Documents

Hwang 2015

Description
Description:
Categories
Published
of 7
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
Share
Transcript
  Original article Predictors of unprovoked seizure after febrile seizure: Short-termoutcomes Gul Hwang, Hyun Sik Kang, Sin Young Park, Kyoung Hee Han, Seung Hyo Kim ⇑ Department of Pediatrics, Jeju National University College of Medicine, Republic of Korea Received 7 February 2014; received in revised form 14 May 2014; accepted 4 June 2014 Abstract Introduction:  We performed this study to confirm the known risk factors and to identify possible new risk factors for subsequentunprovoked seizure after febrile seizure (FS) on Jeju Island, South Korea. Methods:  A population-based retrospective study of 204 children with FS, whose first FS developed between March 2003 andAugust 2011, and who were seen in the Pediatric Department at the Jeju National University Hospital. Results:  Two hundred four children (136 boys and 68 girls) were enrolled in this study. Simple FS was found in 107 children, andcomplex FS was found in 97 children. The average age at the first FS was 18.9 months. The average total number of FSs was 4.3. Afamily history of FS or epilepsy was found in 29.4% and 7.8% of patients, respectively. Abnormal findings of EEG were observed in35.8%. Complex features in the first FS were noted in 28.9%. Subsequent unprovoked seizures occurred in 23.0%. Univariate anal-ysis showed that low parental educational level was one of several variables that were significantly related to unprovoked seizure.Parental educational level was not included in the multivariate model because of an insufficient sample size. Multivariate analysisidentified the following factors as significant predictors of unprovoked seizure: late onset of FS at age > 3 years, complex features inthe first FS, family history of epilepsy, and abnormal findings on EEG, and FS developed at a body temperature of <39   C. Conclusions:  We confirmed the known risk factors for subsequent unprovoked seizure and found that low parental educationalstatus may be a new prognostic indicator. However, further investigation using larger populations and a prospective design is neededto confirm that this is a valid prognostic factor for FS.   2014 The Japanese Society of Child Neurology. Published by Elsevier B.V. All rights reserved. Keywords:  Febrile seizure; Unprovoked seizure; Complex febrile seizure 1. Introduction Febrile seizure (FS) is the most common convulsivedisorder of childhood and has been reported in 2–5%of all children in the United States and Western Europe,5–10% in India, 9–10% in Japan, and 14% in Guam[1–4]. The main concern with FS is whether this benignphenomenon progresses to the development of unpro-voked seizure or epilepsy. Some investigations havereported that there is a variable incidence of unpro-voked seizure and epilepsy [1–8]. The variability maybe attributed to differences in the patients’ age, theduration of follow-up, the method of study such as pop-ulation-based or hospital-based study, and the studydesign such as a retrospective or prospective method.Following a first FS, 2–4% of children develop at leastone unprovoked seizure [3,5,6]. http://dx.doi.org/10.1016/j.braindev.2014.06.0030387-7604/   2014 The Japanese Society of Child Neurology. Published by Elsevier B.V. All rights reserved. ⇑ Corresponding author. Address: Department of Pediatrics, JejuNational University College of Medicine, Ara-1 Dong, Jeju-si, JejuSpecial Self-Governing Province, Jeju-do 690-767, Republic of Korea.Tel.: +82 10 9024 2947; fax: +82 64 717 1097. E-mail address:  shped@jejunu.ac.kr (S.H. Kim). www.elsevier.com/locate/braindev Brain & Development xxx (2014) xxx–xxx Please cite this article in press as: Hwang G et al. Predictors of unprovoked seizure after febrile seizure: Short-term outcomes. Brain Dev (2014),http://dx.doi.org/10.1016/j.braindev.2014.06.003  Some cohort studies and retrospective investigationshave confirmed the risk factors for unprovoked seizureafter FS; these include complex FS, neurodevelopmentalabnormalities, onset of FS at an early age, epileptiformactivity shown by electroencephalography (EEG), and afamily history of epilepsy [5–8]. However, these studieshave not added any new risk factors. There have beenfew reports on the comparison of risk factors for theoccurrence of unprovoked seizure after the classificationof FS according to simple FS or complex FS and includ-ing EEG data.We tried to evaluate the prognostic factors for unpro-voked seizure after FS in relation to the type of FS andto identify any new prognostic factors in children livingon Jeju Island, with a homogeneous population. 2. Patients and methods The Jeju National University Hospital (JNUH) is atertiary hospital located on Jeju Island, which is locatedat the southernmost part of South Korea. The popula-tion of Jeju Island was about 600,000 inhabitants inAugust 2013. This is the only university hospital on JejuIsland providing EEG for children or pediatric neurol-ogy consultation.We reviewed retrospectively the computerized medi-cal records of 323 patients whose first FS developedbetween 3 months and 5 years of age among the patientswho visited the outpatient clinic or were admitted to thepediatric department at the JNUH because of their firstFS attack between May 2003 and August 2011. The dataobtained from the records included demographic vari-ables, seizure semiology, family history, birth history,EEG findings, the family’s economic status, and paren-tal educational background.We excluded patients with evidence of intracranialinfection or afebrile seizure that had developed beforethe occurrence of FS. Patients with abnormal resultsfrom their neurological examination at presentationwere excluded. Of this population, all neurologicallynormal patients with at least 2 years of follow-up andEEG data were included in this study to evaluate theprognostic indicators, especially EEG findings.A pediatric neurologist determined the patients’developmental status by performing a neurologicalexamination and obtaining a history of developmentalmilestones. The pediatric neurologist also recorded themedical history, such as seizure semiology, family his-tory, and other variables obtained mainly from themothers at the outpatient clinic or ward at admission.We contacted parents by phone in cases of missinginformation such as the mother’s age at birth, monthlyincome, and parental educational background.EEG studies were performed using the 10–20 interna-tional system with bipolar and referential montages.EEG was performed within 7 days after the FS for14 patients, between 7 and 14 days after the FS for33 patients, between 14 and 21 days for 114 patients,and more than 21 days after the FS for 43 patients. Thefirst EEG was performed at the first FS for 96 patients,at the second FS for 32 patients, at the third FS for28 patients, and at the time of other FSs for patientswho experienced more than three attacks of FS. Theduration of the recording was 20–30 min. Each EEGwasinterpretedbyonepediatricneurologistwhoiscertif-icated by the Korean Child Neurology Society. Weclassified spike or sharp wave discharges on EEG asabnormalfindings.Imagingstudies,includingbraincom-puted tomography (CT) or magnetic resonance imaging(MRI), were performed when considered necessary.FS was defined as a provoked seizure where the acuteprovocation was fever (temperature > 38.3   C or 101   F)documented in the emergency department in a patientwith no evidence of an acute central nervous systeminfection. Prolonged (>15 min), focal convulsive activityor repeated occurrence within 24 h, was defined as acomplex FS. FS plus was defined as FS continuingbeyond 6 years of age. Nonfebrile seizure was definedas a seizure that developed after at least one episodeof FS. We defined low educational level for each parentas having high school graduation as the highest levelattained. We defined low economic status of the familyas having a monthly income less than 2,000,000 Koreanwon, based on the classification suggested by the Minis-try of Employment and Labor, 2013.Statistical analysis was performed using the  t  test forcontinuous variables, and Pearson’s chi-square andFisher’s exact test for dichotomous variables. Multivar-iate logistic regression was performed for the variableswith a  P   value < 0.05 in the univariate analysis. Oddsratios and 95% confidence intervals were also calculated.The statistics package SPSS for Windows (version 18.0;SPSS Inc., Chicago, IL) was used.The design of this study was approved by the institu-tional ethical review committee at JNUH. 3. Results We obtained information about FS for 323 patientsamong the 628 patients who visited our outpatient clinicor who were admitted to the ward because of their firstFS. Demographic variables such as age and sex did notdiffer significantly between the study population(323 patients) and patients who were not included in thestudy (the other 305 patients among the total populationof628 patients).Themale/femaleratioswere127/196and110/195 ( P   = 0.401), and the mean ages were 5.08 and4.89 years( P   = 0.384),respectively.WeincludedpatientswhosefirstFSdevelopedbetween3 monthsand5 yearsof age,thosewithatleast2 yearsfollow-upafterfirstFS,andthose who had EEG data, resulting in 246 patients.Forty-two patients were excluded because of suspected 2  G. Hwang et al./Brain & Development xxx (2014) xxx–xxx Please cite this article in press as: Hwang G et al. Predictors of unprovoked seizure after febrile seizure: Short-term outcomes. Brain Dev (2014),http://dx.doi.org/10.1016/j.braindev.2014.06.003  neurodevelopmental delay at the start of study ( n  = 33),unprovoked seizure before FS ( n  = 5), meningitis( n  = 3),orbraintumor( n  = 1).Finally,204 patientswereincluded in this study.Seventy-three percent (149/204) of the families of both parents have lived on Jeju Island for over threegenerations. Thus, our study population was relativelyhomogeneous and came from families that have livedon Jeju Island for a long period.Clinical and demographic characteristics classifiedaccording to the type of FS are summarized in Table 1.The mean follow-up duration after the first FS was49.4 ± 29.2 months. The ratio of simple FS versus com-plex FS was 1.1:1. The male:female ratio was 2:1. Clas-sification of the type of first FS showed that 171 (83.8%)comprised generalized seizure, and the other first FSscomprised partial seizure.The average age at the first FS was 19.2 ±11.3 months: 35.8% developed FS before 1 year of age,44.1% between 1 and 2 years of age, and 92.6% by3 years of age, and 7.4% showed late onset of FS at>3 years of age. Patients with complex FS developedthe first FS at a younger age than those with simpleFS ( P   = 0.006).The average number of episodes of FS in the groupwith complex FS was 5.4 ± 3.8 up to August 2013( P   = 0.001). The overall frequencies of attacks were asfollows: one episode of FS in 44 patients, two episodesin 30 patients, three episodes in 32 patients, four epi-sodes in 27 patients, five episodes in 17 patients, andmore than five episodes in 54 patients. Imaging studieswere performed in 73/204 (35.8%), of whom, brain CTand brain MRI were performed in seven and 66 patients,respectively. Twelve of the 73 patients showed nonspe-cific hyperintensity in the white matter, and one patientshowed a small arachnoid cyst.The incidence of complex features was as follows:29 patients had seizures lasting >15 min, 35 patientshad focal seizures, and 59 patients had  P 2 seizureswithin 24 h. 80 patients showed one complex feature,16 patients showed two complex features, and onlyone patient showed three complex features in a singleFS episode. Sixty-one percent of 97 patients exhibitedcomplex features as the first FS. Of the whole sample,information about socioeconomic status was availablefor 161 of the patients with FS. Low economic statuswas observed more often in patients with complex FScompared with those with simple FS ( P =  0.002). Paren-tal educational level, maternal age at birth, cesarean sec-tion delivery, preterm delivery (<37 weeks), and lowbody weight (<2500 g) at birth did not differ signifi-cantly between the groups with simple FS and complexFS.Table 2 shows the clinical and demographic charac-teristics of FS in patients grouped according to whetherpatients had or had not experienced unprovoked seizure.The average interval between the first FS and the firstunprovoked seizure was 34.1 ± 28.6 months. Variablesthat differed significantly were family history of epilepsy,FS plus, late-onset FS at >3 years of age, high frequencyof FS (>5 attacks), complex FS as their first FS, abnor-mal EEG findings, FS developed at a body temperature Table 1Clinical and demographic characteristics according to type of febrileseizure.Simple FS ComplexFS P  value n  = 107  n  = 97Sex (male/female),  n  77/30 59/38 0.092Mean age (months) 69.4 ± 31.0 67.6 ± 30.6 0.942Mean age at first FS (months) 21.2 ± 12.8 16.9 ± 8.9 0.006 * Mean frequency of FS,  n  3.2 ± 2.3 5.4 ± 3.8 0.001 * FS at <1 year of age,  n  (%) 31 (29.0) 42 (43.3) 0.033 * FS plus,  n  (%) 17 (15.9) 25 (25.8) 0.081Family history of FS,  n  (%) 27 (25.2) 33 (34.0) 0.169Family history of epilepsy,  n  (%) 8 (7.5) 8 (8.2) 0.838Abnormal EEG findings,  n  (%) 40 (37.4) 33 (34.0) 0.617Unprovoked seizure,  n  (%) 21 (19.6) 26 (26.8) 0.224FS within 24 h after fever,  n  (%) 74 (69.2) 78 (80.4) 0.065FS at < 39   C,  n  (%) 56 (52.3) 71 (73.2) 0.002 * Low SES,  n  = 161 (%) 8 (9.9) 23 (28.8) 0.002 * Low educational status (father), n  = 159 (%)18 (22.5) 24 (30.4) 0.260Low educational status (mother), n  = 159 (%)23 (28.8) 25 (31.6) 0.691 * Significant at  P   < 0.05.Table 2Clinical and demographic characteristics of patients with FS accordingto whether the patients exhibited or did not exhibit unprovokedseizure.WithoutunprovokedseizureWith firstunprovokedseizure P  value n  = 157  n  = 47Sex (male/female),  n  108/49 28/19 0.240Mean age (months) 61.0 ± 26.0 93.6 ± 32.4 0.001 * Mean age at first FS (months) 18.8 ± 10.7 20.4 ± 13.2 0.437Mean frequency of FS,  n  3.8 ± 2.5 5.6 ± 4.8 0.017 * FS at <1 year of age,  n  (%) 54 (34.4) 19 (40.4) 0.449FS at >3 years of age,  n  (%) 18 (11.5) 11 (23.4) 0.040 * FS plus,  n  (%) 26 (16.6) 16 (34.0) 0.009 * Family history of FS,  n  (%) 49 (31.2) 11 (23.4) 0.303Family history of epilepsy,  n  (%) 8 (5.1) 8 (17.0) 0.013 * Complex FS,  n  (%) 71 (45.2) 26 (55.3) 0.224Complex FS as first FS,  n  (%) 38 (24.2) 21 (44.7) 0.007 * Abnormal EEG findings,  n  (%) 41 (26.1) 32 (68.1) 0.001 * FS within 24 h after fever,  n  (%) 115 (73.2) 37 (78.7) 0.450FS at <39   C,  n  (%) 91 (58.0) 36 (76.6) 0.021 * Low SES,  n  = 161 (%) 23 (19.0) 8 (20.0) 0.890Low educational status (father), n  = 159 (%)25 (21.0) 17 (42.5) 0.008 * Low educational status (mother), n  = 159 (%)30 (25.2) 18 (45.0) 0.018 ** Significant at  P   < 0.05. G. Hwang et al./Brain & Development xxx (2014) xxx–xxx  3 Please cite this article in press as: Hwang G et al. Predictors of unprovoked seizure after febrile seizure: Short-term outcomes. Brain Dev (2014),http://dx.doi.org/10.1016/j.braindev.2014.06.003  of <39   C, and low educational level of both parents.Although low educational level of both parents was sig-nificant in the group of 159 patients whose parents pro-vided this information, this variable was not included inthe multivariate model because this number was notconsidered representative of the entire sample. In thisgroup, only one father had graduated from middleschool, 41 fathers had graduated from high school,and 117 fathers had graduated from college. In thisgroup, 48 mothers had graduated from high school,and 111 mothers had graduated from college.Low family economic status, maternal age at birth,cesarean section delivery, preterm delivery, and lowbody weight (<2500 g at birth) did not differ signifi-cantly between the groups with and without unprovokedseizure.Among the 47 patients with unprovoked seizure,15 patients had only one unprovoked seizure before thecurrent analysis started, and 32 patients had two ormore, thereby meeting the definition of epilepsy. Themean age at the first FS was 22.8 months in 21 patientswith partial seizure and 18.5 months in 26 patients withgeneralized seizure. Thirty-four percent had the firstunprovoked seizure within 1 year of the first FS, 59.6%had their first unprovoked seizure within 3 years afterthe first FS, and 19% and 21% had the first unprovokedseizure 3–5 years and 6–8 years after the first FS, respec-tively. 36.2% of 47 patients developed unprovokedseizure at <3 years of age, and 63.8% (30/47) showedunprovoked seizure at  P 3 years of age. Of the32 patients who developed epilepsy after FS, the classifi-cation of epilepsy syndrome was as follows: 13 patientshad generalized tonic–clonic seizures, six had frontallobe epilepsy, four had benign rolandic epilepsy, fourhad childhood absence epilepsy, two had temporal lobeepilepsy, two had Panayiotopoulos syndrome, and onehad occipital lobe epilepsy.We used univariate analysis to identify the signifi-cance of three individual complex features that wereclassified as complex FS as any FS, and as complexFS as the first FS (Table 3). Among the three individualfeatures of complex FS, only repeated occurrence of FSwithin 24 h correlated significantly with unprovoked sei-zure after FS.Table 4 shows the results of the multivariate analysis.The significant predictors of unprovoked seizure afterFS were (in decreasing order of odds ratio): abnormalEEG findings, family history of epilepsy, late-onset FSat >3 years of age, FS developing at a body temperatureof <39   C, and complex FS as the first FS. 4. Discussion The findings from our retrospective analysis in thissubstantially homogeneous population on Jeju Islandare similar to those of previous investigations in different Table 3Univariate analysis of the risk of developing unprovoked seizure between simple FS and individual complex features, classified as complex FS as anyFS and complex FS as the first FS.Simple FS ( n  = 107) vs complex FS as any FS ( n =  97) Simple FS ( n  = 107) vs complex FS as the first FS ( n  = 59)Overlappingfeatures * ( n )Seizure ** ( n ) P   Singlefeature ( n )Seizure( n ) P   Overlappingfeatures ( n )Seizure( n ) P   Singlefeature ( n )Seizure( n ) P  Focal (35) 10 0.266 Focal (16) 4 0.739 Focal (21) 8 0.086 Focal (13) 5 0.152Prolonged (29) 6 0.899 Prolonged(16)3 1 Prolonged (18) 6 0.219 Prolonged(11)3 0.693Multiple  (59) 17 0.177 Multiple(40)13 0.099 Multiple (33) 12 0.048 Multiple(25)10 0.030 * Defined as a single complex feature or a single complex feature plus other complex features in a single episode of FS.  Defined as repeated occurrence of FS within 24 h. ** Defined as an unprovoked seizure developed after FS.Table 4Independent risk factors for unprovoked seizure after FS in all patients and in subgroups with simples FS and complex FS.Overall group ( n =  204) SubgroupSimple FS ( n =  107) Complex FS ( n =  97) P   Adjusted OR 95% CI  P   Adjusted OR 95% CI  P   Adjusted OR 95% CIHigh frequency of FS(>5 attacks) 0.086 2.21 0.9–5.5 NS 0.005 6.4 1.7–23.9Complex FS as the first FS 0.045 2.37 1.1–5.5 NS 0.013 7.6 1.5–37.2Family history of epilepsy 0.019 5.0 1.3–19.0 NS 0.074 11.8 0.8–175.2Abnormal EEG findings 0.001 5.93 2.7–13.0 0.003 5.9 1.8–18.7 0.010 5.1 1.5–17.6FS at > 3 years of age 0.010 4.99 1.4–11.4 0.009 5.7 1.6–20.8 NSFS at < 39   C 0.035 2.57 1.1–6.2 0.041 3.6 1.1–12.5 NSOR, odds ratio; CI, confidence interval, NS, not significant.4  G. Hwang et al./Brain & Development xxx (2014) xxx–xxx Please cite this article in press as: Hwang G et al. Predictors of unprovoked seizure after febrile seizure: Short-term outcomes. Brain Dev (2014),http://dx.doi.org/10.1016/j.braindev.2014.06.003
Search
Tags
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks
SAVE OUR EARTH

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!

x