Childhood Poisoning at a Tertiary Hospital in South West Nigeria

Address for correspondence: Dr. OLATUNYA Oladele Simeon Department of Paediatrics Ekiti State University/Teaching Hospital P.O. BOX, 2267, Ado Ekiti, Ekiti State, Southwest Nigeria Tel +2348038617705 E-mail: ladeletunya@yahoo.com 1Oladele Simeon OLATUNYA, Lecturer, Department of Paediatrics, Ekiti State University/ Teaching Hospital, Ado Ekiti, Ekiti State, Nigeria, 2Ayodeji Olusola Isinkaye, Federal Medical Centre Ido Ekiti, Department of Community Medicine, Nigeria, 3Ezra Olatunde Ogundare, Dept of Paediatrics, Ekiti State University/Teaching Hospital, Ado Ekiti, Nigeria, 4Isaac Oludare Oluwayemi, Department of Paediatrics, Ekiti State University Teaching Hospital, Nigeria, 5Foluso John Akinola, Department of Paediatriics, Ekiti State University Teaching Hospital, Nigeria. Abstract


Introduction
P oisoning in children is a preventable cause of morbidity and mortality that is s ll very common worldwide 1 .The developed countries of the world are containing it through adequate data gathering and other well entrenched ins tu onalised policies 2,3,4,5 .In contrast, poisoning profi les in developing countries are poorly documented 6,7 .The high contribu ons of malnutri on and infec ous diseases to the overbearing burden of childhood morbidity and mortality in these countries have the tendency to divert a en on from other preventable causes of child deaths 8 .There are few studies from Nigeria on childhood poisoning, 9,10,11,12,13,14,15 some of them were restricted to a few targeted poisoning agents 9,10 while some are quite dated 11,12 and none was from the current study area.This paper examines childhood poisoning at a ter ary hospital in south west Nigeria.The study was conducted with the aims of iden fying the demographic features of children involved with poisoning, their clinical features, types of poisoning agents, home interven ons and the outcome of hospital treatment.Informa on from the current study will update data on childhood poisoning in Nigeria and help stakeholders in planning appropriate preven ve and management strategies.

Material and Methods
This study was conducted at The Eki State University Teaching Hospital (EKSUTH), Ado Eki , the capital of Eki State in South west Nigeria.The hospital receives referrals from public and private hospitals within the state and from neighbouring towns in other adjoining states like Ondo, Osun, Kogi and Kwara.New paediatric pa ents are usually fi rst seen at the paediatric out-pa ent department (POPD) before being triaged.The POPD and the children emergency ward (CEW) are both housed in the same building.They are co-managed as a joint en ty by six doctors and fi een nurses.These are made up of fi ve paediatric registrars, three paediatric nurses, twelve nurse-midwives and a supervising consultant paediatrician.Records of all children admi ed into the CEW of EKSUTH for poisoning between January 1, 2011 and December 31, 2014 were reviewed.A pre-tested review chart form was used to extract informa on on pa ents` socio-demographic characteris cs, admission date, me poisoning occurred, poison agent, poisoning circumstances, dura on of poisoning before presenta on, home interven on(s), symptoms, hospital interven ons, admission dura on, outcome and number of follow up visit.Cases of poisoning from animal bites and food contamina on were not included.Parental socio-economic status was determined using their occupa on and levels of educa on 16 .
Approval for the study was obtained from the Ethics Commi ee of Eki State University Teaching Hospital.Sta s cal analysis was conducted with IBM SPSS so ware version 20.Descrip ve sta s cs were performed using simple propor ons while Fisher's exact test was carried out to detect rela onship between variables.

Results
A total of 5,256 children were admi ed into the Children Emergency ward (CEW) during the study period.Eighty-one (81) of them, represen ng 1.54% were cases of child poisoning.They were made up of 47 (58.0%) boys and 34 (42.0%)girls.Their median age was 36 months with a range from six weeks to 15 years.Most of the children, 53 (65.4 %) were between one and fi ve years of age, followed by those aged above fi ve years, 20 (24.7 %), while the remaining 8 (9.9 %) were aged less than one year.The children were more or less evenly distributed across the three family social classes with 25 (30.9%),27 (33.3%)and 21 (25.9%)belonging to the lower, middle and upper classes respec vely.Eight (9.9%) of the children had no documented informa on necessary for compu ng their family social classes.
Regarding the circumstances surrounding poisoning incidence and home interven ons given before presenta on at the hospital, 69 (85.2%) of the cases of poisoning in the pa ents were accidental and the ages of pa ents with accidental poisoning ranged between six months and 84months with a mean of 36 months.This is followed by twelve (14.8%) inten onal poisoning that included three suicidal a empt cases.The pa ents with inten onal poisoning were aged between 10 and 15 years with a mean age of 12.8 years.None was homicidal.Most of the poisoning cases 76 (93.8%), occurred during the day and majority 72 (88.9%), took place at home.The largest propor on of the children (25.9%), ingested the poisoning agents kept in used water bo les.Most of the poisoning agents were kept at home in the living room (50.6%) and the kitchen (18.5%).
Poisoning occurred as a results of self-inges on by 65 (80.2%) of the children, three of which were cases of a empted suicide by adolescent girls aged 13, 14 and 15 years who had ingested sodium hypochlorite solu on, a roden cide (pes cide) and a corrosive agent respec vely following disagreements with their parents.Two of them had psychiatric evalua on done by the mental health physicians.Pa ent's mother administered the poisoning agents in six (7.4%) cases, grandmother and other close rela ves in four (4.9%) cases, and a crèche's member of staff in one (1.2%)case.The remaining fi ve (6.2%) children were cases of inhala on of emissions from electricity genera ng sets while sleeping.There were no cases of inges on of mul ple agents.Palm oil (oil from Elais guineensis) was the commonest home remedy administered in 56 (69.1%) cases, followed by induced vomi ng (38.3%), herbal concoc ons (23.5%), milk (4.9%) and enema (3.7%) Table1.The mean dura on before presen ng at the hospital a er poisoning, was 7.9 ±8.3hours (range 30 minutes to 3 days) with 20 (24.7%), 55 (67.9%), and 6 (7.4%) pa ents presen ng within 6 hours, between 6 to 12 hours, and at greater than 12 hours respec vely.Respiratory system related symptoms of dyspnoea (61.7%) and cough (43.2%) cons tuted the commonest presen ng complaints.Haemograms and serum electrolytes were the most common inves ga ons requested in the cases.None of the case had toxic levels of agents assayed as there were no facili es to do so.Intravenous fl uids (67.9%), an bio cs (66.7%), and intranasal oxygen (51.9%) were the common hospital interven on administered.Most pa ents, 40 (49.4%) were on hospital admission for 24 -36 hours and only 15 (18.5%) returned for at least one follow-up visit (Table 2).
Most of the cases, 69 (85.2%) recovered fully, fi ve (6.2%) discharged against medical advice (DAMA) and seven children died.This is equivalent to a case fatality rate of 8.6%.The results of bivariate analysis for rela onship between poisoning agent and age and pa ent outcome are shown in Table 3. Kerosene was the commonest poisoning agent (37.0%), followed by drugs (22.2%), alcohol based concoc ons (19.8%), pes cides / herbicides (8.6%), caus c soda (6.2%) and carbon monoxide (6.2%).Infants were more likely to have poisoning from alcohol based concoc ons (Fisher's exact p < 0.001).There was no sta s cally signifi cant associa on between pa ent outcomes and poisoning agents (Fisher's exact p > 0.05).
Of the seven children that died, one (a ten month old boy) was forcefully administered with overdose of unspecifi ed an -malarial syrup by a caregiver at the crèche.He was admi ed with dyspnoea no ced immediately a er the drug administra on, fever, unconsciousness, coarse crepita ons, severe anaemia (haematocrit was 14%) and malaria parasitaemia but died about one and half hours later while undergoing blood transfusion.A second 10 month old boy who was administered with herbal concoc on by his grandmother presented three days a er with seizures, unconsciousness and dyspnoea.He had received palm oil, enema and more herbal concoc ons as home remedies but died within 48 hours of admission.
The third and fourth mortali es were cases of self-inges on of kerosene by two boys aged 15 and 22 months.They presented with unconsciousness, focal seizures, dyspnoea, hypoglycaemia and malaria parasitaemia.Each of them died within 20 hours of admission.The fi h case involved a 3 year old boy who accidentally self-ingested an organophosphate pes cide (Gamalin-20, the trade name of a popular organophosphate based pes cide used by farmers in Nigeria) on the farm.He presented 18 hours later at the hospital with dyspnoea, fever, seizures and unconsciousness a er receiving some herbal concoc on, palm oil and induced vomi ng at home.He also died within 20 hours of admission.The sixth case involved a four year old girl (as well as her mother) who had carbon monoxide poisoning from exhaust fumes from an electricity genera ng set placed overnight near their window.They were brought to the hospital by neighbours more than ten hours a er exposure.She presented unconscious and died within six hours of admission.
The last case was a twelve year old boy who ingested a locally available alcohol-based aphrodisiac herbal mixture, Osomo® kept at home by his stepfather.He presented at the hospital a er 16 hours with dyspnoea, unconsciousness and seizures.He was discharged against medical advice by his caregivers a er three days of admission and got re-admi ed two days later with features of acute renal failure alongside his earlier signs and symptoms.He died within twentyfour hours of his re-hospitaliza on at the intensive care unit while being worked up for haemodialysis.

Discussion
The determinants of poisoning outcome in children includes: the physiology of the child, child`s health status, type of agents involved, dura on of symptoms before presenta on and the quality of care received 6,7,17 .In this study, 1.54% of all admissions at the CEW were due to poisoning.This propor on is higher than 0.2% to 0.94% previously reported from hospital based studies in Nigeria 11,12,13,15 .It is an indica on that the burden of childhood poisoning is high in the study area.
Except for the even distribu on of the pa ents across the social divides in the current study, other demographic a ributes of the pa ents as found, had been described locally 9,10,11,12,13,14 .and interna onally 2,3,6,7,18,19,20 .This a ests to the global similari es in the demographic picture of children with poisoning.The referral status of the current study centre, coupled with the metropolitan nature of the study centre loca on could have been responsible for the observed social class admixture as previous Nigerian studies indicated a higher preponderance of childhood poisoning in the lower class 9,10,11,12,13,14,15 .Our fi ndings could also be a hint at possible changes in the epidemiology of the condi on in the country.Although one study from Nigeria 15 found a female preponderance, other Nigerian and interna onal studies are in agreement with the male preponderance observed in this study 2,6,7,11,12,13,19,20 .Also, the observa on that most cases were uninten onal, especially among the under-fi ves, is not restricted to this study 2,3,4,5,6,7,9.10,11,12, 13,14,15 .Children in this age group are in their forma ve years when they are usually more adventurous 15,17,21 and so, may inadvertently ingest the agents.The bimodal nature of the cases in this study, with noninten onal poisoning involving young children and inten onal cases involving adolescents, conforms with previous reports 2,4,7,11,15 .Inten onal poisoning involving inges on of poisonous substances with suicidal intent is especially noted with adolescents 17,22,23 .This tendency was thought to be peculiar to the developed world, but it is now being increasingly recognised in the third world 11,7,15,19 .These behaviours among adolescents could refl ect their maladjustments as they transit to adulthood 23.24,25,26 .Two of the three adolescents girls with inten onal poisoning for suicidal purposes in this review, had psychiatric evalua on conducted on them by mental health physicians.Kerosene (paraffi n), a dangerous hydrocarbon derived from petroleum dis llates is the leading poisoning agent in this review and this conforms with some earlier Nigerian reports 11,12 .Similar trends have also been observed in neighbouring African countries 6,27 .In Nigeria, kerosene is a major source of household fuel used for cooking and ligh ng 9,10,11,12,13,14,21 .It is also stocked for business purposes by many traders and households in unmarked familiar containers that are usually very a rac ve to children.The high vola lity of its vapour makes it diff use readily to children's airway upon inges on, thereby causing irrita on of the airways 17,21 .This perhaps, explains the high propor on of pa ents with respiratory symptoms in this review.There were higher preponderances of drugs and alcohol poisoning in this study compared to most Nigerian studies 11,13,15 .This may connote a gradual shi towards poisoning pa erns in other parts of the world 6,[20][21] and may be a refl ec on of the double burden of childhood poisoning in the study area as epidemiology of poisoning varies with me and place 1,6,7,17,21 .The incessant use of alcohol under the guise of herbal remedies among adult Nigerians 28 might have contributed to this pa ern as children like to copy adults 17,21 .
Like a previous Nigerian report where mortality was recorded 15 , one of the fi ve cases of carbon monoxide (CO) poisoning in this review died.Un l recently 15 , CO poisoning was rare in Nigeria 11,12,13 .The epilep c power supply bedevilling the country has led many households to depend on electricity generators for their power supply 29 and this may fuel an upsurge in the number of new cases of CO poisoning if not checked.Most of these household generators are usually poorly maintained leading to their emission of CO from incomplete combus ons.CO becomes hazardous when excessive accumula on occurs and its toxic threshold is reached or exceeded thereby causing health hazards.It displaces oxygen from its binding sites in haemoglobin to form Carboxy-haemoglobin (COHb) which impairs oxygen delivery to body ssues 17,30 .Expectedly, vic ms suff er ssue hypoxia and they o en present with a barrage of symptoms ranging from headache, coma, convulsions and irregular breathing pa erns 15,17,[30][31] .Some of the cases of CO poisoning involved in this study were admi ed with convulsions, coma and irregular breathings.The four surviving cases are being followed up for the manifesta ons of some recognised sequelae of CO poisoning which includes impaired cogni on, progressive brain damage and chronic lung disease 15,17,30,31 .
The high rates of induc on of vomi ng and administra on of herbal concoc ons as home interven ons refl ects the poor knowledge of parents in the study area on the health dangers inherent in such prac ces 6,11,17,21 .Africans in rural se ngs s ll have strong faiths in the use of tradi onal herbal agents as home remedies in poisoning cases 6,11 These agents have been found to contribute signifi cantly to mortality in childhood poisoning 6,11 The need for more educa on of parents on the use of only safe home interven ons for children with poisoning cannot be overemphasised.One way of doing this, is through the establishment of poison informa on and control centres in the country where parents or caregivers can readily call to obtain informa on and seek advice on the poisoning situa ons involving their wards as obtainable in other countries 3,7,18,20 .The rela vely high ownership rate of mobile telephones lines in Nigeria will readily support this services if adopted 32 .The use of jingles in mass media and regular counselling of caregivers on poisoning preven on during clinic visits are also recommended.
A cri cal look at the interven on in the hospital revealed that they consisted mostly of suppor ve care and ordering of rou ne inves ga ons as none of the pa ents had assays of the toxic levels of the poisoning agents determined or their arterial blood gases measured where applicable.Also, an dotes were not given for some specifi c cases except for the fi ve cases of CO poisoning.This could be a ributable to the level of severity of the cases or non-availability of more advance tools for the care of poisoned children in the study area as stated.Inadequate knowledge on the part of the health professionals a ending to the pa ents could also be a factor.
An appraisal of the other remaining deaths in this review, depict very interes ng scenarios refl ec ng the turbulent milieu within which children in the developing world operate.For the ten month old child poisoned at the crèche, the constella on of the symptoms and signs suggests a child who had been unwell for quite some me but being forced to be cared for by an unskilled crèche member staff .She masqueraded as an health worker but could not iden fy the danger signs in the child only to fi nally pull the death trigger by giving overdose of drugs albeit forcefully.This case also highlights the impact of co-morbid condi ons on the outcome of childhood poisoning 6,17 .In the cases of kerosene and direct herbal concoc on poisonings, the deaths might be a ributable to the high hazard factor of the agents they ingested 6 .The twelve year old boy might have been cajoled by the alleged mys cal powers of the ingested herbal concoc on as being touted in its many widely publicised adverts across the country as an herbal medicine and food supplement.In Nigeria, herbal remedies are registered by The Na onal Agency for Food, Drug Administra on and Control (NAFDAC) for public consump on 33 .Although the ingested herbal remedy was not available for analysis in this review, nevertheless, studies have shown that some herbal remedies being marketed in Nigeria lack standardised methods of prepara on as s pulated by standard interna onal best prac ces 28,34,35 .As a result, they have wide varia ons in the contents of their ac ve ingredients even within cohorts of the same batch and are laden with hazardous materials that are injurious to health 28,34 .These observa ons may not be totally divulged from the spectral of the clinical manifesta ons in the child before his demise.The pa ent with organophosphate inges on and poisoning on the farm also reinforces the need for more educa on and enlightenment programmes for farmers on safe storage, usage and handling of herbicides and pes cides.This will prevent children from having easy access to chemicals used for agricultural purposes.
The 8.6% mortality rate in this study is higher than the 0.0% and 3.8% reported from Maiduguri 13 and Jos 15 in northern Nigeria, but lesser than 11.9% and 20.0% earlier reported in Ile-Ife 11 and Calabar 12 respec vely both from the southern parts of the country where the current study was conducted.This may suggest a possible regional dichotomy and or, a decrease in the burden of childhood poisonings in the southern part of the country.However, these fi gures need to be interpreted with cau on as the previous studies from southern Nigeria are quite older.Nevertheless, the mortality rate in the current study is far above what is being reported from other parts of the world 3,7,18,19,20 .This indicates the need for improved care and outcome for children with poisoning in the study area and Nigeria by extension.Upgrading implements for the care of children with poisoning and regular in-service trainings or refresher courses for health workers in the country will update them on current trends in poisoning management such as the use of Poison Scores to triage, classify and plan management of poisoning pa ents 6,36 .

Conclusion
This study has highlighted the burden of childhood poisoning in the study area and iden fi ed challenges facing its management.It also raises ques ons about factors that sustain the current status.Establishment of poison informa on and control centres, focused health educa on on ra onal use of drugs, herbal mixtures, and regula on of their sales with appropriate legisla on will help combat the burden of childhood poisoning in developing countries.

Table 2 :
Symptoms and Hospital interven ons