|Year : 2021 | Volume
| Issue : 2 | Page : 128-134
Childhood brain tumors in Southern Nigeria: A multicenter experience
Ndubuisi Anyele Uwaezuoke1, Enoch Ogbonna Uche2, Christian B Eke1, Ann Ebele Aronu1, Francis I Ukekwe3, Joy N Eze1, Ngozi Ojinnaka1
1 Department of Paediatrics, College of Medicine, University of Nigeria/Teaching Hospital, Ituku/Ozalla, Enugu State, Nigeria
2 Department of Neurosurgery, College of Medicine, University of Nigeria/Teaching Hospital, Ituku/Ozalla, Enugu State, Nigeria
3 Department of Histopathology, College of Medicine, University of Nigeria/Teaching Hospital, Ituku/Ozalla, Enugu State, Nigeria
|Date of Submission||03-May-2020|
|Date of Decision||29-Aug-2020|
|Date of Acceptance||15-Sep-2020|
|Date of Web Publication||29-Jan-2021|
Ndubuisi Anyele Uwaezuoke
Department of Paediatrics, College of Medicine, University of Nigeria, Ituku-Ozalla Campus, Enugu.
Source of Support: None, Conflict of Interest: None
Background: Childhood brain tumors (CBTs) constitute 5% of all childhood tumors and can involve any part of the brain disrupting brain structure, form, and function with associated high morbidity and mortality. The objectives of the study were to determine the clinicopathologic pattern and treatment outcome of brain tumor in children in our setting. Materials and Methods: A 3-year prospective multicenter observational study of children with brain tumors was undertaken. Relevant patients’ clinical data sought, included histological diagnosis, World Health Organization grade and survival pattern. Data were analyzed using SPSS version 21.0, whereas the level of statistical significance was set at P < 0.05. Results: Study participants were aged 2–16 (mean 7.93 ± 4.21) years. Fifteen were female (51.7%). The duration between; illness to presentation, presentation to confirmation of diagnosis, and from presentation to specific treatment were 4.40 ± 4.40, 5.0 ± 4.0, and 6.3 ± 6.0 months, respectively.Cerebellar tumors occurred most commonly 15 (51.7%). Majority of the cases, 21 (72.4%) were low grade. Central nervous tumor was first suspected in eight cases (27.6%) at the referring center. Eleven (37.9%) and ten (34.5%) patients, respectively, survived at 1 year and 3 years of follow-up post-diagnosis. The most common complication was hydrocephalus 12(41.4%). Mortality occurred in seven (21.4%) patients arising from their brain tumors. Conclusion: Poor knowledge of health care providers in referring centers, along with delay in initiating treatment contributed to mortality and morbidity.
Keywords: Childhood brain tumor, outcomes, presentation
|How to cite this article:|
Uwaezuoke NA, Uche EO, Eke CB, Aronu AE, Ukekwe FI, Eze JN, Ojinnaka N. Childhood brain tumors in Southern Nigeria: A multicenter experience. Int J Med Health Dev 2021;26:128-34
|How to cite this URL:|
Uwaezuoke NA, Uche EO, Eke CB, Aronu AE, Ukekwe FI, Eze JN, Ojinnaka N. Childhood brain tumors in Southern Nigeria: A multicenter experience. Int J Med Health Dev [serial online] 2021 [cited 2022 Aug 12];26:128-34. Available from: https://www.ijmhdev.com/text.asp?2021/26/2/128/308249
| Introduction|| |
Primary central nervous system tumors account for approximately 5% of all childhood tumors with a high morbidity and mortality.
There appears to be an increase in incidence of childhood brain tumor (CBT) worldwide with about 2000 deaths occurring in Nigeria in 2016. The prevalence and outcome of CBT vary between different regions of the world with high-income regions having better survival and Disability Adjusted Life Years (DALY).
A hospital-based survey performed in two centers from different regions of Nigeria showed varied patterns of CBT occurrence suggesting a higher incidence of other pediatric tumors in the southwest, but a significantly increased relative incidence of CBT in the northeast region.
The difference in presentation as reported in the comparative study above may suggest interplay of genetic and environmental factors in the aetiopathogenesis of pediatric tumors.
Generally, the main brain tumor phenotype in children include astrocytomas (38%– 50%), ependymomas (8%–14%), primitive neuroectodermal tumors (PNETs), which include medulloblastoma (16%–25%) amongst other tumors, and non-PNETs such as gliomas (4%–16%).
A previous report categorizing childhood brain tumors by anatomic location among male and female children showed that; 24% occur in the cerebellum, 22% in the cerebrum, 20% in the brain stem, and 16% in both the infra and supratentorial locations (transtentorial). Sixteen percent occurred in not otherwise specified areas. This may also suggest that infratentorial tumors occur more commonly in children than supratentorial tumors.
In the pediatric age group, the characterization of brain tumors into benign and malignant neoplasms is a thin line as all CNS neoplasms are capable of causing both fatal raised intracranial pressure as well as severe neurologic sequelae by impairment of eloquent brain function without prejudice to their degree of histological differentiation and biologic behavior.
The exact burden of CBT in sub-Saharan African countries including Nigeria is not known owing to lack of national cancer registries, poor diagnostic facilities, and poor reporting of cases.
The objectives of this study were to describe the pattern of brain tumor in children in our setting as well as evaluate the outcome of treatment.
| Patient and Methods|| |
This was a multicenter study conducted in four tertiary health facilities in southeast Nigeria; University of Nigeria Teaching Hospital, Ituku-Ozalla, Enugu State, Federal Medical Centre, Umuahia, Abia State, Enugu State University of Science and Technology Teaching Hospital, Park Lane, Enugu and Federal Teaching Hospital, Abakaliki, Ebonyi State.
It was a prospective observational study conducted on children and adolescents with confirmed brain tumor in the four study locations between January 1, 2016 to December 31, 2018.
Ethical approval for the study was obtained from the Health Research Ethics Committee of University of Nigeria Teaching Hospital, Ituku- Ozalla, Enugu State, whereas further permissions to embark on the study were obtained from the other health facilities.
All cases of suspected CBT presenting to the Department of Paediatrics and/or Sub-division of Neurosurgery of the four study tertiary health facilities were selected consecutively during the study period (2016–2018). Follow-up was between 1 and 3 years.
Relevant clinical details studied included- sociodemographic characteristics (age, sex, place of domicile, socioeconomic status of parents, whether residing in rural, semi-urban or urban areas).
Duration of the illness prior to presentation in months, interval between presentation, and confirmatory diagnosis (in months), family history of previous CNS tumor.
Common presenting symptoms including altered consciousness, vomiting, headache, limb weakness, deterioration school/academic performance, gait disturbances, irritability, unsteadiness, poor growth, progressive head enlargement, visual hallucination, behavioral problems, among others.
Signs recorded included features of raised intracranial pressure such as pupillary dilatation and papilledema. Other signs were hypertonia, hyperreflexia, long tract signs, visual defects, cranial nerve deficits, cerebellar signs, macrocephaly, and endocrinopathy.
CBT’s were classified using the 2007 World Health Organization (WHO) classification of tumors of the CNS following histological diagnosis.
Available diagnostic neuroimaging modalities were MRI, computerized tomography (CT) scan with contrast, combined skull X-ray plus CT scan, transfontanel ultrasound plus CT scan, CT scan with MRI brain, skull X-ray plus CT brain combined with MRI.
For our index study, only subjects who satisfied all CBT diagnostic triad below were enrolled.
i. Clinical features suggestive of intracranial mass lesion based on neurological evaluation.
ii. Neuroimaging (brain CT scan or MRI scan or a combination) evaluation showing an intracranial mass lesion.
iii. Confirmatory histopathological diagnosis of tumor biopsies.
We adopted the WHO 2007 classification of CNS tumors because of the unavailability of facilities for genetic and molecular evaluation of brain tumor biopsies in our study locations.
Cases of suspected brain tumors that did not satisfy all diagnostic criteria above were excluded.
Treatment of CBT
The mainstay of treatment of CBT in our study centers is microsurgical resection. Other treatment modalities include radiotherapy alone or in combination with surgery and adjunctive treatment/palliative treatment. Chemotherapy as a modality was not given to our patients because the protocols for CBT are still being developed for the study locations.
Outcome parameters: treatment outcomes were evaluated at 1 year, 3–5 years follow-up. Other features studied include; interval between initial presentation and treatment, interval between presentation and confirmatory diagnosis, survival outcomes and complications such as hydrocephalus, cerebral edema, brain herniation, midline brain shift, obstructive hydrocephalus, and mortality rate.
Determination of social class
Social class determination of the study subjects was obtained using the method proposed by Oyedeji in Ilesha among hospitalized Nigerian children. Here, the social class of each child was based on the highest educational attainment and occupational status of both parents. The average of the sum for the father and mother rounded off to the nearest whole number was assumed as the social class of each child selected for the study.
The highest score is 1, whereas the least is 5. Social classes 1 and 2, 3, 4, and 5 were, respectively, classified as high, middle, and low socioeconomic classes. In situation whereby one of the parents is late, the social class of the child was determined with the highest educational attainment and current occupation of the surviving parent.
Data were analyzed using the IBM SPSS for windows version 21.0 (IBM, Armonk, New York).
Categorical variables were presented as frequencies and percentages, whereas statistical significance of association between variables was determined using Chi-square test. The level of statistical significance was set at P < 0.05.
| Results|| |
A total of 29 (100%) children and adolescents aged between 2 and 16 years (mean: 7.93 ± 4.21); median age: 10 years who presented and were managed during the period were studied.
Majority were females, 15(51.7%) and residing in urban setting, 23(79.3%). In addition, majority of the cases 15 (51.7%) were of low social class [Table 1].
The mean time of presentation to any of the four specialist facilities used from the onset of the illness was 4.40 ± 4.40 months. The interval between patients’ presentation and arriving at the different confirmatory diagnosis was 5.0 ± 4.0 months, whereas the duration between presentation and commencement of specific treatment was 6.26 ± 5.96 months.
Three (10.3%) of the cases reported positive family history of CNS tumor (two cases (66.7%) in one of the parents and one case (33.3%) in a sibling) [Table 2].
The predominant presenting symptoms in the study patients were as follows: altered consciousness 26 (89.7%) vomiting 22(75.9%), headache 22 (75.9%), limb weakness 22(75.9%), deteriorating school performance 17(58.6%), and seizures 15(51.7%) among others.
Similarly, the commonest signs were papilledema 25(86.2%), hyperreflexia 25(86.2%), hypertonia 24(82.8%), focal motor deficits 23(79.3%), and raised intracranial pressure 21(72.4%), unconsciousness 18(62.1%), among others [see [Table 2]].
CT scan combined with magnetic resonance imaging were applied for diagnosis in sixteen (55.2%) of the cases, whereas CT scan (CT brain) with contrast was used in making the diagnosis in 7 (24.1%) children, making CT scan the single most used diagnostic modality among study subjects. Most of the subjects studied had tumors with WHO grade I, 14(48.3%) had grade II, 7(24.1%) grade III and grade IV, 6(20.7%) [Table 3].
|Table 3: Diagnostic tools and World Health Organization staging of the CNS tumor (n = 29)|
Click here to view
Following initial clinical evaluation on presentation, 8(27.6%) was suspected clinically to harbor CNS tumor prior to the definitive investigation. In terms of the final/confirmatory diagnosis of the studied cases, cerebellar tumors other than medulloblastomas 15(51.7%) followed by medulloblastomas 4(13.8%), craniopharyngioma 3(10.3%), optic nerve glioma 1 (3.4%), among other cases were the common tumor phenotypes seen among study subjects.
Observed complications were hydrocephalus 12(41.4%), brain herniation 2(6.9%), cerebral edema 1(3.4%), midline brain shift 1(3.4%), among others. A total of eleven patients (34.5%) were alive at 1-year follow-up, whereas 10(34.5%) survived to 3 years, 8(27.6%) sought discharge against medical advice and the details of their outcomes missed. Only two subjects (6.9%) survived without complications, 20(69%) with complications, 7(24.1%) died and 8(27.6%) were discharged against medical advice. Mortality was reported in seven patients. All CBT-related deaths occurred within 12 months of presentation and majority, 6(86.7%) occurred within 6 months. Most of the patients that survived with complications and those without complications with regards to interval between presentation and confirmatory diagnosis (P = 0.06). Similarly, most of the patients that survived with complications had their treatment outcome made within 1–4 months of presentation, 10(62.5%). There was also no significant difference between subjects that survived with and those without complications with regards to interval between presentation and determination of treatment outcome (P = 0.840).
| Discussion|| |
In our index study, we have prospectively evaluated 29 patients with CBT-treated across four tertiary hospitals in southern Nigeria.
The mean age was 7.93 ± 4.2 years, ranging from 2 to 16 years, with a slight female preponderance. Primary CNS tumor was previously reported to be more common in males due to preponderance of medulloblastoma and ependymomas among teenage boys. The slight female preponderance in this study may reflect a change in the attitude of parents to the girl child with brain tumor leading to reduced gender bias or discrimination and ultimately improved presentation for treatment. It may also partly be due to the low incidence of ependymoma (n = 1). Majority of the twenty-nine children lived in urban areas. Based on the classification of socioeconomic status of parents as described by Oyedeji, those from low and middle social economic class recorded the highest incidence of CBT. Infratentorial tumors were more frequent with cerebellar tumors being most frequent and constituting 65.5% of all tumors followed up over the study period [Table 4]. The childhood brain tumor consortium reported slight infratentorial predominance for central nervous tumors; 43.2% of which were infratentorial in location, whereas 40.9% were located in the supratentorial region, 4.9% in the spinal cord and 11% in multiple sites. Our finding may appear skewed considering the large number of cerebellar tumors. Cerebellar tumors including medulloblastomas and cystic cerebellar astrocytomas are among the most common childhood brain tumors reported in literature. Our findings are in agreement with these reports.,,
As expected the most commonly used radiological tool was a combination of MRI and CT scan and this was used in more than 50% of the patients. The least used single radiological investigation were skull X-ray as well as MRI each having a frequency of 3.4%. We believe many referring physicians are beginning to understand that skull x ray alone is not very helpful in CBT diagnosis and are now requesting for advanced imaging techniques. Also neuroimaging modalities such as MRI scan which were poorly accessible previously are now more readily available. Currently most patients present to the neurosurgeon with an already reported CT scan and sometimes also with a brain MRI scan. The cost of the CT and MRI range from 138 US dollars to 223 US dollars. This sum is by far more than the national minimum wage of fifty dollars per month during the period of our study. We believe the high cost of CT and MRI as well as lack of adequate health insurance coverage may have contributed immensely to the delay in intervention. Only three patients had a close relative with history of CNS tumor in the past and were presented by their parents to the pediatric neurologist/ neurosurgeon months earlier than those who had no close family member with the brain tumor.
The most common symptoms [Table 5] were altered consciousness, vomiting, headache, limb weakness, deteriorating school performance and limb weakness. Symptoms experienced by patients were variable but consistent with clinical expectations regarding to tumor location in the brain. Cerebellar tumor mostly presented with gait impairment, poor coordination, nystagmus, dystonia, and in some cases hypotonia. Seizures were more frequent among patients with meningioma located over the cerebral convexity. Half of the patients had seizures, generalized seizures being more common than partial seizures. These seizures responded better to antiepileptic drugs (AEDs) when compared to brain stem seizures which were stereotypical, more frequent but short lasting.
The most common signs seen in about 25 of the 29 patients were papilledema and hyperreflexia, hypertonia, focal neurological deficits. Features of elevated ICP was seen in 21 of the patients, reduced level of consciousness and long tract signs in 18, cranial nerve deficit in 11, cerebellar sign in eight, whereas microcephaly and endocrinopathies in three patients [Table 5].
Despite presenting with well-defined and clinically recognizable localizing neurologic symptoms and signs, only just about a quarter of all of the 29 cases were initially suspected to be CBTs. The presence of headache alone was not helpful in diagnosis as malaria and a family history of migraine are mimics to CBT.
Delayed presentation continues to negatively impact disease outcomes in resource-poor nations where medical insurance coverage is still in single digits (5%) and out of pocket spending for health-care predominates. The average time spent before presenting to a health facility (from onset of symptoms to presentation to any medical facility) of up to four and half months is alarming. On the average it took a further 2 months after presentation to commence specific treatment modality. The wide variation in the time from presentation to commencing treatment was mainly due to financial constraint, reflecting the severe disparity in social class, with those in the lower social economic class having by far greater delay in receiving treatment. Delay in carrying out radiological and other laboratory investigations was also longer among patients of lower socioeconomic class when compared to those of middle or high socioeconomic class. Like in most published data on CBT most of the tumors were low grade tumor. More than 70% of all of the tumors graded in this study, with the WHO staging of the tumor, were low grade with only about a quarter being high grade. Tumors with high grade features were those graded 3 and 4 and are associated with worse prognosis. Histological verification was performed by Haematoxilin and Eosin (H&E) staining method which accesses tumor phenotype (2007 CNS WHO). Genomic categorization of brain tumors using genetic and molecular markers was not done as required by the 2016 review of the WHO classification of tumors of the CNS because the special stains and reagents for immunohistochemical evaluation of relevant biologic markers as well as techniques for genetic studies were not available in the study locations at the time of the study.
Misdiagnosis and use of alternative medicine contributed to the delay in hospital presentation. The delay experienced by CBT patients in getting specific treatment was even longer, with a mean duration of 6 months. We are of the opinion that this scale of delay has negatively affected patients’ survival. Further treatment time of 8 weeks from onset of symptoms to definitive treatment was observed. This finding has been corroborated in studies from southeast and southwest Nigeria.,, Presentation was much earlier in USA with better outcome even for infratentorial tumors.
Seventy percent of the mortality in our series occurred within the first 4 months of onset of symptoms before diagnosis, with one death occurring in a patient with the disease for up to 4–6 months [Table 6]. This tells the impact of delayed presentation and the effect of delayed diagnosis on the sufferers. One mortality occurred beyond the average duration for making a diagnosis which is 6 months, showing that if efforts are made to shorten diagnosis and intervention, the gap in CBT survival outcome between resource poor nations including ours and civilized nations may be narrowed. Survival indices after commencement of treatment shows that only one mortality occurred after the third year of follow-up.
Our opinion is that delay in presentation contributed to the profile of complication suffered among those patients treated for CBT who survived [Table 7]. Only two patients (7%) survived without complications. In the light of overall presentation of CBTs in our environment, our survival indices may summarize the expectations for the quality of life achievable by patients treated for CBT. However, our study is limited by its small sample size and may not be truly representative as we may have missed out many CBT’s especially those without histological proof. This may include brainstem and pineal region tumors as well as tumors in difficult to access intracranial sites. Furthermore, many CBT patients whose tumor histopathology were evaluated in private pathology laboratories and those who transferred their care to peripheral hospitals during the many labor strikes were also not accounted for.
The 2018 WHO fact sheet for 2018 gives a four times increased chance of being cured of brain tumor in high-income nations when compared to middle- and low-income nations. Early and accurate diagnosis followed with effective treatment are known to improve outcome. Significant improvement has been seen in medulloblastoma and low-grade glioma, whereas little progress has been made in treatment of other high grade tumor within the past three decades. The significance of this is noted as most gliomas in children are low grade and even high grade gliomas in children differ biologically from those seen in adults.,
| Conclusion|| |
Early presentation, a heightened attitude of clinical suspicion by health-care professionals at primary centers will help in ensuring prompt referral; reduce delay in diagnosis and treatment. This gap in knowledge may be narrowed by incorporating brain tumor diagnosis in continuing medical education for primary health workers. We believe this will lead to improved outcome of CBTs in a health system already burdened by more than its fair size of challenges. The future may appear brighter if multidisciplinary efforts at diagnosis and care of CBT are synergized.
Investing in newer modalities of diagnosis such as genetic and molecular markers will present to our centers, the proper tools required to implement current guidelines for brain tumor diagnosis, risk categorization, and prognostication. This will help us compare our experiences with locations defined by similar socioeconomic status as well as the entire global community.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| References|| |
Steliarova-Foucher E, Colombet M, Ries LAG, Moreno F, Dolya A, Bray F, et al
; IICC-3 contributors. International incidence of childhood cancer, 2001-10: A population-based registry study. Lancet Oncol 2017;18:719-31.
Patel AP, Fisher JL, Nichols E, Abd-Allah F, Abdela J, Abdelalim A, et al
. Global, regional, and national burden of brain and other CNS cancers, 1990–2016: A systemic analysis for Global burden of disease study. Lancet 2019;18:376-93.
Shehu UA, Adegoke SA, Abdulsalam U, Ibrahim M, Oyelami OA, Adeodu OA Pattern of childhood malignant tumors in two tertiary teaching hospitals in Nigeria! A comparative study. Niger J Paed 2013;40:175-8.
Roberta MC, Pedram R, Preston- Martin S Trends in childhood brain tumor incidence. J Neurooncol 2013;115:153-60.
Sarnaik AP, Kopec J, Moylan P, Alvarez D, Canady A Role of aggressive intracranial pressure control in management of pediatric craniocerebral gunshot wounds with unfavorable features. J Trauma 1989;29:1434-7.
Ogunbiyi J, Fabowale A, Ladipo A Cancer incidence and ten top cancers in eleven local government areas in Ibadan and its environs, 2004–2008. Ibadan Cancer Registry. Nigeria Technical report. Avaialble from: https://afcrn.org>index.php> 88-ibcr. [Last accessed 2020 Apr 5].
Oyedeji GA Socio-economic and cultural background of hospitalized children in Ilesha. Niger J Paediatr 1985;12:111-7.
Obalum DC, Fiberesima F Nigerian NHIS an overview. Nigerian Postgraduate Medical Journal 2012;19:167-74.
Ndubuisi CA, Ohaegbulam SC, Ejembi GO Paediatric brain tumours managed in Enugu, Southeast Nigeria. Review of one center experience. Niger Postgrad Med J 2018;25:186-90.
] [Full text]
Ogun GO, Adeleye AO, Babatunde TO, Ogun AO, Salami A, Brown BJet al
. Central nervous system tumors in children in Ibadan, Nigeria, a histhopathologic study. Pan Afr Med J 2016;24:34-45.
Uche EO, Shokunbi MT, Malomo AO, Akang EEU, Lagunju I, Amanor-Boadu SD Pediatric brain tumors in Nigeria, clinical profile, management strategies and outcome. Childs Nerv Syst 2013;29:1131-35.
Rosemberg S, Fujiwara D Epidemiology of pediatric tumors of the nervous system according to the WHO 2000 classification: A report of 1,195 cases from a single institution. Childs Nerv Syst 2005;21:940-4.
US Childhood Cancer Statistics. The American childhood cancer. Available from: https://www.acco.org> us-childhood. [Last accessed 2019 Dec 24].
Roger JP Childhood brain tumors, accomplishments and ongoing challenges. J Child Neurol 2008;23:1122-27.
Gupta S, Howard SC, Hunger SP Treating childhood cancers in low and middle income countries. In: Disease control priorities. Vol 3. Available from: http//dcp-3.org/chapter900/treating childhood cancers-low-and-middle-income-countries. [Last accessed on 2019 Dec 24].
Pollack IF, Finkelstein SD, Woods J, Burnham J, Holmes EJ, Hamilton RL, et al
; Children’s Cancer Group. Expression of p53 and prognosis in children with malignant gliomas. N Engl J Med 2002;346:420-7.
Sobel EL, Gilles FH, Leviton A, Tavaré CJ, Hedley-Whyte ET, Rorke LB, et al
. Survival of children with infratentorial neuroglial tumors. The childhood brain tumor consortium. Neurosurgery 1996;39:45-54; discussion 54-6.
[Table 1], [Table 2], [Table 3], [Table 4], [Table 5], [Table 6], [Table 7]