Seizure 55 (2018) 36–47

Contents lists available at ScienceDirect

Seizure
journal homepage: www.elsevier.com/locate/yseiz

Investigating the prevalence of febrile convulsion in Kayseri, Turkey:

An assessment of the risk factors for recurrence of febrile convulsion
and for development of epilepsy
Mehmet Canpolata,* , Huseyin Perb , Hakan Gumusc , Ferhan Elmalid, Sefer Kumandase,*
a
Erciyes University Medical School, Department of Pediatrics, Division of Pediatric Neurology, Talas, Kayseri, Turkey
b
Erciyes University Medical School, Department of Pediatrics, Division of Pediatric Neurology, Talas, Kayseri, Turkey
c
Erciyes University Medical School, Department of Pediatrics, Division of Pediatric Neurology, Talas, Kayseri, Turkey
d
Erciyes University Medical School, Department of Biostatistics, Talas, Kayseri, Turkey
e
Erciyes University Medical School, Department of Pediatrics, Division of Pediatric Neurology, Talas, Kayseri, Turkey

A R T I C L E I N F O A B S T R A C T

Article history: Purpose: The purpose of this study was to investigate the prevalence and recurrence of febrile convulsion
Received 25 July 2016 (FC) and risk factors for development of epilepsy in school children throughout in the Kayseri provincial
Received in revised form 19 December 2017 center.
Accepted 8 January 2018
Method: Ten thousand individuals selected using “stratified cluster sampling” from a student population
Available online xxx
of 259,428 inside the Kayseri Urban Municipality represented the study sample. Fifteen thousand
questionnaires were distributed, of which 10,742 (71.6%) were returned. Telephone interviews were
Keywords:
performed with the families of the students reported as having undergone FC, and the medical records of
Child
Epilepsy
patients with a history of hospitalization were evaluated. Data were analyzed on IBM SPSS Statistics 22.0
Febrile convulsions package program. Significance was set at p < 0.05.
Prevalence Results: Prevalence of FC was 4.2% in girls and 4.3% in boys, with a total prevalence of 4.3%. Recurrence of
Risk factors FC was observed in 25.4% of cases. Risk of recurrence increased 7.1 times in subjects with a history of FC in
Kayseri first and second degree relatives, 17.8 times in those with fever interval <1 h before convulsion and 17.6
times in those with pre-convulsion body temperature <39  C. Epilepsy developed in 33 (7.2%) cases.
Neurodevelopmental abnormality was the most important risk factor for epilepsy (21.1-fold risk
increase).
Conclusions: Analysis revealed that FC with a good prognosis had a high rate of recurrence and a higher
risk of epilepsy than in the general population. The prevalence of FC in the province of Kayseri was closer
to that in developed rather than developing countries.
© 2018 Published by Elsevier Ltd on behalf of British Epilepsy Association.

1. Introduction FC is the most common childhood seizure disorder [3].

Febrile convulsion (FC) is described by the International League
Against Epilepsy (ILAE) as convulsion occurring during febrile
disease in children aged over one month with no previous afebrile
convulsion, with no infection involving the central nervous system
(CNS), and with no identified cause such as electrolyte imbalance,
metabolic disorder, intoxication or trauma [1–3]. Onset after the
age of 7 years is rare [1].
* Corresponding authors.
E-mail addresses: mcanpolat@erciyes.edu.tr, drmehmetcanpolat@gmail.com
(M. Canpolat), hper@erciyes.edu.tr (H. Per), hakgumus@erciyes.edu.tr (H. Gumus),
elmali@erciyes.edu.tr (F. Elmali), skumandas@erciyes.edu.tr,
skumandas@hotmail.com (S. Kumandas).
Incidence and prevalence of FC is thought to vary depending on
geographic, socioeconomic variations, and genetic disposition.
Previous studies have reported prevalence of FC up to the age of
7 between 2% and 8% [3–8]. The prevalence in the USA and Europe
is reported at 2–5%, at 6–9% in Japan and 14% in Guam [9–12]. The
prevalence in Turkey is uncertain, although prevalence of FC have
been reported in primary school children at Diyarbakır and final
year students at Istanbul University 8,9%, 5,5% respectively. Aydın
et al. [15] reported a lifetime prevalence of FC of 9,7% in a study
from the province of Izmir
FC recurs in approximately 25–50% of children with FC [1–3].
Recurrence in cases with onset below the age of 1 year is
approximately 50%, and approximately 10% in cases with onset
after the age of 3 [3]. The main factors increasing the risk of
recurrence of FC are seizure onset below the age of 1 year, family

https://doi.org/10.1016/j.seizure.2018.01.007
1059-1311/© 2018 Published by Elsevier Ltd on behalf of British Epilepsy Association.
 M. Canpolat et al. / Seizure 55 (2018) 36–47
history, low fever during FC and short-lived fever before seizure
[3,16,17].
Progression to epilepsy in FC is a controversial subject. The risk
of progression to epilepsy in single FC is regarded as no different to
that in the general population (approximately 0,5%), although a
reported rise to 2–10% as risk factors increase [3]. Risk factors for
epilepsy in the literature include a family history of epilepsy,
neurodevelopmental abnormality, complex FC and short fever
interval before seizure [1,3,18,19].
This study was planned as the second stage of our work
assessing the prevalence of epilepsy in school children aged 7–17
in the Kayseri provincial center in the 2010–2011 academic year
[20], and it was performed in districts in the Kayseri provincial
center in order to determine the prevalence of FC in 7–17 years old
children who had experienced FC when they were 1 month-old to 6
year-old. The results were analyzed for assessing the risk factors for
recurrence of FC, and development of epilepsy in patients
diagnosed with FC, and monitoring of FC.
2. Materials and methods
This study was approval by the Local Ethics Committee of
Erciyes University (2015/404).
2.1. Determination of sample size
The population of this cross-sectional study consisted of
259,428 students attending a total of 725 primary and high school
or equivalent institutions inside the borders of the Kayseri
Metropolitan Municipality, 184,281 (71.03%) receiving primary
education and 75,147 (28.96%) secondary education (high schools)
[21]. The stratified cluster sampling method was used.
2.2. Selection of individuals
In the light of the approximately 75–80% return rates in
previous prevalence studies, 15,000 questionnaires were distrib-
uted. Primary (71.03% of the population, sample size 10,655
students) and high (28.96% of the population, sample size 4345
students) schools were grouped separately to determine their
weights within the population, the proportions of which were
represented within the population. Schools were stratified
according to socioeconomic level, 20% high, 60% medium and
20% low, from Kayseri Provincial Education Directorate Statistical
Office data proportional to national socioeconomic group classi-
fications and socioeconomic groups in the population. Each school
within these strata was regarded as a cluster. Seven schools
(clusters) of the 145 schools with a low socioeconomic level, 21 of
the 435 schools with a moderate socioeconomic level and 7 of the
145 schools with a high socioeconomic level were determined
using simple random sampling. Schools in each set were numbered
on the basis of alphabetic order, and schools were selected on a
random basis with the help of a computer program. A study
population of 15,000 students was established using random
sampling in such a way as to include 3000 high socioeconomic
level students, 9000 medium socioeconomic level students and
3000 low socioeconomic level students, equally divided between
males and females, from classes in the 35 identified schools.
2.3. Study protocol
Following receipt of approval from the Provincial Education
Directorate, a questionnaire consisting of 40 items prepared in line
with prevalence study criteria for developing countries
37
recommended by the World Health Organization (WHO) and
the Guidelines for Epidemiologic Studies recommended by
International League Against Epilepsy (ILAE) – Commission of
Epidemiology and Prognosis (CEP)-1993 was distributed to schools
for the purpose of determining the prevalence of febrile convul-
sion. “Having experienced FC” represented the dependent variable
of the research. Age, sex, social security, consanguineous marriage,
residence, parental education level, history of FC in the family and
family history of epilepsy were independent variables of the
research into FC. The first 22 questions concerned the subjects’
sociodemographic characteristics. The remaining 18 questions
were drawn up in line with the Epidemiological Studies Guidelines
recommended by ILAE–CEP–1993. Cases were reassessed at the
2nd and 3rd states of the study using a separate patient assessment
form consisting of 35 questions in order to elicit detailed
information in patients who had undergone FC. Students described
as having undergone FC according to the results of the 2nd stage
questionnaire were reassessed on the basis of school counseling
service records or telephone interviews with families. In order to
confirm diagnosis of FC in the telephone questionnaire, we
inquired whether the case had been diagnosed with FC by a
health institution. Students monitored at any health institution
with a diagnosis of FC and agreeing to interview were invited for
check-up at our clinic in the 3rd stage. Seizure semiology, physical
examination, drugs used for treatment, recurrence of FC and risk
factors for epilepsy were evaluated. Available imaging records and/
or reports for first electroencephalography (EEG) performed after
FC attack and EEGs performed after 6, 12 or 24 months were
reassessed, and the results were recorded as normal or pathologi-
cal. Findings for patients who had undergone cranial imaging
[computed tomography (CT) and/or magnetic resonance imaging
(MRI)] and whose imaging records were available were reassessed.
Results were recorded as normal or pathological, and no additional
cranial imaging and EEG were performed.
2.4. Statistical analysis
Data were analyzed on the IBM SPSS Statistics 22.0 package
program (IBM Corp., Armonk, New York, USA). Descriptive
statistics were expressed as number (n), percentage (%) and
mean  standard deviation (x  sd). Normal distribution of nu-
merical variables were investigated using the Shapiro-Wilk test
and Q–Q plots. The two group independent samples t test was used
in two-group comparisons. Categoric variables were analyzed
using the chi square test. Binary logistic regression analysis with
Wald backward elimination was used to determine risk factors.
Significance was set at p < 0.05. Levels of significance were derived
as for two-tailed tests.
3. Results
The study group consisted of 10,742 children, with a return rate
of 71.6%.
3.1. General demographic findings
A total of 10,742 individuals, with a mean age of 12.013.16
years, 5312 (49.5%) male and 5430 (50.5%) female responded to the
questions in the questionnaire (p = 0.112). Completed question-
naires were received for analysis from 7790 (72.5%) primary school
students and 2952 (27.5%) from high school students. Examination
of the economic status of the cases comprising the sample revealed
that 20.8% had a good level of income, 58.3% a moderate level and
20.9% a low level.
 38
Table 1
Sociodemographic characteristics of students with febrile convulsion, risk factors for first febrile convulsion, risk factors for recurrence of febrile convulsion-1 and risk factors for development of epilepsy in febrile convulsion-1.

Question Answer Risk factors for first febrile convulsion Risk factors for recurrence in febrile convulsion 1 Risk factors for development of epilepsy in febrile convulsion
1

With febrile convulsion Without febrile p Recurrence of febrile Recurrence of febrile p Development of epilepsy Development of epilepsy p
convulsion convulsion Yes convulsion No after febrile convulsion after febrile convulsion
Yes No

n Percentage n Percentage n Percentage n Percentage n Percentage n Percentage

(%) (%) (%) (%) (%) (%)
Sex Male 230 50.3 5082 49.4 0.702 62 53.4 168 49.3 0.437 20 60.6 210 49.5 0.220
Female 227 49.7 5203 50.6 54 46.6 173 50.7 13 39.4 214 50.5

Type of delivery Cesarean 75 16.47 2187 21.3 0.013 18 15.5 57 16.7 0.763 6 18.2 69 16.3 0.776
Normal 382 83.6 8098 78.7 98 84.5 284 83.3 27 81.8 355 83.7
vaginal
delivery

Time of birth Premature 37 8.1 587 5.7 0.073 12 10.3 25 7.3 0.105 6 18.2 31 7.3 0.022
Term 410 89.7 9521 92.6 99 85.3 311 91.2 25 75.8 385 90.8
Late 10 2.2 177 1.7 5 4.3 5 1.5 2 6.1 8 1.9

M. Canpolat et al. / Seizure 55 (2018) 36–47

Birth Weight < 2500 gr 67 14.7 1205 11.7 0.118 20 17.2 47 13.8 0.467 5 15.2 62 14.6 0.853
2500– 289 63.2 6891 67.0 68 58.6 221 64.8 22 66.7 267 63.0
3500 gr
>3500 gr 101 21.3 2189 21.3 28 24.1 73 21.4 6 18.2 95 22.4

Consanguineous Yes 87 19.0 1562 15.2 0.027 22 19.0 65 19.1 0.982 11 33.3 76 17.9 0.030
marriage
No 370 81.0 8723 84.8 94 81.0 276 80.9 22 66.7 348 82.1

Degree of 3rd degree 65 74.7 1185 75.9 0.667 10 45.5 55 84.6 0.002 8 72.7 57 75.0 0.672
consanguinity
between mother
and father
4th degree 13 14.9 268 17.2 7 31.8 6 9.2 1 9.1 12 15.8
Other 9 10.3 109 6,9 5 22.7 4 6.2 2 18.2 7 9.2

Mother’s age 38.38  6.65 37.70  6.01 0.032 36.91  6.36 38.88  6.68 0.006 37.45  7.00 38.46  6.62 0.406
(x  sd) age age

Mother’s Illiterate 20 4.4 316 3.1 0.001 3 2.6 17 5.0 0.184 2 6.1 18 4.2 0.823
education level
Literate 16 3.5 245 2.4 4 3.4 12 3.5 2 6.1 14 3.3
Primary 263 57.5 5266 51.2 69 59.5 194 56.9 20 60.6 243 57.3
school
Middle 56 12.3 1299 12.6 12 10.3 44 12.9 2 6.1 54 12.7
schools
High school 78 17.1 2136 20.8 17 14.7 61 17.9 5 15.2 73 17.2
Higher 24 5.3 1023 9.9 11 9.5 13 3.8 2 6.1 22 5.2
education

Mother’s Unemployed 426 93.2 9326 90.7 0.262 108 93.1 318 93.3 0.718 30 90.9 396 93.4 0.571
occupation
Teacher 14 3.1 356 3.5 5 4.3 9 2.6 1 3.0 13 3.1
Health 5 1.1 174 1.7 1 0.9 4 1.2 0 0.0 5 1.2
worker
Other 12 2.6 429 4.2 2 1.7 10 2.9 2 6.1 10 2.4
Father’s age 42.07  6.42 41.75  6.34 0.294 40.41  6.27 42.64  6.38 0.001 40.21  6.95 42.22  6.36 0.084
(x  sd) age age

Father’s education Illiterate 6 1.3 79 0.8 0.004 3 2.6 3 0.9 0.262 2 6.1 4 0.9 0.210
level
Literate 10 2.2 107 1.0 4 3.4 6 1.8 1 3.0 9 2.1
Primary 185 40.5 3501 34.0 42 36.2 143 41.9 11 33.3 174 41.0
school
Middle 77 16.8 1818 17.7 15 12.9 62 18.2 5 15.2 72 17.0
schools
High school 102 22.3 2691 26.2 29 25.0 73 21.4 7 21.2 95 22.4
Higher 77 16.8 2089 20.3 23 19.8 54 15.8 7 21.2 70 16.5
education

Father’s Unemployed 6 1.3 76 0.7 0.540 1 0.9 5 1.5 0.897 2 6.1 4 0.9 0.039
occupation

M. Canpolat et al. / Seizure 55 (2018) 36–47

Teacher 21 4.6 462 4.5 6 5.2 15 4.4 3 9.1 18 4.2
Health 6 1.3 111 1.1 1 0.9 5 1.5 0 0.0 6 1.4
workers
Other 424 92.8 9636 93.7 108 93.1 316 92.7 28 84.8 396 93.7

Family’s income Good 73 16.0 2160 21.0 0.001 17 14.7 56 16.4 0.136 4 12.1 69 16.3 0.662
level
Medium 259 56.7 6005 58.4 59 50.9 200 58.7 18 54.5 241 56.8
Low 125 27.4 2120 20.6 40 34.5 85 24.9 11 33.3 114 26.9

Residence Own home 286 62.6 6487 63.1 0.169 66 56.9 220 64.5 0.066 19 57.6 267 63.0 0.086
Lodging 7 1.5 304 3.0 0 0.0 7 2.1 2 6.1 5 1.2
Rented 164 35.9 3494 34.0 50 43.1 114 33.4 12 36.4 152 35.8

Number of people 3 or less 40 8.8 1014 9.9 0.206 8 6.9 32 9.4 0.763 2 6.1 38 9.0 0.887
in household
4 183 40.0 4077 39.6 47 40.5 136 39.9 15 45.5 168 39.6
5–6 201 44.0 4670 45.4 54 46.6 147 43.1 14 42.4 187 44.1
7 or more 33 7.2 524 5.1 7 6.0 26 7.6 2 6.1 31 7.3

39
40 M. Canpolat et al. / Seizure 55 (2018) 36–47
3.2. Febrile convulsion
Second and third stage analysis on the 868 (8.1%) cases reported
as having experienced FC revealed that 397 (3.7%) cases had
experienced febrile disease only and had no history of FC, that 14
(0.1%) cases had presented due to FC and been diagnosed with
meningitis and that 457 (4.3%) cases had experienced FC. The FC
patient group thus consisted of 457 patients.
3.3. Febrile convulsion prevalence findings
A total of 457 (4.3%) subjects with a mean age of 12.64  3.23
months had experienced FC, 230 (50.3%) males and 227 (49.7%)
females. The male female ratio was 1:1 (p = 0.702). The prevalence
of FC was 4.3% in boys and 4.2% in girls, for a total prevalence of
4.3%.
3.4. Febrile convulsion general demographic findings and risk factors
in first febrile convulsion
The socioeconomic characteristics of the participants are
shown in Table 1. Prevalence of FC was 3.3% in those with a good
level of income and 5.6% in those with low income (p = 0.001).
Examination of parental education levels in subjects with FC
revealed a prevalence of FC of 7.1% in those whose fathers were
illiterate and a prevalence of 3.6% in children of university
graduates. The difference was significant (p = 0.001). Prevalence of
FC was 6.0% in children of illiterate mothers and 2.3% in children of
university graduates (p = 0.001). Prevalence of FC in subjects born
by cesarean delivery was 3.3%, and 4.5% in those born by normal
vaginal delivery (p = 0.013).
FC was most commonly experiences between the ages of 18–24
months; 126 cases (27.6%) were identified in that age group. FC
attacks were observed in 54 cases (11.8%) before the age of 6
months, in 75 cases (16.4%) aged 6–12 months, in 28 cases (6.1%)
aged 12–18 months, in 124 cases (27.1%) aged 24–60 months and in
41 cases (9.0%) aged 6 years or more. Age at FC could not be
remembered in 9 cases (2.0%).
Cause of fever in subjects with FC was upper respiratory tract
infection in 221 cases (48.4%), lower respiratory tract infection in
86 cases (18.8%), urinary infection in 37 cases (8.1%), acute
gastroenteritis in 66 cases (14.4%), acute otitis media in 35 cases
(7.7%), post-vaccination temperature elevation in 10 cases (2.2%)
and FC associated with other nonspecific infections such as viral
exanthematous disease in 2 cases (0.4%). Pre-FC febrile disease was
present in 410 cases (89.7%).
A history of FC was determined in the mothers of 45 cases (9.8%)
with FC, in the fathers of 36 cases (7.9%) and in the siblings of 189
cases (41.4%). A history of FC was present in more than one family
member in 9 cases (1.9%). FC was determined in the families of 261
cases (57.1%) with FC. The most common age at undergoing FC in
the siblings of subjects with FC was 18–24 months, in 61 cases
(32.3%). Body temperature was below 39  C in 113 cases (24.7%) in
the seizure period and above 39  C in 344 cases (75.3%). A history of
short fever lasting less than 1 h before seizure was determined in
115 cases (25.2%) and of prolonged fever lasting more than 1 h in
342 cases (74.8%). A family history of epilepsy was present in 27
cases (5.9%), a history of attending a crèche in the period when FC
was experience in 18 (3.9%), neurodevelopmental abnormality in
24 (5.3%), hospitalization exceeding 30 days in the neonatal unit in
17 (3.7%) and a low APGAR score (<7) in 114 (24.9%). Thirty-four
cases (7.4%) with FC had received mother’s milk for a short time,
less than 6 months, 333 cases (72.9%) for 6–12 months and 66 cases
(14.4%) for 12–24 months, while 24 cases (5.3%) no mother’s milk.
Simple FC had been experienced in 373 cases (81.6%) and
complex FC in 84 (18.4%). Forty-seven cases (10.3%) experienced
focal convulsion and 410 (89.7%) experienced generalized convul-
sion. Two hundred forty-three cases (53.2%) experienced tonic
convulsion, 71 (15.5%) clonic convulsion, 55 (12.0%) tonic-clonic
and 88 (19.3%) atonic convulsion. Mean length of febrile episode
was 9.23  8.91 min (min–max: 1–60 min). Duration of episode
was less than 15 min in 399 cases (87.3%) and longer in 58 (12.7%).
EEG was performed after FC in 103 cases (22.5%), comprising 40
cases (10.7%) presenting to hospital with simple FC and 63 (75.0%)
presenting with complex FC. A total of 67 (65.0%) first EEGs were
pathological, 21 (52.5%) in cases of simple FC and 46 (73.0%) in
cases of complex FC. Final EEG was normal in 101 (98.1%) of these
cases, while final EEG was pathological in 2 cases (1.9%) of complex
FC.
Cranial imaging was performed in 105 cases (23.0%), 43 (11.5%)
of simple FC and 62 (73.8%) of complex FC. One hundred four
(99.9%) of these cases were reported to be normal. Lumbar
puncture (LP) was performed in 31 cases (6.8%), and meningitis
was diagnosed in none of these cases. Of the 31 cases in which LP
was performed, 14 (45.1%) were aged under 12 months at time of
LP, 9 (29.1%) were aged 12–24 months and 8 (25.8%) were aged over
24 months.
One hundred twelve cases (24.5%) received antiepileptic
therapy due to FC, 47 (12.6%) with simple FC and 65 (77.4%) with
complex FC, while antiepileptic therapy was not recommended in
345 cases (75.5%). Sixty (53.6%) cases received phenobarbital
therapy, 22 (19.6%) with simple FC and 38 (33.9%) with complex FC.
Thirty-three cases received sodium valproate therapy, 8 (7.1%) with
simple FC and 25 (38.5%) with complex FC, while intermittent
rectal diazepam therapy only was advised in 19 cases (17.0%), 17
(36.2%) with simple FC and 2 (3.1%) with complex FC. Mean length
of treatment was 30.02  20.14 months in simple FC and
34.20  18.95 months in complex FC. We learned that no
significant drug side-effects had been observed.
3.5. Febrile convulsion recurrence
Recurrence of FC was observed in 116 cases (25.4%), 62 boys
(53.4%) and 54 girls (46.6%). Of the cases in which FC recurred,
recurrence was observed once in 54 cases (46.5%), twice in 27 cases
(23.3%), 3 times in 18 cases (15.5%) and 4 or more times in 17
(14.7%). Recurrence after first FC was observed in less than 1 month
in 10 cases (8.6%), in 1–3 months in 40 cases (34.5%), in 3–6 months
in 53 cases (45.7%), in 6–12 months in 7 cases (6.0%), in 12–24
months in 3 cases (2.6%) and after 24 months in 3 cases (2.6%).
Recurrence of FC took place in the first 6 months in 88% of cases.
Age at final FC was below 6 months in 2 cases (1.7%), 6–12 months
in 11 (9.5%), 12–18 months in 23 (19.8%), 18–24 months in 16
(13.8%), 24–60 in 51 (44.0%) and 72 months of more in 13 (11.2%).
Mean age at final convulsion was 27.95  19.22 months (min–max:
2–108 months). The risk factors associated with FC recurrence are
presented in Tables 1 and 2.
Logistic regression analysis was performed in order to examine
the effects of more than one variable together. A final model
containing 3 variables (history of FC in the family, duration of fever
pre-convulsion and pre-convulsion temperature) was obtained as
result of the analysis. Accordingly, subjects with a history of FC in
first and second-degree relatives had a 7.1-fold greater risk of
recurrence, those with pre-convulsion duration of fever <1 h a
17.8-fold greater risk of recurrence and those with pre-convulsion
fever <39  C a 17.6-fold greater risk of recurrence (Table 3).
3.6. Risk factors affecting development of epilepsy in febrile convulsion
Epilepsy was determined in 33 (7.2%) of the patients presenting
with FC (457 subjects), 20 (60.6%) male and 13 (39.4%) female. Total
prevalence of epilepsy after FC was 7.2%, 5.7% in girls and 8.7% in
 M. Canpolat et al. / Seizure 55 (2018) 36–47 41

Table 2
Risk factors for recurrence of febrile convulsion-2.

Recurrence of Febrile Convulsion Yes Recurrence of Febrile Convulsion No p

n Percentage (%) n Percentage (%)

Age at first FC 6 months or less 15 12.9 39 11.4 0.216
6–12 months 20 17.2 55 16.1
12–18 months 10 8.6 18 5.3
18–24 months 36 31.0 90 26.4
24–60 months 29 25.0 95 27.9
72 or more 6 5.2 35 10.3
Do not remember 0 0.0 9 2.6

FC in family Yes 45 38.8 93 27.3 0.026

No 71 61.2 248 72.7

Body temperature pre-FC <39  C 77 66.4 36 10.6 <0.001

>39  C 39 33.6 305 89.4

Duration of fever pre-FC < 1h 67 57.8 48 14.1 <0.001

> 1h 49 42.2 293 85.9

Epilepsy in family Yes 8 6.9 19 5.6 0.601

No 108 93.1 322 94.4

Type of FC Simple 70 60.3 303 88.9 <0.001

Complicated 46 39.7 38 11.1

Characteristic of seizure Focal 26 22.4 21 6.2 <0.001

Generalized 90 77.6 320 93.8

Form of seizure Tonic 72 62.1 171 50.1 0.054

Clonic 15 12.9 56 16.4
Tonic – clonic 7 6.0 48 14.1
Atonic-Hypomotor 22 19.0 66 19.4

Recurrence within the same day Yes 14 12.1 11 3.2 <0.001

No 102 87.9 330 96.8

Length of seizure <15 min 86 74.1 313 91.8 <0.001

>15 min 30 25.9 28 8.2

More than one finding of complex FC Yes 11 9.5 15 4.4 0.041

No 105 90.5 326 95.6

Sex Female 62 53.4 168 49.3 0.437

Male 54 46.6 173 50.7

Neurodevelopmental abnormality Yes 12 10.3 12 3.5 0.004

No 104 89.7 329 96.5

Use of mother’s milk 0–6 months 11 9.5 23 6.7 0.345

6–12 months 87 75.0 246 72.1
12–24 months 15 12.9 51 15.0
None 3 2.6 21 6.2

Low APGAR score or history of resuscitation Yes 34 29.3 80 23.5 0.208

No 82 70.7 261 76.5

Long-term treatment Yes 61 52.6 51 15.0 <0.001

No 55 47.4 290 85.0

Hospitalization Yes 100 86.2 241 70.7 0.001

No 16 13.8 100 29.3

Length of hospitalization <1 day 49 42.2 99 29.0 0.012

1–7 50 43.1 134 39.3
7–14 1 0.9 4 1.2
15–30 0 0.0 3 0.9
>30 days 0 0.0 1 0.3
None 16 13.8 100 29.3

Hospitalization in the neonatal ICU exceeding 30 days Yes 14 12.2 3 0.9 <0.001
No 101 87.8 338 99.1

EEG performed post-FC Yes 64 55.2 39 11.4 <0.001

No 52 44.8 302 88.6

First post-FC EEG Normal 19 29.7 17 43.6 0.151

42 M. Canpolat et al. / Seizure 55 (2018) 36–47
Table 2 (Continued)
Recurrence of Febrile Convulsion Yes
n Percentage (%)
Pathological
Final post-FC Normal
Pathological
EEG: Electroencephalography, FC: febrile convulsion, ICU: Intensive care unit, APGAR: Appearance, Pulse, Grimace, Activity, Respiration.
Table 3
Logistic regression analysis results for risk factors for recurrence of febrile convulsion (FC).
Logistic regression model
History of FC in the family Other
With history of FC in first and second degree relatives
Pre-FC body temperature >39  C
<39  C
Duration of fever pre-FC >1 h
<1 h
boys. There was no significant difference between boys and girls in
terms of epilepsy after FC (p = 0.220). Risk factors affecting
development of epilepsy in febrile convulsion are presented in
Tables 1 and 4.
Logistic regression analysis was performed in order to examine
the effect of more than one variable. Variables identified as being
associated with epilepsy at single-variable analysis were included
in this model. Following analysis, a final model containing two
variables (first EEG after FC and neurodevelopmental abnormality)
was obtained. According to this model, the risk of developing
epilepsy was 4.5 times higher in subjects with pathological EEG
after FC and 21.1 times higher in those with neurodevelopmental
abnormality (Table 5).
3.7. Prognosis in febrile convulsion and types of epilepsy
Four hundred twenty-three (92.6%) of the 457 cases diagnosed
with FC resolved in a healthy condition. Sequelae developed
involving epilepsy in 33 cases (7.2%) and somnambulism in one
case. Types of epilepsy occurring in FC based on the ILAE 1989
classification are shown in Table 6. Of the 19 cases developing
temporal, frontal, parietal and occipital lobe epilepsy, temporal
epilepsy occurred in 15 (78.9%), frontal in 2 (10.5%), parietal in 1
(5.3%) and occipital lobe epilepsy in 1 (5.3%). Nine (27.3%) of the 10
patients in the “Other generalized idiopathic epilepsies” group
were identified as being in the “generalized epilepsy with febrile
seizures plus (GEFS+)” group according to the ILAE 2001
classification of epileptic seizures.
4. Discussion
Febrile convulsions are the most common cause of convulsion
in the pediatric age group [22–26]. They are reported in 2–5% of
children [7,24–26]. FCs are generally regarded as benign, although
they also involve risks of recurrence and leading to afebrile
convulsions, and are therefore of considerable importance [3,24–
26]. Publications have reported a higher prevalence of FC in
developing countries than in developed ones. The prevalence of FC
in the pediatric population has been reported as 2–5% in Europe
and America, 6–9% in Japan, 0.5–1.5% in China, 4.1% in Switzerland,
2.9% in Greece and 5–10% in India, while an incidence as high as
14% has been reported in a study from Guam in the Pacific islands
Recurrence of Febrile Convulsion No p
n Percentage (%)
45 70.3 22 56.4
64 100.0 37 94.9 0.067
0 0.0 2 5.1
Logistic regression model %95 CI for odds
(Odds Ratio)
1 –
7.1 1.6–31.3
1 –
17.6 3.6–84.9
1 –
17.8 3.6–86.6
[9–12,27–32]. Baldin et al. [33] reported a prevalence of FC of 1.5%
in a study of school children aged 6–16 in the USA in 2012. The
prevalence of FC in Turkey has been reported as 3.5–9.7% [13–
15,34]. Total prevalence of FC in our study was 4.3%, 4.2% in girls
and 4.3% in boys. Comparing our findings with those from studies
investigating the prevalence in the same age group, they are close
to those from developed countries. The lower prevalence of FC in
developed countries may be associated with the higher socioeco-
nomic levels in those countries, a more aware patient and family
profile, patients presenting to hospital earlier and hospital records
being better and more accurate.
FC is more common in boys than in girls. Özmen et al. [35]
reported a ratio of 1.36/1 in a study from Turkey. However, the
difference was not significant in several studies. Pavlidou et al. [28]
reported a male/female ratio of 1/1. The male/female ratio in our
study was 230/227 (1.01/1), with gender-specific prevalence of
4.2% in girls and 4.3% in boys. Prevalence rates were slightly higher
in boys than girls, but the difference was not significant (p = 0.702).
The prevalence of FC in cases with a good level of income was
3.3%, compared to 5.6% in those with poor income (p = 0.001).
Prevalence of FC in subjects with illiterate fathers was 7.1%, and
3.6% in those with university graduate fathers (p = 0.001).
Prevalence of FC in subjects with illiterate mothers was 6.0%,
and 2.3% in those with university graduate mothers (p = 0.001).
Aydın et al. [15] investigated the relationship between prevalence
of FC and sociodemographic factors in Turkey and reported that FC
was approximately 2.13 times more prevalent in families with a
low socioeconomic and cultural level and emphasized that the
main reasons for this were families lacking information about FC
and having inadequate access to health services. Similarly, a higher
prevalence of FC has been reported in developing countries
compared to developed countries [36–40].
FC is observed in the 1st month at the earliest. It has no upper
age limit. Onset is within the first 2 years in approximately 50% of
children, is most common at months 18–24 and rarely emerges
after the age of 7 years and is generally seen between the ages of 3
months and 5 years [3,41–47]. In this study, in agreement with the
literature, most common age at FC was between 18 and 24 months
in 27.6% of cases. FC began below the age of 2 in 61.9% of cases.
The origin of FCs is associated with viral infections in
approximately 80% of cases; it may be associated with various
causes such as upper respiratory tract infection or pharyngitis,
 M. Canpolat et al. / Seizure 55 (2018) 36–47 43

Table 4
Risk factors affecting development of epilepsy in febrile convulsion-2.

Development of epilepsy Yes Development of epilepsy No p

n Percentage (%) n Percentage (%)

Age at first FC 6 months or less 6 18.2 48 11.3 0.163
6–12 months 2 6.1 73 17.2
12–18 months 1 3.0 27 6.4
18–24 months 7 21.2 119 28.1
24–60 months 11 33.3 113 26.7
72 months or over 6 18.2 35 8.3
Do not remember 0 0.0 9 2.1

FC in the family Yes 19 57.6 242 47.1 0.955

No 14 42.4 182 42.9

Body temperature pre-FC <39  C 16 48.5 97 22.9 0.001

>39  C 17 51.5 327 77.1

Duration of fever pre-FC <1 h 12 36.4 103 24.3 0.124

>1 h 21 63.6 321 75.7

Epilepsy in the family Yes 1 3.0 26 6.1 0.467

No 32 97.0 398 93.9

Type of FC Simple 15 45.5 358 84.4 <0.001

Complicated 18 54.5 66 15.6

Characteristic of seizure Focal 12 36.4 35 8.3 <0.001

Generalized 21 63.6 389 91.7

Form of seizure Tonic 25 75.8 218 51.4 0.059

Clonic 2 6.1 69 16.3
Tonic – clonic 2 6.1 53 12.5
Atonic-Hypomotor 4 12.1 84 19.8

Recurrence of seizure within the same day Yes 4 12.1 21 5.0 0.081
No 29 87.9 403 95.0

Length of seizure <15 min 29 87.9 370 87.3 0.919

>15 min 4 12.1 54 12.7

More than one finding of complex FC Yes 3 9.1 23 5.4 0.381

No 30 90.9 401 94.6

Neurodevelopmental abnormality (Previous neurological finding) Yes 14 42.4 10 2.4 <0.001

No 19 57.6 414 97.6

Use of mother’s milk 0–6 months 3 9.1 31 7.3 0.512

7–12 months 26 78.8 307 72.4
13–24 months 4 6.1 62 14.6
None used 0 0.0 24 5.7

Low APGAR score and history of resuscitation Yes 6 18.2 108 25.5 0.351
No 27 81.8 316 74.5

Use of cigarettes by mother during pregnancy Yes 11 33.3 67 15.8 0.010

No 22 66.7 357 84.2

Long-term antiepileptic treatment Yes 27 81.8 85 20.0 <0.001

No 6 18.2 339 80.0

Hospitalization Yes 32 97.0 309 72.9 0.002

No 1 3.0 115 27.1

Length of hospitalization <1 day 16 48.5 132 31.1 0.049

1–7 16 48.5 168 39.6
7–14 0 0.0 5 1.2
15–30 0 0.0 3 0.7
>30 days 0 0.0 1 0.2
None 1 3.0 115 27.1

EEG performed post-FC Yes 28 84.8 75 17.7 <0.001

No 5 15.2 349 82.3

First EEG post-FC Normal 3 10.7 33 44.0 0.002

Pathological 25 89.3 42 56.0

Final EEG post-FC Normal 27 96.4 74 98.7 0.464

44 M. Canpolat et al. / Seizure 55 (2018) 36–47

Table 4 (Continued)
Development of epilepsy Yes Development of epilepsy No p

n Percentage (%) n Percentage (%)

Pathological 1 3.6 1 1.3

Recurrence of FC Yes 23 69.7 93 21.9 <0.001

No 10 30.3 331 78.1

Number of recurrences of FC 1 6 26.1 45 48.4 0.199

2 7 30.4 20 21.5
3 4 17.4 14 15.1
4 or more 6 26.1 14 15.1

EEG: Electroencephalography, FC: febrile convulsion, APGAR: Appearance, Pulse, Grimace, Activity, Respiration.

Table 5
Logistic regression analysis results for risk factors affecting development of epilepsy in febrile convulsion.

Logistic regression model Logistic regression model %95 CI for odds

(Odds Ratio)

First EEG Normal 1 –

Pathological 4.5 1.2–18.2
Neurodevelopmental abnormality No 1 –
Yes 21.1 4.1–107.0

EEG: Electroencephalography.

Table 6
Types of epilepsy developing in febrile convulsion according to the ILAE 1989 classification.

1989 ILAE classification of epilepsies n %

Localization-related 20 60.6
Generalized seizures 12 36.4
Uncertain whether focal or generalized 1 3.0

Total 33 100

According to the 1989 ILAE classification subtypes of epilepsy n %

Localization-related Childhood epilepsy with occipital paroxysms 1 3.0

Temporal, frontal, parietal, and occipital lobe epilepsy 19 57.6
Generalized Epilepsy with grand mal (GTCS) seizures on awakening 2 6.1
Other generalized idiopathic epilepsies 10 30.3
Uncertain whether focal or generalized Epilepsies and syndromes uncertain whether focal or generalized 1 3.0

Total 33 100

acute otitis media, lower respiratory tract infection, urinary tract
infection, acute gastroenteritis, roseola infantum and non-infec-
tious diseases [24,28,41–45]. Convulsions appearing with 2 weeks
of vaccination exhibit similar characteristics to FC, and frequently
develop in association with high fever after diphtheria-tetanus-
pertussis vaccination and measles vaccination [3]. In this study,
and in agreement with the literature, 48.4% of patients were
identified as experiencing FC in association with upper respiratory
tract infection, 18.8% with lower respiratory tract infection, 8.1%
with urinary tract infection, 14.4% with acute gastroenteritis, 7.7%
with acute otitis media, 2.2% with high fever after vaccination and
0.4% in association with other nonspecific infections, such as viral
exanthematous disease.
FC is simple in 80–85% of cases and complex in 10–15%
[3,43,44]. Seizures in FC are generally brief, generalized, tonic-
clonic, tonic, atonic and rarely partial [3]. In a study of 482 cases,
Sfaihi et al. [41] observed 55.6% tonic-clonic seizures, 16.8% tonic,
12.4% atonic, 9.1% clonic and 6.2% focal, and reported a mean length
of seizure of 7 min (min–max:1–30 min), with 68% of cases lasting
less than 5 min. In our study, and in agreement with the literature,
81.6% of cases underwent simple FC and 18.4% complex FC, with
10.3% focal convulsion and 89.7% generalized convulsion. Tonic
convulsion was determined in 53.2% of cases, clonic in 15.5%, tonic-
clonic in 12.0% and atonic convulsion in 19.3%. Mean duration of
febrile seizure was 9.23  8.91 min (min–max: 1–60 min), and
seizures lasted less than 15 min in 87.3% of cases.
The most important risk factors in FC are temperature and age
of the child. The higher the temperature, the greater the risk of FC
[3]. Four risk factors have been reported to affect first FC. A history
of FC in a first degree relative, hospitalization in the neonatal unit
exceeding 30 days, neurodevelopmental abnormality and crèche
attendance are regarded as the main risk factors. FC develops in
28% of cases if at least two of these risk factors for first FC are
present [3]. One study of 10,224 children with a history of FC in a
sibling reported that number of hospitalizations, gestational age
and low birth weight increase the risk of FC [46]. Studies in recent
years, however, report that the association is not that explicit [24].
In this study, 3.9% of patients attended crèche during the period
when FC was experienced, 5.3% had neurodevelopmental abnor-
mality and 3.7% of patients were hospitalized in the neonatal unit
for more than 30 days. Of the cases presenting with FC, 74.6% were
monitored in hospital and 25.4% were monitored without being
 M. Canpolat et al. / Seizure 55 (2018) 36–47
hospitalized. Premature birth was present in 8.1% of cases with FC
and 5.5% of those without FC, and gestational age had no effect on
FC (p = 0.073).
A history of FC in a first degree relative and elevated
temperature during a febrile illness are known to play an
important role in the development of first FC [3]. At the same
time, genetic factors are also known to be important in FC. A
history of FC in the family has been reported as a risk factor for first
and recurrent FCs, presence of FC in the family is cited at 26–54% in
the literature [3,34,36–39]. In our study, too, FC was present in the
family of 57.1% of cases with FC; a history of FC was present at levels
of 9.8% in mothers, 7.9% in fathers, 41.4% in siblings and 1.9% in
more than one relative. Consanguineous marriage was determined
in 15.2% of cases without FC and 19.0% of those with FC (p = 0.027).
This indicates the importance of genetic factors in the appearance
of FC. Despite the powerful family history in FC, the mode of
transmission is unknown. It is usually multifactorial, with
autosomal dominant inheritance in a smaller group. Gene loci in
chromosomes 6, 8 and 19 have been implicated [3].
Studies have reported that the prevalence of epilepsy in the
families of children with FC is greater than that in the normal
population, ranging between 1.6% and 9% [36]. Sfaihi et al. [41]
reported a history of epilepsy in the family at 5.6%. In this study,
and in agreement with the literature, a family history of epilepsy
was determined in 5.9% of cases.
The accepted body temperature threshold in FC is reported as
38 C in some studies and 38.5  C and above in others. Sfaihi et al.
[41] reported that the level was FC was higher above a mean body
temperature of 39  C, Okumura et al. [48] of 39.4  C and Knudsen
[49] of 39.5  C. FC generally developed within 1–24 h following
elevated body temperature [47]. In this study, in agreement with
the literature, body temperature was below 39  C in 24.7% of cases
at time of seizure and above 39  C in 75.3% of cases. A short pre-
convulsion history of fever lasting less than 1 h was determined in
25.2% of cases and of fever lasting more than 1 h in 74.8% of cases.
The diagnostic value of EEG is limited in children with FC. It is
normal in 60% of simple FCs. However, EEG abnormality is reported
at a level of 2–86% [3]. There are no FC-specific EEG findings. EEG is
not generally recommended in FC. In complex FC, EEG abnormality
may occur in subjects with neurodevelopmental abnormality and a
history of FC in the family, although the clinical significance is
controversial [3,50,51]. In this study, EEG was performed post-
seizure in 22.5% of all cases, 10.7% of those with simple FC and
75.0% of cases of complex FC. Sixty-five percent of first EEGs were
pathological, while 98.1% of final EEGs were normal. This supports
the idea that routine control EEGs are not useful guides in
monitoring and treatment.
CT and MRI are not essential in FC and are not routinely
recommended, although there is broad consensus regarding their
performance in cases with focal neurological deficit [3]. In this
study, too, cranial imaging was performed in 23.0% of cases, of
which 99.% were normal. In agreement with the literature, this
supports the idea that imaging is not essential but may be
considered when indicated, but is not required as a routine
procedure.
Lumbar puncture (LP) in FC is controversial, but it advised in
cases with suspicion of meningitis and patients aged under 12
months. In this study, LP was performed in 6.8% of the 457 cases of
FC, and meningitis was diagnosed in none of these. In addition,
45.1% of those receiving LP were aged under 12 months.
Phenobarbital or sodium valproate are preferred in long-term
prophylactic treatment of FC [3]. In this study, long-term treatment
with phenobarbital or sodium valproate was given to 24.5% of cases
with FC, and no significant drug-related side-effect was deter-
mined.
45
4.1. Recurrence of febrile convulsion
FC recurs in 25–50% of children. Recurrence occurs in the first
year in approximately 75% and in the first 2 years in 90%. The risk of
recurrence increases with the number of risk factors. The risk of
recurrence in subjects with no risk factors is approximately 14.3%,
rising to 50–60% in those with more than 3 risk factors risk
[3,35,51–55]. Single recurrence level is cited as 40–64%, and the
level for 3 or more recurrences is 9–27% [36,41,55]. In this study, in
agreement with the literature, recurrence was observed in 116
cases (25.4%). Recurrence was determined once in 46.5% of cases,
twice in 23.3%, 3 times in 15.5% and 4 or more times in 14.7%.
Recurrence of FC occurred in the first 6 months in 88.8% of cases.
Age less than 18 months (or under 12 months), history of FC in
the family, low fever and a history of fever lasting less than 1 h
before seizure are known certain risk factors for recurrence of FC. A
history of epilepsy in the family is regarded as a probable risk
factor. Neurodevelopmental abnormality, complex FC, more than
one characteristic of complex FC, sex and race are widely regarded
as not representing risk factors [3,41,52]. In this study, comparison
of first FC in the first 24 months or the subsequent period as age
threshold for recurrence of FC revealed greater recurrence in the
first 24 months (59.7%) and that recurrence was higher in subjects
experiencing first FC at early ages, although there was no
statistically significant difference in terms of age at first attack
compared with cases with no recurrence (p = 0.216). History of FC
in the family, pre-FC fever below 39  C, first FC being observed in
less than 1 h after body temperature elevation, complex seizure,
focal seizure, recurrence within the same day, seizure duration
>15 min, more than one finding of complex FC, neurodevelop-
mental abnormality and long-term hospitalization on the neonatal
ICU were increased risk for recurrence of FC. (p values 0.026,
<0.001, <0.001, <0.001, 0.001, <0.001, <0.001, 0.041, 0.004 and
<0.001, respectively) (Table 2). Ages of mother and father in cases
with recurrence of FC were lower than those with no recurrence (p;
0.001 and 0.006, respectively) (Table 1). Logistic regression
analysis performed in order to examine the effect of more than
one variable on recurrence of FC revealed, in agreement with the
literature, that the risk of recurrence increased 7.1 times in subjects
with a history of FC in first and second degree relatives, a 17.8 times
greater risk in those with pre-FC duration of fever <1 h and a 17.6
times greater in those with pre-FC body fever <39 C (Table 3).
In this study, in agreement with the literature, recurrence of FC
was not correlated with such factors as sex, type of delivery, time of
birth, birth weight, consanguineous marriage, educational level of
parents, parental occupations or family income (p > 0.05) (Table 1).
No statistical correlation was also determined between recurrence
of FC and first and final electroencephalography findings, presence
of epilepsy in the family, use of mother’s milk, a low APGAR score
or a history of neonatal resuscitation (p > 0.05) (Table 2).
4.2. Risk factors affecting development of epilepsy in febrile convulsion
The risk of progression to epilepsy in FC is a subject of debate.
The risk in children with a single simple FC is regarded as low and
no different to that in the general population [3]. The most
important risk factors for epilepsy after FC are regarded as
neurodevelopmental abnormality, complex FC, and duration of
fever before FC seizure [3,29,41,53]. Presence of more than one
feature of complex FC is also regarded as a risk factor [3]. One study
reported that multiple episodes of FC and onset of FC episodes after
the age of 3 as risk factors [28]. In their study of 428 cases of FC,
Sfaihi et al. [41] reported a post-FC prevalence of epilepsy of 1.2%.
They also showed an increased risk of epilepsy in subjects with
fever <39 C during seizure and those with complex FC. Other risk
46 M. Canpolat et al. / Seizure 55 (2018) 36–47
factors they cited were focal, prolonged seizures, seizures
recurrent in the time of the same febrile illness, neurodevelop-
mental abnormality and a positive family history of epilepsy. They
described neurodevelopmental abnormality and the most signifi-
cant prognostic factor. Hwang et al. [53] assessed 204 cases and
observed unprovoked seizure in 23%. They reported onset of FC
seizures after the age of 3, a positive family history, abnormal EEG
findings and fever <39 C as risk factors for epilepsy. They also
reported no increased risk associated with low socioeconomic
status, age of mother at birth, cesarean delivery, low birth weight
or premature birth. Pavlidou et al. [29] reported that epilepsy
developed in 5.4% of 501 cases of FC, and that 4 or more
recurrences of FC and focal seizures when duration of fever was
short represented a high risk of progression to epilepsy.
In our previous study, we determined a prevalence of epileptic
seizure in school children in Kayseri of 0.6% in girls and 0.9% in
boys, for a total prevalence of 0.8% [20]. In this study, the post-FC
prevalence of epilepsy was 5.7% in girls and 8.7% in boys, a total
prevalence of 7.2%. In agreement with the literature, this finding
shows that the risk of epilepsy increases in FC. In agreement with
the literature, risk factors for epilepsy developing post-FC were
body temperature below 39  C, complex FC, first EEG being
pathological, seizures being focal and neurodevelopmental abnor-
mality (p = 0.001, <0.001, 0.002, <0.001 and <0.001, respectively).
Logistic regression analysis revealed a 4.5-fold greater risk of
epilepsy in subjects with first EEG after FC being pathological, and
a 21-fold greater risk in those with neurodevelopmental abnor-
mality (Table 5). In contrast to Pavlidou et al.’s [29] study,
recurrence of FC (independent of number of recurrences) was
significant as a risk factor for epilepsy (p < 0.001). Again in contrast
to that study, no statistical correlation was observed between
duration of fever before FC and development of epilepsy. In
contrast to Hwang et al.’s [53] study, no correlation was
determined between a history of epilepsy in the family and
development of epilepsy. This may be due to a positive family
history being present in only one case with epilepsy after FC in our
study group and to such cases being inadequately represented in
the study population.
Hwang et al. [53] emphasized that onset of FC seizures after the
age of 3 years increased the risk 4.99-fold. In our study, first FC in
the cases subsequently developing epilepsy was after the age of 2
in 51.5% of cases, and in 35% of those with no epilepsy. This finding
supports the idea of a high risk of progression to epilepsy of FC
seizures observed at a later age. However, no statistically
significant correlation was determined between age at first FC
and development of (p = 0.163).
The view that EEG is not indicative of prognosis in FC was
widely accepted in the past [3]. However, there have been studies
reporting the opposite in recent years. Hwang et al. [53] reported
that abnormal EEG findings increased the risk of epilepsy 5.95
times. In this study, a significant correlation was determined
between first EEG and risk of epilepsy (p = 0.002). Epilepsy was
observed in 89.3% of cases with pathological first EEG and in 10.7%
of those with normal EEG. The risk was 4.5 times greater in cases
with pathological first EEG. However, EEG was only performed in
103 cases, and the other cases’ not undergoing EEG may have been
misleading. No correlation was determined between final EEG and
risk of epilepsy. Epilepsy developed in 81.8% of cases receiving
long-term treatment after FC, but in only 18.2% of those not
receiving long-term therapy. This shows that long-term treatment
has a protective treatment has no protective effect against
development of epilepsy.
Wide population studies confirm that long-term prognosis is
good in most patients with febrile episodes [3,27]. In agreement
with the literature, 92.6% of patients monitored with a diagnosis of
FC were cured completely. In terms of sequelae, epilepsy was
determined in 7.2% of cases and somnambulism in one case.
Generalized, absence and partial type episode have been described
in FCs [3]. In this study, too, 19 (57.6%) of the 33 cases in which
epilepsy developed post-FC were complex partial in character,
while 12 (36.4%) were generalized. Epileptic episodes were
compatible with GEFS+ in 9 cases in which generalized episodes
were observed.
The main limitation of this study is that the questionnaire
return rate was 71.6%. This may be due to the study being planned
as a questionnaire study, to its being based on voluntary
participation and to families with no history of epilepsy or FC
showing insufficient interest in the study. Another limitation is
that no information was obtained concerning whether families
with no history of epilepsy or FC in their children had members
with epileptic seizure or FC and the impossibility of accessing
demographic and clinical data if individuals with FC and/or
epileptic episode existed.
In conclusion, the prevalence of FC in Kayseri, 4.3%, is closer to
the prevalences in developed countries. The most important risk
factors for recurrence of FC are a history of FC in first and second
degree relatives, pre-convulsion fever lasting less than 1 h and pre-
convulsion fever <39 C. Epilepsy occurs at a level of 7.2% in FC, and
the risk is higher than in the general population. The most
significant risk factor in terms of development of epilepsy is
neurodevelopmental abnormality.
Contributors
Canpolat M and Kumandas S were proposed the study and
wrote the first draft. Kumandas S helped the writing. All authors
contributed to the design and interpretation of the study.
Kumandas S, Gumus H, Elmali F, Per H, and Canpolat M are the
guarantor.
Competing interest
No conflict of interest.
Funding
None.
References
[1] Shinnar S, Glauser TA. Febrile seizures. In: Pellock JM, Bourgeois BFD, Dodson
WE, editors. Petiatric epilepsy. 3rd ed. New York: Demos Medical Publishing;
2008. p. 293–305.
[2] Guidelines for Epidemiologic Studies on Epilepsy. Commission on epidemiol-
ogy and prognosis, international league against epilepsy. Epilepsia
1993;34:592–6.
[3] Yakut A. Febril Konvulziyon. Türkiye Klinikleri 2003;119–27.
[4] Sadleir LG, Scheffer IE. Febrile seizures. BMJ 2007;334:307–11.
[5] Nelson KB, Ellenberg JH. Predictors of epilepsy in children who have
experienced febrile seizures. N Engl J Med 1976;295:1029–33.
[6] Duffner PK, Baumann RJ. A synopsis of the American Academy of Pediatrics’
practice parameters on the evaluation and treatment of children with febrile
seizures. Pediatr Rev 1999;20:285–7.
[7] Freeman JM, Vinig EP. Decision making and the child with febrile seizures.
Pediatr Rev 1992;13:298–304.
[8] Kjeldsen MJ, Kyvik KO, Friis ML, Christensen K. Genetic and environmental
factors in febrile seizures: a Danish population-based twin study. Epilepsy Res
2002;51:167–77.
[9] Verity CM, Butler NR, Golding J. Febrile convulsions in a national cohort
followed up from birth. II—medical history and intellectual ability at 5 years of
age. Br Med J (Clin Res Ed) 1985;290:1311–5.
[10] Annegers JF, Blakley SA, Hauser WA, Kurland LT. Recurrence of febrile
convulsions in a population-based cohort. Epilepsy Res 1990;5:209–16.
[11] Tsuboi T. Epidemiology of febrile and afebrile convulsions in children in Japan.
Neurology 1984;34:175–81.
[12] Stanhope JM, Brody JA, Brink E, Morris CE. Convulsions among the chamorro
people of Guam, Mariana islands. II. Febrile convulsions. Am J Epidemiol
1972;95:299–304.
 M. Canpolat et al. / Seizure 55 (2018) 36–47
[13] Yayla V, Çakmak G, Apak I, _ Ateş U, Erdog _
an F. Ilkokul Çocuklarında Geçirilmiş
Febril Konvülziyon Prevalansı. Yeni Symp 1997;35:11–3.
[14] Gökyig _
it A, Çalışkan A. Istanbul Tıp Fakültesi son sınıf ög rencilerinde febril
_
konvülzüyon prevalansı. Istanbul Tıp Fakültesi Mecmuası 1985;48:102–7.
[15] Aydin A, Ergor A, Ozkan H. Effects of sociodemographic factors on febrile
convulsion prevalence. Pediatr Int 2008;50:216–20.
[16] Offringa M, Bossuyt PM, Lubsen J, Ellenberg JH, Nelson KB, Knudsen FU, et al.
Risk factors for seizure recurrence in children with febrile seizures: a pooled
analysis of individual patient data from five studies. J Pediatr 1994;124:574–
84.
[17] Berg AT, Shinnar S, Hauser WA, Leventhal JM. Predictors of recurrent febrile
seizures: a metaanalytic review. J Pediatr 1990;116:329–37.
[18] Verity CM, Golding J. Risk of epilepsy after febrile convulsions: a national
cohort study. BMJ 1991;303:1373–6.
[19] Annegers JF, Hauser WA, Shirts SB, Kurland LT. Factors prognostic of
unprovoked seizures after febrile convulsions. N Engl J Med 1987;316:493–8.
[20] Canpolat M, Kumandas S, Poyrazoglu HG, Gumus H, Elmali F, Per H. Prevalence
and risk factors of epilepsy among school children in Kayseri City Center, an
urban area in Central Anatolia, Turkey. Seizure 2014;23:708–16.
_
[21] Türkiye Istatistik _
Kurumu. (TÜIK), BÜlgesel GÜstergeler. TR 72 Kayseri, Sivas,
_
Yozgat: Türkiye Istatistik Kurumu; 2009 Yayın No: 3378.
[22] Capovilla G, Mastrangelo M, Romeo A, Vigevano F. Recommendations for the
management of “febrile seizures”: ad hoc task force of LICE guidelines
commission. Epilepsia 2009;50:2–6.
[23] Fisher RS, Acevedo C, Arzimanoglou A, Bogacz A, Cross JH, Elger CE, et al. ILAE
official report: a practical clinical definition of epilepsy. Epilepsia
2014;55:475–82.
[24] Fetveit A. Assessment of febrile seizures in children. Eur J Pediatr
2008;167:17–27.
[25] Hirtz D. Febrile seizures. Pediatr Rev 1997;18:5–9.
[26] Hauser WA. The prevalence and incidence of convulsive disorders in children.
Epilepsia 1994;35:1–6.
_
[27] Yeni Epilepsi SN. Insidansı, _ Yeni SN,
Prevalansı ve Risk Faktörleri. In: Bora I,
_
Gürses C, editors. Epilepsi. Istanbul: Nobel Tıp Kitapevleri; 2008. p. 65–73.
[28] Pavlidou E, Hagel C, Panteliadis C. Febrile seizures: recent developments and
unanswered questions. Childs Nerv Syst 2013;29:2011–7.
[29] Pavlidou E, Panteliadis C. Prognostic factors for subsequent epilepsy inchildren
with febrile seizures. Epilepsia 2013;54:2101–7.
[30] Fu Z, Lavine L, Wang Z, Cheng X, Li S, Emoto S, et al. Prevelance and incidence of
febril seizures (FBS) in China. Neurology 1987;37:149.
[31] Bauman R, Marx M, Leinidakis M. Epilepsy in rural Kentucky: prevelance in a
population of school age children. Epilesiae 1987;19:75–80.
[32] Offringa M, Hazebroek-Kampschreur AA, Derksen Lubsen G. Prevelance of
febril seizures in Dutch school children. Paediatr Peerinat Epidemiol
1991;5:181–8.
[33] Baldin E, Ludvigsson P, Mixa O, Hesdorffer DC. Prevalence of recurrent
symptoms and their association with epilepsy and febrile seizure in school-
aged children: a community-based survey in Iceland. Epilepsy Behav
2012;23:315–9.
47
[34] Yakinci C, Kutlu NO, Durmaz Y, Karabiber H, Eg ri M. Prevelance of febrile
convulsion in 3637 children of primary school age in the province of Malatya,
Turkey. J Trop Pediatr 2000;46:249–50.
[35] Özmen M, Çaliskan M. Febril konvülziyonlar. Ist _ Çocuk Klin Derg 1995;30:116–
21.
[36] Verity CM, Butler NR, Golding J. Febrile convulsions in a national cohort
followed up from birth. I-prevelance and recurence in the first five years of life.
BMJ 1985;290:1307–10.
[37] Wallace SJ. Febrile seizures. Epilepsia 1996;2:28–33.
[38] Ling SG. Febrile convulsions: acute seizures characteristics and anti-
convulsant therapy. Ann Trop Paediatr 2000;20:227–30.
[39] Kolfen W, Pehle K, Konig S. Is the long-term outcome of following febrile
convulsions favorable? Dev Med Child Neurol 1998;40:667–71.
[40] Visser AM, Jaddoe VW, Hofman A, Moll HA, Steegers EA, Tiemeier H, et al. Fetal
growth retardation and risk of febrile seizures. Pediatrics 2010;126:919–25.
[41] Sfaihi L, Maaloul I, Kmiha S, Aloulou H, Chabchoub I, Kamoun T, et al. Febrile
seizures: an epidemiological and outcome study of 482 cases. Childs Nerv Syst
2012;28:1779–84.
[42] Østergaard JR. Febrile seizures. Acta Paediatr 2009;98:771–3.
[43] Shimony A, Afawi Z, Asher T, Mahajnah M, Shorer Z. Epidemiological
characteristics of febrile seizures–comparing between Bedouin and Jews in
the southern part of Israel. Seizure 2009;18:26–9.
[44] Esmaili Gourabi H, Bidabadi E, Cheraghalipour F, Aarabi Y, Salamat F. Febrile
seizure: demographic features and causative factors. Iran J Child Neurol
2012;6:33–7.
[45] Delpisheh A, Veisani Y, Sayehmiri K, Fayyazi A. Febrile seizures: etiology,
prevalence, and geographical variation. Iran J Child Neurol 2014;8:30–7.
[46] Vestergaard M, Basso O, Henriksen TB, et al. Risk factors for febrile convulsions.
Epidemiyology 2002;13:282–7.
[47] Berg AT, Shinnar S, Hauser WA, Alemany M, Shapiro ED, Salomon ME, et al. A
prospective study of recurrent febrile seizures. N Engl J Med 1992;327:1122–7.
[48] Okumura A, Uemura N, Suzuki M, Itomi K, Watanabe K. Unconsciousness and
delirious behavior in children with febrile seizures. Pediatr Neurol
2004;30:316–9.
[49] Knudsen FU. Febrile seizures-treatment and outcome. Brain Dev 1996;18:438–
49.
[50] Maytal J, Steele R, Eviatar L, Novak G. The value of early postictal EEG in
children with complex febrile seizures. Epilepsia 2000;41:219–21.
[51] Sofijanov N, Emoto S, Kuturec M, Dukovski M, Duma F, Ellenberg JH, et al. Febrile
seizures: clinical characteristics and initial EEG. Epilepsia 1992;33:52–7.
[52] Pavlidou E, Tzitiridou M, Kontopoulos E, Panteliadis CP. Which factors
determine febrile seizure recurrence? A prospective study. Brain Dev
2008;30:7–13.
[53] Hwang G, Kang HS, Park SY, Han KH, Kim SH. Predictors of unprovoked seizure
after febrile seizure: short-term outcomes. Brain Dev 2015;37:315–21.
[54] Fallah R, Karbasi SA. Recurrence of febrile seizure in Yazd, Iran. Turk J Pediatr
2010;52:618–22.
[55] Kolfen W, Pehle K, Konig S. Is the long-term outcome of following febrile
convulsions favorable? Dev Med Child Neurol 1998;40:667–71.