabowJD, Groover RV,
great importance. In making a de-
seizures. Neurology. 1987;37:1837- et al. Seizures after head trauma: a
cision, the physician must take 1844.
into account not only the risk of population study. Neurology. 1980;
6. Sugimoto T, Woo M, Okazaki H. Neu- 30:683-689.
seizure recurrence in this popula- rological outcome in full term infants
tion and its consequences but also 19. Hauser WA, Ramirez-Lassepas M,
with intraventricular hemorrhage.
the risks of antiepileptic drug Rosenstein R. Risk for seizures and
Acta PediatrJpn. 1987;29:749-752.
epilepsy following cerebrovascular in-
therapy. There is a broad range of 7. Legido A, Clancy RR, Berman PH. Re-
sults. Epilepsia. 1984;25:666 (abstract).
practices among neonatologists cent advances in the diagnosis, treat-
20. Hauser WA, Rich SS, AnnegersJF, An-
and pediatric neurologists regard- ment, and prognosis of neonatal
derson VE. Seizure recurrence after a
ing prophylactic antiepileptic seizures. PediatrNeurol. 1988;4:79-86.
first unprovoked seizure: an extended
drug therapy in infants with his- 8. Mizrahi EM. Diagnosis and treatment
follow-up. Neurology. 1990;40:1163-
tory of acute neonatal seizures. of neonatal seizures. Int Pediatr.
1170.
These practices range from a 1989;4:180-187.
9. Scher MS, Painter MJ. Controversies 21. Bergman I, Painter MJ, Hirsh RP, et al.
brief course of medication in the Outcome in neonates with convul-
nursery to extending treatment
concerning neonatal seizures. Pediatr
Clin North Am. 1989;36:281-310. sions treated in an intensive care unit.
up to 2 years of age.7'12,22,23 Since Ann Neurol. 1983;14:642-647.
10. Andre M, Matisse P, Debruille Vert
the overall incidence of remote 22. Boer HR, Gal P. Neonatal seizures: a
Ch. Neonatal seizures-recent as-
seizures in infants with IVH is low, pects. Neuropediatrics. 1988;19:201- survey of current practice. Clin Pedi-
the use of long-term prophylactic 207. atr. 1982;21:453-457.
antiepileptic drug therapy may 11. Hellstrom-Westas L, Blennow G, Lin- 23. Ellison PH, LargentJA, BahrJP. A
not be justified in most of these droth M, et al. Low risk of seizure re- scoring system to predict outcome fol-
patients. Long-term antiepileptic currence after early withdrawal of lowing neonatal seizures. J Pediatr.
therapy should be limited to in- antiepileptic treatment in the neona- 1981;99:455-459.

NOVEMBER 1997
NOVEMBER
Downloaded from cpj.sagepub.com at UNIV OF PENNSYLVANIA on June 29, 2015 CLINICAL PEDIATRICS
CLINICAL PEDIATRICS
647
647