SAJCH Vol 11, No 3 (2017) by HMPG

CHILD HEALTH SOUTH AFRICAN JOURNAL OF

October 2017

Volume 11

Number 3

• Outcomes of children with West syndrome in KwaZulu-Natal, SA • Soaps and cleansers for atopic eczema: What every paediatrician should know about pH • Determinants of antibiotic prescription in Mozambique • Factors associated with severity of motor impairment in CP children in Nigeria • Parents' knowledge of ASD treatments

CHILD HEALTH SOUTH AFRICAN JOURNAL OF

OCTOBER 2017

Volume 11

Number 3

CONTENTS Short Report

108 Reflexivity: Parent of a child with ASD

Research

109 Determinants of antibiotic prescription in paediatric patients: The case of two hospitals in Maputo, Mozambique

L G S Monteiro, A Chaúque, M P Barros, T R Irá

112 Factors associated with the severity of motor impairment in children with cerebral palsy seen in Enugu, Nigeria

S O Iloeje, C C Ogoke,

117 The views and knowledge of parents of children with autism spectrum disorder on a range of treatments

EDITOR J M Pettifor FOUNDING EDITOR N P Khumalo EDITORIAL BOARD Prof. M Adhikari (University of KwaZuluNatal, Durban) Prof. M Kruger (Stellenbosch University) Prof. H Rode (Red Cross War Memorial Children's Hospital, Cape Town) Prof. L Spitz (Emeritus Nuffield Professor of Paediatric Surgery, London) Prof. A Venter (University of the Free State, Bloemfontein) Dr T Westwood (Red Cross War Memorial Children's Hospital, Cape Town) Prof. D F Wittenberg (University of Pretoria) HEALTH & MEDICAL PUBLISHING GROUP: CEO AND PUBLISHER Hannah Kikaya EXECUTIVE EDITOR Bridget Farham MANAGING EDITOR Naadia van der Bergh TECHNICAL EDITORS Claudia Naidu

V Wetherston, S Gangat, N Shange, K Wheeler, S B Sayed Karrim, J Pahl

PRODUCTION MANAGER Emma Jane Couzens

122 Toilet training practices in Nigerian children A U Solarin, O A Olutekunbi, A D Madise-Wobo, I Senbanjo

DTP AND DESIGN Travis Arendse Clinton Griffin

129 The barriers that women face when choosing food for their primary school children: A case study in the Western Cape Province, South Africa

CHIEF OPERATING OFFICER Diane Smith | Tel. 012 481 2069 Email: dianes@hmpg.co.za

Y Smit, S Kassier, D Nel, N Koen

135 The clinical profile and outcome of children with West syndrome in KwaZulu-Natal Province, South Africa: A 10-year retrospective review

A Keshave, N Yende-Zuma, L Mubaiwa, M Adhikari

141 Fanconi anaemia in South African patients with Afrikaner ancestry C Feben, T Haw, D Stones, C Jacobs, C Sutton, J Kromberg, A Krause

146 Soaps and cleansers for atopic eczema, friends or foes? What every South African paediatrician should know about their pH

N C Dlova, T Naicker, P Naidoo

Case Report

149 Acute poisoning in children from Jatropha curcas seeds

M C Moshobane, C Wium, L V Mokgola

151

CPD

ONLINE SUPPORT Gertrude Fani | Tel. 021 532 1281 Email: publishing@hmpg.co.za FINANCE Tshepiso Mokoena HMPG BOARD OF DIRECTORS Prof. M Lukhele (Chair), Dr M R Abbas, Dr Mrs H Kikaya, Dr M Mbokota, Dr G Wolvaardt HEAD OFFICE Block F, Castle Walk Corporate Park, Nossob Street, Erasmuskloof Ext. 3, Pretoria, 0181 EDITORIAL OFFICE Suite 11, Lonsdale Building, Lonsdale Way, Pinelands, 7405 Tel. 021 532 1281 Email: publishing@hmpg.co.za ISSN 1999-7671

Use of editorial material is subject to the Creative Commons Attribution – Noncommercial Works Licence. http://creativecommons.org/licenses/by-nc/4.0

ublished by the Health and Medical Publishing Group, P Suite 11, Lonsdale Building, Lonsdale Way Pinelands 7405

Cover: Anako, Red Cross War Memorial Children's Hospital Primary School

apers for publication should be addressed to the Editor, P via the website: www.sajch.org.za Tel: 021 532 1281 E-mail: publishing@hmpg.co.za ©Copyright: Health and Medical Publishing Group (Pty) Ltd

SHORT REPORT

This open-access article is distributed under Creative Commons licence CC-BY-NC 4.0.

Reflexivity: Parent of a child with ASD To the editor: As a parent of a child with autism spectrum disorder (ASD), I felt it necessary to write this reflexivity piece about my research journey around the topic of ASDs in the South African (SA) context. I have also chosen to include this section of my research that draws from the following statement by Guillemin and Gillam:[1] ‘Our research interests and the research questions we pose, as well as the questions we discard, reveal something about who we are.’ I was asked by the majority of my study participants why I had chosen to conduct a study of this nature. I initially found this question difficult and avoided answering it, but I soon realised that the question was unavoidable. As I could not divulge that I am a parent of a child with ASD I answered the question as follows: ‘This topic is very important to me.’ One important aspect that interested me, was understanding how parents from disadvantaged backgrounds cope with the costs associated with raising a child with ASD. I have personally experienced the daunting process of consulting many different medical specialists without obtaining a diagnosis for my boy. I spent vast amounts on therapies and medications, and wanted to understand how a parent who was less privileged would cope with raising a child with ASD. Most of all, I needed to understand how the child’s development might be affected by limited financial resources in the household. Through conducting this study, I have gained immense knowledge and information on the topic of ASD within the SA context. The first discovery was that the SA government has created special grants for children and parents who are less privileged. Through these grants, parents who were less advantaged may be able to cater for the needs of their children. The second discovery was that the SA government has opened schools that cater specifically for the educational needs of children diagnosed

108

SAJCH

with ASD. However, there are too few schools available for all the children who have been diagnosed. Some schools have long waiting lists of parents who need a place for their children to be schooled. As a final comment, embarking on a personal topic is an emotional process. There were moments in the field where I felt overwhelmed with emotions which were aroused by parents who were misinformed, lacked knowledge and sometimes were frustrated by the process of trying to gain services or even obtaining their child’s diagnosis. Some of the struggles that these parents were experiencing were struggles that I personally relate to and experienced. Most of these parents were very keen to take part in the study with the hope that they might find answers or information to ease their frustration. Their anxiety was disheartening and left me with an even greater desire to pursue more research in the area of ASD in the SA context.

Ayanda Purity Simelane

Department of Psychology, Faculty of Humanities, The Wits Reproductive Health and HIV Institute, University of the Witwatersrand, Johannesburg, South Africa simelane.ayanda1@gmail.com S Afr J Child Health 2017;11(3):108. DOI:10.7196/SAJCH.2017.v11i3.1422

1. Guillemin M, Gillam L. Ethics, reflexivity, and ‘ethically important moments’ in research. Qual Inq 2004;10(2):261-280. https://doi. org/10.1177/1077800403262360 2. Ryan L, Golden A. ‘Tick the box please’: A reflexive approach to doing quantitative social research. Sociology 2006;40(6):1191-1200. https://doi. org/10.1177/0038038506072287

OCTOBER 2017 Vol. 11 No. 3

RESEARCH

This open-access article is distributed under Creative Commons licence CC-BY-NC 4.0.

Determinants of antibiotic prescription in paediatric patients: The case of two hospitals in Maputo, Mozambique L G S Monteiro,1 MSc; A Chaúque,2 BSc; M P Barros,2 BSc; T R Irá,1,2 BSc 1 2

Faculty of Medicine, Eduardo Mondlane University, Maputo, Mozambique Faculty of Sciences, Eduardo Mondlane University, Maputo, Mozambique

Corresponding author: L G S Monteiro (govilee@gmail.com) Background. The need for healthcare in paediatric patients is often due to respiratory diseases, acute diarrhoea and viral fever, which suggests a limited need for the use of antibiotics. Objectives. To identify the determinants of antibiotic prescription in hospitalised paediatric patients in Mozambique. Methods. A cross-sectional study was conducted between January and June 2015. A total of 454 medical prescriptions and clinical records of children aged 0 - 14 years from Hospital Central de Maputo (HCM) and Hospital Geral de Mavalane (HGM) were analysed. Results. Antibiotics were used in 97.6% of the patients, with no significant differences (p>0.05) in the prescription rates of the hospitals. The most commonly used antibiotics were beta-lactams (57.3%), aminoglycosides (28.3%) and co-trimoxazole (9.4%). Antibiotics were prescribed in all cases of bronchopneumonia, fever, sepsis and acute gastroenteritis. For malaria and undefined diagnoses, antibiotics were prescribed 97.8% and 99.3% of cases, respectively. It was clear that most severe clinical conditions (odds ratio (OR) 9.06; 1.13 - 12.14) and age <5 years (OR 5.47; 1.54 - 7.60) were treated with antibiotics. Conclusion. The prescription of antibiotics for paediatric patients at both HCM and HGM was largely influenced by patients’ clinical condition and age. It showed that physicians used an empirical approach, in the absence of laboratory tests, often leading to unnecessary antibiotic treatments with negative causative effects. Physicians should be encouraged to use an evidence-based approach for managing the cases correctly. S Afr J Child Health 2017;11(3):109-111. DOI:10.7196/SAJCH.2017.v11i3.1224

Disease occurs at all ages, with pneumonia, malaria and diarrhoea being the most prevalent among paediatric cases.[1,2] These and other childhood diseases can be caused by different pathogens including viruses, bacteria, fungi and protozoa and identifying the aetiological agents to establish the appropriate therapeutic approach is essential.[3-6] According to van Buul et al.,[7] a physician’s decision to use antibiotics is guided by six factors: the patient’s clinical condition, advance care plans, using diagnostic resources, physician-perceived risks, the influence of family members, and the influence of the environment. Good prescription practice relies on determining the infection aetiology. Lack of resources can result in errors like prescribing antibiotics in the absence of a bacterial infection.[8,9] The use of antibiotics for viral diseases and indeterminate fever is associated with poor clinical outcomes and increased costs for both patients and health systems, and it can result in resistance to antibiotics.[2,10-12] To avoid antibiotic prescription errors and their consequences, physicians should consider the following: clinical and physiological condition of the patient, presence of bacterial infection, antibiotic spectrum, route of administration, pharmacokinetic and side-effects, as well as sociodemographic characteristics.[13] In Mozambique, little is known about the factors that influence antibiotic prescriptions in paediatric patients, and therefore this study was conducted to identify the determinants of antibiotic prescription in hospitalised paediatric patients in Mozambique.

Methods

Between January and June 2015, a cross-sectional study was conducted in two hospitals in Maputo, Mozambique: the Hospital Central de Maputo (HCM), a level-4 hospital (national reference), and Hospital Geral de Mavalane (HGM), a level-3 hospital with a wide coverage range for Maputo. Although each facility has a microbiology laboratory to support diagnoses, the HGM has particularly limited resources and services. The paediatrics department of HCM has 338 beds, while HGM has 80 beds. Data from medical prescriptions and clinical records of children

109

SAJCH

admitted to the emergency services were analysed to screen the use and the reasons for prescribing antibiotics. According to the World Health Organization (WHO), a statistically viable antibiotic prescription analysis requires a minimum of 100 prescriptions.[13] Therefore, a total of 454 patient records, with and without antibiotic prescriptions, were randomly selected from every third new child until the sample number conformed to WHO regulations. The information collected included demographic data, clinical conditions (as perceived by the physician and referred to in the medical records), common diseases, drugs and route of administration. Data analyses were performed using the Statistical Package for Social Sciences version 19 (IBM Corp., USA). A logistical regression model was constructed to measure the influence of clinical and demographic characteristics on antibiotic use. Descriptive statistics were used to characterise the patients’ antibiotic exposures and common routes of administration. This study was approved by Comité Institucional de Bioética em Saúde da Faculdade de Medicina & Hospital Central de Maputo (CIBS FM & HCM). Additional authorisations were obtained from HGM and HCM.

Results

A total of 454 (229/665 from HCM and 225/860 from HGM) inpatient medical records were analysed. The majority (81.6%) of patients were <5 years old and male (Table 1). Bronchopneumonia was diagnosed in 36.6%, followed by undefined diagnoses (30.4%) (Table 2). Culture and sensitivity tests were performed on 16.2% (n=37) of HCM patients, but on no patients from HGM. Antibiotic prescription rates did not differ significantly (p>0.05) between the two hospitals, with an average prescription rate of 97.6%. The most commonly used antibiotics were beta-lactams (57.3%), aminoglycosides (28.3%), co-trimoxazole (9.4%), chloramphenicol (2.2%) and macrolides (1.0%). Antibiotics were prescribed for children with bronchopneumonia (100%),

OCTOBER 2017 Vol. 11 No. 3

RESEARCH Table 1. Patientsâ&#x20AC;&#x2122; demographic characteristics (N=454)

Table 4. Factors that influence the use of antibiotics

Gender

Factor (reference)

95% CI

p-value

Age, years

Male, n (%)

Female, n (%) Total, N (%)

Hospital level (HCM)

2.576

(0.654 - 6.37)

0.18

â&#x2030;¤2

139 (57.0)

131 (62.4)

270 (59.5)

Clinical condition (severe)

9.056

(1.134 - 12.328)

0.03

3-5

60 (24.6)

36 (17.1)

96 (21.1)

Gender (male)

1.244

(0.359 - 4.313)

0.73

5.496

(1.541 - 7.595)

0.009

6 - 10

34 (13.9)

28 (13.3)

62 (13.7)

Age (<5 years)

11 - 14

11 (4.5)

15 (7.1)

26 (5.7)

OR = odds ratio; CI = confidence interval; HCM = Hospital Central de Maputo.

Table 2. Common diagnoses referred to in medical records (N=454) Hospital Diagnosis

HCM, n (%)

HGM, n (%)

Total, N (%)

Bronchopneumonia

60 (26.2)

106 (47.1)

166 (36.6)

Malaria

20 (8.7)

35 (15.6)

55 (12.1)

Fever

14 (6.1)

9 (4.0)

23 (5.1)

Sepsis

14 (6.1)

7 (3.1)

21 (4.6)

Acute gastroenteritis

17 (7.4)

3 (1.3)

20 (4.4)

Multiple (undefined)

84 (36.7)

54 (24.0)

138 (30.4)

Other*

20 (8.7)

11 (4.9)

31 (6.8)

Total

229 (100)

225 (100)

454 (100)

HCM = Hospital Central de Maputo; HGM = Hospital Geral de Mavalane. *Tonsillitis, kwashiorkor, tuberculosis, marasmus, asthma, varicella, otitis, meningitis.

malaria (97.8%), fever (100.0%), sepsis (100.0%), acute gastroenteritis (100.0%) and undefined diagnoses (99.3%) (Table 3). The main route of antibiotic administration was parenteral (52.9%), and the oral route was never exclusively used. The influence of the hospital environment, age of the patient, and clinical condition on antibiotic use were analysed. Both age and clinical condition were associated with the use of antibiotics (Table 4).

Discussion

This study indicates the excessive use of antibiotics in paediatric patients in Mozambique, regardless of the presence of bacterial infection, such as in malaria cases. A similar picture was found in Ethiopia[14] and Nigeria.[15] Possible explanations include drug availability, the presence of nonspecific infections, such as upper respiratory tract infections,

empirical treatment, and prophylactic use.[17-19] Empirical treatment and prophylactic use seem to be an effective solution for treatment, but they are frequently associated with poor clinical responses and the selection of antibiotic-resistant bacteria.[20] The most common clinical conditions were bronchopneumonia, malaria, fever, sepsis and acute gastroenteritis. It is important to note the high proportion of undefined diagnoses due to limited laboratory studies. Confirming the presence of a bacterial infection is essential to guiding the use of antibiotics and improve clinical responses. Fisher et al.[21] noted that the absence of laboratory studies to distinguish viral from bacterial infections may promote continued antibiotic use. Physicians should be encouraged to distinguish between viral and bacterial infections through laboratory testing to verify their diagnoses. Beta-lactams and aminoglycosides were among the most commonly used antibiotics, particularly the combination of ampicillin and gentamicin. The Mozambican Ministry of Health guidelines recommend this ampicillin/gentamicin combination in children <2 months old, and penicillin/chloramphenicol for children >2 months old. The ampicillin/gentamicin combination is recognised as a useful first-line treatment to manage bacterial sepsis in children <5 years old.[22] Despite the fact that the oral route of administration is recommended for children, this was never used exclusively in either health facility.[13,23] The reasons for parenteral use included the clinical condition of the patient, the need for rapid action, the belief that injectable antibiotics are more potent, and the availability of injectable antibiotics.[14,16,23,24] Children <5 years old received antibiotics more often than older children, possibly owing to more severe clinical conditions in younger patients.[25] In addition, regardless of laboratory confirmation, clinical conditions played an important role in the use of antibiotics. Physiciansâ&#x20AC;&#x2122; decisions were based on clinical conditions because of long turnaround times as a result of delayed responses from the laboratory. Physician-perceived risks associated with the need for rapid assistance may also have influenced the decisions to use antibiotics at HCM and HGM.[7] Even though these perceptions were not explored, our findings highlight the excessive use of antibiotics in paediatrics patients, regardless of verified bacterial infections.

Table 3. Antibiotics used by clinical condition Antibiotics Clinical condition

Betalactams, n (%)

Aminoglycosides, n (%)

Chloramphenicol, n (%)

TetracyQuinolones, cline, n (%) n (%)

Co-trimoxa- Macrolides, zole, n (%) n (%)

Total

Bronchopneumonia

166 (100)

82 (49.4)

7 (4.2)

0 (0.0)

1 (0.6)

18 (10.8)

4 (2.4)

166

Malaria

45 (97.8)

4 (8.7)

1 (2.2)

0 (0.0)

4 (8.7)

0 (0.0)

Fever

23 (100)

11 (47.8)

1 (4.3)

0 (0.0)

1 (4.3)

0 (0.0)

Sepsis

21 (100)

16 (76.2)

1 (4.8)

0 (0.0)

1 (4.8)

5 (23.8)

1 (4.8)

Acute gastroenteritis

19 (100)

11 (57.9)

0 (0.0)

1 (5.3)

0 (0.0)

3 (15.8)

0 (0.0)

Multiple (undefined)

136 (99.3)

78 (56.9)

6 (4.4)

5 (3.6)

36 (26.3)

3 (2.2)

137

Other*

31 (100)

16 (51.6)

1 (3.2)

0 (0.0)

5 (16.1)

0 (0.0)

Total

441 (99.5)

218 (49.2)

17 (3.8)

7 (1.6)

72 (16.3)

8 (1.8)

443

*Tonsillitis, kwashiorkor, tuberculosis, marasmus, asthma, varicella, otitis, meningitis.

110

SAJCH

OCTOBER 2017 Vol. 11 No. 3

RESEARCH Conclusion

The prescription of antibiotics for paediatric patients at both HCM and HGM was largely influenced by patients’ clinical condition and age. It showed that physicians used an empirical approach, in the absence of laboratory tests, often leading to unnecessary antibiotic treatments with negative causative effects. Actions need to be taken to encourage physicians to use evidence-based approaches for managing the cases correctly. Acknowledgements. To the Paediatrics Departments of HCM and HGM for providing us the data needed, and to all our colleagues. Author contributions. LGSM, AC, MPB and TRI contributed equally to the research and writing up of the review. Funding. None. Conflicts of interest. None. 1. Feleke M, Yenet W, Lenjisa JL. Prescribing pattern of antibiotics in pediatric wards of Bishoftu Hospital, East Ethiopia. Int J Basic Clin Pharmacol 2013;2(6):718-722. https://doi.org/10.5455/2319-2003.ijbcp20131209 2. Kardas-Sloma L, Boelle PY, Opatowski L, Brun-Boisson C, Guilllemot D, Temime L. Impact of antibiotic exposure patterns on selection of communityassociated methicillin-resistant Staphylococcus aureus in hospital settings. Antimicrob Agents Chemother 2011;55(10):4888-4895. https://doi. org/10.1128/aac.01626-10 3. Kieninger E, Fuchs O, Latzin P, Frey U, Regamey N. Rhinovirus infections in infancy and early childhood. Eur Resp J 2013;41(2):443-452. https://doi. org/10.1183/09031936.00203511 4. O'Callaghan-Gordo C, Bassat Q, Diez-Padrisa N, et al. Lower respiratory tract infections associated with rhinovirus during infancy and increased risk of wheezing during childhood. A cohort study. PLoS ONE 2013;8(7):e69370. https://doi.org/10.1371/journal.pone.0069370 5. Saha SK, Al Emran HM, Hossain B, et al. Streptococcus pneumoniae serotype-2 childhood meningitis in Bangladesh: A newly recognized pneumococcal infection threat. PLoS ONE 2012;7(3):e32134. https://doi.org/10.1371/journal. pone.0032134 6. Barry MA, Weatherhead JE, Hotez PJ, Woc-Colburn L. Childhood parasitic infections endemic to the United States. Pediatr Clin North Am 2013;60(2):471485. https://doi.org/10.1016/j.pcl.2012.12.011 7. Van Buul LW, van der Steen JT, Doncker SM, et al. Factors influencing antibiotic prescribing in long-term care facilities: A qualitative in-depth study. BMC Geriatr 2014;14(1):136-147. https://doi.org/10.1186/1471-2318-14-136 8. Korhonen L, Kondrashova A, Tauriainen S, et al. Enterovirus infections in early childhood and the risk of atopic disease – a nested case-control study. Clin Exp Allergy 2013;43(6):625-632. https://doi.org/10.1111/cea.12068 9. Lee GC, Reveles KR, Attridge RT. Outpatient antibiotic prescribing in the United States: 2000 to 2010. BMC Med 2014;12(1):96. https://doi. org/10.1186/1741-7015-12-96

111

SAJCH

10. Bharathiraja R, Sridharan S, Chelliah LR, Suresh S, Senguttuvan M. Factors affecting antibiotic prescribing pattern in pediatric practice. Indian J Pediatr 2005;72(10):877-879. https://doi.org/10.1007/bf02731121 11. Al-Jeraisy MI, Alanazi MQ, Abolfotouh MA. Medication prescribing errors in a pediatric inpatient tertiary care setting in Saudi Arabia. BMC Res Notes 2011;4:294-300. https://doi.org/10.1186/1756-0500-4-294 12. Llor C, Bjerrum L. Antimicrobial resistance: Risk associated with antibiotic overuse and initiatives to reduce the problem. Ther Adv Drug Saf 2014;5(6):229241. https://doi.org/10.1177/2042098614554919 13. World Health Organization. Medicines: Medicines for Children. Geneva: WHO, 2010. 14. Alemnew G, Atnafie SA. Assessment of the pattern of antibiotics use in pediatrics ward of Dessie referral hospital, North East Ethiopia. Int J Med Med Sci 2015;7(1):1-7. https://doi.org/10.5897/ijmms2014.1101 15. Fadare J, Olatunya O, Oluwayemi O, Ogundare O. Drug prescribing pattern for under-fives in a paediatric clinic in south-western Nigeria. Ethiop J Health Sci 2015;25(1):73-78. https://doi.org/10.4314/ejhs.v25i1.10 16. Del Fiol FdeS, Lopes LC, Barberato-Filho S, Motta CdeC. Evaluation of the prescription and use of antibiotics in Brazilian children. Braz J Infect Dis 2013;17:332-337. https://doi.org/10.1016/j.bjid.2012.10.025 17. Zoorob R, Sidani MA, Fremont RD, Kihlberg C. Antibiotic use in acute upper respiratory tract infections. Am Fam Physician 2012;86(9):817-822. 18. Greenfield SP. Antibiotic prophylaxis in pediatric urology: An update. Curr Urol Rep 2011;12(2):126-131. https://doi.org/10.1007/s11934-010-0164-y 19. Song S-H, Kim KS. Antibiotic prophylaxis in pediatric urology. Indian J Urol 2008;24(2):145-149. https://doi.org/10.4103/0970-1591.40605 20. Bryce A, Hay AD, Lane IF, Thornton HV, Wootton M, Costelloe C. Global prevalence of antibiotic resistance in paediatric urinary tract infections caused by Escherichia coli and association with routine use of antibiotics in primary care: Systematic review and meta-analysis. BMJ 2016;352:i939. https://doi. org/10.1136/bmj.i93921. 21. Fisher BT, Meaney PA, Shah SS, Steenhof AP. Short Report: Antibiotic Use in pediatric patients admitted to a referral hospital in Botswana. Am J Trop Med Hyg 2009;81(1):129-131. 22. Bibi S, Chisti MJ, Fau-Akram F, Akram F, Fau-Pietroni MAC, Pietroni MA. Ampicillin and gentamicin are a useful first-line combination for the management of sepsis in under-five children at an urban hospital in Bangladesh. J Health Popul Nutr 2012;30(4):487-490. https://doi.org/10.3329/ jhpn.v30i4.13418 23. Teni FS, Surur AS, Getie A, Alemseged A, Meselu M. Medication prescribing pattern at a pediatric ward of an Ethiopian hospital. Int J Pediatr 2014;2(4.2):2330. https://doi.org/10.22038/ijp.2014.3342 24. Janjua NZ, Hutin YJ, Akhtar S, Ahmad K. Population beliefs about the efficacy of injections in Pakistan's Sindh province. Public Health 2006;120(9):824-833. https://doi.org/10.1016/j.puhe.2006.05.004 25. Krishnan A, Amarchand R, Gupta V, et al. Epidemiology of acute respiratory infections in children ‑ preliminary results of a cohort in a rural north Indian community. BMC Infect Dis 2015;15(1):1-10. https://doi.org/10.1186/s12879015-1188-1

OCTOBER 2017 Vol. 11 No. 3

RESEARCH

This open-access article is distributed under Creative Commons licence CC-BY-NC 4.0.

Factors associated with the severity of motor impairment in children with cerebral palsy seen in Enugu, Nigeria S O Iloeje,1 MB BCh, FWACP (Paed); C C Ogoke,2 MBBS, FWACP (Paed) 1 2

Department of Pediatrics, University of Nigeria Teaching Hospital, Ituku-Ozalla, Enugu, Nigeria Department of Pediatrics, Federal Medical Center, Owerri, Imo State, Nigeria

Corresponding author: C C Ogoke (chrischikere@yahoo.com) Background. Cerebral palsy (CP) is a heterogeneous condition that is well known to cause impairments with varying degrees of severity. The gross motor function classification system (GMFCS) is widely used to assess ambulatory function in CP, but little is known about the factors that account for the variations in gross motor function in children. The purpose of this study was to assess the relation between the severity of gross motor dysfunction (GMD) and certain factors such as the type of CP, aetiology of CP, nutrition, socioeconomic class (SEC), and the frequency of these accompanying impairments like visual, auditory, cognitive and speech impairments. Methods. This was a cross-sectional observational study of 100 consecutively recruited CP patients aged 9 - 96 months, who attended the paediatric neurology clinics (PNCs) in Enugu between April and October 2010. Each patient’s clinical history was recorded, a neurological examination conducted and GMFCS level ascertained. Statistical analyses were done to determine the association between the categorical variables. Results. The type of CP (p=0.000), aetiological factors (p=0.016), the presence of malnutrition (p=0.004) and the frequency of accompanying impairments (p=0.001) were significantly associated with the severity of GMD, while SEC (p=0.649) had no significant association. Conclusion. The type of CP, aetiological factors, the presence of malnutrition and the number of accompanying physical, mental or physiological impairments, were positively associated with the severity of GMD and walking ability in children with CP. S Afr J Child Health 2017;11(3):112-116. DOI:10.7196/SAJCH.2017.v11i3.1246

Cerebral palsy (CP) is a neurodevelopmental disability. It is the most common cause of physical disability in childhood and occurs worldwide with a prevalence of 2 - 2.5 per 1 000 live births in the Western world.[1] The prevalence of CP in Nigeria is unknown and the rate of children attending neurology clinics in Nigeria varies broadly from 16% to 50.3%.[2–4] The University of Nigeria Teaching Hospital (UNTH), Enugu, reported CP to be the second most frequent neurological disorder seen in the paediatric neurology clinic (PNC);[2] however, there is a paucity of studies on gross motor dysfunction (GMD) in CP in Africa. The most current definition of CP is, ‘CP describes a group of permanent disorders of the development of movement and posture causing activity limitation that are attributed to non-progressive disturbances that occurred in the developing fetal or infant brain. The motor disorders of cerebral palsy are often accompanied by disturbances of sensation, perception, cognition, communication and behaviour, by epilepsy and secondary musculoskeletal problems.’[5] These impairments are experienced to varying degrees and the severity of CP is based on a functional classification of these impairments.[1,5,6] There are various functional scales, with the most popular one being the Gross Motor Function Classification System (GMFCS). GMFCS quantifies a child’s gross motor function (ambulatory function) into five levels of severity, ranging from walking without restrictions (level I), to total dependence for ambulation (level V).[5,6] Vasconcellos et al.[7] in Brazil reported 34.3% of the study population as ambulatory while Nordmark et al.[8] in southern Sweden reported a much higher proportion of ambulatory patients (73%). Medical research hasn't been able to explain this huge discrepancy. The aim of this study was to identify the factors that are associated with the severity of motor function, as determined by the GMFCS in children with CP in Enugu, Nigeria.

Methods

Study sites

The study was carried out in the PNCs of the UNTH, Ituku-Ozalla, and Enugu State University Teaching Hospital (ESUTH). The UNTH is

112

SAJCH

one of the first-generation teaching hospitals in Nigeria. It has multidisciplinary departments and caters for patients from predominantly the south-eastern region of the country. The PNC of UNTH operates once a week and caters primarily for children who have neurological conditions and have been referred. ESUTH is a smaller and relatively new teaching hospital. Its PNC also runs once a week (on a different weekday).

Sampling method

This was a cross-sectional observational study. Sample selection was non-randomised and the subjects were consecutively recruited until the desired sample size was reached. Patients aged between 9 and 96 months with a diagnosis of CP were recruited. Patients with other movement disorders, physical disabilities and motor abnormalities, such as muscular dystrophies, paralytic poliomyelitis and spina bifida (myelomeningocele) were excluded.

Ethical approval and consent

This research was approved by the UNTH Research Ethics Committee and this was accepted by the ESUTH. Informed consent was obtained from the parents/guardians of the patients before recruitment and confidentiality of the patients’ information was ensured.

Study population

The cohort consisted of CP patients, with a median age of 32 months (range 9 - 36), who attended the PNCs of the abovementioned teaching hospitals between April and October 2010. This age bracket was chosen in view of the age distribution found in a previous study of CP patients in Enugu.[9] It is necessary to mention that spastic quadriplegics classified in this study as ambulatory included children walking with hand-held mobility devices and those who could propel a manual wheelchair (level III), and therefore should not be misconstrued only as those walking unaided.

OCTOBER 2017 Vol. 11 No. 3

RESEARCH For each child, sociodemographic data, including parental level of education and occupation were ascertained and data were recorded. Relevant history about the antenatal, perinatal and postnatal periods and developmental milestones was recorded. The gross motor developmental milestones indicated a delay in acquisition of motor milestones in all cases without a regression or loss of milestones already achieved. History also indicated motor impairment and other CP-related symptoms such as epilepsy, visual, speech, cognitive and hearing impairments in the first 2 - 3 years of life. A consistent history and descriptions of motor findings, consistent with the most recent definition of CP, were used to determine case status.[5] Anthropometry was done to determine the nutritional state of each child. A flexible inelastic tape measure was used to measure the occipitofrontal circumference (OFC) to the nearest 0.1 cm. An infantometer was used to measure the length of children <3 years old and children who could not stand. The length of older, non-ambulatory patients (GMFCS level IV and V) was measured using a flexible, inelastic tape measure. An assistant helped to hold their legs straight by pressing on the child’s chest and knees to enable measurement from the top of the head to the soles. None of the patients had severe lower-limb contractures that required length measurement in segments. Children who could stand flat-footed and bear their own weights had their weights and heights measured with a standing scale equipped with a stadiometer (SM–160, Surgifield Medical, England). The basinet scale (Model 180, Salter, Australia) was used to weigh patients ≤15 kg. None of the patients had a feeding gastrostomy. Detailed age-adjusted neurological examinations were conducted and showed abnormalities of movement, posture, tone and reflexes (deep tendon and primitive). The number of impairments in each child was recorded and included speech, auditory and visual impairments, undernutrition, epilepsy, microcephaly, drooling and contracture. The current motor abilities of each child in the home setting were ascertained from the parent/guardian. Subsequently, for each child the following motor abilities were assessed: head control and observed lying, sitting, crawling, standing, walking, running a few meters and climbing stairs. Each child’s present ability and limitations were determined. Children who were <18 months old were assessed for head control, floorsitting with or without hand support, and whether the child could crawl or pull to stand. Using the GMFCS, each child was classified into one of five levels of severity by cross-checking the findings in the appropriate age band of the child.[10] All findings were carefully recorded in the proforma.

Data analysis

Data were analysed using the statistical package for the social sciences (SPSS version 15; IBM Corp., USA.) The categorical variables (GMFCS) levels, type of CP, socioeconomic class (SEC), aetiological factors, presence or absence of malnutrition, frequency of accompanying impairments) were further examined for association with GMD severity using Fisher’s exact test; (p<0.05). The SEC was obtained by calculating the mean value (to the nearest whole number) of the scores for the occupation and level of education of their parents or substitutes as recommended by Oyedeji.[11] Based this information, the subjects were classified into upper class (I and II), middle class (III) and lower class (IV and V). The upper class correlated with high-income earners with formal education and adequate health knowledge, while the lower SEC represented lowincome earners who lacked formal education and social amenities. The middle classes were intermediate between high- and low-income earners. The OFC of each child was plotted on the World Health Organization (WHO) head circumference-for-age child growth chart and those below the third percentile for their age were noted to have microcephaly.[12] The following anthropometric indices of the WHO child growth standards were used for analysis: weight-for-height Z scores (WHZ) and height-for-age Z scores (HAZ).[12,13] Wasting and stunting were reported following the internationally recognised definitions proposed by WHO/UNICEF. Wasting (acute malnutrition) was defined as WHZ <–2 and severe

113

SAJCH

wasting as WHZ <–3. Stunting (chronic malnutrition) was defined as HAZ <–2 and severe stunting as HAZ <–3.[12,13] The distribution and types of malnutrition in our cohort is shown in Fig. 1.

Results

Biodemographic characteristics of study population

Out of the 100 consecutively recruited patients, there were 58 males and 42 females, giving a male:female gender ratio of 1.4:1. The mean (SD) age of the children was 32.0 (22.7) months. The mean (SD) age of males and females was 27.95 (18.3) and 37.69 (26.9) months, respectively. The patients comprised 43% high SEC (I and II), 25% middle SEC and 32% low SEC individuals (Table 1).

Association between severity of GMD and categorical variables SEC and severity of GMD

The severity of GMD in terms of ambulatory and non-ambulatory status was analysed in the context of the SEC of the study subjects. Differences in the distribution were not statistically significant (p=0.649) (Table 1).

Type of CP and severity of GMD

The severity of GMD in relation to the type of CP was statistically significant, indicating an association between the type of CP and the severity of GMD (p=0.000) (Table 2). Spastic quadriplegia showed the highest correlation with severe GMD. Hemiplegic and hypotonic CP were associated with the lowest GMD.

Aetiological factors and severity of GMD

There was no significant association between the occurrence of single, multiple or unknown aetiological factors and the severity of GMD, (p=0.791) (Table 3). However, different aetiological factors analysed in isolation showed a statistically significant correlation between the type of factor and the severity of GMD (p=0.0159). Children with postnatal central nervous system (CNS) infection as a single aetiological factor all had severe GMD, indicating the highest correlation with severe GMD (Table 4).

Malnutrition and severity of GMD

A greater proportion (65.1%) of patients with normal nutritional status were ambulatory, while a greater proportion (64.9%) with malnutrition were non-ambulatory. GMD and the presence of malnutrition showed a statistically significant association (p=0.004) (Table 5).

Frequency of accompanying impairments and severity of GMD

The frequency of associated impairments was proportionate to the severity of the GMD. All patients with five or six associated impairments had severe GMD showing a statistically significant association (p=0.001) (Table 6). 70

60 50

Patients, %

Study protocol

40 30

20 9

0 Normal

Wasted

Stunted

Wasted and stunted

Fig. 1. Distribution and types of malnutrition in the patients.

OCTOBER 2017 Vol. 11 No. 3

RESEARCH Table 1. Severity of GMD by SEC of patients (N=100)

Table 4. Severity of GMD and type of aetiological factor among patients with single aetiological factor

GMFCS severity

Severity of GMD

Socioeconomic class

Ambulatory (n=54), n (%)

Nonambulatory (n=46), n (%)

SEC I

8 (50.0)

SEC II

15 (59.3)

12 (40.7)

Type of aetiological factor

Ambulatory (n=23), n (%)

Nonambulatory (n=46), n (%)

Total

Severe birth asphyxia

13 (61.9)

8 (38)

Total

SEC III

12 (48.0)

13 (52.0)

Prematurity

2 (50)

SEC IV

16 (53.3)

14 (46.7)

Kernicterus

7 (38.9)

11 61.1)

SEC V

2 (100)

Postnatal CNS infection

5 (100)

Infantile status epilepticus

1 (10)

9 (90)

GMD = gross motor dysfunction; SEC = socioeconomic class; GMFCS = gross motor function classification system. 2 χ =2.475, df=4, p=0.649.

GMD = gross motor dysfunction. Fisher’s exact test p=0.01.

Table 2. Severity of GMD by type of CP (N=100) GMFCS Severity

Type of CP

Ambulatory (n=54), n (%)

Nonambulatory (n=46), n (%)

Total

Spastic quadriplegia

10 (23.8)

32 (76.2)

Spastic hemiplegia

17 (100)

Spastic diplegia

3 (60.0)

2 (40.0)

Extrapyramidal

10 (50.0)

Hypotonic

7 (100)

Mixed

7 (77.8)

2 (22.2)

Table 5. Severity of GMD and malnutrition (N=100) GMFCS severity

Malnutrition

GMD = gross motor dysfunction; CP = cerebral palsy; GMFCS = gross motor function classification system. Fisher’s exact test 38.105, df=5, p=0.000.

Ambulatory (n=54) n (%)

Absent

41 (65.1)

22 (34.9)

13 (35.1)

24 (64.9)

GMD = gross motor dysfunction; GMFCS = gross motor function classification system. 2 χ =8.414, df=1, p=0.004.

Table 6. Severity of GMD by frequency of accompanying impairments (N=100) GMFCS severity Number of other impairments

Ambulatory (n=54), n (%)

Nonambulatory, (n=46), n (%)

Total

Single aetiological factor

31 (57.4)

27 (58.7)

Multiple aetiological factors

21 (38.9)

16 (34.8)

Unknown

2 (3.7)

3 (6.5)

Total

Present

Table 3. Aetiological factors and severity of GMD (N=100) Severity of GMD

Nonambulatory (n=46) n (%)

Ambulatory (n=54), n (%)

Nonambulatory (n=46), n (%)

One

19 (82.6)

4 (17.4)

Two

18 (69.2)

8 (30.8)

Three

10 (38.5)

16 (61.5)

Four

7 (36.8)

12 (63.2)

Five

0 (0.0)

5 (100)

Six

0 (0.0)

1 (100)

GMD = gross motor dysfunction. Fisher’s exact test p=0.791.

GMD = gross motor dysfunction; GMFCS = gross motor function classification system. Fisher’s exact test p=0.01.

Discussion

This study has shown that the factors associated with the severity of GMDs in CP patients are the neurological subtypes of CP, the aetiological factors, malnutrition, and the number of accompanying impairments. Further studies may show the impact of reducing these factors on ambulatory function in CP patients. Spastic quadriplegia, postnatal CNS infections, malnutrition and highly prevalent accompanying impairments were most notably associated with more severe GMD in our patients. These findings are all consistent with previously published data on CP.[7,8,14,15] However, this study showed no correlation between SEC status and the severity of GMD, irrespective of the fact that severely affected children survive longer if given superior support compared with those from low SEC environments who usually seek unorthodox interventions, or none at all.

114

SAJCH

This is surprising because patients from high SEC environments typically have normal nutritional statuses, they can commence early rehabilitation such as physiotherapy, and they have access to multidisciplinary hospital treatments – factors expected to positively affect gross motor abilities. The reason for the negative association is unclear and may require further investigation. It is not surprising that spastic quadriplegia caused severe GMD; whole-body involvement usually results from extensive brain damage such as a diffuse CNS infection and severe birth asphyxia, resulting in severe hypoxic-ischaemic encephalopathy. Vasconscellos et al.[7] and Nordmark et al.[8] reported spastic quadriplegic CP as predominantly nonambulatory, which is consistent with findings in this study. Interestingly,

OCTOBER 2017 Vol. 11 No. 3

RESEARCH 24% of the spastic quadriplegics were ambulatory. Those children with spastic quadriplegia (severe and equal spasticity of all four limbs) differed clinically from the spastic diplegics (more severe lower- limb involvement) and had different aetiological factors (mostly severe birth asphyxia and postnatal CNS infections for spastic quadriplegics v. premature birth for spastic diplegics). However, this clinical assessment is subjective and the need to improve the reliability of CP classifications has given birth to the classification by surveillance for CP in Europe (SCPE).[16] It is necessary to mention that spastic quadriplegics classified in this study as ambulatory included children walking with hand-held mobility devices and those who could propel a manual wheelchair (level III), and therefore should not be misconstrued only as those walking unaided. Furthermore, a greater proportion of the population studied (52%) were <24 months old and some of the spastic quadriplegics (most prevalent type) could have been classified to better functioning levels because the GMFCS is less precise in infants <24 months old.[17] This study, by classifying all children with hypotonic CP (7%) as ambulatory, differs significantly from the study by Pfeifer et al.,[14] who reported all children with hypotonic CP (3%) as nonambulatory. The reason for this finding is unclear. Due to a larger cortical representation of the hand and arm than the leg, upper-limb function is more impaired than lower-limb function and explains why patients with hemiplegia usually ambulate. All hemiplegic patients in this study were ambulatory and in agreement with the report by Gorter et al.,[15] who observed that the majority (87.8%) of hemiplegic CP was classified as level I (mild GMD). Similar reports were made by Vasconcellos et al.[7] and Nordmark et al.[8] The findings of this study are consistent with previous studies, reinforcing that children with hemiplegic CP are not likely to require mobility aids, while those patients with spastic quadriplegia are more likely to need augmentative interventions.[7,8,14,15] Though a small number (5%) of our patients had postnatal CNS infections (bacterial meningitis) as a single aetiological factor, it is interesting to note that all of them had severe GMD. This occurred because the infections cause widespread brain damage resulting in spastic quadriplegia, which is frequently associated with severe GMD and an inability to ambulate independently. The most recent definition of CP highlights the importance of accompanying physiological impairments and comorbidities in CP, by incorporating their occurrence as part of the definition of CP.[5] We found that all of our patients had at least one accompanying impairment. We sought to ascertain whether there was an association between the number of impairments and the severity of GMD, and not the effect of individual disorders such as epilepsy, on GMD. We hypothesised that a child with CP with a greater number of other impairments would have more severe GMD. As hypothesised, GMD became more severe with an increasing number of accompanying disorders (p=0.001), suggesting that the total number of associated deficits in any child with CP may give an indication of the functional abilities of that child. This finding is supported by Vasconcellos et al.,[7] who observed that associated impairments were mainly evidenced at GMFCS levels IV/V. This finding suggests more severe damage to the developing motor areas of the brain with an increasing number of other impairments. Conversely, it is possible that the presence of severe GMD in a child with CP could be predisposed to a higher number of accompanying impairments. For instance, a child with spastic quadriplegia on GMFCS level V is at a greater risk of having hip subluxation/dislocation, contractures and, if adequately fed, develops obesity (malnutrition) because of immobilisation. In addition, oromotor dysfunction, sucking and swallowing difficulties in this child may result in undernutrition and increased risk of aspiration and pneumonia. However, another child with spastic quadriplegia who is on level I (mild GMD) faces less risk of these impairments by being independently mobile and having less severe motor dysfunction. An earlier study in Enugu found that malnutrition was more prevalent (36%) in children with CP than their controls.[18] This prevalence rate was similar to our finding. Severe malnutrition negatively affects functional

115

SAJCH

motor skills by impairing brain maturation, reducing motor activity and causing muscle weakness. Due to the relatively high prevalence of malnutrition in children with CP and its worsening effect on existing GMD, paediatricians should pay attention to the nutrition of newly diagnosed children. It should be noted that severe GMP, as seen in spastic quadriplegia, could manifest with oromotor dysfunction (due to pseudobulbar palsy) causing sucking and swallowing difficulties, which results in undernutrition. In this study, it does seem likely that poor nutrition was secondary to motor dysfunction, since most of the malnourished patients were non-ambulatory with severe GMD. This emphasises the need for alternative feeding modalities like gastrostomies, which are largely unavailable in Nigeria. Nasogastric tube feeding is an easier alternative feeding intervention and should be encouraged in those children with swallowing difficulties. The statistically significant association between the severity of GMD and malnutrition (p=0.004) in this study is a valid finding, but with the realisation that this association could be bidirectional. Longitudinal or controlled studies are needed to learn more about this association.

Limitations

Some neurometabolic/neurodegenerative disorders present with motor dysfunction. None of the patients in this study had serum amino acids, organic acids, lactate, ammonia or urinary glycosaminoglycans measured (metabolic screen). It must therefore be acknowledged that the profile of aetiological factors presented in this study could have been influenced by both the investigative facilities available, e.g. no genetic/chromosomal studies or magnetic resonance imaging, as well as recall bias owing to the cross-sectional design of this study.

Conclusion

The type of CP, aetiological factors, malnutrition and the number of accompanying impairments correlated statistically significantly with the severity of GMD. SEC did not correlate with GMD severity as expected and further clinical research is required to elucidate the reasons for this finding. Ultimately, there is a need for further research on the factors causing variations in GMD in different populations of children with CP. Acknowledgements. We are grateful to our patients and their parents for participating in this study. We thank the staff of the Paediatric Neurology Clinics (PNCs) of UNTH and ESUTH, Enugu for their support. Author contributions. SOI was involved in the conception and design of the topic; he revised the article critically for important intellectual content and gave the final approval of the version to be published. CCO made substantial contribution to the conception and design of the topic; he was involved in the acquisition of data, data analysis, interpretation and drafting of the article. Both authors read and agreed to the final manuscript. Funding. None. Conflicts of interest. None. 1. Rosenbaum P. Cerebral palsy: What parents and doctors want to know. Br Med J 2003;326(7396):970-974. https://doi.org/10.1136/bmj.326.7396.970 2. Izuora GI, Iloeje SO. A review of neurological disorders seen at the Paediatric Neurology Clinic of the University of Nigeria Teaching Hospital Enugu. Ann Trop Paediatr 1989;9(4):185-189. https://doi.org/10.1080/02724936.1989.11748629 3. Frank-Briggs AI, Alikor EAD. Pattern of paediatric neurological disorders in Port Harcourt, Nigeria. Int J Biomed Sci 2011;7(2):145-149. 4. Ogunlesi T, Ogundeyi M, Ogunfowora O, Olowu A. Socio-clinical issues in cerebral palsy in Sagamu, Nigeria. S Afr J Child Health 2008;2(3):120-124. 5. Bax M, Goldstein M, Rosenbaum P, Paneth N. Proposed definition and classification of cerebral palsy. Dev Med Child Neurol 2005;47:571-576. https:// doi.org/10.1017/s001216220500112x 6. Palisano R, Rosenbaum P, Walter S, Russell D, Wood E, Gluppi B. Development and reliability of a system to classify gross motor function in children with cerebral palsy. Dev Med Child Neurol 1997;39(4):214-223. http://dx.doi. org/10.1111/j.1469-8749.1997.tb07414.x 7. Vasconcellos RLM, Moura TL, Campos TF, Lindquist ARR, Guerra RO. Functional performance assessment of children with cerebral palsy according to motor impairment levels. Rev Bras Fisioter 2009;13(5):390-397. https:// doi.org/10.1590/s1413-35552009005000051

OCTOBER 2017 Vol. 11 No. 3

RESEARCH 8. Nordmark E, HĂ¤gglund G, Lagergren J. Cerebral palsy in southern Sweden 1II Gross motor function and disabilities. Acta Paediatr 2001;90(11):12771282. https://doi.org/10.1111/j.1651-2227.2001.tb01575.x 9. Iloeje SO, Ejike-Orji I. Compliance by cerebral palsy patients attending a child neurology service in a developing country. A preliminary study. W Afr J Med 1993;12(1):1-5. 10. Palisano RJ, Rosenbaum P, Bartlett D, Livingston MH. Content validity of the expanded and revised Gross Motor Function Classification System. Dev Med Child Neurol 2008;50(10):744-750. 18834387. http://dx.doi.org/10.1111/ j.1469-8749.2008.03089.x 11. Oyedeji GA. Socioeconomic and cultural background of hospitalised children in Ilesa. Nig J Paediatr 1985;12(4):111-117. 12. WHO/UNICEF. WHO child growth standards and the identification of severe acute malnutrition in infants and children. A joint statement by the WHO & UNICEF Geneva: WHO, 2009. http://www.who.int/nutrition/publications/ severemalnutrition/9789241598163/en/ (accessed 10 May 2015). 13. UNICEF. Tracking progress on child and maternal nutrition; a survival and development priority. Geneva: WHO, 2009. https://www.unicef.org/

116

SAJCH

publications/files/Tracking_Progress_on_Child_and_Maternal_Nutrition_ EN_110309.pdf (accessed 10 May 2015). 14. Pfeifer LL, Silva DBR, Funayama CAR, Santos JL. Classification of cerebral palsy: Association between gender, age, motor type, topography and gross motor function. Arq Neuropsiquiatr 2009;67(4):1057-1061. https://doi. org/10.1590/s0004-282x2009000600018 15. Gorter JW, Rosenbaum PL, Hanna SE, et al. Limb distribution, motor impairment and functional classification of cerebral palsy. Dev Med Child Neurol 2004;46(7):461-467. https://doi.org/10.1017/s0012162204000763 16. Surveillance of cerebral palsy in Europe. SCPE Collaborative Group: A collaboration of cerebral palsy surveys and registers. Dev Med Child Neurol 2000;42(12):816-824. https://doi.org/10.1017/s0012162200001511 17. Gorter JW, Ketelaar M, Rosenbaum P, Helders PJM, Palisano R. Use of the GMFCS in infants with CP: The need for reclassification at 2 years or older. Dev Med Child Neurol 2008;51(1):46-52. https://doi.org/10.1111/j.14698749.2008.03117.x 18. Okeke IB, Ojinnaka NC. Nutritional status of children with cerebral palsy in Enugu, Nigeria. Euro J Sci Res 2010;39(4):505-513.

OCTOBER 2017 Vol. 11 No. 3

RESEARCH

This open-access article is distributed under Creative Commons licence CC-BY-NC 4.0.

The views and knowledge of parents of children with autism spectrum disorder on a range of treatments V Wetherston, BCom Path (SLP), MSLT; S Gangat, BCom Path (SLP); N Shange, BCom Path (SLP); K Wheeler, BCom Path (SLP); S B Sayed Karrim, MCom Path (SLP); J Pahl, BSc, MA, Dip Ed Discipline of Speech-Language Pathology, School of Health Sciences, University of KwaZulu-Natal, Durban, South Africa Corresponding author: J Pahl (pahlj@ukzn.ac.za) Background. Autism spectrum disorder (ASD) is a neurodevelopmental disorder that appears before the age of 3 years. Symptoms reflect delayed or abnormal social interaction and communication skills, with restricted or repetitive behaviour warranting the need for early intensive treatment. Methods. The aim of the study was to investigate the knowledge and views of parents regarding treatments for their children, aged between 5 and 9 years old with ASD, in eThekwini Metropolitan Municipality, South Africa. An embedded mixed methods research design was utilised. Nonrandom purposive sampling was used to select 46 parents of children with ASD. A 42-item questionnaire was used and the data were interpreted using descriptive statistics and thematic analysis. Results. More than half of the parents (53%) were unfamiliar with or had only heard of treatments in question, while 13.4% had a practical understanding of the treatments. Of all the treatments, parents rated their knowledge of speech-language therapy (SLT) most highly. The majority (68%) stated that they had difficulties accessing ASD treatment facilities and healthcare professionals, and perceived treatments as being costly. Even so, 74% of parents reported that they had a good relationship with their healthcare professional. Conclusion. The above findings should be viewed as motivation for health professionals to share information regarding the range of ASD treatments. They can assist parents in accessing appropriate facilities, recommend treatments that are supported by research, and update their knowledge on advances in ASD treatment. S Afr J Child Health 2017;11(3):117-121. DOI:10.7196/SAJCH.2017.v11i3.1274

Autism spectrum disorder (ASD) is a severe, lifelong neurodevelopmental condition without a cure. In a plethora of available and new treatments there is no exclusively accepted treatment. The variability and severity of symptoms differ widely between children with ASD, and over time, treatments advised by professionals have varied. Therefore, it is imperative that parents collaborate with health professionals to select and design treatment programmes. Parents of children with ASD can provide significant data on the effects of a wide variety of treatments. Parents need accurate knowledge of the treatments, as they play a vital role in selecting and implementing them. This is done by the health professionals who provide counselling and information to the parents about available and evidence-based treatments. In this study, treatment was defined as any medical or therapeutic intervention for children with ASD. The Diagnostic and Statistical Manual of Mental Disorders - IV was used to describe the characteristics and diagnostic criteria of the children in this study.[1] Two overlapping categories of treatment were used in this study. The first grouping is in accordance with the National Standards Report, which was in operation at the time of the study, where treatment is classified according to its evidence basis as established, emerging or unestablished.[2] Established treatments are treatments for which there is evidence of a beneficial effect.[2] Examples of established treatments include pivotal response training, discrete trial instruction, functional communication training and social stories. Emerging treatments are those supported by one or more studies that suggest the intervention may produce positive outcomes; however, not enough studies have consistently shown these benefits.[2] Treatments such as augmentative and alternative communication, picture exchange communication system (PECS), and developmental, individual differences and relationship-based model (DIR)/floortime, are classified as emerging treatments. As there is no evidence to support the success of the aforementioned treatments, they are referred to as unestablished treatments.

117

SAJCH

Studies do not report ineffective or harmful treatments as researchers tend to change the focus of their studies to treatments which may be effective as soon as they suspect a treatment to be harmful or ineffective.[2] The second category of ASD treatments comprised developmental approaches, naturalistic behavioural approaches and controversial and alternative medicines. Developmental approaches comprise a combination of features, such as focusing interventions on the impairments associated with ASD, e.g. impairments in social interaction, verbal and nonverbal communication, and behaviour. Other features focus on including the family in decision-making and ensuring that interventions are based on the familyâ&#x20AC;&#x2122;s requirements, concerns and priorities, ensuring that intervention occurs within natural environments and providing intervention through natural methods.[3] Examples of developmental approaches include DIR/floortime,[4] and relationship development intervention.[5] Behavioural interventions are based on operant conditioning techniques to facilitate learning. Examples include naturalistic behavioural approaches, which contain features such as following the childâ&#x20AC;&#x2122;s lead, providing intervention in natural environments and the use of highly motivating rewards.[6] Other treatments are based on the principles of applied behavioural analysis (ABA), such as pivotal response training and incidental teaching. Other behavioural approaches include milieu teaching, intensive behaviour intervention or early behaviour intervention. Intensive behaviour intervention, uses ABA techniques to improve socialisation and communication skills.[7,8] Some treatments do not fall into the above categories. They address specific aspects such as social interaction and communication. Examples of these treatments include speech-language therapy (SLT) and augmentative and alternate communication. Controversial and alternative medicines (CAM) are classified as unestablished treatments.[2] Children who participate in CAM interventions usually do so in conjunction with a more conventional therapy, e.g. behavioural and educational services.[9]

OCTOBER 2017 Vol. 11 No. 3

RESEARCH Common types of CAM include sensory motor therapies like sensory integration therapy, and certain other forms like vitamin therapies, weighted vests, chelation therapy and special diets.[10] With the number of interventions discussed above, the role of the parent or caregiver in treatment is vital for continuity in the home context as many autism interventions take place at a therapist’s office or at school. With training, families can extend SLT, occupational, and physical therapies in the home setting, with involvement in goal setting, being part of the team, helping to deliver the programme and being part of the programme evaluation.[11] Research has shown that parent involvement increases the effectiveness of the treatment over time.[12] Parents of children with ASD believed that treatments were costly and inaccessible. It was found that there was a lack of skilled professionals in the area of ASD and this affected parents’ experiences and access to quality treatment.[13] Long waiting lists for accessing services, lack of government support to cover costly treatment and participating in numerous treatments are tiresome.[14] While research has shown that intensive early intervention (<3 years old) has significantly improved outcomes, these services can be extremely expensive.[14-16] Children with ASD are often required to participate in a number of different treatments, which can be time consuming and costly. Relationships with healthcare professionals can be difficult and it has been found that most parents have negative feelings towards their healthcare professional.[17] Parents felt as though healthcare professionals did not share the parents’ concerns in treatment of ASD and did not use ‘the information that parents provided in a collaborative nature’.[18]

Rationale for the study

Parents of children with ASD often manage a complex array of treatments and therapies for their children.[17] Little is known about parents’ satisfaction with the treatments and they may withdraw their child from treatment if no progress is observed.[19,20] In addition, in South Africa (SA) the treatment needs of children with ASD are not met.[21] Further, Deyro et al.[22] found professional advice to be the most influential source in guiding parents in ASD treatment selection for their children. It is therefore important for health professionals to take parents’ views into account when suggesting treatments by sharing knowledge or increasing awareness of the treatments available and understanding the parents’ views to improve the quality of services.[23] Furthermore, there is little research in the area of parents’ views and knowledge of treatment for children with ASD in SA.

Context

This study was conducted in the eThekwini Metropolitan Municipality of KwaZulu-Natal Province (KZN), SA. eThekwini is one of 11 districts in KZN with an estimated population of 3 090 126, the majority of whom speak isiZulu.[24] As ASD is not associated with culture, socioeconomic status or geographical region, the ASD prevalence statistics of 1:68 are taken to apply to SA.[25] In SA, services for ASD are obtained from governmental sources, via the Department of Health, for children between 0 and 6 years old; certain special schools treat children from 6 years to 18 years old, and the private healthcare industry and education services support families with medical aid (insurance) or sufficient funds. There is no set policy or set of guidelines for the identification and management of children with ASD. There is also a paucity of recordkeeping or database management of children with ASD in KwaZuluNatal.

Methods

Study design

A mixed methodology was used with an embedded research design.

Study participants

A non-random purposive sampling technique was used to recruit parents or caregivers of children with a diagnosis of ASD in the eThekwini district in KZN. Parents were recruited via schools or

118

SAJCH

centres who offer placement for children with ASD, via speech-language therapists, and via ASD support group co-ordinators. Children were required to be 5 - 9 years old, have no co-occurring syndromes, have received or were currently receiving treatment for ASD for a minimum of 6 months. The participants included 40 mothers and 6 fathers. The median age of their children was 7 years, 1 month old. Their children fell into different ethnic groups, which included 19 black, 18 indian, 7 white, 1 asian, and 1 coloured. The majority (67%) of the children were English first-language speakers; 15 children were isiZulu first-language speakers. The median age of diagnosis was 3 years, 6 months old.

Data collection

Data were obtained using a questionnaire, which was either paper-based or electronic (sent via e-mail). The first 30 questions pertained to the participants’ knowledge of treatments on an ordinal five-point Likert scale, developed and expanded from a questionnaire by Sansosti.[18] The next 12 questions pertained to the views that participants had about these treatments. A pilot study was conducted with two parents of children with ASD. Minor adjustments were made to the layout of the questionnaire following the pilot study.

Analysis

Quantitative data were analysed using the Statistical Package for Social Sciences (SPSS) version 21.0 (IBM Corp., USA) and thematic analysis. Qualitative data were analysed by reviewing open-ended questions and noting emerging themes.

Reliability, validity and trustworthiness

Reliability and validity were addressed by ensuring that all participants met the inclusion criteria; that data were collected, analysed, represented and interpreted in the same manner, using the same data collection instrument. A thorough description of the research context and any assumptions that were central to the research were recorded so that they could be applied to the greater population. In terms of credibility, a sample that reflected the range of ethnicities and languages of the eThekwini district was obtained.

Ethical issues

Ethical clearance was obtained from the relevant university ethics committee (ref. no. SHSEC 047/13). The principles of ethical clearance, protection from harm, informed consent, confidentiality, the right to privacy and data management were considered.

Results and discussion

Parents’ knowledge of the different treatments

Of all the treatments used, 53.3% were rated as being unfamiliar or just heard of (Fig. 1). A third of the treatments were rated as being familiar, with a limited understanding of the treatment. The parents indicated that they had a practical understanding of only 13.4% of the treatments. The five treatment options that parents rated themselves most knowledgeable in were SLT, physiotherapy, occupational therapy, parentinvolvement and social stories. It was not surprising that SLT was well known, as it addresses core deficits in ASD.[18,26,27] The first three treatments would be offered in special schools and at hospitals in SA, also contributing to parents’ knowledge of them. The data indicated that 78.2% of parents were not informed about the different treatment options prior to use of the treatment (Fig. 2). Families indicated a belief that professionals themselves are not informed about the range of ASD treatments. This could indicate a lack of guidance from professionals regarding which treatments to select.[18]

Parents’ access to treatments

Most of the participants (68.6%) stated that they had difficulty accessing treatment facilities and health professionals specialising in

OCTOBER 2017 Vol. 11 No. 3

RESEARCH 60 53.3%

50 Respondents, %

31.4% Difficult to access

40 33.3%

Accessible

68.6%

30 20 13.4%

Fig. 3. Parents’ access to treatment (N=35).

Unfamiliar

Familiar

Practical understanding

13.2%

Fig. 1. Level of knowlegde of treatments (N=46). 13.2%

78.2%

Subsidised by government Reasonable

73.6%

Costly

Percentage of parents

60 50 Fig. 4. Costs involved in treatment for autistic spectrum disorder (ASD) as perceived by the parents (N=38).

40 30

21.7%

6.4%

20 10 0

19.4%

Yes

Mixed feelings

Fig. 2. Number of parents who were informed of the various treatments prior to the use of the chosen treatment (N=46).

ASD (Fig. 3). These findings correlate with a study of participants from the USA, Canada, Australia, New Zealand, England and Ireland.[14] Challenges included limited numbers of trained health professionals and treatment facilities, long waiting lists and long distances travelled by parents to receive treatment. One parent stated that ‘there is so much on the internet yet so few treatments are available in SA’. Another parent mentioned that ‘it is extremely hard to find one (a treatment) in your mother (home) language’. In SA, the commonly used treatments were traditional therapies such as SLT and occupational therapy, applied behavioural analysis, dietary and biomedical interventions. Many of these treatments are only available in the private sector.

The cost involved in various treatments as perceived by the parents

The data showed that 73.6% of parents felt that the treatments for ASD were costly (Fig. 4). This parallels previous research where parents stated that services for ASD can be extremely expensive and that there is a lack of government support. A parent declared that ‘it is very expensive and there is no government funding for external (out of school) therapy and

119

Poor

SAJCH

Good 74.2%

Fig. 5. Parents’ relationship with healthcare professionals (N=31).

medical aid does not cover most of the costs as autism is not recognised as a primary medical benefit.’ Parents (13.2%) who received government subsidised treatment from schools or hospitals reported that the costs of the treatments were reasonable. Accessing school-based services is consistent with reports by Patten et al. [27] and Mire et al. [28]

Parents’ perspective on the effectiveness of treatments

As found previously, a third of parents believed that a combination of treatments had the best outcomes in their children.[17,26] According to 43.6% of participants SLT was the most effective treatment for their child. For example, one parent reported that SLT was effective as ‘it gave my child confidence and encouragement when it came to attempting new words both in home language isiZulu and English.’ Another parent stated that ‘before speech therapy my son could not talk at all, now he is

OCTOBER 2017 Vol. 11 No. 3

RESEARCH 70

65.2%

3.3 10

Assist in facilitating treatment

Assist with transport

Percentage

Provide support and encouragement Involved in decision-making process

34.8%

76.7

30 20 Fig. 7. Roles family members play in treatment (N=30).

10 0

Yes

Fig. 6. Family involvement in the treatment (N=46).

saying a few words, which I can understand, where before he was only pointing at things.’ Other treatments that parents found effective included applied behaviour analysis (28.2%), occupational therapy (25.6%), augmentative and alternative communication (18%), and drug therapies (12.8%). The use of drug therapies could be linked to the high rate of comorbidity.[29] There were mixed views on the use of CAM treatments and some parents perceived these to be most effective, despite the lack of evidence to support them, e.g. chelation therapy and a gluten-free, casein-free diet. A previous study supported these findings.[30] According to 69% of participants, the reason for ending the use of a particular treatment was because the treatment was ineffective.

Parents’ relationship with service providers

Contrary to previous research, which found that that most parents had negative feelings towards their health professionals, this study found that 74.2% parents had a good relationship with their health professional (Fig. 5).[17] This was due to good communication between the health professionals and the parents, the health professionals’ willingness to assist where needed, health professionals working as a team with the parents, and parents feeling that they were constantly aware of what was occurring in treatment. Parents (6.4%) who viewed their relationship with their healthcare professional as being poor, commented on a lack of communication and reluctance to help, inappropriate recommendations and a lack of individual care. The remaining 19.4% experienced mixed feelings regarding their relationship with professionals.

Families’ involvement in the treatment process

Nearly two-thirds of parents (65.2%) indicated that their families were involved in treatment by facilitating treatment, providing transport, support and encouragement, and in decision-making (Figs 6 and 7).

Study limitations

Limitations include a small sample size due to the low response rate. The results are only representative of one geographic location and all of the information was provided from the parents by self-report. The questionnaire measured the parents’ perception of their knowledge rather than their actual knowledge and the researchers were speech-language therapists which could have impacted parents’ response to SLT-related questions.

Conclusion

Parents of children with ASD in eThekwini have some degree of knowledge of the different treatments available. The findings

120

SAJCH

highlighted that parents perceived treatments as being expensive, not easily accessible, and viewed some treatments as more effective than others. Parents were not informed of options pre treatment, or did not receive guidance regarding available evidence-based treatments and facilities. In this study, most parents stated that they had a good relationship with their healthcare professional. The parents also ranked their knowledge of SLT as highest compared with other treatments, reporting that it was the most effective treatment. These findings should motivate healthcare professionals involved in ASD diagnosis and management to advocate the provision of information and guidance regarding the range of ASD treatments, and to improve access to appropriate facilities through public service treatment provision. Professionals should acknowledge the use of combination treatments including established, emerging, unestablished or controversial treatments. Professionals should be aware of information overload and factor in overlapping treatments, e.g. SLT and picture exchange communication system (PECS), treatments for comorbid conditions, and psychopathology. Ultimately, the type of treatment is affected by the child’s age and by the cultural and linguistic background of the family e.g. internet access and the availability of treatments in the child’s first language, which is complicated in the SA context. Therefore, it is imperative that healthcare professionals engage in continuing professional education to update their knowledge on the latest evidence-based treatments Acknowledgements. None. Author contributions This article is based on an Honours study conducted by VW, SG, NS and KW, who were supervised by JP, with conceptual and writing input from SK. Funding. None. Conflicts of interest. None. 1. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders DSM-IV-TR (text revised). Arlington: APA, 2000. 2. Wilczynski S, Green G, Ricciardi J, et al. National Standards Report: The National Standards Project: Addressing the Need for Evidence-based Practice Guidelines for Autism Spectrum Disorders. Randolph: The National Autism Center, 2009. 3. Wetherby AM, Woods JU. Developmental approaches to treatment. In: Chawarska K, Klin A, Volkmar FR, eds. Autism Spectrum Disorders in Infants and Toddlers: Diagnosis, Assessment, and Treatment. London: The Guilford Press, 2008:170-206. 4. Peloquin LJ. Connecting with your child through DIR: The Developmental, Individual Difference, Relationship Based Approach. In V. Zysk, ed. The Best of Autism Digest Magazine volume 1. Texas: Future Horizons Inc., 2005. 5. Gutstein SE. Autism Aspergers, Solving the Relationship Puzzle: A New Developmental Program that Opens the Door to Lifelong Social and Emotional Growth. Texas: Future Horizons Inc., 2000. 6. Koegel LK, Koegel RL, Fredeen RM, Gengoux GW. Naturalistic behavioral approaches to treatment. In: Chawarska K, Klin A, Volkmar FR, eds. Autism Spectrum Disorders in Infants and Toddlers: Diagnosis, Assessment, and Treatment. London: The Guilford Press, 2008;207-242.

OCTOBER 2017 Vol. 11 No. 3

RESEARCH 7. Skinner BF. The experimental analysis of operant behavior. In: Rieber RW and Salzinger K, eds Psychology: Theoretical-historical Perspectives. New York: Academic Press, 1980:190-202. 8. Prior M. Intensive behavioural intervention in autism. J Paediatr Child Health 2004;40(9‐10):506-507. https://doi.org/10.1111/j.1440-1754.2004.00453.x 9. Smith TR, Wick JE. Controversial treatments. In: Chawarska K, Klin A, Volkmar FR, eds. Autism Spectrum Disorders in Infants and Toddlers: Diagnosis, Assessment, and Treatment. London: The Guilford Press, 2008:243273. 10. Robledo SJ, Ham-Kucharski D. The Autism Book: Answers to Your Most Pressing Questions. New York: Penguin Group, 2005:4-7. 11. Prior M, Roberts JM, Rodger S, Williams K, Sutherland R. A Review of the Research to Identify the Most Effective Models of Practice in Early Intervention for Children with Autism Spectrum Disorders. Australian Government Department, FaHCSIA, 2011. 12. Oono IP, Honey EJ, McConachie HH. Parent‐mediated early intervention for young children with autism spectrum disorder (ASD). Cochrane Rev J 2013;8(6):2380-2479 https://doi.org/10.1002/ebch.1952 13. Dymond SK, Gilson CL, Myran SP. Services for children with autism spectrum disorders. J Disabl Policy Stud 2007;18(3):133-147. https://doi.org/10.1177/10 442073070180030201 14. Goin-Kochel RP, Myers BJ, Mackintosh VH. Parental reports on the use of treatments and therapies for children with autism spectrum disorders. Res Autism Spectr Disord 2007;1(3):195-209. https://doi.org/10.1016/j. rasd.2006.08.006 15. Mackintosh VH, Goin-Kochel RP, Myers BJ. “What do you like/dislike about the treatments you’re currently using?” A qualitative study of parents of children with autism spectrum disorders. Focus Autism Dev Disabl 2012;27(1):51-60.. https://doi.org/10.1177/1088357611423542 16. Smith T. Outcome of early intervention for children with autism. Clin Psychol Sci Pract 1999;6(1):33-49. https://doi.org/10.1093/clipsy.6.1.33 17. Cimera RE, Cowan RJ. The costs of services and employment outcomes achieved by adults with autism in the US. Autism 2009;13(3):285-302. https:// doi.org/10.1177/1362361309103791 18. Sansosti FJ, Lavik KB, Sansosti JM. Family experiences through the autism diagnostic process. Focus Autism Dev Disabl 2012;27(2):81-92. https://doi. org/10.1177/1088357612446860

121

SAJCH

19. Lanners R, Mombaerts D. Evaluation of parents' satisfaction with early intervention services within and among European countries: Construction and application of a new parent satisfaction scale. Infants Young Children 2000;12(3):61-70. https://doi.org/10.1097/00001163-200012030-00009 20. Rodger S, Keen D, Braithwaite M, Cook S. Mothers’ satisfaction with a home based early intervention programme for children with ASD. J Appl Res Intellect Disabl 2008;21(2):174-182. https://doi.org/10.1111/j.1468-3148.2007.00393.x 21. Bowker A, D’Angelo NM, Hicks R, Wells K. Treatments for autism: Parental choices and perceptions of change. J Autism Dev Disord 2011;41(10):13731382. https://doi.org/10.1007/s10803-010-1164-y 22. Deyro MC, Simon EW, Guay J. Parental awareness of empirically established treatments for autism spectrum disorders. Focus Autism Dev Disabl 2016;31(3):184-195. 10.1177/1088357614559210 23. Shyu YI, Tsai JL, Tsai WC. Explaining and selecting treatments for autism: Parental explanatory models in Taiwan. J Autism Dev Disord 2010;40(11):1323-1331. 24. Statistics South Africa (SSA). Census 2011 Statistical release. Pretoria: SSA, 2012:17. 25. DMNSY Principal Investigators. Prevalence of autism spectrum disorder among children aged 8 years – Autism and developmental disabilities monitoring network, 11 sites, United States, 2010. MMWR 2014;63(2):1-21. 26. Green VA, Pituch KA, Itchon J, Choi A, O’Reilly M, Sigafoos J. Internet survey of treatments used by parents of children with autism. Res Dev Disabl 2006;27(1):70-84. https://doi.org/10.1016/j.ridd.2004.12.002 27. Patten E, Baranek GT, Watson LR, Schultz B. Child and family characteristics influencing intervention choices in autism spectrum disorders. Focus Autism Dev Disabl 2013;28(3):138-46. https://doi.org/10.1177/1088357612468028 28. Mire SS, Gealy W, Kubiszyn T, Burridge AB, Goin-Kochel RP. Parent perceptions about autism spectrum disorder influence treatment choices. Focus Autism Dev Disabl 2015:1-14. https://doi.org/10.1177/1088357615610547 29. Matson JL, Adams HL, Williams LW, Rieske RD. Why are there so many unsubstantiated treatments in autism? Res Autism Spectr Disord 2013;7(3):466474. https://doi.org/10.1016/j.rasd.2012.11.006 30. Senel HG. Parents’ views and experiences about complementary and alternative medicine treatments for their children with autistic spectrum disorder. J Autism Dev Disord 2010;40(4):494-503. https://doi.org/10.1007/s10803-0090891-4

OCTOBER 2017 Vol. 11 No. 3

RESEARCH

This open-access article is distributed under Creative Commons licence CC-BY-NC 4.0.

Toilet training practices in Nigerian children A U Solarin,1 MPhil, Cert Nephrol (SA) Paed, FWAC Paed, MBBS; O A Olutekunbi,2 FWAC Paed, MBBS; A D Madise-Wobo,1 FMC Paed, MSc, MBBS; I Senbanjo,1 FWAC Paed, MSc, MB ChB 1 2

Department of Paediatrics, Lagos State University Teaching Hospital, Lagos, Nigeria Department of Paediatrics, Gbagada General Hospital, Lagos, Nigeria

Corresponding author: A U Solarin (asolar234@gmail.com) Background. This study reports on toilet training with a focus on the effect of age, methods used, and factors that can affect urinary incontinence in Nigerian children. Methods. This was a cross-sectional hospital-based study carried out in public and private hospitals in South-Western Nigeria. A questionnaire was used to obtain information about toilet training practices from 350 adults, who toilet trained 474 children. Results. The adults had previously toilet trained children 1 - 18 years old. In this study, toilet training commenced at ≤12 months, during the day and night in 40.6% and 33.4% of children, respectively. Of the 350 parents/guardians, 141 (47.7%) commenced toilet training by waking children from their afternoon nap. The most common method was allowing the child to urinate at fixed time intervals, while the least common was a reward/punishment system. Furthermore, age was considered as the most common indicator to commence toilet training. For 36.9% of the children, training lasted 1 - 6 months. Daytime continence was achieved by 33.4% of children at ≤12 months old, and night-time continence was achieved in 29.7% of children between 12 and 18 months old. By 30 months, 91.1% and 86.9% had attained day- and night-time continence, respectively, and only 8.6% of the children were incontinent at night. Conclusion. Assisted infant toilet training is still practised among Nigerian parents despite the influence and the trends in the developed countries. The age at initiation and completion of toilet training was lower than those reported for developed countries. S Afr J Child Health 2017;11(3):122-128. DOI:10.7196/SAJCH.2017.v11i3.1287

Toilet training is an important aspect of early childhood developmental milestones, and it may be quite challenging. It involves a complex integration of neurological, muscular, and behavioural mechanisms.[1,2] Toilet training is also influenced by physiological, psychological and sociocultural factors.[3,4] Failure of toilet training may result in significant physical and psychological consequences like a sense of failure through partial loss of autonomy.[5] Different opinions on the optimal time to initiate toilet training cut across different cultures and beliefs.[2,6] Five to six decades ago, toilet training in Western countries was initiated at an earlier age compared with the current initiation age.[2,4] For example, in the USA, the median age of initiating toilet training ranged from 25 to 27 months in the 80s; it had increased to n (SD) of 36.8 (6.1) months in 2003.[7] In a report, toilet training before the age of 18 months had already started ~60 years ago, while modern parents usually start the training after 18 months.[4] One of the reasons for the later initiation age of toilet training may be because of the recommendations by the American Academy of Pediatrics (AAP). According to the AAP, starting toilet training before the age of 2 years is not recommended because the readiness skills and physical abilities required only develop between age 18 and 30 months.[8] In addition to the AAP’s recommendation, the introduction of disposable diapers, more efficient laundry facilities and both parents working may contribute to the later age of commencing toilet training.[9] There are arguments against early and late initiation of toilet training. A delay in toilet training was considered to be related to increased frequency of dysfunctional voiding in children.[4,10] For example, in the United Kingdom, a cohort study showed that when toilet training commenced after 24 months of age there was an association with diurnal enuresis and delayed acquisition of bladder control.[11] There were also fears that toilet training at an earlier age may result in voiding dysfunction.[12] However, Duong et al.[13] dispelled that fear in a study that investigated early initiation of toilet training in Vietnamese girls. In that study, it was noted that toilet training that was initiated at <12 months did not result in voiding dysfunction.

122

SAJCH

There are different types of toilet training methods available. In the Western communities two predominant methods are used; the ‘childoriented’ method of Brazelton et al.[14] and the Azrin and Foxx method.[15] Other methods include variations of operant conditioning, assisted infant toilet training, and the Spock method.[9] In 1962 Brazelton et al. [14] described the child-oriented approach. It is based on the principle that the child must gently but systematically be encouraged to experiment toileting behaviour. The Azrin and Foxx[15] method, on the other hand, is more intensive and structured. However, the method has been associated with reported side-effects of temper tantrums, hitting and avoidance behaviour. Less known, but applied for many centuries in China, India, Africa, South and Central America, is assisted infant toilet training. The caregivers play a key role by observing the child’s evacuation signals and when they occur, place them in a special position. It is important to note that although recent guidelines of the AAP[16,17] and the Canadian Pediatric Society[18] are based on the child-oriented approach[9] and Azrin and Foxx method,[15] no efficacy studies or randomised controlled trials have been conducted. Recent studies have identified incorrect toilet training as being predictive of persistent urinary symptoms, such as urinary incontinence, enuresis, recurrent urinary tract infection and childhood constipation.[19–21] Urinary incontinence has an impact on both the child and family. It affects the self-esteem, interpersonal relationships and school performance of the children, as described in detail by Mota and Barros[22] in a previous review. Primary enuresis is related to the presence of nocturnal polyuria, difficulties waking from sleep and reduced bladder capacity, whereas secondary enuresis is more related to urinary infections, diabetes mellitus and emotional disorders.[22-25] Coercive or permissive methods of toilet training may be associated with the development of enuresis and encopresis.[14,26,27] In an analysis of the prevalence of enuresis according to the age of acquisition of daytime urinary continence, Chiozza et al.[28] observed that, among children who achieved bladder and bowel control after 36 months, the prevalence of enuresis was 17.1%, whereas children who achieved control before 25 months and between 25 and 36 months had prevalence rates of 2.7% and 5.8%, respectively. These findings

OCTOBER 2017 Vol. 11 No. 3

RESEARCH suggest that starting toilet training later may promote enuresis. Certain interventions to treat enuresis employ techniques of toilet retraining and provide guidance on regularity of elimination habits.[29,30] In Africa there is limited literature focusing specifically on the age aspect of toilet training practices. This study aimed to report on methods at the time of initiation and the time of completion of toilet training, as well as any relationships among these factors and enuresis in our setting.

Method

Study design

This was a cross-sectional hospital-based study designed to answer three major research questions: • At what age do parents/caregivers commence toilet training on their children/wards? • What is the age at attainment of day- and night-time urinary continence? • What is/are the toilet training method/s used by parents/caregivers on their children?

Study setting

The study was carried out in private and government hospitals in two states in South-Western Nigeria. The participants from the public hospital were residents of Lagos State, while those from the private hospital were residents of Ogun State. Lagos State is a densely populated cosmopolitan urban setting, while Ogun State is suburban and less densely populated. The study was conducted between April and July 2016.

Data acquisition

A tested questionnaire was used to obtain information from the participants. The questionnaire was adapted from a previous questionnaire which had been validated and used by Bakker and Wyndaele.[4]

Participants

The participants were parents and caregivers who presented to the hospital for various reasons. The aim of the study was explained to the parents/ caregivers of the children and those who gave verbal consent were included in the study. The parents were recruited consecutively until the minimum sample size was achieved. The socioeconomic classes of the participants were documented using the Oyedeji classification.[31] Those in social classes 1 and 2 were regarded as upper class, those in class 3 as middle class, while those in classes 4 and 5 were in a lower socioeconomic class. The following participants were excluded: parents of children with neurological problems, e.g. spinal dysraphism, hydrocephalus, and cerebral palsy, as well as those who had children with urogenital abnormalities.

Variables

The tested questionnaire was used to obtain relevant information from the participants. The outcome variables were: age at attainment of day- and night-time urinary continence; toilet training methods used; duration of toilet training; and factors which affected the age at attainment of urinary continence. A potential challenge was failure to understand the questions in the questionnaire. To avoid this, the questionnaire was pretested in a different subset of caregivers to avoid ambiguity.

Data measurement

The source of the data was from the questionnaire completed by the participants. The variables that were normally distributed were summarised with mean and standard deviation. For description of the first two outcome variable outcomes, mean and mode were used to analyse those variables. The χ2 test was used to analyse the relationship between variables to ascertain which factors affected the age of attainment of urinary continence. An inter-group comparison was also done using χ2 test; p<0.05 indicated statistical significance.

Table 1. Sociodemographic characteristics of parents (N=350) Type of centre Public (n=250), n (%)

Private (n=100), n (%)

Total, n (%)

p-value

Father’s level of education None Primary Secondary BSc/HND Postgraduate

3 (1.2) 9 (3.6) 51 (20.4) 176 (70.4) 11 (4.4)

1 (1.0) 3 (3.0) 24 (24.0) 64 (64.0) 8 (8.0)

4 (1.1) 12 (3.4) 75 (21.4) 240 (68.6) 19 (5.4)

0.616

Mother’s level of education Primary Secondary BSc/HND Postgraduate

3 (1.2) 67 (26.8) 174 (69.6) 6 (2.4)

2 (2.0) 27 (27.0) 66 (66.0) 5 (5.0)

5 (1.4) 94 (26.9) 240 (68.6) 11 (3.1)

0.574

Father’s occupation None Civil servant Artisan Businessman Student Professional

6 (2.4) 64 (25.6) 41 (16.4) 85 (34.0) 0 (0.0) 54 (21.6)

1 (1.0) 20 (20.0) 11 (1.0) 31 (31.0) 0 (0.0) 37 (37)

7 (2.0) 84 (24.0) 52 (14.9) 116 (33.1) 0 (0.0) 91 (26.0)

0.621

Mother’s occupation None Civil servant Artisan Businesswoman Student Professional

12 (4.8) 81 (32.4) 16 (6.4) 101 (40.4) 9 (3.6) 31 (12.4)

3 (3.0) 18 (18.0) 18 (18.0) 21 (21.0) 2 (2.0) 38 (38.0)

15 (4.3) 99 (28.2) 34 (9.7) 122 (34.9) 11 (3.1) 69 (19.7)

0.432

HND = higher national diploma.

123

SAJCH

OCTOBER 2017 Vol. 11 No. 3

RESEARCH Table 2. Method and age of commencement of toilet training including types of diapers and reasons for commencement (N=350)

Number of respondents, n

400

88.3%

300

200

100

4.3% 0

Parent

4.9%

Grandparent

Nanny

2.6% Relations

Fig. 1. Person in charge of toilet training (N=350).

Number of respondents, n

150

100

0 On demand His/her of the age child

School

Comment of others/ family

Season

When he/ Saving she was dry diaper during the afternoon nap

Could not remember

Reason

Fig. 2. Reasons for initiating toilet training during the day (N=350).

Results

Sociodemographic characteristics of the study participants

A total of 350 parents/guardians participated in the study. The sociodemographic characteristics of the caregivers who participated in the study were defined (Table 1). The mean (SD) ages of the caregivers were 42 (6.80) and 36.41 (7.15) for the males and females, respectively. The majority of the parents (284 (81.9%)), were Christian, while 66 (18.1%) practised Islam. Most of the parents had a tertiary level of education and there was no significant difference in the level of education of the parents in the different hospitals (p>0.05). The majority of the parents were either businessmen or businesswomen and professionals, with no significant difference in the occupations of the parents in the different hospitals (p>0.05). The social class of the parents in both group of hospitals was social class 4 (lower social class based on educational level and occupation of both parents). All of the study participants had children. Of the 350 parents/ caregivers who participated in the study, information was obtained for 474 children. A total of 22 (6.3%) participants had grandchildren. The children were between 1 and 18 years old. The majority of the

124

SAJCH

Variable

n (%)

How grandparent participated in toilet training By giving advice In keeping the children during the day During a stay

275 (78.6) 39 (11.1) 36 (10.2)

Age of commencement of day-time toilet training (months) ≤12 13 - 18 19 - 24 25 - 30 ≥31 Yet to start Do not remember

142 (40.6) 80 (22.9) 74 (21.1) 31 (9.0) 12 (3.4) 0 11 (3.1)

Age of commencement of toilet training at night (months) ≤12 13 - 18 19 - 24 25 - 30 ≥31 Do not remember Yet to start

124 (35.4) 79 (22.6) 86 (24.6) 30 (8.6) 21 (6.0) 7 (2.0) 3 (0.9)

Dryness during afternoon nap Yes No Do not remember

153 (43.7) 119 (34.1) 78 (22.3)

Type of diaper used Cotton Flannel Disposable Plastic pants

25 (6.6) 3 (0.9) 317 (83.7) 4 (1.1)

Method used Urinate at fixed time Remove the diaper On demand of the child Reward Punish Imitation of parent or older sibling Do not remember

163 (46.6) 105 (30.0) 47 (13.4) 8 (2.2) 8 (2.3) 9 (2.6) 10 (2.9)

participants (67.7%, n=237) had children who were ≤5 years old, 30.6% (n=107) were between 5 and 10 years old, 19.1% (n=67) were 10 - 15 years old, and 11.7% (n=41) were >15 years old.

Method and age at commencement of toilet training

A total of 309 (88.3%) of the respondents participated in the toilet training of their own children. The remaining respondents left the toilet training to either a grandparent, caregiver or family relative (Fig. 1). A total of 275 (78.6%) participated in toilet training by giving advice while the others participated in the toilet training of their grandchildren when the children were in their custody during the day or during a brief visit. Table 2 shows the method and age at commencement of toilet training of the children. The parents considered the age of the child as the most common reason for commencement of toilet training (Fig. 2). Toilet training was commenced at ≤12 months in the majority of the children during the day and night at 40.6% and 35.4%, respectively. A total of 153 (43.7%) of the respondents commenced toilet training

OCTOBER 2017 Vol. 11 No. 3

RESEARCH Table 3. Age of attainment of continence, location and duration of training (N=350)

Table 4. Comparison between public and private hospitals in relation to toilet training

n (%)

Variable Duration of toilet training for continence (months) ≤1 1-6 7 - 12 ≥12 Still wet

Types of tertiary centres

55 (15.7) 129 (36.9) 84 (24.0) 75 (21.4) 7 (2.0)

Where the training mainly took place At home With grandparent In reception class In crèche

310 (88.6) 8 (2.3) 7 (2.0) 25 (7.1)

Age of attainment of night-time continence (months) <12 12 - 18 19 - 30 31 - 60 >61 Still wet

70 (20.0) 104 (29.7) 84 (24.0) 46 (13.1) 16 (4.6) 30 (8.6)

Age of attainment of day-time continence (months) <12 12 - 18 19 - 24 25 - 30 ≥31 Do not remember Yet to start

117 (33.4) 68 (19.4) 104 (29.7) 30 (8.6) 21 (6.0) 7 (2.0) 3 (0.9)

during an afternoon nap. Disposable diapers were the most commonly used diapers. The most common method of toilet training was allowing the child to urinate at fixed time intervals (46.6%) and the least common method was by either reward or punishment.

Attainment of continence

The duration of toilet training was 1 - 6 months for 36.9% (n=129) of the children; the shortest duration of training was <1 month. Training took place at home in 88.6% (n=310) of the children, with 7.1% (n=25) at the crèche. Continence was described for day- and night-time. The modal age at attainment of daytime continence in the wards of the participants was <12 months of age, closely followed by children age 19 - 24 months (33.4 and 29.7%, respectively). By 30 months of age 91.1% (n=319) of the children had attained daytime continence. Contrary to the daytime continence, 29.7% (n=104) attained night-time continence between 12 and 18 months of age. Similar to the finding of the daytime continence, by 30 months of age the majority of the children, 86.9% (n=304) had attained night-time continence. A total of 4.6% (n=16) achieved night time continence after 5 years of age while 30 (8.6%) had not achieved continence at the time of this study. Among the 30 children who were still incontinent at night, the age at commencement of their toilet training was 18 - 24 months (13.3%), >30 months (13.3%), <1 year (16.7%), 12 - 18 months (23.3%) and 24 - 30 months (33.3%) Table 3.

Comparison between private and public hospitals

The age at attainment of night-time continence, place of training and who was in charge of the training were compared for private and public hospitals. There were no significant differences in results between private and public hospitals (p>0.61). However, there was a significant difference in the age of attainment of night-time continence (p<0.05) in both study centres. While most of the children from the public hospital

125

SAJCH

Public (n=250), n (%)

Private (n=100), n (%)

Total, n (%)

p-value

Who was in charge of toilet training? Parent

217 (88.8) 92 (92.0)

Grandparent

11 (4.4)

4 (4.0)

309 (88.3) 0.427

15 (4.3)

Caregiver

15 (6.0)

2 (2.0)

17 (4.9)

Relations

7 (2.8)

2 (2.0)

9 (2.6)

≤12

57 (22.8)

13 (13.0)

70 (20.0)

13 - 18

82 (32.8)

22 (22.0)

104 (29.7) 0.002*

19 - 24

47 (18.8)

37 (37.0)

84 (24.0)

Age at attainment of night-time continence (months)

25 - 30

31 (12.4)

15 (15.0)

46 (13.1)

≥31

14 (5.6)

2 (2.0)

16 (4.6)

Still wet

19 (7.6)

11 (11.0)

30 (8.6)

Place of toilet training At home

219 (87.6) 91 (91.0)

310 (88.6)

With grandparent

7 (2.8)

1 (1.0)

8 (2.7)

In reception

6 (2.4)

1 (1.0)

7 (2.0)

At crèche

18 (7.2)

7 (7.0)

25 (7.1)

92 (36.8)

32 (32.0)

124 (35.4)

0.613

Age when toilet training commenced (months) ≤12 13 - 18

59 (23.6)

20 (20.0)

79 (22.6)

19 - 24

55 (22.0)

31 (31.0)

86 (24.6)

25 - 30

20 (8.0)

10 (10.0)

30 (8.6)

≥31

17 (6.8)

4 (4.0)

21 (6.0)

Still wet

4 (1.6)

3 (3.0)

7 (2.0)

Do not remember

3 (1.2)

3 (0.9)

0.414

*Statistically significant.

achieved continence between 12 and 18 months, most of the children at the private hospital attained continence at a more advanced age of 18 - 30 months. Also, the age at which toilet training was commenced in the public hospital was mainly ≤12 months. The children from the private hospital commenced training at either ≤12 months or 19 - 24 months (Table 4).

Test of associations

The test of association was computed for variables such as who was in charge of toilet training, age at which the child commenced training, method used and duration of toilet training for continence against age at

OCTOBER 2017 Vol. 11 No. 3

RESEARCH Table 5. Test of association among different variables and age of attainment of day-time continence Age of attainment of day time continence (months) <12

12 - 18

19 - 30

31 - 60

>61

Still wet

Yet to start

Age of commencement of day-time toilet training (months) ≤12

114 (97.4) 26 (38.2)

1 (1.0)

1 (3.3)

13 - 18

42 (61.8)

29 (27.9)

2 (6.7)

7 (100)

19 - 24

66 (63.5)

3 (10.0)

5 (23.8)

25 - 30

6 (5.8)

23 (76.7)

2 (9.5)

≥31

9 (42.9)

3 (100)

Do not remember

3 (2.6)

2 (1.9)

1 (3.3)

5 (23.8)

36 (30.8)

12 (17.6)

1 (1.0)

4 (19.0)

2 (66.7)

1-6

81 (69.2)

34 (50.0)

2 (1.9)

7 (33.3)

4 (57.1)

1 (33.3)

7 - 12

9 (13.2)

15 (14.4)

28 (93.3)

10 (47.6)

>12

13 (19.1)

82 (78.8)

2 (6.7)

Still wet

4 (3.8)

3 (42.9)

χ2=781.377; p=0.000

Duration of toilet training for continence (months)

χ2=467.232; p=0.000

Who was in charge of training Parents

107 (91.5) 63 (92.6)

90 (86.5)

19 (63.3)

21 (100)

6 (85.7)

3 (100)

Grandparents

6 (5.1)

1 (1.5)

7 (6.7)

1 (14.3)

Caregiver

4 (3.4)

3 (4.4)

7 (6.7)

3 (10.0)

Relatives

1 (1.5)

0 (0.0)

8 (26.7)

2 χ =88.611; p=0.000

Method of toilet training Urinate at fixed intervals

75 (64.1)

21 (30.9)

54 (51.9)

2 (6.7)

8 (38.1)

2 (28.6)

1 (33.3)

Remove diaper

19 (16.2)

35 (51.5)

41 (39.4)

2 (6.7)

5 (23.8)

3 (42.9)

On demand of the child

23 (19.7)

4 (5.9)

2 (1.9)

8 (26.7)

8 (38.1)

2 (28.6)

Reward

7 (23.3)

1 (33.3)

Punishment

4 (5.9)

4 (3.8)

Imitation of parent/older sibling

4 (5.9)

5 (16.7)

Do not remember

0 (0.0)

3 (2.9)

6 (20.0)

1 (33.3)

2 χ =248.912; p=0.000

continued...

attainment of both day- and night-time continence. It was observed that there was significant association among all those variables tested against age at attainment of day- and night-time continence (p≤0.05). Mothers and grandparents were in charge of toilet training in the majority of children who were dry during the night before 12 months of age. Toilet training was commenced in children within 18 months of birth in the majority of children who were continent before 30 months. The most common method used in children who were continent before 18 months, was urinating at fixed time points. A toilet training period of 1 - 6 months was noticed most among children who became continent at night before 18 months (Table 5).

Discussion

There are limited reports on toilet training in children in Africa. This study was conducted to describe the toilet training practices and factors, if any, which influenced the age of continence in Nigerian children. In this study, toilet training was initiated at ≤12 months in most of the subjects. The findings in this regard contradict reports from previous studies where toilet training started at a later age of 18 - 24 months.[4,7]

126

SAJCH

The reason for this difference was not obvious. Possible explanations may be the difference in race, culture and beliefs among the subjects. This is because the present study included an African population group, while previous studies had been conducted in developed countries. It has been shown in a previous report that, among other factors, race affects the age at which toilet training is initiated.[2,32] To our knowledge, there are no reports on the association, if any, between race and the age at initiation of toilet training. Different methods have been used to toilet train children. In the present study, children were asked to urinate at a fixed time. This involved removing the diapers and allowing for urination at fixed times. This bears semblance with the assisted infant toilet training method – a parent-oriented training method.[33] This method existed centuries before the Western methods were described. It is popular among populations in China, India, South and Central America and less popular in North America and European countries. De Vries et al.[6] studied this method among the Digo tribe in East Africa and noted that it was effective and resulted in early achievement of continence. The implication of this finding is that the advent of westernisation has not affected the age-old traditional method of toilet training in Nigeria.

OCTOBER 2017 Vol. 11 No. 3

RESEARCH Table 5. (continued) Test of association among different variables and age of attainment of night-time continence Age of attainment of night-time continence (months) <12

12 - 18

19 - 30

31 - 60

>61

Still wet

≤12

64 (91.4)

58 (55.8)

1 (1.2)

1 (2.2)

13 - 18

46 (44.2)

19 (22.6)

7 (15.2)

7 (23.3)

19 - 24

63 (75.0)

11 (23.9)

10 (62.5)

2 (6.7)

25 - 30

16 (34.8)

6 (37.5)

8 (26.7)

Age of commencement of toilet training at night (months)

≥31

11 (23.9)

10 (33.3)

Do not remember

6 (8.6)

1 (1.2)

Yet to start

3 (10.3)

<1 month

24 (34.3)

23 (22.1)

1 (1.2)

2 (4.3)

5 (31.2)

1 - 6 months

46 (65.7)

33 (31.7)

29 (34.5)

7 (43.8)

14 (46.7)

7 - 12 months

48 (46.2)

22 (26.2)

4 (8.7)

4 (25.0)

6 (20.0)

>12 months

28 (33.3)

40 (87.0)

7 (23.3)

Still wet

4 (4.8)

3 (10.0)

Parents

58 (82.9)

93 (89.4)

77 (91.7)

35 (76.1)

16 (100)

30 (100)

Grandparents

8 (11.4)

7 (6.7)

χ2=533.445; p=0.000

Duration of toilet training for continence

χ2=273.882; p=0.000

Who was in charge of training?

Caregiver

4 (5.7)

3 (2.9)

7 (8.3)

3 (6.5)

Relatives

1 (1.0)

8 (17.4)

28 (40.0)

51 (49.0)

53 (63.1)

9 (19.6)

8 (50.0)

14 (46.7)

χ2=70.181; p=0.000

Method of toilet training Urinate at fixed intervals Remove diaper

19 (27.1)

41 (39.4)

23 (27.4)

8 (17.4)

0 (0.0)

14 (46.7)

On demand of the child

23 (32.9)

4 (3.8)

0 (0.0)

10 (21.7)

850.(0)

2 (6.7)

Reward

8 (17.4)

Punishment

4 (3.8)

4 (4.8)

Imitation of parent or older sibling

4 (3.8)

5 (10.99)

Do not remember

4 (4.8)

6 (13.0)

The duration of toilet training in most of the respondents in the present study was 1 - 6 months. The finding in this regard is incongruent with report from previous studies.[2,4,34] The general trend noted from previous reports was that training is completed within 1 year of commencement. It was therefore not surprising that 70% of the children were continent within 1 year of commencement of training in the present study. In the present study, the age at attainment of day- and night-time continence was <12 and 12 - 18 months, respectively. This finding was contrary to reports from previous studies. Daytime continence was achieved at 32.5 and 35 months in boys and girls, respectively, in the report by Schum et al.[35] Also, a recent report has shown that only 40 - 60% of children completed training by 36 months. The general trend noted from previous reports was that complete continence was attained at 30 - 40 months – much later than the age reported in the present study. A possible reason may be because of the earlier timing of initiation of toilet training and the method used, as noted earlier. It has been observed that age at attainment of day- or night-time continence is dependent on factors such as timing of initiation and method used. Even if the age at

127

SAJCH

χ2=185.931; p=0.000

attainment of continence in the present study is earlier than in previous reports, the finding in this regard is not far-fetched. This is because bladder development and maturation occur at about 18 months of age, which was the time reported by the majority of the respondents in the present study. Attempts have been made to document factors which affect toilet training from previous studies. In the current study, factors which affected the time of attainment of continence were the individuals responsible for toilet training, time of initiation of toilet training and the method of toilet training. Most of the children who achieved continence within 1 year of commencement were trained by their mothers. This is not surprising because the mothers are more patient or tolerant compared with the fathers or any other relatives/persons. Children who commenced toilet training within 1 year of life also achieved continence earlier. It has been shown that the duration of toilet training is related to the age at initiation of training.[36] Children who commenced training earlier achieved continence earlier.[4,36] But this is not always the case, because some children may develop problems along the line of training and if this not handled properly by

OCTOBER 2017 Vol. 11 No. 3

RESEARCH the guardian/parent, it may hamper training and prolong the duration of training and time to achieve continence. Children who were trained by urinating at fixed intervals achieved continence earlier. There are no randomised studies on the assisted infant training method, but studies that have been done with the other training methods have shown that those methods affected the time of continence.[33] There are, however, conflicting reports in this regard. The present study compared findings from caregivers who attended private and public tertiary hospitals. The sociodemographic characteristics were similar between the respondents of both hospitals. It was therefore not surprising that there were similar findings in both children.

Conclusion

Nigerian children are being toilet trained at an earlier age compared with children in developed countries. Also, the age-old traditional method of toilet training is still practised in Nigeria despite the influence of the Western world on our way of life. The age of attaining continence is also lower than for developed countries. Acknowledgements. Sincere appreciation to the parents of the children. Author contributions. AUS developed the concept, literature review, data collection and analysis and discussion. OAO did data collection and analysis. ADM helped with the literature review, analysis and discussion. IS contributed to the concept and review of manuscript. Funding. None. Conflicts of interest. None. 1. Brazelton TB, Christopher E, Frauman A, et al. Instruction, timeliness and medical influences affecting toilet training. Pediatrics 1999;103(6):1353-1358. 2. Yang SS-D, Zhao L-L, Chang S-J. Early initiation of toilet training for urine was associated with early urinary continence and does not appear to be associated with bladder dysfunction. Neurourol Urodyn 2011;30(7):1253-1257. https://doi. org/10.1002/nau.20982 3. Hinde M, Hjertonsson M, Broberg A. [Low self-esteem of children with enuresis. Mental and social health compared in different groups]. Lakartidningen 1995;92(36):3225-3229. 4. Bakker E, Wyndaele J. Changes in the toilet training of children during the last 60 years : The cause of an increase in lower urinary tract dysfunction ? BJU Int 2000;86(3):248-252. https://doi.org/10.1046/j.1464-410x.2000.00737.x 5. Kinservik M, Friedhoff M. Control issues in toilet training. Pediatr Nurs 2000;26(3):267-272. 6. De Vries M, De Vries M. Cultural relativity of toilet training readiness: A perspective from East Africa. Pediatrics 1977;60(2):170-177. 7. Blum NJ, Taubman B, Nemeth N. Why is toilet training occurring at older ages? A study of factors associated with later training. J Pediatr 2004;145(1):107-111. https://doi.org/10.1016/j.jpeds.2004.02.022 8. Toilet Training Guidelines: Parents – The Role of the Parents in Toilet Training. Pediatrics 1999;103(6 Pt 2):1362-1363. 9. Vermandel A, Van Kampen M, Van Gorp C, Wyndaele J-J. How to toilet train healthy children? A review of the literature. Neurol Urodyn 2008;27(3):162-166. https://doi.org/10.1002/nau.20490 10. Hellström A. Influence of potty training habits on dysfunctional bladder in children. Lancet 2000;356(9244):1787. https://doi.org/10.1016/s01406736(00)03228-1 11. Joinson C, Heron J, von Gontard A, Butler U, Emond A, Golding J. A prospective study of age at initaition of toilet training and subsequent day time bladder control in school-aged children. J Dev Behav Pediatr 2009;30(5):385-393. https:// doi.org/10.1097/dbp.0b013e3181ba0e77 12. Mota DM, Barros AJ. Toilet training: situation at 2 years of age in a birth cohort. J Pediatr (Rio J) 2008;84(5):455-462. https://doi.org/10.2223/jped.1832

128

SAJCH

13. Duong T, Jansson U, Holmdahl G, Sillen U, Hellstrom A. Development of bladder control in the first year of life in children who are potty trained early. J Pediatr Urol 2010;6(5):501-505. https://doi.org/10.1016/j.jpurol.2009.11.002 14. Brazelton TB. A child-oriented approach to toilet training. Pediatrics 1962;29(1):121-128. 15. Foxx RM, Azrin NH. Dry pants: A rapid method of toilet training children. Behav Res Ther 1973;11(4):435-442. https://doi.org/10.1016/0005-7967(73)90102-2 16. Stadtler AC, Gorski PA, Brazelton TB. Toilet training methods, clinical interventions and recommendations. Pediatrics 1999;103(supp 3):1359-1368. 17. American Academy of Pediatrics. Toilet Training. Guidelines for Parents. Illinois: AAP, 1998. 18. Community Paediatrics Committe C. Toilet learning: Anticipatory guidance with a child-oriented approach. J Paediatr Child Health 2000;5(6):333-335. https://doi.org/10.1093/pch/5.6.333 19. De Paepe H, Hoebeke P, Renson C, Van Laecke C, Raes A, Van Hoecke E. Pelvic-floor therapy in girls with recurrent urinary tract infections and dysfunctional voiding. Br J Urol 1998;81(S3):109-113. https://doi.org/10.1046/ j.1464-410x.1998.00021.x 20. Polaha J, Warzak W, Dittmer-Mcmahon K. Toilet training in primary care: current practice and recommendations from behavioural pediatrics. J Dev Behav Pediatr 2002;23(6):424-429. https://doi.org/10.1097/00004703-200212000-00005 21. Schmitt B. Toilet training: Getting it right the first time. Contemp Pediatr 2004;21:105-122. 22. Mota DM, Barros AJD. Toilet training : Methods , parental expectations and associated dysfunctions. Jornal de Pediatria 2008;84(1):9-17. https://doi. org/10.2223/jped.1752 23. Caldwell P, Edgar D, Hodson E, Craig J. Bedwetting and toileting problems in children. Med J Aust 2005;182:190-195. 24. Hjalmas K. Still much ignorance about the fact that children with enuresis need treatment. Lakartidningen 2004;101:276-280. 25. Yeung C. Nocturnal enuresis (bedwetting). Curr Opin Urol 2003;13(4):337-343. https://doi.org/10.1097/00042307-200307000-00011 26. Abramovitch IB, Abramovitch HH. Enuresis in cross-cultural perspective: a comparison of training for elimination control in three Isreali ethnic groups. J Soc Psychol 1989;129(1):47-56. https://doi.org/10.1080/00224545.1989.9711698 27. Fishman L, Rappaport L, Cousineau D, Nurko S. Early constipation and toilet training in children with encopresis. J Pediatr Gastroenterol Nutr 2002;34(4):385388. https://doi.org/10.1097/00005176-200204000-00013 28. Chiozza ML, Bernardinelli L, Caione P, et al. An Italian epidemiological multicentre study of nocturnal enuresis. Br J Urol 1998;81(s3):86-89. https://doi. org/10.1046/j.1464-410x.1998.00015.x 29. De Paepe H, Renson C, Van Laecke E, Raes A, Vande Walle J, Hoebeke P. Pelvicfloor therapy and toilet training in young children with dysfunctional voiding and obstipation. Br J Urol 2000;85(7):889-893. https://doi.org/10.1046/j.1464410x.2000.00664.x 30. Hellstrom A, Hjalmas K, Jodal U. Rehabilitation of the dysfunctional bladder in children: Method and 3-year follow up. J Urol 1987;138(4):847-849. https://doi. org/10.1016/s0022-5347(17)43395-7 31. Oyedeji GA. Socio economic and cultural background of hospitalised children in Ilesha. Niger J Paediatr 1985;12(4):111-117. 32. Horn I, Brenner R, Rao M, Cheng T. Beliefs about the appropriate age for initiating toilet training: are there racial and socioeconomic diffrences? J Pediatr 2006;149(2):165-168. https://doi.org/10.1016/j.jpeds.2006.03.004 33. Choby BA, George S. Toilet training. Am Fam Physician 2008;78(9):1059-1064. 34. Bloom DA, Seely WW, Ritchey ML, McGuire EJ. Toilet habits and continence in children: An opportunity in search of normal parameters. J Urol 1993;149(5):10871090. https://doi.org/10.1016/s0022-5347(17)36304-8 35. Schum TR, Kolb TM, McAuliffe TL, Simms MD, Underhill RL, Lewis M. Sequential acquisition of toilet-training skills: a descriptive study of gender and age differences in normal children. Pediatrics 2002;109(3):e48-49. https://doi. org/10.1542/peds.109.3.e48 36. Blum NJ, Taubman B, Nemeth N. Relationship between age at initiation of toilet training and duration of training: A prospective study. Pediatrics 2003;111(4):810-814. https://doi.org/10.1542/peds.111.4.810

OCTOBER 2017 Vol. 11 No. 3

RESEARCH

This open-access article is distributed under Creative Commons licence CC-BY-NC 4.0.

The barriers that women face when choosing food for their primary school children: A case study in the Western Cape Province, South Africa Y Smit,1 BSc, M Nutr; S Kassier,2 PhD; D Nel,3 PhD; N Koen,1 BSc, M Nutr Division of Human Nutrition, Faculty of Medicine and Health Sciences, Stellenbosch University, Cape Town, South Africa Dietetics and Human Nutrition, University of KwaZulu-Natal, Pietermaritzburg, South Africa 3 Centre for Statistical Analyses, Stellenbosch University, Stellenbosch, South Africa 1 2

Corresponding author: Y Smit (yolandes@sun.ac.za) Background. Unhealthy food choices made by mothers can impact negatively on child health and may lead to unhealthy eating behaviours that persist into adulthood. Choosing food is a complex process influenced by many factors. Objectives. To determine the factors that influence mothersâ&#x20AC;&#x2122; food choices and to investigate barriers to purchasing healthy food. Methods. A cross-sectional, descriptive study, with an analytical component, was conducted. Mothers (N=476) were recruited from three randomly selected primary schools, from a low, middle and high quintile. A self-administered questionnaire was used to collect data on demographics, knowledge, attitude and practices of mothers. Six focus groups (FG) (two per school) were conducted to investigate the factors that prevent mothers from making healthy dietary decisions. Results. The mean nutrition knowledge score for the group was 68.6%. Nutrition knowledge was significantly lower (p<0.01) in mothers from the lower quintile school (64.0%). Primary factors influencing food purchases were cost (60%), nutritional value (37%) and time constraints (29%). Primary sources of nutrition information included magazines and health professionals at 62% and 44%, respectively. Time constraints resulted in mothers purchasing convenience foods more often (p=0.001). The main barriers identified were mixed media messages, the school environments and supermarket layouts. Conclusion. Nutrition education campaigns should include practical advice, e.g. the preparation of economical, wholesome meals. Policymakers should monitor increasing prices of healthy foods. School and supermarket environments, as well as the food industry, can play a pivotal role in facilitating mothers to make healthy food choices. S Afr J Child Health 2017;11(3):129-134. DOI:10.7196/SAJCH.2017.v11i3.1292

Mothers of young children have a primary influence and control over the food their children eat and have access to.[1] Their role is fundamental in promoting a healthy lifestyle and behaviour in children.[2-4] Dietary habits, including healthy food choices acquired during childhood, often persist into adulthood and lay the foundation for adult health and quality of life.[5,6] Childhood obesity is a growing phenomenon in South Africa, is a risk factor for non-communicable diseases (NCDs), and often tracks into adulthood, which is associated with growing up in an obesogenic environment.[5,7] Although the home environment is the logical place in which to foster healthy eating habits, studies have shown that food choices are complex and influenced by several factors including knowledge, socioeconomic status, cost, taste, child preference, urbanisation, and culture.[3,8,9] These factors ultimately impact on the food that is available at home. The Metro North Education District (MNED) in the Western Cape (WC) Province, South Africa, is an urban district with diverse living conditions ranging from wealthy suburbs to underdeveloped informal urban areas. Diverse sociodemographic profiles are prominent. It is therefore important to recognise the impact that socioeconomic status can have on food choices made by mothers. Nutrition research should not lose touch with reality and ought to be tailored for specific target audiences, including more vulnerable, lowincome groups.[10,11] A better understanding of these elements can impact future health education strategies and social marketing campaigns targeting mothers and childrenâ&#x20AC;&#x2122;s food choices, and may play a role in curbing the growing epidemic of childhood obesity. The objectives of the

129

SAJCH

study were to: (i) investigate the factors that influence food choices of mothers with children attending primary schools in the MNED of the WP; and (ii) to identify barriers to making healthy food choices.

Methods

A mixed-method study design employing triangulation of data was conducted. A self-administered questionnaire (SAQ) was used to collect demographic data, data related to factors and barriers influencing food choices, and to identify mothers for focus groups (FGs) to further explore these factors and barriers. Approvals to conduct the study were obtained from the Health Research Ethics Committee of Stellenbosch University (S/3/10/210), the Western Cape Education Department (WCED) and the respective school principals.

Sampling of schools

A list of schools in the MNED was obtained from the WCED website.[12] Public primary schools were stratified into the five national quintiles (NQ). The excel random generation function was used to randomly select one school from each quintile. Quintiles 1 and 5 represented lower and higher socioeconomic groups, respectively. Schools from quintile 1 to 3 were grouped together as they qualified for exemption of school fees and represented disadvantaged communities. Hence, 3 schools were randomly selected; school A (NQ 1 - 3), school B (NQ 4) and school C (NQ 5). Two schools were selected for the pilot study, 1 school from NQ 1 to 3 and 1 from NQ 4. Schools were contacted by the researcher to gain permission to conduct the study. If it was not granted, the random selection process was repeated.

OCTOBER 2017 Vol. 11 No. 3

RESEARCH Sampling of mothers

Primary schools included learners from grades 1 to 7. Two classes per grade were randomly selected. Children received a SAQ to take home to their mothers. Since older children have a greater influence on mothers’ decision-making, to keep the sample homogenous, children who had siblings in high school were excluded from this study.[13] To recruit FG participants, the last page of the questionnaire included a tear-off slip for mothers to indicate their willingness to participate in a FG. All mothers who supplied their contact details were contacted telephonically and depending on their willingness or availability, were invited to participate in the FG.

the Mann-Whitney or Kruskal-Wallis tests. Relations between nominal variables were investigated with contingency tables and likelihood ratio χ2 tests. Voice recordings were transcribed verbatim by one researcher and reread several times to ensure accuracy of the transcriptions. Thematic content analysis was performed by the same researcher. Major themes were identified based on the study objectives, manually coded and a code list was compiled. Quotes pertaining to each theme were transferred from the original text and copied to the most applicable code. New emerging themes were also deduced from the transcriptions.[15,18]

Preparation for the study

Results

Prior to the study, the researcher visited the school principals to explain both the purpose of the study, and to discuss the language proficiency of the parents. School principals confirmed that the language used for communication with parents was Afrikaans and English. The study was piloted at an NQ 3 and an NQ 4 school. These schools did not form part of the main study. During the pilot study, the face validity of the questionnaire was assessed and feedback from the parents relating to readability and understanding of the SAQ was received to ensure that participants could fill in the questionnaire.[14,15] Content validity was evaluated by three dietitians with experience in the relevant field.[14,15]

Data collection tools

Phase 1: Self-administered questionnaire

The SAQ was developed to identify factors influencing food choices based on an existing, validated, nutrition knowledge questionnaire for adults.[16] This paper reports results pertaining to section (i) demographics, (ii) nutrition knowledge and (v) barriers. Questionnaires were distributed to the children in unsealed envelopes. Parents were asked to return the questionnaire in the sealed envelope. Consent forms for participation in the main study were provided in duplicate – one copy for the parents’ reference and the other to be returned to the school with the completed questionnaire. Consent forms were available in Afrikaans and English and the researcher’s contact details were provided in case participants had any queries. Questionnaires were coded (A, B or C) in order to identify the school they originated from and for data entry and coding purposes.

Phase 2: Focus groups

Thirty-seven mothers participated in the FG discussions (FGDs) to obtain a deeper understanding of the factors influencing food choices and to explore barriers to making healthy food choices. Participation in the FGs varied from 5 to 10 participants per FG.[17] The discussions were conducted in a pre-arranged, suitable, school-based venue. One female researcher used a discussion guide to facilitate the FGs. Ten questions were included to explore the influence of knowledge, employment status, family preference, barriers and school environment to making healthy food choices. A list of probes, e.g. ‘tell me more’, ‘why do you feel that way?’, and ‘explain further’, were included in the discussion guide. An observer who was a qualified nutritionist, made notes regarding participant interaction and nonverbal cues. No language barriers were experienced during the FGDs. The researcher translated questions or replies into either Afrikaans or English if there was a need. The researcher ended off the discussion by asking the participants if they had additional comments and then summarised the main responses.

Data analysis

Microsoft Excel 2007 was used for data entry and STATISTICA version 12 (StatSoft Inc., USA) for data analysis. Summary statistics were used to describe sample characteristics. Independent variables were compared with appropriate analysis of variance (ANOVA). Bonferroni multiple comparisons identified significant differences between groups. Dependent variables were compared with independent variables using

130

SAJCH

A 50% response rate yielded a study sample of N=476. Sociodemographic characteristics of the study sample were defined according to the SAQ (Table 1). The mean (SD) age of participants was 37.45 (7.07). Six FGs (N=37) were conducted, two per school. Participation in the FG was as follows: 16 participants from NQ 1 - 3, 11 participants from NQ 4, and 10 participants from NQ 5. Mothers were mixed race (n=34) or black (n=3).

Nutrition knowledge

Participants were asked a series of nutrition-related questions (Table 2). The questions with the lowest mean scores were ‘white bread is healthier when it is toasted’, ‘low fat products contain less than 3 g fat per 100 g’ and ‘brown sugar and honey is healthier than white sugar.’ The nutrition knowledge of mothers from school A (64.0%) was significantly lower (p<0.01) compared with mothers from schools B (70.3%) and C (74%). The mean (SD) nutrition knowledge score for the whole group (N=476) was 68.6%. During the FGs, all mothers from the lower socioeconomic group agreed that nutrition knowledge could influence food choices. Mothers from the higher socioeconomic groups agreed that other factors such as habit, tradition or children’s preferences also influence food choice. Inadequate nutrition knowledge resulted in misperceptions regarding healthy food choices. The majority of mothers from schools A and B were of the opinion that ‘brown sugar is healthier than white sugar’ and that they ‘have to buy it for their kids’. Mothers from the lower socioeconomic group seemed to be aware that carbonated beverages have a high sugar content and explained that they would rather opt for a cool drink that ‘can be diluted with water or ice because it absorbs the sugar’ and is an alternative to drinking water. However, even though they were aware of the high sugar content, the gas in the carbonated drinks was perceived as the main reason for making them unhealthy, thereby placing less emphasis on the high sugar content.

Sources of nutrition information

Mothers source nutrition information from various platforms (Fig. 1). During FGs, the workplace, hospitals, child-care centres and clinics were mentioned as additional sources of nutrition information by mothers from schools A and B. Only mothers from school C indicated that they used the internet as a source of information. Mothers from school A explained that they listened to advice given by nurses at clinics and applied the healthy eating guidelines to themselves and their families. Most mothers were aware of the nutritional information on food products; however, they all agreed that it did not influence their purchasing decisions. Mothers from schools B and C had the perception that the products with logos on are more expensive. Despite any differences, all of the FG participants expressed the need to understand and use the available nutrition information on food labels to guide their food choices.

Other factors influencing food choice

Numerous important factors influence the choices mothers make when buying food (Fig. 2). The cost of a product was significantly more important (p<0.001) for mothers from school A, (72%; n=145) compared with mothers from schools B and C, (56%; n=83) and (50%; n=64),

OCTOBER 2017 Vol. 11 No. 3

RESEARCH Table 1. Sociodemographic information of mothers (N=476) All schools (NQ 1 - 5); N=476, n (%)*

School A (NQ 1 - 3); n=202, n (%)

School B (NQ 4); n=147, n (%)

School C (NQ 5); n=127, n (%)

Mother

447 (94)

188 (93)

133 (90)

126 (99)

Grandmother

8 (2)

5 (2.5)

2 (1.3)

1 (0.01)

Foster mother

21 (4)

9 (4.5)

12 (8)

0 (0)

Relationship to child

Race

(N=475)

(n=202)

(n=146)

(n=127)

Black

52 (11)

21 (10.4)

17 (12)

14 (11)

Coloured

406 (85)

180 (89.1)

117 (80)

109 (86)

Indian

4 (0.1)

1(0.5)

0 (0 0)

3 (2.4)

White

13 (3)

0 (0.0)

12 (8)

1 (0.01)

Employment status

(N=467)

(n=198)

(n=143)

(n=126)

Working

316 (68)

105 (53)

109 (76)

102 (81)

Not working

151 (32)

93(47)

34 (24)

24 (19)

(N=470)

(n=198)

(n=145)

(n=127)

Grade ≤7

60 (13)

52 (26)

8 (6)

0 (0)

Grade 8 - 11

154 (33)

108 (55)

37 (25)

9 (7)

Grade 12

146 (31)

32 (16)

69 (48)

45 (35)

Diploma or higher degree

110 (23)

6 (3)

31 (21)

73 (57)

(N=435)

(n=181)

(n=133)

(n=121)

<1 000

76 (17)

65 (36)

11 (8)

0 (0)

1 001 - 2 500

71 (16)

58 (32)

13 (9)

0 (0)

2 501 - 3 500

33 (8)

26 (14)

7 (5)

0 (0)

3 501 - 5 500

47 (11)

19 (11)

19 (14)

9 (7)

5 501 - 9 000

44 (10)

11 (6)

21 (16)

12 (10)

9 001 - 12 500

39 (9)

2 (1)

21(16)

16 (13)

12 501 - 16 500

38 (9)

0(0)

14 (11)

24 (19)

>16 500

87 (20)

0 (0)

27 (20)

60 (46)

Level of education

Level of income (ZAR per month)

ZAR = South African Rand; low income = <ZAR3 200/month; middle income = ZAR3 210 - ZAR12 500/month; high income = >ZAR12 501/month. *N and n-values for each criterion differed as the respondents did not always answer all of the questions.

p=0.01 74*

p=0.01 59*

58.4

53.1

49.6

44.5 40.2

42.9

33.7

30.2

71.78*

% of subjects’ responses

Subjects who responded affirmatively, %

56.5 50.4

44.09*

41.73*

32.28* 27.9

30 20

36.7

33.07*

29.2

23.1

18.3

22.5

16.3

14.8

5.9

7.8

3.4

Lack of time

Price

Taste

0 Magazines

Friends

Children

Television

School B (n=147)

Preference of children

Factors influencing food choice

Doctors

Sources of nutrition information School A (n=202)

Nutritional value

School A (n=202)

School C (n=127)

School B (n=147)

School C (n=127)

*p<0.05 X statistics Fig. 1. Sources of nutrition information as indicated by mothers from 3 primary schools representing different socioeconomic groups (N=476; *p<0.05, χ2 statistics).

Fig. 2. Comparison of factors influencing food choices of mothers from 3 primary schools representing different socioeconomic groups (N=476, *p<0.05, χ2 statistics).

respectively. During the FGs, all mothers agreed that they shop where the bargains are and would rather choose products that are on special.

Mothers from the lower socioeconomic group expressed their concern regarding the high cost of the fruit and vegetables that they buy from

131

SAJCH

OCTOBER 2017 Vol. 11 No. 3

RESEARCH Table 2. Percentage score of nutrition knowledge questions answered correctly (N=476) All schools (N=476), n (%)

Statement

School A (n=202), n (%)

School B (n=147), n (%)

School C (n=127), n (%)

White bread is healthier when toasted

197 (41.3)

52 (25.7)

68 (46.3)

77 (60.6)

Growing children need a lot of sugar

315 (66.1)

90 (44.5)

115 (78.2)

110 (86.6)

Children need to eat fruit and vegetables daily

459 (96.4)

193 (95.5)

143 (97.2)

123 (96.8)

If children eat a healthy diet there is no need for them exercise

430 (90.3)

167 (82.6)

138 (93.8)

125 (98.4)

A glass of fruit juice is healthier than a fresh fruit

365 (76.6)

130 (64.3)

122 (82.9)

113 (88.9)

Fruit and vegetables are fat-free items

309 (64.9)

146 (72.2)

87 (59.8)

76 (59.0)

Red meat is a good source of iron

311 (65.3)

118 (58.4)

100 (73.2)

93 (73.2)

Baked beans are a good source of protein

361 (75.8)

153 (75.7)

106 (80.3)

102 (80.3)

Fried eggs are healthier than boiled eggs

440 (92.4)

186 (92.0)

134 (94.4)

120 (94.4)

Low-fat products contain <3 g fat per 100 g

178 (37.3)

84 (41.5)

53 (32.2)

41 (32.2)

Gas cooldrinks are healthy drinks

456 (95.8)

188 (93.0)

144 (97.6)

124 (97.6)

Brown sugar and honey are healthier than white sugar

22 (46.2)

11 (54.4)

3 (20.4)

8 (62.9)

Coffee creamers are just as healthy as milk

406 (85.2)

164 (81.18)

131 (89.1)

111 (87.4)

Table 3. Comparison of responses from mothers at different schools pertaining to barriers to making healthy food choices and food preparation practices Disagree, n (%)

Agree, n (%)

Strongly agree, n (%)

p-value 0.32*

Barrier

School

Strongly disagree, n (%)

I do not know how to prepare healthy meals (N=470)

198

77 (38.9)

96 (48.4)

24 (12.1)

1 0.5

146

59 (40.4)

73 (50 )

11 (7.5)

3 (2.0)

126

Average Fast food shops close to home (N=471)

13 (10.3)

1 (0.8)

48 (10.2)

5 (1.1)

199

99 (49.8)

87 (43.7)

11 (5.5)

2 (1.0)

146

75 (51.4)

67 (45.8)

2 (1.3)

126

199

64 (50.8)

55 (43.6)

3 (2.3)

4 (3.1)

238(50.6)

209(44.3)

16 (3.3)

8 (1.6)

20 (10.0)

54 (27.1)

114 (57.3)

11( 5.5)

147

16 (10.8)

54 (36.7)

70 (47.6)

7 (4.8)

126

12 (9.5)

58 (46.0)

52 (41.3)

4 (3.2)

48 (10.2)

66 (35.2)

236 (50)

22 (4.0)

15 (7.5)

55 (27.3)

121 (60.2)

10 (4.9)

Average I add sugar and margarine to vegetables to make them tasty (N=474)

51 (40.4) 220(46.2)

Average I use oil regularly when preparing meals (N=472)

61 (48.4) 197(42.5)

201

147

18 (12.2)

52 (35.3)

75 (51)

2 (1.2)

126

13 (10.3)

48 (38.1)

58 (46)

7 (5.5)

46 (9.7)

155(37.2)

254(53.6)

19 (4.0)

Average

0.30*

0.02†‡

0.01†§

*p>0.05 χ statistics does not indicate statistical significance. † p<0.05 Kruskal-Wallis test indicates statistically significant differences. ‡ Significant difference between school A and C. Bonferroni p<0.04. § Significant difference between school A and B. Bonferroni p<0.02.

informal vendors, sometimes the only suppliers they have access to. Furthermore, these mothers agreed that unhealthy food options such as sweets and chips are more affordable when compared with purchasing fresh fruit or healthy food. Significantly more mothers (p=0.02) from school C (44%; n=56) listed nutritional value as an influencing factor compared with mothers from school A (29%; n=59). Information deduced from the FGs indicated that mothers, regardless of their socioeconomic status, viewed fruit and vegetables as an essential part of a healthy diet. They also mentioned low-

132

SAJCH

fat products and foods containing roughage as being healthier, confirming that mothers were aware of the nutritional value of these foods. More mothers (p=0.000) from school C (42%; n=53) reported that a lack of time influenced their food choices compared with mothers from school A (18%; n=37) and mothers from school B (27%; n=41). During the FGs, mothers from school C, the higher socioeconomic group, explained that time constraints, mostly as a result of them working, resulted in opting for time-saving convenience options, even if they were not the healthiest.

OCTOBER 2017 Vol. 11 No. 3

RESEARCH Table 4. Comparison of the responses of mothers from different schools towards the statement: ‘The messages through TV and radio influence my food choice’ (N=473) Strongly disagree, n (%)

Disagree, n (%)

Agree, n (%)

Strongly agree, n (%)

p-value

School A (n=201)

26 (12.9)

59 (29.4)

94 (46.7)

22 (10.9)

0.26*

School B (n=145)

10 (6.9)

50 (34.5)

74 (51.0)

11 (7.9)

School C (n=127)

9 (7.1)

48 (37.8)

57 (44.9)

13 (10.2)

p>0.05 χ statistics indicate no statistically significant difference.

Taste and child preference were significantly (p=0.003) more important factors for mothers from school C (32.3%; n=41) and (33.1%, n=42), respectively. During FGs, mothers from school A felt strongly that children must eat what is served compared with mothers from school C who said, ‘I buy what my kids want even if I know it is not the healthiest option.’

Barriers to making healthy food choices

A lack of cooking skills and accessibility to numerous fast-food outlets were not viewed as a barrier when food choices were made (Table 3). However, more than half of the study sample that completed the SAQ indicated that they regularly made use of unhealthy meal preparation methods through the addition of unhealthy ingredients like oil, sugar and margarine. Three main barriers identified during the FGs are discussed below.

Inconsistent media messages

Several mothers from school C voiced their frustration with the mixed messages resulting in confusion. They explained that children are vulnerable, easy to influence and aware of marketing strategies. More than half (57%; n=271) of the participants agreed that media messages influence their food choices (Table 4).

School environment

School tuck shops (a small kiosk providing snacks on the school premises) were discussed during the FGs. Schools B and C had tuck shops and mothers from school A reported that informal vendors would sell chips and sweets outside the school premises. Almost all of the mothers agreed that tuck shops and informal vendors sell predominantly unhealthy food, but were unsure if fresh fruit was sold at the tuck shop. Some mothers expressed the desire to be involved in deciding what was sold at tuck shops. All mothers agreed that schools should create more awareness and encourage children to bring more healthy food to school. All FG participants acknowledged the powerful impact that teachers have on their children, because the children idealised their teachers. Mothers from the lower socioeconomic group agreed that it was the school’s responsibility to teach children about healthy eating.

Supermarket layout

Mothers from schools B and C agreed that if aisles with healthier options were more prominently positioned and convenient to access, it would assist them and their children to make healthier food choices. Mothers from school A, who often bought food from spaza shops, explained that ‘if fruit and vegetables were more visible, the children might also choose to buy it more often.’

Discussion

Higher levels of parental education were positively associated with higher nutrition knowledge scores.[10,11] In turn, higher levels of nutrition knowledge were positively associated with healthy diets.[16] This study identified gaps in knowledge related to fibre, fat and sugar intake, which could potentially have a negative impact on childhood health. Diets low in fibre and high in fat and sugar are indicators of the nutrition transition taking place in South Africa.[19] Questions with the best scores directly

133

SAJCH

related to the messages conveyed through the South African Food-Based Dietary Guidelines (FBDG).[20] Even though mothers knew that fried foods were less healthy, they still used oil and margarine regularly in the preparation of meals. This strengthens the notion that knowledge does not necessarily translate into healthy food choices or preparation practices.[21 This study emphasised the importance of nutrition education, especially for mothers from lower socioeconomic groups. Furthermore, the study found that the media was a powerful medium to convey scientifically accurate information to the lay public and should be better utilised.[22] The cost of food was identified as the strongest determining factor influencing food choices, regardless of the school quintile. The effect of food cost on food choices and healthy eating cannot be underestimated; it calls for drastic measures to be implemented by the SA government and the food industry to curb the increasing cost of healthy food. Less healthy, energy-dense food is more affordable, making it a more desirable purchasing option, especially among lower socioeconomic groups.[10,23,24] It is therefore evident that it is not only the high cost of healthy food, but also the affordability of unhealthy food that seems to be a barrier when making healthier food choices.[23] A third of the study sample listed nutritional value as a factor influencing food choice compared with the 14.3% of participants in the SANHANES-1 study.[8] It is concerning that mothers potentially do not have adequate nutrition knowledge to identify foods with high nutritional value, as identified by the knowledge questions and FGs in this study. Similar to the findings of Machin et al.,[25] mothers found it difficult to interpret nutrition information tables, even though they expressed a need to understand them. Mothers in this study therefore did not benefit optimally from this source of information. Simplified nutrition labels or logos might be more effective in influencing food choices, especially among those individuals in lower-income groups. Vastly higher percentages of mothers, although significantly different between the three schools (Fig. 2), reported that time constraints influenced their food choices, compared with the 9.6% in the SANHANES-1 study.[7] Time constraints resulted in the regular purchases of convenience foods, especially in school C, the quintile 5 school that also had a higher complement of working mothers. Unfortunately, these foods are often high in fat and sugar and children become accustomed to them, hence developing a preference.[26,27] This could result in the younger generations acquiring fewer cooking skills from their mothers and, in turn, a reliance on convenience foods. Nutritional advice often centres around what to eat, with less emphasis being placed on how to incorporate the advice into a busy lifestyle.[26] Not only is there a need for healthier pre-prepared and convenience meals, but also for the education of mothers on ways to prepare quickand-easy nutritious homemade meals. The food industry has responded and capitalised on this need, but the majority of options still focus on convenience and not on health.[27,28] The value of changing supermarket layout to enhance the visibility of healthier items justifies further exploration. Participants in a study conducted by Zacchary et al.[29] suggested having taste test sections to increase consumer confidence in a product. By changing supermarket layout, the principle of making healthier choices the easiest choice, can be supported and implemented.[30] In addition, creating a healthy school

OCTOBER 2017 Vol. 11 No. 3

RESEARCH environment and utilising teachers as role models and advocates for healthy eating can influence children’s attitudes towards healthy eating, creating a positive spillover into the home environment.[31] Increased parental involvement in decision-making regarding the food items sold at school tuck shops can ensure that healthy nutrition principles are implemented consistently at school and in the home environment.[3] This study was the first in SA that focused specifically on mothers of primary school children, and comparing NQs with each other. The study provided new insights into the needs and challenges that exist between different socioeconomic groups, strengthening the notion that nutrition education and health promotion should be tailor-made for diverse groups. These results could be of value when developing interventions that involve mothers as important change agents, aimed at the prevention of childhood obesity.

Study limitations

The sociodemographic distribution of race was not representative, as the study sample included a limited number of white and black participants. The researchers acknowledge the potential of acquiescence and social desirability bias with FG participants, as well as the Hawthorne effect with the mothers who completed the SAQ. This may have influenced the extrapolation and generalisations of the study findings.

Conclusion

Greater emphasis should be placed on imparting basic nutrition knowledge to diverse target audiences. This includes empowering mothers to interpret food labels and demonstrating healthy and economical food preparation techniques. The school environment can be used effectively to change children’s perceptions regarding healthy food, and support mothers to implement their nutrition knowledge in the home environment. This justifies further investigations to effectively identify specific target groups within the larger public.[10] Acknowledgements. Participants, school principals and the Harry Crossley Foundation. Author contributions. YS did the primary data collection and analyses. DN did the statistical analyses. Funding. The Harry Crossley Foundation. Conflicts of interest. None. 1. Johnson CM, Sharkey JR, Dean WR, Kubena KS. It’s who I am and what I eat. Mothers’ food-related identities in family food choice. Appetite 2011;57(1):220228. https://doi.org/10.1016/j.appet.2011.04.025 2. Lindsay AC, Sussner KM, Kim J, Gortmaker S. The role of parents in preventing childhood obesity. Future of Children 2006;16(1):169-186. https:// doi.org/10.1353/foc.2006.0006 3. Patrick H, Nicklas TA. A review of family and social determinants of children’s eating patterns and diet quality. J Am Coll Nutr 2005;24(2):83-92. https://doi. org/10.1080/07315724.2005.10719448 4. Scaglioni S, Arrizza G, Vecchi F, Tedeschi S. Factors of children's eating behavior. Am J Clin Nutr 2011;94(6):2006s-2011s. https://doi.org/10.3945/ ajcn.110.001685 5. Black RE, Victora CG, Walker SP, et al. Maternal and child undernutrition and overweight in low-income and middle-income countries. Lancet 2013;382(9890):427-451. https://doi.org/10.1016/s0140-6736(13)60937-x 6. Nossle C. Childhood obesity on the increase. The Mercury. http://iol. co.za/mercury/childhood-obesity-in-the-increase-1.1213585 (accessed 12 September 2013). 7. Schonfeldt HC, Gibson N. Healthy eating in the South African context. J Food Comp Anal 2009;22:68-73. https://doi.org/10.1016/j.jfca.2009.01.005 8. Shisana O, Labadarios D, Rehle T, et al., and the SANHANES-1 Team. South African National Health and Nutrition Examination Survey (SANHANES-1): 2014 Edition. Cape Town: HSRC Press, 2014. http://www.hsrcpress.ac.za/ product.php?productid=2314&cat=0&page=1&featured&freedownload=1 (accessed 12 September 2014).

134

SAJCH

9. Cowburn G, Stockley L. Consumer understanding and use of nutrition labeling: a systematic review. Public Health Nutr 2005;8(1):21-28. https://doi. org/10.1079/phn2004666 10. Darmon N, Drewnowski A. Does social class predict diet quality? Am J Clin Nutr 2008;87(5):1107-1117. 11. Turrell G, Kavanagh AM. Socio-economic pathways to diet: modeling the association between socio-economic position and food purchasing behavior. Public Health Nutr 2005;9(3):375-383. https://doi.org/10.1079/phn2006850 12. Western Cape Education Department website. Find-A-School. http://wcedemis. pgwc.gov.za/wced/findaschool.html (accessed 12 October 2013). 13. Mangleburg TF. Children’s influence in purchase decisions: A review and critique. Adv Consum Res 1990;17:813-825. 14. Gleason PM, Harris J, Sheaan PM, Boushey CJ, Bruemmer B. Publishing nutrition research: Validity, reliability and diagnostic test assessment in nutrition-related research. J Am Diet Assoc 2010;110:409-419. https://doi. org/10.1016/j.jada.2009.11.022 15. Skinner D. Qualitative research methodology: An introduction. In: Joubert G, Ehrlich R, eds. Epidemiology: A Research Manual for South Africa. Cape Town: Oxford University Press; 2008:318-326. 16. Parmenter K, Wardle J. Development of a general nutrition knowledge questionnaire for adults. Europ J Clin Nutr 1999;53:298-308. https://doi. org/10.1038/sj.ejcn.1600726 17. Rabiee F. Focus group interviews and data analysis. Proceedings of the Nutrition Society. 2004;63:655-660. https://doi.org/10.1079/pns2004399 18. Mitchell K, Branigan P. Using focus groups to evaluate health promotion interventions. Health Educ 2000;100(6):261-268. https://doi. org/10.1108/09654280010354887 19. Abrahams Z, Mchisa Z, Steyn NP. Diet and mortality rates in Sub-Saharan Africa: Stages of the nutrition transition. BMC Public Health 2011;11(1):801. https://doi.org/10.1186/1471-2458-11-801 20. Vorster HH. Badham JB, Venter CS. An introduction to the revised food-based dietary guidelines for South Africa. S Afr J Clin Nutr 2013;26(3):5-12. 21. Wardle J, Parmenter K, Waller J. Nutrition knowledge and food intake. Appetite 2000;34:268-275. https://doi.org/10.1006/appe.1999.0311 22. Charlton KE, Brewitt P, Bourne LT. Sources and credibility of nutrition information among black urban South African women, with a focus on messages related to obesity. Public Health Nutr 2004;7(6):808-811. https://doi. org/10.1079/phn2004611 23. Temple NJ, Steyn NP. The cost of a healthy diet: A South African perspective. Nutrition 2011;27(5):505-508. https://doi.org/10.1016/j.nut.2010.09.005 24. Fitzgerald N, Spaccarotella K. Barriers to a healthy lifestyle: From individual to public policy – an ecological perspective. J Extension 2009;47(1):1-8. 25. Machín L, Giménez A, Curutchet MR, Martinez J, Ares G. Motives underlying food choice for children and perception of nutritional information among low-income mothers in a Latin American country. J Nutr Educ Behav 2016;48(7):478-485. https:// doi.org/10.1016/j.jneb.2016.04.396 26. Slater J, Sevenhuysen G, Edginton B, O’Nell J. Trying to make it all come together: structuration and employed mothers’ experience of family food provisioning in Canada. Health Prom Int 2012;27(3):405-415. https://doi.org/10.1093/heapro/dar037 27. Devine CM, Jastran M, Jabs JA, Wethington E, Farrell TJ, Bisogni CA. A lot of sacrifices: Work-family spillover and the food choice coping strategies of low wage employed parents. Soc Sci Med 2006;63(10):2591-2603. https://doi.org/10.1016/j. socscimed.2006.06.029https://doi.org/10.1016/j.socscimed.2006.06.029 28. Jabs J, Devine CM. Time scarcity and food choices: An overview. Appetite 2006;47(2):196-204. https://doi.org/10.1016/j.appet.2006.02.014 29. Zachary DA, Palmer AM, Beckham SW, Surken PJ. A Framework for understanding grocery purchasing in a low-income urban environment. Qual Health Res 2013;35(5):665-678. https://doi.org/10.1177/1049732313479451 30. World Health Organization. Milestones in Health Promotion. Statements from Global Conferences. Geneva. Switzerland. 2009. http://who.int/healthpromotion/ Milestones-Health-Promotion-05022010.pdf (accessed 12 September 2013). 31. Global Nutrition Report. From Promise to Practise. UNICEF DATA 2016. https://data.unicef.org/wp-content/uploads/2016/06/130565-1.pdf http://dx.doi. org/10.2499/9780896295841. 32. Wiles NL, Green JM, Veldsman FJ. Tuck shop purchasing practices of Grade 4 learners in Pietermaritzburg and childhood overweight and obesity. S Afr J Clin Nutr 2013;26(1):37-42. https://doi.org/10.1080/16070658.2013.11734438

OCTOBER 2017 Vol. 11 No. 3

RESEARCH

This open-access article is distributed under Creative Commons licence CC-BY-NC 4.0.

The clinical profile and outcome of children with West syndrome in KwaZulu-Natal Province, South Africa: A 10-year retrospective review A Keshave,1 MB ChB, MMed, FC Paed, Cert Paed Neuro; N Yende-Zuma,2 BSc, BSc Hons, MSc; L Mubaiwa,1 MB ChB, FC Paed, MA; M Adhikari,1 MB ChB, FC Paed, PhD 1 2

Department of Paediatric Neurology, Nelson R Mandela School of Medicine, University of KwaZulu-Natal, Durban, South Africa Centre for the AIDS Programme of Research in South Africa (CAPRISA), University of KwaZulu-Natal, Durban, South Africa

Corresponding author: A Keshave (amith.keshave@gmail.com) Background. West syndrome (WS) is a rare epileptic encephalopathy of infancy. There is currently no research on the incidence or prevalence of WS in Africa. Methods. We aimed to describe the outcome of children with WS at a quaternary-level hospital in KwaZulu-Natal, South Africa (SA). This was a retrospective chart review conducted on patients diagnosed with WS over a 10-year period. Eight children (males, n=7; African, n=6; Asian, n=2) identified with WS out of 2 206 admitted with epilepsy. The median age (range) at diagnosis was 7.5 (1 - 9) months. The average time between onset of epileptic spasms and diagnosis was 3.1 months. Results. Six patients had abnormal neuroimaging (atrophy (n=2); corpus callosum agenesis (n=2); tuberous sclerosis (n=1); focal dysplasia (n=1)). Drug management included sodium valproate (n=8), topiramate (n=7) and levetiracetam (n=3). Subsequent definitive treatment was intramuscular adrenocorticotrophic hormone (n=3), vigabatrin (n=2) and oral prednisone (n=4). Four (50%) patients had complete seizure remission (neuromigratory disorder (n=2); tuberous sclerosis (n=1); and idiopathic (n=1)) and 4 had partial remission (neonatal complications (n=3); idiopathic (n=1)). Discussion. Most of our patients had symptomatic WS, with 50% remission on treatment. Outcomes were poorer in our study when compared with those in published data. Conclusion. Further collaborative studies are still needed to evaluate the true impact and prevalence of WS in SA. S Afr J Child Health 2017;11(2):135-140. DOI:10.7196/SAJCH.2017.v2i3.1300

West syndrome (WS) is a rare epileptic disorder of infancy or early childhood. Dr William James West first described the syndrome in 1841 when it affected his son.[1] It is an epileptic encephalopathy characterised by epileptic spasms, electroencephalographic (EEG) evidence of hypsarrhythmia and developmental delay or regression.[2] Infantile spasms involve the neck, trunk and extremities. Spasms are classified as either flexors (flexion of the arms, legs and neck with contraction of the abdominal muscles in a jack knife or salaam attack), extensors (extension of the neck and trunk with abduction or adduction of the arms and legs), or mixed flexor-extensor spasms (flexion of the neck, trunk and arms with extension of the legs).[3] Spasms usually appear in the first 2 years of life, with a peak incidence between 4 and 6 months, and sometimes continue until adolescence.[2] Hypsarrhythmia is described as highvoltage, random, slow and spike waves in the cortex that vary in duration and location, and occasionally may become generalised.[4] The incidence of WS is well documented in developed countries, and a study in Finland reported an estimated incidence of 30.7 per 100 000 live births.[5] The incidence and prevalence of aetiological factors of WS in Africa are unknown. Successful treatment of WS improves morbidity and mortality outcomes. In a long-term-outcome study of 214 Finnish children with WS over a 35-year period, a third of the patients died, of whom a third died before the age of 3 years.[6] Significant contributing factors to the poor prognosis are the lack of standardised treatment guidelines, and delays in commencing treatment. This was evident after the results of studies from the Pellock et al.,[2] Lux et al.,[7] Ito et al.,[8] and Wilmshurst et al.[9] highlighted the treatment gap for WS. It was noted that further research was needed to formulate appropriate treatment protocols for WS with respect to pharmacological and non-pharmacological therapies, such as the ketogenic diet and neurosurgical interventions, such as vagal nerve stimulation, deep brain stimulation and epileptic surgery. There

135

SAJCH

are numerous publications of studies regarding the classification, treatment, and prognosis of WS in developed countries.[10] Knowledge regarding the epidemiology and outcomes of children with WS in developing countries and in resource-constrained settings is limited.[11] There are no descriptive studies on WS in Africa. We hypothesised that the incidence would be higher and treatment outcomes poorer for WS in our setting than in developed countries. The aim of the study was to evaluate the clinical presentation and treatment outcomes of children with WS in KwaZulu-Natal Province, South Africa (SA), and to identify factors related to current practices in diagnosing and treating WS in the Paediatric Neurology Department at Inkosi Albert Luthuli Central Hospital. Ethical approval was obtained from the Biomedical Research Ethics Committee of the University of KwaZulu-Natal (ref. no. BE419/15). Gatekeeper approval for use of the patientsâ&#x20AC;&#x2122; records and data was obtained from the Department of Health and Inkosi Albert Luthuli Central Hospital.

Methods

The study was conducted at the Paediatric Neurology Department at Inkosi Albert Luthuli Central Hospital, a quaternary care centre in Durban, SA. The hospital serves the general population of KwaZuluNatal. The patients are of mixed socioeconomic backgrounds.

Study design

Data were collected using a retrospective chart review of all patients diagnosed with WS over a 10-year period, from January 2005 to August 2015. Patients were identified from a data base of all patients <12 years of age admitted to the paediatric neurology ward with seizures or epilepsy. We identified 8 patients diagnosed with WS, from a total of 2 206 patients with seizures.

OCTOBER 2017 Vol. 11 No. 3

RESEARCH We extracted the following data from the patient files: sex, race, age of onset of infantile spasms, age at diagnosis, number of clusters per day, number of seizures per cluster, perinatal events, mode of delivery, retroviral status, aetiology, relation to tuberous sclerosis, anthropometry and developmental assessment, neuroimaging with magnetic resonance imaging (MRI) and electrophysiological studies with EEG. Data on treatment were also captured, including the type of antiepileptic drug used, the maximum dose and side-effects, and the response to treatment. The outcome was determined by the degree of epileptic spasm reduction prior to discharge, and on follow-up. The response to treatment was categorised into three groups: complete spasm cessation, partial response (>50% reduction in spasms) and no response (<50% reduction in spasms). Information regarding seizure semiology, EEG findings, anthropometry and neurodevelopment that was recorded in patient files on clinic visits was also extracted.

diameter <2 standard deviations (SD) for age, and 37.5% (n=3) were underweight for their age, and stunted, with length-for-age <2 SD. All patients presented with developmental delay, and the majority (n=6) had global developmental delay, affecting gross and fine motor skills (87.5% and 85.7%, respectively).

Investigations

EEG findings on admission were modified hypsarrhythmia (37.5%, n=3) and classic hypsarrhythmia (62.5%, n=5). All patients had MRI of the brain on admission, and the findings were: normal (n= 2), global atrophy (n=2), corpus callosum agenesis (n=2), features of tuberous sclerosis (hamartoma and sub-ependymal giant cell astrocytoma) (n=1) and left focal cortical dysplasia (n=1). The causes associated with WS in our cohort of patients were cryptogenic (n=2), neuromigratory disorder (n=2), neonatal hypoglycaemia (n=1), kernicterus (n=1), tuberous sclerosis (n=1) and neonatal hypoxic ischaemic encephalopathy (n=1) (Table 3).

Data analysis

Treatment

Continuous data were summarised using means, and categorical data using proportions. Fisherâ&#x20AC;&#x2122;s exact test was used to test for association between categorical variables. Statistical analyses were conducted using SAS version 9.4 (SAS Institute, USA).

At our institution, the first-line therapy for patients presenting with suspected WS is sodium valproate. The second-line therapy is topiramate or lamotrigine, followed by escalation with sequential use of levetiracetam. Once a diagnosis of WS is confirmed, adrenocorticotropic hormone (ACTH) and vigabatrin are prescribed. Oral prednisone is used as an alternative if ACTH is not available. Patients are screened for tuberculosis, which is endemic in our population, prior to the use of ACTH and prednisone. None of our patients who received ACTH or prednisone had a positive screen for tuberculosis. A total of 11 drugs were used in different combinations for all 8 patients: sodium valproate (n=8), topiramate (n=7), clonazepam (n=5), prednisone (n=4), ACTH (n=3), levetiracetam (n=3), lamotrigine (n=3), vigabatrin (n=2), lorazepam (n=1), phenytoin (n=1) and phenobarbital (n=1) (Table 4). Three patients had adverse reactions to sodium valproate and topiramate. The adverse effects of sodium valproate were hyperammonia (n=3) and thrombocytopenia (n=1). The adverse effects occurred with a prescribed average dose of sodium valproate of 44.7 mg/kg/day, and after a mean duration of treatment of 396 days. The main side-effect with topiramate was hypercarbia (n=1), with the average dose of topiramate being 9 mg/kg/day, and the average duration of treatment was 30 days. Only 1 patient in the study group died, as a result of comorbid sepsis and fulminant liver failure.

Results

Demographics

The WS patients were predominantly male (87.5%, n=7). The majority (75%, n=6) were black African, and the others Indian (25%, n=2).

Perinatal history

Table 1 shows the demographics of the perinatal history of the children in our patient cohort. Six patients (75%) were born by caesarean section. Four patients (50%) had abnormal perinatal courses. The remaining 4 patients had the following perinatal comorbidities: hypoglycaemia (n=1), kernicterus (n=1), neonatal seizures (n=1) and low birth weight (n=1). An abnormal Apgar score was defined as a score <6 at 5 minutes or 10 minutes, and this was noted in 1 patient (12.5%), who subsequently developed neonatal encephalopathy. Three patients (37.5%) were born to mothers infected with HIV, but were not infected themselves.

Presentation

Table 2 shows a summary of the presentation of our patients with WS. The mean age of epileptic spasm onset was 4.4 (range 1 - 9) months, with the mean age at diagnosis being 7.5 (range 3 - 18) months. The average duration between onset of spasms and diagnosis was 3.1 months. Patients had on average 5 clusters per day, with 7 spasms per cluster. Three patients (37.5%) had mixed seizures, consisting of spasms and myoclonic seizures. A total of 66.7% (n=4) had microcephaly, with an occipital-frontal

Final outcomes

An uneventful perinatal course was associated with a better outcome (75%), compared with the group that had an underlying perinatal event (25%). The mode of delivery and HIV status did not determine

Table 1. Demographics of West syndrome in our patient cohort Patient 1

Race

Indian

African

Indian

African

Sex

Male

Female

Perinatal course

Hypoglycaemia

Kernicterus

LBW

Uneventful

Neonatal seizures + MSL

Uneventful

Mode of delivery

NVD

Apgar score

Normal

Abnormal

Normal

HIV status

Unexposed

Exposed

Unexposed

Exposed

Unexposed

Exposed

Unexposed

LBW = low birth weight; MSL = meconium-stained liqor; CS = caesarean section; NVD = normal vaginal delivery.

136

SAJCH

OCTOBER 2017 Vol. 11 No. 3

RESEARCH Table 2. Summary of presentation of our cohort of patients with West syndrome Patient 1

Age (months) of spasms

Age (months) at diagnosis

Clusters per day

Unknown

Spasms per cluster

Unknown

Seizure type on admission

Spasms

Spasms and myoclonic

Spasms

Spasms and myoclonic

Spasms

Spasms and myoclonic

Occipital frontal circumference

Unknown

Microcephaly

Macrocephaly

Unknown

Microcephaly

Macrocephaly

Weight-for-age

Underweight

Normal

Underweight

Overweight

Normal

Underweight

Normal

Height-for-age

Normal

Stunted

Normal

Stunted

Normal

Developmental assessment on presentation

Global delay (no grasping + no babbling + sits only with support)

Global delay (sitting with support only + no transfer)

Global delay (no sitting even with support + no babbling + no grasping)

Global delay Normal (crawling only + only coos + no pincer grasping)

Global delay (no sitting with support + no grasping + no babbling)

Delay (no rolling)

Global delay (no babbling + no head control + poor grasping)

Anthropometry

poor outcome. Outcome measures were: spasm cessation, cognitive development, epilepsy evolution, morbidity and mortality. An earlier age of epileptic spasm presentation (<5 months) was associated with a better prognosis, with 60% of these children having spasm cessation, as compared with older children (>5 months), who had no spasm cessation (33.3%). There was no correlation between the age of diagnosis, head circumference, clusters per day and spasms per cluster with final outcomes. Only 1 patient had improved developmental milestones on follow-up. The patient, on presentation at 9 months old, was not able to grasp objects or babble, and sat only with support. At the follow-up visit at 35 months of age, the patient could speak full sentences, had a pen grip, could copy a circle and self-feed with a spoon. Patients who presented with modified hypsarrhythmia on EEG had slightly better outcomes with respect to seizure cessation than those with classic hypsarrhythmia (66.7% v. 40%). Neuroimaging findings did not correlate with prognosis. A total of 37.5% (n=3) were prescribed ACTH, and 50% (n=4) had oral prednisone. The best treatment response, as indicated by spasm remission, was with vigabatrin (50%), followed by ACTH, lamotrigine and levetiracetam (33.3%). Our facility has access only to tetracosactide â&#x20AC;&#x201C; a long-acting corticotrophin preparation administered intramuscularly.[12] Patients received 20 units (0.25 mg), which was appropriate for the age and weight of patients <1 year of age, daily for the first 3 days, and then on alternate days for 5 doses. A total of 14 days of treatment accounts for a single course. This low-dose regimen induced a good response to treatment, with spasm cessation in 2 out of 3 patients, and spasm cessation occurred within a week of commencing ACTH. One patient failed to respond owing to an incomplete course of ACTH, as the drug was not available. The oral dose for prednisone used was 2 mg/kg daily, tapered after 2 weeks. Four patients who received oral prednisone had the following outcomes: 2 were spasm-free, 1 had a partial response and 1 had no response. The favourable response was noted after a day of treatment.

137

SAJCH

Three patients who were not prescribed ACTH or prednisone were on dual therapy of sodium valproate and topiramate, and their outcomes were as follows: 2 had complete cessation of spasms, and 1 had a partial response. The patients with spasm cessation were both on sodium valproate and topiramate, while the 1 with partial response was on sodium valproate and lamotrigine.

Follow-up consultations

The patients with WS were regularly reviewed at the outpatients department after discharge, for an average of 29.4 (range 2 - 120) months. Patients presented with the following seizures: spasms (n=1), generalised tonic clonic (n=2), myoclonic (n=1), mixed seizures (n=1), i.e. both spasms and myoclonic seizures, and seizure-free (n=2). The patients with WS showed improvement in gross and fine motor skills on clinical assessment, but a worsening of language and social skills on follow-up. The EEG findings on follow-up, after a median (range) of 25.5 (2 - 108) months were: normal (n=2), general spike wave (n=3), generalised epileptiform activity (n=1), slow for age and right focus (n=1).

Discussion

This retrospective study describes the profile and outcomes of patients with WS, who accounted for 0.4% of all patients admitted with seizures at our institution. There was a male predominance, in keeping with studies in Sweden and Iceland that showed a 1:3 female to male ratio.[13,14] Patients who had a normal perinatal course had a better outcome, with 75% having spasm remission, as compared with 25% with an abnormal perinatal course. The cause of WS related to a perinatal insult was 50% in our study population. This is higher than that documented in studies from developed countries, but is similar to that found in other developing countries.[7,11] We also noted that there was no correlation between HIV exposure and outcome. The mean age of onset and diagnosis of WS was similar to that in both developed and developing countries, where a peak incidence

OCTOBER 2017 Vol. 11 No. 3

RESEARCH Table 3. Summary of the investigations, aetiology and final outcomes of our cohort of patients with West syndrome Patient 1

3 (demised)

Admission EEG

Hyps

Modified hyps

Hyps

Modified hyps

Hyps

Modified hyps

Hyps

Neuroimaging results

Global atrophy

Corpus callosum agenesis

Hamartomas and subependymal astrocytoma (TS)

Normal

Corpus callosum agenesis

Left focal cortical dysplasia

Normal

Tuberous sclerosis

Yes

Aetiology

Neonatal hypoglycaemia

Kernicterus

Neuromigratory disorder

Tuberous sclerosis

Idiopathic

Neuromigratory disorder

HIE

Idiopathic

Final outcomes of spasms

Partial response

Spasm cessation

Partial response

Seizure type on followup

Spasms and GTC

GTC

Tonic

Spasms

Seizure-free

Unknown semiology

Seizure-free

Myoclonic

Time to follow-up seizures (months)

120

Follow-up EEG

Generalised spike wave

Epileptiform activity

Normal

Generalised spike wave

Not conducted

Generalised spike wave

Slow for age â&#x20AC;&#x201C; right focus

Time to follow-up EEG (months)

108

Developmental assessment on followup

Delay: sits with support

Global delay: single words + sits with support

Global delay: not babbling + no grasp

Global delay: only coos + not smiling

Global delay: babbling only + not walking

Global Delay: only smiles + not sitting with support

Global delay: sitting with support + not reaching + no words

Global delay: sits without support + no pincer grasp + babbles only

EEG = electroencephalogram; Hyps = hypsarrhythmia; TS = tuberous sclerosis; HIE = hypoxic ischaemic encephalopathy; GTC = generalised tonic clonic seizures.

of epileptic spasms occurs between 3 and 7 months.[2,11] The average length of time between spasm onset and diagnosis was 3.1 months. This was greater than that in developed countries, which showed a lag time of 25 - 45 days, but less than in other developing countries, at 7.9 months.[11,15] There was no correlation between the number of spasms per cluster or clusters per day and final neurological outcome in our study. The number of spasms at presentation was similar to that recorded in other studies.[4] All our patients had resolution of hypsarrhythmia when assessed on follow-up visits. Two patients had normal EEG pattern, and 6 patients had evolution to other epileptiform activities. The natural progression of hypsarrhythmia is to resolve with brain maturation.[16] Evolution of hypsarrhythmia to other interictal patterns has been reported, and was noted in our study.[4] In our study, 2 patients had seizure remission on follow-up consultation, and 2 had persistent infantile spasms. The remainder had evolved to other seizure types. Spasms in WS rarely persist after 5 years of age. Fifty percent of epileptic spasms cease at ~3 years of age. In 60% of patients with WS, the spasms may evolve into other seizure types.[17] Studies show that 27% of patients with WS evolve to Lennox-Gastaut syndrome (LGS), and that 40% of patients with LGS had WS.[18] This evolution of WS to LGS was not evident in our study, since the follow-up period of patients was rather short, and we had few patients.

138

SAJCH

Results of published studies have revealed that 70% of patients with WS have symptomatic WS and an identifiable aetiology on neuroimaging.[19] This was similar in our study, with 6/8 patients having an underlying cause. Previous studies have shown that patients with identified normal neuroimaging have a better clinical outcome.[20] However, this was not evident in our study, which showed a 50% partial response to spasms in patients with a normal MRI. The 2010 US Consensus Report on infantile spams concluded that treatment with steroids was beneficial.[2] The overall response rate in our study was in keeping with other studies[2,9] that recommended the use of ACTH and prednisone for treatment of spasms in WS. The patient who did not respond to prednisone had focal cortical dysplasia on MRI. None of our patients who received either ACTH or prednisone developed sideeffects to the drugs. This was not in keeping with studies that showed an 85% incidence of adverse effects in WS patients treated with steroids, with an increased risk at higher doses (maximum doses of prednisone 60 mg/day and ACTH 60 IU on alternate days). The adverse effects reported included increased appetite, irritability, hypertension and glycosuria.[7,21] The low dose used in our patients possibly minimised the adverse effects. Vigabatrin was prescribed for 2 patients, and there was a 50% cessation of spasms. This was in keeping with a placebo-controlled trial that showed a 68% spasm cessation rate.[22] Vigabatrin has been shown to be the treatment of choice in patients with tuberous sclerosis as a cause

OCTOBER 2017 Vol. 11 No. 3

RESEARCH Table 4. Summary of treatments and doses used in our patients with West syndrome

Topiramate Vigabatrin (mg/kg/ (mg/kg/ Levetiracetam day) day) (mg/kg/day)

Lamotrigine (mg/kg/day)

ACTH (mg) daily for 3 days then alternate days for 5 days

Patient

Final outcomes

Sodium valproate (mg/kg/ day)

Partial response

8.3

0.25

Partial response

2.6

Spasm cessation (demised)

0.25

Spasm cessation

1.7

Spasm cessation

24.5

Spasm cessation

2.8

Partial response

0.25

Partial response

7.4

2.1

Prednisone (mg/kg/ day)

NU = not used.

for WS. Several studies have demonstrated that patients with tuberous sclerosis treated with vigabatrin had complete remission of spasms.[2,9,23] Vigabatrin has not been found to be superior to ACTH or prednisone. Vigabatrin is not the first-line therapy in WS patients without tuberous sclerosis, because of the associated adverse effects of irreversible retinal dysfunction and concentric visual-field constriction.[9,24] The treatment-algorithm strategy in our institution for WS was initiation with sodium valproate, followed by adjunctive therapy of topiramate or lamotrigine, and escalation to levetiracetam until a definitive diagnosis of WS is made. Studies on the use of sodium valproate for WS show inconsistent efficacious outcomes. The current literature shows a positive response rate of 73% to sodium valproate therapy within 2 weeks.[25] Topiramate as add-on therapy was used in an open-label study in 11 patients with refractory WS.[26] The results of the study showed that 50% had spasm cessation, and an additional 4 patients showed a 50% reduction in spasm frequency. Research evidence on the use of levetiracetam is limited, and inconclusive on its efficacy as monotherapy in WS.[27,28] In our study, we found that all 3 patients treated with levetiracetam continued to have seizures, and were subsequently treated with ACTH.

Study limitations

The limitations of the study are that this was a retrospective study with a limited number of patients from a single centre. We found a male predominance, with a higher incidence of perinatal comorbidity, in our cohort of patients. Epileptic spasms had an earlier age of onset, with a longer duration from onset of spasms to time of diagnosis, as compared with developed countries. This can be attributed to the poor clinical identification of WS, and hence the delay in referral to our centre for further investigations and treatment. Most of our patients had an underlying diagnosis and abnormal neuroimaging, with neuromigratory disorders being the most common finding. Half of our patients responded to our treatment algorithm. Lack of access to ACTH for treatment limited the chances of a favourable outcome in our patients. Only 2 patients (25%) remained seizure-free on follow-up, compared with studies from the USA (46%)[29] and the UK (75%).[7] The mortality rate was 12.5% in this study.

Conclusion

WS is a rare epileptiform encephalopathy, and a high index of suspicion is needed for identification. This is the first published study to describe the

139

SAJCH

clinical profile and outcome of children with WS in Africa. The diagnosis of WS is usually delayed, owing to late referral to specialist care and lack of resources, and treatment options are limited by the availability of medication. Prospective studies with larger patient numbers are required to evaluate better treatment practices and outcome measures in Africa. This will allow for the formulation of standardised treatment guidelines for resource-poor settings, and improve the outcomes of children with WS. Acknowledgments. The Department of Paediatric Neurology medical staff at Inkosi Albert Luthuli Central Hospital for the management of the patients. Author contributions. Data collection, extraction and drafting of manuscript was done by AK. LM and MA supervised the project, and read and corrected the manuscript. Data analysis was performed by NYZ, a biostatistician. Drafts were then read and approved by all authors. Funding. None. Conflict of interest. None. 1. West WJ. On a peculiar form of infantile convulsions. Letter to the editor. Lancet 1841;1:724-725. 2. Pellock JM, Hrachovy R, Shinnar S, et al. Infantile spasms: A US consensus report. Epilepsia 2010;51(10):2175-2189. https://doi.org/10.1111/j.15281167.2010.02738.x 3. Kellaway P, Hrachovy RA, Frost JD Jr, Zion T. Precise characterization and quantification of infantile spasms. Ann Neurol 1979;6(3):214-218. https://doi. org/10.1002/ana.410060306 4. Hrachovy RA, Frost JD Jr. Infantile epileptic encephalopathy with hypsarrhythmia (infantile spasms/West syndrome). J Clin Neurophysiol 2003;20(6):408-425. https://doi.org/10.1097/00004691-200311000-00004 5. Riikonen R. Epidemiological data of West syndrome in Finland. Brain Dev 2001;23(7):539-541. https://doi.org/10.1016/s0387-7604(01)00263-7 6. Riikonen R. Long-term outcome of patients with West syndrome. Brain Dev 2001;23(7):683-687. https://doi.org/10.1016/s0387-7604(01)00307-2 7. Lux AL, Edwards SW, Hancock E, et al. The United Kingdom Infantile Spasms Study (UKISS) comparing hormone treatment with vigabatrin on developmental and epilepsy outcomes to age 14 months: A multicentre randomised trial. Lancet Neurol 2005;4(11):712-717. https://doi.org/10.1016/s1474-4422(05)70199-x 8. Ito M, Seki T, Takuma Y. Current therapy for West syndrome in Japan. J Child Neurol 2000;15(6):424-428. https://doi.org/10.1177/088307380001500615 9. Wilmshurst JM, Gaillard WD, Vinayan KP, et al. Summary of recommendations for the management of infantile seizures: Task Force Report for the ILAE Commission of Pediatrics. Epilepsia 2015;56(8):1185-1197. https://doi. org/10.1111/epi.13057 10. Young C. National survey of West syndrome in Taiwan. Brain Dev 2001;23(7):570574. https://doi.org/10.1016/s0387-7604(01)00271-6

OCTOBER 2017 Vol. 11 No. 3

RESEARCH 11. Kaushik JS, Patra B, Sharma S, Yadav D, Aneja S. Clinical spectrum and treatment outcome of West Syndrome in children from Northern India. Seizure 2013;22(8):617-621. https://doi.org/10.1016/j.seizure.2013.04.014 12. Hamano S, Yamashita S, Tanaka M, Yoshinari S, Minamitani M, Eto Y. Therapeutic efficacy and adverse effects of adrenocorticotropic hormone therapy in West Syndrome: Differences in dosage of adrenocorticotropic hormone, onset of age, and cause. J Pediatr 2006;148(4):485-488. https://doi.org/10.1016/j. jpeds.2005.11.041 13. Sidenvall R, Eeg-Olofsson O. Epidemiology of infantile spasms in Sweden. Epilepsia 1995;36(6):572-574. https://doi.org/10.1111/j.1528-1157.1995. tb02569.x 14. LuoAvigsson P, Olafsson E, Sigurthardottir S, Hauser WA. Epidemiologic features of infantile spasms in Iceland. Epilepsia 1994;35(4):802-805. https:// doi.org/10.1111/j.1528-1157.1994.tb02514.x 15. Lagae L, Verhelst H, Ceulemans B, et al. Treatment and long term outcome in West syndrome: The clinical reality. A multicentre follow up study. Seizure 2010;19(3):159-164. https://doi.org/10.1016/j.seizure.2010.01.008 16. Wong M, Trevathan E. Infantile spasms. Pediatric Neurol 2001;24(2):89-98. https://doi.org/10.1016/s0887-8994(00)00238-1 17. Riikonen R. A long-term follow-up study of 214 children with the syndrome of infantile spasms. Neuropediatrics 1982;13(1):14-23. https://doi. org/10.1055/s-2008-1059590 18. Rantala H, Putkonen T. Occurrence, outcome, and prognostic factors of infantile spasms and Lennox-Gastaut syndrome. Epilepsia 1999;40(3):286-289. https://doi.org/10.1111/j.1528-1157.1999.tb00705.x 19. Wirrell EC, Shellhaas RA, Joshi C, et al. How should children with West syndrome be efficiently and accurately investigated? Results from the National Infantile Spasms Consortium. Epilepsia 2015;56(4):617-625. https://doi. org/10.1111/epi.12951

140

SAJCH

20. Saltik S, Kocer N, Dervent A. Informative value of magnetic resonance imaging and EEG in the prognosis of infantile spasms. Epilepsia 2002;43(3):246-252. https://doi.org/10.1046/j.1528-1157.2002.14001.x 21. Riikonen R, Donner M. ACTH therapy in infantile spasms: Side effects. Arch Dis Child 1980;55(9):664-672. https://doi.org/10.1136/adc.55.9.664 22. Appleton RE, Peters AC, Mumford JP, Shaw DE. Randomised, placebocontrolled study of vigabatrin as first-line treatment of infantile spasms. Epilepsia 1999;40(11):1627-1633. https://doi.org/10.1111/j.1528-1157.1999.tb02049.x 23. Granstrom ML, Gaily E, Liukkonen E. Treatment of infantile spasms: Results of a population-based study with vigabatrin as the first drug for spasms. Epilepsia 1999;40(4):950-957. https://doi.org/10.1111/j.1528-1157.1999.tb00802.x 24. Koul R, Chacko A, Ganesh A, Bulusu S, Al Riyami K. Vigabatrin associated retinal dysfunction in children with epilepsy. Arch Dis Child 2001;85(6):469473. https://doi.org/10.1136/adc.85.6.469 25. Siemes H, Spohr HL, Michael T, Nau H. Therapy of infantile spasms with valproate: Results of a prospective study. Epilepsia 1988;29(5):553-560. https:// doi.org/10.1111/j.1528-1157.1988.tb03760.x 26. Glauser TA, Clark PO, Strawsburg R. A pilot study of topiramate in the treatment of infantile spasms. Epilepsia 1998;39(12):1324-1328. https://doi. org/10.1111/j.1528-1157.1998.tb01331.x 27. Gümüş H, Kumandaş S, Per H. Levetiracetam monotherapy in newly diagnosed cryptogenic West syndrome. Pediatr Neurol 2007;37(5):350-353. https://doi. org/10.1016/j.pediatrneurol.2007.06.019 28. Mahmoud AA, Rizk TM, Mansy AA, Ali JA, Al-Tannir MA. Ineffectiveness of topiramate and levetiracetam in infantile spasms non-responsive to steroids. Open labeled randomised prospective study. Neuroscience 2013;18(2):143146. https://doi.org/10.1016/j.jns.2013.07.2038 29. Knupp KG, Coryell J, Nickels KC, et al. Response to treatment in a prospective national infantile spasms cohort. Ann Neurol 2016;79(3):475-484. https://doi. org/10.1002/ana.24594

OCTOBER 2017 Vol. 11 No. 3

RESEARCH

This open-access article is distributed under Creative Commons licence CC-BY-NC 4.0.

Fanconi anaemia in South African patients with Afrikaner ancestry C Feben,1 MB BCh, DCH, MMed, FCMG (SA); T Haw,1 MSc (Med); D Stones,2 FCPaed (SA); C Jacobs,3 MB MCh; C Sutton,4 MB MCh, DTM&H, DPH, DCH, FCPaed (SA); J Kromberg,1 PhD; A Krause,1 PhD ivision of Human Genetics, National Health Laboratory Service, and Division of Human Genetics, School of Pathology, Faculty of Health D Sciences, University of the Witwatersrand, Johannesburg, South Africa 2 Department of Paediatrics, Universitas Hospital; and University of the Free State, Bloemfontein, South Africa 3 Unitas Hospital, Pretoria, South Africa 4 Department of Paediatrics, Polokwane Mankweng Hospital Complex, Polokwane and University of Limpopo, Polokwane, South Africa 1

Corresponding author: C Feben (candice.feben@nhls.ac.za) Background. Fanconi anaemia (FA) is a rare genetic disorder of impaired DNA repair that results in physical and haematological consequences in affected individuals. In South Africa (SA), individuals with Afrikaner ancestry are at an increased risk of inheriting disease-causing FA mutations, owing to the three common FANCA (FA, complementation group A) founder mutations present in this population subgroup. Objectives. To describe the physical phenotype of SA patients with FANCA mutations for the purpose of recommending appropriate care for affected individuals. Methods. A structured clinical examination and file-based review were used to evaluate the physical phenotype of 7 patients with compound heterozygous and homozygous FANCA founder mutations, and 1 patient with confirmed FANCA complementation analysis. Descriptive statistical analysis was used to determine the frequency of physical anomalies in Afrikaner patients and to compare the described phenotype to other FA cohorts, including a previously clinically characterised black SA FA cohort. Results. An earlier age of diagnosis of FA in Afrikaner patients, a high frequency of somatic anomalies and a higher-than-expected incidence of the VACTERL/H phenotype were noted. Conclusions. Based on our findings, recommendations for the care of FA patients with Afrikaner ancestry are made, including renal ultrasound evaluation at diagnosis and hearing screening. S Afr J Child Health 2017;11(2):141-145. DOI:10.7196/SAJCH.2017.v11i2.1312

Fanconi anaemia (FA) is a rare genetic condition of impaired DNA repair mechanisms and chromosomal instability. Although most cases are inherited in an autosomal recessive manner, both autosomal dominant and X-linked recessive patterns of inheritance have been described.[1-3] In general, the FA phenotype is characterised by a broad spectrum of congenital anomalies, principally involving growth, skin pigmentation and dysmorphogenesis of the skeletal, cardiovascular, genito-urinary, gastrointestinal and central nervous systems (CNS).[4] Internationally, major congenital malformations are reported in over two-thirds of patients.[5] Affected individuals are further at risk for childhood-onset haematological disease, which is characterised by initial thrombocytopenia and macrocytosis and progression to aplastic anaemia, acute myeloid leukaemia (AML) or myelodysplastic syndrome (MDS).[5] The incidence of solid tumours, particularly of squamous-cell lineage (including cancers of the head and neck, oesophagus, vulva and cervix), is also significantly increased in affected patients. These tumours are recognised particularly in developed nations, where improved treatments of the haematological complications of the disease, including haematopoietic stem-cell transplantation, are readily available.[5,6] While FA is a genotypically heterogeneous disorder, caused by mutations in at least 21 different FA genes (FANCA, B, C, D1, D2, E, F, G, I, J, L, M, N, O, P, Q, R, S, U, V),[2,3,7-9] mutations in 3 of these genes account for a high proportion of diagnosed cases â&#x20AC;&#x201C; FANCA, 60.5%; FANCC, 16%; and FANCG 10%.[5] Furthermore, certain population groups (e.g. Ashkenazi Jews, Spanish Gypsies, Japanese) have been shown to harbour founder mutations that account for a high percentage of cases in these specific groups.[10-12] In South Africa (SA), founder mutations have been characterised in both the black and caucasian Afrikaans-speaking populations.[13,14]

141

SAJCH

Black SA patients with FA have been shown to carry a founder mutation in the FANCG gene (c.637_643delTACCGCC â&#x20AC;&#x201C; a 7 base-pair (bp) deletion mutation).[14] This founder mutation, in the homozygous state, accounts for ~80% of affected black patients, and has resulted in an estimated birth incidence of FA in black South Africans nearing 1/40 000.[14] The physical and haematological phenotypes of black patients, homozygous for the 7 bp deletion mutation, have been characterised previously.15,16] In SA individuals with Afrikaner ancestry, three null mutations in FANCA (del E12-31 (deletion mutation of exons 12 to 31); del E11-17 (deletion mutation of exons 11 to 17) and 3398delA (point mutation at position 3398)) have previously been shown to account for ~80% of FA cases.[13] Molecular and genealogical evidence has confirmed that these founder FANCA mutations were probably introduced into SA following the 17th century migration of the French Huguenots to the Cape.[13] Based on birth incidence and point-prevalence data, respectively, the prevalence of FA in individuals with Afrikaner ancestry has previously been estimated to approximate between 1/22 000 and 1/26 000.[17] Despite the predicted relative high prevalence of FA in Afrikaner individuals, there is little published work on the physical and haematological phenotypes of these patients.[17,18] The most comprehensive study was performed in 1994,[18] prior to the introduction of molecular testing methods to characterise specific mutations and therefore no detailed mutation-specific genotype-phenotype correlations could be drawn. The aims of the present study were firstly, to describe the physical phenotype of affected Afrikaner patients with confirmed founder FANCA mutations, and secondly, to compare this phenotype with the physical phenotype in black South African patients with FA caused by the homozygous FANCG deletion mutations described by Feben et al.[15] and

OCTOBER 2017 Vol. 11 No. 3

RESEARCH other FA cohorts with mutations in FANCA. An improved recognition of the physical phenotype of FA in patients with Afrikaner ancestry may assist in improving the early recognition and diagnosis of the condition, and subsequently, the care offered to affected patients.

Methods

Patients for this study were recruited from tertiary-level haematology/ oncology clinics in the cities of Pretoria, Bloemfontein and Polokwane, which are situated in 3 of the 9 provinces of SA, between October 2009 and July 2012. Altogether, 8 (self-reported ancestry) Afrikaner patients were recruited, of whom 7 were homozygous or compound heterozygous for the FANCA founder mutations. The 8th patient had positive chromosome-breakage testing with confirmed FANCA complementation analysis, but did not consent to further molecular testing. Molecular genetic testing, combining polymerase chain reaction (PCR) and multiplex ligation probe specific amplification (MLPA) techniques, is used as the initial testing method in Afrikaner patients suspected to have FA, in view of the high percentage of cases caused by founder mutations in SA (personal communication, Prof. A Krause, January 2016). Molecular genetic testing of the 7 patients was performed by the Molecular Genetics Laboratory at the National Health Laboratory Services in Johannesburg, SA, with assistance from a collaborating laboratory in one case (Sheffield Diagnostic Genetics Services, UK). A comprehensive clinical examination and a concurrent review of each patientâ&#x20AC;&#x2122;s hospital records were completed. The clinical examination aimed to document growth measurements (current weight, height and head circumference); upper-limb and lower-limb anomalies, with particular focus on abnormalities of the radial ray and hands, and skin-pigmentary abnormalities. All measurements were plotted on relevant growth charts for comparison with the age-related mean. Major malformations of the renal, genital and cardiovascular systems were documented, as well as the results of hearing-screening tests. Information from the patientsâ&#x20AC;&#x2122; hospital records was used to document the age of presentation at the haematology/oncology clinic, and the clinical symptoms at initial presentation. Descriptive statistical analysis of the data was performed. The median current age and median age of presentation with symptoms in keeping with FA were calculated. The frequency of somatic anomalies and growth disturbances were recorded, and compared with those of other cohorts with documented FANCA mutations, as well as with the Rosendorff et al.[17] and MacDougall et al.[18] cohorts, which represent the only major study groups comparable with the present cohort. Descriptive comparisons were also made with the previously published data on the physical phenotypic characteristics in black South African patients with FA caused by a homozygous FANCG founder mutation.[15] A modified International FA Registry (IFAR) score (using growth retardation, renal anomalies, thumb and/or radius anomalies, microphthalmia and birthmarks (each scoring positive 1 if present) and other skeletal anomalies (scoring negative 1 if present, for a maximum score of 5)) was calculated for each patient. The IFAR score was initially described by Auerbach et al.[19] as a screening tool to predict the likelihood of an individual having FA. The original score comprises the above five criteria, as well as low platelet count (score positive 1 if present) and developmental delay (score negative 1 if present), giving a maximum score of 6.[19] An adjusted score was used, as we did not have platelet counts and neurodevelopmental assessments for each patient. Additionally, a modified pigmentation, small head, small eyes, CNS (not hydrocephalus), otology, and short stature (PHENOS) score was calculated for each patient. The PHENOS score, developed by Alter and Giri,[20] allocates a score of positive 1 for each of: pigmentation anomaly, microcephaly (small head), microphthalmia (small eyes), CNS anomalies (other than hydrocephalus), otologic anomalies (structural ear anomalies or hearing loss) and short stature, with a maximum score of 6. The score is used to identify patients with VACTERL/H association

142

SAJCH

(which includes at least three of the following cardinal abnormalities: vertebral, ano-rectal, cardiac, tracheo-oesophageal, renal, limb defects and hydrocephalus), who should be tested for FA.[20] We used a modified score as we did not have CNS imaging in our patients. The study was approved by the research ethics committees of the University of the Witwatersrand (ref. no. M090681), the University of the Free State (ref. no. ETOVS NR 52/2010) and the University of Limpopo (ref. no. PMREC-19).

Results

The median age of the 8 patients attending the haematology/oncology clinics at the time of the study was 11 years 11 months (range 4 years 1 month - 17 years 3 months). The initial presenting complaints were variable, and included bleeding diatheses (epistaxis, haematemesis), fatigue and somatic abnormalities (particularly involving the thumbs), with the median age at presentation being 3 years (range birth - 12 years). The physical phenotype of each patient (N=8) was recorded (Fig. 1). The most frequently occurring anomalies were head circumference below the 10th centile for age (75%), with head circumference below the 3rd centile for age in 3 of these patients (37.5%); radial-ray anomalies (87.5%); and skin-pigmentation abnormalities (87.5%). Two patients met the criteria for the diagnosis of VACTERL/H association. The median modified PHENOS score was 2 (range 1 - 3), with the 2 patients who met criteria for VACTERL/H association scoring 3 (patient 2) and 2 (patient 7). The median modified IFAR score was 3 (range 2 - 4). Although this score cannot be directly used to predict the likelihood of having FA, given that we did not calculate the full score (in the absence of haematological data and a neurodevelopmental assessment), the high median value points to the high frequency of congenital malformations in the present cohort. A score of 3 using the complete IFAR score would correlate with a >92% risk of having FA. The physical phenotype of the 8 patients in the present cohort was compared with the physical phenotypes of the patients in the Rosendorff et al.[17] (N=18) and MacDougall[18] (N=28) cohorts (Table 1). No significant differences were found in the frequency of anomalies between the groups (using two-tailed Fisherâ&#x20AC;&#x2122;s exact test). Most patients in the Rosendorff et al.[17] cohort had an IFAR score >4, also indicating the high number of somatic anomalies in each patient. Given that we now know the molecular basis of FA in Afrikaner patients, we can assume that at least 80% of the patients in the above 2 cohorts would have carried founder FANCA mutations as the cause of their FA. A comparison to a large (N=170) international FANCA cohort also showed no significant differences in the frequency of all evaluated anomalies (Table 1).[21] The Afrikaner FA physical phenotype was compared with that previously detailed in black SA FA patients who are homozygous for a FANCG 7 bp deletion founder mutation (N=35) (Table 2).[15] No significant differences were noted between the 2 groups regarding the frequency of the anomalies; however, the sample size of the present cohort was small. The trend appears to indicate that physical anomalies (radial defects, renal malformations) may be more common in Afrikaner FA patients with FANCA mutations, and suggests that analysis of a larger Afrikaner cohort may be beneficial in determining whether this trend is significant. The age of presentation with symptoms or signs suggestive of FA was significantly lower in Afrikaner patients than in black FA patients (median age at presentation 3 years v. 7 years 8 months, p<0.001).[15] Additionally, 3 of the Afrikaner patients presented with a diagnosis of FA before 6 months of age, on the basis of congenital anomalies, whereas none of the black patients presented before 2 years 11 months, suggesting that the physical anomalies in Afrikaner patients were clinically more apparent or more severe in presentation.[15] This is further evidenced by the differences in the description of the radial anomalies in the two groups. The radial-ray anomalies noted in black patients were described as subtle, with none of the patients showing radius hypoplasia or aplasia, whereas the Afrikaner patients had very easily identifiable radius and radial-ray defects, 2 cases having bilaterally absent thumbs.[15]

OCTOBER 2017 Vol. 11 No. 3

RESEARCH Table 1. Physical phenotypic characteristics in Afrikaner patients with FA (N=8) No. Sex

Head circumference FANCA mutation for age

Renal anomaly

Cardiac anomaly

GIA

Radial ray defect

Pigment anomaly

Hearing loss

Complementation group A; DEB+

>10th centile

Left horseshoe kidney; right renal agenesis

Absent right 1st metacarpal, hypoplastic right thumb

CAL

Bilateral severe mixed loss

Exon 12-31 del homozygous

<3rd centile

Left renal agenesis

Tracheooesophageal fistula

Bilateral hypoplastic radii, bilateral absent 1st metacarpal and thumb

Bilateral severe conductive

Exon 12-31 del; C2533_2536 del

>10th centile

Bilateral hypoplastic 1st metacarpals

CAL; hyperpigmented macules

Bilateral moderate conductive

Exon 12-31 del homozygous

<3rd centile

Ventricular septal defect

Bilateral absent thumbs

Hypopigmented macules

Exon 11-17 del; 3398delA

3rd - 10th centile

Bilateral hypoplastic 1st metacarpals and proximally inserted thumbs

CAL; hypopigmented macules

Exon 12-31 del homozygous

<3rd centile

CAL; hypo-and hyperpigmented macules

Exon 12-31 del homozygous

3rd - 10th centile

Bilateral pelvic kidneys, vesicoureteric reflux

Patent ductus arteriosus

Hypoplastic left radius and left thumb

CAL; hypopigmented macules

Exon 11-17 del homozygous

3rd - 10th centile

Pelvic kidney

Bilateral hypoplastic thumbs

CAL; hyperpigmented macules

FA = Fanconi anaemia; GIA = Gastrointestinal anomaly; M = male; DEB+= diepoxybutane positive (chromosome breakage); del = deletion; - = not observed; - = not observed; CAL = cafĂŠ au lait macule; del = deletion; F = female. *Patients meeting criteria for VACTERL association.

Discussion

FANCA mutations account for the highest percentage of FA patients worldwide as documented by the IFAR. Numerous common and familyspecific deleterious mutations have been reported in this gene, which appears to function as part of the core nuclear complex in the FA pathway.[22] Previous research efforts have variably focused on complementationspecific and mutation-specific genotype-phenotype correlations in order to better elucidate the physical and haematological consequences of mutations in FANCA. Clinical phenotyping aims to expedite diagnosis and improve care and management of affected individuals. Two large international studies characterised the clinical phenotype of individuals with different mutations in the FANCA gene. The initial study carried out in 2000 found that FANCA individuals with homozygous null mutations had a higher frequency of somatic anomalies than individuals with FANCC mutations, but a similar frequency to those with FANCG mutations. FANCA individuals were also at increased risk for severe haematological disease. A number of the patients in this study were of Afrikaner ancestry and had at least one null del1231 mutation.[21] In 2011, research into the functional role and clinical impact of FANCA mutations challenged the earlier findings. They

143

SAJCH

demonstrated no correlation between mutation type and physical or haematological outcome, indicating that mutation-specific genotypephenotype correlations may not be useful in individuals with FANCA mutations. The authors concluded that mutation type had little prognostic value in FANCA patients and that other factors, including genetic background, ethnicity and environmental exposure were more important in determining the clinical outcome.[12] Despite the different conclusions drawn in these two studies, they both support the importance of clinical phenotyping in individuals with FANCA mutations. More specifically, the later study suggested that characterisation of the FA phenotype based on ethnicity may be as important as mutation-specific phenotype correlations, and that population-specific phenotypes may exist independently of the specific mutation/s.[12] A similar focus on the underlying importance of ethnicity was highlighted by Futaki et al.[23] The study showed that a specific mutation (IVS4+4A>T) in the FANCC gene caused a significantly milder clinical phenotype in Japanese patients than in Ashkenazi Jewish patients. Previous SA research studies documented the physical phenotype of Afrikaner patients with FA based on home language and self-reporting; however, at this point molecular pathogenesis of the condition had not

OCTOBER 2017 Vol. 11 No. 3

RESEARCH Table 2. Phenotype comparison between present Afrikaner cohort, two other Afrikaner FA cohorts and an international FANCA cohort Anomaly

Afrikaner FA cohort (N=8), n (%)

Rosendorff cohort[17] (N=18), n/N (%)

MacDougall cohort[18] (N=28), n (%)

Faivre FANCA cohort[20] (N=170), n/N (%)

Pigmentary anomalies

7 (87.5)

16/18 (89)

25 (89)

124/167 (74)

Short stature

5 (62.5)

12/14 (86)

20 (71)

98/164 (60)

Microcephaly

6 (75)

9/15 (60)

22 (78)

Genito-urinary anomalies

4 (50)

4/11 (36)

7 (24)

33/170 (19)

Radial ray anomalies

7 (87.5)

12/18 (67)

24 (85)

83/170 (49)

Cardiac defects

2 (25)

3 (11)

16/170 (9.4)

Hearing loss

3 (37.5)

19/159 (11)

FA = Fanconi anaemia; FANCA = FA, complementation group A.

been characterised.[17,18] It is now known that the majority of Afrikaner patients with FA harbour founder FANCA mutations as the cause of their condition. The results of the present study confirm the high number of somatic anomalies found in Afrikaner patients with FA due to founder mutations. They also show that Afrikaner patients are diagnosed at a significantly younger age than black patients in SA. While this may partially reflect social factors like access to tertiary healthcare, it is important to note that the somatic anomalies in Afrikaner patients are often recognisable and obvious at birth, and that these anomalies should prompt consideration of an FA diagnosis and further investigations.[15] The results support the use of the IFAR scoring system as an important screening tool for FA in Afrikaner patients. The use of screening tools is particularly important in countries, such as SA, where access to genetic services and particularly molecular genetic testing is limited by poor resources and funding. Once an individual is suspected of having FA, the results in the present cohort suggest that these individuals, at least, should be offered renal ultrasound screening and hearing tests as part of their care. Two patients met the clinical criteria for a diagnosis of VACTERL/H association. Previous reports differed vastly in their reporting of the percentage of patients with FANCA mutations who meet the diagnostic criteria for VACTERL/H association, with figures of between 3% and 20%.[20,24] Although the VACTERL/H phenotype appears to be more often associated with mutations in FANCD1 (33%), E (40%) and F (30%), more recent work by Alter and Giri[20] suggests that the co-occurrence of this phenotype with FA is under-recognised.[24] The present sample size is however small and the results suggest that Afrikaner patients with a VACTERL/H phenotype should always be tested for FA. Future studies with a larger cohort will allow a more accurate frequency measurement of the VACTERL/H phenotype in this population subgroup. Further, the small sample size questions the viability of the PHENOS score in Afrikaner patients. Even so, this score may be useful in stratifying patients with the VACTERL/H phenotype for FA testing in future studies. Despite the high frequency of FA and high incidence of congenital anomalies in the Afrikaner population subgroup, anecdotal evidence suggests that very few affected patients ever attend a genetic clinic or have their clinical diagnosis confirmed with a molecular test in SA. Certainly, in our study, we were only able to ascertain 7 patients with molecular confirmation of their condition despite accessing three tertiary healthcare centres in three cities in SA. The reason for this underrecognition and under-diagnosis of FA in the Afrikaner population in SA remains unclear. It is concerning that the patient referral rate to genetic services, tertiary haematology and oncology services is so low. It is possible that a more severe end of the spectrum exists, with affected individuals dying at a young age, before consideration of the diagnosis of FA (possible misdiagnosis of isolated VACTERL/H association); or

144

SAJCH

Table 3. Comparison between Afrikaner and black FA phenotype in South Africa

Anomaly

Afrikaner FA cohort (N=8), n/N (%)

Black FA cohort[15] (N=35), n /N (%)

p-value

Radial ray anomalies

7/8(87.5)

26/35 (74)

0.6563

Pigmentary anomalies

7/8 (87.5)

34/35(97)

0.3411

Renal malformations

4/8 (50)

12/31 (37)

0.6937

Hearing loss

3/8 (37.5)

1/9 (11)*

0.2941

Age at presentation (median)

3 years

7 years, 8 months

<0.001

FA = Fanconi anaemia. *Unpublished data.

that a much milder undiagnosed phenotype exists with few congenital anomalies and limited haematological complications. An area for future research includes a molecular screen for FANCA mutations in Afrikaner patients with VACTERL/H association.

Conclusion

The results indicate that Afrikaner patients with FA have a high incidence of congenital anomalies and yet very few have their diagnoses confirmed on a molecular level. As such, we recommend that all children with one or more significant or major congenital anomaly (a birth defect that may cause death or disability) be referred for a tertiary evaluation by a paediatrician or medical geneticist. The diagnosis of FA should be considered on the basis of the congenital malformation, especially in the presence of other clinical features (particularly skin pigmentary anomalies) to diagnose FA prior to the onset of haematological disease. In the SA setting, molecular FANCA analysis of the three founder mutations would be the first line of investigation to confirm the diagnosis in an individual with Afrikaner heritage. Through earlier diagnoses, we can improve care and surveillance, provision of genetic counselling (including recurrence risk counselling) and provide an opportunity for the affected patient to be considered for haematopoietic stem cell transplantation. Acknowledgements. The authors wish to thank the clinical, counselling, laboratory and support staff in the Division of Human Genetics (National Health Laboratory Service and The University of the Witwatersrand), clinical and support staff at Universitas Hospital, Polokwane/Mankweng Hospital Complex and Unitas Hospital and the patients and their families who agreed to participate in this research. We thank our partners at the Sheffield Regional Genetics Laboratory for their assistance.

OCTOBER 2017 Vol. 11 No. 3

RESEARCH Author contributions. Study design and conception: CF, AK, TH, JK; data collection and analysis: CF, AK, TH, JK, DS, CJ, CS; clinical collaboration: DS, CJ, CS; writing of article: CF, JK, AK; editing: CF, JK Funding. Medical Research Council of South Africa. Conflicts of interest. None. 1. Meetei AR, Levitus M, Xue Y, et al. X-linked inheritance of Fanconi anaemia complementation group B. Nat Genet 2004;36:1219-1224. https://doi. org/10.1038/ng1458 2. Alter B, Kupfer G. Gene Reviews: Fanconi Anemia. Bethesda: National Center for Biotechnology Information, 2013. https://www.ncbi.nlm.nih.gov/books/ NBK1401/ (accessed 15 July 2016). 3. Ameziane N, May P, Haitjema A, et al. A novel Fanconi anaemia subtype associated with a dominant-negative mutation in RAD51. Nat Commun 2015;6:8829. https://doi.org/10.1038/ncomms9829 4. Shimamura A, Alter B. Pathophysiology and management of inherited bone marrow failure syndromes. Blood Rev 2010;24(3)101-122. https://doi. org/10.1016/j.blre.2010.03.002 5. Auerbach A. Fanconi anaemia and its diagnosis. Mut Res 2009;668(1-2):4-10. https://doi.org/10.1016/j.mrfmmm.2009.01.013 6. Rosenberg PS, Greene MH, Alter BP. Cancer incidence in persons with Fanconi anaemia. Blood 2003;101(3):822-826. https://doi.org/10.1182/blood-2002-05-1498 7. Sawyer SL, Tian L, Kahkonen M, et al. Biallelic mutations in BRCA1 cause a new Fanconi anaemia subtype. Cancer Discoveries 2015;5(2):135-142. https:// doi.org/10.1158/2159-8290.cd-14-1156 8. Park JY, Virts EL, Jankowska A, et al. Complementation of hypersensitivity to DNA interstrand crosslinking agents demonstrates that XRCC2 is a Fanconi anaemia gene. J Med Genet 2016;53(10):672-680. https://doi.org/10.1136/ jmedgenet-2016-103847 9. Bluteau D, Masliah-Planchon J, Clairmont C, et al. Biallelic inactivation of REV7 is associated with Fanconi anaemia. J Clin Invest 2016;126(9):3580-3584. https://doi.org/10.1172/jci88010 10. Whitney MA, Saito H, Jakobs PM, Gibson RA, Moses RE, Grompe M. A common mutation in the FACC gene causes Fanconi anaemia in Ashkenazi Jews. Nat Genet 1993;4(2):202-205. https://doi.org/10.1038/ng0693-202 11. Yagasaki H, Oda T, Adachi D, et al. Two common founder mutations of the fanconi anaemia group G gene FANCG/XRCC9 in the Japanese population. Hum Mut 2003;21(5):555. https://doi.org/10.1002/humu.9142 12. Castella M, Pujol R, Callen E, et al. Origin, functional role and clinical impact of Fanconi anaemia FANCA mutations. Blood 2011;117(14):3759-3769. https://doi.org/10.1182/blood-2010-08-299917

145

SAJCH

13. Tipping AJ, Pearson T, Morgan NV, et al. Molecular and genealogical evidence for a founder effect in Fanconi anaemia families of the Afrikaner population of South Africa. Proc Natl Acad Sci USA 1993;98(10):5734-5739. https://doi. org/10.1073/pnas.091402398 14. Morgan NV, Essop F, Demuth I, et al. A common Fanconi anaemia mutation in black populations of sub-Saharan Africa. Blood 2005;105(9):3542-3544. https://doi.org/10.1182/blood-2004-10-3968 15. Feben C, Kromberg J, Wainwright R, et al. Phenotypic consequences in black South African Fanconi anaemia patients homozygous for a founder mutation. Genet Med 2014;16(5):400-406. https://doi.org/10.1038/gim.2013.159 16. Feben C, Kromberg J, Wainwright R, et al. Haematological consequences of a FANCG founder mutation in black South African Patients with Fanconi anaemia. Blood Cells Mol Dis 2015;54(3):270-274. https://doi.org/10.1016/j. bcmd.2014.11.011 17. Rosendorff J, Bernstein R, MacDougall L, Jenkins T, Opitz JM, Reynolds JF. Fanconi anaemia: Another disease of unusually high prevalence in the Afrikaans population of South Africa. Am J Med Genet 1987;27(4):793-797. https://doi.org/10.1002/ajmg.1320270408 18. MacDougall LG, Rosendorff J, Poole J, Cohn RJ, McElligott SE. Comparative study of Fanconi anaemia in children of different ethnic origin in South Africa. Am J Med Genet 1994;52(3):279-284. https://doi.org/10.1002/ajmg.1320520306 19. Auerbach AD, Rogatko A, Schroeder-Kurth TM. International Fanconi Anaemia Registry: Relation of clinical symptoms to diepoxybutane sensitivity. Blood 1989;73:391-396. 20. Alter B, Giri N. Thinking of VACTERL-H? Rule out Fanconi anaemia according to PHENOS. Am J Med Genet 2016;170(6):1520-1524. https://doi. org/10.1002/ajmg.a.37637 21. Faivre L, Guardiola P, Lewis C, et al. Association of complementation group and mutation type with clinical outcome in Fanconi anaemia. Blood 2000;96:4064-4070. 22. Levran O, Diotti R, Pujara K, Batish SD, Hanenberg H, Auerbach AD. Spectrum of sequence variations in the FANCA gene: An International Fanconi Anaemia Registry (IFAR) study. Hum Mut 2005;25(2):142-149. https://doi. org/10.1002/humu.20125 23. Futaki M, Yamashita T, Yagasaki H, et al. The IVS4+4A>T mutation of the Fanconi anaemia gene FANCC is not associated with a severe phenotype in Japanese patients. Blood 2000;95:1493-1498. 24. Faivre L, Portnoi MF, Pals G, et al. Should chromosome breakage studies be performed in patients with VACTERL association? Am J Med Genet A 2005;137:55-58. https://doi.org/10.1002/ajmg.a.30853

OCTOBER 2017 Vol. 11 No. 3

RESEARCH

This open-access article is distributed under Creative Commons licence CC-BY-NC 4.0.

Soaps and cleansers for atopic eczema, friends or foes? What every South African paediatrician should know about their pH N C Dlova,1 MB ChB, FCDerm, PhD; T Naicker,2 PhD; P Naidoo,2 MMedSc, PhD 1 2

Dermatology Department, Nelson R Mandela School of Medicine, University of KwaZulu-Natal, Durban, South Africa Discipline of Pharmaceutical Sciences, School of Health Science, Westville Campus, Durban, South Africa

Corresponding author: N C Dlova (dlovan@ukzn.ac.za) Background. Knowledge of the pH level of soaps and cleansers used by patients with atopic eczema and sensitive skin is crucial, as high-alkalinity products are irritants and impair the normal skin barrier, so interfering with the adequate control of atopic eczema. Objectives. The aim of this study was to assess the pH of various bar soaps and cleansers that are usually recommended and used by patients with atopic diseases and dry, sensitive skin in South Africa. Methods. Forty-nine commercial soap bars and cleansers were randomly selected for pH analysis. The samples were prepared as 8% emulsions in tap water. Nine undiluted liquid facial cleansers were also evaluated. Deionised water was used as a negative control. The pH of each emulsion or liquid cleanser was recorded in duplicate using a Metrohm pH meter model 827 (Metrohm, Herisau, Switzerland). Results. Of the 49 samples analysed, 34 (69.4%) were alkaline with a pH ranging from 9.3 - 10.7. Two samples (4.1%) were within the acceptable range of (5.4 - 5.9), and 2 samples (4.1%) had pH levels of below 5. In total, 5 samples (10.2 %) had a pH of 4 - 6. Conclusion. The majority of soaps and cleansers analysed in this study were alkaline, with only 2 falling in the acceptable pH range of 5.4 - 5.9 and 5 within the pH range of 4 - 6, thus raising concerns regarding the optimal management of atopic eczema patients. S Afr J Child Health 2017;11(3):146-148. DOI:10.7196/SAJCH.2017.v11i3.1325

Atopic eczema is a chronic, inflammatory disease of the skin, distinguished by xerosis, pruritus, and erythematous lesions often resulting from a defective skin barrier, usually measured as increased transepidermal water loss (TEWL) with prevalence ranging from 10% to 20% in the First-World countries and some urban African countries.[1-5] A study undertaken in Cape Town, South Africa (SA), on atopic children, showed a prevalence rate of 8.3%, with 2.3% of children presenting with severe disease symptoms.[6] In a recent study on the epidemiology of skin conditions in 6 664 African patients in KwaZulu-Natal province, SA, eczemas were the second-most common conditions seen (15.9%), with atopic dermatitis (AD) the most common in children (7.2%).[7] A complex interaction of genetic, environmental and immunological factors has been implicated in the pathogenesis of atopic eczema.[5] Studies have demonstrated the pivotal role of epidermal barrier dysfunction in AD; it results in the down-regulation of cornified envelope genes, reduced ceramide levels in the stratum corneum (SC), elevated levels of endogenous proteolytic enzymes, and increased TEWL.[11-13] It is compounded by a lack of endogenous protease inhibitor, which perpetuates a cycle of barrier destruction.[1,2,4] Certain soaps and detergents increase endogenous protease activity, elevating SC pH levels and causing barrier dysfunction.[1,2] Knowing the pH levels of the soaps and cleansers used by patients with sensitive skin is crucial, as alkaline pH products are skin irritants. The irritation often leads to an impairment of the normal skin barrier, interfering with adequate control of atopic eczema. The skin pH values vary from 4.0 to 7.0. Although the bodyâ&#x20AC;&#x2122;s internal pH tends to be neutral to slightly alkaline, the normal adult SC is decidedly acidic, with reported values ranging from 4 - 6.[14] It has been shown that skin with pH values of <5.0 is in a better condition regarding the skin barrier function, moisturisation and scaling.[15] Lambers et al.,[15] assessed the impact of pH on adhesion of normal bacterial flora and showed that an acidic skin (pH 4.0 - 4.5) maintained the attachment of the skin commensals, whereas an alkaline skin (pH 8.0 - 9.0) encouraged the dispersal from the skin, confirming that a skin surface pH<5.0 is beneficial for resident flora.[15]

146

SAJCH

The use of detergents to clean human skin is a widespread phenomenon. It works by emulsifying the skin-surface lipids that are subsequently removed using water. Most soaps and cleansers sufficiently remove skin surface dirt; however, this may cause long-term interference of the skin-barrier function, manifesting as dry, red, itchy and inflamed skin, particularly in atopic and elderly patients.[16] Although some detergents are safe to use, highly alkaline detergents affect the physiologically protective 'acid mantle' of the skin by decreasing the fat content.[16,17] Soaps and detergents can increase skin pH, disrupting the SC and inducing irritant contact eczema and pruritus, which seem to be worse during winter, particularly in patients with atopic eczema and those with dry and sensitive skin.[2,16] The cutaneous changes are usually cumulative and indiscernible, having a greater impact on the elderly and atopic individuals.[19,20] Diligent use of moisturisers, soap substitutes and replacement of irritating wash products with moisturising ointments and oils form the backbone and are first-line therapy for atopic eczema.[21] Special soaps and cleansers purported to be safe for use in patients with AD and dry sensitive skin are readily available on the market. However, the veracity of their claims has not been investigated, as the majority of these soaps and shampoos do not disclose their pH values.[22] In SA, both paediatricians and patients recommend and purchase soaps in supermarkets and pharmacies without background knowledge of product pH levels. The aim of this study was to assess the pH of a group of soap bars and cleansers commonly used by patients with atopic and dry sensitive skin in SA. This was to provide clinicians and patients with the knowledge to make informed choices regarding their skin-care products.

Methods

Commercial soap bars (n=38) and cleansers (n=11) were randomly selected for pH analysis from the shelves of supermarkets, pharmacies, cosmetics shops, as well as those sold by hawkers on the streets of Durban, SA.

OCTOBER 2017 Vol. 11 No. 3

RESEARCH Selection criteria

Table 1. Soap brand names and pH values Brand name

pH value

1 Green bar soap 2 PnP green beauty soap 3 Breeze bar soap 4 Sunlight bar soap 5 Pond’s facial bar 6 Nivea cream soap 7 Gentle-magic skincare soap 8 Savlon hygiene soap 9 Bee Natural rich care family soap 10 Elizabeth Anne’s baby aq. cream bar 11 Beauty Magic facial soap bar 12 Vinolia luxury body soap 13 Lux beauty soap 14 Cuticura hygiene soap 15 Skin beauty soap 16 Dettol hygiene soap 17 Protex for Men AntiGerm soap 18 Clean & Clear facial cleansing bar soap 19 Germex hygiene soap 20 Mvelo magic treatment 21 Lemon-Lite complexion soap 22 Lifebuoy hygiene soap 23 Lifebuoy (clini-care) advanced hygiene soap 24 AquaBar 25 Pure soap 26 Palmolive naturals bar soap 27 Epiwash 28 Oh So Heavenly milk proteins moisturising glycerine bar 29 Hydra (aq. Cream & Glycerine) soap 30 Aqua-bar cleansing bar 31 Pears transparent soap 32 Protex AntiGerm soap 33 Tru Essentials bar soap 34 Clinique 35 Garnier smoothing facewash 36 Dove beauty cream bar soap 37 Cetaphil 38 Dove 4 Men body & face bar soap 39 Johnson's facial wash 40 Clean & Clear facial wash 41 Aqueous moisturising cream used as soap substitute 42 Cetaphil gentle cleanser 43 Pond's face wash 44 Cuticura face wash 45 Gill face wash 46 Garnier deep clean face wash 47 Himalaya Herbals gentle 48 Bioclear purifying face wash 49 Clearasil gel wash

10.75 10.66 10.36 10.34 10.30 10.27 10.26 10.25 10.22 10.19 10.18 10.16 10.10 10.10 10.08 10.05 10.04 9.99 9.99 9.98 9.93 9.93 9.89 9.80 9.71 9.71 9.70 9.65 9.57 9.42 9.42 9.41 9.38 9.36 8.41 7.21 7.26 7.11 6.89 6.86 6.79 6.44 6.39 5.92 5.85 5.84 5.28 4.68 3.74

aq. = aqueous.

147

SAJCH

• Word-of-mouth recommendations for soap and cleansing liquids for sensitive or atopic skin from pharmacies or beauty shops. • Soap bars or cleansers labelled as safe for dry or sensitive skin. • Prescribed products by healthcare professionals. • All atopic patients interviewed at a major referral skin centre at King Edward Hospital. In the interview, patients stated their personal brand-name soap or cleanser. The majority of interviewees (70%) used the green sunlight laundry soap. All of the other soaps and cleansers mentioned were also purchased for the analysis.

Soap emulsions

The samples were prepared as 8% emulsions in tap water.[23] Nine undiluted liquid facial cleansers were also included in the study and deionised water was used as a negative control.

Determination of pH

The pH of each emulsion or liquid cleanser was recorded in duplicate using a Metrohm pH meter model 827 (Herisau, Switzerland), according to manufacturer’s instructions.

Results

The pH of the 38 soap bars and 11 cleansers ranged from 7.11 - 10.75 and 3.74 - 6.89, respectively. Thirty-four soap bars had pH values ranging from 9.36 - 10.75. Two cleansers had pH of 5.84 and 5.83 (range between 5.4 - 5.9). Two cleansers had a pH below 5, with a total of 5 cleansers within the normal skin pH range. The results of the soap and cleanser analyses are presented in decreasing order of pH values (Table 1).

Discussion

The results of this study showed that only 5 of the samples had a pH in the 4 - 6 range, with only 2 samples below pH 5, and 2 between pH 5.4 and 5.9. The majority of the soaps had an alkaline pH. The soaps and shampoos commonly used by the studied population have a pH outside the range of normal skin pH.[15] The importance of soap pH and its potential to irritate the skin is an area that has been underscored in the medical fraternity. Furthermore, the lack of proper product labels makes it difficult to get this information from the packaging. It is essential that prior to recommending a soap to a patient, due consideration is given to the pH factor. Manufacturers must be encouraged to declare the pH of soaps and cleansers, thereby assisting consumers and healthcare workers to make informed choices. The green sunlight laundry soap was the most (70%) frequently used soap and it showed one of the highest pH values of 10.34. It should therefore be discouraged for use by atopic eczema patients and those with sensitive skin. It was interesting to note that some of the soaps with persuasive package labels like Oh So Heavenly milk proteins moisturising glycerine bar (9.6), Hydra (aq. cream & glycerine) soap (9.5) and Aqua-bar cleansing bar (9.4), which are targeted at atopic and sensitive skins, had high pH values. Aqueous cream which is used commonly by some as a moisturiser, and as a soap substitute by others, had a pH of 6.7. None of the soaps analysed in this study had the pH value displayed on the package insert. The most commonly used soaps were also found to be less expensive per gram. Affordability may explain their frequent use, as few patients were able to afford the more desirable yet more expensive soaps.

Limitations of the study

A number of soaps and cleansers are manufactured under different brand names. Hence we cannot conclude that the findings are representative of all of the soaps/cleansers on the market. The absence of the pH value on

OCTOBER 2017 Vol. 11 No. 3

RESEARCH the soap labels made comparison with our results difficult. It would have been useful to compare the findings from the study with the actual pH value on the labels of the soap/cleanser.

Conclusion

The majority of soaps and cleansers analysed in this study were alkaline. Only 2 were in the acceptable pH range of 5.4 - 5.9, and 5 within the pH range of 4 - 6, thus raising some concerns with regards to optimal management of atopic eczema patients. Patient education programmes and information on what soaps to avoid should be made easily available in order to enlighten the general population and clinicians. Better regulation of advertisement specifications, including the pH level and type of cleanser contained is necessary for the majority of soaps and cleansers.[16] We hope that the compiled list of analysed soaps will help as an easy desk reference and assist paediatricians to make better choices when recommending soaps and cleansers to patients with atopic eczema and to elderly patients (Table 1). 1. Cork MJ, Danby SG, Vasilopoulos Y, et al. Epidermal barrier dysfunction in atopic eczema. J Inves Dermatol 2009;129(8):1892-1908. https://doi. org/10.1038/jid.2009.133 2. Cork MJ, Robinson DA, Vasilopoulos Y, et al. New perspectives on epidermal barrier dysfunction in atopic eczema: Gene–environment interactions. J Allergy and Clin Immunol 2006;118(1):3-21. https://doi.org/10.1016/j. jaci.2006.04.042 3. Bloomfield SF, Stanwell-Smith R, Crevel RWR, Pickup J. Too clean or not too clean: The hygiene hypothesis and home hygiene. Clin Exp Allergy 2006; 36(4):402-425. https://doi.org/10.1111/j.1365-2222.2006.02463.x 4. Williams H, Robertson C, Stewart A, et al. Worldwide variations in the prevalence of symptoms of atopic eczema in the International Study of Asthma and Allergies in Childhood. J Allergy Clin Immunol 1999;103(1):125138. https://doi.org/10.1016/s0091-6749(99)70536-1 5. Akdis CA, Akdis M, Bieber T, et al. Diagnosis and treatment of atopic eczema in children and adults: European Academy of Allergology and Clinical Immunology/American Academy of Allergy, Asthma and Immunology/ PRACTALL Consensus Report. Allergy 2006;61(8):969-987. https://doi. org/10.1111/j.1398-9995.2006.01153.x 6. Ait‐Khaled N, Odhiambo J, Pearce N, et al. Prevalence of symptoms of asthma, rhinitis and eczema in 13‐to 14‐year‐old children in Africa: the International Study of Asthma and Allergies in Childhood Phase III. Allergy 2007;62(3):247-258. https://doi.org/10.1111/j.1398-9995.2007.01325.x 7. Dlova NC, Mankahla A, Madala N, Grobler A, Tsoka‐Gwegweni J, Hift RJ. The spectrum of skin diseases in a black population in Durban, KwaZulu‐Natal, South Africa. Int J Dermatol 2014;54(3):279-285. https://doi. org/10.1111/ijd.12589

148

SAJCH

8. Elias PM, Wood LC, Feingold KR. Epidermal pathogenesis of inflammatory dermatoses. Eczema 1999;10(3):119-126. https://doi.org/10.1097/01206501199909000-00001 9. Taïeb A. Hypothesis: From epidermal barrier dysfunction to atopic disorders. Contact Eczema 1999;41(4):177-180. https://doi. org/10.1111/j.1600-0536.1999.tb06125.x 10. Muto T, Hsieh S, Sakurai Y, et al. Prevalence of atopic eczema in Japanese adults. Br J Dermatol 2003;148(1):117-121. https://doi.org/10.1046/j.13652133.2003.05092.x 11. Palmer CN, Irvine AD, Terron-Kwiatkowski A, et al. Common loss-offunction variants of the epidermal barrier protein filaggrin are a major predisposing factor for atopic eczema. Nature Genet 2006;38(4):441-446. https://doi.org/10.1038/ng1767 12. Sandilands A, Sutherland C, Irvine AD, McLean WI. Filaggrin in the frontline: Role in skin barrier function and disease. J Cell Sci 2009;122(9):1285-1294. https://doi.org/10.1242/jcs.033969 13. Tupker R, Pinnagoda J, Coenraads P, Nater J. Susceptibility to irritants: Role of barrier function, skin dryness and history of atopic eczema. Br J Dermatol 1990;123(2):199-205. https://doi.org/10.1111/j.1365-2133.1990.tb01847.x 14. Panther DJ, Jacob SE. The importance of acidification in atopic eczema: An underexplored avenue for treatment. J Clinic Med 2015;5:970-978. https:// doi.org/10.3390/jcm4050970 15. Lambers HI, Piessens S, Bloem A, Pronk H, Finkel P. Natural skin surface pH is on average below 5, which is beneficial for its resident flora. Int J Cosmet Sci 2006;28(5):359-370. https://doi.org/10.1111/j.1467-2494.2006.00344.x 16. Baranda LI, González-Amaro R, Torres-Alvarez B, Alvarez C, Ramírez V. Correlation between pH and irritant effect of cleansers marketed for dry skin. Int J Dermatol 2002;41(8):494-499. https://doi.org/10.1046/j.13654362.2002.01555.x 17. Gfatter R, Hackl P, Braun F. Effects of soap and detergents on skin surface pH, stratum corneum hydration and fat content in infants. Dermatol 1997;195(3):258-262. https://doi.org/10.1159/000245955 18. Cowley N, Farr P. A dose-response study of irritant reactions to sodium lauryl sulphate in patients with seborrhoeic eczema and atopic eczema. Acta Derm Venereol 1992;72(6):432. http://dx.doi.org/10.1016/S0190-9622(96)90495-8 19. Resnick B. Dermatologic problems in the elderly. Lippincotts Prim Care Pract 1997;1(1):14-30. 20. Wortzman MS. Evaluation of mild skin cleansers. Dermatol Clin 1991;9(1):35-44. 21. Danby SGI, Al-Enezi T, Sultan A, Chittock J, Kennedy K, Cork MJ. The effect of aqueous cream BP on the skin barrier in volunteers with a previous history of atopic eczema. Br J Dermatol 2011;165(2):329-334. https://doi. org/10.1111/j.1365-2133.2011.10395.x 22. Tarun J, Susan J, Suria J, Susan VJ, Criton S. Evaluation of pH of bathing soaps and shampoos for skin and hair care. Ind J Dermatol 2014;59(5):442444. https://doi.org/10.4103/0019-5154.139861 23. Frosch PJ, Kligman AM. The soap chamber test: A new method for assessing the irritancy of soaps. JAAD 1979;1(1):35-41.

OCTOBER 2017 Vol. 11 No. 3

CASE REPORT

This open-access article is distributed under Creative Commons licence CC-BY-NC 4.0.

Acute poisoning in children from Jatropha curcas seeds M C Moshobane,1 MSc; C Wium,2 MSc; L V Mokgola,3 BSW Directorate of Biological Invasions, South African National Biodiversity Institute, Pretoria, South Africa Poison Information Centre, Division Clinical Pharmacology, Faculty of Medicine and Health Sciences, Stellenbosch University, Tygerberg, Cape Town, South Africa 3 Limpopo Department of Social Development, ga-Kgapane, South Africa 1

Corresponding author: M C Moshobane (moshobanemc@gmail.com) The semi-evergreen shrub, Jatropha curcas is native to Central and South America, but now occurs worldwide. Four children suffered severe symptoms of abdominal pain, nausea and vomiting after ingesting the seeds of J. curcas. These cases support the listing of J. curcas as a noxious weed. As a result of this, and a few other incidents, municipal authorities are urged to discourage the use of highly toxic plants such as J. curcas for hedges and garden plants, and to monitor the occurrence of such species. We present a case report about J. curcas poisoning. S Afr J Child Health 2017;11(3):149-150. DOI:10.7196/SAJCH.2017.v11i3.1401

Case report

Poisoning from plants is common in South Africa (SA).[1â&#x20AC;&#x201C;5] Children are regularly exposed to toxic plants and usually poisoned when plants are mistaken for common, edible plants e.g., Jatropha curcas is often mistaken for nuts.[3-7] Similar cases have been reported in Thailand, India and Israel.[8-10] All of these studies show that children are more susceptible to J. curcas poisoning because of their highly curious nature, which often leads to ingestion of the seeds. There are fewer datasets on paediatric poisoning from developing countries than from developed countries, likely due to underreporting of cases and lack of proper data management systems.[11] This report deals with four cases of acute J. curcas poisoning in children in Limpopo Province, SA. The immature fruit is small, capsule-like, round and green and becomes dark brown as it matures (Fig. 1). Each fruit contains three black seeds resembling nuts, e.g. peanuts/cashew nuts, that have a sweet taste.[12] Four children aged between 2 and 6 years were playing close to an abandoned yard where the plant grew. A fruit-bearing branch of J. curcas was hanging over the fence and the children collected the fruit and ate the seeds. Shortly after ingesting the fruit, they developed abdominal pain and were rushed to hospital. The lag time before the onset of gastrointestinal (GI) complications varied between 90 and 120 minutes. The GI symptoms that developed included abdominal pain, nausea, vomiting and diarrhoea, as well as a burning sensation in the throat. The patients were treated with intravenous hydration and anti emetic medication and the symptoms resolved within 12 - 14 hours. Three of the 4 children were observed for 24 hours and discharged the following day. The 2-year-old child was hospitalised for 7 days owing to prolonged diarrhoea. There is no specific antidote and the lethal dose in humans has not been established.[13]

Fig. 1. The fruit of the Jatropha curcas plant. (Photograph courtesy of Annah Ngobeni.)

149

SAJCH

Discussion

J. curcas L is a member of the family Euphorbiaceae and also known as a physic nut, purging nut, Barbados nut and purgeerboontjie. It is native to Central and South America, but is now widely distributed in most African countries, India and South East Asia.[14] In SA, it is found mainly in the northern and eastern parts of the country, especially in Limpopo, Mpumalanga and KwaZulu-Natal provinces.[15] It is a drought-resistant perennial plant, widely used as hedges and traditional medicine for ailments such as malaria, oedemas, etc.[6] Research has also recently investigated its potential use in biodiesel production.[16,17] Rural communities cultivate this plant as hedges around gardens and fields. Because of its unpalatability it is not grazed on by cattle.[3] J. curcas seeds contain a toxalbumin, curcin, which is 1 000 times less toxic than ricin found in Jatropha multifida seeds.[10] The GI effects (purgative activity) are caused by the diterpenoids and curcanoleic acid in the seed oil.[8] GI symptoms vary in severity and include abdominal pain, nausea, vomiting, diarrhoea and burning sensations in the throat.[8-10] Mastication of the seed may also play an important role in the extent of its toxicity.[18] The most effective treatment is symptomatic and supportive care. Most reported cases of J. curcas poisoning occurred in children, with toddlers being at higher risk of severe toxicity, although no fatalities have been reported in humans.[3,8-10] Despite the toxicity of the plant, it is cultivated for various uses, which includes its use as a biofuel. [16] The increased interest and cultivation of J. curcas will eventually increase the likelihood of accidental poisoning of children, particularly those in rural areas.[19-22]

Conclusion

This case highlights the need to consider primary public awareness of the toxicity of J. curcas. Furthermore, children in areas where the plants occur should be prevented from ingesting the seeds and educated about the dangers of ingesting anything unauthorised by their parents. This case report supports the previous evidence suggesting that J. curcas should be considered a noxious weed. We urge municipal agencies to discourage the use of highly toxic plants as hedges or in gardening. In cases where legislation controls the use of this plant, more emphasis should be placed on monitoring and educating people about the dangers of J. curcas.[23] Therefore, increased monitoring and surveillance of J. curcas populations is recommended.

OCTOBER 2017 Vol. 11 No. 3

CASE REPORT Acknowledgements. The South African National Department of Environment Affairs (DEA) through its funding for the South African National Biodiversity Institute (SANBIs)'s Directorate of Biological Invasions supported this work. We thank Mrs Annah Ngobeni for kindly providing logistical information and Dr Samuel Adu-Acheampong for proofreading and correcting this manuscript. Author contributions. Authors contributed equally to the manuscript. Funding. This study was funded by the SANBI. Conflicts of interest. None. 1. Balme K, Clare Roberts J, Glasstone M, Curling L, Mann MD. The changing trends of childhood poisoning at a tertiary children’s hospital in South Africa. S Afr Med J 2012;102(3):142-146. https://doi.org/10.7196/samj.5149 2. Du Plooy WJ, Jobson MR, Osuch E, Mathibe L, Tsipa P. Mortality from traditional-medicine poisoning: A new perspective from analysing admissions and deaths at Ga-Rankuwa Hospital. S Afr J Sci 2001;97(3-4):70. 3. Joubert PH, Brown JM, Hay IT, Sebata PD. Acute poisoning with Jatropha curcas (purging nut tree) in children. S Afr Med J 1984;65(18):729-730. 4. Marks CJ, van Hoving DJ. A 3-year survey of acute poisoning exposures in infants reported in telephone calls made to the Tygerberg Poison Information Centre, South Africa. S Afr J Child Health 2016;10(1):43-46. https://doi. org/10.7196/sajch.2016.v10i1.1045 5. Van Wyk B-E, Van Heerden FR, Van Oudtshoorn B. Poisonous Plants of South Africa. Pretoria: Briza Publications, 2002. 6. Mampane KJ, Joubert PH, Hay IT. Jatropha curcas: Use as a traditional Tswana medicine and its role as a cause of acute poisoning. Phyther Res 1987;1(1):5051. https://doi.org/10.1002/ptr.2650010112 7. Wanzala W, Wanjala CCW. Discovering poisonous plants by tasting: The case of children in Mumias Sub-County, Kenya. Arab J Med Aromat Plants 2016;2(2):99-110. 8. Chomchai C, Kriengsunthornkij W, Sirisamut T, Nimsomboon T, Rungrueng W, Silpasupagornwong U. Toxicity from ingestion of Jatropha curcas (‘saboo dum’) seeds in Thai children. Southeast Asian J Trop Med Public Health 2011;42(4):946-950. 9. Singh RK, Singh D, Mahendrakar AG. Jatropha poisoning in children. Med J Armed Forces India 2010;66(1):80-81. https://doi.org/10.1016/s03771237(10)80106-6

150

SAJCH

10. Levin Y, Sherer Y, Bibi H, Schlesinger M, Hay E. Rare Jatropha multifida intoxication in two children. J Emerg Med 2000;19(2):173-175. https://doi. org/10.1016/s0736-4679(00)00207-9 11. Pocock SJ, Collier TJ, Dandreo KJ, et al. Issues in the reporting of epidemiological studies: A survey of recent practice. BMJ 2004;329(7471):880-883. https://doi. org/10.1136/bmj.38250.571088.55 12. Begg J, Gaskin T. Jatropha curcas L. Poisonous Plants. 1994. http://www. inchem.org/documents/pims/plant/jcurc (accessed 23 November 2016). 13. Micromedex database. Jatropha curcas. Micromedex: Truven Health Analytics, 2017. https://www.micromedexsolutions.com/micromedex2/librarian/ssl/true (accessed 21 July 2017). 14. Fairless D. Biofuel: The little shrub that could – maybe. Nature 2007;449(7163):652-655. https://doi.org/10.1038/449652a 15. Henderson L. South Afr Plant Invaders Atlas (SAPIA). Appl Plant Sci 1998;12:31-32. 16. Pramanik K. Properties and use of Jatropha curcas oil and diesel fuel blends in compression ignition engine. Renew Energy 2003;28(2):239-248. https://doi. org/10.1016/s0960-1481(02)00027-7 17. Achten WMJ, Verchot L, Franken YJ, et al. Jatropha bio-diesel production and use. Biomass and Bioenergy 2008;32(12):1063-1084. https://doi.org/10.1016/j. biombioe.2008.03.003 18. Hirai T, Kang Y, Koshino H, et al. Occlusal-masticatory function and learning and memory: Immunohistochemical, biochemical, behavioral and electrophysiological studies in rats. Japan Dent Sci Rev 2010;46(2):143-149. https://doi.org/10.1016/j.jdsr.2009.12.002 19. Sarin R, Sharma M, Sinharay S, Malhotra RK. Jatropha-palm biodiesel blends: An optimum mix for Asia. Fuel 2007;86(10):1365-1371. https://doi. org/10.1016/j.fuel.2006.11.040 20. Van Eijck J, Romijn H, Smeets E, et al. Comparative analysis of key socioeconomic and environmental impacts of smallholder and plantation based Jatropha biofuel production systems in Tanzania. Biomass Bioenergy 2014;61:25-45. https://doi.org/10.1016/j.biombioe.2013.10.005 21. Shah S, Sharma A, Gupta MN. Extraction of oil from Jatropha curcas L. seed kernels by combination of ultrasonication and aqueous enzymatic oil extraction. Bioresour Technol 2005;96(1):121-123. https://doi.org/10.1016/ s0140-6701(05)82158-7 22. Yue GH, Sun F, Liu P. Status of molecular breeding for improving Jatropha curcas and biodiesel. Renew Sustain Energy Rev 2013;26:332-343. https://doi. org/10.1016/j.rser.2013.05.055 23. Moshobane MC. Inaugural meeting of the Alien Species Risk Analysis Review Panel. S Afr Netw Coast Ocean Res 2017;214:8.

OCTOBER 2017 Vol. 11 No. 3

CPD October 2017 The CPD programme for SAJCH is administered by Medical Practice Consulting. CPD questionnaires must be completed online at www.mpconsulting.co.za

True (T) or false (F): Regarding antibiotic prescription in children admitted to the emergency wards in Maputo 1. Malaria was the most common reason for antibiotic usage. 2. Ampicillin was the most commonly used antibiotic. Regarding motor impairment in children with cerebral palsy in Nigeria 3. A quarter of patients with spastic quadriplegia were ambulatory. 4. Spastic diplegia was more commonly associated with prematurity than spastic quadriplegia. Regarding autism spectrum disorder (ASD) in KwaZulu-Natal (KZN) 5. The majority of parents interviewed had a practical understanding of available treatments for ASD. 6. Two-thirds of families assisted the parents with the treatment of their children with ASD. Regarding toilet training and enuresis 7. Less than 15% of caregivers commenced toilet training of their infants before 12 months of age in Nigeria. 8. By 30 months of age, >85% of infants had attained day- and nighttime continence. Regarding barriers to heathy food choices among children in the Western Cape 9. In the lower socioeconomic groups, mothers thought that the gas in carbonated drinks was unhealthy. 10. The nutritional value of a food was considered to be a factor influencing food choice by the majority of mothers interviewed. 11. The vast majority of mothers believed vegetables and fruits are important in a childâ&#x20AC;&#x2122;s diet.

Regarding West syndrome in children in KZN 12. The features of West syndrome include epileptic spasms, hypsarrhythmia on electroencephalogram and developmental delay or regression. 13. A number of studies have shown a female predominance. Regarding Fanconi anaemia in South Africa (SA) 14. In SA, the prevalence of Fanconi anaemia is more common in children of Jewish ancestry than other ethnic groups. 15. The genetic abnormality causes impaired DNA repair and chromosome instability. 16. In the Afrikaner community, the prevalence of Fanconi anaemia is estimated to be approximately 1/22 000. Regarding soaps used for atopic eczema 17. Soaps with an alkaline pH are beneficial in the treatment of atopic eczema. 18. The vast majority of soaps available in SA have a pH that is detrimental for the management of atopic eczema. Regarding poisoning by Jatropha curcas seeds 19. The plant Jatropha curcas, although not indigenous, is found in the northern and eastern parts of SA. 20. Abdominal pain, diarrhoea and vomiting are the most common symptoms when the fruit is ingested.

A maximum of 3 CEUs will be awarded per correctly completed test. CPD questionnaires must be completed online via www.mpconsulting.co.za. After submission you can check the answers and print your certificate. Accreditation number: MDB015/172/02/2017 (Clinical)

151

SAJCH

OCTOBER 2017 Vol. 11 No. 3