Abstrak Journal Reading

Oleh :

Ayudita Silvia Hasibuan 2040312117

Preseptor :
Dr. dr. Yusri Dianne Jurnalis, Sp.A (K)

BAGIAN ILMU KESEHATAN ANAK

RSUP DR.M.DJAMIL PADANG FAKULTAS KEDOKTERAN UNVERSITAS
ANDALAS

2021
Abstrak Jurnal 1

Kejang Berulang pada Anak Setelah Menghentikan Pengobatan Antiepilepsi:

Follow-up 5 Tahun

Inn-Chi Lee, Shuan-Yow Li, Yung-Jung Chen

Latar Belakang: Kami ingin mengidentifikasi faktor-faktor yang terkait dengan kontrol
kejang dan kekambuhan setelah remisi 2 tahun pada anak dengan epilepsi.

Metode: Kami melakukan follow-up 5 tahun terhadap anak dengan epilepsi yang pengobatan
antiepilepsinya telah dihentikan. Analisis bivariat dan multivariat digunakan untuk
membandingkan fitur elektroensefalogram (EEG) dan temuan klinis. Dalam studi ini, 43 dan
64 pasien dengan dan tanpa kekambuhan kejang/ seizure recurrence (SR) yang terdaftar.

Hasil: Gambaran klinis yang sangat terkait dengan SR dalam analisis univariat termasuk
etiologi gejala untuk kejang, riwayat status epileptikus, durasi pengobatan sebelum
menghentikan obat antiepilepsi, dan temuan EEG abnormal pada saat menghentikan obat
antiepilepsi.

Kesimpulan: Kami menemukan bahwa riwayat status epileptikus, gejala epilepsi parsial,
durasi pengobatan sebelum menghentikan obat antiepilepsi, dan EEG abnormal saat
pengobatan dihentikan merupakan prediktor penting dari SR. Faktor risiko SR setelah
penghentian obat antiepilepsi telah diteliti dalam beberapa penelitian. Namun, riwayat status
epileptikus sebagai faktor prediktif jarang disebutkan.

Kata Kunci: Obat anti epilepsi, anak-anak, epilepsi, hasil, kambuh, remisi.
 Pediatrics and Neonatology (2017) 58, 338e343

Available online at www.sciencedirect.com

ScienceDirect

jo u rn a l h o m e p a g e : h tt p : // w w w . p e d i a t r -n e o n a to l .c o m

ORIGINAL ARTICLE

Seizure Recurrence in Children

after Stopping Antiepileptic
Medication: 5-YearFollow-Up
a,b
Inn-Chi Lee , Shuan-Yow Li b,c,*, Yung-Jung Chen d,*
a
Division of Pediatric Neurology, Department of Pediatrics, Chung Shan Medical University Hospital,
Taichung, Taiwan
b
Institute of Medicine, School of Medicine, Chung Shan Medical University, Taichung, Taiwan
c
Genetics Laboratory and Department of Biomedical Sciences, Chung Shan Medical University,
Taichung, Taiwan
d
Department of Pediatrics, College of Medicine, National Cheng Kung University
Hospital, Tainan,Taiwan

Received Apr 15, 2016; received in revised form Jul 6, 2016; accepted Aug 1, 2016
Available online 18 December 2016

Key Words Background: We wanted to identify in children with epilepsy the factors
antiepileptic drug; associated with seizure control and recurrence after a 2-year remission.
children; Methods: We did a 5-year follow-up of epileptic children whose antiepileptic
epilepsy; medication had been stopped. Bivariate and multivariate analyses were used
outcome; to compare features of electro- encephalograms (EEGs) and clinical findings.
recurrence; In this study, 43 patients with and 64 without a seizure recurrence (SR) were
remission enrolled.
Results: Clinical features strongly associated with SR in the univariate
analysis included a symptomatic etiology for seizures, a history of status
epilepticus, treatment duration before stopping antiepileptic drugs, and
abnormal EEG findings at the time of stopping antiepileptic drugs.
Conclusion: We found that a history of status epilepticus, symptomatic
partial epilepsy, treat- ment duration before stopping antiepileptic drugs,
and an abnormal EEG when the medication was stopped are important
predictors of SR. The risk factors of SR after discontinuing antiep- ileptic
drugs have been investigated in several studies. However, a history of status
epilepticus as a predictive factor is rarely mentioned.

Copyright ª 2017, Taiwan Pediatric Association. Published by Elsevier

Taiwan LLC. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/ by-nc-nd/4.0/).
* Corresponding authors. National Cheng Kung University Hospital, Tainan, Taiwan (Y.-J. Chen); Genetics Laboratory and

http://dx.doi.org/10.1016/j.pedneo.2016.08.005 1875-9572/Copyright ª 2017, Taiwan Pediatric Association. Published by

Elsevier Taiwan LLC.
Abstrak Jurnal 2

Mengevaluasi Efek Probiotik pada Pediatrik dengan Sakit Perut Berulang

Parisa Rahmani, Azin Ghouran-Orimi, Farzaneh Motamed, Alireza Moradzadeh

Latar Belakang: Nyeri perut berulang / Reccurent Abdominal Pain (RAP) merupakan salah
satu keluhan yang sering muncul dalam praktek umum, terutama pada pediatri dan
merupakan salah satu penyebab umum rujukan ke gastroenterologiklinik.

Tujuan: Penelitian ini dirancang untuk menyelidiki efek probiotik untuk pengobatan RAP
dan hasil yang diharapkan dari terapi.

Metode: Seratus dua puluh lima anak dengan diagnosis RAP menurut kriteria Rome III untuk
sindrom iritasi usus besar / irritable bowel syndrome (IBS), nyeri perut fungsional / functional
abdominal pain (FAP), dispepsia fungsional (FD), dan migrain perut (AM), terdaftar dalam uji
coba terkontrol acak dobleblind ini.

Hasil: Enam puluh lima subjek menerima probiotik, dan lainnya menerima pengobatan
plasebo selama 4 minggu. Terapi dengan Lactobacillus reuteri efektif pada 32 pasien
dibandingkan dengan 8 pasien, menanggapi pengobatan plasebo. Dibandingkan dengan garis
dasar, semua variabel yang berhubungan dengan nyeri menunjukkan penurunan yang
signifikan untuk IBS dan FD pada akhir minggu ke-4. Namun, hal itu tidak merespon dengan
baik di kelompok FAP dan AM. Hasil terkait nyeri seperti, frekuensi nyeri, keparahan, dan
durasi nyeri menurun setelah pengobatan probiotik. Tidak ada respon terapeutik yang terlihat
pada kelompok AM setelah pemberian probiotik. L. reuteri secara signifikan menyebabkan
pereda nyeri pada populasi secara keseluruhan, dan juga pada subkelompok FAP, FD, dan
IBS.

Kesimpulan: L. reuteri probiotik cenderung meredakan RAP dan dapat direkomendasikan

untuk pengobatan fungsional gangguan pencernaan.

Kata Kunci: Lactobacillus reuteri, Recurrent abdominal pain (RAP), Anak.

 Vol. 63, No. 12, 485–490, 2020 Original article
CEP https://doi.org/10.3345/cep.2019.01613

Evaluating the effects of probiotics in

pediatrics with recurrent abdominal
pain

Parisa Rahmani, MD1, Azin Ghouran-Orimi, MD2, Farzaneh Motamed, MD1, Alireza Moradzadeh, MD3
1Pediatric Gastroenterologyand Hepatology Research Center, Tehran University of Medical Sciences, Tehran, Iran; 2Iran university of Medical
Sciences, Tehran, Iran;
3Department of Pediatric Gastroenterology, Children’s Medical Center, Tehran University of Medical Sciences, Tehran, Iran

Background: Recurrent abdominal pain (RAP) is one of the frequent complaints in general practice, particularly in pedi- atrics
and is among the common cause of referral to gastroen-terology clinics.
Purpose: This study is designed to investigate the effects of probiotics for the treatment of RAP and desired therapeutic
outcomes.
Methods: One hundred twenty-five children with the diag- nosis of RAP according to Rome III criteria for irritable bowel
syndrome (IBS), functional abdominal pain (FAP), functional dyspepsia (FD), and abdominal migraine (AM), were enrolled in this
double-blind randomized controlled trial.
Results: Sixty-five subjects received probiotics, and othersreceived placebo treatment for 4 weeks. Lactobacillus reuteriwas
therapeutically effective in 32 patients compared to 8 pa-tients, responding to the placebo treatment. Compared to base-line, all
pain-related variables showed a significant reduction forthe IBS and FD at the end of the 4th week. However, it did notrespond well in
FAP and AM groups. Pain-related outcomessuch as, frequency of the pain, severity, and duration of thepain were decreased
following the probiotic treatment. Notherapeutic response was seen in AM group after the admini-stration of probiotics. L. reuteri
significantly led to pain relief inthe overall population, and also in FAP, FD, and IBS subgroups. Conclusion: L. reuteri probiotics are
likely to lead to RAPrelief and can be recommended for the treatment of functionalgastrointestinal disorders.

Key words: Lactobacillus reuteri, Recurrent abdominal pain,Child

Corresponding author: Alireza Moradzadeh, MD. Department of Pediatric Gastroenterology, Children’s Medical Center, Tehran University of Medical
Sciences, Tehran, Iran E-mail: md.moradzadeh.a@gmail.com, https://orcid.org/0000-0003-3355-7086
Received: 17 December, 2019, Revised: 29 April, 2020, Accepted: 20 May, 2020
This is an open-access article distributed under the terms of the Creative Commons Attribution Non-Commercial License
(http://creativecommons.org/licenses/by- nc/4.0/) which permits unrestricted non-commercial use, distribution, and reproduction in any medium,
provided the original work is properly cited.
Copyright © 2020 by The Korean Pediatric Society
Abstrak Jurnal 3

Prediktor Awal Hiperbilirubinemia Neonatal Pada Bayi Baru Lahir Cukup Bulan

May Ahmed Khairy, Walaa Alsharany Abuelhamd, Ismail Mohamed Elhawary, Ahmed Said
Mahmoud Nabayel

Latar belakang : Penanda prediktif yang dapat diandalkan dokter untuk mengidentifikasi
bayi baru lahir mana yang akan mengalami hiperbilirubinemia signifikan menjadi kewajiban
untuk pencegahan hiperbilirunemia berat. Penelitian ini bertujuan untuk menentukan kadar
kritis serum bilirubin dan albumin serta rasio bilirubin / albumin pada tali pusat sedini
mungkin sebagai penanda yang dapat diandalkan.

Desain penelitian : Penelitian prospektif ini melibatkan 175 neonatus cukup bulan.
Pengukuran bilirubin, albumin dan rasio bilirubin / albumin tali pusat dilakukan untuk
memprediksi hiperbilirubinemia yang bermakna pada bayi baru lahir cukup bulan
berdasarkan pengukuran serum bilirubin yang dilakukan dalam 5 hari kehidupan.

Hasil : Sebagian besar kasus yang berkembang menjadi hiperbilirubinemia neonatal

bermakna (67,9%) memiliki kadar albumin tali pusat ≤ 2,8 gm / dl. Nilai cut off rasio
Bilirubin / albumin tali pusat > 0,61 memiliki nilai prediktif yang baik dengan sensitivitas
100% dan spesifisitas 88,4%, serta nilai cut off albumin serum tali pusat ≤ 3,0 mg / dl juga
memiliki nilai prediktif yang baik dengan sensitivitas 85,7 % dan spesifisitas 67,3%.

Analisis kurva ROC terhadap total bilirubin tali pusat menunjukkan nilai cut off ≥ 1,84 mg /
dl memiliki nilai prediktif yang baik dengan sensitivitas 100,0% dan spesifisitas 87,1%.

Kesimpulan : Rasio bilirubin / albumin, bilirubin serum dan albumin tali pusat dapat
menjadi prediktor awal hiperbilirubinemia neonatal.

Kata kunci : Rasio bilirubin /albumin tali pusat, albumin serum tali pusat, bilirubin serum
tali pusat, neonatal, hiperbilirubinemia.
 Pediatrics and Neonatology (2019) 60, 285e290

Available online at www.sciencedirect.com

ScienceDirect

jo u rn a l h o m e p a g e : h t tp :/ / w w w . p e d i a t r - n e o n a t o l. c o m

Original Article

Early predictors of neonatal

hyperbilirubinemia in full term
newborn
May Ahmed Khairy a, Walaa Alsharany Abuelhamd a,*,
Ismail Mohamed Elhawary a, Ahmed Said Mahmoud Nabayel b
a
Department of Pediatrics, Faculty of Medicine, Cairo University, Cairo, Egypt
b
Department of Pediatrics, Ministry of Health, Cairo, Egypt

Received Jan 13, 2018; received in revised form Mar 5, 2018; accepted Jul 18, 2018
Available online 26 July 2018

Background: Reliable predictive markers enabling physicians to identify which

Key Words newborns will develop significant hyperbilirubinemia have become mandatory for
cord bilirubin/ prevention of severe hy- perbilirunemia. We aimed at determining the critical
albumin ratio; cord serum bilirubin and albumin levels and bilirubin/albumin ratio early as
cord serum albumin; reliable markers.
cord serum bilirubin; Study design: This prospective study included 175 full-term neonates.
neonatal Measurement of cord bilirubin, albumin and bilirubin/albumin ratio was done to
hyperbilirubinemia predict significant hyperbilirubine- mia in healthy term newborns based on serum
bilirubin measurements made within 5 days of life.
Results: Most cases that developed significant neonatal hyperbilirubinemia (67.9%)
had cord albumin level ≤ 2.8 gm/dl. Cord Bilirubin/albumin ratio cut off value >
0.61 had a good pre- dictive value with a sensitivity of 100% and specificity of
88.4%, and cord serum albumin cut off value ≤ 3.0 mg/dl also had a good predictive
value with a sensitivity of 85.7% and specificity of67.3%.
ROC curve analysis of cord total bilirubin demonstrated that a cut off value of
≥1.84 mg/dl had a good predictive value with a sensitivity of 100.0% and
specificity of 87.1%.
Conclusion: Cord bilirubin/albumin ratio, serum bilirubin and albumin could be
early predic- tors for neonatal hyperbilirubinemia.

Copyright ª 2018, Taiwan Pediatric Association. Published by Elsevier Taiwan LLC.

This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/ by-nc-nd/4.0/).* Corresponding author.
Pediatric Department, Faculty of Medicine, Cairo University, New Children Hospital, (Abu El
Rish), Cairo UniversityHospitals. https://doi.org/10.1016/j.pedneo.2018.07.005
1875-9572/Copyright ª 2018, Taiwan Pediatric Association. Published by Elsevier Taiwan
LLC.
Abstrak Jurnal 4

Pengaruh Anemia Defisiensi Besi Terhadap Perkembangan Bahasa

Pada Anak-Anak Prasekolah Mesir

Mervat A.M. Youssef, Eman S. Hassan, Dalia G. Yasien

Latar Belakang : Anemia defisiensi besi (Iron deficiency anemia / IDA) merupakan
defisiensi gizi yang paling sering terjadi terutama di negara berkembang.

Tujuan : Studi ini mengevaluasi pengaruh IDA pada perkembangan bahasa anak-anak
prasekolah.

Metodologi : Penelitian ini adalah studi cross-sectional komparatif multisenter yang

melibatkan 226 anak-anak antara usia 4–6 tahun. Anak-anak diklasifikasikan menjadi dua
kelompok, anemia (pasien) dan non anemia (kontrol) menurut kadar hemoglobin. Semua
anak yang anemia menjalani pemeriksaan zat besi lengkap termasuk; Zat besi serum, total
iron binding capacity (TIBC), Kadar feritin serum, untuk memastikan diagnosis anemia
defisiensi besi. Penilaian kognitif dilakukan dengan menggunakan skala kecerdasan
terjemahan bahasa Arab Stanford Binet, versi empat yang terdiri dari empat skor area
kognitif; penalaran visual, penalaran verbal, penalaran kuantitatif dan memori jangka pendek.
Pengukuran IQ dan usia mental dihitung untuk setiap anak. Evaluasi bahasa dilakukan
dengan menggunakan tes Bahasa Arab. Kecerdasan bahasa reseptif, kecerdasan bahasa
ekspresif dan kecerdasan bahasa total dihitung untuk setiap anak.

Hasil : 122 anak anemia dan 90 non anemia dengan kadar hemoglobin masing-masing 10,65
dan 11,96 g / dL (P <0,000). Anak-anak anemia memiliki serum feritin yang lebih rendah
secara signifikan (p <0,0001), dan serum besi (p <0,0001) dibandingkan dengan kontrol.
Kedua kelompok dibandingkan dalam hal usia, jenis kelamin, tingkat sosial ekonomi dan
tingkat pendidikan orang tua. Tidak ada perbedaan signifikan yang diamati mengenai IQ, usia
mental, bahasa reseptif, ekspresif dan kecerdasan bahasa total antara anak-anak anemia dan
non-anemia.

Kesimpulan : IDA tampaknya tidak berpengaruh pada perkembangan bahasa pada anak
prasekolah di Mesir. Studi terkontrol besar di masa depan dengan waktu tindak lanjut yang
lama untuk kelompok usia yang lebih muda diperlukan untuk menentukan apakah ada
hubungan antara IDA dengan perkembangan bahasa.

Kata kunci : Bahasa, Anemia defisiensi zat besi, Pengartian, Anak-anak Mesir
 International Journal of Pediatric Otorhinolaryngology 135 (2020) 110114

Contents lists available at ScienceDirect

International Journal of Pediatric Otorhinolaryngology

journal homepage: www.elsevier.com/locate/ijporl

2016 ESPO Congress

Effect of iron deficiency anemia on language development in preschool

Egyptian children
Mervat A.M. Youssefa,∗, Eman S. Hassanb, Dalia G. Yasienc
a - Pediatric Hematology Unit, Children Hospital, Faculty of Medicine, Assiut University, Assiut, Egypt
b - Phoniatrics Unit, ENT Department, Assiut University Hospital, Faculty of Medicine, Assiut University, Assiut, Egypt
c - Phoniatrics Unit, ENT Department, Helwan University Hospital, Faculty of Medicine, Helwan University, Cairo, Egypt

A R T I C LE IN FO A BS T RA CT

Keywords: Background: Iron deficiency anemia (IDA) is the most common nutritional deficiency primarily in developing
Language countries.
Iron deficiency anemia Objective: This study evaluates the effect of IDA on language development in preschool children.
Cognition Methodology: The study is a multicenter, comparative cross-sectional study included 226 children between
Egyptian children ages 4–6 years. The children were classified into two groups’ anemic (patients) and non anemic (controls)
according to the hemoglobin level. All anemic children subjected to complete iron study including; Serum iron,
total iron binding capacity (TIBC), Serum ferritin level, to confirm the diagnosis of iron deficiency anemia.
Cognitive assessment was done using the Arabic translation Stanford Binet intelligence scale, version four which
comprised of four cognitive area scores; visual reasoning, verbal reasoning, quantitative reasoning and short-
term memory. Measurement of IQ and mental age were calculated for each child. Language evaluation was
done using the Arabic Language test. Receptive language quotient, expressive language quotient and total
language quotient were calculated for each child.
Results: 122 children were anemic and 90 were non-anemic with hemoglobin level 10.65 and 11.96 g/dL,
re- spectively (P < 0.000). Anemic children had significantly lower serum ferritin (p < 0.0001), and serum
iron (p < 0.0001) compared to the controls. Both groups were matched as regards age, sex, socioeconomic
levels and parental educational level. No significant differences observed regarding IQ, mental age,
receptive, ex- pressive and total language quotients between anemic and non-anemic children.
Conclusions: IDA does not seem to have an effect on language development in preschool Egyptian
Children. Future large controlled studies with long follow-up time for the younger age group are needed to
determine whether there are existent associations between IDA with language development.

∗ Corresponding author. Children Hospital, hematology unit, Faculty of medicine, Assiut University, Egypt.
E-mail address: mamuosif2000@gmail.com (M.A.M. Youssef).

https://doi.org/10.1016/j.ijporl.2020.110114
Received 30 October 2019; Received in revised form 10 May 2020; Accepted 10 May 2020
Av ailableonline13M a y2 020
016 5 -5
8 76/© 2020E lse
v ie
rB .V.A llrig
h tsreserv
ed.
Abstrak Jurnal 5

Prevalensi dan Manifestasi Klinis Makrolida yang Resisten terhadap Pneumonia

Mycoplasma pada Anak-anak di Korea

Eun Lee, Hyun-Ju Cho, Soo-Jong Hong, Jina Lee, Heungsup Sung, Jinho Yu

Tujuan: Tingkat resistensi makrolida dari Mycoplasma pneumoniae telah meningkat pesat
pada anak-anak. Studi tentang gambaran klinis antara macrolide susceptible-M. pneumoniae
(MSMP) dan macrolide resistant-M. pneumoniae (MRMP) kurang. Tujuan dari penelitian
ini adalah untuk mengidentifikasi tingkat resistensi makrolida M. pneumoniae pada anak di
Korea dengan M. pneumoniae penupmonia tahun 2015 dan membandingkan manifestasi
antara MSMP dan MRMP.

Metode: Di antara 122 anak (0-18 tahun) yang didiagnosis M. pneumoniae pneumonia, 95
anak dengan hasil uji sensitivitas makrolida diikutsertakan dalam penelitian ini. Manifestasi
klinis diperoleh dengan menggunakan rekam medis retrospektif.

Hasil: Tingkat resistansi makrolida dari M. pneumoniae adalah 87,2% (82 dari 94 pasien)
pada anak-anak dengan M. pneumoniae radang paru-paru. Satu pasien menunjukkan tipe
campuran dari tipe liar dan mutasi A2063G pada 23S rRNA M. pneumoniae. Tidak ada
perbedaan yang signifikan dalam temuan klinis, laboratorium, dan radiologis antara
kelompok MSMP dan MRMP pzada kunjungan pertama ke rumah sakit kami. Interval
waktu antara inisiasi makrolida dan defervesensi secara signifikan lebih lama pada
kelompok MRMP (4,9 ± 3,3 vs 2,8 ± 3,1 hari, P = 0,039).

Kesimpulan: Tingkat resistensi makrolida dari M. pneumoniae sangat tinggi pada anak-
anak dengan M. pneumoniae pneumonia di Korea. Manifestasi klinis MRMP mirip dengan
MSMP kecuali untuk periode penundaan setelah pemberian makrolida. Pemantauan terus
menerus dari kejadian dan kerentanan antimikrobial MRMP diperlukan untuk mengontrol
penyebarannya dan menetapkan strategi untuk mengobati infeksi M. pneumoniae resisten
antibiotik lini kedua.

Kata kunci: Anak, Macrolide, Mycoplasma pneumoniae, Resistensi obat

Original article
Korean J Pediatr 2017;60(5):151-157
https://doi.org/10.3345/kjp.2017.60.5.151 Korean J Pediatr 2017;60(5):151-157
pISSN 1738-1061•eISSN 2092-7258
Korean J Pediatr

Prevalence and clinical manifestations of

macrolide resistant Mycoplasma
pneumoniaepneumonia in Korean children
Eun Lee1, Hyun-Ju Cho2, Soo-Jong Hong2, Jina Lee2, Heungsup Sung3, Jinho Yu2
1
Department of Pediatrics, Chonnam National University Hospital, Gwangju, Departments of 2 Pediatrics and 3 Laboratory Medicine, Asan Medical Center, University
of Ulsan College of Medicine, Seoul, Korea

Corresponding author: Jinho Yu, MD, PhD

Purpose: Macrolide resistance rate of Mycoplasma pneumoniae has rapidly increased in children.
Department of Pediatrics, Asan Medical Center, Uni-
Studies on the clinical features between macrolide susceptible-M. pneumoniae (MSMP) and macrolide versity of Ulsan College of Medicine, 88 Olympic-ro
resistant-M. pneumoniae (MRMP) are lacking. The aim of this study was to identify the macrolide 43-gil, Songpa-gu, Seoul 05505, Korea
resistance rate of M. pneumoniae in Korean children with M. pneumoniae penupmonia in 2015 and Tel: +82-2-3010-3922
Fax: +82-2-473-3725
compare manifestations between MSMP and MRMP.
E-mail: jyu3922@gmail.com
Methods: Among 122 children (0–18 years old) diagnosed with M. pneumoniae pneumonia, 95 children Co-corresponding author: Heungsup Sung, MD,
with the results of macrolide sensitivity test were included in this study. Clinical manifestations were PhD
acquired using retrospective medical records. Department of Laboratory Medicine, Asan Medical
Center, University of Ulsan College of Medicine, 88
Results: The macrolide resistant rate of M. pneumoniae was 87.2% (82 of 94 patients) in children with
Olympic-ro 43-gil, Songpa-gu, Seoul 05505, Korea
M. pneumoniae pneumonia. One patient showed a mixed type of wild type and A2063G mutation in Tel: +82-2-3010-4499
23S rRNA of M. pneumoniae. There were no significant differences in clinical, laboratory, and radiologic Fax: +82-2-2045-4081
findings between the MSMP and MRMP groups at the first visit to our hospital. The time interval between E-mail: sung@amc.seoul.kr
initiation of macrolide and defervescence was significantly longer in the MRMP group (4.9±3.3 vs.
Received: 8 September, 2016
2.8±3.1 days, P=0.039). Revised: 1 December, 2016
Conclusion: The macrolide resistant rate of M. pneumoniae is very high in children with M. pneumoniae Accepted: 21 December, 2016
pneumonia in Korea. The clinical manifestations of MRMP are similar to MSMP except for the deferve-
scence period after administration of macrolide. Continuous monitoring of the occurrence and antimi-
crobial susceptibility of MRMP is required to control its spread and establish strategies for treating
second-line antibiotic resistant M. pneumoniae infection.

Key words: Child, Macrolide, Mycoplasma pneumoniae, Drug resistance

Copyright © 2017 by The Korean Pediatric Society

This is an open-access article distributed under the

terms of the Creative Commons Attribution Non-
Commercial License (http://creativecommons.org/
licenses/by-nc/4.0/) which permits unrestricted non-
commercial use, distribution, and reproduction in any
medium, provided the original work is properly cited.

https://doi.org/10.3345/kjp.2017.60.5.151 151
https://doi.org/10.3345/kjp.2017.60.5.151 339
Pediatrics and Neonatology (2017) 58, 338e343

Available online at www.sciencedirect.com

ScienceDirect

journal homepage: http://www.pediatr-neonatol.com

ORIGINAL ARTICLE

Seizure Recurrence in Children after

Stopping Antiepileptic Medication: 5-Year
Follow-Up
Inn-Chi Lee a,b, Shuan-Yow Li b,c,*, Yung-Jung Chen d,*

a
Division of Pediatric Neurology, Department of Pediatrics, Chung Shan Medical University Hospital,
Taichung, Taiwan
b
Institute of Medicine, School of Medicine, Chung Shan Medical University, Taichung, Taiwan
c
Genetics Laboratory and Department of Biomedical Sciences, Chung Shan Medical University,
Taichung, Taiwan
d
Department of Pediatrics, College of Medicine, National Cheng Kung University Hospital, Tainan,
Taiwan

Received Apr 15, 2016; received in revised form Jul 6, 2016; accepted Aug 1, 2016
Available online 18 December 2016

Key Words Background: We wanted to identify in children with epilepsy the factors associated with
antiepileptic drug; seizure control and recurrence after a 2-year remission.
children; Methods: We did a 5-year follow-up of epileptic children whose antiepileptic medication had
epilepsy; been stopped. Bivariate and multivariate analyses were used to compare features of electro-
outcome; encephalograms (EEGs) and clinical findings. In this study, 43 patients with and 64 without a
recurrence; seizure recurrence (SR) were enrolled.
remission Results: Clinical features strongly associated with SR in the univariate analysis included a
symptomatic etiology for seizures, a history of status epilepticus, treatment duration before
stopping antiepileptic drugs, and abnormal EEG findings at the time of stopping antiepileptic
drugs.
Conclusion: We found that a history of status epilepticus, symptomatic partial epilepsy, treat-
ment duration before stopping antiepileptic drugs, and an abnormal EEG when the medication
was stopped are important predictors of SR. The risk factors of SR after discontinuing antiep-
ileptic drugs have been investigated in several studies. However, a history of status epilepticus
as a predictive factor is rarely mentioned.
Copyright ª 2017, Taiwan Pediatric Association. Published by Elsevier Taiwan LLC. This is an
open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/
by-nc-nd/4.0/).

* Corresponding authors. National Cheng Kung University Hospital, Tainan, Taiwan (Y.-J. Chen); Genetics Laboratory and Department of
Biomedical Sciences, Chung Shan Medical University, Taichung, Taiwan (S.-Y. Li).
E-mail addresses: i0000528@ms12.hient.net (S.-Y. Li), pcyj@mail.ncku.edu.tw (Y.-J. Chen).

http://dx.doi.org/10.1016/j.pedneo.2016.08.005
1875-9572/Copyright ª 2017, Taiwan Pediatric Association. Published by Elsevier Taiwan LLC. This is an open access article under the CC BY-
NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
Seizure Recurrence in Children
1. Introduction
Approximately 70% of epileptic children who are seizure
free for more than 2e4 years while on antiepileptic drugs
will remain seizure free after withdrawing from their
antiepileptic drugs.1e3 Because epilepsy is a heterogeneous
disorder (i.e., it can be lesional or genetic), rather than a
single disease entity, some patients will require drugs for
continued seizure control, but others might not. There is no
single, accepted duration for defining remission.4 Stopping
antiepileptic drugs early is not recommended as a standard
practice in children, even in those who rapidly respond to
the medication. However, many physicians believe that it is
necessary to stop medication after 2 years.1,5e7 Dis-
continuing antiepileptic drugs early might prevent unnec-
essarily prolonged treatment, or it might increase the risk
of seizure recurrence (SR). Hence, it might be appropriate
to consider in which patients medication can be safely
stopped rather than just when it can be stopped.8 In
addition, predictors of prognosis after discontinuation of
treatment would be of considerable clinical value.
Shorvon and Sander9 reported that 13 (12%) of 108 pa-
tients with epilepsy had an intermittent pattern of SR after
certain periods of being seizure free. More than half of the
recurrences occurred in children who were tapering or had
completely tapered their medications or who had never
taken medications.10 The risk of a recurrence after dis-
continuing antiepileptic drugs in unselected groups of
seizure-free patients is on the order of 25% in the 1st year
and 29% after 2 years.11
Many factors appear to affect the degree of risk for SR
after stopping medication: epileptic syndromes, age at
onset of epilepsy, underlying etiology, an abnormal elec-
troencephalogram (EEG), epilepsy severity, individual drug
side effects, and a prediction of recurrence.11 However,
there is no general agreement on the risk factors for SR
after stopping antiepileptic drugs. This is most likely
because studies often have different inclusion criteria,
which affect the composition of the groups investigated. In
addition, the duration of follow-ups and methodologies
often differ.
In this study, we wanted to identify the risk factors of
recurrence when discontinuing antiepileptic drugs after
2 years of seizure remission; thus, we compared, in a 5-year
follow-up, the clinical features between children with and
without SR.
2. Patients and Methods
We retrospectively reviewed medical records in our epi-
lepsy database from the Department of Pediatric Neurology
at National Cheng Kung University Hospital and the
Department of Pediatric Neurology at Chung Shan Medical
University Hospital. We identified 125 children with epi-
lepsy who had been treated at our hospital and met the
inclusion criteria: (a) onset of epilepsy at 15 years of age or
younger; (b) seizure free for 2 years or more; (c) antiepi-
leptic drugs discontinued; (d) information about seizure
manifestation, medication, blood level of antiepileptic
drugs, and EEG findings available. Informed consent was
obtained from all parents. Relevant clinical information
339
was obtained from the medical records, from a question-
naire completed by the parents, and from a structured
telephone interview every 3 months. The children were
followed periodically after the medications had been dis-
continued. If a seizure did occur, their antiepileptic drug
was restarted at the previous dose. Based on the outcome
after stopping the medication, the patients were divided
into two groups: (i) SR groupdSR during the 5-year follow-
up, and (ii) no SR (NSR) groupdno SR during the 5-year
follow-up.
We recorded each patient’s sex, age at onset of epi-
lepsy, types of seizures and epilepsy syndromes, number of
seizures before treatment, seizure frequency (low: one
every 4e6 months; moderate: one to two every
1e3 months; and high: >1/month), presence of status
epilepticus, febrile seizures, family history of epilepsy,
presence of abnormal neurological findings (e.g., delayed
cognitive development), motor deficits, duration of treat-
ment, number of antiepileptic drugs used, duration of the
seizure-free period before stopping antiepileptic drugs, and
interval between stopping antiepileptic drugs and SR. In
addition, we looked at the serial EEG findings when epilepsy
was diagnosed, when antiepileptic drugs were stopped,
during the follow-up after antiepileptic drugs had been
discontinued or when seizures recurred, and at the end of
the 2-year follow-up. An abnormal EEG was defined as a
specific focal or generalized epileptiform or slow-wave
abnormality. The types of seizures and epileptic syn-
dromes were classified according to the Commission on
Classification and Terminology of the International League
Against Epilepsy.12,13
For patients with SR, outcomes were classified as good
(seizures remitted for 1 year after antiepileptic drug
restart), fair (seizures remitted for <1 year after antiepi-
leptic drug restart), and deteriorated (seizures more
frequent than before stopping antiepileptic drugs and not
seizure free even with polytherapy).
2.1. Statistical analysis
The risk of SR was calculated using KaplaneMeier statistics
and is shown in survival curves. The ManneWhitney test was
used for univariate analyses of continuous variables; a c2 or
Fisher exact test was used for bivariate analyses of
dichotomous variables; and a stepwise Cox hazard-function
test was used for multivariate analyses to determine which
factors were most strongly related to predicting SR. Sig-
nificance was set at p < 0.05.
3. Results
A total of 125 eligible patients were identified. Of these,
five declined to participate and 13 were unreachable. The
remaining 107 patients were recruited for the study. Sixty-
four (60%) remained seizure free and 43 (40%) had recurrent
seizures (Table 1). There was a significantly higher rate of
status epilepticus and a longer follow-up period in patients
with SR than in those without.
After the initiation of antiepileptic drug therapy,
remission occurred in 13.6  18.2 months in the SR group
and in 12.5  15.5 months in the NSR group. There were 26
340 I.-C. Lee et al

EEG at the time of antiepileptic drug discontinuation: 30

Table 1 Comparison of seizure recurrence group and no
(70%) of 43 in the SR group and 31 (48%) of 64 in the NSR
seizure recurrence group children at the end of the 5-year
group. In the last EEGdon each patient’s last visit to the
follow-up.
outpatient clinicd48 of the 107 patients (45%) had an
Characteristics SR group* NSR group* p abnormal EEG: 22 (51%) in the SR group and 26 (41%) in the
(n Z 43) (n Z 64) NSR group. Thus, the number of abnormal EEGs fell in both
groups during the follow-up. A significant number of chil-
Sex (male/female) 21/22 35/29 NS
dren with an abnormal EEG when their antiepileptic drugs
Family history of epilepsy 5 (11.6) 6 (9.3) NS
were discontinued had a recurrence of seizures, and the
Age at onset of epilepsy
rate of abnormal EEGs was higher in the SR group than in
<2 y 6 (14) 3 (5) NS
the NSR group (p Z 0.045; odds ratio, 2.45; 95% confidence
2e12 y 32 (74) 57 (89) NS
interval, 1.08e5.55).
>12 y 5 (12) 4 (6) NS
There were significantly higher rates of status epi-
Type of seizures
lepticus and longer treatment duration before stopping
Partial 24 (56) 26 (40)
antiepileptic drugs for patients with an SR than for those
Generalized 19 (44) 38 (60)
without (Tables 1 and 2).
History of febrile seizures 5 (11.6) 8 (12.5) NS
The cumulative time-dependent probability of remain-
History of status epilepticus 6 (13.9) 1 (1.6) 0.016
ing seizure free was significantly different in the two groups
EEG when epilepsy was diagnosed
(Figure 1). The overall KaplaneMeier estimate of SR was 0.3
Epileptiform EEG 38 (88.4) 61 (95.3) NS
at 6 months, 0.48 at 12 months, and 0.51 at 24 months. The
Slowing EEG 12 (28) 9 (14) NS
mean time to SR was 10.8  6.2 months, and the median
Abnormal EEG 30 (70) 31 (48) 0.045
was 12 months. The risk of SR was greatest in the first few
when AEDs stopped
months after the antiepileptic drugs had been dis-
Abnormal EEG at 22 (51) 26 (41) NS
continued: seizures recurred in 22 (51%) of the 43 patients
the end of follow-up
in the SR group within 12 months (mean, 5.7  3.2 months;
Initial seizure frequency
median, 6 months), and in 21 (49%), within 24 months
High (>1 every mo) 10 (23.2) 13 (20.3) NS
(mean, 16.1  3.2 months; median, 16 months; range,
Moderate 9 (21) 8 (12.5) NS
13e24 months). Terminal remission after SR occurred in 33
(1e2 every 1e3 mo)
(76.8%) of the 43 patients. The remaining 10 patients
Low (1 every 4e6 mo) 24 (55.8) 43 (67.2) NS
showed no remission, and six of them had more seizures
Total no. of seizures before control
than before the recurrence. More patients in the SR group
2 3 (7) 17 (26.5) <0.05
had symptomatic focal epilepsy than did patients in the NSR
3e10 30 (70) 33 (51.5) NS
group (p Z 0.007; Tables 3 and 4).
>10 10 (23) 14 (22) NS
The most important predictors of SR were a history of
No. of antiepileptic drugs
status epilepticus, symptomatic partial epilepsy, treatment
1 30 (69.8) 51 (80) NS
duration before stopping antiepileptic drugs, and an
2 8 (18.6) 9 (14) NS
Withdrawal duration
6 mo 18 (42) 28 (44)
>6 mo 25 (58) 36 (56) Table 2 Severity of epilepsy in SR group and NSR group
Cognitive developmental 11 (25.6) 8 (12.5) NS children after stopping antiepileptic drugs.
delay (IQ < 70) Variables SR group* NSR group* p
Neurological abnormality 12 (25.6) 8 (12.5) NS (n Z 43) (n Z 64)
Neuroimaging abnormality 3/16 (19) 6/17 (35) NS
Cognitive developmental 11 (25.6) 8 (12.5) NS Time between seizure onset 10.2  22.8 6.5  13.3 NS
delay (IQ < 70) and treatment, mo
Time to seizure control, mo 13.6  18.2 12.5  15.5 NS
AED Z antiepileptic drug; EEG Z electroencephalogram;
Treatment duration before 71.3  37.5 55.0  21.8 0.005
IQ Z intelligence quotient; NSR Z no seizure recurrence;
stopping antiepileptic
SR Z seizure recurrence.
* Unless otherwise indicated, data are n (%). drugs, mo
Age at last seizure before 7.4  3.9 7.8  3.2 NS
stopping antiepileptic
drugs, y
(60%) seizure-free children in the SR group and 40 (62%) in
Seizure-free duration 4.9  1.9 5.0  2.1 NS
the NSR group within 12 months of starting antiepileptic
before stopping
drug treatment (early remission). The remaining 41 chil-
antiepileptic drugs, y
dren, 17 in the SR group and 24 in the NSR group, were
Age when antiepileptic 11.3  4.3 11.1  3.3 NS
seizure free after more than 1 year of treatment (late
drugs were stopped, y
remission).
Age at end of follow-up, y 16.1  5.9 15.7  3.3 NS
On the initial EEG, when each of our patients was
diagnosed with epilepsy, 102 of 107 patients (95%) had an NS Z not significant; NSR Z no seizure recurrence;
abnormal EEG: 39 (91%) in the SR group and 63 (98%) in the SR Z seizure recurrence.
NSR group. Sixty-one of 107 patients (57%) had an abnormal * Data are mean  standard deviation.
Seizure Recurrence in Children 341

Table 4 Risk factors for seizure recurrence in children

after stopping antiepileptic drugs.
Risk factors Hazard 95% p
ratio confidence
interval
Early onset of epilepsy (<2 y) 1.85 1.08e3.17 NS
History of status epilepticus 2.32 1.55e3.44 <0.02
Initial seizure frequency
High (>1 every mo) 1.10 0.64e1.89 NS
Seizure frequency
Moderate (1e2 every 1e3 mo) 1.32 0.83e2.09 NS
to high (>1 every month)
Seizure type (partial vs. general) 1.44 0.90e2.29 NS
Symptomatic partial epilepsy 1.95 1.26e3.01 <0.05
Abnormal EEG when antiepileptic 1.74 1.02e2.94 <0.05
Figure 1 KaplaneMeier curve showing the probability of drugs were stopped
remaining seizure free after discontinuing antiepileptic drugs Slowing EEG 1.58 0.99e2.52 NS
in children with seizures after 2 years of being seizure free: Cognitive developmental delay 1.59 0.99e2.55 NS
overall recurrence risk. Time ‘0’ (lower left) refers to when (intelligence quotient < 70)
antiepileptic drugs were stopped. EEG Z electroencephalogram; NS Z not significant.

abnormal EEG when antiepileptic drugs were discontinued

(Tables 2 and 4).
Medication was used in children without an SR: 50 (78%) antiepileptic drugs. In the SR group, 30 children (69%) were
were treated with a single antiepileptic drug, nine (14%) treated with a single antiepileptic drug, eight (19%) with
with two antiepileptic drugs, and four (6%) with three two antiepileptic drugs, three with three antiepileptic
drugs, one with four antiepileptic drugs, and one with five
antiepileptic drugs.
Table 3 Epilepsy syndrome distribution in patients with
childhood-onset epilepsy.
4. Discussion
Epilepsy syndrome SR group NSR group p
(n Z 43) (%) (n Z 64) (%) We found significant risk factors related to SR after dis-
Localization related 39 48 continuing antiepileptic drugs: a history of status epi-
Idiopathic 10 (23.2) 27 (42.2) NS lepticus, symptomatic partial epilepsy, treatment duration
Rolandic 3 22 before stopping antiepileptic drugs, and an abnormal EEG
Occipital 7 5 at the time the antiepileptic drugs were stopped. The
Lesional (symptomatic) 11 (25.6) 5 (7.8) 0.007 longer duration of treatment in the SR group is probably
Temporal lobe 6 3 because of the severity of the disease, which will normally
Frontal lobe 4 1 influence a physician to not discontinue antiepileptic drug
Occipital lobe 1 1 treatment. Once antiepileptic drugs are started, many
Cryptogenic 18 (41.8) 16 (25) NS children with epilepsy will become seizure free, and after a
(probable lesional) period the antiepileptic drugs can be withdrawn with good
Generalized 3 14 results. Although the risk of recurrent seizures after dis-
Idiopathic 3 (6.9) 14 (21.8) NS continuing the antiepileptic drugs is relatively low, there is
Childhood absence 1 6 no guarantee that their seizure freedom will be long last-
Juvenile myoclonic 0 1 ing. After antiepileptic drugs are stopped, as many as
Generalized 2 6 21e37% of patients with childhood-onset epilepsy will un-
toniceclonic dergo recurrent seizures.5,10,11,14e16 The risk factors of SR
seizures on awakening after discontinuing antiepileptic drugs have been investi-
Other idiopathic 0 1 gated in several studies.5,11,14e17 However, a history of
generalized status epilepticus as a predictive factor is rarely
epilepsy/unclassified mentioned.
idiopathic generalized Most studies have found that epilepsy with an onset in
epilepsy childhood has a more favorable prognosis than does epi-
Undetermined whether 1 (2.3) 2 (3) lepsy with an onset in adolescence (age > 10e12 years).11
focal or generalized An adolescent onset is consistently associated with a high
Total 43 64 risk of SR after stopping antiepileptic drugs because the risk
of recurrence is more a function of the age at epilepsy
NS Z not significant; NSR Z no seizure recurrence;
onset rather than the age at which antiepileptic drugs are
SR Z seizure recurrence.
discontinued.11,18 A younger age at onset is also associated
342
with a higher risk of SR after stopping antiepileptic
drugs,18,19 but this appears to be limited to patients with a
remote symptomatic etiology.18 Hence, a younger age at
onset is frequently a marker for more severe neurological
impairment. In our study, the age at onset was correlated
with a risk of SR after stopping antiepileptic drugs. Recur-
rence before 2 years even in those aged <2 years occurred
in six of nine of children who were younger than 2 years of
age at seizure onset, which was more often than in children
who were older than 2 years of age at seizure onset.
Patients with remote symptomatic epilepsy are less
likely to enter remission than are those with idiopathic or
cryptogenic epilepsy. Once in remission, they are about 50%
more likely to have recurrent seizures if their antiepileptic
drugs are stopped. In one study,18 12 of 22 (55%) patients
with intelligence quotients (IQs) less than 70, and pre-
sumably with remote symptomatic epilepsy, had recurrent
seizures, compared with six of 45 (13%) patients with IQs
(>70; p < 0.001). In our study, after antiepileptic drugs
were discontinued, 11 of 43 (26%) children who had recur-
rent seizures also had cognitive developmental delays, but
only eight of 64 (13%) children without recurrent seizures
had cognitive developmental delays. Focal slowing in EEG
could suggest a structural basis for brain.20 Berg and Shin-
nar11 reported that after discontinuing antiepileptic drugs,
the risk factors of SR included status epilepticus, symp-
tomatic etiology, and focal slow EEGs. In our study, the
ratio of focal slow EEGs was greater in the SR group than in
the NSR group, but not significant, probably because of the
small number of cases number in our study.
Verrotti et al6 evaluated the predictive value of inter-
ictal EEG findings after the antiepileptic drugs had been
discontinued. Two-thirds of the children in the SR group had
EEG abnormalities at the time they stopped taking their
antiepileptic drugs, but only 10% of the children in the NSR
group did. Verrotti et al6 concluded that the reappearance
of EEG abnormalities after discontinuing antiepileptic drugs
is a risk factor for SR. Shinnar et al18 reported that focal
slowing on an EEG is a risk factor for recurrence of seizures
after withdrawing from antiepileptic drugs. In this study,
we found that an abnormal EEG when discontinuing anti-
epileptic drugs was associated with an increased risk of SR,
which is consistent with Shinnar et al18 and Verrotti et al.21
Focal slowing on EEGs was more common in the SR group
than in the NSR group, which is compatible with Shinnar
et al,18 despite the smaller number of cases and the
absence of a significant difference.
Most SR happens early: almost half occurs within
6 months, and 60e80% occurs within 1 year after dis-
continuing antiepileptic drugs. Although late recurrence
does happen, it is rare. The British Medical Research
Council study22,23 of antiepileptic drug withdrawal found
that the increased risk of SR attributable to withdrawal was
limited to the first 2 years after discontinuing antiepileptic
drugs. The risk of SR was approximately twice as high in
patients who had discontinued their antiepileptic drugs
than in those who had not, but that difference was seen
only for the first 2 years after stopping antiepileptic drugs,
after which the risk of subsequent recurrence was the same
for both groups.24
Shinnar et al18 reported that if seizures recur after dis-
continuing antiepileptic drugs, the prognosis is generally
I.-C. Lee et al
favorable and that the majority of patients become seizure
free again after restarting antiepileptic drugs. The prog-
nosis for long-term remission appears to be primarily a
function of the underlying epilepsy syndrome. In our study,
we found that although 33 of 43 (77%) patients with SR
reentered remission during the 5-year follow-up, 10 pa-
tients did not. Six of these 10 patients had more seizures
than before discontinuing antiepileptic drugs, even on
antiepileptic drug polytherapy. Camfield and Camfield16
reported that 1% of children who became seizure free and
then discontinued antiepileptic drugs had recurrent sei-
zures that could not be pharmacologically controlled again.
In this study, six of 43 patients (14%) in the SR group had an
increase in seizure frequency, and even status epilepticus,
after they had discontinued their antiepileptic drugs.
This study has some limitations. First, only patients
whose seizures had been in remission for 2 years were
enrolled. Thus, our series might include cases that are less
severe than those in other studies. However, the purpose of
this study was to assess the risk factors of recurrence in
patients after they had discontinued their antiepileptic
drugs. Second, this was not a population-based study. Our
patients probably presented more severe varieties of epi-
lepsy than the typical pediatric patient with epilepsy in the
general population; thus, our findings are most likely
inapplicable in any context other than a referral center.
Remission rates after discontinuing antiepileptic drugs
would presumably have been even higher in a more repre-
sentative sample population. In this study, we excluded
cases with status epilepticus caused by poor patient
compliance. The etiologies of status epilepticus are multi-
ple and have traditionally been classified as acute or
remote symptomatic, and idiopathic.25,26 We, however, did
not classify status epilepticus as idiopathic or secondary;
rather, classified the etiology as epilepsy syndrome itself.
In conclusion, our results indicate that the following
might lead to SR: a history of status epilepticus, symp-
tomatic partial epilepsy, treatment duration before stop-
ping antiepileptic drugs, and an abnormal EEG at the time
of antiepileptic drug discontinuation.
Conflicts of Interests
The authors declare that they have no conflicts of interest
with respect to the authorship or publication of this article
and confirm that they have read the journal’s position on
issues involved in ethical publication and affirm that this
report is consistent with those guidelines.
Acknowledgments
This work was supported by Chung Shan Medical University
Hospital grant CSH-2016-C-010. Ethical approval of the
study was provided by the hospital’s IRB (CS13036).
References
1. Shinnar S, Vining EP, Mellits ED, D’Souza BJ, Holden K,
Baumgardner RA, et al. Discontinuing antiepileptic medication
Seizure Recurrence in Children
in children with epilepsy after two years without seizures. A
prospective study. N Engl J Med 1985;313:976e80.
2. Arts WF, Visser LH, Loonen MC, Tjiam AT, Stroink H,
Stuurman PM, et al. Follow-up of 146 children with epilepsy
after withdrawal of antiepileptic therapy. Epilepsia 1988;29:
244e50.
3. Camfield PR, Camfield CS. Antiepileptic drug therapy: when is
epilepsy truly intractable? Epilepsia 1996;37:S60e5.
4. Guidelines for epidemiologic studies on epilepsy. Commission
on Epidemiology and Prognosis, International League Against
Epilepsy. Epilepsia 1993;34:592e6.
5. Verrotti A, Morresi S, Basciani F, Cutarella R, Morgese G,
Chiarelli F, et al. Discontinuation of anticonvulsant therapy in
children with partial epilepsy. Neurology 2000;55:1393e5.
6. Verrotti A, D’Egidio C, Agostinelli S, Parisi P, Spalice A,
Chiarelli F, et al. Antiepileptic drug withdrawal in childhood
epilepsy: what are the risk factors associated with seizure
relapse? Eur J Paediatr Neurol 2012;16:599e604.
7. Li W, Si Y, Zou XM, An DM, Yang H, Zhou D. Prospective study on
the withdrawal and reinstitution of antiepileptic drugs among
seizure-free patients in west China. J Clin Neurosci 2014;21:
997e1001.
8. Geerts AT, Niermeijer JM, Peters AC, Arts WF, Brouwer OF,
Stroink H, et al. Four-year outcome after early withdrawal of
antiepileptic drugs in childhood epilepsy. Neurology 2005;64:
2136e8.
9. Shorvon SD, Sander JW. Temporal patterns of remission and
relapse of seizures in patients. In: Schmidt D, Morselli PL, ed-
itors. Intractable epilepsy. New York: Raven Press; 1986.
p. 13e23.
10. Berg AT, Shinnar S, Levy SR, Testa FM, Smith-Rapaport S,
Beckerman B, et al. Two-year remission and subsequent
relapse in children with newly diagnosed epilepsy. Epilepsia
2001;42:1553e62.
11. Berg AT, Shinnar S. Relapse following discontinuation of anti-
epileptic drugs: a meta-analysis. Neurology 1994;44:601e8.
12. Proposal for revised clinical and electroencephalographic
classification of epileptic seizures. From the Commission on
Classification and Terminology of the International League
Against Epilepsy. Epilepsia 1981;22:489e501.
13. Proposal for revised classification of epilepsies and epileptic
syndromes. Commission on Classification and Terminology of
the International League Against Epilepsy. Epilepsia 1989;30:
389e99.
14. Sillanpää M, Schmidt D. Prognosis of seizure recurrence after
stopping antiepileptic drugs in seizure-free patients: a long-
343
term population-based study of childhood-onset epilepsy. Ep-
ilepsy Behav 2006;8:713e9.
15. Bouma PA, Peters AC, Brouwer OF. Long term course of child-
hood epilepsy following relapse after antiepileptic drug with-
drawal. J Neurol Neurosurg Psychiatry 2002;72:507e10.
16. Camfield P, Camfield C. The frequency of intractable seizures
after stopping AEDs in seizure-free children with epilepsy.
Neurology 2005;64:973e5.
17. Gherpelli JL, Kok F, dal Fomo S, Elkis LC, Lefevre BH,
Diament AJ. Discontinuing medication in epileptic children: a
study of risk factors related to recurrence. Epilepsia 1992;33:
681e6.
18. Shinnar S, Berg AT, Moshé SL, Kang H, O’Dell C, Alemany M,
et al. Discontinuing antiepileptic drugs in children with epi-
lepsy: a prospective study. Ann Neurol 1994;35:534e45.
19. Emerson R, D’Souza BJ, Vining EP, Holden KR, Mellits ED,
Freeman JM. Stopping medication in children with epilepsy:
predictors of outcome. N Engl J Med 1981;304:1125e9.
20. Smith SJ. EEG in neurological conditions other than epilepsy:
when does it help, what does it add? J Neurol Neurosurg Psy-
chiatry 2005;76:ii8e12.
21. Dooley J, Gordon K, Camfield P, Camfield C, Smith E. Discon-
tinuation of anticonvulsant therapy in children free of seizures
for 1 year: a prospective study. Neurology 1996;46:969e74.
22. Bonnett LJ, Shukralla A, Tudur-Smith C, Williamson PR,
Marson AG. Seizure recurrence after antiepileptic drug with-
drawal and the implications for driving: further results from
the MRC Antiepileptic Drug Withdrawal Study and a systematic
review. J Neurol Neurosurg Psychiatry 2011;82:1328e33.
23. Strozzi I, Nolan SJ, Sperling MR, Wingerchuk DM, Sirven J. Early
versus late antiepileptic drug withdrawal for people with epi-
lepsy in remission. Cochrane Database Syst Rev 2015;(2):
CD001902.
24. Arts WF, Brouwer OF, Peters AC, Stroink H, Peeters EA,
Schmitz PI, et al. Course and prognosis of childhood epilepsy:
5-year follow-up of the Dutch study of epilepsy in childhood.
Brain 2004;127:1774e84.
25. Gaspard N, Foreman BP, Alvarez V, Cabrera Kang C,
Probasco JC, Jongeling AC, et al. New-onset refractory status
epilepticus: etiology, clinical features, and outcome.
Neurology 2015;85:1604e13.
26. Spatola M, Novy J, Du Pasquier R, Dalmau J, Rossetti AO. Status
epilepticus of inflammatory etiology: a cohort study.
Neurology 2015;85:464e70.
 Original article
CEP Vol. 63, No. 12, 485–490, 2020
https://doi.org/10.3345/cep.2019.01613

Evaluating the effects of probiotics in pediatrics with

recurrent abdominal pain
Parisa Rahmani, MD1, Azin Ghouran-Orimi, MD2, Farzaneh Motamed, MD1, Alireza Moradzadeh, MD3
1
Pediatric Gastroenterology and Hepatology Research Center, Tehran University of Medical Sciences, Tehran, Iran; 2Iran university of Medical Sciences, Tehran, Iran;
3
Department of Pediatric Gastroenterology, Children’s Medical Center, Tehran University of Medical Sciences, Tehran, Iran

Background: Recurrent abdominal pain (RAP) is one of the
frequent complaints in general practice, particularly in pedi­
atrics and is among the common cause of referral to gastroen­
terology clinics.
Purpose: This study is designed to investigate the effects of
probiotics for the treatment of RAP and desired therapeutic
outcomes.
Methods: One hundred twenty-five children with the diag­
nosis of RAP according to Rome III criteria for irritable bowel
syndrome (IBS), functional abdominal pain (FAP), functional
dyspepsia (FD), and abdominal migraine (AM), were enrolled
in this double-blind randomized controlled trial.
Results: Sixty-five subjects received probiotics, and others
received placebo treatment for 4 weeks. Lactobacillus reuteri
was therapeutically effective in 32 patients compared to 8 pa­
tients, responding to the placebo treatment. Compared to base­
line, all pain-related variables showed a significant reduction for
the IBS and FD at the end of the 4th week. However, it did not
respond well in FAP and AM groups. Pain-related outcomes
such as, frequency of the pain, severity, and duration of the
pain were decreased following the probiotic treatment. No
therapeutic response was seen in AM group after the admini­
stration of probiotics. L. reuteri significantly led to pain relief in
the overall population, and also in FAP, FD, and IBS subgroups.
Conclusion: L. reuteri probiotics are likely to lead to RAP
relief and can be recommended for the treatment of functional
gastrointestinal disorders.
Key words: Lactobacillus reuteri, Recurrent abdominal pain,
Child
Key message
Question: ecurrent abdominal pain (RAP) is a chief complaint
among pediatrics and is associated with reduced quality of
life, for both parent and child, and economic burden. Does
probiotics reduce the frequency of RAP among children?
Finding: This study reported the effects of Lactobacillus reuteri
probiotics among children with RAP as a result of multiple
etiologies.
Meaning: The administration of probiotic supplements is sig­
nificantly associated with pain relief among RAP children
presented with functional abdominal pain, irritable bowel
syndrome, and functional dyspepsia.
Introduction
Gut microbiota is an important determinant of gastrointestinal
health. Alterations in this diverse collection of microbes can lead
to several gastric and extragastric diseases. Probiotics are living
microorganisms that can be beneficial to the health of the host,
if provided in an appropriate amount.1) These microorganisms
include lactic acid bacilli species (such as Lactobacillus and
Bifidobacterium), nonpathogenic Escherichia coli species (e.g.
E. coli strain Nissle 1917), Clostridium butyricumi, Strepto­
coccus salivarius, and Saccharomyces boulardi (noninfectious
yeast species). Genetically-engineered bacteria have immunomo­
dulatory characteristics such as stimulating the production of
interleukin-10 (IL-10) and the trefoil factor, which have bene­
ficial effects on the immune system.2) Several studies have shown
that probiotics are effective against numerous pathological con­
ditions.3)
Recurrent abdominal pain (RAP) is one of the most common
consequences of pediatric gastrointestinal tract infections,4)
which affects 10% of all children and the highest incidence is
among 7 and 12 years old children.5) Four to twenty percent of
school-going children are presented with the complaint of RAP
that prevents their daily activities6) such as schooling7) perhaps
due to inadequate treatment.8) Anxiety in parents and frequent
doctor referral imposes an economic and emotional burden, and
results from quantitative studies have shown that children with
RAP are unlikely to respond to any definitive treatment plan.
Owing to the beneficial effects of probiotics on the gastrointes­
tinal tract, immune system, the aim of this study is to investigate
the effects of L. reuteri in the treatment of RAP in children aged 6

Corresponding author: Alireza Moradzadeh, MD. Department of Pediatric Gastroenterology, Children’s Medical Center, Tehran University of Medical Sciences, Tehran, Iran
E-mail: md.moradzadeh.a@gmail.com, https://orcid.org/0000-0003-3355-7086
Received: 17 December, 2019, Revised: 29 April, 2020, Accepted: 20 May, 2020
This is an open-access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-
nc/4.0/) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.
Copyright © 2020 by The Korean Pediatric Society

Probiotics (Lactobacilus reuteri) treatment (4 weeks)
on recurrent abdominal pain
• Frequency
• Severity
• Colic pattern
• Duration
• Dysfunction
“ Lactobacillus reuteri probiotics was significantly effective for the
treatment of recurrent abdominal pain in our study.”
Graphical abstract
to 16 years old, referred to the gastroenterology clinic.
Methods
This is a double-blinded clinical trial that was conducted from
June 2017 to June 2018. A total of 198 children with RAP signs
referred to the gastroenterology clinic were randomly divided
into case and control groups. Children who were included in the
study were between the ages of 6 and 16 years, and the diagnosis
of nonorganic RAP was performed according to ROME III
criteria.9)
Data collection method: The information collected included
age, sex and weight of the child, the frequency of repetitive pain,
the severity of pain, the duration of each episode of pain, the
number of days drugs were administered to treat the pain, num­
ber of days of disturbance in daily activity, the type of primary
disease and the pattern of pain mentioned by the doctor and
recorded on a daily basis by the parents in the notebook.
The following were excluded from the study: the presence
of any one of the red flag items, use of antibiotics in the last 1
month, organic disorder based on clinical and paraclinical find­
ings, and participants or parents who did not co-operate in re­
gards with medications and referrals.
At the time of referral, the children were physically examined
and the following information was registered at the time of regi­
stration and end of the study by single physician: pain intensity
based on the Wang-Baker FACES Pain Rating Scale (WBFPRS)
2, frequency of pain and recurrence, duration of each episode of
pain, pattern of pain (colic Crampi or permanent), the number of
days affected by day-to-day activities (such as school absenteeism)
and the need for other medications to relieve pain. The children
were randomly assigned into 2 groups as quadruple blocks of
case and control using Block-Randomization method. They were
486 Rahmani P, et al. Probiotics and recurrent abdominal pain
Double-blind randomized control study (n=125)
treated for 4 weeks with probiotic chewable tablets (containing
108 colony forming units L. reuteri) or placebo for 2 times a day.
The boxes containing placebo and probiotic were similar in the
shape, size, and taste.
The patients’ parents were provided with the notebook to
record information such as pain intensity based on WBFPRS
criteria, frequency of repetitive pain per day, duration of each
episode of pain, type of the pain, interruption of daily activities,
and the need for the use of another drug.
Patients were examined every 2 weeks and the changes in the
cases recorded by the parents were examined where, parents
were questioned regarding the relief in the pain. According to
the basic principles of the Helsinki Declaration, the information
was recorded confidentially; complete details of the study were
given to the patients and written consent was obtained for each
participating individual. Patients’ data were kept confidential
and were not disclosed to any individual or legal person. No
intervention was none to harm the patient and no additional
costs were imposed on patients. This study was approved by the
Research Ethics Board of Tehran University of Medical Sciences
(TUMS.91/D/130/378).
The data was computerized and analyzed using IBM SPSS ver.
18.0 (IBM Co., Armonk, NY, USA). The mean of quantitative
data, as the central index and standard deviation, was reported
as a dispersion index. Mann-Whitney-Wilcoxon test was used
for comparative analysis for the following parameters; age,
weight, frequency of repetitive pain, pain intensity, duration of
episode of pain, number of days drugs were used and number of
days needed to use the drug and the number of days of disturbed
daily activities, between the 2 groups. Chi-square or Fisher exact
test was used to determine the relationship between the type of
primary disease and the pattern of pain in the case and control
groups. P value of <0.05 was considered significant.
www.e-cep.org
Results
After the following the exclusion criteria and inability of pati­
ents to adhere to the study, a total of 125 subjects were studied
where, 57 children had FAP, 29 had FD, 30 had IBS, and 9 were
presented with AM. Overall, 65 patients with an average age
of 7.3±1.7 years in case group and 60 patients with an average
age of 7.7±2.1 years were in the control group. The mean body
weight in the case group was 23.1±7.6 kg and in control group
was 23.5±5.6 kg. The male to female ratio was 27/38 in the
case group and 30/30 in the control group. Above mentioned
variables (age, weight, and sex) were not significantly different
between the 2 groups (P>0.05). The frequency of repeated initial
pain in the group receiving probiotic was 6.8±6.8 and in the
placebo (control) group was 7.8±5.9, which was not statistically
significant (P=0.59). Frequency of repetitive pain 4 weeks after
these 2 groups was 3.6±2.2 and 4.6±4.9, respectively, which
was found to be statistically significant (P<0.001).
In this study, success in the treatment that was defined as pain
case group was 0.3±0.05 days a week and in the placebo was
0.6±0.2 days per week, which was marked with statistical signi­
ficance (P=0.04).
Two patients in the case group with an average of 0.6±0.1
days/wk required drugs for the pain management, at the beginn­
ing of the study whereas, a week after the commencement of the
treatment, 1 case for 0.2±0.03 days/wk needed drugs for pain
management. There was no significant difference in the need
for medication before and after treatment between the 2 groups
(P>0.05).
The success rate in treatment was 32 (49%) in the case group
and 8 (13%) in the control group, which was significantly differ­
ent in the 2 groups (P<0.001). All data and statistical analysis are
summarized in Table 1 for all participants in the study at the time
of entry and after 4 weeks.
Table 1. Characteristics of all study participants by study group
Probiotic Placebo
Characteristic P value
(n=65) (n=60)

intensity=0 (soft-faced child or facial scan=0) was shown as the Age (yr) 7.3±1.7 7.7±2.1 0.43
result of L. reuteri probiotics intake in the study population. Body weight (kg) 23.1±7.6 23.5±5.6 0.26

The severity of the primary pain prior to the treatment in the Sex, male:female 27:38 30:30 0.37
Frequency of pain
case and control groups was 3.3±0.9 and 3.1±0.6, respectively,
At the baseline 8.6±6.0 7.8±5.9 0.59
which did not show a statistically significant difference between
After 4 weeks 2.2±3.6 4.9±4.6 0.00
the 2 groups (P=0.19). However, there was a significant differ­
Severity of pain
ence between the severity of pain 4 weeks after the treatment in
At the baseline 3.3±0.9 3.1±0.6 0.19
the 2 groups (P<0.001); 1.3±1.1 in the case group compared to
After 4 weeks 1.1±1.3 2.0±1.0 0.00
1±2 in the control group.
Pattern of pain, colic:continiuse
Sixty-two patients in the case group and 58 patients in the
At baseline 62:3 58:2 1.00
control group, had the colic pain and 3 cases in the case group
After 4 weeks 31/2 50/2 <0.001
and 2 in the control group had constant pain, which was not
Duration of pain (min/day)
statistically significant between the 2 groups.
At baseline 22.5±50.8 20.5±52.6 0.25
Four weeks after the start of the treatment, 31 patients in the After 4 weeks 5.6±15.2 15.4±42.4 <0.001
case group and 50 in the control group had colic pain, and 2 in Dysfunction, presence:absence
each group had continuous pain. The pain pattern following At baseline 13:52 8:52 0.34
4 weeks was significantly different between the 2 groups (P< After 4 weeks 1/64 6/54 0.054 (fisher)
0.001). Dysfunction (day/wk)
Duration of the pain in the 2 groups was 50.8±22.5 and At baseline 0.5±1.2 0.2±0.7 0.25
52.6±20.5 min/day in the case and control groups, respectively, After 4 weeks 0.05±0.3 0.2±0.6 0.04
which was not statistically significant (P=0.25); while the dura­ Use another drug, used:not used
tion of pain after 4 weeks of the treatment in these 2 groups were At baseline 2/63 0/60 0.49
15.2±5.6 and 42.4±15.4, respectively, which was statistically After 4 weeks 1/64 0/60 1.00 (Fisher)
significant (P<0.001). Use another drug (day/wk)
Thirteen patients in the case group and 8 patients in the con­ At baseline 0.1±0.6 0 0.17
trol group were reported to present disturbances in their daily After 4 weeks 0.03±0.2 0 0.3
activities before treatment, which was not significantly different Initial diagnosis
(P=0.34), whereas, 4 weeks following the treatment, 1 patient Functional abdominal pain 28 (43) 29 (48) 0.59
in the case group and 6 in the control group continued to have Functional dyspepsia 16 (24) 13 (21) 0.83
disturbed daily activities. This difference was not significant Irritable bowel syndrome 15 (23) 15 (25) 0.83
between the 2 groups (P=0.054). Patients in the pretreatment Abdominal migraine 6 (9) 3 (5) 0.49
group had an average daily activity of 1.2±0.5 days per week, Treatment success 32 (49) 8 (13) <0.001
which was 0.7±0.2 in the control group. The difference was Values are presented as mean±standard deviation or number (%).
P values were derived using the Mann-Whitney, Wilcoxon, chi-square, or
not statistically significant (P=0.25). Following the 4 weeks of Fisher exact test.
the treatment the amount of distraction in daily activities in the Boldface indicates a statistically significant difference with P<0.05.

www.e-cep.org https://doi.org/10.3345/cep.2019.01613 487

 1. Functional abdominal pain verity, and duration of pain were significantly reduced in treat­
Of 57 children presented with FAP, 28 children received ment group, as compared to placebo (P<0.001, P<0.001, and
probiotics, and 29 received placebo. Overall, treatment was P<0.001, respectively) (Table 4).
successful in 13 patients in probiotic group and 8 patients in
placebo group (P=0.17). Frequency and duration of pain were 4. Abdominal migraine
not significantly reduced in probiotic group after 4 weeks of the Of 9 patients presented with AM, 6 children were in treatment
treatment (P=0.051 and P=0.054, respectively). Severity of pain group and 3 children were placebo. Overall, treatment was
was significantly different in the 2 groups (P<0.001) (Table 2). successful in 2 patients in probiotic group and none in placebo
group, which was not statistically different (P=0.5). Similarly,
2. Functional dyspepsia after 4 weeks, frequency, duration, and severity of pain were also
Of 29 patients presented with FD, 16 children received pro­ not statistically significant in the 2 groups (P>0.05) (Table 5).
biotic and 13 received placebo. Overall, 11 children (68%)
showed a significant response to probiotic treatment, compared
to the placebo group (P<0.001). Duration of pain, severity, and Discussion
frequency were significantly reduced in the treatment group
(P<0.001, P<0.001, and P<0.001, respectively) (Table 3). Findings from this study reveal that probiotic treatment (L.
reuteri) significantly improved the intensity and the duration of
3. Irritable bowel syndrome RAP. All pain-related characteristics, such as the frequency of
Of 30 children presented with IBS, 15 were in probiotic group days of pain, the severity of pain, the duration of pain, and its
and 15 were in placebo group. Six patients were successfully pattern in the case group, were significantly reduced upon the
treated in the treatment group, while none of the patients in treatment in comparison with the control group following the 4
placebo group showed improvement (P=0.01). Frequency, se­ weeks of the treatment. Also, the number of days of disturbance
in daily activities, such as the absence of children from the school,
Table 2. Characteristics of patients with the initial diagnosis of
functional abdominal pain by study group Table 3. Characteristics of patients with the initial diagnosis of
Case Control functional dyspepsia by study group
Characteristic P value
group group Characteristic Probiotic Placebo P value
Treatment success 13 (46) 8 (27) 0.17 Treatment success 11 (68) 0 (0) <0.001
Frequency of pain Frequency of pain
At the baseline 10.4±7.0 7.4±5.1 0.13 At the baseline 6.7±4.7 9.7±7.3 0.28
After 4 weeks 2.1±2.7 4.1±4.4 0.051 After 4 weeks 1.6±3.0 6.0±5.0 0.00
Severity of pain Severity of pain
At the baseline 3.5±1.0 3.1±0.5 0.14 At the baseline 3±0.8 3±0.4 0.65
After 4 weeks 1.1±1.3 2.0±1.0 <0.001 After 4 weeks 0.8±1.5 2.0±0.6 <0.001
Pattern of pain, colic:continiuse Pattern of pain, colic:continiuse
At baseline 26:2 28:1 0.61 At baseline 15:1 12:1 1.00
After 4 weeks 13:2 20:1 0.22 After 4 weeks 5:0 12:1 <0.001
Duration of pain (min/day) Duration of pain (min/day)
At baseline 36±75.6 21.3±54 0.16 At baseline 14.5±10.8 32.8±80.3 0.94
After 4 weeks 5.6±15.2 15.4±42.4 0.054 After 4 weeks 2.3±4.4 24.9±64.7 <0.001
Dysfunction, presence:absence Dysfunction, presence:absence
At baseline 7:21 3:26 0.17 At baseline 1:15 1:12 1.00
After 4 weeks 0:28 1:28 1.00 (Fischer) After 4 weeks 0:16 1:12 0.44
Dysfunction (day/wk) Dysfunction (day/wk) 0.1
At baseline 0.6±1.4 0.2±0.6 0.14 At baseline 0.2±0.5 0.08±0.2 1.00
After 4 weeks 0 0.07±0.3 0.32 After 4 weeks 0 0.08±0.2 0.74
Use another drug, used:not used Use another drug, used:not used
At baseline 1:27 0:29 0.49 At baseline 1:15 0:13 Invalid
After 4 weeks 0:28 0:29 Invalid After 4 weeks 0 0 Invalid
Use another drug (day/wk) Use another drug (day/wk)
At baseline 0.1±0.9 0 0.31 At baseline 0.1±0.5 0 0.77
After 4 weeks 0 0 Invalid After 4 weeks 0 0 Invalid
Values are presented as number (%) or mean±standard deviation. Values are presented as number (%) or mean±standard deviation.
P values were derived using the Mann-Whitney, Wilcoxon, chi-square, or P values were derived using the Mann-Whitney, Wilcoxon, chi-square, or
Fisher exact test. Fisher exact test.
Boldface indicates a statistically significant difference with P <0.05. Boldface indicates a statistically significant difference with P <0.05.

488 Rahmani P, et al. Probiotics and recurrent abdominal pain www.e-cep.org

Table 4. Characteristics of patients with the initial diagnosis of Table 5. Characteristics of patients with the initial diagnosis of
irritable bowel syndrome by study group abdominal migraine by study group
Characteristic Probiotic Placebo P value Characteristic Case group Control group P value
Treatment success 6 (40) 0 (0) 0.01 Treatment success 2 (50) 0 (0) 0.5
Frequency of pain Frequency of pain
At the baseline 9.0±5.3 8.3±6.1 0.77 At the baseline 4±2 1.3±0.5 0.02
After 4 weeks 3.7±5.5 6.3±4.5 0.01 After 4 weeks 1.1 ± 0.9 1.3±0.5 1.00
Severity of pain Severity of pain
At the baseline 3.4±0.8 3.2±0.8 0.46 At the baseline 3.1±0.9 2.6±0.5 0.54
After 4 weeks 1.4±1.4 2.8±0.8 0.01 After 4 weeks 1.3±1.5 2.3±0.5 0.26
Pattern of pain, colic:continiuse Pattern of pain, colic:continiuse
At baseline 15:0 15:0 Invalid At baseline 6:0 3:0 invalid
After 4 weeks 9:6 15:0 0.01 After 4 weeks 4:0 3:0 0.5
Duration of pain (min/day) Duration of pain (min/day)
At baseline 10.2±4.6 10.6±4.9 0.87 At baseline 11.4±6.0 8.3±2.6 0.26
After 4 weeks 2.9±4.0 10.6±4.9 <0.001 After 4 weeks 11±24 8.3±2.8 0.16
Dysfunction, presence:absence Dysfunction, presence:absence
At baseline 3:12 4:11 1.00 At baseline 2:4 0:3 0.4
After 4 weeks 1:14 4:11 0.33 After 4 weeks 0:6 0:3 invalid
Dysfunction (day/wk) 0.6 Dysfunction (day/wk)
At baseline 0.7±1.3 0.6±1.1 0.87 At baseline 0.8±1.3 0 0.54
After 4 weeks 0.2±0.7 0.6±1.1 0.38 After 4 weeks 0 0 1.00
Use another drug, used:not used Use another drug, used:not used
At baseline 0 0 Invalid At baseline 0:6 0:3 invalid
After 4 weeks 0 0 Invalid After 4 weeks 1:5 0:3 0.66
Use another drug (day/wk) Use another drug (day/wk)
At baseline 0 0 Invalid At baseline 0 0 Invalid
After 4 weeks 0 0 Invalid After 4 weeks 0.3±0.8 0 0.74
Values are presented as number (%) or mean±standard deviation. Values are presented as number (%) or mean±standard deviation.
P values were derived using the Mann-Whitney, Wilcoxon, chi-square, or P values were derived using the Mann-Whitney, Wilcoxon, chi-square, or
Fisher exact test. Fisher exact test.
Boldface indicates a statistically significant difference with P <0.05. Boldface indicates a statistically significant difference with P <0.05.

was also reduced in the treatment group.
A recent study conducted by Royan et al.10) in Iran reported
that different strains of L. reuteri extracted from poultry ducks
have immune-protective role in boiler chicken and balance serum
lipid levels. Furthermore, Liu et al.11) reported that L. reuteri
strains DSM 17938 and ATCC PTA 4659 have therapeutic
efficacy against necrotizing enterocolitis as they promote anti-
inflammatory response by upregulating the production of IL-10
and reducing the expression of inflammatory cytokines such
as; IL-6, tumor necrosis factor-alpha, toll-like receptor 4, and
nuclear factor kappa-light-chain-enhancer of activated B cell.
A recent Cochrane review update has shown low-to-moderate
evidences based on the findings that probiotic treatment of RAP
in children is associated with a significant reduction in the fre­
quency and the intensity of the pain. This improvement was
reported greatest in the children presented with IBS. Further­
more, Martens et al.12) reported that Symbioflor2 containing E.
coli significantly improves the symptoms of IBS among child­ren
aged 4–18 years.
In a meta-analysis, Lactobacillus rhamnosus GG (LGG) have
been reported to reduce the intensity of the pain in children
presenting pain-associated functional gastrointestinal diseases.13)
Three clinical trials with a total population of 290 patients
were evaluated in the study. The results of this study showed
success in treatment and a moderate reduction in the severity
and frequency of the pain and the frequency of pain days in
IBS patients. Since there are no established reports on the use
of certain drugs in RAP pain management, this study reported
different criteria for the treatment of RAP.14)
Our study results were similar to those of Gawrońska et
al.15) concerning the variables studied, the type of evaluation of
pain-related functional disorders and the definition of pain se­
verity; nevertheless, in this study, there was a relative success in
probiotic-based treatment without significant improvement in
recovery. Differences between the strains of bacteria used is a
possible explanation of the discrepancies.
Romano et al.16) L. reuteri DSM 17938 probiotic reduce the
intensity of the pain among 6–16 years old children presented
with functional abdominal pain.
L. reuteri is a lactic acid bacteria that has been shown to pro­
mote health by improving the movement, function, and modula­
tion of pain signals via the neurotransmitter of the middle cere­
bral cortex.17,18) It has been shown that L. reuteri has a high
tendency to colonize in gastrointestinal mucus.19)
A clinical trial study in 2010 by Savino et al.20) was conducted
to evaluate the effect of L. reuteri DSM 17938 for the treatment

www.e-cep.org https://doi.org/10.3345/cep.2019.01613 489

 of neonatal colic. Our study is the first clinical trial to report high-
dose L. reuteri treatment for abdominal incontinence in children.
This result was obtained not only in our study population with
abdominal remission but also in each subgroup including FAP,
FD, and IBS. In abdominal migraine, the difference between the
studied variables in the case and control groups was not signi­
ficant, as one of the causes was the low number of patients in this
subgroup (9 subjects).
In conclusion, owing to the facts that RAP is one of the com­
mon disorders in children, especially at school-going age, and
can disrupt daily activities, leading to fewer children attending
in the community activities, and also adding to anxiety for the
parents along with socioeconomic pressure, therapeutic mea­
sures to for RAP stand great demand. Based on the results of this
study, L. reuteri probiotic treatment significantly improves the
incidence of abdominal pain in children. Considering the results
of this study and its significant effects, especially on IBS and FD,
the use of probiotic L. reuteri can be recommended as one of the
RAP therapies.
Given that there is no known mechanism of action for L. reuteri
that makes L. reuteri superior to LGG and other probiotics,
detailed and comparative case-control trials are recommended to
draw a clinically advantageous conclusion.
Further evidence-based studies on the therapeutic use of
probiotics in patients with a larger sample size along with their
dosage and the duration of the treatment are also recommended.
Conflicts of interest
No potential conflict of interest relevant to this article was
reported.
Acknowledgments
We greatly thank "Tehran University of Medical Sciences,
Tehran, Iran" for their cooperation and support.
References
1. Rahmani P, Moradzadeh A, Farahmand F. Giving probiotics to your
children for gastrointestinal problems: In the light of scientific findings.
PharmaNutrition 2019;10:100164.
2. de Moreno de LeBlanc A, Del Carmen S, Chatel JM, Miyoshi A, Azevedo
V, Langella P, et al. Current review of genetically modified lactic acid
bacteria for the prevention and treatment of colitis using murine models.
Gastroenterol Res Pract 2015;2015:146972.
3. Peeters B. Gastrointestinal motility disorders in children: etiology and
associated behaviors [dissertation]. Amsterdam: University of Amsterdam,
2013.
490 Rahmani P, et al. Probiotics and recurrent abdominal pain
4. Salem AE, Singh R, Ayoub YK, Khairy AM, Mullin GE. The gut micro­
biome and irritable bowel syndrome: State of art review. Arab J Gastro­
enterol 2018;19:136-41.
5. Marcdante K, Kliegman RM. Nelson essentials of pediatrics e-book: with
student consult online access: Philadelphia (PA): Saunders/Elsevier, 2014.
6. Newlove-Delgado TV, Martin AE, Abbott RA, Bethel A, Thompson-
Coon J, Whear R, et al. Dietary interventions for recurrent abdominal
pain in childhood. Cochrane Database Syst Rev 2017;3:CD010972.
7. Bufler P, Gross M, Uhlig HH. Recurrent abdominal pain in childhood.
Dtsch Arztebl Int 2011;108:295-304.
8. Rose MA. Lactobacillus rhamnosus GG reduces frequency and severity of
abdominal pain compared with placebo in children with irritable bowel
syndrome. Evid Based Med 2011;16:141-2.
9. Shih DQ, Kwan LY. All roads lead to Rome: update on Rome III criteria
and new treatment options. Gastroenterol Rep 2007;1:56-65.
10. Royan M, Hashemi M, Seighalani R. Effect of isolates of lactobacillus
reuteri and lactobacillus salivarius isolated from the gastrointestinal tract
of native poultry of northern of iran on performance, serum lipids and
immune parameters of broiler chickens. Res Anim Prod 2019;10:18-26
11. Liu Y, Fatheree NY, Mangalat N, Rhoads JM. Lactobacillus reuteri strains
reduce incidence and severity of experimental necrotizing enterocolitis via
modulation of TLR4 and NF-κB signaling in the intestine. Am J Physiol
Gastrointest Liver Physiol 2012;302:G608-17.
12. Martens U, Enck P, Zieseniss E. Probiotic treatment of irritable bowel
syndrome in children. Ger Med Sci 2010;8:Doc07.
13. Horvath A, Dziechciarz P, Szajewska H. Meta-analysis: Lactobacillus
rhamnosus GG for abdominal pain-related functional gastrointestinal
disorders in childhood. Aliment Pharmacol Ther 2011;33:1302-10.
14. Francavilla R, Miniello V, Magistà AM, De Canio A, Bucci N, Gagliardi F,
et al. A randomized controlled trial of Lactobacillus GG in children with
functional abdominal pain. Pediatrics 2010;126:e1445-52.
15. Gawrońska A, Dziechciarz P, Horvath A, Szajewska H. A randomized
double-blind placebo-controlled trial of Lactobacillus GG for abdominal
pain disorders in children. Aliment Pharmacol Ther 2007;25:177-84.
16. Romano C, Ferrau' V, Cavataio F, Iacono G, Spina M, Lionetti E, et al.
Lactobacillus reuteri in children with functional abdominal pain (FAP). J
Paediatr Child Health 2014;50:E68-71.
17. Wang B, Mao YK, Diorio C, Wang L, Huizinga JD, Bienenstock J, et al.
Lactobacillus reuteri ingestion and IK(Ca) channel blockade have similar
effects on rat colon motility and myenteric neurones. Neurogastroenterol
Motil 2010;22:98-107, e33.
18. Indrio F, Riezzo G, Raimondi F, Bisceglia M, Cavallo L, Francavilla R.
The effects of probiotics on feeding tolerance, bowel habits, and gastro­
intestinal motility in preterm newborns. J Pediatr 2008;152:801-6.
19. Valeur N, Engel P, Carbajal N, Connolly E, Ladefoged K. Colonization
and immunomodulation by Lactobacillus reuteri ATCC 55730 in the
human gastrointestinal tract. Appl Environ Microbiol 2004;70:1176-81.
20. Savino F, Cordisco L, Tarasco V, Palumeri E, Calabrese R, Oggero R, et al.
Lactobacillus reuteri DSM 17938 in infantile colic: a randomized, double-
blind, placebo-controlled trial. Pediatrics 2010;126:e526-33.
How to cite this article: Rahmani P, Ghouran-orimi A,
Motamed F, Moradzadeh A. Evaluating the effects of probiotics
in pediatrics with recurrent abdominal pain. Clin Exp Pediatr
2020;63:485-90. https://doi.org/10.3345/cep2019.01613
www.e-cep.org
Pediatrics and Neonatology (2019) 60, 285e290

Available online at www.sciencedirect.com

ScienceDirect

journal homepage: http://www.pediatr-neonatol.com

Original Article

Early predictors of neonatal

hyperbilirubinemia in full term newborn
May Ahmed Khairy a, Walaa Alsharany Abuelhamd a,*,
Ismail Mohamed Elhawary a, Ahmed Said Mahmoud Nabayel b

a
Department of Pediatrics, Faculty of Medicine, Cairo University, Cairo, Egypt
b
Department of Pediatrics, Ministry of Health, Cairo, Egypt

Received Jan 13, 2018; received in revised form Mar 5, 2018; accepted Jul 18, 2018
Available online 26 July 2018

Key Words Background: Reliable predictive markers enabling physicians to identify which newborns will
cord bilirubin/ develop significant hyperbilirubinemia have become mandatory for prevention of severe hy-
albumin ratio; perbilirunemia. We aimed at determining the critical cord serum bilirubin and albumin levels
cord serum albumin; and bilirubin/albumin ratio early as reliable markers.
cord serum bilirubin; Study design: This prospective study included 175 full-term neonates. Measurement of cord
neonatal bilirubin, albumin and bilirubin/albumin ratio was done to predict significant hyperbilirubine-
hyperbilirubinemia mia in healthy term newborns based on serum bilirubin measurements made within 5 days of
life.
Results: Most cases that developed significant neonatal hyperbilirubinemia (67.9%) had cord
albumin level  2.8 gm/dl. Cord Bilirubin/albumin ratio cut off value > 0.61 had a good pre-
dictive value with a sensitivity of 100% and specificity of 88.4%, and cord serum albumin cut off
value  3.0 mg/dl also had a good predictive value with a sensitivity of 85.7% and specificity of
67.3%.
ROC curve analysis of cord total bilirubin demonstrated that a cut off value of 1.84 mg/dl
had a good predictive value with a sensitivity of 100.0% and specificity of 87.1%.
Conclusion: Cord bilirubin/albumin ratio, serum bilirubin and albumin could be early predic-
tors for neonatal hyperbilirubinemia.
Copyright ª 2018, Taiwan Pediatric Association. Published by Elsevier Taiwan LLC. This is an
open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/
by-nc-nd/4.0/).

* Corresponding author. Pediatric Department, Faculty of Medicine, Cairo University, New Children Hospital, (Abu El Rish), Cairo University
Hospitals, Ali Basha Ebrahim, PO Box 11562, Cairo, Egypt.
E-mail addresses: maykhairy@yahoo.com (M.A. Khairy), walaa_alsharany@hotmail.com (W.A. Abuelhamd), Elhawary_20@yahoo.com
(I.M. Elhawary), DrAhmedSaid87@gmail.com (A.S. Mahmoud Nabayel).

https://doi.org/10.1016/j.pedneo.2018.07.005
1875-9572/Copyright ª 2018, Taiwan Pediatric Association. Published by Elsevier Taiwan LLC. This is an open access article under the CC BY-
NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
286
1. Introduction
Severe hyperbilirubinemia can occur without apparent
reason in healthy infants, and some may develop ker-
nicterus.1 The prevention of poor outcomes necessitates
early detection of neonates who are at risk of developing
significant hyperbilirubinemia.
Early discharge of healthy term newborns has become a
common practice because of advantages including preven-
tion of nosocomial infections, encouragement of early
maternal-infant bonding and also lower cost. The American
Academic of Pediatrics (AAP) recommends that newborns
discharged within 48 h should have a follow-up visit after
48e72 h for any significant jaundice or other problems.2 In
developing countries, the value of follow-up visits after
early discharge is questionable as many mothers do not
return owing to the distance they need to travel.
Reliable predictive markers enabling physicians to
identify which of the newborns discharged early are at
increased risk for significant hyperbilirubinemia has
become mandatory. Several studies have been performed
to assess the ability of cord bilirubin and albumin and first-
day bilirubin levels to be tools for screening of subsequent
neonatal hyperbilirubinemia.3e6
The bilirubin/albumin (B/A) ratio is considered a surro-
gate parameter for free bilirubin (Bf) and an interesting
additional parameter in the management of hyper-
bilirubinemia.7 It offers the clinician a reasonable measure
of bilirubin binding to albumin until unbound bilirubin or
albumin binding reserve can be measured clinically with
accuracy and precision.8
To our knowledge, there are no documented studies
evaluating the significance of cord B/A ratio in predicting
neonatal hyperbilirubinemia. In the present study we aimed
at determining the critical cord serum bilirubin and albumin
level and bilirubin/albumin ratio that predict significant
hyperbilirubinemia in healthy term newborns based on
serum bilirubin measurements made within 5 days of life.
2. Materials and methods
This is a prospective cohort study that included 175 neo-
nates born in Cairo University Hospital, from July 2015 to
June 2016. All neonates involved were full term (gesta-
tional age ranging from 37 to 41 weeks) of either gender
without any significant illness or major congenital malfor-
mations. The neonates with conditions that could aggravate
hyperbilirubinemia (sepsis, respiratory distress syndrome,
asphyxia, diabetic mothers, or intrauterine growth retar-
dation) or cholestatic jaundice were excluded from the
study.
Participants were subjected to detailed history-taking
including maternal medical diseases, consanguinity, siblings
by hyperbilirubinemia, mode of delivery, Apgar score,
oxytocin use, and type of feeding. Thorough physical ex-
amination of neonates was done with assessment of
gestational age by new Ballard score and birth weight by
growth curves. The use of phototherapy or exchange
transfusion were recorded as indicated by the American
Academy of Pediatrics guidelines for management of
neonatal hyperbilirubinemia.2
M.A. Khairy et al
In the delivery room, 3e5 ml of cord venous blood
were collected after clamping the umbilical cord with 2
clamps, the second placed 4e6 inches away from the
first. The collected sample was tested for serum albumin
level using bromocresol green method (BCG), serum bili-
rubin (total and direct) level by Colormetric method,
hemoglobin concentration using cell counter T 660, and
reticulocytic count done manually after staining with
brilliant cresyl blue and examined under oil emersion
lens. Blood groups (ABO, Rh) were determined for new-
borns and mothers.
Follow up was done on days one, three and five of life by
assessment of serum bilirubin level for all cases. Significant
hyperbilirubinemia was defined as the need of photo-
therapy or exchange transfusion based on the American
Academy of Pediatrics guidelines for management of
neonatal hyperbilirubinemia.2
2.1. Sample size
Sample size calculation was based on the sensitivity of cord
blood albumin and bilirubin in predicting the occurrence of
indirect neonatal hyperbilirubinemia. We planned to study
the independent cases and controls with 1 control(s) per
case. Prior data indicated that the average sensitivity of
cord blood albumin in predicting later occurrence of
neonatal hyperbilirubinemia was approximately 77%3 while
that of cord blood bilirubin was 85%.4 If the true sensitivity
differed by 10%, we needed to study 21 cases and 21 control
subjects to be able to reject the null hypothesis for cord
blood albumin and we needed to study 18 cases and 18
control subjects to be able to reject the null hypothesis for
cord blood bilirubin with 80% power. Thus, we took cord
blood samples from all newborns until we had a minimum of
21 cases with significant indirect neonatal hyper-
bilirubinemia. The remaining samples without significant
indirect neonatal hyperbilirubinemia were considered
controls with a minimum of 21 babies. We used uncorrected
chi-squared statistics to evaluate this null hypothesis with
setting type I error probability to 0.05. Calculations were
done using Flahault equation.9
2.2. Statistical analysis
Data were entered on the computer using Microsoft Office
Excel Software program (2010) for Windows, then trans-
ferred to the Statistical Package of Social Science Soft-
ware (SPSS) program, version 23 to be statistically
analyzed. Data were summarized using range, mean,
standard deviation, median and percentiles for quantita-
tive variables or frequency and percentage for qualitative
ones. Comparison between groups was performed using
ManneWhitney test for quantitative variables while com-
parison for qualitative variables was performed through
Chi square or Fisher’s exact test. Receiver operating
characteristics (ROC) curve analysis was performed to
explore the discriminant ability of different cord measures
in predicting neonatal jaundice. P values less than 0.05
were considered statistically significant. Graphs were used
to illustrate some information.
Predictors of neonatal hyperbilirubinemia 287

3. Results Total serum bilirubin levels measured on days 1, 3 and

5 were significantly higher (P < 0.001) in group 1
The study population included 175 neonates with a mean [(8.8  1.5 mg/dl versus 4.1  1.2 mg/dl)], [(15.4  2.5 mg/dl
gestational age of 37.9 (0.9) weeks and a mean birth versus 6.7  2.8 mg/dl)] and [(13.8  5.1 mg/dl versus
weight of 2.9 (0.3) kg. Of these, 28 neonates (16%) 4.4  2.4 mg/dl)] respectively. However, the hemoglobin
developed significant hyperbilirubinemia (group 1) and 147 concentration showed no significant difference between
neonates (84%) did not (group 2). groups.
Among the 28 neonates who developed significant The cases with significant hyperbilirubinemia were all
neonatal hyperbilirubinemia, 16 were males and 12 were managed with phototherapy whether intensive (82.1%) or
females; 21 cases were delivered by caesarean section and conventional (17.9%). Intravenous immunoglobulin (IVIG)
6 cases received oxytocin for induction of labour. As regards was indicated in 14.3% of the cases and none required ex-
the 147 neonates who did not develop significant neonatal change transfusion.
hyperbilirubinemia, 73 were males and 74 were females; Among the neonates that developed significant hyper-
105 cases were delivered by caesarean section and 29 cases bilirubinemia, 67.9% had low cord serum albumin <2.8 mg/dl,
received oxytocin for induction of labour. 25% had it ranging between 2.8 and 3.3 mg/dl and only 7% had
Group 1 infants had statistically significant higher cord a level over 3.3 mg/dl.
reticulocytic count [(3.4  1.3%) versus (2.1  0.8%)] than For the prediction of significant neonatal hyper-
those in group 2 (p < 0.001) with a diagnosis of Rh in- bilirubinemia, a cut off value of cord serum bilirubin of
compatibility in 10.7% of the cases (3 patients out of 28) 1.84 mg/dl was chosen on the basis of the ROC curve
and ABO incompatibility in 14.3% of the cases (4 out of 28 analysis. The area under the curve was 0.95 indicating high
patients) in group 1 versus 1.4% and 4.1% respectively in significance. The cord serum bilirubin of 1.84 had a sensi-
group 2. tivity of 100%, specificity of 87.1%, positive predictive value
Cases with significant neonatal hyperbilirubinemia (group of 59.6% and negative predictive value of 100% in the pre-
1) had statistically significantly higher cord total bilirubin diction of neonatal hyperbilirubinemia (Fig. 1) (Table 2).
[(2.4  0.2 mg/dl) versus (1.4  0.4 mg/dl)] (p < 0.001), The optimum cut off value for cord serum albumin as
significantly lower cord albumin [(2.8  0.3 gm/dl) versus shown by the ROC curve for neonates with significant in-
(3.3  0.5 gm/dl)] (p < 0.001), and significantly higher cord direct hyperbilirubinemia was 3 gm/dl with a sensitivity of
B/A ratio [(0.86  0.14) versus (0.44  0.19)] with p < 0.001 85.7% and specificity of 67.3%, negative predictive value of
(Table 1). 96.1% and positive predictive value of 33.3% (area under
the curve of 0.825) (Fig. 2) (Table 2). Also, the cord B/A
ratio cut off value of >0.61 had a good predictive value for
neonates that developed significant neonatal hyper-
Table 1 Comparison of laboratory data of cases with
bilirubinemia with a sensitivity of 100%, specificity of
significant and insignificant neonatal hyperbilirubinemia.
88.4%, positive predictive value of 62.2% and negative
Group 1 Group 2 P value predictive value of 100.0%. The area under the curve is
(n Z 28) (n Z 147) 0.936 indicating high significance (Fig. 3) (Table 2).
Cord hemoglobin (gm/dl)
Range 15.3 18.5 14 18.2 0.070
Mean  SD 16.5  1 16  0.9 4. Discussion
Median 16.3 16.2
Cord reticulocytes The need for early prediction of jaundice has become
Range 1 7 1 4 <0.001# increasingly important for identifying those babies at risk of
Mean  SD 3.4  1.3 2.1  0.8 neonatal hyperbilirubinemia considering the severe neuro-
Median 3 2 logical morbidities caused by bilirubin toxicity.
Cord total bilirubin (mg/dl) The current AAP guidelines for managing healthy jaun-
Range 1.9 3 0.6 2.6 <0.001# diced term and near-term newborns recommends the use of
Mean  SD 2.4  0.2 1.4  0.4 the total bilirubin concentration/albumin ratio in addition
Median 2.4 1.3 to the TBC; however, it has not been widely used by clini-
Cord direct bilirubin (mg/dl) cians.2 In the current study the mean cord B/A ratio was
Range 0.2 0.7 0.1 0.5 0.029# significantly higher in neonates who developed neonatal
Mean  SD 0.3  0.1 0.3  0.1 hyperbilirubinemia than in those who did not (0.86  0.14
Median 0.3 0.2 versus 0.44  0.19). ROC curve analysis of cord B/A ratio
Cord albumin (gm/dl) demonstrated that a cut off value  0.61 mg/dl had a good
Range 2.4 3.6 1.8 4.8 <0.001# predictive value with a sensitivity of 100.0%, specificity of
Mean  SD 2.8  0.3 3.3  0.5 88.4%, and positive predictive value of 62.2%.
Median 2.7 3.3 Ahlfors et al. suggested that, in the absence of an
Cord bilirubin/Albumin ratio available assay for free bilirubin (Bf), the bilirubin/albumin
Range 0.62 1.2 0.14 1 <0.001# ratio B/A might provide a better estimate of Bf because it
Mean  SD 0.86  0.14 0.44  0.19 contains 2 of the 3 factors determining Bf (TSB, albumin
Median 0.88 0.4 and the albumin binding affinity).10 However, the value of
B/A ratio may be reduced because of some factors that
SD Z standard deviation, significant P value < 0.05.
influence the intrinsic albumin-bilirubin binding constant (it
288 M.A. Khairy et al

Figure 1 ROC curve analysis to explore the discriminant ability of cord total bilirubin in predicating significant
hyperbilirubinemia.

predicting the risk of neonatal hyperbilirubinemia.5 Knüp-

Table 2 Comparison between ROC curves.
fer et al. reported that a cord bilirubin cut off level of
Cord total Cord albumin Cord B/A Ratio 1.76 mg/dl for predicting hyperbilirubinemia had a sensi-
bilirubin tivity of 70.3% and a negative predictive value of 65.6% and
Sensitivity 100.0% 85.7% 100.0% they concluded that cord blood bilirubin could be used as
Specificity 87.1% 67.3% 88.4% an early predictor of neonatal jaundice.14 Dwarampudi and
PPV 59.6% 33.3% 62.2% Ramakrishna reported that neonates with cord bilirubin
NPV 100.0% 96.1% 100.0% level less than 2 mg/dl were in a safe zone with respect to
Cut-off 1.84 3.0 0.61 development of subsequent hyperbilirubinemia.15
P value <0.001 <0.001 <0.001 Albumin helps in hepatic transportation of bilirubin and its
clearance. Low serum albumin level decreases bilirubin
PPV Z positive predictive value, NPV Z negative predictive
clearance and thus increases significant hyper-
value, significant P value < 0.05.
bilirubinemia.16 The ability of cord albumin to act as a tool for
predicting neonatal jaundice was assessed in the current
may be decreased by drugs (e.g, ceftriaxone)7 and the study. There was a significantly higher cord serum albumin
presence of other plasma constituents that bind unconju- level in neonates who did not develop neonatal hyper-
gated bilirubin (as apolipoproteins and alfa fetoprotein).11 bilirubinemia in comparison to those who did. 67.9% of cases
The current study demonstrated a highly significant that developed significant neonatal hyperbilirubinemia had
difference between the cord serum bilirubin levels in the 2 cord albumin level 2.8 gm/dl. Burtis et al.17 stated that the
studied groups. They were lower in neonates who did not lower normal limit for cord serum albumin in term babies was
develop significant hyperbilirubinemia (1.4  0.4 mg/dl) 2.8 gm/dl; and Reshad et al.18 found that in the term group,
versus those who developed it (2.4  0.2 mg/dl). This is 19 (61.2%) newborns with cord serum albumin <2.8 g/dl
consistent with Ipek et al. who found that the mean cord developed neonatal hyperbilirubinemia. Moreover, similar
serum bilirubin was also lower in babies who developed results were reported by several researchers3,4,19 that found
neonatal hyperbilirubinemia versus those who did not cases with low cord albumin <2.8 gm/dl developed more
(1.64  0.41 mg/dl versus 2.05  0.9 mg/dl).12 Several significant hyperbilirubinemia requiring phototherapy and
studies reported lower mean cord serum bilirubin levels in exchange transfusion.
neonates who developed hyperbilirubinemia.3,4,13 Receiver operating characteristics (ROC curve) analysis
ROC curve analysis of cord total bilirubin demonstrated demonstrated that cord serum albumin cut off value  3.0 mg/
that a cut off value  1.84 mg/dl had a good predictive dl had a good predictive value with a sensitivity of 85.7% and
value with a sensitivity of 100.0%, specificity of 87.1%, and specificity of 67.3%. However, Rajpurohit et al. reported a
positive predictive value of 59.6%. Rajpurohit et al. re- lower cord blood albumin level (2.6 gm/dl) to have a sensi-
ported a cord blood bilirubin cut off value > 2 mg/dl had a tivity of 80% and specificity of 86.67% in predicting the risk of
sensitivity of 90%, specificity of 53.89%, positive predictive neonatal hyperbilirubinemia.5 Aiyappa and colleagues found
value of 17.8% and negative predictive value of 98% in the sensitivity of cord albumin to detect hyperbilirubinemia to
Predictors of neonatal hyperbilirubinemia 289

Figure 2 ROC curve analysis to explore the discriminant ability of cord albumin in predicating significant hyperbilirubinemia.

Figure 3 ROC curve analysis to explore the discriminant ability of B/A ratio (cord) in predicating significant hyperbilirubinemia.

be 71.8%, while specificity was 65.1%.6 Similarly, Pahuja et al. were probably safe to discharge a neonate in respect to the risk
stated a fair predictive value of cord albumin for development of development of neonatal hyperbilirubinemia.15
of neonatal hyperbilirubinemia of 75%.20 Also Dwarampudi and With lack of studies done on cord B/A ratio as an early
Ramakrishna suggested that cord albumin levels (>2.8 gm/dl) predictor of significant hyperbilirubinemia, this work opens
290
the window for further studies to be performed in this field
and we are aware that larger scale trials including preterm
neonates are needed.
In this study cord serum B/A ratio proved to predict the
development of significant neonatal hyperbilirubinemia. In-
fants with either cord serum total bilirubin 1.84 mg/dl, cord
serum albumin 3.0 gm/dl or cord serum B/A ratio 0.61,
were at risk of developing significant indirect neonatal hyper-
bilirubinemia needing interventions. These can be considered
possible early predictors for neonatal hyperbilirubinemia.
We recommend measurement of cord serum albumin,
albumin and bilirubin/albumin ratio in all healthy term
babies at delivery to prevent dangerous consequences of
hyperbilirubinemia as acute bilirubin encephalopathy.
Conflict of interest
Nil.
Ethical approval
The study was approved by the Ethics Committee of Pedi-
atric Department, Faculty of Medicine, Cairo University.
References
1. Ahire N, Sonawane R, Gaikwad R, Patil S, Sonawane T. Study of
correlation of cord blood bilirubin with neonatal hyper-
bilirubinemia. MVP J Med Sci 2016;3:60e6.
2. American Academy of Pediatrics Subcommittee on Hyper-
bilirubinemia. Management of hyperbilirubinemia in the
newborn infant 35 or more weeks of gestation. Pediatrics 2004;
114:297e316.
3. Trivedi DJ, Markande DM, Vidya BU, Bhat M, Hegde PR. Cord
serum bilirubin and albumin in neonatal hyperbilirubinemia.
Int J Int Sci Inn Tech Sec A 2013;2:39e42.
4. Venkatamurthy M, Murali SM, Mamatha S. A comparison study:
cord serum albumin is compared with cord serum bilirubin as a
risk indicator in predicting neonatal jaundice. JEMDS 2014;3:
4017e22.
5. Rajpurohit N, Kumar S, Sharma D, Choudhary M, Purohit S. To
assess predictive value of cord blood bilirubin and albumin for
significant neonatal hyperbilirubinemia: a prospective study
from India. J Pediatr Neonatal Care 2015;2:60.
M.A. Khairy et al
6. Aiyappa GKC, Shriyan A, Raj B. Cord blood albumin as a pre-
dictor of neonatal hyperbilirubinemia in healthy neonates. Int
J Contemp Pediatr 2017;4:503e6.
7. Hulzebos CV, van Imhoff DE, Bos AF, Ahlfors CE, Verkade HJ,
Dijk PH. Usefulness of the bilirubin/albumin ratio for predict-
ing bilirubin-induced neurotoxicity in premature infants. Arch
Dis Child Fetal Neonatal Ed 2008;93:F384e8.
8. Bhutani VK. Kernicterus as a ’never-event’: a newborn safety
standard? Indian J Pediatr 2005;72:53e6.
9. Flahault A, Cadilhac M, Thomas G. Sample size calculation
should be performed for design accuracy in diagnostic test
studies. J Clin Epidemiol 2005;58:859e62.
10. Ahlfors CE, Wennberg RP, Ostrow JD, Tiribelli C. Unbound
(free) bilirubin: improving the paradigm for evaluating
neonatal jaundice. Clin Chem 2009;55:1288e99.
11. Wennberg RP. The bloodebrain barrier and bilirubin encepha-
lopathy. Cell Mol Neurobiol 2000;20:97e109.
12. Ipek IO, Bozaykut A, Çagrıl SC, Sezer RG. Does cord blood
bilirubin level help the physician in the decision of early
postnatal discharge? J Matern Fetal Neonatal Med 2012;25:
1375e8.
13. Sun G, Wang YL, Liang JF, Du LZ. Predictive value of umbilical
cord blood bilirubin level for subsequent neonatal jaundice.
Zhonghua Er Ke Za Zhi 2007;45:848e52 [Article in Chinese].
14. Knüpfer M, Pulzer F, Gebauer C, Robel-Tillig E, Vogtmann C.
Predictive value of umbilical cord blood bilirubin for postnatal
hyperbilirubinaemia. Acta Paediatr 2005;94:581e7.
15. Dwarampudi GS, Ramakrishna N. Cord blood albumin and bili-
rubin levels as predictors in neonatal hyperbilirubinemia. Int J
Pharm Biol Sci 2015;6:273e9.
16. Meena KJ, Singh S, Verma RC, Sharma R. Utility of cord blood
albumin as a predictor of significant neonatal jaundice in
healthy term newborns. Ped Oncall 2015;12:66.
17. Burtis CA, Ashwood AR, Bruns DE. Tietz Textbook of clinical
chemistry and molecular diagnostics. 4th ed. Amsterdam:
Elsevier; 2008. p. 2254.
18. Reshad M, Ravichander B, Raghuraman TS. A study of cord
blood albumin as a predictor of significant neonatal hyper-
bilirubinemia in term and preterm neonates. Int J Res Med Sci
2016;4:887e90.
19. Sahu S, Abraham R, John J, Mathew AA, George AS. Cord blood
albumin as a predictor of neonatal jaundice. Int J Biol Med Res
2011;2:436e8.
20. Pahuja M, Dhawan S, Chaudhary SR. Correlation of cord blood
bilirubin and neonatal hyperbilirubinemia in healthy newborns.
Int J Contemp Pediatr 2016;3:926e30.
 International Journal of Pediatric Otorhinolaryngology 135 (2020) 110114

Contents lists available at ScienceDirect

International Journal of Pediatric Otorhinolaryngology

journal homepage: www.elsevier.com/locate/ijporl

2016 ESPO Congress

Effect of iron deficiency anemia on language development in preschool T

Egyptian children
Mervat A.M. Youssefa,∗, Eman S. Hassanb, Dalia G. Yasienc
a
- Pediatric Hematology Unit, Children Hospital, Faculty of Medicine, Assiut University, Assiut, Egypt
b
- Phoniatrics Unit, ENT Department, Assiut University Hospital, Faculty of Medicine, Assiut University, Assiut, Egypt
c
- Phoniatrics Unit, ENT Department, Helwan University Hospital, Faculty of Medicine, Helwan University, Cairo, Egypt

A R T I C LE I N FO A B S T R A C T

Keywords: Background: Iron deficiency anemia (IDA) is the most common nutritional deficiency primarily in developing
Language countries.
Iron deficiency anemia Objective: This study evaluates the effect of IDA on language development in preschool children.
Cognition Methodology: The study is a multicenter, comparative cross-sectional study included 226 children between ages
Egyptian children
4–6 years. The children were classified into two groups’ anemic (patients) and non anemic (controls) according
to the hemoglobin level. All anemic children subjected to complete iron study including; Serum iron, total iron
binding capacity (TIBC), Serum ferritin level, to confirm the diagnosis of iron deficiency anemia. Cognitive
assessment was done using the Arabic translation Stanford Binet intelligence scale, version four which comprised
of four cognitive area scores; visual reasoning, verbal reasoning, quantitative reasoning and short-term memory.
Measurement of IQ and mental age were calculated for each child. Language evaluation was done using the
Arabic Language test. Receptive language quotient, expressive language quotient and total language quotient
were calculated for each child.
Results: 122 children were anemic and 90 were non-anemic with hemoglobin level 10.65 and 11.96 g/dL, re-
spectively (P < 0.000). Anemic children had significantly lower serum ferritin (p < 0.0001), and serum iron
(p < 0.0001) compared to the controls. Both groups were matched as regards age, sex, socioeconomic levels
and parental educational level. No significant differences observed regarding IQ, mental age, receptive, ex-
pressive and total language quotients between anemic and non-anemic children.
Conclusions: IDA does not seem to have an effect on language development in preschool Egyptian Children.
Future large controlled studies with long follow-up time for the younger age group are needed to determine
whether there are existent associations between IDA with language development.

1. Introduction
Anemia is a disorder characterized by decreased number and
oxygen-carrying capacity of the red blood cells (RBCs). Universally,
Iron deficiency anemia (IDA) is the most common nutritional problem
affecting about 2 billion of the world's individuals; most of them (89%)
are in developing countries [1,2]. IDA affects about 300 million chil-
dren worldwide, aged from six months to five years [3,4]. In developing
nations, IDA is a common health problem affecting infants, per -school
and school children due to rapid growth rate combined with exhaustion
of iron storage, adverse living conditions and inadequate diets [5].
The diagnosis of IDA is usually established when the child is already
in a progressive stage of iron deficiency. Thus, it is critically important
to enhance iron storage and treat iron deficiency, specially in the first
two years of live, the sensitive period of rapid cerebral development, it
is also the period of greatest prevalence of iron deficiency anemia [6].
The contributing association between IDA and language and cog-
nitive development impairment cannot be established in the literatures
yet. Some studies considered IDA as a major risk factor for impaired
cognitive skills and delayed achievement of motor coordination.
Moreover, It affects language, behavior development, and the learning
process [7,8]. Additionally, some studies have recognized an associa-
tion between iron deficiency and delays in cognitive and psychomotor
development in early childhood, school children and adolescents
[8–11]. Anemic children over two years of age exhibit reduced cogni-
tive achievement when compared with non-anemic children, even
though there is an obvious improvement with proper treatment [12].
Iron deficiency decreases the iron-dependent enzymes activity that are

∗
Corresponding author. Children Hospital, hematology unit, Faculty of medicine, Assiut University, Egypt.
E-mail address: mamuosif2000@gmail.com (M.A.M. Youssef).

https://doi.org/10.1016/j.ijporl.2020.110114
Received 30 October 2019; Received in revised form 10 May 2020; Accepted 10 May 2020
Available online 13 May 2020
0165-5876/ © 2020 Elsevier B.V. All rights reserved.
M.A.M. Youssef, et al. International Journal of Pediatric Otorhinolaryngology 135 (2020) 110114

necessary for synthesis and function of neurotransmitters which may Table 1

lead to central nervous system myelination abnormalities [13,14]. Demographic characteristics of anemic and non anemic groups.
Furthermore, neurocognitive abnormalities may occur as a con- Non-anemic group Anemic group P value
sequence of iron deficiency, even in the absence of anemia [15]. On the (n = 90) (n = 122)
other hand, many authors reviewed the literatures and they could not
Sex
find a clear support for direct relation between IDA and cognitive de-
Males 56 (62.2%) 73 (59.8%) NS
velopment [16,17]. Even supplementation of iron in the preschool Females 34 (37.8%) 49 (40.2%) NS
found to have no valuable effects on the long term cognitive function Age in Years (Mean ± SD) 5.01 ± 0.50 4.94 ± 0.43 NS
[18,19]. Likewise, there are a limited data supporting the effect of IDA Hemoglobin (gm) 11.96 ± 0.4 10.65 ± 0.52 0.0001
on cognitive development in the preschool children [20].
Duration of breast- feeding 13.6 ± 3.35 14.13 ± 3.8 NS
Little is known about the effects of anemia on the language and
in months
cognitive development in Egyptian children. Therefore the objective of
Mother's education level in 11.7 ± 1.59 12.1 ± 2.05 NS
this study is to investigate the effects of IDA on language development
years
in preschool children.
A chi - square test used to compare between categorical variables and t-teat
2. Methods used to compare between continuous variables. P value < 0.05 is
significant.Ns, non significant.
2.1. Participants and data collection
Table 2
This study is a multicenter, comparative cross-sectional study in- Hemoglobin level and iron study for anemic and non -anemic children.
cluded 226 children (89 females, 137 males) enrolled from Helwan Non-anemic (n = 90) Anemic (n = 122) P value
university hospital, Faculty of Medicine, Helwan University, and Assiut
university hospital, Faculty of Medicine, Assiut University. Hb (g/dl) 11.96 ± 0.4 10.65 ± 0.52 0.0001
MCV(fl) 74.29 ± 2.76 62.2 ± 4.72 0.0001
The exclusion criteria included patients with congenital disorders or
MCH(pg) 26.01 ± 1.35 21.67 ± 1.95 0.0001
history of any disorder that may affect their cognition such as pre-, MCHC(g/dl) 32.97 ± 1.57 29.4 ± 1.7 0.001
peri-, post-natal hypoxia, motor speech disorders, emotional disorders, Serum ferritin (ng/ml) 76.73 ± 14.49 9.68 ± 1.89 < .0001
and neurological symptomatology. Additionally, patients’ chronic sys- Serum Iron (mg/dl) 74.07 ± 19.84 12.6 ± 4.89 < 0.0001
temic diseases or acute inflammatory processes that alter the he- TIBC(mcg/dl) 293.4 ± 34.66 499.02 ± 26.3 < 0.0001

moglobin level were excluded.

t-test used. P value < 0.05 indicated a significant difference. MCV: Mean
Informed consent was gained from the parents of the children before
corpuscular volume, MCH: Mean corpuscular hemoglobin MCHC: Mean cor-
enrollment in the study. All procedures fulfilled the Health and Human puscular hemoglobin, TIBC: Total iron-binding capacity.
Ethical Approval Committee guidelines for Clinical Research at Helwan
and Assiut University.
Table 3
All enrolled children were subjected to a detailed history including Cognitive assessment.
the neonatal, nutritional, socioeconomic, cultural level and parental
Non-anemic (n = 90) Anemic (n = 122) P. value
education. Also the children underwent general, neurological and
psychometric evaluation. I.Q 106.08 ± 10.41 105.19 ± 9.9 0.528
Mental Age (years) 5.1 ± 0.84 5.09 ± 0.84 0.528
2.2. Hematological assessment
P value < 0.05 indicated a significant difference.
Initially, venous blood samples were taken from the children in
order to determine hemoglobin levels (Hg). Children were defined as the child obtained in each test)/(the total score) × 100.
anemic if their hemoglobin level two standard deviations below the
mean for age [21]. According to the hemoglobin levels, the children 2.4. Statistics
were classified into two groups; anemic (patients) and non anemic
(controls). All anemic patients were subjected to Complete iron study The data were tested for normality using the Kolmogorov-Smirnov
under supervision of pediatric hematologist including; Serum iron, total test and for homogeneity variances prior to further statistical analysis.
iron binding capacity (TIBC), Serum ferritin level, to confirm the di- Categorical variables were described by number and percent (N, %),
agnosis of iron deficiency anemia and to exclude any other causes of where continuous variables described by mean and standard deviation
microcytic hypochromic anemia. The laboratory results were compared (Mean, SD, Median). Chi-square test was used to compare between
to the normal reference rang for IDA in children from 4 to 6 years [22]. categorical variables while T-test was used to compare between con-
Children who proved to have other types of microcytic anemia were tinuous variables. All analyses were performed with the IBM SPSS 20.0
excluded from the study. software. P value < 0.05 indicated a significant difference.

2.3. Cognitive and language assessment 3. Results

Cognitive assessment by measuring IQ and mental age was done for
each child using the Arabic translation Stanford Binet intelligence scale
version four [23]. This scale comprised of four cognitive areas; visual,
verbal and quantitative reasoning and short-term memory. Language
evaluation was done using the Arabic Language test [24]. It is valid and
reliable test for evaluation of language development of Arabic speaking
children in the age from 2 to 8 years old. The Arabic language test has
been widely used in Egypt since 1995 and it measures all the domains of
language [25]; Receptive language quotient, expressive language quo-
tient and total language quotient were calculated by dividing (the score
One hundred thirty six (136) of the 226 children enrolled in the
study had microcytic hypochromic anemia. Ninety children had normal
hemoglobin levels for age and they considered as a control (non-anemic
group). After iron study, 122 children (anemic group) of 136 were
found to have IDA while 14 children diagnosed as thalassemia trait by
hemoglobin electrophoresis and they were excluded from the study
(Table 1) showed the demographic data and the mothers' educational
level, and duration of breastfeeding for the patients and control. No
significant differences were observed between patient and control re-
garding age, sex, mothers' educational level.

2
M.A.M. Youssef, et al.
Table 4
Language assessment.
Number of children diagnosed as Delayed Language development
Rec. Q%
Exp. Q %
Total Q%
P value < 0.05 indicated a significant difference.
The hemoglobin level and iron study of patients and control were
shown in (Table 2). The hemoglobin level, MCV, MCHC, MCH, serum
ferritin, and serum iron were significantly lower in anemic patients
compared to the controls. Although, TIBC was significantly higher in
anemic patients (p < 0.0001). (Table 3) reveals the cognitive assess-
ment of both groups. No significant differences regarding IQ, mental
age between anemic and non-anemic children. (Table 4) reveals the
language assessment of both groups. The total number of children re-
cognized to have delayed language development were 34, (13 non-
anemic and 21 anemic).No significant differences could be detected
regarding receptive, expressive and total language quotients between
both groups.
4. Discussion
Iron deficiency anemia is the most common nutritional deficiency
worldwide, affecting approximately a quarter of the world's population,
primarily in developing countries [26]. Although few studies were
found in the literature specifically evaluating language in children
suffering from IDA, numerous studies reveal the effects of IDA on the
cognitive processes that make language learning possible. IDA has been
found to reduce the brain oxygenation and affect the neurotransmission
and myelination processes [1,13]. Moreover, Studies demonstrate that,
iron deficiency can cause abnormalities in the structure of hippocampus
pyramidal cell dendrites, within the limbic system which is an im-
portant seat of memory [13,27].
Our study focused on the impact of IDA on language development in
preschool children. Language evaluations using a standardized and
valid test for language assessment of children showed equal scores on
the index of perceptive, expressive and total language quotients for
anemic and non-anemic children.
A causal relationship between anemia and worse language and
cognitive development cannot be established in the literatures. Glazer
et al. [16], reviewed 10 longitudinal and clinical trials, and they could
not evaluate a causal relationship between iron deficiency anemia, and
cognitive, and mental development in childhood which play a basic role
in language development. Thomas et al. [28], reviewed the literatures
concerning iron and iron deficiency and its impact on psychological and
neurobiological consequences in both humans and experimental ani-
mals. They emphasis the role of iron during pregnancy and infancy and
the potential short and long term effect on neurobehavioral develop-
ment. They also could not find a clear support for a direct contributing
association between ID and abnormal development even with the clear
illumination of the neural mechanics of ID [17].
Additionally, supplementation of iron in the preschool found to
have no beneficial effects on the long term cognitive performance
[18,19]. Also, the clinical evidence does not appear to confirm that, the
long-term cognitive development can be enhanced by antenatal iron
supplementation [29]. Dissanayake et al. [30], conducted a study to
recognize the association of IDA with educational performance and
intelligence of 188 students. They concluded that IDA does not play a
major role in educational performance and intelligence of school-going
adolescent, asserting that numerous factors rather than ID affect edu-
cational performance and intelligence.
Thompson et al. [20] conducted a systematic review and meta-
3
International Journal of Pediatric Otorhinolaryngology 135 (2020) 110114
Non-anemic (n = 90) Anemic (n = 122) P. value
13 (14.4%) 21 (17.2%) 0.719
88.27 ± 10.76 86.6 ± 12.25 0.303
82.66 ± 7.54 82.07 ± 7.99 0.587
85.47 ± 8.19 84.34 ± 8.95 0.347
analysis study, and they revealed only limited data from randomized
controlled trials on the effects of iron on cognitive development in 2–5
years old.
Nevertheless, there are many studies have a reverse result.
Gunnarsson et al. [31], reported worse comprehension abilities in
children of mothers suffered from iron deficient. Likewise, the study
conducted by Santos et al. [32], they observed a difference in the lan-
guage evaluation between anemic and non -anemic children aged from
3 to 5 years, the worst performance was in the anemic group. However,
they could not confirm this observation statistically because the number
of children evaluated in this age was smaller than that has been eval-
uated in our study.
The relation between cognitive development, which is mandatory
for normal language development and psychomotor development
among anemic children has been evaluated in numerous studies. Some
of them support the suggestion that, the anemic children have delayed
cognitive and neuropsychomotor development [8,9]. While others
considered that the relation between nutrition and mental development
were interceded by motor activity and motor development [33,34].
The motely active child has more affluent experiences, more sti-
mulating circumstances, and more interactions with others which may
explain the positive impacts of the motor development on the cognitive
performance [35].
Conclusion: No influence of the IDA on the language development in
preschool children has been found. Future large controlled studies with
long follow-up time for younger age groups are needed to determine
whether there are existent associations between IDA with language and
to further clarification of the mechanisms underlying these associations.
Declaration of competing interest
The authors disclose no conflicts of interest.
References
[1] World Health Organization, Prevention of Iron Deficiency Anemia in Adolescents.
Role of Weekly Iron and Folic Acid Supplementation, (2011).
[2] N.J. Kassebaum, The global burden of anemia, Hematol. Oncol. Clin. N. Am. 30
(2016) 247–308.
[3] World Health Organization, The Global Prevalence of Anaemia in 2011, World
Health Organization, Geneva, 2015.
[4] WHO/UNICEF/UNU: Iron Deficiency Anaemia: Assessment, Prevention, and
Control. A Guide for Program Managers, World Health Organization, Geneva, 2001.
[5] M.C. Assunçäo, I.S. Satnos, Effect of food fortification with iron in childhood an-
emia: a review study, Cad. Saúde Pública 23 (2007) 269–281.
[6] S.E. Cusick, M.K. Georgieff, The role of nutrition in brain development: the golden
opportunity of the “first 1,000 days, J. Pediatr. 175 (2016) 16–21.
[7] J.C. McCann, B.N. Ames, An overview of evidence for a causal relation between iron
deficiency during development and deficits in cognitive or behavioral function, Am.
J. Clin. Nutr. 85 (2007) 931–945.
[8] B. Lozoff, J. Beard, J. Connor, B. Felt, M. Georgieff, T. Schallert, Long-Lasting neural
and behavioral effects of iron deficiency in infancy, Nutr. Rev. 64 (2006) 34–91.
[9] B. Lozoff, A.W. Wolf, E. Jimenez, Iron-deficiency anemia and infant development:
effects of extended oral iron therapy, J. Pediatr. 129 (1996) 382–389.
[10] J.S. Halterman, J.M. Kaczorowski, C.A. Aligne, P. Auinger, P.G. Szilagyi, Iron de-
ficiency and cognitive achievement among school-aged children and adolescents in
the United States, Pediatrics 107 (2001) 1381–1386.
[11] B. Lozoff, E. Jimenez, J.B. Smith, Double burden of iron deficiency in infancy and
low socioeconomic status: a longitudinal analysis of cognitive test scores to age 19
years, Arch Pediatr Adolesc 13 (2006) 1108–1113.
[12] S. Grantham-McGregor, C. Ani, A review of studies on the effect of iron deficiency
M.A.M. Youssef, et al.
on cognitive development in children, J. Nutr. 131 (2001) 649S-68.
[13] B. Lozoff, M.K. Georgieff, Iron deficiency and brain development, Semin. Pediatr.
Neurol. 13 (2006) 158–165.
[14] L.1 Pivina, Semenova Y1, M.D. Doşa, M. Dauletyarova, G. Bjørklund, Iron defi-
ciency, cognitive functions, and neurobehavioral disorders in children, J. Mol.
Neurosci. 68 (2019) 1–10.
[15] B.S. Gunnarsson, I. Thorsdottir, G. Palsson, S.J. Gretarsson, Iron status at 1 and 6
years versus developmental scores at 6 years in a well-nourished affluent popula-
tion, Acta Paediatr. 96 (2007) 391–395.
[16] Y. Glazer, N. Bilenko, Effect of iron deficiency and iron deficiency anemia in the
first two years of life on cognitive and mental development during childhood,
Harefuah 149 (2010) 309–314 335.
[17] I. Jáuregui-Lobera, Iron deficiency and cognitive functions, Neuropsychiatric Dis.
Treat. 10 (2014) 2087–2095.
[18] L.E. Murray-Kolb, S.K. Khatry, J. Katz, B.A. Schaefer, P.M. Cole, S.C. LeClerq,
M.E. Morgan, J.M. Tielsch, P. Christian, Preschool micronutrient supplementation
effects on intellectual and motor function in school-aged Nepalese children, Arch
Pediatr Adolesc 166 (2012) 4 04–410.
[19] T. Pongcharoen, A.M. DiGirolamo, U. Ramakrishnan, P. Winichagoon, R. Flores,
R. Martorell, Long-term effects of iron and zinc supplementation during infancy on
cognitive function at 9 y of age in northeast Thai children: a follow-up study, Am. J.
Clin. Nutr. 93 (2011) 636–643.
[20] J.1 Thompson, B.A. Biggs, S.R. Pasricha, Effects of daily iron supplementation in 2-
to 5-year-old children: systematic review and meta-analysis, Pediatrics 131 (2013)
739–753.
[21] J. Flerlage, B. Engorn (Eds.), The Harriet Lane Handbook: A Manual for Pediatric
House Officers, twentieth ed., Saunder/Elsevier, Philadelphia, Pa, 2015, p. 305.
[22] N.1 Özdemir, Iron deficiency anemia from diagnosis to treatment in children, Turk
Pediatri Ars 50 (2015) 11–19.
[23] M.A. Hanoura, Stanford-binet Intelligence Scale: Arabic Version of the Fourth
Revision, Anglo-Egyptian Library, Cairo, Egypt, 2003.
[24] N. Rifaie, S. Hassan, The Arabic language test revised, Benha Med J 21 (2004)
205–216 (2004).
4
International Journal of Pediatric Otorhinolaryngology 135 (2020) 110114
[25] M.N. Kotby, A. Khairy, M. Barakah, et al., Language testing of Arabic speaking
children, Proceeding of the XXIII World Congress of the International Association of
Logopedics. Cairo, 6–10th August, 1995.
[26] R.F. Luo, X. Wang, J.X. Zhang, C.F. Liu, Y.J. Shi, G. Miller, High anaemia prevalence
in Western China, The Southeast Journal of tropical medicine and puplic health 42
(2011) 1204–1213.
[27] B. Lozoff, I. De Andraca, M. Castillo, J.B. Smith, T. Walter, P. Pino, Behavioral and
developmental effects of preventing iron-deficiency anemia in healthy full-term
infants, Pediatrics 112 (2003) 846–854.
[28] D.G.1 Thomas, S.L. Grant, Aubuchon-Endsley, The role of iron in neurocognitive
development, Dev. Neuropsychol. 34 (2009) 196–222.
[29] L.M. Larson, K.S. Phiri, S.-R. Pasricha, Iron and cognitive development: what is the
evidence? Ann Nutr Metab 71 (2017) 25–38.
[30] D.S. Dissanayake, P.V.R. Kumarasiri, D.B. Nugegoda, et al., The association of iron
status with educational performance and intelligence among adolescents, Ceylon
Med. J. 54 (2009) 75–79.
[31] I. Gunnarsson Bs Thorsdottir, G. Palsson, S.J. Gretarsson, Iron status at 1 and 6
years versus developmental scores at 6 years in a well-nourished affluent popula-
tions, Acta Pediatr 96 (2007) 391–395 2007.
[32] J.N. Santos, S.M. rates, S.M. Lemos, J.A. Lamounier, Consequences of anemia on
language development of children from a public day care center, Rev. Paul. Pediatr.
Rev. paul. pediatr. 27 (1) (2009) São Paulo Mar. 2009 Print, version ISSN 0103-
0582On-line version ISSN 1984-0462.
[33] D.K. Olney, P.K. Kariger, R.J. Stoltzfus, S.S. Khalfan, N.S. Ali, J.M. Tielsch,
S. Sazawal, R. Black, L.H. Allen, E. Pollitt, Developmental effects of micronutrient
supplementation and malaria in Zanzibari children, Early Hum. Dev. 89 (2013) 6
67–674.
[34] E. Pollitt, A. Jahari, H. Walka, A developmental view of the effects of an energy and
micronutrient supplement in undernourished children in Indonesia, Eur. J. Clin.
Nutr. 54 (2000) S107–S113.
[35] K.E. Adolph, C.S. Tamis-LeMonda, The costs and benefits of development: the
transition from crawling to walking, Child Dev Perspect 8 (2014) 187–192.
Original article
Korean J Pediatr 2017;60(5):151-157 Korean J Pediatr 2017;60(5):151-157
https://doi.org/10.3345/kjp.2017.60.5.151
pISSN 1738-1061•eISSN 2092-7258
Korean J Pediatr

Prevalence and clinical manifestations of ma­

crolide resistant Mycoplasma pneumoniae
pneumonia in Korean children
Eun Lee1, Hyun-Ju Cho2, Soo-Jong Hong2, Jina Lee2, Heungsup Sung3, Jinho Yu2
1
Department of Pediatrics, Chonnam National University Hospital, Gwangju, Departments of 2Pediatrics and 3Laboratory Medicine, Asan Medical Center, University
of Ulsan College of Medicine, Seoul, Korea

Purpose: Macrolide resistance rate of Mycoplasma pneumoniae has rapidly increased in children. Corresponding author: Jinho Yu, MD, PhD
Studies on the clinical features between macrolide susceptible-M. pneumoniae (MSMP) and macrolide Department of Pediatrics, Asan Medical Center, Uni-
versity of Ulsan College of Medicine, 88 Olympic-ro
resistant-M. pneumoniae (MRMP) are lacking. The aim of this study was to identify the macrolide 43-gil, Songpa-gu, Seoul 05505, Korea
resistance rate of M. pneumoniae in Korean children with M. pneumoniae penupmonia in 2015 and Tel: +82-2-3010-3922
compare manifestations between MSMP and MRMP. Fax: +82-2-473-3725
Methods: Among 122 children (0–18 years old) diagnosed with M. pneumoniae pneumonia, 95 children E-mail: jyu3922@gmail.com
Co-corresponding author: Heungsup Sung, MD,
with the results of macrolide sensitivity test were included in this study. Clinical manifestations were PhD
acquired using retrospective medical records. Department of Laboratory Medicine, Asan Medical
Results: The macrolide resistant rate of M. pneumoniae was 87.2% (82 of 94 patients) in children with Center, University of Ulsan College of Medicine, 88
M. pneumoniae pneumonia. One patient showed a mixed type of wild type and A2063G mutation in Olympic-ro 43-gil, Songpa-gu, Seoul 05505, Korea
Tel: +82-2-3010-4499
23S rRNA of M. pneumoniae. There were no significant differences in clinical, laboratory, and radiologic Fax: +82-2-2045-4081
findings between the MSMP and MRMP groups at the first visit to our hospital. The time interval between E-mail: sung@amc.seoul.kr
initiation of macrolide and defervescence was significantly longer in the MRMP group (4.9±3.3 vs.
2.8±3.1 days, P=0.039). Received: 8 September, 2016
Revised: 1 December, 2016
Conclusion: The macrolide resistant rate of M. pneumoniae is very high in children with M. pneumoniae Accepted: 21 December, 2016
pneumonia in Korea. The clinical manifestations of MRMP are similar to MSMP except for the deferve­
scence period after administration of macrolide. Continuous monitoring of the occurrence and antimi­
crobial susceptibility of MRMP is required to control its spread and establish strategies for treating
second-line antibiotic resistant M. pneumoniae infection.

Key words: Child, Macrolide, Mycoplasma pneumoniae, Drug resistance

Introduction
Mycoplasma pneumoniae can cause a variety of respiratory tract diseases, such as upper
respiratory infection and atypical pneumonia1). The clinical course of M. pneumoniae
infection is diverse and ranges from self-limiting to severe pneumonia with extrapul­
monary complications2). Among the diverse clinical presentations, lower respiratory tract
infections with pneumonia most commonly require clinical attention.
Copyright © 2017 by The Korean Pediatric Society
Macrolide is considered the first-line treatment for M. pneumoniae infection3). Transi­
tional mutations in 23S rRNA of M. pneumoniae were reported in erythromycin-resistant This is an open-access article distributed under the
terms of the Creative Commons Attribution Non-
M. pneumoniae in 1995.4) Thereafter, especially since 2000, the prevalence of macrolide- Commercial License (http://creativecommons.org/
resistant M. pneumoniae (MRMP) infection has rapidly increased, with variations according licenses/by-nc/4.0/) which permits unrestricted non-
commercial use, distribution, and reproduction in any
to region and study population5). The macrolide resistance rate of M. pneumoniae is much medium, provided the original work is properly cited.

https://doi.org/10.3345/kjp.2017.60.5.151 151
Lee E, et al. • Macrolide-resistant Mycoplasma pneumoniae pneumonia
higher in East Asia than in Europe and North America, with up to
87.1% in Japanese children in 20115-8). In recent M. pneumoniae
epidemics in Korea, the macrolide resistance rate has markedly
increased from 2.9% in 2003 to 62.9% in 20115).
In cases of MRMP infection, secondary treatment options are
limited due to adverse effects of tetracycline or fluoroquinolones,
especially in children9). In addition, resistance to second-line
therapy is a concern given the rapid increase in MRMP preval­
ence. Therefore, continuous survey on the prevalence of MRMP
and surveillance on the prescription for M. pneumoniae are in­
evitably needed in the prevailing state of MRMP.
Previous studies comparing the clinical manifestations of
macrolide susceptible M. pneumoniae (MSMP) and MRMP show­
ed inconclusive results10-13). Although a high macrolide resistance
rate of M. pneumoniae has been reported, studies on the treat­
ment patterns of MRMP pneumonia in children are lacking. Fur­
ther investigation is needed to develop appropriate treatment
strategies and monitor the emergence of second-line therapy
resistant M. pneumoniae.
The aim of this study was to identify the prevalence of macro­
lide resistance in children with M. pneumoniae pneumonia in
2015 and compare the clinical features and treatment patterns of
MSMP and MRMP in these children.
Materials and methods
1. Study population
This study enrolled patients aged between 0–18 years old, who
were diagnosed with community-acquired pneumonia due to M.
pneumoniae who visited our tertiary hospital in Seoul between
April 2015 and November 2015. All of the present study patients
underwent chest radiography and either blood tests including
specific IgM against M. pneumoniae using a LIAISON Mycoplas­
ma pneumoniae IgM kit (DiaSorin, Dublin, Ireland) or polymerase
chain reaction (PCR) for M. pneumoniae using the AmpliSens
Mycoplasma pneumoniae/Chlamydia pneumoniae-FRT PCR kit
(InterLabService Ltd., Moscow, Russia) at the initial visit to the
hospital.
During the study period, 122 children were diagnosed with M.
pneumoniae pneumonia on the basis of either specific IgM
positivity in a blood test or positive PCR result combined with
chest radiography and physical examination14). Four children re­
ceived only serologic testing for specific IgM against M. pneu­
moniae and showed positivity. Eight children underwent only
PCR analysis of their sputum for M. pneumoniae and showed a
positive result. Ninety-two children showed both specific IgM and
PCR positivity for M. pneumoniae. The remaining 18 children
were tested for both specific IgM and PCR for M. pneumo­niae,
but showed a positive result only for PCR. Among the 118
152 https://doi.org/10.3345/kjp.2017.60.5.151
children with positive result by PCR for M. pneumoniae, macro­
lide resistance tests were performed for 95 children with available
samples.
All of the chest radiographs were reviewed by an experienced
radiologist. Infiltration on the chest radiography was defined as
any poorly defined opacity in the lung field and consolidation
was defined as air-space opacification. Information on clinical
manifestations and prescribed medicine during the disease course
was obtained using a retrospective chart review. Fever was de­
fined as a body temperature above 38°C. This study was approved
by the Institutional Review Board of Asan Medical Center
(approval number: 2015-1400).
2. PCR for identification of macrolide resistance
During the study period, PCR for M. pneumoniae was perform­
ed in children with pneumonia diagnosed on the basis of chest
radiography and physical examination. This analysis was done
using nasopharyngeal aspirates collected upon visiting to the
hospital. For detection of M. pneumoniae, our previously reported
procedure was applied15). Evaluations of macrolide resistance
were performed in children with a positive PCR result for M.
pneumoniae. A total of 95 M. pneumoniae isolates, including one
case of a mixed type of MSMP and MRMP, were obtained from
sputum samples. Domain V of the 23S rRNA gene was amplified
using previously described primer pairs (GenBank accession no.
X68422)16). Nested PCR primers and the conditions described by
Oh et al.17) were used for the specimens. PCR products were
purified using a Power Gel Extraction kit (TaKaRa Bio Inc., Shiga,
Japan). The purified templates were sequenced using an ABI
Prism BigDye Terminator v3.1 cycle sequencing kit (Applied
Biosystems, Foster City, CA, USA) and analyzed on an ABI 3730xl
DNA analyzer (Applied Biosystems).
3. Detection of respiratory virus
Nasopharyngeal swabs were taken by flocked swab and sub­
mitted in Universal Transport Medium (Copan Italia S.p.A.,
Brescia, Italy). Viral RNA was extracted from the swabs with
NucliSENS easyMAG (bioMerieux, Marcy I’Etoile, France). cDNA
was synthesized using a Revert Aid First Standard cDNA Syn­
thesis kit (Fermentas, York, UK), and each cDNA preparation was
subjected to three sets of real-time multiplex PCR with an
Anyplex II RV16 Detection kit (Seegene, Seoul, Korea); this kit
targets 16 respiratory viruses, including respiratory syncytial
viruses A and B, adenovirus, rhinovirus, parainfluenza viruses 1
to 4, influenza viruses A and B, metapneumovirus, bocavirus,
corona viruses OC43, 229E, and NL63, and enterovirus. These 16
viruses cause the most common respiratory infections in Korea
according to weekly monitoring by the Korea Centers for Disease
Control & Prevention18).
 Korean J Pediatr 2017;60(5):151-157

Table 1. Clinical characteristics of study participants with Mycoplasma pneumoniae pneumonia according to macrolide susceptibility
Characteristic MSMP (n=12) MRMP (n=82) Total group (n=94) P value
Age (yr) 7.6±3.1 5.1±2.6 5.4±2.8 0.001*
Male sex 3/12 (25.0) 34/82 (41.5) 37/94 (39.4) 0.353
Admission rate 7/12 (58.3) 62/82 (75.6) 69/94 (73.4) 0.206
Total fever duration (day) 8.0±6.0 8.2±3.2 8.2±3.6 0.884
Respiratory rate at the time of visit (/min) 25.7±4.4 (20–34) 27.4±6.0 (20–52) 27.2±5.9 (20–52) 0.486
Heart rate at the time of visit (/min) 118.9±13.8 122.9±16.4 122. 5±16.1 0.508
Positive IgM against M. pneumoniae 8/9 (88.9) 64/77 (83.1) 72/86 (83.7) 0.207
Whole blood cells (cells/mm3) 7,500±2,670 8,926±4,801 8,717±4,575 0.319
Platelet count (/mm3) 248.5±59.2 308.3±106.1 300.9±103.1 0.072
Neutrophil (%) 69.1±7.7 61.0±13.3 62.0±13.0 0.054
Lymphocytes (%) 21.1±6.0 28.6±11.9 27.7±11.6 0.003*
Eosinophil (%) 1.5±2.2 2.1±2.1 2.0±2.1 0.456
CRP (mg/dL) 5.9±7.3 (0.4–21.2) 4.8±5.0 (0.1–26.5) 4.9±5.3 (0.1-26.5) 0.539
AST (IU/L) 648.7±2,153.5 (23–7,487) 82.3±420.5 (17–3,746) 159.6±878.7 (17–7,487) 0.380
ALT (IU/L) 720.6±2354.0 (6–7,818) 59.2±379.1 (6–3,362) 140.9±896.3 (6–7,818) 0.374
Coinfection with virus 5/6 (83.3) 30/54 (55.6) 35/60 (58.3) 0.190
Values are presented as a mean±standard deviation (SD), number (%), mean±SD (range),
MRMP, macrolide-resistant Mycoplasma pneumoniae; MSMP, macrolide-susceptible Mycoplasma pneumoniae; CRP, C-reactive protein; AST, aspartate
aminotransferase; ALT, alanine aminotransferase.
*P<0.05.
95.1%
100 (n=39/41)
Macrolide resistance rate of M. pneumoniae (%)

4. Statistical analysis 81.8%

(n=36/44) 77.8%
To compare the clinical and radiologic features and treatment (n=7/9)
80
regimen between MSMP and MRMP, a t test, Mann-Whitney U
test, chi-square test, and Fisher exact test were used, as appro­
60
priate. To control for the confounding factors, logistic regression
analysis was performed. P values less than 0.05 were considered
40
statistically significant. All statistical analyses were performed
using IBM SPSS Statistics ver. 21.0 (IBM Co., Armonk, NY, USA).
20

0
Results 0–4 Years 5–9 Years 10–18 Years

Fig. 1. Rate of macrolide-resistant Mycoplasma pneumoniae in children

1. Macrolide resistance rate of M. pneumoniae and its distribu­
tion according to age
The macrolide resistant rate of M. pneumoniae was 87.2% (82
of 94) in children with M. pneumoniae pneumonia (Table 1).
Those with MRMP were significantly younger than those with
MSMP (MSMP, 7.6±3.1 years; MRMP, 5.1±2.6 years; P=0.001) in
the present study. When stratified according to age, the decreas­
ing pattern of the prevalence of MRMP was observed with a weak
trend significance (P=0.052) as follows: 0–4 years, 95.1% (39 of
41); 5–9 years, 81.8% (36 of 44); 10–18 years, 77.8% (7 of 9) (Fig.
1). The macrolide resistance rate of M. pneumoniae was higher in
late summer and fall (Fig. 2).
2. Detection of 23S rRNA gene mutations of M. pneumoniae
Among the 95 M. pneumoniae-positive samples, 82 cases were
according to age. P for trend=0.052.
diagnosed with A2063G mutation. No other known mutations,
such as A2064, in the 23S rRNA gene of M. pneumoniae were
identified in the present study. The other 12 samples carried a
wild type of 23S rRNA gene. The remaining one case showed a
mix of wild type and A2063G mutation.
3. Comparison of clinical findings between MSMP and MRMP in
children with M. pneumoniae pneumonia
There were no significant differences in total fever duration and
respiratory rate at the initial hospital visit between the MSMP and
MRMP groups. The admission rate due to M. pneumoniae pneu­
monia was 72.6% (69 of 95) in the total population. The hospitali­
zation rate was higher in the MRMP group compared with the

https://doi.org/10.3345/kjp.2017.60.5.151 153
Lee E, et al. • Macrolide-resistant Mycoplasma pneumoniae pneumonia

MSMP group without statistical significance (75.6% vs. 58.3%; P=
0.206) (Table 1). Blood lymphocytes were significantly in­creased
in the MRMP group compared with the MSMP group (P=0.003),
although this difference was not significant after controlling for
age. Among the children with M. pneumoniae pneumonia, 58.3%
showed coinfection with respiratory viruses such as rhinovirus
MSMP
35
30
25
Number of cases
and parainfluenza virus without significant differences between
the MSMP (83.3%, 5 of 6) and MRMP (55.6%, 30 of 54) groups.
4. Comparison of radiologic findings between MSMP and MRMP
in children with M. pneumoniae pneumonia
Chest radiography indicated that consolidation (MSMP, 10 of
12 vs. MRMP, 46 of 82) and effusion (MSMP, 3 of 12 vs. MRMP, 9
of 82) were commonly involved in M. pneumoniae pneumonia
MRMP
regardless of macrolide resistance (Table 2). There were no statis­
tical differences in the prevalence of consolidation or effusion
between the MSMP and MRMP groups.

20 5. Comparison in the treatment and clinical outcome of M.

15
pneumoniae pneumonia according to macrolide sensitivity
The total duration of antibiotic administration, including non­
10
macrolide antibiotics (beta-lactam and cephalosporin) and
5 macrolide, was slightly longer in the MRMP group than in the
0
MSMP group, without statistical significance. The total number of
administered antibiotics was higher in the MRMP group than in
the MSMP group (P=0.046) (Table 3). The most commonly
prescribed initial antibiotic was macrolide in both the MSMP
Fig. 2. Monthly distribution of the occurrence of MSMP and MRMP
pneumonia in Korean children in 2015. MSMP, macrolide susceptible- (80.0%) and MRMP (68.4%) groups, without significant signi­
Mycoplasma pneumoniae; MRMP, macrolide resistant-M. pneumoniae. ficance. Tetracycline or fluoroquinolone were administered due to
unresponsiveness to macrolide in both the MSMP (25.0%, 3 of
Table 2. Comparison of radiologic features between children infected 12) and MRMP (29.3%, 24 of 82) groups. Changes of antibiotics
with MSMP and MRMP to other antibiotics among macrolides (azithromycin, clarithro­
MSMP (n=12) MRMP (n=82) Total (n=94) P value mycin, or roxithromycin) were more common in the MRMP
Infiltration 12 (100) 82 (100) 94 (100) NA group compared with the MSMP group without statistical signifi­
Consolidation 10 (83.3) 46 (56.1) 56 (59.6) 0.073 cance (41.8% vs. 18.2%). The time intervals between the initiation
Effusion 3 (25.0) 9 (11.0) 12 (12.8) 0.174 of macrolide and defervescence were significantly longer in the
Values are presented as number (%). MRMP group compared with the MSMP group (4.9±3.3 vs. 2.8±
MSMP, macrolide-susceptible Mycoplasma pneumoniae; MRMP, macrolide- 3.1 days, P=0.039). There was no significant difference in the
resistant Mycoplasma pneumoniae; NA, not applicable.

Table 3. Comparison of treatment patterns and response to macrolides in children with Mycoplasma pneumoniae pneumonia according to macrolide
sensitivity
Variable MSMP (n=12) MRMP (n=82) Total (n=94) P value
Total duration of antibiotics (day) 12.7±6.4 13.5±5.2 13.4±5.3 0.632
Total number of antibiotics used (day) 1.6±0.7 2.1±0.8 2.00±0.8 0.046*
Initially prescribed antibiotics 0.671
Nonmacrolide 2/10 (20.0) 24/76 (31.6) 26/86 (30.2)
Azithromycin 6/10 (60.0) 23/76 (30.3) 29/86 (33.7)
Clarithromycin 2/10 (20.0) 24/76 (31.6) 26/86 (30.2)
Roxithromycin 0/10 (0) 5/76 (6.5) 5/86 (5.8)
Number of changes in antibiotics from macrolide to tetracycline or fluoroquinolone 3/12 (25.0) 24/82 (29.3) 27/94 (28.7) 0.833
Antibiotics changes within the macrolide 2/11 (18.2) 33/79 (41.8) 35/90 (38.9) 0.133
Time to defervescence after initiation of the first macrolide (day) 2.8±3.1 (0–9) 4.9±3.3 (0–15) 4.7±3.3 0.039*
Values are presented as a mean±standard deviation (SD), number (%), mean±SD (range),
Nonmacrolide antibiotics included β-lactam and cephalosporin.
MRMP, macrolide-resistant Mycoplasma pneumoniae; MSMP, macrolide-susceptible Mycoplasma pneumoniae.
*P<0.05.

154 https://doi.org/10.3345/kjp.2017.60.5.151

period between onset of fever and start of macrolide admini­
stration (MSMP, 6.6±7.0 days; MRMP, 4.1±3.2 days; P=0.254).
Fever was subsided after 1.7 days from the start of administration
of tetracycline or fluoroquinolone. There were no side effects of
the treatment with tetracycline or fluoroquinolone. None of the
variables listed in Table 3 were confounded by age. All patients
hospitalized due to M. pneumoniae pneumonia were discharged
in a defervescent state with partial or total improvement in chest
radiography compared with administration. There was no signi­
ficant difference in the hospitalization duration between the
MSMP (mean±standard deviation, 7.7±5.6 days) and MRMP (6.3±
3.4 days) groups. There were also no cases in our current series
who needed ventilator care or transfer to an intensive care unit
due to the pneumonia.
6. Mixed wild type and A2063G mutant case
A mix of wild type and A2363G mutant M. pneumoniae 23S
rRNA was detected in a 4-year-old girl, who had presented with
fever and cough 4 days earlier. She was prescribed with a 3-day
regimen of clarithromycin before sputum sample collection for
sequencing of the 23S rRNA gene.
Discussion
In our current study, we have identified a macrolide resistance
rate of 87.2% in children diagnosed with M. pneumoniae in 2015.
All cases of this macrolide-resistant strain showed an A2063G
point mutation in the 23S rRNA gene. MRMP was detected in
younger children with a higher prevalence. There were no signifi­
cant differences in the clinical, laboratory, and radiologic findings
between MSMP and MRMP groups. Administration of macrolide
led to more rapid defervescence in the MSMP group compared
with the MRMP group.
The prevalence of MRMP observed in our present study was
higher than that reported for 2000 to 2011 in Korea. This pre­
valence increased over time and exceeded a peak of 62.9% in
20115), and was similar to that reported in Japanese children in
2011 (87.1%)8) and Chinese children in 2008–2009 (90.0%)13).
However, the prevalence of MRMP in Europe has been reported
to be less than 26%19,20). Recent studies on the prevalence of
MRMP infection are lacking, and our present analysis is signifi­
cant because it reports on the recent macrolide resistance rate of
M. pneumoniae with increasing pattern in Korean children.
We found no significant differences in clinical manifestations
or laboratory findings between the MRMP and MSMP groups.
Even in the MSMP group, complications of M. pneumoniae infec­
tion, such as hepatitis, high C-reactive protein levels, long-term
fever, and consolidation and effusion in chest radiography, were
similar to those in the MRMP group. In addition, no differences in
Korean J Pediatr 2017;60(5):151-157
the clinical manifestations between the 2 groups were found to be
associated with the administration of tetracycline or fluoro­
quinolone before identification of macrolide sensitivity, even in
the MSMP group. Previous studies on the comparisons of clinical
manifestations between MSMP and MRMP groups also reported
no significant differences12,21). However, the mean duration from
the start of macrolide treatment to defervescence was longer in
the MRMP group compared with the MSMP group in our present
study, which is similar to the results of the previous study11). The
relatively long-term period of fever in M. pneumoniae infection
might be partially attributable to the immune reaction in asso­
ciated with M. pneumoniae infection regardless of the macrolide
resistance22). Large-scale studies on the clinical course of these
infections are needed in the future to compare clinical manife­
station between MSMP and MRMP infection.
In previous studies, the most common macrolide resistance
mutation (up to 97.5%) was the A2063G mutation in the 23s
rRNA5,21), which we also observed in our present study. Macrolide
inhibits protein synthesis by binding to domain V of 23S RNA at
nucleotide positions 2063 and 206421). Mutations at these sites
enable protein synthesis that promotes M. pneumoniae survival.
The minimum inhibitory concentration (MIC) of macrolides
differs according to the specific point mutation7,8,21). A2063G and
A2064G confer the most resistance to macrolides and also pro­
duce resistance to 14-ring macrolides, such as clarithromycin
(MIC, 8 to >128) and roxithromycin (MIC, 0.008 to 128), and 15-
ring macrolides such as azithromycin (MIC, 1 to 64)5,7,23). Com­
pared to clarithromycin, azithromycin and roxithromycin have
lower MIC levels and are preferred as an initial treatment option
for M. pneumoniae infection with unidentified macrolide resis­
tance. As widespread macrolide usage is associated with the
occurrence of MRMP, continuous monitoring of the MICs for
each macrolide and secondary line therapy against M. pneu­
moniae infection are needed to identify the advent of M. pneu­
moniae stains that are resistant to other antibiotics and establish
treatment strategies for MRMP infection.
We identified one case with mixed A2063G and wild type 23S
rRNA. Although most macrolide resistance is detected at the start
of the disease course7), a conversion from MSMP to MRMP is also
possible during clarithromycin treatment24). Possible underlying
mechanisms of mixed type of macrolide resistance in M. pneu­
moniae include selected outgrowth of MRMP resulting from
administration of clarithromycin. The aforementioned case might
support the outgrowth of MRMP during M. pneumoniae treat­
ment with macrolide.
In our present study series, 30.2% of the children with M.
pneumoniae pneumonia were initially prescribed nonmacrolide
anti­biotics. Although M. pneumoniae is known to cause pneu­
monia in older children14), it can also cause lower respiratory tract
infections including pneumonia in children as young as 6 months
https://doi.org/10.3345/kjp.2017.60.5.151 155
Lee E, et al. • Macrolide-resistant Mycoplasma pneumoniae pneumonia
old25). Therefore, M. pneumoniae can be considered as a pathogen
for respiratory infections, even in young children, and especially
during an epidemic of M. pneumoniae.
Our present study is significant because it has compared the
manifestations of MRMP and MSMP in children in a high macro­
lide resistance period for M. pneumoniae. However, it also had
several limitations of note. All of the patients analyzed visited our
tertiary hospital, and this population may therefore have included
some very severe M. pneumoniae cases. However, we found no
significant differences between the clinical, radiologic, and
laboratory findings for the MSMP and MRMP groups analyzed.
Our sample size was relatively small, and the study duration was
relatively short. Therefore, our analysis lacked an evaluation of
the full spectrum of M. pneumoniae pneumonia in relation to
macrolide resistance. Also, there was a significant difference in
age distribution between the 2 groups, which caused a selec­tion
bias. However, the prevalence of consolidation and effusion on
chest radiography, which might suggest more severe pneu­monia,
was similar between our 2 study groups even after adjust­ment for
age (data not shown). For the diagnosis of M. pneu­moniae
infection, serological assays and PCR using sputum samples are
widely used. However, these tests have limitations in that false
responses can be obtained depending on sample collec­tion time
and remote infection in serology tests and colonization in airways
in PCR.
In conclusion, there was a high macrolide resistance rate of M.
pneumoniae (87.2%) in Korean children with M. pneumoniae
pneumonia in 2015. MRMP pneumonia occurred across all ages,
including infants. Although there were no significant differences
in the clinical, laboratory and radiologic findings between the
MSMP and MRMP groups, MRMP is associated with persistence
of fever during its clinical course. Further large-scale, nation-
wide studies are required to control the spread of MRMP and
establish strategies for treatment of MRMP infection.
Conflict of interest
No potential conflict of interest relevant to this article was
reported.
Acknowledgments
This study was supported by the BioNano Health-Guard Re­
search Center funded by the Ministry of Science, ICT & Future
Planning (MSIP) of Korea as a Global Frontier Project (grant
number: H-GUARD_ERND2-5).
156 https://doi.org/10.3345/kjp.2017.60.5.151
References
1. Eun BW, Kim NH, Choi EH, Lee HJ. Mycoplasma pneumoniae in
Korean children: the epidemiology of pneumonia over an 18-year
period. J Infect 2008;56:326-31.
2. Defilippi A, Silvestri M, Tacchella A, Giacchino R, Melioli G, Di
Marco E, et al. Epidemiology and clinical features of Mycoplasma
pneumoniae infection in children. Respir Med 2008;102:1762-8.
3. Kwiatkowska B, Ma śli ńska M. Macrolide therapy in chronic
inflammatory diseases. Mediators Inflamm 2012;2012:636157.
4. Lucier TS, Heitzman K, Liu SK, Hu PC. Transition mutations in the
23S rRNA of erythromycin-resistant isolates of Mycoplasma
pneumoniae. Antimicrob Agents Chemother 1995;39:2770-3.
5. Hong KB, Choi EH, Lee HJ, Lee SY, Cho EY, Choi JH, et al. Macro­
lide resistance of Mycoplasma pneumoniae, South Korea, 2000-
2011. Emerg Infect Dis 2013;19:1281-4.
6. Dumke R, von Baum H, Lück PC, Jacobs E. Occurrence of macro­
lide-resistant Mycoplasma pneumoniae strains in Germany. Clin
Microbiol Infect 2010;16:613-6.
7. Cao B, Zhao CJ, Yin YD, Zhao F, Song SF, Bai L, et al. High pre­
valence of macrolide resistance in Mycoplasma pneumoniae iso­
lates from adult and adolescent patients with respiratory tract
infection in China. Clin Infect Dis 2010;51:189-94.
8. Okada T, Morozumi M, Tajima T, Hasegawa M, Sakata H, Ohnari S,
et al. Rapid effectiveness of minocycline or doxycycline against
macrolide-resistant Mycoplasma pneumoniae infection in a 2011
outbreak among Japanese children. Clin Infect Dis 2012;55:1642-
9.
9. Principi N, Esposito S. Macrolide-resistant Mycoplasma pneu­
moniae: its role in respiratory infection. J Antimicrob Chemother
2013;68:506-11.
10. Miyashita N, Kawai Y, Akaike H, Ouchi K, Hayashi T, Kurihara T, et
al. Macrolide-resistant Mycoplasma pneumoniae in adolescents
with community-acquired pneumonia. BMC Infect Dis 2012;12:
126.
11. Suzuki S, Yamazaki T, Narita M, Okazaki N, Suzuki I, Andoh T, et
al. Clinical evaluation of macrolide-resistant Mycoplasma pneu­
moniae. Antimicrob Agents Chemother 2006;50:709-12.
12. Matsubara K, Morozumi M, Okada T, Matsushima T, Komiyama O,
Shoji M, et al. A comparative clinical study of macrolide-sensitive
and macrolide-resistant Mycoplasma pneumoniae infections in
pediatric patients. J Infect Chemother 2009;15:380-3.
13. Liu Y, Ye X, Zhang H, Xu X, Li W, Zhu D, et al. Characterization of
macrolide resistance in Mycoplasma pneumoniae isolated from
children in Shanghai, China. Diagn Microbiol Infect Dis 2010;67:
355-8.
14. McIntosh K. Community-acquired pneumonia in children. N Engl
J Med 2002;346:429-37.
15. Yoo SJ, Kim HB, Choi SH, Lee SO, Kim SH, Hong SB, et al. Differ­
ences in the frequency of 23S rRNA gene mutations in Mycoplas­
ma pneumoniae between children and adults with community-
acquired pneumonia: clinical impact of mutations conferring
macrolide resistance. Antimicrob Agents Chemother 2012;56:
6393-6.
16. Wolff BJ, Thacker WL, Schwartz SB, Winchell JM. Detection of
macrolide resistance in Mycoplasma pneumoniae by real-time
PCR and high-resolution melt analysis. Antimicrob Agents
Chemother 2008;52:3542-9.
17. Oh CE, Choi EH, Lee HJ. Detection of genetic mutations associated
with macrolide resistance of Mycoplasma pneumoniae. Korean J
Pediatr 2010;53:178-83.

18. Korea Centers for Disease Control and Prevention. Acute infectious
agents laboratory surveillance reports [Internet]. Cheongju: Korea
Centers for Disease Control and Prevention; c2012 [cited 2014 Oct
2]. Available from: http://www.cdc.go.kr.
19. Chironna M, Sallustio A, Esposito S, Perulli M, Chinellato I, Di Bari
C, et al. Emergence of macrolide-resistant strains during an out­
break of Mycoplasma pneumoniae infections in children. J Anti­
microb Chemother 2011;66:734-7.
20. Pereyre S, Charron A, Renaudin H, Bébéar C, Bébéar CM. First
report of macrolide-resistant strains and description of a novel
nucleotide sequence variation in the P1 adhesin gene in Myco­plas­
ma pneumoniae clinical strains isolated in France over 12 years. J
Clin Microbiol 2007;45:3534-9.
21. Morozumi M, Takahashi T, Ubukata K. Macrolide-resistant Myco­
plasma pneumoniae: characteristics of isolates and clinical aspects
Korean J Pediatr 2017;60(5):151-157
of community-acquired pneumonia. J Infect Chemother 2010;16:
78-86.
22. Yang J, Hooper WC, Phillips DJ, Talkington DF. Cytokines in
Mycoplasma pneumoniae infections. Cytokine Growth Factor Rev
2004;15:157-68.
23. Ye Y, Li S, Yajun L, Li Y, Ren T, Liu K. Mycoplasma pneumoniae
23S rRNA gene mutations and mechanisms of macrolide resis­
tance. Lab Med 2013;44:63-8.
24. Cardinale F, Chironna M, Dumke R, Binetti A, Daleno C, Sallustio
A, et al. Macrolide-resistant Mycoplasma pneumoniae in pae­
diatric pneumonia. Eur Respir J 2011;37:1522-4.
25. Sun H, Chen Z, Yan Y, Huang L, Wang M, Ji W. Epidemiology and
clinical profiles of Mycoplasma pneumoniae infection in hospi­
talized infants younger than one year. Respir Med 2015;109:751-7.
https://doi.org/10.3345/kjp.2017.60.5.151 157