Menu
Menu
Menu
Menu
Background The brain volume loss also known as brain atrophy is increasingly observed among children in the course of performing neuroimaging using CT scan and MRI brains. While severe forms of brain volume loss are frequently associated with neurocognitive changes due to effects on thought processing speed, reasoning and memory of children that eventually alter their general personality, most clinicians embark themselves in managing the neurological manifestations of brain atrophy in childhood and less is known regarding the offending factors responsible for developing pre-senile brain atrophy. It was therefore the goal of this study to explore the factors that drive the occurrence of childhood brain volume under the guidance of brain CT scan quantitative evaluation. Methods This study was a case-control study involving 168 subjects with brain atrophy who were compared with 168 age and gender matched control subjects with normal brains on CT scan under the age of 18 years. All the children with brain CT scan were subjected to an intense review of their birth and medical history including laboratory investigation reports. Results Results showed significant and influential risk factors for brain atrophy in varying trends among children including age between 14-17(OR = 1.1), male gender (OR = 1.9), birth outside facility (OR = 0.99), immaturity (OR = 1.04), malnutrition (OR = 0.97), head trauma (OR = 1.02), maternal alcoholism (OR = 1.0), antiepileptic drugs & convulsive disorders (OR = 1.0), radiation injury (OR = 1.06), space occupying lesions and ICP (OR = 1.01) and birth injury/asphyxia (OR = 1.02). Conclusions Pathological reduction of brain volume in childhood exhibits a steady trend with the increase in pediatric age, with space occupying lesions & intracranial pressure being the most profound causes of brain atrophy.
Four facilities were included in the study, namely Agakhan Health Center, the Arusha Lutheran medical Center, Afyamax polyclinic all in Arusha region and the Kilimanjaro Christian Medical Center in Kilimanjaro region of Tanzania. The centers were picked intentionally for individuals residing inside Northern Tanzania on the basis of availability and access to CT scan imaging services. During this study period, the two locations, namely Tanga and Manyara, did not have functioning CT scans. Their patients were however sampled by referral system set by the national health care network system. Charan & Biswas, 2013, derived the sample size from a statistical formula, by the guide that part of the investigation concerned quantitative variables in case control [19]. Where as: Standard variable Z1-α /2 standard (the error is 1.96 at 5% type 1 (P<0.05). The standard deviation value was derived from Holger Schmidt’s previous work and 2.3 was found [20]. d = The expected mean difference between case and control but in this case, half the value of the SD was considered to be 1.15. r = Ratio of control to cases, 1 for equal number of Case and Control. Z = Standard normal power variate; the value is 1.28 for 90% power. In this study, the 90% power was selected with SD (2.3) and d values (1.15), the sample size was calculated to be 168 and when the ratio of 1:1 [21] was considered the study then recruited 168 cases and 168 controls. Cases and controls were included after meeting inclusion criteria by being (i) less than 18 years of age; (ii) have performed brain CT scan. (iii) Accessibility and availability of quality CT scan image of patients. (iv) Presence of the biological mother for the birth and early infancy history. The exclusion criteria of the study involved (i) Individuals of 18 years and above. Children who were not born and raised in Northern Tanzania were excluded. The matching criteria involved age and sex. Case-control ratio of 1:1 with age interval of 2 years was considered. A radiologist examined the Brain CT-scan images, which evaluated sulcal width, lateral ventricular body width and the Evans index using the three familiar radiological linear procedures. Diagonal multi-linear (DBF) approach was calculated to identify the presence or absence of brain atrophy for those who qualified for the use of this technique [22]. The collection of data included questionnaires designed to assess the presence of 10 non- demographic categories of risk factors using primary information from mothers of children covering medical history from antenatal to adolescents, examination of obvious risk factors through image analysis such as tumors, intracranial pressure, hypoxic-ischemic encephalopathy as well as head injury was conducted (Fig 1). Additional data as age and sex were considered from hospitals data base. After parents’ approval, children whose results for HIV tests were not found were subjected to such tests. A: Volume rendering 3D CT scan image of a 3 months’ child with birth related head injury presenting with cephalohematoma at the vertex. B: Axial CT scan image of the same child A showing diffuse reduction in brain volume in cortical regions as evidenced by prominent sulci. C: Volume rendering 3D image of 1-year male child showing premature closure of the sagittal suture with overall increase in anterior-posterior diameter of the head(craniosynostosis). D: Sagittal CT brain image of the same child C showing loss of brain volume in the frontal lobes showing gross deviation from the calvarium with prominent CSF spaces. E: A 2 months’ female child presenting with diffuse reduction in cerebral hemispheric density sparing the cerebellum (White cerebella sign) representing hypoxic ischemic encephalopathy with early bi-temporal cerebral atrophy. F: A 4 months’ male child showing increased intracranial pressure evidenced by effacement of sulci and gyri as a result of ballooning of the lateral and 3rd ventricles. The 4th ventricle is small due to obstructive hydrocephalus following aquiductal stenosis. The descriptive statistics such as frequency, percentage and graphical visualization (bar graphs) were produced. Some inferential statistics were also performed. A simple linear regression model was produced to assess the impact of risk factors such as head trauma, birth asphyxia among others toward the response variable (brain atrophy). The significant risk factors were then exported to the multiple linear regression analysis. All liner regression assumptions were checked including normality of the continuous variables. Shapiro-Wilk test and variance inflation factor (VIF) were used to check for normality and collinearity tests respectively. Variables with VIF above 10 were removed from the model otherwise they were retained in the model development. All tests were performed at 5% confidence level. Also, the coefficient of determination (R-square) was used to check for model goodness of fit. The R-statistical software version 4.0.1 was used for all analyses. For ethical approval of this study, the Northern Zone Health Research Ethics Committee (KNCHREC) was consulted, and a research permit number KNCHREC 0010 dated 20th December 2019 was obtained. Covering letters were then sent to each health facility, requesting permission to collect and use data from their archives within the parameters of this study at no additional cost. Once the study was explained to the participants, their parents or guardians freely consented to their participation by signing pre-written forms. Patients’ names and other personal identifiers were not recorded in order to maintain confidentiality. The retrospective data components were anonymized.
N/A
