New Series - hkjpaed.org

New Series

Offical Publication of:Hong Kong College of PaediatriciansHong Kong Paediatric Society

HK J Paediatr (New Series) Vol 24. No. 3 July 2019

ISSN 2309-5393 (onl ine)ISSN 2309-5393 (onl ine)ISSN 2309-5393 (onl ine)ISSN 2309-5393 (onl ine)ISSN 2309-5393 (onl ine)ISSN 1013-9923 (pr in t)ISSN 1013-9923 (pr in t)ISSN 1013-9923 (pr in t)ISSN 1013-9923 (pr in t)ISSN 1013-9923 (pr in t)

Indexed in EMBASE/Excerpta Medica,Indexed in EMBASE/Excerpta Medica,Indexed in EMBASE/Excerpta Medica,Indexed in EMBASE/Excerpta Medica,Indexed in EMBASE/Excerpta Medica,Science Citation Index Expanded (SCIE) and ScopusScience Citation Index Expanded (SCIE) and ScopusScience Citation Index Expanded (SCIE) and ScopusScience Citation Index Expanded (SCIE) and ScopusScience Citation Index Expanded (SCIE) and ScopusFull text online at www.hkjpaed.orgFull text online at www.hkjpaed.orgFull text online at www.hkjpaed.orgFull text online at www.hkjpaed.orgFull text online at www.hkjpaed.org

EditorialIntegrating Neonatal Research and Neonatal 118Practice: A Crucial Step ForwardLam

Original ArticlesImpact of Minor Blood Group Incompatibility 120Versus ABO and Rh Blood Group Incompatibilityin Newborns with Indirect Hyperbilirubinaemia:A Single-Centre Clinical ExperienceÇelik, Aydin Koker, Ö zkul, Olukman, Karapinar,Vergin, Çalkavur

Escherichia coli Meningitis in Neonatal Intensive 127Care Units: A Five Years StudyChik, To, Luk, Lam, Shek, Tse, Ng

Non-nutritive Sucking and Nesting Greatly 135Reduces Pain During Retinopathy Screeningin Premature InfantsLiao, Xu, Zhou, Tian, Fang, Ding, Zeng, Wu

Clinical Manifestations and Outcomes of Cardiac 140Tumours in ChildrenKim, Lee, Ko, Byun, Lee, Sung, Kim, Choi

Case ReportsChinese Boy with Normal Initial Peroxisomal 147Blood Assays: A Diagnostic Pitfall in theWorkup for Infantile Refsum DiseaseCheung, Ko, Lee, Mak

What Can It Be If Not a Simple Haemangioma? 151Yu, Luk, Lo

Spontaneous Acute Epidural Haematoma in a 155Paediatric Patient with CongenitalAfibrinogenaemiaOcak, Eser Ocak, Gungorer

Cervical Thymic Cyst: Case Report 158Vu kovi , S Radovi , Sorat, Filipovi

CommentaryThe Role of Clinical Neuropsychology in thePaediatric Epilepsy Management 161Cheung

Clinical QuizWhat is the Diagnosis? 164Fung, Rethanavelu, Chung

Abstracts of Articles in Chinese 165

MCQs 169

HONG KONG JOURNAL OF PAEDIATRICS

Medcom Limited

Blank

BLANK

HK J Paediatr (new series) 2019;24:118-119

Hong Kong Journal of Paediatrics

(New Series)An Official Publication ofHong Kong College of Paediatricians &Hong Kong Paediatric Societyc/o Hong Kong College of Paediatricians, Room 801,Hong Kong Academy of Medicine Jockey Club Building,99 Wong Chuk Hang Road, Aberdeen,Hong Kong.

Editorial Board

Chief EditorCHEUNG Yiu Fai Associate EditorsCHUNG Hon Yin LAM Hung San LEUNG Ting Fan Honorary SecretaryCHAN Ching Ching MembersCHAN Chi Fung *CHAO Sih Yin CHEUK Ka Leung FUNG Po Gee HON Kam Lun HUI Wun Fung IP Patrick KWAN Yin Wah KWONG Ling LEE Pui Wah LI Albert Martin LIU Kam Wing SIU Luen YeeTSAO Yen Chow TSE Kei Chiu WONG Hiu Lei YEUNG Chap Yung YEUNG Wai Lan

Honorary Advisors to the EditorialBoardAndrew BUSH, United KingdomDon M. ROBERTON, AustraliaDavid K. STEVENSON, USAGUI Yong-Hao, China

Business ManagerTSOI Nai Shun **

*Representing HK College of Paediatricians**Representing HK Paediatric Society

PublisherHong Kong Journal of Paediatrics is publishedby Medcom Ltd, Flat E8, 10/F, Ka Ming Court,688-690 Castle Peak Road, Kowloon, HongKong SAR. Tel: (852) 2578 3833, Fax: (852)2578 3929, Email: [email protected] in EMBASE/Excerpta Medica, ScienceCitation Index Expanded (SCIE) and ScopusWebsite: www.hkjpaed.org

ISSN 1013-9923

Editorial

Of the paediatric population as a whole (children or adolescents lessthan 18 years of age), neonates (infants 28 days of age or less) accountonly for a tiny percentage (<1%). In contrast, within the eight publichospitals with deliveries in Hong Kong, perhaps a quarter to a third of thedoctors working in paediatrics devote their time to neonatal medicine. Thisapparently disproportionate workload and interest dedicated to the firstmonth of life is also reflected by the high proportion of paediatricpublications devoted to newborn infants.

Clinicians who take care of newborn infants must be able to handle all aspectsrelated to the care of the baby. From life-threatening illnesses to the medicalproblems of relatively healthy newborns, from respiratory diseases to infectionsto haematological problems. In this issue of the Hong Kong Journal ofPaediatrics, we gather several papers1-4 written by authors from around theworld. Each paper focuses on an important research field within neonatology.

In the first article,1 Çelik et al shed light on the haemolytic causes of neonataljaundice and argue that minor blood group incompatibility is an important factorto consider in babies admitted for neonatal jaundice, especially in those whopresent relatively late and have rebound jaundice after termination of therapy.In the second article,2 Chik et al describe the characteristics of newborn infantsdiagnosed with Escherichia coli meningitis over a five-year period and providesus with insights about the antibiotic resistance patterns within this population.This provides additional data which can help neonatal clinicians when decidingwhich antibiotics to use empirically. In the third article,3 Liao et al describe theresults of a randomised controlled trial that investigates the impact of non-nutritive sucking and nesting as non-pharmacological pain interventions inpreterm infants requiring eye examination. The data suggests that these relativelyeasy to implement measures could significantly improve validated pain scoresand provides valuable data upon which units may base quality improvementprogrammes targeted at pain management. In the fourth article4 by Kim et al, acohort of young children with cardiac tumours is described. While not allpresenting within the neonatal period, many of the children were diagnosedantenatally and the authors believe that early diagnosis facilitates timelyintervention during early infancy, or even the neonatal period, which couldhopefully decrease the potential complications that may arise.

As can be seen from these articles, an important function of neonatal researchis to improve our ability to care for our newborn patients. The investigatorsfeatured in this issue of our Journal have achieved this by painstakingly recordingcharacteristics of patient cohorts with relatively well-defined conditions ofinterest and then analysing and presenting their data so that clinicians from aroundthe world can learn and reflect on how to apply their findings to their ownnewborn patients. However, despite the overwhelming explosion of availabledata derived from studies from around the world, the data available are not always

Integrating Neonatal Research andNeonatal Practice: A Crucial Step Forward

119 Integrating Neonatal Research and Neonatal Practice

applicable to our own population, whether because ofdifferences in the clinical setting or the variations incharacteristics between the study's patients and the patientsunder our own care. Hence even with good quality data,applying evidence to one's own patients may not bestraightforward and it is often necessary to implement thefindings in a systematic way by measuring outcomes afterimplementation and modifying the approach accordingly.

Another way that good practice can be disseminated andlarge sets of clinically relevant data collected, is by poolingdata of clinically important neonatal characteristics andoutcomes between large numbers of neonatal units. In thisway, some of the limitations of small, single centre studiescan be overcome. It has increasingly been recognised that itis an important part of continuous improvement for neonatalunits to routinely measure important clinical outcomes forbenchmarking in order to identify areas to focus qualityimprovement and research on. To this end, it has becomeincreasingly apparent that joining well-designed neonatalnetworks with an emphasis on quality improvement5-7 may bethe way forward as these multi-centre networks not onlypave the way for big data analyses to tackle clinical importantproblems, but also serve as robust platforms for large-scalestudies such as point-of-care trials8 which may provideinsights that have eluded us despite decades of neonatalresearch.

HS LamAssociate Editor

References

1. Çelik K, Aydin Koker S, Özkul MT, Olukman Ö, Karapinar TH,Vergin RC, Çalkavur S. Impact of Minor blood groupincompatibility versus ABO and Rh blood group incompatibility innewborns with indirect hyperbilirubinaemia: a single centre clinicalexperience. HK J Paediatr (new series) 2019;24:120-6.

2. Chik KK, To WK, Luk K, Lam B, Shek CC, Tse C, Ng D.Escherichia coli meningitis in neonatal intensive care units: A fiveyears study. HK J Paediatr (new series) 2019;24:127-34.

3. Liao WH, Xu HY, Zhou X, Tian JL, Fang XY, Ding J, Zeng C, WuHL. Non-nutritive sucking and nesting greatly reduces pain duringretinopathy screening in premature infants. HK J Paediatr (newseries) 2019;24:135-9.

4. Kim G, Lee N, Ko H, Byun JH, Lee HD, Sung SC, Kim H, ChoiKH. Clinical manifestations and outcomes of cardiac tumours inchildren. HK J Paediatr (new series) 2019;24:140-6.

5. Battersby C, Statnikov Y, Santhakumaran S, et al; UK NeonatalCollaborative and Medicines for Neonates Investigator Group. TheUnited Kingdom National Neonatal Research Database: A validationstudy. PLoS One 2018;13:e0201815.

6. Lee SK, Aziz K, Singhal N, Cronin CM. The Evidence-basedPractice for Improving Quality method has greater impact onimprovement of outcomes than dissemination of practice changeguidelines and quality improvement training in neonatal intensivecare units. Paediatr Child Health 2015;20:e1-9.

7. Shah PS, Lee SK, Lui K, et al; International Network for EvaluatingOutcomes of Neonates (iNeo). The International Network forEvaluating Outcomes of very low birth weight, very pretermneonates (iNeo): a protocol for collaborative comparisons ofinternational health services for quality improvement in neonatalcare. BMC Pediatr 2014;14:110.

8. Gale C, Modi N; WHEAT trial development group. Neonatalrandomised point-of-care trials are feasible and acceptable in theUK: results from two national surveys. Arch Dis Child Fetal NeonatalEd 2016;101:F86-7.


Impact of Minor Blood Group Incompatibility Versus ABO andRh Blood Group Incompatibility in Newborns with IndirectHyperbilirubinaemia: A Single-Centre Clinical Experience

K ÇELIK, S AYDIN KOKER, MT ÖZKUL, Ö OLUKMAN, TH KARAPINAR, RC VERGIN, ÇALKAVUR

Abstract Introduction: Haemolytic disease of the fetus and newborn (HDFN) is caused by the destruction of redblood cells of the neonate or fetus by maternal immunoglobulin G (IgG) antibodies. Alloimmune HDFNprimarily involves the major blood groups of Rhesus (Rh), A, B, AB, and O, although minor blood groupincompatibilities (Kell, Duffy, MNS, P, and Diego systems) can also result in significant disease. Objective:The main objective was to provide insight about the impact of minor blood group incompatibility againstABO and Rh blood group incompatibility in newborns with hyperbilirubinaemia in terms of the demographicdata, laboratory values, clinical course, and responses to treatment. Methods: Three hundred fifty-fiveinfants hospitalised with the diagnosis of unconjugated hyperbilirubinaemia (UHB) had data collectedretrospectively. The mother and the baby's blood group type as C, c, E, e, Kell, D, d, A, B, O was detectedby gel centrifugation method. Patients were divided into 4 groups: group 1, ABO blood groupincompatibility; group 2, Rh blood group incompatibility; group 3, ABO+Rh blood group incompatibility;and group 4, minor blood group incompatibility. A total of 355 patients of which 230 (64.7%) were ingroup 1, 68 (19.1%) were in group 2, 15 (4.2%) were in group 3, and 42 (12%) were in group 4 wereevaluated, respectively. Results: Among those with minor blood group incompatibility, 12 (28%), 11(26%), 9 (21%), 6 (14%), and 4 (9%) patients had "C," "c," "E+c," "E," and "Kell" incompatibilities,respectively. The mean age of diagnosing jaundice (8±7.5 days) was significantly higher (p=0.015), andthe mean haemoglobin and the mean haematocrit levels at admission were lower (p=0.007, p=0.041) ingroup 4. The rebound elevation of serum bilirubin was remarkably high in group 2 and group 4(p=0.025). The requirement of intravenous immunglobulin, exchange-erythrocyte transfusions, as well asrehospitalisation for phototherapy, after discharge was significantly higher in group 4. Conclusions: Weshould keep mind the minor blood group incompatibility in infants who have late hospital admission,prolonged jaundice, prominent anaemia, rebound elevation of serum bilirubin after treatment termination,requirement of exchange and/or erythrocyte transfusions, and increased rehospitalisation rate.

Key words ABO; Indirect hyperbilirubinaemia; Neonatal; Rh and minor blood group incompatibility

Department of Neonatology, Dr. Behçet Uz Children'sHospital, Izmir, Turkey

K ÇELIK MDÖ OLUKMAN MD, Associate Professor ÇALKAVER MD, Associate Professor

Department of Pediatrics, Dr. Behçet Uz Children's Hospital,Izmir, Turkey

MT ÖZKUL MD

Original Article

Department of Pediatric Hematology, Dr. Behçet UzChildren's Hospital, Izmir, Turkey

S AYDIN KOKER MDTH KARAPINAR MD, Associate ProfessorRC VERGIN MD, Associate Professor

Correspondence to: Dr S AYDIN KOKER

Email: [email protected]

Received October 10, 2017

ABO, RhD and Minor Blood Group Incompatibility121

Introduction

Neonatal jaundice is the most common cause ofhospitalisation in the neonatal period. It is estimated that60% of term newborns develop jaundice and 2% reach asevere hyperbilirubinaemia. In 2006, Sgro et al publishedan incidence of severe hyperbilirubinaemia of 1/2480 (notfor haemolytic anaemia);1 in 2007, it decreased to 1/8600after the implementation of a Guideline of ClinicalPractice aimed at improving the management of severeneona t a l j aund ice . 2 , 3 R i sk f ac to r s fo r s eve rehyperbilirubinaemia include malnutrition, infection,history of jaundice in sibling(s), the presence of a largecephalhaematoma, premature labour, sequestration,polycythemia, isoimmunisation, erythrocyte enzymatic andstructural defects, and disorder of conjugation.1,4 The mostcommon cause of severe early jaundice is fetal-maternalblood group incompatibility with resulting isoimmunisation.Maternal immunisation develops when erythrocytes leakfrom fetal to maternal circulation. Fetal erythrocytescarrying different antigens are recognised as foreign by thematernal immune system that forms antibodies againstthem (maternal sensitisation). These antibodies (IgGimmunoglobulins) cross the placental barrier into the fetalcirculation and bind to fetal erythrocytes.5 This disease ofthe fetus caused by a maternal response to pregnancy iscalled the haemolytic disease of the fetus and newborn(HDFN). Although Rhesus (Rh) blood group incompatibilityis well known in HDFN, the ABO and minor blood groupincompatibility have become the increased frequent forHDFN in developed countries by virtue of the common usingof anti-D immunoglobulin. In Rh incompatibility,sequestration and destruction of the antibody-coatederythrocytes take place in the reticuloendothelial system ofthe fetus. In ABO incompatibility, haemolysis isintravascular, complement-mediated, and usually not assevere as Rh disease. Significant haemolysis can also resultfrom incompatibilities between minor blood group antigens(e.g., Kell).6

In this study, we aimed to compare the demographic data,laboratory values, clinical course, and responses fortreatment between minor blood, Rh, and ABO blood groupantigens incompatibility in newborns with HDFN, whowere hospitalised due to jaundice.

Materials and Methods

A total of 355 patients who have been hospitalised

between January 1, 2013, and December 31, 2015, inneonatal service of Dr. Behçet Uz Children's Hospital withindirect hyperbilirubinaemia defined as >95p belowaccording to nomogram based on the hour-specific serumbilirubin values recommended in the 2004 guideline of theAmerican Academy of Pediatrics (AAP)3 for neonates witha gestational age of 35 weeks or more and detected to bewith Rh, ABO, or minor blood group incompatibility wereincluded in the study. The study data were collected via aretrospective review of the respective patient files whichwas carried out with the approval of the local ethicscommittee (14.01.2016-01.06.2016). Patients with agestational age of less than 37 weeks and with glucose 6-phosphate dehydrogenase (G6PDH) deficiency wereexcluded. Demographic data, such as sex, gestationalweek at delivery, mode of delivery, age at admission,birth weight, body weight at admission, maternal age, thepresence of dehydration, and laboratory values, such astotal and direct bilirubin levels, blood groups of motherand baby, direct Coombs test, haemolysis findings onperipheral blood smear, reticulocyte, haemoglobin,haematocrit, aspartate aminotransferase (AST), alanineaminotransferase (ALT), total bilirubin/albumin ratio, andresponses to treatment and clinical courses such as levelat discontinuation of phototherapy, decreased in totalserum bilirubin (TSB) level per hour, phototherapy time,rebound bilirubin level after phototherapy, othertreatments (intravenous immunoglobulin (IVIG),erythrocyte transfusion, exchange transfusion, intravenousfluid) and rehospitalisation need, were reviewed in thepatient files. Dehydration (severe) was defined the lossof >10% in body weight. Rehospitalisation werehospitalised again within 30 days of newborn's indexadmission. Bilirubin rebound was described as post-phototherapy bilirubin level needing reinstitution ofphototherapy. Phototherapy and/or exchange transfusionprocedure were carried out in patients according to curvesrecommended in the 2004 guideline of the AAP for neonateswith a gestational age of 35 weeks or more.7 In line with therecommendations of the AAP, an IVIG treatment at a doseof 1 g/kg was administered in case of a continuing increasein total bilirubin level or the presence of a bilirubin levelwithin the range of ± 2 to 3 mg/dl to exchange transfusionl imit in HDFN despi te intensive phototherapy.An intravenous fluid replacement was administeredto l e tha rg ic pat ients reluctant to oral feedingwith a weight loss of >10%. ABO and Rh blood groupincompatibility, dehydration, sepsis, haematoma, thepresence of a level above the exchange transfusion limit

Çelik et al 122

without G6PDH deficiency at admission, the absence of theexpected response to phototherapy or reticulocytosis,anaemia, and C, c, E, e, Kell, Duffy, Diego, Kidd, MNSsantigens in both mothers and babies in the presence of directCoombs test positivity were studied in patients. The motherand the baby's blood group type as C, c, E, e, Kell, D, d, A,B, O was detected by gel centrifugation method. Patients asonly ABO, only Rh, both ABO and Rh blood groupincompatibilities, or minor blood group incompatibility wereseparated to group 1, group 2, group 3, or group 4,respectively. Demographic data, laboratory values,responses to treatment, and clinical courses were comparedbetween groups. The statistical analysis was carried out byusing SPSS for Windows, version 20.0. For comparingnumerical variables between groups, ANOVA test (if thedata fit a normal distribution) and Kruskal Wallis test forthe nonparametric were used. For comparing ratios betweengroups, the chi square test was used. The statisticalsignificance was accepted as p<0.05.

Results

A total of 355 patients, of which 230 (64.7%) were ingroup 1, 68 were (19.1%) in group 2, 15 (4.2%) were ingroup 3, and 42 (12%) were in group 4 were evaluated.Among those with minor blood group incompatibility, 12

(28%), 11 (26%), 9 (21%), 6 (14%), and 4 (9%) patientshad "C," "c," "E+c," "E," and "Kell" incompatibilities,respectively. When groups were compared in regard todemographic data, no significant difference was detectedwith respect to sex, gestational week at delivery, birth order,and the presence of dehydration. Birth weight in group 4was lower (3085±440 g) compared with other groups(p=0.001). Mean diagnosis age for all groups was 5.4±4.5days, and postnatal age at admission in group 4 (8±7.5 days)was significantly higher compared with other groups(p=0.015). The detailed distribution of demographic datais shown in Table 1 both in general and separately.

When laboratory data obtained from patients atadmission were evaluated, it was found that haemoglobinand haematocrit values in group 4 were statisticallysignificantly lower compared with the other groups(p=0,007 and p=0.041, respectively). There was nosignificant difference between groups regarding otherlaboratory data. However, TSB, direct bilirubin, level atdiscontinuation of phototherapy, reticulocyte, Haemolysisfinding on peripheral blood smear were shown to be higherin group 4, although not statistically significant. The detaileddistribution of laboratory data is given in Table 2 both ingeneral and separately.

No significant difference was found between groups inregard to total phototherapy time applied to patients as fromthe first hospitalisation, decrease rate of bilirubin, and IVIG

Table 1 Distribution of demographic data according to patient groups

General Group 1 Group 2 Group 3 Group 4 p

% (n) % (n) % (n) % (n) % (n)

Sex Boy 53.2 (189) 53 (122) 54 (37) 53 (8) 52 (22)Girl 46.7 (166) 47 (108) 46 (31) 47 (7) 48 (20)

Mode of deliveryNSVD 36.9 (131) 32 (74) 35 (24) 60 (9) 57 (24)C/S 63.1 (224) 68 (156) 65 (44) 40 (6) 43 (18)

DehydrationYes 9.6 (34) 7 (18) 11.7 (8) 13 (2) 14 (6) 0.345No 90.4 (321) 93 (212) 88.3 (60) 87 (13) 86 (36)

*Gestational week at delivery 38.2±1.2 38.4±1.2 38.4±1.1 39±0.8 38.1±1.5

*Birth weight (g) 3321±440 3324±409 3425±461 3403±556 3085±440 0.001

*Diagnosis age (days) 5.4±4.5 5.1±3.5 5±3.9 4.1±3.3 8±7.5 0.015

*Maternal age 28.2±5.2 27.9±5.1 29.9±5.3 26.2±5 27.8±5.7 0.015

*Pregnancy number 2.2±1.3 2.2±1.2 2.4±1.4 2.4±1.3 2±1.2

NSVD=Normal spontaneous vaginal delivery; C/S=Cesarean section; *Data are shown as mean ± standard deviation.


need. However, the need for IVIG in group 4 wasnumerically higher compared with the other groups.Although exchange transfusion was administered to 1.6%(n=6) of all patients, the rate of exchange transfusionadministration in the minor blood group incompatibility(group 4) was 11.9 % (n=5) of the total study patients. Theneed of exchange transfusion and transfusion witherythrocyte suspension in Group 4 was statisticallysignificantly higher compared with the other groups(p=0.001 and p<0.001, respectively).

Rebound bilirubin levels after phototherapy weresignificantly higher in patients with RhD incompatibility andminor blood group incompat ibi l i ty (p=0.025) .Rehospitalisation due to unconjugated hyperbilirubinaemia(UHB) after discharge was significantly more commonamong patients with minor blood group incompatibility(p=0.005). Responses to treatment and rehospitalisationrates of patients are given in Table 3.

Discussion

The most common cause of Haemolytic disease ofthe newborn is Rh, ABO, and minor blood groupincompatibilities, resulting from antibodies formed in themother against neonatal erythrocyte antigens. Before the

administration of anti-D immunoglobulin, the frequency ofRh alloimmunisation was about 16%, and it was the mostfrequently encountered cause of mortality- and morbidity-related Haemolytic disease in fetal and neonatal periods.By virtue of administration of anti-D immunoglobulin inthe antepartum and postpartum period, the frequency ofalloimmunisation has decreased to 0.17% to 0.28%, and thefrequency of Haemolytic disease-related mortality hasdecreased from 46/100,000 to 1.6/100,000.8,9 Anti-D is stillregarded as one of the critical antibodies because ofconditions, such as unmonitored pregnancies, suboptimalresponses to immunoprophylaxis, and the like. In additionto RhD antigen, erythrocyte antigens related with Haemolyticdiseases include non-D Rh antigens (C, c, E, e), ABOsystem antigens, Kell, and, as more rarely seen ones, Duffy,Kidd, and MNS antigens.10 In our study, the most commonlyreported causes were ABO blood group incompatibilitywith the ratio of 64.7%, Rh blood group incompatibilitywith the ratio of 19.1%, minor blood group incompatibilitywith the ratio of 12%, and ABO plus Rh blood groupincompatibility with the ratio of 4.2%, respectively. In thestudy by Karagöl et al in which 106 patients with minorblood group incompatibility were evaluated, mean postnatalage at admission was reported as 6.1±5.2 days.11 The meandiagnosis age of all patients was found as 5.4±4.5 days inour study. The mean diagnosis age in patients with minor

Table 2 Comparison of patient groups for laboratory values


(mean±±±±±SD) (mean±±±±±SD) (mean±±±±±SD) (mean±±±±±SD) (mean±±±±±SD)

Total bilirubin (mg/dl) 19.5±4.9 19.6±4 18.1±4.8 20.7±6.3 21.1±7.6 0.106

Direct bilirubin (mg/dl) 0.6±0.5 0.6±0.2 0.5±0.16 0.5±0.14 1±1.5 0.168

Level at discontinuation of 10.9±5.5 11.3±6.2 10.3±1.9 9.5±1.4 11.5±5.9 0.468phototherapy (mg/dl)

Reticulocyte (%) 3.8±4.2 3.6±3.9 3.2±3 3.3±2.8 5.2±6.7 0.083

Haemoglobin (g/dL) 15.4±2.9 15.5±2.4 16.3±3.2 15.3±2.2 14.6±4.5 0.007

Haematocrit (%) 46.1±8.4 45.9±6.9 48.2±9.4 46.1±6.8 44.5±12.6 0.041aAST (IU/L) 51±30.6 50±26.5 57±45 60.7±20.7 48.1±19.5 0.162bALT (IU/L) 18.6±12.7 18.5±113.7 19.8±12 17.6±5.6 17.8±10.4 0.837

Total bilirubin/albumin 5.5±1.4 5.5±1.2 5.1±1.3 5.8±1.8 5.8±2.5 0.067

*Direct Coombs positivity 12.9 (46) 10.4 (24) 16.1 (11) 20 (3) 19 (8) 0.378

*Haemolysis finding on 20.5 (73) 19.5 (45) 19.1 (13) 6.6 (1) 33.3 (14) 0.148peripheral blood smear

*Normal spontaneous vaginal delivery; a Cesarean section; b Data are shown as mean ± standard deviation.

Çelik et al 124

blood group incompatibility was as 8±7.5 days, which wasstatistically significantly higher compared with the otherstudy groups. The underlying reason for this significantdifference may be the increased awareness both in familiesand health institutions as a result of routine scans forABO and Rh blood group incompatibilities carried outwithin the scope of antepartum blood group analyses.In some countries, scans for non-anti-D erythrocytealloimmunisation, with respect to minor blood groupincompatibility, are carried out in pregnant women due tothe risk of severe Haemolytic disease, even though the riskis very low.8 However, there are no such routine scans inTurkey.

T h e A A P g u i d e l i n e o n m an a g e m e n t o fhyperbilirubinaemia highlights the importance of thepresence of Haemolysis, with respect to hyperbilirubinaemiaand neurotoxicity, and states that haemolytic disease shouldbe primarily considered if jaundice is detected and thepredischarge hour-specific TSB nomogram shows a TSBvalue within the high risk zone, especially in the first 24hours.12,13 Because haemolytic and nonhaemolytic factorsare generally seen together in neonatal period, assessingTSB, reticulocyte, direct Coombs test, complete bloodcount, and erythrocyte morphology on peripheral bloodsmear may be insufficient to evaluate Haemolytic diseasein these patients.14 In another study carried out by Bolat etal on hyperbilirubinaemic neonates with ABO or RhDincompatibility, the researchers have determined that therate of severe hyperbilirubinaemia in patients with ABOincompatibility was significantly higher.15 In our study, nosignificant difference was found in laboratory findingsbetween groups, except lower haemoglobin and haematocrit

levels reported in patients with minor blood groupincompatibility. These findings suggest that newborns withminor blood group incompatibility are exposed toHaemolysis for a longer time because this patient groupapplies to a hospital later compared with the other groups.

The most valid treatment method in UHB isphototherapy. In high TSB levels unresponsive tophototherapy, exchange transfusion should be urgentlyadministered. Other medical treatment options apart fromIVIG administration, such as phenobarbital andmetalloporphyrins, are less frequently preferred.Particularly, in the presence of haemolytic disease of thenewborn, IVIG is recommended if there is a TSB levelwithin the limits of exchange transfusion or an increasedTSB level despite phototherapy.16 In the study by Bolat etal, the need for IVIG administration was more frequentlyreported among pat ients with Rh blood groupincompatibility,15 and in an another study by Nasseri et al, ashorter phototherapy time, a lower rate of exchangetransfusion requirement, and a shorter hospitalisation periodwere reported for neonates receiving IVIG with a diagnosisof haemolytic disease of the newborn.17 According to Bekenet al, IVIG therapy did not decrease phototherapy norhospitalisation duration in infants with Haemolytic disease-related ABO blood group incompatibilities. IVIG is notsuccessful for decreasing haemolysis but IVIG preventsrequiring exchange transfusion.18 Likewise, in the Demirelet al's study, IVIG therapy, single or multiple, did notaffect exchange transfusion, need of erythrocytetransfusion, and hospitalisation when used in combinationwith phototherapy.19 In the study by Karagöl et al, inwhich the researchers have evaluated 106 patients with

Table 3 Responses to treatment and clinical course in patient groups


*Phototherapy time (hours) 37.1±20.1 37.6±21.5 35.1±16.6 34.6±19 38.8±17.8 0.729

*Decrease in total serum 0.37±0.17 0.41±0.21 0.24±0.15 0.33±0.19 0.28±0.16 0.903

bilirubin level per hour

Rebound bilirubin 18 (64) 14.4 (33) 22 (15) 13.3 (2) 33.3 (14) 0.025

Intravenous fluid need 13.5 (48) 9.1 (21) 8.8 (6) 40 (6) 35 (15) 0.000

**IVIG need 16 (57) 14.3 (33) 14.7 (10) 6 (1) 30 (13) 0.067

Exchange transfusion need 1.6 (6) 0.4 (1) − − 11.9 (5) 0.001

Erythrocyte transfusion need 4.5 (16) 0.8 (2) 4.4 (3) 6.6 (1) 23.8 (10) <0.001

Rehospitalisation 6.2 (22) 5.2 (12) 1.4 (1) 6.6 (1) 19 (8) 0.005

*Data are shown as mean ± SD and % (n); **Intravenous immunoglobulin


minor blood group incompatibility, 20.8% of all patientshave been subjected to exchange transfusion.11 In our study,the rate of exchange transfusion requirement was 1.6% forall groups, and the rate of exchange transfusionadministration in the minor blood group incompatibility(group 4) was 11.9% (n=5) of the total study patients. Inour study, there was no significant difference between groupswith respect to the need for IVIG, whereas in patients withminor blood incompatibility, the need for exchangetransfusion and erythrocyte transfusion was significantlyhigher compared with the other patients. In patients whohave been subjected to IVIG treatment, no difference wasfound compared with the other patients regardingphototherapy time, exchange transfusion requirement, andrebound increase in TSB.

In our study, the rate of intravenous fluid need was higherin groups with minor blood group incompatibility andABO+Rh blood group incompatibility compared with theother groups. This may be the result of the presence of ahigher rate of dehydration in these two groups, although itwas not statistically significant. Furthermore, for patientswith minor blood group incompatibility, the higher rate ofintravenous fluid use may be associated with theintroduction of intravenous fluid administration withdiscontinuation of oral feeding which occurred because ofthe presence of a higher need for exchange transfusion anderythrocyte transfusion in the patient group.

Bilirubin levels may increase again after phototherapy,and a postdischarge rehospitalisation may be observedas a result thereof. The presence of haemolytic jaundiceis a critical risk factor for rebound bilirubin levels. Otherrisk factors include premature labour, the onset time andseverity of jaundice, and G6PDH deficiency and diet.12,20

In a study by Bansal et al, the researchers have reportedrebound bilirubin levels in 7.3% of patients withhyperbilirubinaemia.21 In our study, the frequency valuewas found to be 18% for all patients, and patients withRh or minor blood group incompatibility were morefrequently associated with rebound bilirubin levels. Thefrequency of postdischarge rehospitalisation was reportedas 13.3% in a study by Kaplan et al.20 In our study, thefrequency value was 6.2% in general and was significantlyhigher among patients with minor blood groupincompatibility. This overall high rate reported in ourstudy was thought to be dependent on the fact that unlikeother studies, we have only included the patients withHaemolytic disease of the newborn in our study. Also,the patients with minor blood group incompatibility was

associated with lower haemoglobin and haematocrit levels.The authors have probably checked potential

incompatibility at other blood group antigens when ABOor RhD incompatibility did not explain the Haemolyticdisease of a newborn. A higher rate of incompatibility couldbe detected, if we have routinely examined suitableparameters in all newborns.

Conclusion

Consequently, it was found that minor blood groupincompatibility constituted 12% of all blood groupincompatibilities among patients included in this study.The older postnatal age at admission, more frequent co-occurrence of anaemia, rebound TSB levels, higher ratesof intravenous fluid, exchange transfusion and erythrocytetransfusion needs, and higher rate of rehospitalisationwere seen in the patients with minor blood groupincompatibilities compared with the other patients.Therefore , when ABO and/or Rh b lood groupincompatibilities are not detected in neonates withjaundice requiring hospitalisation, minor blood groupincompat ib i l i t ies should a lways be taken in toconsideration.

Conflict of Interest

The authors have no conflicts of interest to disclose.

References

1. Sgro M, Campbell D, Shah V. Incidence and causes of severeneonatal hyperbilirubinemia in Canada. CMAJ 2006;175:587-90.

2. Sgro M, Kandasamy S, Shah V, Ofner M, Campbell D. SevereNeonatal Hyperbilirubinemia Decreased after the 2007 CanadianGuidelines. J Pediatr 2016:43-7.

3. Guidelines for detection, management and prevention ofhyperbilirubinemia in term and late preterm newborn infants (35or more weeks' gestation). [French, English]. Paediatr ChildHealth 2007;12:401-18.

4. Maisels MJ, Bhutani VK, Bogen D, Newman TB, Stark AR,Watchko JF. Hyperbilirubinemia in the newborn infant > or =35weeks' gestation: an update with clarifications. Pediatric 2009;124:1193-8.

5. Zipursky A. Mechanisms of hemolysis. Mead Johnson SympPerinat Dev Med 1982;19:17-24.

6. Wenk RE, Goldstein P, Felix JK. Kell alloimmunization,

Çelik et al 126

hemolytic disease of the newborn, and perinatal management.Obstet Gynecol 1985;66:473-6.

7. American Academy of Pediatrics, Steering Committee on QualityImprovement and Management. A taxonomy of recommendations.Pediatrics 2004;114:874-7.

8. Qureshi H, Massey E, Kirwan D, et al. British Society forHaematology. BCSH guideline for the use of anti-Dimmunoglobulin for the prevention of haemolytic disease of thefetus and newborn. Transfus Med 2014;2:8-20.

9. Rath ME, Smits-Wintjens VE, Walther FJ, Lopriore E.Hematological morbidity and management in neonates withhemolytic disease due to red cell alloimmunization. Early HumDev 2011;87:583-8.

10. Fasano RM. Hemolytic disease of the fetus and newborn in themolecular era. Semin Fetal Neonatal Med 2016;21:28-34.

11. Karagol BS, Zenciroglu A, Okumus N, Karadag N, Dursun A,Hakan N. Hemolytic disease of the newborn caused by irregularblood subgroup (Kell, C, c, E, and e) incompatibilities: Report of106 cases at a tertiary-care centre. Am J Perinatol. 2012;29:449-54.

12. American Academy of Pediatrics Subcommittee onHyperbilirubinemia. Management of hyperbilirubinemia in thenewborn infant 35 or more weeks of gestation. Pediatrics 2004;114:297-316.

13. Maisels MJ, Bhutani VK, Bogen D, Newman TB, Stark AR,Watchko JF. Hyperbilirubinemia in the newborn infant ≥35 weeksgestation: an update with clarifications. Pediatrics 2009;124:1193-8.

14. Kaplan M, Bromiker R, Hammerman C. Hyperbilirubinemia,hemolysis, and increased bilirubin neurotoxicity. Semin Perinatol2014;38:429-37.

15. B o l a t F, U s l u S , B u l b u l A . Ye n i d o g an i n d i r e k thiperbilirubinemisinde ABO ve Rh uygunsuzlugununkars last r lmas . S.E.E.A.H. T p Bulteni 2010;44:156-61.

16. Kaplan M, Wong RJ, Sibley E, Stevenson DK. Neonatal jaundiceand liver diseases. In: Martin RJ, Fanaroff AA, Walsh MC, eds.Fanaroff and Martin's Neonatal-Perinatal Medicine: Diseases ofthe Fetus and Infant. 10th ed. Philadelphia, PA: Elsevier Saunders;2015:chapter 100.

17. Nasseri F, Mamouri GA, Babaei H. Intravenous immunoglobulinin ABO and Rh hemolytic diseases of newborn. Saudi Med J 2006;27:1827-30.

18. Beken S, Hirfanoglu I, Turkyilmaz C, et al. IntravenousImmunoglobulin G Treatment in ABO Hemolytic Disease of theNewborn, Is It Myth or Real? Indian J Hematol Blood Transfus2014;30:12-5.

19. Demirel G, Akar M, Celik IH, et al. Single versus multiple doseintravenous immunoglobulin in combination with LEDphototherapy in the treatment of ABO hemolytic disease inneonates. Int J Hematol. 2011;93:700-3.

20. Kaplan M, Kaplan E, Hammerman C, et al. Post phototherapyneonatal bilirubin rebound: a potential cause of significanthyperbilirubinaemia. Arch Dis Child 2006;91:31-4.

21. Bansal A, Jain S, Parmar Vr, Chawla D. Bilirubin rebound afterintensive phototherapy for neonatal jaundice. Indian Pediatr 2010;47:607-9.


Escherichia coli Meningitis in Neonatal Intensive Care Units:A Five Years Study

KK CHIK, WK TO, K LUK, B LAM, CC SHEK, C TSE, D NG

Abstract Introduction: Neonatal meningitis causes severe morbidity. The incidence of ampicillin-resistantEscherichia coli (E. coli) meningitis was reported to be increasing in overseas neonatal intensive careunits (NICUs) recently. This is the first local study to investigate E. coli neonatal meningitis. Choice ofempirical antibiotics for intrapartum period and neonates will be discussed. Methods: Retrospective studywas performed in 2 NICUs between 1st January 2011 and 31st May 2016. Neonates with E. coli meningitiswere included. Demographic data, laboratory results, antibiotics use and antimicrobial susceptibilities werereviewed. Results: Eleven neonates were identified and 55% had early-onset (EO) meningitis. Ninety-onepercent of E. coli isolates were ampicillin-resistant. Twenty-seven percent were extended-spectrum-beta-lactamase (ESBL) producing and all were isolated in EO meningitis group. Use of intrapartum ampicillinand ESBL-E. coli infection in neonates showed no statistical significance. Conclusions: The risk ofampicillin-resistant E. coli meningitis in neonates continues despite of using intrapartum antibiotics inhigh risk group. Choice of intrapartum antibiotic and empirical treatment of neonatal sepsis has an urgentneed to be reviewed.

Key words Antibiotic resistance; Escherichia coli; Neonatal meningitis

Department of Pathology, Princess Margaret Hospital,2-10 Princess Margaret Hospital Road, Lai Chi Kok,Kowloon, Hong Kong, China

KK CHIK FHKAM(Path), HKCPath, FHKAM(Paed)WK TO FHKAM(Path), FHKCPathK Luk FHKAM (Path), FHKCPathB LAM FHKAM (Path), FHKCPath

Department of Paediatrics, Princess Margaret Hospital,2-10 Princess Margaret Hospital Road, Lai Chi Kok,Kowloon, Hong Kong, China

CC SHEK FHKAM(Paed), FHKCPaed

Department of Pathology, Kwong Wah Hospital ,25 Waterloo Road, Yaumatei, Kowloon, Hong Kong, ChinaC TSE FHKAM(Path), FHKCPath

Department of Paediatrics, Kwong Wah Hospital,25 Waterloo Road, Yaumatei, Kowloon, Hong Kong, China

D NG FHKAM(Paed), FHKCPaed

Correspondence to: Dr KK CHIK


Received November 21, 2017

Original Article

Introduction

Neonatal meningitis is a serious infection causing highmortality and morbidity including neurological disability.The major pathogens causing neonatal meningitis areStreptococcus agalactiae (GBS) and Escherichia coli(E. coli). The United States started the national screeningprogramme for GBS in pregnancy in 1996, since then theincidence of early onset group B streptococcus diseasedecreased from 1.7 per 1000 live births to 0.6 per 1000.1,2

However, the incidence of neonatal E. coli sepsis andmeningitis showed an increasing trend.3,4 There were alsocase reports of ampicillin-resistant E. coli neonatal infectionafter implementing the universal intrapartum antibioticprophylaxis to maternal Group B Streptococcus carriers.3,5

Not just ampicillin-resistance, infection caused by extendedspectrum β-lactamases (ESBL) producers are also spreadingrapidly. ESBL enzyme results in resistance to manycommonly used beta-lactam antibiotics, including ampicillin,first, second and third generation cephalosporin. This mayaffect the choice of empirical antibiotic in neonatal sepsis.Hong Kong started the antenatal screening and treatment

Chik et al 128

programme for maternal GBS in 2012. There was no localstudy on neonatal E. coli meningitis all along. To addressthis issue, we carried out the first local retrospective studyto describe the demographic data of E. coli neonatalmeningitis and the antibiotic susceptibility pattern of theisolates between 2011 and 2016. We investigate andanalyse the association between demographic data,laboratory findings, use of intrapartum ampicillin anddrug resistance E. coli neonatal infection, also reviewthe choice of empirical intrapartum antibiotic andempirical neonatal antibiotics.

Methods

Study DesignA retrospective study performed in two Hong Kong

neonatal intensive care units (NICU). The demographic datawere reviewed including the gestational age, days ofsymptoms onset, gender, birth weight, maternal andneonates' laboratory results, intrapartum antibiotic andempirical antibiotic use in neonates. Their neurologicaloutcomes were also reviewed. In addition, the antimicrobialsusceptibilities of the isolates were analysed. Data werecollected between 1st January 2011 and 31st May 2016via Clinical Data Analysis and Reporting System of HongKong Hospital Authority.

PatientsNeonates </=28 days of age and microbiological

confirmed E. coli meningitis were included. The diagnosisof meningitis was based on positive culture of E. coli fromcerebrospinal fluid (CSF) and/or from blood cultureassociated with pleocytosis (>/=10 cells/mm3) in CSF. Theincluded patients were categorised into early onset neonatalmeningitis and late onset neonatal meningitis according totheir gestational age and the time of symptoms onset. Earlyonset neonatal meningitis was defined as onset of bacterialmeningitis at ≤72 hours in preterm infants and ≤7 days interm infants. Late onset meningitis was defined as the onsetoccurred after 72 hours of life in preterm babies and >7days up to 90 days in term infants.

Statistical AnalysisExcel spreadsheets were used for tabulation of data.

Statistical analysis was performed using the StatisticalPackage for the Social Sciences (Windows version 17;SPSS Inc, Chicago [IL], US) Fisher's exact test and Mann-Whitney U test were used for categorical variables and

continuous variables respectively to analyse the associationbetween study factors and ESBL E. coli infection amongneonates. P values of <0.05 were considered statisticallysignificant.

Ethics ReviewThis proposal was reviewed and approved by the

Research Ethics Committee of the Kowloon West Cluster(Kowloon West; REC [KWC REC No.: 107-07]).

Results

Demographics and Maternal Peripartum Risk Factorsfor Neonatal Sepsis

Between 2011 and 2016, the total number of live bornin Hong Kong public hospitals was around 227, 300 and58, 113 babies were born in the two study hospitals. Itaccounted for 25% of all living newborns in all Hong Kongpublic hospitals. With eleven cases of neonatal E. colimeningitis were identified in this study, the prevalence rateof E. coli meningitis infection in neonates of the studyhospitals was estimated to be 0.02% (11/58113). This wasbased on the assumption that all cases delivered in these 2hospitals would be admitted to the same hospital's NICUwhile babies with E. coli meningitis delivered in otherhospitals were not admitted to these 2 NICUs. Demographicdata were shown in Table 1a, 1b and Table 2. Six of themwere male (male to female ratio 1.2:1). Mean gestationalage was 34 weeks (range 29-40 weeks). Fifty-five percentwere preterm babies. The mean age of onset was 10.7 days(range 0-28 days). Six had early onset infection and 5 hadlate onset. Mean birth weight was 2.31 kg (range 1.36-3.61kg). Among those with early onset disease, 83% hadmaternal risk factors for neonatal sepsis, includingprematurity, premature rupture of membranes, maternalintrapartum fever and acute chorioamnionitis confirmed byplacental histology. Although intrapartum intravenousampicillin was given as per protocol, their babies still hadearly onset meningitis with ampicillin-resistant E. coli.

Clinical Presentation and Laboratory ResultsEight neonates presented with fever, 2 had recurrent

apnoea and 1 had malaise and feeding intolerance. Majoritydid not have congenital abnormalities except one hadHirschsprung's disease presented with abdominal distensionand intestinal obstruction. For the laboratory results, all hadelevated C-reactive protein level (range 48.3-173 mg/L).For the cerebrospinal fluid (CSF) study, 72.7% had

E. coli Neonatal Meningitis129

polymorph predominant pleocytosis, CSF glucose level wasranged from 0.6 to 6.3 mmol/L and 54.5% had low CSFglucose level which was defined as </=1.1 mmol in pretermand </=2.5 mmol/L in term infant.6-8 CSF protein level wasranged from 1.02 to 6.47 g/L and seven of them had abnormalresult which was defined as >1.50 g/L in preterm and >1 g/Lin term infants.6-8 (Table 2) Among those six with abnormalCSF glucose level, they also had elevated CSF protein(Figure 1). Five neonates had Gram-negative bacilli detectedon the Gram stain of the CSF and all of them had low CSFglucose level and elevated CSF protein level. Among thesefive babies, 80% had positive result on E. coli K1 latexagglutination test. None had positive agglutination resultwhile the CSF Gram stain was negative. Three babies hadE. coli isolated in the urine specimens with samesusceptibility pattern as the CSF culture isolate. Ninety-onepercent had coexisting E. coli bacteraemia.

Antibiotic SusceptibilityAntibiotic susceptibility patterns of the isolates were

determined according to the Clinical & Laboratory StandardsInstitute guideline. Ninety-one percent of the isolates wereresistant to ampicillin. One isolate (9.1%) was resistant togentamicin. Three isolates (27.3%) were ESBL producingand resistant to third generation cephalosporins. Otherwise,all tested isolates were susceptible to amoxicillin-clavulanate; meropenem, and amikacin (Table3).

ESBL E. coli Isolate and Intrapartum AntibioticProphylaxis (IAP)

The demographic data, laboratory results and antibioticsuse among neonates with and without ESBL - E. coli infectionwere analysed. It was identified that all cases infected withneonatal ESBL E. coli were presented with early onsetmeningitis and were given intrapartum ampicillin. There was

Figure 1 Cerebrospinal fluid results among neonates with E. coli meningitis. Normal CSF protein: <1 g/L in preterm and <1 g/L interm infants; Normal CSF glucose concentration: </=1.1 mmol/L in preterm and </= 2.5 mmol/L in term infants.

Chik et al 130

no ESBL E. coli infection among the group withoutintrapartum ampicillin given. After analysis with Fisher'sexact test for categorical variables and Mann-Whitney Utest for continuous variables, there was no statisticalsignificance identified between intrapartum ampicillin useand neonatal ESBL E. coli infection (p value: 0.061). Therewas also no significance association between ESBL E. coliinfection and other demographic data and laboratory results(Table 4).

Neurological OutcomeThree babies had developmental delay, one of them had

hydrocephalus and seven had normal development onfollow-up. One baby defaulted follow-up (Table 2). Therewas no statistical association between neonatal ESBLE. coli infection and delay in neurological development(Table 4).

Discussion

Hong Kong does not have a registry on the incidence ofneonatal E. coli infection. To our knowledge; this is thefirst local study of neonatal E. coli meningitis in NICUs.The study hospitals accounted for more than 25% of totallive births in Hong Kong public hospitals; therefore theresult is representative to Hong Kong situation. Theestimated prevalence of neonatal E. coli meningitis was0.02%, with the significant mortality and morbidity, it is aproblem worthy of further investigations. There is no doubtthat use of IAP, i.e. ampicillin in pregnant women with GBScolonisation can reduce early-onset neonatal GBS infection.The reported odd ratio was 0.17.1,2 However, some overseasstudies reported that the incidence of neonatal E. coli sepsisand meningitis had increased after this practice, togetherwith the emergence of ampicillin-resistant E. coli infectionin neonates.3-5 A Taiwanese study also described an increasein the percentage of early onset neonatal E. coli sepsisfrom 40.9% to 70% since the implementation of theguideline on prevention of perinatal group B streptococcaldisease by Center for Disease Control and Prevention.4 HongKong has started the GBS screening in 2012 but we lack ofdata on this.

Concerning the microbiological aspect and thepathogenesis, E. coli is a normal flora of the humangastrointestinal tract. It has diverse and complex antigenicstructures. Some strains carry a specific virulence factorK1 capsular antigen, a polysialic acid, which impairs host'sopsonophagocytic killing. The amount and persistence of

K1 antigen in the CSF is closely linked to neonatalmeningitis and the disease severity. Infants infected withK1 antigenic strains have increased morbidity and mortalitycompared to those infected with other strains.Unfortunately, up to 50% of E. coli colonised in maternalvagina can express K1 antigen which poses risk for verticaltransmission of this virulence strain.9 In the past, researcherssuggested the E. coli strains with K1 capsular antigen, i.e.virulent strain, were able to cause invasive infection butthey were more susceptible to antimicrobial drugs, as theycould not possess resistant genes. While the moreantimicrobials resistant organisms, the more difficult forthem to reconcile bacterial virulence and thus could onlycause noninvasive infection.9 Later on, this postulationwas being challenged by the case reports of ESBL E. colineonatal meningitis.10 Our findings also opposed the

Table 1a Demographic data of neonates with early onset andlate onset sepsis

Total cases Early onset Late onset

11 6 5

Mean age of symptoms Day 10.7 Day 3.17 Day 19.4onset in days (range) (0-28) ( 0-6) ( 11-28)

SexM:F 1.2:1 2:1 1:1

Mean birth weight 2.31 kg 2.29 kg 2.86 kg(range) (1.36-3.61) (1.13-3.61) (1.49-3.39)

Mean gestation 34 weeks 32.8 weeks 37.2 week(range) (29-40) (29-37) (30-40)

Full term 5 1 4

Preterm 6 5 1

Table 1b Demographic data of preterm and term neonates

Preterm Full term

Early Late Early Late

onset onset onset onset

Number of cases 5 1 1 4

Mean gestation 32 30 37 39(range) (29-36) (38-40)

SexM:F 3:2 1:0 1:0 1:1

Mean birth weight 2.0 1.49 3.61 3.2(range) (1.36-3.55) (2.82-3.35)

Mean age of symptoms Day 3.2 Day 11 Day 3 Day 24onset in days (range) (0-6) (17-29)

E. coli Neonatal Meningitis131T

able

2D

emog

raph

ic d

ata,

inve

stig

atio

n re

sults

, tre

atm

ent a

nd o

utco

me

Ges

tatio

nSe

xIP

AB

Mat

erna

lPl

acen

taB

irth

•••• •Sy

mpt

oms o

nset

Max

Blo

od•••• •

CSF

WB

C/

•••• •G

ram

Susc

eptib

ility

ESB

LIn

itial

Subs

eque

ntO

ther

Neu

rolo

gica

l

risk

hist

olog

yw

eigh

t•••• •

EO

vs L

O/

CR

Pcu

lture

/po

lym

orph

(%)

stai

npa

ttern

antib

iotic

antib

iotic

inve

stig

atio

nsou

tcom

e

fact

ors

(kg)

•••• •Pr

esen

ting

(mg/

L)

CSF

•••• •R

BC

•••• •L

atex

resu

lts

for

peri

part

um sy

mpt

oms

cultu

re•••• •

Prot

ein

(g/L

)ag

glut

inat

ion

infe

ctio

n•••• •

Glu

cose

(m

mol

/L)

129

wee

kF

Amp

PPRO

MAc

ute

1.36

•D6

79E.

col

i/•

140/

99%

•GN

BAm

p R;

Posit

ive

Aug

Mem

Ear s

wab

& G

/L:

Delay

chor

io-

•EO

E. c

oli

•78

0•

Posit

ive

Aug

S;&

Gen

ESBL

E. c

oli;

amni

oniti

s•

Feve

r•

6.47

CTX

R;Ur

ine:

no g

rowt

h;•

<0.7

Gen

S;Ur

ine W

BC 1

0/m

m3

Mem

S;

Levo

S

230

MAm

pPP

ROM

Acut

e1.

45•

DO

173

E. c

oli/

•16

40/9

0%•

GNB

Amp

R;Ne

gativ

eAu

gCT

XG

/L: E

. col

i &De

laywe

ek 6

d+

Gen

mate

rnal

chor

io-

•EO

E. c

oli

•79

0•

Posit

ive

Aug

S;&

Gen

K. p

num

onia

e

feve

ram

nion

itis

•Fe

ver

•4.

32CT

X S;

•1.1

Gen

S;Le

vo S

338

wee

kM

Nil

Nil

ND

3.39

•D1

965

E. c

oli/

•>1

0000

/97%

•GN

BAm

p R;

Nega

tive

Aug

CTX

Urin

e:No

rmal

•LO

E. c

oli

•64

0•

Posit

ive

Aug

S;&

CTX

E. c

oli,

S. b

ovis

•Fe

ver

•2.

11CT

X: S

;Ur

ine W

BC•

2.3Ge

n S;

10-1

00/m

m3

Levo

R

438

wee

kF

Nil

Nil

ND

2.82

•D2

857

E. c

oli/

•42

80/9

5%•

NA

SAm

p R;

Nega

tive

Aug

CTX

Urin

e:No

rmal

•LO

E. c

oli

•39

6•

Nega

tive

Aug

S;&

CTX

no g

rowt

h•

Feve

r•

4.2CT

X S;

Urin

e WBC

•0.6

Gen

S;<1

0/m

m3

Levo

S

536

wee

kM

Amp

PPRO

MAc

ute

2.66

•D5

51E.

col

i/•

2930

/75%

•GN

BAm

p R;

Nega

tive

Amp

CTX

Urin

e: E.

col

iN

orm

al

chor

io-

•EO

E. c

oli

•10

5•

Posit

ive

Aug

S;&

Gen

Urin

e WBC

amni

oniti

s•

Poor

•13

.4CT

X S;

<10/

mm

3

feed

ing

•1.1

Gen

S;Le

vo S

;M

em S

Am

p: a

mpi

cilli

n, A

ug: a

ugm

entin

, CR

P: C

reac

tive

prot

ein,

CSF

: cer

ebro

spin

al fl

uid,

CSU

: cat

hete

r sav

ed u

rine,

CTX

: cef

otax

ime;

EO

: ear

ly o

nset

, ESB

L: e

xten

ded

spec

trum

bet

a la

ctam

ase,

F: f

emal

e, F

EP:

cefe

pim

e, G

BS:

gro

up B

stre

ptoc

occu

s car

rier,

Gen

: gen

tam

icin

, G/L

: Gas

tric

Lava

ge, I

PAB

: int

rapa

rtum

antib

iotic

, Lev

o: le

voflo

xaci

n ; L

O: l

ate

onse

t, G

NB

: Gra

m n

egat

ive

baci

lli, M

: mal

e, M

em: m

erop

enem

,M

et: m

etro

nida

zole

, NA

S: n

o ab

norm

ality

see

n, N

D: n

ot d

one,

Pen

: pen

icill

in, P

PRO

M: p

rete

rm p

rela

bour

rupt

ure

of m

embr

ane,

R: r

esis

tant

, S: s

ensi

tive;

WB

C: w

hite

blo

od c

ell (c

ontin

ued

on p

age

132)

Chik et al 132T

able

2D

emog

raph

ic d

ata,

inve

stig

atio

n re

sults

, tre

atm

ent a

nd o

utco

me

(con

t'd)

Ges

tatio

nSe

xIP

AB

Mat

erna

lPl

acen

taB

irth

•••• •Sy

mpt

oms o

nset

Max

Blo

od•••• •

CSF

WB

C/

•••• •G

ram

Susc

eptib

ility

ESB

LIn

itial

Subs

eque

ntO

ther

Neu

rolo

gica

l

risk

hist

olog

yw

eigh

t•••• •

EO

vs L

O/

CR

Pcu

lture

/po

lym

orph

(%)

stai

npa

ttern

antib

iotic

antib

iotic

inve

stig

atio

nsou

tcom

e

fact

ors

(kg)

•••• •Pr

esen

ting

(mg/

L)

CSF

•••• •R

BC

•••• •L

atex

resu

lts

for

peri

part

um sy

mpt

oms

cultu

re•••• •

Prot

ein

(g/L

)ag

glut

inat

ion

infe

ctio

n•••• •

Glu

cose

(mm

ol/L

)

629

wee

kM

Amp

PPRO

MAc

ute

1.13

•D

O11

5E.

col

i/•

Bloo

d sta

ined

•N

DAm

p R;

Posit

ive

Aug

Mem

Eye s

wab

&No

rmal

mate

rnal

chor

io-

•EO

E. c

oli

(WBC

and

•Ne

gativ

eAu

g S;

& G

enG/

L &

ear

feve

ram

nion

itis

•Ap

noea

RBC

coun

t not

CTX

R;sw

ab: E

SBL

perfo

rmed

)Ge

n S;

E. c

oli

•1.

43Le

vo R

;•

3M

em S

735

wee

kF

Nil

Nil

ND3.

55•

D514

8E.

col

i/•

0/0%

•N

DAm

p R;

Nega

tive

Aug

Mem

Urin

e: no

Defa

ulted

•EO

E. c

oli

•17

1•

ND

Aug

S;&

Gen

grow

thfo

llow

up•

Feve

r•

1.02

CTX

S;Ur

ine W

BC•

2.8Ge

n S

nil

8Fu

ll ter

mF

Nil

Nil

No3.

26•

D17

56.1

E. c

oli/

•57

60/5

7%•

GNB

Amp

R;Ne

gativ

eAm

pCT

X+Ge

nUr

ine:

E c

oli

Norm

al•

LOE.

col

i•

640

•Ne

gativ

eAu

g S;

& C

TX&

ent

eroc

occu

s

•Fe

ver

•2.

65CT

X S;

•1

Gen

S

9Fu

ll ter

mM

Amp

Mate

rnal

ND

3.61

•D3

56E.

col

i/•

0/0%

•N

DAm

p R;

Posit

ive

Amp

Mem

+Gen

Perit

onea

lDe

layfe

ver

•EO

E. c

oli

•42

5•

ND

Aug

S;&

CTX

swab

:•

Feve

r•

0.86

CTX

R;&

Met

ESBL

E c

oli

•6.3

Gen

S;

(Hirs

chru

ng's

dise

ase)

Levo

R;

Mem

S

1030

wee

kM

Nil

Nil

No1.

49•

D11

154

E. c

oli/

•2/

ND•

NA

SAm

p R;

Nega

tive

Pen

CTX

Urin

e &No

rmal

•LO

E. c

oli

•53

5•

Nega

tive

Aug

S;&

Gen

cent

ral l

ine

•Ap

noea

•1.

24CT

X S;

tip: n

o•

4.2Ge

n R

grow

th

11Fu

ll ter

mF

Nil

Nil

No3.

35•

D22

48.3

No g

rowt

h/•

320/

94%

•N

AS

Amp

S;Ne

gativ

eAm

pAm

pUr

ine:

Nor

mal

•LO

E. c

oli

•24

80•

Nega

tive

Aug

S;&

CTX

no g

rowt

h•

Feve

r•

1.37

CTX

S;•

2.9Ge

n S;

Mem

SA

mp:

am

pici

llin,

Aug

: aug

men

tin, C

RP:

C re

activ

e pr

otei

n, C

SF: c

ereb

rosp

inal

flui

d, C

SU: c

athe

ter s

aved

urin

e, C

TX: c

efot

axim

e; E

O: e

arly

ons

et, E

SBL:

ext

ende

d sp

ectru

m b

eta

lact

amas

e, F

: fem

ale,

FEP

:ce

fepi

me,

GB

S: g

roup

B st

rept

ococ

cus c

arrie

r, G

en: g

enta

mic

in, G

/L: G

astri

c La

vage

, IPA

B: i

ntra

partu

m an

tibio

tic, L

evo:

levo

floxa

cin

; LO

: lat

e on

set,

GN

B: G

ram

neg

ativ

e ba

cilli

, M: m

ale,

Mem

: mer

open

em,

Met

: met

roni

dazo

le, N

AS:

no

abno

rmal

ity s

een,

ND

: not

don

e, P

en: p

enic

illin

, PPR

OM

: pre

term

pre

labo

ur ru

ptur

e of

mem

bran

e, R

: res

ista

nt, S

: sen

sitiv

e; W

BC

: whi

te b

lood

cel

l

E. coli Neonatal Meningitis133

Table 4 Statistical analysis of demographic data, laboratoryresults, intrapartum antibiotic (IAP) use, neurological outcomein associate with neonatal ESBL E. coli infection

ESBL negative ESBL positive P value*

(n=8) (n=3)

Male 4 2 1.00

Ampicillin use (IAP) 2 3 0.061

Maternal HVS/blood 0 0 NDgrew ESBL-E. coli

Placental swab grew 0 2 NDESBL-E. coli

Acute chorioamnionitis 2 2 0.491

Gestation (week) 36±3.8 32.7±6.4 0.376

Birth weight (kg) 2.7±0 .8 2±1.4 0.497

Symptoms onset 13.4±9.7 3.0±3.0 0.133(days of life)

Maximum CRP 94.1±53.9 83.3±29.7 1.00(mg/L)

Delay in neurological 1 2 0.183development

*Fisher exact test for categorical variables, Mann-Witney U test forcontinuous variables

Table 3 Antibiotic susceptibility

Ampicillin Amoxcillin- Cefotaxime Meropenem Gentamicin Amikacin ESBL+

Clavulanic acid

E. coli (susceptible / total isolate) 1/11 11/11 8/11 11/11 10/11 11/11 3/11

EO (susceptible / total isolate) 0/6 6/6 3/6 6/6 6/6 6/6 3/6

LO (susceptible / total isolate) 1/5 5/5 5/5 5/5 4/5 5/5 0/5

ESBL: Extended spectrum beta lactamase; EO: Early onset meningitis; LO: Late onset meningitis

postulation, as among four E. coli strain with positive K1antigen detected in CSF, one of them was ESBL producing.

The global burden of neonatal ESBL producingEnterobacteriaceae infection is high. An overseas studyreported 17% of organisms causing neonatal sepsis wereESBL producing Enterobacteriaceae among all culturepositive cases.11 While our study showed 27.3% of E. colimeningitis were ESBL producing. This was much higherthan the overseas study which probably due to small samplesize and different in the denominator. The known risk factorsof neonatal ESBL producing pathogens infection wereprematurity, and extremely low birth weight.4 There is acontroversy about the association between the use ofintrapartum ampicillin and neonatal ESBL- Gram negative

organism infection. Cordero et al12 reported there was noincrease in neonatal sepsis caused by drug-resistant Gram-negative pathogen after national maternal GBS screeningand prophylaxis, while other study suggested IAP might playa role in neonatal ESBL Gram negative organisms' infection.5Though our study did not show a statistical significantrelationship between IAP and neonatal ESBL E. colimeningitis, further study and monitoring deemed necessaryon this issue. Meanwhile, conclusion is difficult to draw.

Besides the above identifiable risk factors for neonatalinfection by ESBL producing organism, the global increasein the prevalence of ESBL producers infection amonggeneral population also played a part. In United States, 36%of all community-acquired E. coli infections were ESBLproducing,13 while our local data in 2015 was 26%.14 Thehigh rate of ESBL E. coli infection in our community hasbeen postulated to the overuse of antibiotics in livestockfarming. One study identified 53.6% of the local live pigshad faecal carriage of ESBL-producing E. coli.15 Thisemergence of ESBL-producing organisms in the communityposes a major challenge to the choice of empirical antibioticin newborns. According to the international guideline,intrapartum intravenous benzylpenicillin to mother orcombination of ampicillin and gentamicin to the neonatesshould be used to prevent early-onset neonatal infection inthose mothers had risk factors of neonatal sepsis. If there ismicrobiological evidence of neonatal Gram-negativebacterial sepsis or meningitis; the guideline suggestsbenzylpenicillin/ampicillin, gentamicin and cefotaxime untilfurther results are available. This is a widely adapted regimewith good coverage for GBS and E. coli, and evenL. monocytogenes infection.16-18 However, this empiricalantibiotic regime cannot treat meningitis caused by ESBLproducing Enterobacteriaceae as it is intrinsically resistantto third generation cephalosporin and benzylpenicillin/ampicillin. Even adding gentamicin is inadequate as E. colican also be resistant to gentamicin as shown in one of ourpatient. Furthermore aminoglycoside (gentamicin andamikacin) is not a good agent to penetrate the blood brain

Chik et al 134

barrier. Nevertheless, empirical use of broad-spectrumantibiotics such as carbapenem is not a way of solving thisproblem. Broad-spectrum antibiotic would alter the neonatalintestinal flora colonisation patterns and thus increase therisk of necrotising enterocolitis and candidaemia. Thedilemma thus comes up. However, at the time of writing,there are no well-powered randomised trials on empiricalantibiotic therapy for neonatal sepsis. Further studies aredeemed necessary to determine the most appropriate choiceof empirical intrapartum antibiotic and empirical antibioticfor neonatal sepsis in the era of increasing drug resistanceorganisms especially when Gram negative organismmeningitis is suspected.

The limitation of our study is the small sample sizedespite of taking five-year data in 2 large neonatalintensive care units. We suggest continuous monitoringof the incidence of Gram negative bacterial infection inneonates and a cross centre study on the issue ofintrapartum antibiotic use and empirical antibiotic use inneonatal sepsis is urgently needed.

Conclusion

The risk of E. coli sepsis among neonates continuesdespite of improvement in perinatal care and empirical useof intrapartum antibiotic. Though this study did not identifyany statistical significance between intrapartum ampicillinuse and ESBL infection, with the widespread drugresistance, choice of intrapartum antibiotic and empiricaltreatment of neonatal sepsis has an urgent need to bereviewed.

Conflict of Interest

The authors have no conflicts of interest to disclose.

References

1. Center for Disease Control and Prevention(CDC). Prevention ofperinatal group B streptococcal disease: a public healthperspective. MMWR 1996;45(RR-7):1-24.

2. Ohlsson A, Shah VS. Intrapartum antibiotics for known maternalgroup B streptococcal colonization. Cochrane Database Syst Rev

2014;6:CD007467.3. Stoll BJ, Hansen NI, Sanchez PJ, et al. Early onset neonatal sepsis:

the burden of group B streptococcal and E. coli disease continues.Pediatrics 127:817-26.

4. Lin CY, Hsu CH, Huang FY, et al. The changing face of early-onset neonatal sepsis after the implementation of a maternalgroup B Streptococcus screening and intrapartum prophylaxispolicy-a study in one medical center. Pediatr Neonatol 2011;52:78-84.

5. Bizzarro MJ, Dembry LM, Baltimore RS, Gallagher PG. Changingpatterns in neonatal Escherichia coli sepsis and ampicillinresistance in the era of intrapartum antibiotic prophylaxis.Pediatrics 2008;121:689-96.

6. Byington CL, Kendrick J, Sheng X. Normative cerebrospinal fluidprofiles in febrile infants. J Pediatr 2011;158:130-4.

7. Chadwick SL, Wilson JW, Levin JE, Martin JM. Cerebrospinalfluid characteristics of infants who present to the emergencydepartment with fever: establishing normal values by week ofage. Pediatr Infect Dis J 2011;30:e63-7.

8. Rodriguez AF, Kaplan SL, Mason EO Jr. Cerebrospinal fluidvalues in the very low birth weight infant. J Pediatr 1990;116:971-4.

9. Lavigne JP, Blanc-Potard AB, Bourg G et al Virulence genotypeand nematode-killing properties of extra-intestinal Escherichiacoli producing CTX-M Beta-lactamases. Clin Microbiol Infect2006;12:1199-206.

10. Dubois D, Prasadarao NV, Mittal R, Bret L, Roujou-Gris M,Bonnet R. CTX-M beta-lactamase production and virulence ofEscherichia coli K1. Emerg Infect Dis 2009;15:1988-90.

11. Vijayakanthi N, Bahl D, Kaur N, Maria A, Dubey NK. Frequencyand characteristics of infections caused by extended-spectrumbetalactamase-producing organisms in neonates: a prospectivecohort study. Biomed Res Int 2013;2013:756209.

12. Cordero L, Sananes M, Ayers LW. Bloodstream infections in aneonatal intensive-care unit: 12 years' experience with an antibioticcontrol program. Infect Control Hosp Epidemiol 1999;20:242-6.

13. Doi Y, Park YS, Rivera JI, et al. Community-associated extended-spectrum beta-lactamase-producing Escherichia coli infection inthe United States. Clin Infect Dis 56:641-8.

14. Hong Kong Princess Margaret Hospital Antibiotic susceptibilityprofile 2015. http://pmh.home (accessible via Hong Kong HospitalAuthority intranet).

15. Ho PL, Chow KH, Lai EL, et al. Extensive dissemination of CTX-M-producing Escherichia coli with multidrug resistance to'critically important' antibiotics among food animals in Hong Kong,2008-10. J Antimicrob Chemother 2011:66:765-8.

16. National Institute for Health and Care Excellence guidelineNeonatal infection (early onset): antibiotics for prevention andtreatment. Clinical guideline 22 August 2012. Available at:nice.org.uk/guidance/cg149.

17. Simonsen KA, Anderson-Berry AL, Delair SF, Davies HD. Early-onset neonatal sepsis. Clin Microbiol Rev 2014;27:21-47.

18. MacGowan A, Wootton M, Bowker K, Holt HA, Reeves D.Ampicillin-aminoglycoside interaction studies using Listeriamonocytogenes. J Antimicrob Chemother 1998;41:417-8.


Non-nutritive Sucking and Nesting Greatly Reduces PainDuring Retinopathy Screening in Premature Infants

WH LIAO, HY XU, X ZHOU, JL TIAN, XY FANG, J DING, C ZENG, HL WU

Abstract Purpose: To evaluate the efficacy of non-nutritive sucking combined with nesting in the reduction of painassociated with fundus oculi screening in premature infants. Methods: A randomised clinical trial involvingfundus oculi screening was carried out in 120 healthy premature infants. Infants assigned to the interventiongroup received non-nutritive sucking and nesting, while the control group simply received routine nursing.Pain was assessed using the Premature Infant Pain Profile before and during the eye examination.Simultaneously, we evaluated a range of physiological parameters and crying time. Results: There were nosignificance differences between the control group and the intervention group in terms of gestational age,corrected gestational age or mean birth weight at retinopathy of prematurity screening. However, theintervention group showed a significantly lower mean Premature Infant Pain Profile score after fundusoculi screening compared to the control group (P<0.05). Significant differences between the two groupswere also identified in crying time, heart rate, breathing rate, oxygen saturation, blood pressure and facialexpression. Conclusion: Non-nutritive sucking, combined with nesting, alleviated pain in premature infantsduring retinopathy of prematurity screening, and appeared to provide infants with a feeling of warmth andsafety. This technique is worth promoting for use during retinopathy of prematurity screening.

Key words Non-nutritive sucking; Nursing; Premature Infant Pain Profile; Retinopathy screening

Department of Neonatology, Nanfang Hospital, SouthernMedical University, Guangzhou, China

WH LIAO BSHY XU MDX Zhou BSJL TIAN BSXY FANG BSJ DING BSC ZENG BSHL WU BS

Correspondence to: Dr WH LIAO


Received January 25, 2018

Original Article

Introduction

Retinopathy of prematurity (ROP) is a developmentalvascular proliferative disease of the retina, which oftenaffects premature infants with low-birth-weight and can

lead to childhood blindness.1 In China, the incidence of ROPis reported to be 17.8% for premature newborns with a birthweight (BW) of 2,000 g or less and/or a gestational age(GA) of 34 weeks or less.2 ROP can be prevented with timelytreatment by the use of appropriate screening programs; assuch, early eye examinations for ROP are extremelyimportant for premature infants.3 However, the routineprocedure used for ROP screening may be painful forpreterm infants and has been linked to transient changes inheart rate, respiratory rate, blood pressure and oxygensaturation.4,5 Pain evaluations, using the Premature InfantPain Profile (PIPP), have also shown that most infants sufferpain during ROP screening.4,5 Therefore, it is critical forphysicians to develop new methods to assess pain andprovide safe and effective pain-relief interventions duringROP screening. In China, it is not routine practice to usenursing methods to provide comfort to premature infantswhen they undergo ROP screening protocols that involvemild to moderate pain.

Liao et al 136

Previous studies have reported that the use of non-nutritive sucking (NNS) and nesting can reduce pain inneonates and premature infants during painful procedures.6-8

However, the effect of NNS, either alone or in combinationwith nesting, as a procedural pain-relief intervention forROP screening remains undefined. In the present study,we investigated the efficacy of NNS, combined withnesting, for the alleviation of pain associated with ROPscreening in premature infants.

Methods

Study DesignThis is a methodological research study aimed at

developing a technique involving NNS, combined withnesting, to alleviate pain in premature infants during painfulprocedures.9

Setting and SampleWe performed a randomised controlled trial, involving

premature infants, between January 2015 and May 2016 inNanfang Hospital. We recruited 120 premature infants, witha GA of ≤32 weeks and divided randomly these cases intotwo groups with a computer based randomisation processaccording to their ROP screening sequence and by sealedopaque envelopes: an interventional group (n=60) and acontrol group (n=60). The interventional group receivedNNS and nesting care, while the control group simplyreceived routine nursing. Nests were made with a cottonbedsheet in a manner suitable for babies.10

Ethical ConsiderationAll protocols were reviewed and approved by the Ethics

Committee of Nanfang Hospital, an affiliate of SouthernMedical University, China. Detailed written informedconsent was obtained from the parents of all eligible infants.

Intervention and Eye ExaminationAll infants were fed 1 hour prior to eye examination. A

topical anaesthetic (Minims® Tetracaine Hydrochloride,0.5% w/v) was applied 30 second before the eye examinationin all infants. For each infant, the study protocol was appliedonly on the first screening examination and all infants wereexamined by the same ophthalmologist. All infantsunderwent ROP screening using RetCam3 retino-digitalphotography (Clarity Medical System Inc., Mohali, Punjab,India).

ProcedureTwo experienced nurses, who had undergone web-based

training using the PIPP, were tasked to evaluate the PIPPscore.11 Only examinations of the first eye were used torecord the PIPP. The final outcome measurement for thePIPP score was a mean value derived from the twoinvestigators. PIPP scores were recorded 5 minutes and 1minute before and after the eye examination. PIPP scores<7 were considered to be indicative of no pain, while PIPPscores of 7 to 12 were classified as intermediate and those>12 were considered to be indicative of significant pain.Data regarding oxygen saturation, heart rate, blood pressure,crying time and facial expression were collected immediatelyafter the eye examination. Secondary outcomemeasurements including tachycardia (>180 bpm),bradycardia (<100 bpm), desaturation (<85% for >10 s),and crying time. All of the babies were video-recordedduring the eye examination. Demographics and outcomemeasurements were recorded on previously-prepared formsfor each infant.

Data AnalysisData are expressed as the arithmetic mean ± standard

deviation (SD). The significance of differences betweengroups was calculated using Student's t test and P<0.05was considered to be statistically significant.

Results

Baseline CharacteristicsThe baseline characteristics of the premature infants

enrolled in this study are presented in Table 1. A total of120 infants were enrolled. Of these, 48.3% were male and51.7% were female; 30.8% were born by normalspontaneous delivery and 69.2% by Caesarean delivery.Mean GA was 31.52±1.8 weeks, corrected GA atexamination was 35.54±1.6 weeks, and mean body weightwas 2,003±178 g. There were 60 infants in the interventiongroup and an equal number in the control group. Thebackground variables related to GA, corrected GA andmean BW were not significantly different when comparedbetween the two groups.

PIPP ScoresPIPP outcome measurements are summarised in Table 2.

There were no significant differences in baseline PIPP scoreswhen compared between the two groups before eye

NNS and Nesting Reduces Pain During ROP137

examination (2.1±0.6 versus 2.2±0.5, P>0.05). However,infants receiving NNS and nesting had significantly lowermean pain scores during examination of the first eye(12.9±2.0 versus 16.5±2.0, P<0.001) (Figure 1). In addition,86.7% of infants in the control group presented with

Table 1 Baseline characteristics of the premature infantsincluded in this study

Variables Intervention Control P valuegroup group

(NNS + nesting)(n = 60) (n=3)

Sex (male, %) 29 (48.33%) 29 (48.33%) −Method of delivery

Normal birth 28 29 −Caesarean 32 31 −

Gestational age 31.3±1.9 31.7±1.6 >0.05(weeks), (mean ± SD)

Birth weight (g), 2026±174 1980±181 >0.05(mean ± SD)

Corrected age at 35.2±1.7 35.7±1.5 >0.05examination (weeks),(mean ± SD)

Figure 1 Box-plot graph showing Premature Infant Pain Scale(PIPP) scores of the intervention and control groups.

Table 2 Premature Infant Pain Scale (PIPP) parameters eachgroup

Relief measures

Variables Intervention Control P value

group group

(NNS + nesting)

Crying intensityNo crying 3 0 <0.001Whimper 28 12 <0.001Vigorous crying 27 48 <0.001

Facial expressionRelaxed muscles 20 2 <0.001Grimace 40 58 <0.001

Breathing patternsRelaxed muscles 32 6 <0.001Alteration in breathing 28 54 <0.001

Pain degreeNo pain (PIPP <7) 2 0 <0.001Intermediate 32 8 <0.001(7 <PIPP <12)Significant pain 26 52 <0.001(PIPP >12)

significant pain compared to only 43.3% of infants in theintervention group. Furthermore, only 53.3% of theintervention group experienced intermediate pain.

Physiological ParametersPhysiological parameters, including heart rate, breathing

rate, oxygen saturation, blood pressure, and crying timeare shown in Table 3. There were no episodes of vomitingor choking observed in any of the study infants. Overall,premature infants who received non-nutritive sucking andnesting showed better outcomes than those receiving routinenursing. The effects upon increased heart rate and bloodpressure after the eye examination were more remarkablein the control group than in the NNS + nesting group(P<0.01). Sustained crying was observed in only 45% ofthe NNS group, but in 80% of the control group.Furthermore, 33% of infants in the NNS + nesting groupexhibited relaxed facial muscles, compared to only 3% ofthose from the control group. In addition, there was asignificant difference in the breathing rate when comparingbetween the NNS + nesting group and the control group(P<0.01).

Discussion

ROP screening procedure can be painful to preterminfants. Furthermore, repeated pain may lead to serious side-effects in the development of neurological and behaviouralmechanisms in infants.12 Consequently, there is a need to

Liao et al 138

develop effective and safe measures to reduce proceduralpain during eye examinations for ROP in premature infants.At present, non-pharmacological interventions to alleviatepain in preterm infants include oral sucrose or glucose, breastfeeding, NNS, music therapy, and also techniques involvingthe mother such as 'wrapping' or 'kangaroo-mother' care.These measures can alleviate pain in premature infants todiffering extents. For example, Dilli et al reported that oralsucrose and NNS went some way to reduce pain during eyeexaminations for retinopathy in premature infants.4 Liaw etal also reported that infants receiving NNS, and facilitatedsucking, had significantly lower mean pain scores duringheel-stick procedures.7 However, O'Sullivan et al showedthat although sucrose, combined with NNS and swaddling,reduced the level of pain, pain scores remained consistently

high.13 On the other hand, pharmacological interventionshave also been reported to alleviate pain in infants.14

However, the side effects of medication upon infantsshould be considered carefully. In addition, Kabata etal showed that oral paracetamol only reduced pain scoresby a modest amount during eye examinations; moreover,there were no significant differences regarding cryingtime or the number of infants experiencing tachycardia/bradycardia and desaturation following the use of oralparacetamol during ROP screening.14

In our study, for the first time, we applied NNS, combinedwith nesting, to relieve pain during ROP screening inpremature infants. Our results showed that this procedurewas effective in reducing pain during eye examinations.Crying time was shorter in infants receiving NNS combinedwith nesting. Lower PIPP scores were also observed in theintervention group. Nesting is based upon creating a warmbox and making a natural environment similar to a nest fornewborns. This natural environment is similar to an artificialuterus, and provides the infant with an environment that isphysiologically similar to the uterus in terms of head flexion,proximity of the jaw to the chest wall, proximity of the upperarms to the chest, elbow flexion and positioning of theforearm towards or parallel to the chest, extreme hip andknee joint flexion, proximity of the thigh to the abdominalwall, and the positioning of the legs. Nesting care fornewborns can exert a significant clinical effect, and caneffectively maintain a stable environment for prematureinfants. The adaptation of premature infants to suchenvironments creates a useful technique for alleviating painand should be introduced into routine clinical practice. NNSand nesting are not pharmacological interventions, and aretherefore simple to use and are not associated with anyadverse effects.

There are some limitations associated with our study thatshould be considered when interpreting our results. Forexample, our sample size was relatively small. Thisprevented us from creating independent groups for NNSand nesting and thus prevented us from comparing therelative effects of these two techniques when used alonerather than in combination.

Conclusion

On the basis of the results above, NNS combined withnesting was shown to reduce PIPP pain scores during eyeexaminations for ROP, and appeared to provide premature

Table 3 Outcome measurements for the study infants

Relief measures, mean (SD)

Variables Intervention Control P value

group Group

(NNS + nesting)

PIPP score before 2.1±0.6 2.2±0.5 >0.05

examination, (mean ± SD)

Heart rate

Before examination 142.5±12 140.8±13 >0.05

After examination 160.8±14 165.2±15 >0.05

Breathing rate

Before examination 44.6±2.5 44.8±2.6 >0.05

After examination 52.2± 3.6 58.6±2.8 <0.001

Oxygen saturation

Before examination 97.1±2.2 95.1±2.6 <0.001

After examination 93.5±2.8 89.2±2.5 <0.001

Blood pressure (mmHg)

Before examination 64.9±6.1/ 65.4±6.3/ <0.001/41.9±5.8 42.1±5.7 >0.05

After examination 73.6±4.4/ 78.6±5.2/ <0.001/45.89±6.0 47.4±6.6 >0.05

Crying time during 48.6±16.6 81.2±25.8 <0.001examination (sec), (mean ± SD)

Duration of pain during 32.5±3.1 54.3±2.8 <0.001examination (sec), (mean ± SD)

PIPP score after 12.9±2.0 16.5±2.0 <0.001examination, (mean ± SD)

PIPP = Premature Infant Pain Profile.

NNS and Nesting Reduces Pain During ROP139

infants with a feeling of warmth and safety. This combinedmethod is simple, safe, and is worth promoting for routineuse during ROP screening in premature infants.

Declaration of Interest

The authors declared no potential conflicts of interest.

Acknowledgments

This work was supported mainly by a grant from theMedical Scientific Research Foundation of GuangdongProvince (A2015098).

References

1. Fierson WM; American Academy of Pediatrics Section onOphthalmology; American Academy of Ophthalmology; AmericanAssociation for Pediatric Ophthalmology and Strabismus;American Association of Certified Orthoptists. ScreeningExamination of Premature Infants for Retinopathy of Prematurity.Pediatrics 2018;142:e20183061.

2. Xu Y, Zhou X, Zhang Q, et al. Screening for retinopathy ofprematurity in China: a neonatal units-based prospective study.Invest Ophthalmol Vis Sci 2013;54:8229-36.

3. Cavallaro G, Filippi L, Bagnoli P, et al. The pathophysiology ofretinopathy of prematurity: an update of previous and recentknowledge. Acta Ophthalmol 2014;92:2-20.

4. Dilli D, larslan NE, Kabata EU, Zenciro lu A, im ek Y, Okumu

N. Oral sucrose and non-nutritive sucking goes some way toreducing pain during retinopathy of prematurity eye examinations.Acta Paediatr 2014;103:e76-9.

5. Kandasamy Y, Smith R, Wright IM, Hartley L.Pain relief forpremature infants during ophthalmology assessment. J AAPOS2011;15:276-80.

6. Psaila K, Foster JP, Richards R, Jeffery HE. Non-nutritive suckingfor gastro-oesophageal reflux disease in preterm and low birthweight infants. Cochrane Database Syst Rev 2014;10:CD009817.

7. Liaw JJ, Yang L, Katherine Wang KW, Chen CM, Chang YC, YinT. Non-nutritive sucking and facilitated tucking relieve preterminfant pain during heel-stick procedures: a prospective,randomised controlled crossover trial. Int J Nurs Stud 2012;49:300-9.

8. Kandasamy Y, Smith R, Wright MR, Hartley L. Pain relief forpremature infants during ophthalmology assessment. J AAPOS2011;15:276-80.

9. Zahed M, Berbis J, Brevaut-Malaty V, Busuttil M, Tosello B,Gire C. Posture and movement in very preterm infants at termage in and outside the nest. Childs Nerv Syst 2015;31:2333-40.

10. Ferrari F, Bertoncelli N, Gallo C, et al. Posture and movement inhealthy preterm infants in supine position in and outside the nest.Arch Dis Child Fetal Neonatal Ed 2007;92:F386-90.

11. Stevens B, Johnston C, Petryshen P, Taddio A. Premature infantpain profile: development and initial validation. Clin J Pain 1996;12:13-22.

12. Grunau R. Early pain in preterm infants. A model of long-termeffects. Clin Perinatol 2002;29:373-94.

13. O'Sullivan A, O'Connor M, Brosnahan D, McCreery K, DempseyEM. Sweeten soother and swaddle for retinopathy of prematurityscreening: a randomised placebo controlled trial. Arch Dis ChildFetal Neonatal Ed 2010;95:F419-22.

14. Kabata EU, Dursun A, Beken S, Dilli D, Zenciro lu A,Okumu N. Efficacy of Single Dose Oral Paracetamol inReducing Pain During Examination for Retinopathy ofPrematurity: A Blinded Randomized Controlled Trial. IndianJ Pediatr 2016;83:22-6.


Clinical Manifestations and Outcomes ofCardiac Tumours in Children

G KIM, N LEE, H KO, JH BYUN, HD LEE, SC SUNG, H KIM, KH CHOI

Abstract Cardiac tumours occur rarely in children. We report on the clinical features, echocardiogram andelectrocardiographic findings, treatment, and outcomes of primary cardiac tumours in 27 children treatedbetween 2009 and 2017 in our single centre. The most common tumour was rhabdomyoma with tuberoussclerosis in 15 children; two malignant tumours-an angiosarcoma and rhabdomyosarcoma-were also found.Seven patients were haemodynamically unstable: five underwent cardiac surgery, one was managed witheverolimus for arrhythmia, and one died of heart failure 1 day after birth despite full medical support. Wediscuss the various treatment options for haemodynamically unstable paediatric patients with a cardiactumour.

Key words Cardiac surgery; Cardiac tumour; Rhabdomyoma

Heart Center, Pusan National University Children's Hospital,Beomeu-li, Mulgum-eup, Yangsan, Keungsangnam-do 626-770, Republic of Korea

G KIM MD, PhDN LEE MDH Ko MDJH BYUN MDHD LEE MD, PhDSC SUNG MD, PhD

H KIM MDKH CHOI MD, PhD

Correspondence to: Dr HD LEE


Received February 25, 2018

Original Article

Introduction

Cardiac tumours are rare in children.1,2 The prevalenceof primary cardiac tumours in autopsy studies of thepaediatric population is 0.0017-0.28%, and the incidenceis 0.14% during fetal life.3 The most frequently encounteredprimary tumours are rhabdomyomas, which are associatedwith tuberous sclerosis, followed by fibromas, whichaccount for 80% of primary cardiac tumours.3,4 In children,cardiac tumours present at different ages and as different

clinical entities according to the histological type, size, andlocation. Most cardiac tumours have benign features butmay also present with heart failure, haemodynamicallyunstable arrhythmia, and sudden death. We have treatedpaediatric patients with various cardiac tumours, whichranged from benign to malignant and which needed cardiacsurgery or medication. Here we present the results of ourevaluation of the clinical presentation, treatment, andoutcomes in paediatric patients with a cardiac tumourtreated in our centre.

Methods

We performed a retrospective observational study ofpatients diagnosed with cardiac tumours at Pusan NationalUniversity Children's Hospital, Pusan, Korea, from January2009 to August 2017. We identified 33 patients who hadbeen diagnosed with a primary cardiac tumour during this9-year period and reviewed their medical records. Six ofthe 33 patients were excluded because of loss to follow-upafter the first diagnosis. We analysed the clinicalpresentation, treatment, outcome, and electrocardiographicfindings for the remaining 27 patients, and we focused onseven haemodynamically unstable patients. We could notanalyse the data according to tumour histological type

Cardiac Tumours in Children141

and two rhabdomyomas (Table 3). One patient with arhabdomyoma developed haemodynamically unstablearrhythmia and was treated with antiarrhythmic medicationin the acute phase followed by everolimus. One patient withmultiple huge tumours died of severe left ventriculardysfunction on postnatal day 1. Excluding the five patientswho received an operation, the outcomes of the tumoursincluded 18 with partial regression, two with completeregression, and two with no change. During the follow-upperiod, for those with partial regression, this occurred at amedian of 25.4 months and for the two with completeregression, this occured at a median of 16.1 months; in twopatients with no change, this was assessed after a median of24.2 months.

The echocardiographic features are presented in Table2. Seventeen patients showed multiple cardiac tumours, and

because this was confirmed in only five patients who hadundergone surgical excision. Of these five patients, onehad a huge fibroma, two had a malignant sarcoma, and twohad confirmed rhabdomyomas. Thirteen other patients hadbenign rhabdomyomas associated with tuberous sclerosis,and nine had a benign course without tuberous sclerosis,except for one who died on postnatal day 1. The indicationfor the surgical removal of a cardiac tumour was anobstructive or haemodynamically unstable mass.

The study was approved by the Pusan NationalUniversity Yangsan Hospital Institutional Review Board.All data were analysed using SPSS for Windows, version21 (IBM Corp. Released 2012. Armonk, NY: IBM Corp.).Continuous variables are expressed as the median (range).

Results

The baseline characteristics of the study group are shownin Table 1. Twenty-seven patients were evaluated: 17 weregirls, the median age at diagnosis was 0.1 month, and themedian body weight was 3.6 kg. Most patients (74.1%) werediagnosed by fetal echocardiography during the antenatalperiod. Other clinical features at the time of diagnosis werecardiac murmur in one patient, chest pain in one patient,and seizure in two patients. There were 15 rhabdomyomas(55.5%), one f ibroma, one angiosarcoma, onerhabdomyosarcoma, and eight unknown tumours, whichwere suspected of being benign tumours such asrhabdomyomas not associated with tuberous sclerosis. Onepatient with an unknown type of tumour died on postnatalday 1. All 15 rhabdomyomas were associated with tuberoussclerosis: two were confirmed by histological examination,and 13 were diagnosed clinically as being associated withtuberous sclerosis. Nine of the 15 patients with arhabdomyoma were confirmed to harbour a tuberoussclerosis complex gene mutation: eight had a mutation inTSC2 and one a mutation in TSC1. Three patients had asmall atrial septal defect or small patent ductal arteriosus.

The median follow-up period was 23.0 months. Onepatient had a family history of cardiac tumours associatedwith tuberous sclerosis. Seven patients presented in ahaemodynamically unstable state. Five patients neededurgent surgery because of a large tumour; two patients hada tumour that caused obstruction of the left ventricularoutflow tract, six patients had severe heart failure, and onepatient had embolic phenomena.

Histological findings for the tumours were as follows;one angiosarcoma, one fibroma, one rhabdomyosarcoma,

Table 1 Patient and tumour characteristics

Variable All tumours

Female, n (%) 17 (62.9)

Age at diagnosis, months, 0.1 (0-128.5)median (range)

Body weight, kg 3.6 (2.3-38.2)

BSA, m2 0.22 (0.21-1.26)

Symptoms and features Fetal diagnosis 20 (74.1)at diagnosis, n (%) Cardiac symptoms 2 (7.4)

Neurological symptoms 2 (7.4)Incidental diagnosis 3 (11.1)

Tumour type, n (%) Rhabdomyoma 15 (55.5)Fibroma 1 (3.7)

Angiosarcoma 1 (3.7)Rhabdomyosarcoma 1 (3.7)

Unknown 9 (33.3)

Follow-up period, months, 23.0 (3.0-59.8)median (range)

Associated anomalies Tuberous sclerosis 16Coincidental cardiac lesion 3

Family history 1

Haemodynamically unstable 7Surgery 5Death 1Medication 1

Regression Partial 18Complete 2No change 2

BSA, body surface area.

Kim et al 142

Table 2 Echocardiographic and electrocardiographic features

Variable All tumours

Echocardiographic findings at diagnosis, n Multiple 17

Largest cardiac tumour size, cm 2.0 (1.0, 6.0)median (range)

Location as a single mass Interventricular septum 5Right atrium 1Mitral valve 1

Left ventricle 2Right ventricle 1

Symptomatic heart failure, n 6

Table 3 Electrocardiographic features by tumour type

Variable All tumours Rhabdomyoma Fibroma Angiosarcoma Rhabdomyosarcoma Unknown

Patients 27 15 1 1 1 9

Cardiac arrest/VF 1 1 − − − −

Atrial flutter 1 1 − − − −

PAC, PVC 4 4 − − − −

QT prolongation 4 2 − 1 − 1

WPW syndrome 1 − 1 − − −

ST depression 4 2 1 − − 1

LBBB, RBBB 3 1 − − − 2

RVH, LVH 2 1 − − − 1

Values are expressed as number of patients. VF, ventricular fibrillation; PAC, premature atrial complex; PVC, premature ventricular complex; WPW,Wolff-Parkinson-White; LBBB, left bundle branch block; RBBB, right bundle branch block; RVH, right ventricular hypertrophy; LVH, leftventricular hypertrophy.

10 had a single tumour located in the interventricular septum(five patients), right atrium (one patient), mitral valve (onepatient), left ventricle (two patients), or right ventricle (onepatient). The median cardiac tumour size was 2.0 cm (range,1.0-6.0 cm). Some patients presented with abnormalities inelectrocardiographic recordings, including three with bundlebranch block, two with ventricular hypertrophy, four withQT prolongation, four with ST depression, and one withWolff-Parkinson-White syndrome. Four patients displayedarrhythmias, which involved cardiac arrest, atrial flutter, andpremature atrial or ventricular complexes. Table 3 showsthe electrocardiographic features by tumour type.

The clinical characterist ics and treatment of

Table 4 Haemodynamically unstable patients

Patient Tumour Age at Clinical Haemodynamic Tumour Treatment Outcome

type diagnosis signs to unstable size and

(months) diagnosis symptom location

1 Angiosarcoma 128 Chest pain Massive 4 cm, Surgery Alive,pericardial effusion right atrium chemotherapy

2 Fibroma 0 Fetal diagnosis Hypotension 1.6 cm, Surgery Alive, right ventriclright ventricle exclusion, BCPS

3 Rhabdomyosarcoma 94 Seizure Brain infarction 2.5 cm, mitral valve Surgery Alive, chemotherapy

4 Rhabdomyoma 0 Fetal diagnosis Hypotension 1.4 cm, LVOT Surgery Alive

5 Rhabdomyoma 0 Fetal diagnosis Hypotension 2.4 cm, LVOT Surgery Alive

6 Rhabdomyoma 0 Fetal diagnosis Atrial flutter 4.5 cm, multiple Everolimus Alive,complete resolution

7 Unknown 0 Fetal diagnosis Severe left 4.5 cm, multiple Deathventricle dysfunction

BCPS, bilateral cavopulmonary shunt; LVOT, left ventricular outflow tract.


haemodynamically unstable patients are listed in Table 4.Patient 1 in Table 4 was diagnosed at 10 years of age andwas referred to our clinic for chest pain. Initially, shepresented with a huge tumour in the right atrium withmassive pericardial effusion. She underwent cardiac surgeryfor mass excision, was diagnosed with angiosarcoma, andwas treated with chemotherapy. She was still alive but in aterminal state at the last follow-up.

Patient 2 was diagnosed in utero. The tumour was a hugesingle mass in the right ventricle, and the patient exhibitedhypotension because the tumour occupied most of the rightventricular cavity. He underwent cardiac surgery. Thetumour was determined to be a benign fibroma; however,because the tumour was intramural, we performed only apartial excision. The tumour became larger 2 months later,and the patient was transferred to another hospital wherehe underwent right ventricular exclusion with a bilateralcavopulmonary shunt at age 7 months. He presented withpalpitations, atrial arrhythmia was confirmed, and he wasgiven an anti-arrhythmic agent at the last follow-up.

Patient 3 presented with a seizure caused by embolicphenomena. He had a mitral valve tumour that wasconfirmed histologically as rhabdomyosarcoma aftersurgical removal. Patients 1 and 3 with malignant sarcomachose to be transferred to another hospital forchemotherapy. Patient 3 was not followed up after surgery.

Patients 4 and 5 were diagnosed prenatally with a leftventricular outflow tract tumour. On the first day of life,these patients were managed with prostaglandin infusion tomaintain cardiac output . Cardiac surgery withcardiopulmonary bypass was undertaken on postnatal day2 in these two patients. Rhabdomyomas were confirmedhistologically and were associated with tuberous sclerosis.Mutation of TSC2 was confirmed in both patients. The twopatients showed good outcomes and presented with nosymptoms and or significant residual tumour at the lastfollow-up.

Patient 6 was diagnosed antenatally with multiple cardiactumours but was asymptomatic. Tuberous sclerosis wasconfirmed in infancy. At age 15 months, he presented withhaemodynamically unstable arrhythmia with hypotensionand atrial flutter with multiform frequent prematureventricular complexes (Figure 1). The patient was treatedinitially with direct current cardioversion and was infusedwith amiodarone and a beta-blocker in the intensive careunit. Because of recurrent arrhythmia, he was treated witheverolimus. The patient also exhibited subependymal giantcell astrocytoma in the brain. Five months later, the cardiactumour had regressed completely after everolimus treatment

(Figure 1).Patient 7 was diagnosed antenatally with multiple huge

cardiac tumours mainly in the left ventricle, which werethought to be rhabdomyomas. He died of severe heart failureon postnatal day 1.

The other 20 patients had no significant clinical featuresor symptoms at the last follow-up.

Discussion

Primary cardiac tumours are rare in infants and children.5Most primary cardiac tumours in children are benign, butabout 10% are malignant. Even some benign tumours requiretreatment because of the symptoms.6 In our patients, mostof the cardiac tumours were rhabdomyomas accompaniedby tuberous sclerosis and were diagnosed in the antenatalperiod by fetal echocardiography. This is a similar findingto that in other reports of cardiac tumours in children, butwhat is unique about our report is that we treated a higherproportion of patients who had been diagnosed in fetal lifethan in other series.7,8 Many patients were referred to ourcentre for a precise diagnosis and further management. Fetaldiagnosis did not influence clinical presentation in ouranalysis. However, fetal diagnosis could be helpful formanaging emergent cases that require surgery in the neonatalperiod. In our centre, fetal echocardiography could detectabout 60% of the patients needing surgery; all were managedin the neonatal period and their outcomes were very good.Two patients who had not been diagnosed by fetalechocardiography had malignant tumours, possibly becausechildren with malignant tumours tend to present at an olderage than those with benign tumours. Fetal diagnosis isimportant for determining prognosis and outcome in patientswith a haemodynamically compromising tumour.

Most rhabdomyomas took a benign course and neededno treatment, but two patients with rhabdomyomasunderwent cardiac surgery because the tumours werelocated in the left ventricular outflow tract.

Rhabdomyomas usually regress and most require nointervention; therefore, the surgical rate for rhabdomyomasis low compared with other types of cardiac tumours.6,9 Asshown in our study, obstructive rhabdomyoma in the neonateis a critical indication for cardiac surgery. This is one reasonfor antenatal diagnosis to exclude a high risk of suchtumours. One of our patients who presented withhaemodynamically unstable arrhythmia had a goodresponse to everolimus, which decreased the tumour sizequickly. Everolimus, an inhibitor of mechanistic target of

Kim et al 144

Figure 1 (A) Atrial flutter with multiform frequent premature ventricular complexes in patient 6. (B) Huge mass in the right atriumand ventricle at age 15 months. (C) Regressed tumours after 1 month on everolimus.

rapamycin (mTOR), is useful in the treatment of patientswith a TSC mutation and subependymal giant cellastrocytoma or renal angiomyolipoma. Recent reports havesuggested that mTOR inhibitors are also effective in causingr e g r e s s i o n o f p r e v i o u s l y u n c h a n g e d c a r d i a crhabdomyomas.10,11

Cardiac surgery is the best option for relieving an outflowtract obstruction. However, surgery might not be as effectiveas medication in patients with significant arrhythmia causedby multiple cardiac tumours. Extensive intramuralmyocardial involvement may have a high mortality risk,

and patients with such involvement are not considered tobe eligible for surgical resection.11 Our patient also hadunchanged multiple cardiac tumours, but everolimus causedmarked regression of the tumours, and the patient no longerexhibited arrhythmia. The patient who died had multiplehuge cardiac tumours, which were thought to berhabdomyomas, although we could not confirm the exacttumour type by histological examination.

Although we could not evaluate the risk factors fortumour mortality because of the small sample population,other data have suggested that the risk factors for mortality


are tumour size, dysrhythmia, hydrops, and early age of onsetof physiological disturbance.6 In haemodynamically unstablepatients, especially neonates, the primary goals of thetreatment are restoration and preservation of sufficient heartfunction. Continuous intravenous prostaglandin E1 infusionand mechanical ventilator support are important forestablishing pulmonary blood flow and haemodynamicstability. The choice of appropriate options for treatment ofrhabdomyoma should be made after considering the tumourcharacteristics.

Of the three other patients who underwent surgery, twohad a malignant sarcoma and one had a benign fibroma.Malignant tumours were not diagnosed in the prenatalperiod but in childhood or adolescence, and were associatedwith symptoms such as chest pain, which was associatedwith pericardial effusion, or seizure associated withsystemic embolisation of the tumour. Cardiac sarcomatreatment includes primarily excision with curative intent,and these patients were managed with complete resectionfollowed by chemotherapy. However, the results for patientswith a malignant tumour are poor; the reported 2-yearactuarial survival probability was 14% in a series of 21patients with primary cardiac sarcomas, and the recurrenceof cardiac sarcoma, which is associated with a very poorprognosis, was fairly common.8,12

Cardiac fibroma typically presents as a single solidtumour, as in our patients, most commonly in theinterventricular septum, but can also be in the wall of anychamber. F ibroma i s a ben ign tumour, bu t i sunencapsulated, invades the myocardium, and averages5 cm in its largest dimension.13 Mortality is relativelyhigh in patients with a large infiltrative right-sidedfibroma. One of our patients had a large and infiltrativetumour in the right ventricle, but he could not undergototal resection. In the follow-up, the tumour grew at arapid rate and caused right ventricular dysfunction. Thepatient underwent right ventricle exclusion withbidirectional cavopulmonary shunt.

The electrocardiographic findings in our study varied,but only one patient died. Cardiac tumours can inducevarious cardiac arrhythmias, including premature beats,supraventricular tachycardia, ventricular tachycardia,second- or third-degree atrioventricular block, sinus nodaldysfunction, or preexcitation syndrome.14 Clinicallysignificant arrhythmias were documented in 24% of thepatients reported by Miyake's group.15 They found that thegreatest arrhythmia burden occurred in the fibroma patients,who were strongly predisposed to ventricular tachycardia.Arrhythmia control can include medication, placement of

an implantable cardioverter defibrillator, or surgicalexcision. Miyake and colleagues have suggested that tumourresection can be an effective option, especially for patientswith ventricular tachycardia. Our rhabdomyoma patientswho had atrial flutter with multiple premature ventricularcontractions responded to medical management includingamiodarone, beta-blocker, and direct current cardioversion.However, frequent relapse of arrhythmia requiredreadmission to the intensive care unit. Althoughimprovement in rhythm status with tumour regression canoccur in patients with rhabdomyoma and severe arrhythmia,medical therapy including everolimus may be useful forsymptom relief.

Most tumours of unknown origin had a benign course,which appeared as no change, partial regression, orcomplete regression, except for the one patient who died.Although there was no proven association with tuberoussclerosis, most of them were considered to be benignrhabdomyoma, fibroma, or hamartoma according to thetumour features.

Conclusion

In conclusion, most of the cardiac tumours treated in ourcentre were rhabdomyomas associated with tuberoussclerosis, although some malignant tumours were found. Thesurgical outcomes for the cardiac tumours were good.Clinicians should consider various treatment options forhaemodynamically unstable paediatric patients with acardiac tumour.


None

References

1. Gunther T, Schreiber C, Noebauer C, Eicken A, Lange R.Treatment strategies for pediatric patients with primary cardiacand pericardial tumors: a 30-year review. Pediatr Cardiol 2008;29:1071-6.

2. Elderkin RA, Radford DJ. Primary cardiac tumours in a paediatricpopulation. J Paediatr Child Health 2002;38:173-7.

3. Uzun O, Wilson DG, Vujanic GM, Parsons JM, De Giovanni JV.Cardiac tumours in children. Orphanet J Rare Dis 2007;2:11.

4. Tao TY, Yahyavi-Firouz-Abadi N, Singh GK, Bhalla S. Pediatriccardiac tumors: clinical and imaging features. Radiographics 2014;34:1031-46.

Kim et al 146

5. Nield LE, Mendelson M, Ahmad N, Manlhiot C, Jaeggi ET,McCrindle BW. Clinical review of obstructive primary cardiactumors in childhood. Congenit Heart Dis 2014;9:244-51.

6. Bielefeld KJ, Moller JH. Cardiac tumors in infants and children:study of 120 operated patients. Pediatr Cardiol 2013;34:125-8.

7. Linnemeier L, Benneyworth BD, Turrentine M, Rodefeld M,Brown J. Pediatric cardiac tumors: a 45-year, single-institutionreview. World J Pediatr Congenit Heart Surg 2015;6:215-9.

8. Tzani A, Doulamis IP, Mylonas KS, Avgerinos DV, Nasioudis D.Cardiac tumors in pediatric patients: a systematic review. WorldJ Pediatr Congenit Heart Surg 2017;8:624-32.

9. Sciacca P, Giacchi V, Mattia C, et al. Rhabdomyomas and tuberoussclerosis complex: our experience in 33 cases. BMC CardiovascDisord 2014;14:66.

10. Tiberio D, Franz DN, Phillips JR. Regression of a cardiacrhabdomyoma in a patient receiving everolimus. Pediatrics 2011;

127:e1335-7.11. Demir HA, Ekici F, Yazal Erdem A, Emir S, Tunc B. Everolimus:

a challenging drug in the treatment of multifocal inoperable cardiacrhabdomyoma. Pediatrics 2012;130:e243-7.

12. Putnam JB Jr, Sweeney MS, Colon R, Lanza LA, Frazier OH,Cooley DA. Primary cardiac sarcomas. Ann Thorac Surg 1991;51:906-10.

13. Ganame J, D'Hooge J, Mertens L. Different deformation patternsin intracardiac tumors. Eur J Echocardiogr 2005;6:461-4.

14. Shiono J, Horigome H, Yasui S, et al. Electrocardiographicchanges in patients with cardiac rhabdomyomas associated withtuberous sclerosis. Cardiol Young 2003;13:258-63.

15. Miyake CY, Del Nido PJ, Alexander ME, et al. Cardiac tumorsand associated arrhythmias in pediatric patients, with observationson surgical therapy for ventricular tachycardia. J Am Coll Cardiol2011;58:1903-9.


Chinese Boy with Normal Initial Peroxisomal Blood Assays:A Diagnostic Pitfall in the Workup for Infantile Refsum Disease

KL CHEUNG, CH KO, HHC LEE , CM MAK

Abstract The first Chinese case of Infantile Refsum disease (IRD), one of less severe form of peroxisomebiogenesis disorders, is illustrated in this case report. The patient presented with global developmentaldelay without facial dysmorphism and his routine metabolic screening was negative. Initial very longchain fatty acids (VLCFA) assays showed marginally elevated C26/C22 ratio and pristanic and phytanicacid levels were within normal limits. Magnetic resonance imaging and nerve conduction velocity findingsrevealed abnormal results and VLCFA assays were subsequently repeated which demonstrated persistentlyelevated C26/C22 ratio and raised pristanic and phytanic acids. The diagnosis of IRD was confirmed withgenetic study. The diagnostic approach and the corresponding investigations, in particular the use ofserial VLCFA assays to improve the diagnostic yield, are highlighted.

Key words Fatty Acids; Genetic testing; Infantile; Peroxisomal biogenesis disorders; Refsum disease

Department of Paediatrics and Adolescent Medicine,Caritas Medical Centre, 111 Wing Hong Street, ShamShui Po, Kowloon, Hong Kong, China

KL CHEUNG MBChB(CUHK)CH KO FHKAM(Paediatrics), FRCP(Glasg)

Department of Pathology, Princess Margaret Hospital,2-10 Princess Margaret Hospital Road, Lai Chi Kok,Kowloon, Hong Kong, China

HHC LEE FRCPath, FHKAM(Pathology)CM MAK MD, PhD

Correspondence to: Dr KL CHEUNG


Received August 21, 2017

Case Report

Introduction

Infantile Refsum disease (IRD) is a rare neuro-metabolicdisease which belongs to the group of peroxisomalbiogenesis disorders (PBD) and is perceived to be a lesssevere form of Zellweger spectrum disorders (ZSD). Dueto multiple defective peroxisomal enzymes, there areaccumulation of phytanic acid and increased levels of verylong chain fatty acids (VLCFA) in the blood and tissues

leading to damage in brain parenchyma, peripheral nerves,liver and retina. The main manifestations include peripheralpolyneuropathy, retinitis pigmentosa, cerebellar ataxia,developmental delay and growth retardation. Other clinicalfindings are craniofacial dysmorphism including highforehead, frontal bossing, antimongolian palpebral slant,sunken eyes, flat nasal root, elongated philtrum andbilateral simian creases, sensorineural deafness, seizures,hepatomegaly and malabsorption. The disease is inheritedin autosomal recessive manner and confirmed by mutationsin PEX1 gene.

The current case illustrates the diagnostic approach tothis rare disorder, and highlights the importance ofrepeating VLFCA assays to improve the diagnostic yieldin suspected cases.

Case Report

A twenty-six-month-old Chinese boy was referred tous for global developmental delay. He was born at full termby uncomplicated spontaneous vaginal delivery with birthweight of 2.95 kg. Antenatal and postnatal histories wereunremarkable. The parents were non-consanguineous andthere was no known family history of genetic diseases. The

Cheung et al 148

boy was the first child of the couple. Assessment at age20 months revealed developmental age corresponding to10 to 12 months in various aspects. Follow-up assessmentat age 3 revealed milestones corresponding to 16 to 19months of developmental age. He also had bilateralmoderate to severe hearing loss and prominent sialorrhoea.General health was satisfactory and there was no recurrentseizure. Physical examination revealed prominentepicanthic folds and slightly low set ears with paucity offacial expression. No other craniofacial dysmorphicfeatures were noted. He had generalised hypotonia andhyporeflexia. There was no skeletal abnormality. Systemicexamination was otherwise unremarkable. Baselineinvestigations, including complete blood picture, liver andrenal function, thyroid function, ammonia, lactate andpyruvate were unremarkable. Metabolic study revealednormal organic acid and amino acid profiles. Computedtomography of brain showed no definite abnormality.

At age two the child could walk on level ground byhimself, but could not manage stairs. He graduallydeteriorated by age 4.5 years, requiring one hand to beheld on walking. He could build tower of three, obeysimple command and speak 5-10 single words. Clinicalexamination showed persistent hypotonia, discrepant lowerlimb weakness with unsteady gait. Nerve conductionvelocity (NCV) showed demyelinating motor neuropathypredominantly affecting bilateral lower limbs. Cranialmagnetic resonance imaging (MRI) at age 4 years 10months revealed T2 hyperintensity involving bilateralparieto-occipital regions & splenium of corpus callosum(Figure 1). Transferrin isoforms, acetylcholine receptor

antibodies, carnitine profiles, morning cortisol level andlipid profiles were normal. Initial VLCFA assays showedmarginally elevated C26/C22 ratio with normal C22:0,C24:0 and C26:0 levels; pristanic and phytanic acid levelswere within normal limits. In view of abnormal MRI andNCV findings, assays were repeated four months laterwhich demonstrated persistently elevated C26/C22 ratioand increasing C26:0 levels (but still within normal limits),with raised pristanic and phytanic acids. The pristanic tophytanic ratio was also elevated (Table 1). Genetic studyrevealed two heterozygous mutations for PEX1(NM_000466.2:c .2348A>G (p.Asp783Gly) andc.2783+2T>C), confirming the diagnosis of PBD (MIM#601539) with phenotype compatible with IRD.

The child was put on a phytanic acid restriction diet,with fat soluble vitamins and docosahexaenoic acid (DHA)supplementation. Despite initial transient improvement inself-care, he ran into relentless deterioration over the next12 months. By age 6 he became minimally conscious,bedridden and required tube feeding. Repeat MRI brain atage 7 showed extensive leukodystrophic changes (Figure2). There was persistent elevation of C26/C22 and C26:0levels, with normalisation of pristanic and phytanic acids(Table 1).

Discussion

Peroxisomes are subcellular organelles present in allhuman tissues other than mature erythrocytes. They aremainly involved in lipid metabolisms. The reactions or

Figure 1 Cranial MRI at age 4 years 10 months showing T2 hyperintensity involving bilateral parieto-occipital regions and splenium of corpuscallosum.

Diagnostic Pitfall: Infantile Refsum Disease149

components of such reactions, namely plasmalogen, bileacid and DHA biosynthesis, and oxidation of pipecolic acidand VLCFA take place exclusively in the perioxisomes.1

Peroxisomal matrix and membrane proteins are initiallysynthesised in free polyribosomes, enter cytoplasm andare directed to the preexisting peroxisomes from whichnew peroxisomes arise by budding. The targetingmechanism to the organelle is guided by peroxisomaltargeting sequences (PTS1 and PTS2). Complementationanalysis identifies up to 23 PEX genes coding for peroxins,which facilitate importation of targeted proteins into theperoxisomes through receptor docking, stability andtranslocation. PEX1 gene encodes AAA - ATPases fornormal import of PTS1 proteins; PEX1 mutations arefound in majority of cases with ZSD.2

Peroxisomal disorders are genetically heterogeneousgroup of disorders, classified into two categories. The firstcategory is PBD, in which peroxisome fails to formnormally. PBD are further sub-classified into ZSD andrhizomelic chondrodysplasia punctate (RCDP).1 Thespectrum of ZSD ranges from severe Zellweger syndrometo more indolent neonatal adrenoleukodystrophy (NALD)and IRD, which share similar clinical and biochemicalfeatures without dist inct phenotypes. RCDP ischaracterised by severe skeletal abnormalities, cataractsand facial dysmorphism resulting from impairment inplasmalogen synthesis.2 The second category is disordersof single peroxisomal protein in which peroxisomestructure is intact but with defect of single peroxisomalenzymes; classical conditions include X-linkedadrenoleukodystrophy (X-ALD) and D-bifunctional

protein (DBP) deficiency.1

Historical cohorts demonstrate nearly half of presumedgenetic leukodystrophies remain unresolved. Peroxisomaldisorders are rare; the collective estimated frequency is 1in 5000 to 10000 live births. X-ALD is the mostcommonly identified subtype, with an estimated incidenceof 1:17000. Detection of elevated VLCFA in plasmasuggests possibility of PBD and biochemical studies onfibroblasts and genetic test are necessary for properdiagnosis. Newborn screening for PBD utilising liquidchromatography and tandem mass spectrometry to detectelevated VLCFA in blood spots is rarely included innewborn metabolic screening worldwide. Pilot expandednewborn metabolic screening programme in Hong Kongoffers screening for multiple aminoacidopathies, organicacidurias, and fatty acid oxidation disorders but does not

Figure 2 Cranial MRI at age of 6 years old and 1 month showing markedly increased extensive leukodystrophic changes.

Table 1 Patient's plasma levels of very long chain fatty acids

Age 4y 8m Age 5 Age 7 Reference SI unit

C22:0 31.5 39.1 53.1 <96.3 umol/L

C24:0 33.5 38.7 49.4 <91.4 umol/L

C26:0 0.87 1.03 1.67↑↑↑↑↑ <1.30 umol/L

C24:0/C22:0 1.06 0.99 0.93 <1.39

C26:0/C22:0 0.028↑↑↑↑↑ 0.026↑↑↑↑↑ 0.032↑↑↑↑↑ <0.023

Pristanic acid 2.23 6.7↑↑↑↑↑ 0.33 <2.98 umol/L

Phytanic acid 5.69 13.7↑↑↑↑↑ 1.28 <9.88 umol/L

Pristanic/Phytanic 0.39 0.49↑↑↑↑↑ 0.26 <0.39

Cheung et al 150

include peroxisomal disorders.3 In contrast, legislation fornewborn screening of X-ALD has been passed in somestates of United States including New York; this allowsearly detection and treatment of peroxisomal disorders.4

To our knowledge, this is the first reported case ofIRD in Hong Kong. One local Chinese case of DBPdefic iency with features mimicking Zel lwegersyndrome has been reported in 2009.5 As IRD runs amore indolent course than classic Zellweger syndrome,detection in early stage may provide a window fortherapeutic intervention. Early diagnosis of IRD ischallenging; the init ial signs can be subtle andnonspecific, with emerging intellectual deficits,progress ive sensor ineura l deafness and motordeterioration. Dysmorphic features can be subtle withabsence of hepatomegaly and the diagnosis relies onidentification of a constellation of clinical, radiologicaland electrophysiological features. The findings of NCVabnormalities and MRI leukodystrophic changes shouldprompt further biochemical confirmation. However,initial VLFCA, pristanic and phytanic acid assays couldyield false negative results. Van der Knapp et alretrospectively reviewed the pattern of six unclassifiedleukoencephalopathies; sequential MRI abnormalitiesstarted from dentate nucleus and superior cerebellarpeduncles, subsequently affecting cerebellar whitematter and brainstem, followed by parietal occipitalwhite matter, splenium and internal capsule posteriorlimbs, with eventual diffuse involvement. Initialperoxisomal blood tests were notably all inconclusive.Diagnoses were eventually substantiated by fibroblastp e r o s i x o m a l im m u n o f l u o r e s c e n c e s t u d i e s ,complementation analysis and molecular studies,confirming three cases of ZSD and 3 DBP deficiencies.6

Berendse et al reviewed and followed up nineteen patientsfor duration of mean 16 years. Three out of nineteen patientshad insignificant peroxisomal blood assays and werediagnosed by complete analysis in skin fibroblasts.7

Ratbi et al identified six cases of Heimler syndromeconfirmed by bi-allelic mutations in PEX1 or PEX6. Thisis defined as a mild variant of PBD with unremarkable bloodand skin fibroblast analyses.8 As illustrated in the presentcase, negative initial peroxisomal blood assaysdo not rule out the diagnosis. Plasmalogen andcomplementation studies are not available locally. In caseswith concordant phenotypic and radiological findings, it ismandatory to repeat biochemical assays to improve thediagnostic yield and guide further targeted molecularstudies.

Treatment of IRD is mainly supportive and phytanic acid-restricted diet remains the mainstay therapy. DHA ispolyunsaturated fatty acid that has important functions inretina and brain synapses. DHA synthesis takes place inthe peroxisome and is deficient in PBD patients. Anecdotalreports revealed improvement in clinical status from DHAsupplementation.9 However, in a randomised controlledtrial of DHA at 100 mg/kg/day for PBD patients, there wasno significant difference in biochemical function,electroretinogram and growth after one year.10 In thepresent case, despite normalisation of pristanic/phytanicacid levels on phytanic acid-restricted diet and DHAsupplementation, the relentless degenerative course hadnot been altered.


The authors declare that they have no financial or otherconflicts of interest in relation to this publication.

References

1. Moser HW. Peroxisomal disorders. Semin Pediatr Neurol 1996;3:298-304.

2. Raymond GV. Peroxisomal disorders. Curr Opin Pediatr 1999;11:572-6.

3. Chong SC, Law LK, Hui J, Lai CY, Leung TY, Yuen YP.Expanded newborn metabolic screening programme in HongKong: a three-year journey. Hong Kong Med J 2017;23:489-96.

4. Braverman NE, Raymond GV, Rizzo WB, et al. Peroxisomebiogenesis disorders in the Zellweger spectrum: An overviewof current diagnosis, clinical manifestations, and treatmentguidelines. Mol Genet Metab 2016;117:313-21.

5. Siu LY, Kwong L, Wong SN, Kwong NS. Neonatal seizure: Arare aetiology easily missed by routine metabolic screening.HK J Paediatr (new series) 2009;14:37-41.

6. Van der Knaap MS, Wassmer E, Wolf NI, et al. MRI asdiagnostic tool in early-onset peroxisomal disorders. Neurology2012;78:1304-8.

7. Berendse K, Engelen M, Ferdinandusse S, et al. Zellwegerspectrum disorders: clinical manifestations in patients survivinginto adulthood. J Inherit Metab Dis 2016;39:93-106.

8. Ratbi I, Falkenberg K, Sommen M, et al. Heimler SyndromeIs Caused by Hypomorphic Mutations in the Peroxisome-Biogenesis Genes PEX1 and PEX6. Am J Hum Genet 2015;97:535-45.

9. Martinez M. Docosahexaenoic acid therapy in docosahexaenoicacid deficient patients with disorders of peroxisome biogenesis.Lipids 1996;31(Suppl):145-52.

10. Paker AM, Sunness JS, Brereton NH, et al. Docosahexaenoicacid therapy in peroxisomal diseases. Neurology 2010 75;9:826-30.


What Can It Be If Not a Simple Haemangioma?

PT YU, HM LUK, IFM LO

Abstract Capillary haemangioma is a common dermatological condition, which present in approximately 1-2% ofinfant. While 30% appear at birth, almost all will be apparent by the age of 6 months. Capillarymalformation-arteriovenous malformation (CM-AVM) syndrome is an autosomal dominant disorder thatcharacterised by the presence of multiple small capillary malformation, formation of high flow vascularanomalies like arteriovenous malformation (AVMs) and / or arteriovenous fistulas (AFVs) which typicallydevelop in skin, muscle, brain, spine and rarely in bone. The prevalence is about 1:100,000.1 Despite itsrarity, it is still an important disease to recognise and differentiate from simple capillary haemangioma. AsCM-AVM syndrome may lead to life threatening complications like congestive heart failure, rupture ofAVM especially intracranial lesion which may leads to devastating neurological consequences. Here wepresent the first reported case of RASA1 related CM-AVM syndrome in Chinese, discuss on its clinicalfeatures and differential diagnosis, together with the indication for genetic referral.

Key words Capillary haemangioma; Capillary malformation-arteriovenous malformation syndrome; CM-AVM;RASA1

Clinical Genet ic Service , Department of Health ,Cheung Sha Wan Jockey Club Clinic, 2 Kwong Lee Road,Shamshuipo, Kowloon, Hong Kong, China

PT YU MBBS(HK), FHKAM(Paed), FHKCPaedHM LUK MD(HK), FHKAM(Paed), FRCPCH(UK)IFM LO MBChB(HK), FHKAM(Paed), FHKCPaed

Correspondence to: Dr HM LUK



Case Report

Case Report

A 2.7 kg female baby was born at 35+5 week ofgestation. Antenatal ultrasound at 34 week of gestationshowed features of hydrops fetalis. Upon delivery, she wasnoted to have generalised lymphoedema with bilateralchylothorax, which resolved and stabilised after drainageand octreotide treatment. Echocardiogram showed normalcardiac structure. Other investigations for hydrops includedcongenital infection screening, microarray and Karyotyping

were all negative. Family history was unremarkable. Ahaemangioma was developed at the left angle of mouthsince age of 1 month old. She was first seen in our geneticclinic at 4 months of age for subtle dysmorphic featuresand perinatal history of hydrops fetalis. Physicalexamination at 4 months old showed she had body weightof 5.8 kg (10th centile), body length 63 cm (25-50th centile)and head circumference of 40 cm (10th centile). A capillaryhaemangioma was noted at the left angle of mouth. Therewas subtle dysmorphic features included hypertelorism,downslanting palpebral fissure, depressed nasal bridge andlow set posteriorly rotated ears (Figure 1). The clinicaldiagnosis of Noonan syndrome was initially made. Genetictesting for RASopathy included PTPN11, BRAF, CBL,HRAS, KRAS, MAP2K1, MAP2K2, NRAS, RAF1, RIT1,SHOC2 and SOS1 genes were all negative. On subsequentfollow up, there was gradual progression of thehaemangioma that extending to the gum which resulted indental malocclusion (Figure 1). Another 2x3 cm paraspinalsubcutaneous haemangioma was also noted and confirmedby ultrasound at 1 year old. MRI brain and head with contrast

Yu et al 152

at 3 years old showed a haemangioma with underlyinghigh flow vessels within the subcutaneous fat over leftangle of mouth. No arteriovenous malformation detectedin brain parenchyma. She got mild developmental delaybut catched up with training. In view of the above vascularmalformation phenotype, capillary malformation-arteriovenous malformation (CM-AVM) syndrome wassuspected and RASA1 {NM_002890.2} gene sequencingwas performed. A de novo heterozygous one base pairdeletion c.482delG in exon 1 of RASA1 gene wasdetected. The molecular diagnosis of RASA1 related CM-AVM syndrome was substantiated. This was also a novelmutation in the literature.

Discussion

Familial multiple capillary haemangioma and/or AVM hasbeen described for many years and the genetic locus was

firstly mapped to chromosome 5q by Breugem et al2 andEerola et al3 in 2002. In subsequent year, RASA1 mutationwas identified to be the cause of CM-AVM and Parkes-Weber syndrome which have CM-AVM and soft tissue orskeletal hypertrophy of the limbs.4 The underlyingpathophysiology of RASA1 mutation that lead to vascularmalformation are remained to be elucidated. But it isproposed that RASA1 encoded for p120-RasGTPase-activating protein (p120-RasGAP), which negativelyregulated the Ras/MAP-kinase pathway. Ras/MAP-kinasepathway is important for endothelial cells development.Therefore, dysregulation of RASA1 function would resultin abnormal microvascular networks formation.5,6 Withhighly selected patient with CM-AVM, the diagnostic yieldof RASA1 testing is around 50-60%.7

Typical features of CM-AVM is the presence of CMs(98.5%) which mainly found on skin and rarely on mucosa.Majority of them (97%) will have multiple cutaneouslesions. Most CMs are round or oval shape in pink colour

Figure 1 (a) and (b) Evolving of left jaw haemangioma at the age of 5 months and 3 years old respectively. (c) and(d) Multifocal capillary haemangiomas with characteristic halo (blue arrows) over the upper arms. (e) At age of 6years old, the vascular malformation over the left angle of mouth has extended into oral cavity that lead to gumswelling and dental malocclusion. (Consent for publication has been obtained).

RASA1 Related Disease in HK153

and some have characteristic surrounding halo. Majority ofthem are present at birth and increased in size and numbergradually. About 60% of patients have fast flow vascularmalformation. These AVMs and AVFs are mainly locatedon the extremities, intracranial (25%), intraspinal (15%),head and neck region (21%). About 5% will present withfeatures of Parkes Weber syndrome. AVMs and AVFs maypresent in neurological symptoms if located intracraniallyor intraspinally that include seizure, migraine or stroke.Large fast flow vascular malformation could alsocomplicate with congestive heart failure. Non immunehydrops have rarely reported for CM-AVM.8

Differential diagnosis of RASA1 related disorder includehereditary haemorrhagic telangiectasia, Klippel Trenaunaysyndrome, Sturge Weber syndrome, PTEN hamartomatumour syndrome, multiple cutaneous and mucosal venousmalformations, hereditary glomuvenous malformations.Clinical presentation of the above entities and theirdistinguishing features from RASA1 related disorder aresummarised in Table 1.

Capillary haemangoima and port wine stain are extremely

common in paediatric population, with a birth incidence of1-2%. Identification of RASA1 related disorder is importantas it may associated with potential sinister and lifethreatening complications. Therefore, we proposed severalindications for referral to dermatologist or clinical geneticist.It includes1. presence of multifocal, atypical pink to reddish brown,

round to oval lesion, with or without a halo surroundingthe lesion;

2. the presence of arteriovenous malformations (AVMs)or arteriovenous fistulas (AVFs) in soft tissue, bone andbrain;

3. the presence of soft tissue and bony overgrowth, whichmay point to the diagnosis of Parkes Weber syndrome;

4. the presence of positive family history.

Regarding for the management of CM-AVM, baselinebrain and spinal imaging is recommended after the initialdiagnosis to look for AVMs and AVFs. However, there is noconsensus for the subsequent radiographic evaluation.Referral to dermatologist for assessment and follow up is

Table 1 Differential diagnosis of RASA1 related disorder

Gene involved CMs AVM Soft tissue Distinguishing features from

overgrowth RASA1 related disorder

Hereditary haemorrhagic ENG − + − - Characteristic lip or tongue telangiectasiateleangiectasia ACVRL1 - AVM in liver or lung

SMAD4 - Epitaxis and abnormal vessels in mucosa is common- GI bleeding

Sturge Weber syndrome GNAQ + − − - Facial cutaneous vascular malformations overophthalmic branch of trigeminal nerve

- Seizures- Glaucoma

Klippel Trenaunay PIK3CA + − + - Vascular malformations are typically low-flowWeber syndrome lesions without high-flow AVMs

PTEN hamartoma tumour PTEN − + +/- - Vascular anomalies are usually intramuscular, aresyndrome associated with ectopic fat, and severely disrupt

tissue architecture- Tumour predisposition (breast and thyroid cancer)- Macrocephaly- Papillomatous papules

Hereditary glomuvenous GLMN + − − - Hyperkeratotic, raised and nodular with amalformations cobblestone surface

- Bluish purple in colour

Multiple cutaneous and TEK + + − - Small, multifocal bluish cutaneous ormucosal venous (especially mucosal venous malformationsmalformations involving muscle) - May invade subcutaneous muscle & cause pain

Yu et al 154

also recommended. Embolisation or surgical treatment maybe occasionally needed depending on the size and locationof the AVMs and AVFs. Furthermore, it is important to watchout heart failure symptom. For Parkes Weber syndrome,input from orthopedic surgeon would be necessary. Finally,genetic counselling is important. Prenatal diagnosis andfamily planning issues can be addressed during theconsultation.

In summary, we report the first case of RASA1 relatedCM-AVM syndrome in Chinese that present initially withhydrops fetalis with subsequent multiple cutaneouscapillary haemangioma. Increase in clinical awareness ofthis disease entity is important due to the possibility ofassociated devastating complications.

Acknowledgement

We are thankful to the family for their consent for thepublication of their clinical photos.


None

References

1. Revencu N, Boon LM, Mulliken JB, et al. Parkes Weber syndrome,vein of Galen aneurysmal malformation, and other fast-flowvascular anomalies are caused by RASA1 mutations. Hum Mutat2008;29:959-65.

2. Breugem CC, Alders M, Salieb-Beugelaar GB, et al. A locus forhereditary capillary malformations mapped on chromosome 5q.Hum Genet 2002;110:343-7.

3. Eerola I, Boon LM, Wantanabe S, Grynberg H, Mulliken JB,Vikkula M. Locus for susceptibility for familial capillarymalformation ('portwine stain') maps to 5q. Eur J Hum Genet2002;10:375-80.

4. Mulliken JB, Young AE, eds. Vascular Birthmarks: Hemangiomasand Vascular Malformations. Philadelphia, PA: WB Saunders Co;1988.

5. Anand S, Majeti BK, Acevedo LM, et al. MicroRNA-132-mediated loss of p120RasGAP activates the endothelium tofacilitate pathological angiogenesis. Nat Med 2010;16:909-14.

6. Eerola I, Boon LM, Mulliken JB, et al., Capillary malformation-arteriovenous malformation, a new clinical and genetic disordercaused by RASA1 mutations. Am J Hum Genet 2003;73:1240-9.

7. Revencu N, Boon LM, Mendola A, et al. RASA1 mutations andassociated phenotypes in 68 families with capillary malformation-arteriovenous malformation. Hum Mutat 2013;34:1632-41.

8. Overcash RT, Gibu CK, Jones MC, Ramos GA, Andreasen TS.Maternal and fetal capillary malformation-arteriovenousmalformation (CM-AVM) due to a novel RASA1 mutationpresenting with prenatal non-immune hydrops fetalis. Am J MedGenet A 2015;167A:2440-3.


Spontaneous Acute Epidural Haematoma in a PaediatricPatient with Congenital Afibrinogenaemia

U OCAK, P ESER OCAK, B GUNGORER

Abstract Congenital afibrinogenaemia, a rare disorder of coagulation, is characterised by the congenital absence offibrinogen. Uncontrolled umbilical cord stump bleeding is usually the first manifestation of the disease.Although various spontaneous haemorrhages may occur during lifetime, spontaneous epidural haematomais rare. We report a case of congenital afibrinogenaemia in a paediatric patient who admitted to theemergency department with a complaint of headache. Computerised tomography scan of the head revealeda right parietal acute epidural haematoma. The patient underwent surgical evacuation subsequent tofibrinogen replacement in the emergency room. We decided to present our case since congenitalafibrinogenaemia is rare in the emergency care. This report underlines congenital afibrinogenaemia as apotential cause of spontaneous acute epidural haematoma in paediatric patients which may require emergentintervention before the development of irreversible neurological conditions.

Key words Congenital afibrinogenaemia; Paediatric emergency; Spontaneous epidural haematoma

Department of Emergency Medicine, Bozok University Schoolof Medicine, Yozgat 66100, Turkey

U OCAK MD

Department of Neurosurgery, Yozgat State Hospital, Yozgat66100, Turkey

P ESER OCAK MD

Department of Emergency Medicine, Yozgat State Hospital,Yozgat 66100, Turkey

B GUNGORER MD

Correspondence to: Dr U OCAK



Case Report

Introduction

Fibrinogen (coagulation factor 1) is a large, fibrousplasma glycoprotein that plays a role in blood clot formationand platelet aggregation after its transformation into fibrinby thrombin.1 Congenital afibrinogenaemia, a rare bleedingdisorder showing an autosomal recessive pattern of

inheritance, is caused due to the congenital absence offibrinogen. Uncontrolled umbilical stump bleeding in anewborn is the most common presentation and usuallyconsidered as diagnostic in congenital afibrinogenaemia.Although patients may present with spontaneoushaemorrhages from variable regions of the body,spontaneous intracranial haemorrhages are rare. Moreover,spontaneous epidural haematomas beyond the neonatalperiod are even rarer.2,3 We report a case of a paediatricpatient with spontaneous epidural haematoma who waspresented to the emergency department and underwentsuccessful surgical evacuation of the haematoma after theadministration of intravenous (i.v.) fibrinogen concentratein the emergency room. The rarity of this condition inemergency care prompted us to present this case.

Case Report

An 8-year-old male patient with a history of congenitalafibrinogenaemia presented to the emergency departmentwith a complaint of headache, which had started severalhours ago and worsened gradually. There was no history of

Ocak et al 156

trauma. The patient had already been diagnosed withcongenital afibrinogenaemia following massive umbilicalstump bleeding in the neonatal period; however, thisdiagnosis was not further confirmed by genetic tests.Parental consanguinity was negative, and he had a healthyolder sister. No other family members showed the signs ofa similar disease. He had experienced frequent events ofgingival bleeding and suffered from two episodes of patellarhaemorrhage requiring conservative treatment 2 and 4 yearsago. His vital signs and systemic physical examination werewithin normal limits. Glasgow coma scale revealed a scoreof 15, and there were no localising signs. He had isochoricpupils, with positive direct and indirect light reflexes.Prothrombin time and partial thromboplastin time levelswere too long to be measured. Fibrinogen level could notbe measured by functional assay. Computerisedtomography scan of the head revealed a right parietal acuteepidural haematoma that was nearly isodense to the brainparenchyma with a midline shift of 0.5 cm (Figure 1a, b).The patient underwent emergent craniotomy for acute, life-threatening epidural haematoma subsequent to i.v.administration of 100 mg/kg fibrinogen in the emergencyroom. Successful haematoma evacuation was confirmedby computerised tomography scan of the head (Figure 1c).The patient made excellent recovery, and he was dischargedhome on post-operative day 3 without any neurologicaldeficits.

Discussion

Congenital afibrinogenaemia is a rare, autosomalrecessive inherited bleeding disorder caused due to low orabsent levels of fibrinogen in the patient’s serum with anincidence of 1–2/1,000,000.1 The clinical picture is variableexhibiting umbilical cord bleeding, haemorrhage into themuscle and joints, gingival haemorrhage, gastrointestinaland genitourinary bleeding, mucosal bleeding, splenicrupture and thrombotic events.2,4 Haemorrhages usuallyfollow minor trauma or surgery, which was not the case inthis patient. Spontaneous haemorrhages may occasionallyoccur. Intracerebral haemorrhage is a rare manifestationof congenital afibrinogenaemia, which can be lethal.5

Coagulation tests such as prothrombin time, partialthromboplastin time, thrombin time and bleeding time showprolonged duration. Fibrinogen is undetectable in patients'plasma (<10 mg/dL), which is diagnostic for the disease.Functional and immunological assays reveal absence orvery low levels of fibrinogen.

Afibrinogenaemia is characterised by a tendency tobleed. However, patients may have long time intervalsbetween bleeding periods. Our patient had experienced twoepisodes of patellar haemorrhage due to minor trauma 2and 4 years ago since the time of diagnosis of congenitalafibrinogenaemia immediately after birth. Although themajority of bleeding events occur subsequent to trauma or

Figure 1 (a) and (b) Computerised tomography scan of the head revealing a right parietal acute epidural haematoma (white arrows).(c) Computerised tomography scan of the head of the same patient demonstrating complete evacuation of the haematoma following surgery.

Paediatric Congenital Afibrinogenaemia157

surgery, spontaneous haemorrhages may be encounteredoccasionally.

In general, epidural haematomas are associated with headtrauma. A spontaneous epidural haematoma is anuncommon occurrence that is primarily observed in patientswith chronic renal failure or during open heart surgery,middle meningeal artery aneurysm, systemic lupuserythematosus,6 sickle cell disease, haemorrhagic tumours,dural vascular malformations, blood coagulation disordersor infections of the adjacent regions such as otitis andparanasal sinusitis.7 Epidural haematomas, occurring dueto any cause, may become life-threatening with rapiddeterioration of the neurological status of the patient, leadingto high morbidity and mortality rates if not recognised intime.

Spontaneous epidural haematoma is a rare manifestationof congenital afibrinogenaemia in the paediatric population,as well. Hence, knowledge regarding spontaneous epiduralhaematomas in these patients does not extend beyond casereports according to the published literature.2,3 To the bestof our knowledge, our case is the third paediatric patientwith congenital afibrinogenaemia who presented withspontaneous epidural haematoma and underwent successfulsurgical evacuation.

Int racerebral haemorrhages in pat ients wi thcoagulation disorders may be undetectable or isodensedue to the clotting defect.8 Therefore, computerisedtomography scans might over look even f rankhaemorrhages in the initial phase, thus requiring furtherevaluation and surveillance of the patient. Absence ofhyperdensity and probability of isodense haemorrhagiclesions on computerised tomography scans occurring dueto the clotting disorder, requiring careful and thoroughexamination of the images, must also be considered inpatients with bleeding disorders.

Fibrinogen replacement is the treatment of choice inpatients with afibrinogenaemia with intracranial bleedingprior to surgery to avoid perioperative haemorrhagesprecluding haemostasis and safe surgery. Plasmafibrinogen levels more than 0.5-1.0 g/L are alsorecommended to be established for 4-14 days followingsurgery.9 Cryoprecipitate and fresh frozen plasma are theagents that must be used in emergent cases whenfibrinogen is not available.10 Our patient was treated withi.v. fibrinogen prior to surgery as it was available.Prophylactic treatment is still of debate due to thepotential thromboembolic complications and side effectsof blood products.

Conclusion

Emergency care providers should be aware of thespontaneous occurrence of intracranial haemorrhages inpatients with congenital afibrinogenaemia. The clinicaldiagnosis largely depends on suspicion and appropriateneuroimaging due to the probability of unexpected and life-threatening intracranial haematomas. Moreover, acutehaematomas may be undetectable or isodense to the brainparenchyma, thus rendering the diagnosis more difficult inafibrinogenaemic patients. Fibrinogen, fresh frozen plasmaor cryoprecipitate may be used in emergent medical care,which may also provide safe surgery. Consultation with apaediatric haematologist may be a reasonable option inplanning the dose and timing of the treatment. Theseapproaches can help in achieving an early diagnosis andsuccessful treatment in paediatric patients with spontaneoushaematomas before the development of irreversibleneurological conditions.

References

1. Acharya SS, Dimichele DM. Rare inherited disorders offibrinogen. Haemophilia 2007;14:1151-8.

2. Pati S, Kombogiordas D, Anwar A, Price RF. Spontaneous extra-axial intracranial hemorrhage followed by thrombosis incongenital afibrinogenemia: perioperative management of this rarecondition. Surg Neurol 2009;71:689-92.

3. Asiyan KY, Yaman Y, Isguder R, et al. Spontaneous epiduraland subdural hematoma in a child with afibrinogenemia andpostoperative management. Blood Coagul Fibrinolysis 2014;25:398-400.

4. Lak M, Keihani M, Elahi F, Peyvandi F, Mannucci PM. Bleedingand thrombosis in 55 patients with afibrinogenaemia. Br JHaematol 1999;107:204-6.

5. Peyvandi F, Haertel S, Knaub S, Mannucci PM. Incidence ofbleeding symptoms in 100 patients with inherited afibrinogenemiaor hypofibrinogenemia. J Thromb Haemost 2011;9:1687-92.

6. Ishige N, Sunami K, Sato A, Watanabe O, Sakurada M, Ishihara K.A case of systemic lupus erythematosus associated with spontaneousbilateral epidural hematomas. No Shinkei Geka 1985;13:345-9.

7. Kelly DL Jr, Smith JM. Epidural hematoma secondary to frontalsinusitis. J Neurosurg 1968;28:67-9.

8. Henselmans JM, Meijer K, Haaxma R, Hew J, van Der Meer J.Recurrent spontaneous intracerebral hemorrhage in a congenitalafibrinogenemic patient: diagnostic pitfalls and therapeuticoptions. Stroke 1999;30:2479-82.

9. Bornikova L, Peyvandi F, Allen G, Bernstein J, Manco-JohnsonMJ. Fibrinogen replacement therapy for congenital fibrinogendeficiency. J Thromb Haemost 2011;9:1687-704.

10. Tziomalos K, Vakalopoulou S, Perifanis V, Garipidou V.Treatment of congenital fibrinogen deficiency: overview andrecent findings. Vasc Health Risk Manag 2009;5:843-8.


Cervical Thymic Cyst: Case Report

L VU KOVI , S RADOVI , N SORAT, A FILIPOVI

Abstract Introduction: Ectopia of thymic tissue in the neck is well known, while the ectopic thymic cysts areextremely rare. Literature enlists frequency of cervical thymic cysts from 0.3% to 1% of all congenitaltumefactions in the neck area. They are usually diagnosed in early childhood. Case report: A nine-year-old girl was hospitalised at Department of Child Surgery, because of the painless cystic tumour at the leftside of the neck. After the performed diagnostic procedures, a surgical exploration was done. The cystictumour was dissected and sent to histopathological verification. During the histological analysis of thecyst wall, in the part that was 6 mm thick, thymic tissue was detected. Conclusion: Cervical thymic cystsmust be considered in the differential diagnosis of cervical cystic formations, especially in child age andpublications of case reports are important in order of caution when cervical cysts are examined.

Key words Cyst; Neck; Thymus

Department of Pathology, Clinical Centre of Montenegro,Faculty of Medicine, University of Montenegro, Podgorica,Montenegro

L VU KOVI MD, Assistant Professor

Institute for Children's Diseases, Clinical Centre ofMontenegro, Podgorica, MontenegroS RADOVI MDN SORAT MD

Department of Endocrine Surgery, Clinical Centre ofMontenegro, Faculty of Medicine, University of Montenegro,Podgorica, MontenegroA FILIPOVI MD, Assistant Professor

Correspondence to: Dr L VU KOVI


Received November 28, 2017

Case Report

Introduction

Pharyngeal system anomalies are common congenitaldisorders. Anomalies of the second pharyngeal archrepresent around 95% of all anomalies created due topharyngeal system development disorder.1

Thymus is developed from the third pharyngeal arch andduring the embryonic development thymus descends fromthe neck towards the anterior upper mediastinum. Theembryonic development of the thymus explains thepossibility of thymic tissue residue in the neck, most oftendiagnosed in childhood.2

Ectopia of thymic tissue in the neck is well known, whilethe ectopic thymic cysts in the neck are extremely rare.3

Literature enlists frequency of cervical thymic cysts from0.3% to 1% of all congenital tumefactions in the neck area.They are usually diagnosed in early childhood, morecommonly in boys. Considering the fact that these are rarelesions, the right diagnosis is not often set preoperatively.Surgical excision and histological analysis usually lead todiagnosis.4,5

In the paper we present a case of a nine-year-old girlwith clinical, radiological and pathohistologicalcharacteristics of a cervical thymic cyst.

Case Report

A nine-year-old girl was hospitalised at the Departmentof Child Surgery, due to diagnostics and treatment of

Cervical Thymic Cyst159

painless cystic tumour at the left side of the neck. The cystictumour was noticed by the parents 4 weeks prior addressingto a doctor, and since then they do not cite significant sizechanges. They deny trauma as an etiological factor, loss ofweight and fever. The cystic tumour was clearly visiblewhen the head of the patient was raised. The cystic tumourwas positioned in the middle third of the neck, in front ofanterior edge of the sternocleidomastoid muscle and it waselastic, painless, and mobile, did not fix the skin and thesurrounding neck structures, and measured 5x3 cm.

The blood test results, as well as the biochemicalanalysis, were within the reference values.

Magnetic resonance showed that in the soft tissues onthe left side of the neck, in the level from the third to theseventh cervical vertebral body, there was an oval, septatedinhomogenic cystic tumour, well circumscripted from thesurrounding neck structures, up to 6 cm in diameter, filledwith liquid content, with suppression of the thyroid glandleft lobe to the front and trachea to the right.

Scintigraphy of the thyroid gland was normal.After the performed diagnostic procedures, a surgical

exploration was done under general anesthesia. The patientwas placed into spinal position with head rotated to theright. An incision was made on the skin, along the frontedge of the left sternocleidomastoid muscle. Subcutaneoustissue has been dissected, as well as the platysma andsuperficial fascia of the neck. Medially from the left externaljugular vein, along the front edge of the sternocleidomastoidmuscle, central parts of the neck have been approached. Acystic tumour mass was detected, with the wider poleextending from the middle part of the cricothyroid cartilage.The tumour was dissected, separated from the musculusomohioideus and neck blood vessels (the left internal carotidartery and the left internal jugular vein). With a further tissuepreparation between the esophagus and the trachea, the leftrecurrent laryngeal nerve was identified, which crosses theinferior left thyroid artery on the front and enters the larynxin the area of the cricothyroid cartilage, thereby preservingits integrity. Then the cyst was mobilised from the leftsuperior thyroid artery which crossed its narrow part. Afterthe separation of left superior thyroid artery, the origin ofcystic tumour in the level of the pre-vertebral fascia of thethird cervical vertebra was dissected.

The cystic tumour was sent to histopathologicalverification (Figure 1).

The pos topera t ive recovery passed wi thoutcomplications.

Using a macroscopic examination, a cystic tumour wasmultilocular, measured 6x3.7x1.5 cm, with smooth externalsurface. Septa were widening from 0.2 to 0.6 cm with smoothinner surface, with lumen filled by yellowish liquid content.

During the histological analysis of the cyst wall, in thepart that was 0.6 cm thick, thymic tissue has been detected,the remaining cyst wall as well as septa, were made frommoderately cellular connective tissue, infiltrated withlymphocytes, while the interior surface of all parts wascoated with flattened regular squamous epithelium (Figure2).

After the cervical thymic cyst was diagnosed, thymictissue has not been detected by the magnetic resonance ofthe thorax.

Figure 1 Resection of neck thymic cyst.

Figure 2 Thymic cyst, microscopic slide, haematoxylin eosin,x10.

Vu kovi et al 160

Discussion

Pharyngeal system is created in the fourth week of theintrauterine development.1

Thymus is made from the ventrolateral surface of thethird pharyngeal arch, in the sixth week of the intrauterinedevelopment. The thymus tissue residue, that may createcysts, can be found from the mastoid region to the anteriorupper mediastinum.5

Thymic cysts are rare, especially with cervicallocalisation. They can be congenital and acquired; also theycan be unilocular or multilocular. Unilocular cysts arecongenital, they have a thin wall, with thymus tissue onthe periphery. Multilocular cysts are acquired; their wall ismade from abundant connective tissue and coated withmultilayered squamous epithelium, single-layered cubical,cylindrical or multilayered cylindrical epithelium. Thethymus tissue residue can be found in the cyst wall, hencefacilitate setting the right diagnosis. The congenital are rarerthan the acquired. The surgical excision is preferredtreatment.3,6

There are numerous pathological changes that can beclinically manifested by cervical tumefaction. In paediatricage, congenital anomalies are the most frequent cause. Indifferential diagnosis, cervical bronchogenic cysts,branchial arches residue, lymphatic malformations,thyroglossal ducts, cystic teratoma, cysts of the parathyroidand thyroid gland and the thymus anomalies are considered.7

Clinically, the cysts are commonly asymptomatic, only6-10% patients have symptoms, such as dysphagia,dyspnoea, stridor, pain, vocal paralysis caused bycompression to the recurrent laryngeal nerve andhoarseness. Thymic cysts in the neck most often appear inthe first decade of life, more commonly in boys, in the lefthalf of the neck, placed deeply behind or in front ofsternocleidomastoid muscle.3,8,9

In about half of cases, the cyst is in contact withmediastinal thymus.10

During respiratory infections, they can be enlarged.Computerised tomography and/or magnetic resonance ofthe neck and thorax, can help in setting the cyst diagnosis,its connection with mediastinum, as well as thymus existencein younger patients of paediatric age, to avoidimmunological dysfunction. Surgical excision could bemethod of choice for treatment, and the final diagnosis isset via pathohistological examination.8

It is important to note that surgical treatment of thethymic cyst in the neck, without adequate radiologicalexamination of the region of the neck and chest, in the adult

age will not lead to disorders in the function of the immunesystem, but in childhood it can lead to immunodeficiency.

Surgical excision of the cervical thymic cyst should beperformed with great precaution due to close anatomicalrelationships with neurovascular elements in the neck suchas: the vagus nerve, the internal jugular vein, carotid artery,as well as the phrenic, hypoglossal and recurrent laryngealnerves.2,5,7

In the given case, the thymus cyst in the neck has beendiagnosed in a girl, nine years old, without connection withmediastinum, while in the subsequent examination thymushas not been found in the mediastinum.

Cervical thymic cysts must be considered in thedifferential diagnosis of cervical cystic formations,especially in child age. Publication of the case report isimportant in order of caution when cervical thymic cystsare examined.

Conflict of Interests

The authors declare that there is no conflict of interests.

References

1. Nayan S, MacLean J, Sommer D. Thymic cyst: a fourth branchialcleft anomaly. The Laryngoscope 2010;120:100-2.

2. Kaufman MR, Smith S, Rothschild MA, Som P. Thymopharyngealduct cyst: an unusual variant of cervical thymic anomalies. ArchOtolaryngol Head Neck Surg 2001;127:1357-60.

3. Chaudhari J, Fernandez G, Naik L, Pirosha A. Intrathyroidalmultiloculated proliferating thymic cyst. Endocr Pathol 2015;26:45-7.

4. Betti M, Hoseini NH, Martin A, Buccoliero A, Messineo A,Ghionzoli M. Cervical thymic cyst in childhood: a case report.Fetal Pediatr Pathol 2015;34:65-9.

5. Michalopoulos N, Papavramidis TS, Karayannopoulou G, et al.Cervical thymic cysts in adults. Thyroid 2011;21:987-92.

6. Suster S, Moran CA. Mediastinum. In: Weidner N, Cote RJ, SusterS, Weiss LM, editors. Modern surgical pathology. 2nd ed.Philadelphia: Saunders Elsevier, 2009:454-516.

7. Terzakis G, Louverdis D, Vlachou S, Anastasopoulos G,Dokianakis G, Tsikou-Papafragou A. Ectopic thymic cyst in theneck. J Laryngol Otol 2000;114:318-20.

8. Mahmodlou R, Gheibi S, Nargesi AA, Mahmoodzadeh R, SalabatiM. Symptomatic cervical thymic cyst: a case report and literaturereview. J Compr Ped 2013;4:147-50.

9. Millman B, Pransky S, Castillo J, Zipfel TE, Wood WE. Cervicalthymic anomalies. Int J Pediatr Otorhinolaryngol 1999;47:29-39.

10. Saggese D, Ceroni Compadretti G, Cartaroni C. Cervical ectopicthymus: a case report and review of the literature. Int J PediatrOtorhinolaryngol 2002;66:77-80.


The Role of Clinical Neuropsychology in the PaediatricEpilepsy Management

TCK CHEUNG

Department of Psychology, University of TorontoScarborough, Ontario, Canada

TCK CHEUNG MSSc Clin Psy(CUHK), MSc Cog Neuropsy(Lond), PG Dip Clin Paed Neuropsy(Lond)

Correspondence to: Mr TCK CheungEmail: [email protected]

Received May 28, 2018

Commentary

Epilepsy is the most common cause of chronicneurological conditions with typical onset around early tomid-childhood.1,2 It is a chronic neurological disorder whichcould impact multiple aspects of a patient's life. Epilepsyis not only medical,1 but also psychological, cognitive andsocial.3

As reported by Professor Paul Salkovskis in his externalconsultancy report in reviewing the Clinical PsychologyService in Hospital Authority, there was not a single childneuropsychologist in the force of over 150 clinicalpsychologists,4 nor there has been any program driven orprotocol based neuropsychological services among publichospitals. Such disparity of service highlights theimpor tance of the deve lopment of paedia t r icneuropsychology.

Neuropsychological and PsychosocialSequalae of Epilepsy

Neuropsychology addresses the clinical fact thatpaediatric patients with epilepsy have long-term cognitiveimpairment mediated by an atypical brain development. Ina paediatric sample with temporal lobe epilepsy, intellectualdysfunction (defined as IQ <79) could be as high as 57%,with possibility of decline over time.5 Cognitive

dysfunctions such as attention, language, visuo-perceptualability, memory and executive functions were alsodemonstrated,6 and could be unfolded as young as frompreschool.7

Neuropsychology addresses the side effect ofantiepileptic drugs (AED) and surgical procedures onneuropsychological functioning. In a randomised controlledclinical trial that using valproic acid is associated with moresignificant attentional dysfunction than ethosuximide orlamotrigine in children with newly diagnosed childhoodabsence epilepsy, even in the cases where seizure freedomwas attained. More importantly, those deficits wentunnoticed by parents.8

In a prospective long study with average follow-upperiod of 37 years, it was showed that about one-third ofpatients with epilepsy did not remit, with 14% having onand off remission and relapses, and 19% no remission atall.9 For those patients with intractable epilepsy,neurosurgery would be considered with goals of makingthe targeted patient seizure-free or reducing the frequencyand severity of the impeding seizures. Gleissiner andcolleagues found that paediatric patients had differentrecovery trajectory comparing to adult controls.10

Controlling pathology, onset of epilepsy, side and type ofsurgery, they reported that although both children and theadult control groups showed deterioration in memory at threemonths post-operation, children showed improvement backto their pre-operation level in terms of verbal memory, andremissions in visual memory as well as attentional function.Those findings on functional recovery, however, were notshown in the adult group. Therefore, planned multiple time-point assessment would be particularly important forchildren undergoing epilepsy surgeries.

In short, it is clear that pre- and post-treatmentneuropsychological assessment is essential. It providesaccessory information about the localisation andlateralisation of the epilepsy-specific cognitive impairment,

Cheung 162

quality and outcome control, as well as estimation of thepostoperative cognitive development.11

Neuropsychology also addresses the impact of epilepsyon children's psychological, social, and academicdevelopment. The neuropsychological deficits could leadto psychiatric and behavioural problems. Namely, parentsof children with idiopathic epilepsy reported morehyperactivity, emotional and conduct problems, and lessprosocial behaviours.12 Turky and colleagues13 found thatfrom parent's reports nearly half (47.9%) of the sampledchildren and adolescents indicated psychiatric criteria andjust over 60% with chronic distress and social impairment.Thome-Souza and colleagues14 reported that depressionfound in up to 36.4% and ADHD 29.1% in a children andadolescent sample. Lower relationship quality betweenparents and child, higher maternal depression, and morefamily problems were reported.15

Recommendations by Clinical Guidelines

The role of paediatric neuropsychology is substantiatedby NICE Guidelines16 (see also The Hong Kong EpilepsyGuideline17). Evaluating the learning disabilities andcognitive dysfunction, especially in language and memory,is recommended. Referral indicators were also clearlystated, including when (i) educational difficulties arise,(ii) MRI finding reveals abnormalities in cognitivelyimportant brain regions, (iii) the patients complain ofmemory or other cognitive deficits and/or decline, and (iv)as part of presurgical evaluation (in Hong Kong Guidelineonly). When there is suspected non-epileptic attack disorderor when seizures are not controlled and/or there is diagnosticuncertainty or treatment failure, referral could also be madefor further investigation and treatment, alongside with soonreferral to tertiary service.

The International League Against Epilepsy PsychologyTask Force recommended routine psychological screeningshould be implemented. Potential side-effect of AED shouldbe evaluated in a standardised way. For children inparticular, multi-informant screenings should be used.When significant symptoms are identified, a formal mentalhealth assessment should be conducted to inform theselection of treatment elements.18

In terms of interventions for seizure control, when thepatients or the specialists consider medical treatment to beinadequate, psychological interventions such as relaxation,cognitive behavioural therapy (CBT), biofeedback couldbe used with aims of improving quality of life in some

patients, but not for reduction in seizure frequency.Relaxation and CBT was also recommended for childrenand young people with drug-resistant focal epilepsy fordepressive and anxiety symptoms.16 Mindfulness trainingon awareness and acceptance of seizure-related symptomsand seizure control, albeit its current weak evidence, wasemerging as an alternative intervention.18

When there are comorbid mood and adjustmentdisorders, or it is found that the mood conditions such asdepression and anxiety disorders are affecting the casemanagement such as drug compliance, psychologicalinterventions could help alleviating the mood symptomsand enhance health promoting behaviours.19

The Future of Paediatric Neuropsychology inHong Kong

The complexity of the neurocognitive and psychosocialdevelopment of children with epilepsy requires a highlyskilled sub-specialty in applied psychology, to join forcewith the multidisciplinary teams in Hospital Authority, forthe betterment of the holistic care of the patients. The articlehas laid out the 'why' of clinical neuropsychology, but whyhas it not been developed in public hospitals at present?

The under-development of the sub-specialty may be dueto institutional factors. At present, paediatric patients withepilepsy in public hospitals are often referred out to theChild Assessment Service (CAS) of the Department ofHealth for preoperative assessments,1,3 but not withinHospital Authority, with its own historical reasons. Thepros are that at least the child is being assessed and initialrecommendation could be formulated. Clinical data couldbe drawn to allow systematic audit for the outcome of thesurgery by the surgical teams. The cons, however, is thatlong term follow up by CAS psychologists on remediationand training are limited by the service positioning. Inaddition, the assessment result in CAS may not be accessibleto other rehabilitation specialists such as paediatricians,nurses, speech therapists, occupational therapists andphysiotherapists who work in the local hospital the childrenreceive step-down care. Psychologists in the local hospitals,likely constrained by the limited resources for the ever-growing service demand, are likely to be reserved ininitiating uncovered new service. The second reason is thatthere is no provision of specialty training in paediatricneuropsychology in Hong Kong, possibly due to thechicken-and-egg problem of zero opening for such sub-specialty.

Neuropsychology in Paediatric Epilepsy Management163

The establishment of Hong Kong's first Children Hospitalcould be a driving force for the development of thesub-specialty. Management may formulate the new servicemodels and clinical protocols, including but not limitedto early screening,20 with active involvement ofneuropsychologists (virtually child clinical psychologists)for the best interest of paediatric patients. Due to the lack oflocal training in such sub-specialty, sponsoring overseastraining in paediatric neuropsychology4 should be seriouslyconsidered in long-term planning.

Acknowledgement

The author expresses deep gratitude to the Li Po ChunCharitable Trust Fund Overseas Postgraduate Study &Professional Training Scholarship, for supporting the PGDip specialist course offered by the UCL Great OrmondStreet Institute of Child Health, University CollegeLondon, UK; and to the former colleagues of theDepartment of Clinical Psychology, United ChristianHospital, Hong Kong S.A.R.

Declaration of Interests

None

References

1. Chak WK. What is the consequence of children with medicallyintractable epilepsy? What are the treatment options? HK MedDiary 2009;14:5-7.

2. Berg AT, Berkovic SF, Brodie MJ, et al. Revised terminologyand concepts for organization of seizures and epilepsies: Reportof the ILAE Commission on Classification and Terminology,2005-2009. Epilepsia 2010;51:676-85.

3. Tsang L. Neuropsychological assessment in paediatric epilepsy.H K Med Dairy 2012;17:14-6.

4. Salkovskis P. Evaluation of Future Directions in ClinicalPsychology in Hong Kong: External Review of Hong Kong

Hospital Authority Clinical Psychology Professional Provision.Hong Kong: Hospital Authority, 2011.

5. Cormack F, Cross JH, Isaacs E, et al. The development ofintellectual disabilities in paediatric temporal lobe epilepsy.Epilepsia 2007;48:201-4.

6. Rzezak P, Valente KD, Duchowny MS. Temporal lobe epilepsyin children: Executive and mnestic impairments. Epilepsy Behav2014;31:117-22.

7. Danielsson J, Petermann F. Cognitive deficits in children withbenign rolandic epilepsy of childhood or rolandic discharges: astudy of children between 4 and 7 years of age with and withoutseizures compared with healthy controls. Epilepsy Behav 2009;16:646-51.

8. Masur D, Shinnar S, Cnaan A, et al. Pretreatment cognitive deficitsand treatment effects on attention in childhood absence epilepsy. Neurology 2013;81:1572-80.

9. Sillanpaa M, Schmidt D. Natural history of treated childhood-onset epilepsy: prospective, long term population based study.Brain 2006;129:617-24.

10. Gleissner U, Sassen R, Schramm J, Elger CE, Helmstaedter C.Greater functional recovery after temporal lobe surgery in children.Brain 2005;128:2822-9.

11. Helmstaedter C. Neuropsychological aspects of epilepsy surgery.Epilepsy Behav 2004;5:45-55.

12. Dafoulis V, Kalyva E. Factors associated with behavioralproblems in children with idiopathic epilepsy. Epilepsy Res 2012;100:104-12.

13. Turky A, Beavis JM, Thapar AK, Kerr MP. Psychopathology inchildren and adolescents with epilepsy: An investigation ofpredictive variables. Epilepsy Behav 2008;12:136-44.

14. Thome-Souza S, Kuczynski E, Assumpção F, et al. Which factorsmay play a pivotal role on determining the type of psychiatricdisorder in children and adolescents with epilepsy? EpilepsyBehav 2004;5:988-94.

15. Rodenburg R, Meijer AM, Dekovic M, Aldenkamp AP. Familyfactors and psychopathology in children with epilepsy: a literaturereview. Epilepsy Behav 2005;6:488-503.

16. Epilepsies: Diagnosis and Management. UK: NICE, 2012.17. Hong Kong Epilepsy Society. The Hong Kong Epilepsy Guideline

2009. HK Med J 2009;15 Suppl 5:6-28.18. Michaelis R, Tang V, Goldstein LH, et al. Psychological treatments

for adults and children with epilepsy: Evidence-basedrecommendations by the International League Against EpilepsyPsychology Task Force. Epilepsia 2018;59:1282-302.

19. Jackson CF, Makin SM, Baker GA. Neuropsychologicaland psychological interventions for people with newly diagnosedepilepsy. Cochrane Database Syst Rev 2015:CD011311.

20. Wagner J, Smith G. Psychological services in a pediatric epilepsyclinic: Referral patterns and feasibility. Epilepsy Behav 2007;10:129-33.

HK J Paediatr (new series) 2019;24:164

What is the Diagnosis?

JLF FUNG, K RETHANAVELU, BHY CHUNG

The clinical quiz was prepared by:JLF FUNG*K RETHANAVELU*BHY CHUNG

Department of Paediatrics & Adolescent Medicine, The Universityof Hong Kong, Queen Mary Hospital, Pokfulam, Hong Kong

*These authors contributed equally to the work.

Case History

This case was initially referred to the genetics clinic bythe obstetrician at 20 weeks of gestation in view ofabnormal ultrasound scan. Increased nuchal translucency(9 mm) and cystic hygroma was noted in 12 weeks and 16weeks scan. Scan at 20 weeks showed cystic hygroma,subcutaneous edema, short long bones, small stomach,clenched hands and some narrowing of pulmonary artery.There is a high risk in Down syndrome screening of 1:2,subsequent chorionic villus sampling showed normalkaryotype and aCGH. After discussion with the obstetricianand geneticist, the couple decided to keep the pregnancywithout further genetic testing for RASopathies.Ultrasound at 35 weeks showed polyhydramnios,rhizomelic shortening, clenched hands, flat forehead, lowset ears and suspected hypospadias.

The baby was delivered using low forceps at 37+4 weekswith BW 3480 gram (50%), BL 49 cm (10%), HC 34 (10-25%). No resuscitation is required. The baby wasdysmorphic with the following features: frontal bossing,tall forehead, sparse eyebrows, anteverted nares, largemouth with thin upper lips, small palpebral fissure withptosis low set and dysplastic ears. Other clinical findingsinclude short proximal limbs, camptodactyly, submucosalcleft palate, grade I subglottic stenosis, ectopic anus,micropenis, failed BAEP bilaterally, and suspected GERD.EDTA blood was sent out for whole exome sequencing andthe report came back negative.

Clinical photos of patient at birth, 6 months old and 2years old are shown in Figure 1.

Answer to "Clinical Quiz" on Pages 172-174N.B. The Editors invite contributions of illustrative clinical casesor materials to this section of the journal.

Clinical Quiz

Figure 1 Clinical photos of the patient. (Consent for publication hasbeen obtained)


Abstracts of Articles in Chinese

ABO Rh

K Çelik, S Aydin Koker, MT Özkul, Ö Olukman, TH Karapinar, RC Vergin, Çalkavur. Impact of Minor Blood Group

Incompatibility Versus ABO and Rh Blood Group Incompatibility in Newborns with Indirect Hyperbilirubinaemia:

A Single-Centre Clinical Experience. HK J Paediatr (new series) 2019;24:120-126

HDFN G IgG

HDFN Rh Rhesus A B AB O Kell

Duffy MNS P Diego ABO Rh

355 UHB C c E e

Kell D d A B O 4 1 ABO 2 Rh 3 ABO+Rh

4 355 1 230 64.7% 2 68

19.1% 15 4.2% 3 42 12% 4 12

28% 11 26% 9 21% 6 (14%) 4 (9%) "C" "c" "E+c" "E" "Kell"

4 (8±7.5 p=0.015

p=0.007, p=0.041 2 4 p=0.025 4

ABO Rh

NICUKK Chik, WK To, K Luk, B Lam, CC Shek, C Tse, D Ng. Escherichia coli Meningitis in Neonatal Intensive Care

Units: A Five Years Study. HK J Paediatr (new series) 2019;24:127-134

NICU

2011.1.1-2016.5.31 2 NICU

11 55% 91%

27% ESBL

ESBL-

166

WH Liao, HY Xu, X Zhou, JL Tian, XY Fang, J Ding, C Zeng, HL Wu. Non-nutritive Sucking and Nesting Greatly

Reduces Pain during Retinopathy Screening in Premature Infants. HK J Paediatr (new series) 2019;24:135-139

120

p<0.05

G Kim, N Lee, H Ko, JH Byun, HD Lee, SC Sung, H Kim, KH Choi. Clinical Manifestations and Outcomes of Cardiac

Tumours in Children. HK J Paediatr (new series) 2019;24:140-146

2009 2017 27

15

7 5

167

Refsum

KL Cheung, CH Ko, HHC Lee, CM Mak. Chinese Boy with Normal Initial Peroxisomal Blood Assays: A Diagnostic

Pitfall in the Workup for Infantile Refsum Disease. HK J Paediatr (new series) 2019;24:147-150

Refsum

C26 C22 MRI

C26 C22

Refsum

Refsum

PT Yu, HM Luk, IFM Lo. What Can It Be If Not a Simple Haemangioma? HK J Paediatr (new series) 2019;24:151-

154

1-2% 30% 6

CM-AVM

AVMs AFVs

1:10

CM-AVM AVM

RASA1 CM-AVM

CM-AVM RASA1

168

U Ocak, P Eser Ocak, B Gungorer. Spontaneous Acute Epidural Haematoma in a Paediatric Patient with Congenital

Afibrinogenaemia. HK J Paediatr (new series) 2019;24:155-157

CT

L Vu kovi , S Radovi , N Sorat, A Filipovi . Cervical Thymic Cyst: Case Report. HK J Paediatr (new series) 2019;

24:158-160

0.3% ~ 1% 9

6mm


MCQs

Instruction:1. Please use pencil to shade the box for the best and correct answer (only one answer for each question).2. Send back the answer sheet (see loose leaf page) to the Hong Kong College of Paediatricians. One point will be awarded

to each article if ≥3 of the 5 answers are correct. The total score of the 4 articles will be 4 CME points.

(A) Impact of Minor Blood Group IncompatibilityVe r s u s A B O an d R h B l o o d G r o u pIncompatibility in Newborns With IndirectHyperbilirubinaemia: A Single-Centre ClinicalExperience

1. What could severe early jaundice associated withisoimmunisation cause in newborn?a. Erythrocyte enzymatic defectb. Fetal-maternal blood group incompatibilityc. Cephalohematomad. Disorder of conjugatione. Polycythemia

2. When do we need to investigate for minor blood groupincompatibility?a. Requirement of exchange and/or erythrocyte

transfusionsb. Prolonged jaundicec. Anaemiad. Rebound elevation of serum bilirubine. All of above

3. According to the management of hyperbilirubinemia,how should haemolytic disease be evaluated in patients?a. Total serum bilirubinb. Reticulocytec. Direct Coombs testd. Erythrocyte morphology on peripheral blood smeare. All of above

4. Which of the following statements is the correctconclusion of this manuscript?a. IVIG is the best treatment for neonatal isoimmunisation

with minor blood group incompatibilityb. Significant difference was found in the laboratory

findings between groupsc. The frequency of minor blood group incompatibility

was 12%d. The rate of intravenous fluid needed was lower in

groups with minor blood group incompatibility andABO+Rh blood group incompatibility when comparedwith other groups.

e. The rate of exchange transfusion requirement was 5%for all groups

5. Which of the following types of minor blood groupincompatibility has the highest frequency?a. Cb. cc. E+cd. Ee. Kell

170

(B) Escherichia coli Meningitis in NeonatalIntensive Care Units: A Five Years Study

1. Which of the following statement is truea. E. coli is always pathogenicb. K1 capsular antigen is present in all E. colic. The amount of K1 antigen in the CSF is linked to the

disease severity in the neonatesd. Only 5% of the E. coli in maternal vagina can express

K1 antigene. E. coli strain with K1 capsular antigen is always more

susceptible to antimicrobial and never has the ESBLenzyme

2. Which of the following is false?a. More than 30% of all community-acquired E. coli

infections were ESBL producing in United Statesb. More than 25% of all community-acquired E. coli

infections were ESBL producing in Hong Kong in2015

c. More than 50% of the local live pigs had faecalcarriage of ESBL-producing E. coli in Hong Kong

d. The use of antibiotics in livestock farming has norelation with the ESBL E. coli infection in thecommunity

e. Global prevalence of infection caused by ESBLproducers is increasing

3. Concerning the neonatal infection: Hong Kong startedto have Universal Group B Streptococcus screeningamong pregnant ladies sincea. 1992b. 1996c. 2002d. 2005e. 2012

4. Which of the following is true?a. Hong Kong has a registry on the neonatal E. coli

infection since 2012b. The estimated prevalence of neonatal E. coli meningitis

in this study was 0.02%c. This study showed less than 20% of the E. coli

meningitis were ESBL producingd. This study showed a statistical significant relationship

between IAP and neonatal ESBL E. coli meningitise. If neonates has microbiological evidence of Gram

negative bacterial meningitis, antibiotic should benarrowed down to gentamicin only

5. In this study, which of the following is truea. E. coli meningitis in neonates always has abnormal CSF

glucose and proteinb. In this study, all neonates with neonatal meningitis

had elevated CRPc. In those neonates suffered from ESBL positive E. coli

infection, the symptoms onset days were later thanthe ESBL negative E. coli infection group in this study

d. None of the mother had ESBL E. coli in the placentalswab in this study

e. Female babies were predominant in this study withstatistical significance

(C) Non-nutritive Sucking and Nesting GreatlyReduces Pain During Retinopathy Screening inPremature Infants

1. In which group of premature infants is incidence of ROPhigh?a. Birth weight <2000gb. Gestational age <35 weeksc. Birth weight >2500gd. Gestational age<34 weekse. A and B

2. Non-pharmacological interventions to alleviate pain inpreterm infants do not include:a. Oral sucroseb. Oral glucosec. Breast feedingd. Keep warme. NNS

3. What items are not included in the PIPP score?a. Behavioural statusb. Blood pressurec. Heart rated. Oxygen saturatione. Gestational age

4. What is the incidence of ROP in China amongst highrisk premature newborns?a. 10.0%b. 15.8%c. 17.8%d. 20%e. 25%

5. Which value was considered to be indicative of significantpain with PIPP evaluation?a. <7b. 7 to 12c. >12d. >15e. None of the above

171

(D) Clinical Manifestations and Outcomes ofCardiac Tumours in Children

1. Which of the following is the most common cardiactumour in children?a. Angiosarcomab. Fibromac. Rhabdomyosarcomad. Rhabdomyomae. Myxoma

2. Which of the following is the most probable cardiactumour that could be associated with tuberous sclerosis?a. Fibromab. Angiosarcomac. Rhabdomyomad. Rhabdomyosarcomae. Fibrosarcoma

3. Which of the following is the most probable drug to beuseful in cardiac tumours associated with tuberoussclerosis?a. Tacrolimusb. Everolimusc. Infliximabd. Imatinibe. Estrogen

4. Which of the following is the most common indicationfor surgery in rhabdomyoma in this study?a. Arrhythmiab. Left ventricular outflow tract obstructionc. Huge pericardial effusiond. Embolic phenomenone. Sepsis

5. Which of the following is the most common feature atdiagnosis in this study?a. Cardiac murmurb. Chest painc. Seizured. Fetal diagnosise. Incidental diagnosis

Answers of April issue 2019

(A) 1. c; 2. a; 3. d; 4. d; 5. c

(B) 1. d; 2. e; 3. e; 4. a; 5. a

(C) 1. c; 2. a; 3. b; 4. c; 5. e

(D) 1. b; 2. e; 3. c; 4. b; 5. e


CLINICAL QUIZ (p164) ANSWER

What is the diagnosis?

Our proband suffers from Pallister-Killian syndrome (PKS). PKS is characterised by craniofacial dysmorphism,hypotonia in infancy and early childhood, developmental delay, intellectual disability and other birth defects.

Patients with PKS have distinctive facial features described as "coarse". It includes features such as prominent ortall forehead, sparse eyebrows/lashes, hypertelorism, epicanthal folds, broad nasal bridge, short nose, antevertednares, large mouth with thin upper lip and/or thick lower lip, high arched or cleft palate, micrognathia, low-set and/or posteriorly rotated ears and short neck. Most of them have sparse hair and areas of alopecia, particularly in thetemples, which may be filled up as the child grows up.

Severity of the disease and additional features are highly variable. Some patients are noted to have vision and/orhearing loss, seizures, congenital diaphragmatic hernia, polydactyly, genital abnormalities, and heart defects. Hyper/Hypo-pigmentation of the skin and accessory nipples can also be observed.

What is the molecular genetics behind PKS?

PKS is caused by mosaic tetrasomy 12p, extra choromosomal materials are detected. Four copies of the 'p' arms ofchromosome 12 are present instead of two, therefore it is described as "tetrasomy 12p". Usually, the two extra 12pform a mirror image known as isochromosome (as shown in Figure 2). Also, this chromosomal abnormality onlypresents in some of the cells but not every cell of the patient thus it is referred as ‘mosaic’.

Tetrasomy 12p is associated with tissue-limited mosaicism, meaning that the chromosomal changes could only bedetected in specific tissues. Often, children with PKS appeared to have normal chromosomes in lymphocytes, onlyby examining buccal mucosa or skin cells would be able to make a diagnosis.1,2

Figure 2 Photo taken from "Pallister-Killian syndrome" (Rare Chromosome Disorder Support Group,2016).3

173 Clincal Quiz Answer

Very interestingly, for our proband, the chromosomal imbalance was identified in the blood sample of but not inthe CVS. Also, this is a very special case that the initial diagnosis was not made using conventional cytogeneticmethod. In fact, the extra chromosome 12 materials were first identified in a retrospective analysis of the exomesequencing data (Figure 3); which is not the perfect/usual tool we use for detecting copy number variants/chromosomalchanges. The finding was confirmed after repeated aCGH. This highlights the diagnostic journey of PKS can becomplicated.

Figure 3 (a) Gain of genetic material is observed in chromosome 12 (indicated by the arrow). (b) Thegain of genetic material is localised in p arm of chromosome 12, and the intensity is twice as the reference,indicating a tetrasomy 12p.

(a)

(b)

174

What is the management?

The management and treatment of a child with PKS depends on the severity and the symptoms presented. Itrequires a multidisciplinary team consisting of developmental paediatrician, rehabilitation team, geneticist,ophthalmologist, otolaryngologists, neurologist, and orthopaedic surgeon. These children may need resuscitation atbirth and special care for the management of hypotonia. There is an increase in the incidence of diaphragmatichernias and fetal hydrops in PKS. These children may benefit from early intervention program and special education.Regular screening for visual and/or hearing problems and aggressive management of seizure, contractures and scoliosisare required.

What is the risk of recurrence?

All PKS cases occur sporadically. The extra isochromosome 12p typically results from nondisjunction of thereproductive cells. During early development of the embryo, some cells lose the isochromosome while others retainit, leading to mosaicism of tetrasomy 12p.4 Parents are very unlikely to be the carriers; hence the recurrent risk islow.

Conclusion/Learning point

In view of the unique genetic aetiology, the genetic diagnosis of PKS can sometimes be difficult. Since theindividual has tissues with normal chromosomal material and abnormal chromosomal material at different distribution,sampling can be challenging and false negative test result may lead to a further delay in genetic diagnosis. Recognisingfeatures of PKS becomes very important and special diagnostic tests may be required to end the diagnostic odyssey.A diagnosis of PKS should be considered in a child with coarse facies, profound intellectual disability and localisedalopecia especially around the temporal region.

Acknowledgments

We would like to express our gratitude to the patient and his family for their contribution. Informed consent wasobtained for publication.

References

1. Peltomäki P, Knuutila S, Ritvanen A, Kaitila I, Chapelle A. Pallister-Killian syndrome: cytogenetic and molecular studies. Clin Genet 1987;31:399-405.

2. Warburton D, Anyane-Yeboa K, Francke U, Reynolds JF. Mosaic tetrasomy 12p: Four new cases, and confirmation of the chromosomalorigin of the supernumerary chromosome in one of the original Pallister-Mosaic syndrome cases. Am J Med Genet 1987;27:275-83.

3. Rare Chromosome Disorder Support Group. Pallister-Killian Syndrome. Rarechromo.org [https://www.rarechromo.org/media/information/Chromosome%2012/Pallister-Killian%20syndrome%20FTNW.pdf]

4. Gajecka M. Unrevealed mosaicism in the next-generation sequencing era. Mol Genet Genomics 2016;291:513-30.


4. SI units should be used or included in parentheses.

Ethical Consideration

For original clinical study, authors must state that theprotocol for the research project has been approved by the EthicsCommittee of the institution within which the work wasundertaken. All investigations on human subjects must includea statement that informed consents have been obtained. Patientanonymity must be preserved. Photographs and video clippingsneed to be prepared to prevent human subjects being recognizedunless prior written permission has been obtained. Whenreporting experiments on animals, authors should indicatewhether the institutional and national guide for the care anduse of laboratory animals was followed.

The manuscript should usually be arranged as follows:Title page

This page should include the full names, and affiliations ofall authors. A short title of no more than 40 characters shouldalso be given. Up to three academic degrees for each author areallowed. If an author’s affiliation has changed since the workwas done, list the new affiliations as well. Limit the number ofauthors to 4 for case reports and clinical quiz.Abstract and Key words

The abstract should be no more than 150 words summarisingthe purpose, methods, findings and conclusions. Authors shouldprovide no more than five key words to assist with cross-indexing of the paper. Key words should be taken from IndexMedicus.IntroductionMethodsResultsDiscussionReferences

Number references in the order they appear in the text.References should follow the Vancouver style and should appearin the text, tables and legends as Arabic numerals in superscript.Journal titles should be abbreviated in accordance with IndexMedicus. List all authors and/or editors up to six; if more thansix, list the first three and "et al".

Examples of References:Articles in Journals1. Standard journal article Greenberg DL, Root RK. Decision

making by analogy. N Engl J Med 1995;332:592-6.2. Organisation as author The Royal Marsden Hospital Bone-

Marrow Transplantation Team. Failure of syngeneic bone-marrow graft without preconditioning in post-hepatitismarrow aplasia. Lancet 1977;2:742-4.

3. No author given Coffee drinking and cancer of the pancreas[editorial]. BMJ 1981;283:628.

4. Issue with supplement Gardos G, Cole JO, Haskell D, MarbyD, Paine SS, Moore P. The natural history of tardive dyskinesia.J Clin Psychopharmacol 1988;8(4 Suppl):31S-37S.

The Hong Kong Journal of Paediatrics (HKJP) is a jointquarterly publication of the Hong Kong College of Paediatricians(HKCPaed) and The Hong Kong Paediatric Society (HKPS).The HKJP publishes original research papers, review articles,case reports, editorials, commentaries, letters to the editor andconference proceedings. Topics of interest will include allsubjects that relate to clinical practice and research in paediatricsand child health.

Manuscripts are accepted on the condition that they aresubmitted solely to the HKJP and have not been publishedelsewhere previously and are not under consideration by anotherjournal. A complete report following presentation or publicationof preliminary findings elsewhere can be considered.

Categories of articles include the following:

Original Articles The text should not usually exceed 5,000words excluding references; the number of tables, figures,or both should normally be not more than six, and referencesnot more than 50.

Review Articles Reviews are usually invited systematic criticalassessments of literature.

Case Reports Length should not exceed 1,500 words; thenumber of tables or figures used should not be more thantwo, and references should not be more than 10. Limit thenumber of authors to 4.

Commentaries Commentary on current topics is welcome.Length should not exceed 1,200 words; no tables or figuresallowed, and references should not be more than 20.

Clinical Quiz The clinical quiz should be educational. It shouldi) include the description of a case in no more than 250 wordsand 3 clinical photos or figures, and ii) provide answers onthe diagnosis, clinical features and findings, and managementof the condition in no more than 1,000 words, 10 references,and 3 photos, figures or tables.

Letters to the Editor Letters discussing a recent article in theHKJP are welcome. Original letters that do not refer to anHKJP article may also be considered. Letters should notexceed 500 words and have no more than five references.Published letters may be edited.

Manuscript Preparation

1. Use Arabic numerals for numbers above nine, for designators(e.g. case 5, day 2, etc.) and for units of measure; numbersshould be spelled out if below 10, at the beginning and endof sentences, and for fractions below one.

2. Manuscripts should be submitted as a Word document inBritish English in the following format: Typed double-spaced, page size 22 cm. x 29 cm. (8 1/2 in. x 11 in.), pagemargins 2.54 cm (1 in), font size 12 pt.

3. Do not use abbreviations in the title or abstract and limittheir use in the text. Standard abbreviations may be used andshould be defined on first mention in the text unless it is astandard unit of measurement.

Instructions to Authors

176

5. Type of article indicated as needed [i] Fuhrman SA, JoinerKA. Binding of the third component of complement C3 byToxoplasma gondii [abstract]. Clin Res 1987;35:475A. [ii]Feldman N. Laparoscopic nephrectomy [letter]. N Engl J Med1991;325:1110.

Books and Other Monographs6. Personal author(s) Colson JH, Armour WJ. Sports injuries

and their treatment. 2nd ed. London: S. Paul, 1986.7. Editor(s), compiler as author Dausset J, Colombani J,

editors. Histocompatibility testing 1972. Copenhagen:Munksgaard, 1973.

8. Chapter in a book Steiner RE, Bydder GM. Clinical nuclearmagnetic resonance imaging. In: Dawaon AM, CompstonND, Besser GM, editors. Recent advances in medicine no.19. Edinburgh: Churchill Livingston, 1984:39-56.

9. Conference paper Harley NH. Comparing radon daughterdosimetric and risk models. In: Gammage RB, Kaye SV,editors. Indoor air and human health. Proceedings of theSeventh Life Sciences Symposium; 1984 Oct 29-31;Knoxville(TN). Chelsea(MI): Lewis, 1985:69-78.

10. Dissertation Cairns R.B Infrared spectroscopic studies ofsolid oxygen [dissertation]. Berkeley, (CA): Univ. ofCalifornia, 1965.

11. Scientific and technical report Akutsu T. Total heartreplacement device. Bethesda (MD): National Institutes ofHealth, National Heart and Lung Institute; 1974 Apr ReportNo: NIH-NHLI-69-2185-4.

Unpublished Material12. In press Lillywhite HD, Donald JA. Pulmonary blood flow

regulation in an aquatic snake. Science. In press.13. A reference to unpublished work which has not yet been

accepted for publication should not appear in the referencelist but should be cited in the text as unpublished data orpersonal communication.

TablesEach table should begin on a separate page. Number tables

consecutively in the order of their first citation in the text andsupply a brief title for each. Give each column a short orabbreviated heading. Place explanatory matter in footnotes,not in the heading. Vertical rules and horizontal rules shouldbe omitted.

Illustrations (Figures)Each illustration must be submitted as a separate figure file.

The file name should be the same as the figure number. Preferredformats for digital artwork submission include EncapsulatedPostScript (EPS), Portable Document Format (PDF), and TaggedImage Format (TIFF). Letters, numbers and symbols should beclear and of sufficient size to retain legibility when reduced.Photographs of persons must be retouched to make the subjectunidentifiable, or be accompanied by written permission fromthe subject to use the photograph.

Number illustrations consecutively according to the orderin which they have been first cited in the text. Titles and detailedexplanations should be confined to legends and not included

in illustrations. The legends should be numbered and providedon a separate double-spaced page.

Supplementary Video ClipsVideo clips can be submitted with your manuscript in MP4

file format with H.264 codec. The size of the video should notexceed 5 MB. Patient anonymity must be preserved unless priorwritten permission has been obtained. If accepted, the video willappear online on the Journal's website, http://www.hkjpaed.org.

Declaration of InterestAll sources of funding for research are to be explicitly stated.

All authors must disclose any financial and personalrelationships with other people or organisations that couldinappropriately influence their work. It is the sole responsibilityof authors to disclose any affiliation with any organisation witha financial interest, direct or indirect, in the subject matter ormaterials discussed in the manuscript that may affect the conductor reporting of the work submitted.

If there are no declarations, authors should explicitly statethat there are none. This must be stated at the point ofsubmission (within the manuscript, after the main text, under asubheading "Declaration of Interest"). Manuscript submissioncannot be completed unless a declaration of interest statement(either stating the disclosures or reporting that there are none)is included. This will be made available to reviewers and willappear in the published article.

The intent of this policy is not to prevent authors with anyparticular relationship or interest from publishing their work,but rather to adopt transparency such that reviewers, editors,the publisher, and most importantly, readers can make objectivejudgments concerning the work product.

CorrespondenceAll manuscripts, correspondence and subscription

should be addressed to Chief Editor, Hong Kong Journalof Paediatrics Limited, c/o Hong Kong College ofPaediatricians, Room 801, HK Academy of Medicine JockeyClub Building, 99 Wong Chuk Hang Road, Aberdeen,Hong Kong, e-mail:[email protected].

DisclaimerThe opinions expressed in the HKJP are those of the authors

and not those of HKCPaed, HKPS or the publisher. The HKJP,HKCPaed and HKPS do not hold any legal responsibility forthe opinions expressed in the articles, unless this is clearlyspecified. The authors should hold legal responsibility for thecontent and opinion expressed. Manuscripts and discs aresubmitted at the sender's risk and the HKJP, HKCPaed andHKPS assume no responsibility for return of material.

OffprintsOffprints can be ordered at a cost at the time of proof-

reading.

DistributionThe HKJP is published quarterly and adopts an open access

policy. Electronic copies of all articles are available online athttp://www.hkjpaed.org. Members of HKCPaed and HKPSwill receive quarterly e-mail notification of the Table ofContents in each of the Journal issues.

New Series - hkjpaed.org

Documents

Transcript of New Series - hkjpaed.org