PDF
Citation
Print
Introduction
A brain computed tomography (CT) scan, is an accessible, simple, and very common medical examination which may provide much information about morphologic alterations of the brain. Although it uses ionizing radiation, the benefits readily outweigh the risks such as cancer (0.001%) if the optimized dose is used and the indications are logical [1]. Regardless of having diagnostic benefits in a wide range of conditions such as head trauma, headache, tumors, stroke, epilepsy, hydrocephalus, aneurysms, and other pathologic conditions, a brain CT scan can be informative of cortical, and subcortical atrophy, physiologic intracranial calcification, and intracranial arterial calcification which may act as predictors of certain future cognitive conditions or stroke [2-5].
Cortical and subcortical atrophy, as predictors of cognitive conditions such as Alzheimer’s disease, can be detected in brain CT scan by measuring the sulci widening and ventricular indices, respectively [5]. Studies investigating these two factors in normal participants report significant changes in cortical atrophy—sulci widening [5]—, and subcortical atrophy—increased ventricular indices such as Evans’ index (EI) [4]—in an age-dependent manner.
In this study, a whole generation (aged 1–99 years) of normal participants who have done brain CT scans because of head trauma or headache was investigated to reveal the common pattern of age-related morphologic alterations of the brain including cortical and subcortical atrophy.
Materials and Methods
Participants and computed tomography
This cross-sectional retrospective study was conducted after Institutional Review Board approval (IR.AJAUMS.REC.1402.049). Brain CT scans of 400 patients were investigated retrospectively for one year (2020 to 2022) in the medical imaging center of Imam Reza AJA Hospital, Tehran, Iran. The study included 20 groups each containing 20 patients: group 1 (1–9 years old male), group 2 (1–9 years old female), group 3 (10–19 years old male), group 4 (10–19 years old female), group 5 (20–29 years old male), group 6 (20–29 years old female), group 7 (30–39 years old male), group 8 (30–39 years old female), group 9 (40–49 years old male), group 10 (40–49 years old female), group 11 (50–59 years old male), group 12 (50–59 years old female), group 13 (60–69 years old male), group 14 (60–69 years old female), group 15 (70–79 years old male), group 16 (70–79 years old female), group 17 (80–89 years old male), group 18 (80–89 years old female), group 19 (90–99 years old male), group 20 (90–99 years old female).
The CT imaging device was Hitachi-Supria 16/32 with 51 kW power, 75 cm gantry bore, 180 cm scan range, 5 mega-hit unit X-ray tube, and 0.675 mm minimum slice thickness [2, 6].
Before entering the study, the medical history of patients was reviewed and patients with cognitive disease, severe diabetes, previous stroke, and renal failure were excluded from the study and cases with head trauma and headache who had no other pathological conditions were entered the study. Patients were supposed to stay calm to avoid motion artifacts during the CT scan procedure. Images were acquired from the foramen magnum to vertex (5 mm thickness and 5 mm interval), then the images were reconstructed with desired thickness and interval in various planes of axial, and sagittal to analyze the variables of the study more accurately. Cases were reviewed by two blind radiology residents and pathological cases and disagreements were removed from the study.
Subcortical atrophy
In this study, subcortical atrophy contains two categories of parameters: firstly, the major ventricular parameters, and secondly, derived ventricular parameters (ventricular indices). Major ventricular parameters include, A1: maximum width of frontal horns, A2: maximum inner diameter of cranium, A3: maximum inner diameter of cranium at the level of A1, B1: minimum bicaudate width (BCW) at the level of A1, B2: maximum inner diameter of cranium at the level of B1, C1: bioccipital width of occipital horns of lateral ventricles, C2: maximum inner diameter of cranium at the level of C1, D1: maximum width of third ventricle, D2: maximum inner diameter of cranium at the level of D1, E1: bitemporal width of temporal horns of lateral ventricles, E2: maximum inner diameter of cranium at the level of E1, F1: maximum width of lateral ventricles at the level of cella media, F2: maximum inner diameter of cranium at the level of F1, G1: maximum length of lateral ventricles at the level of cella media, G2: maximum inner diameter of cranium at the level of G1, H: the width of neural tissue between lateral ventricles at the level of cella media, J: maximum width of left temporal horn of lateral ventricles, K: maximum width of right temporal horn of lateral ventricles (Fig. 1). And derived parameters include EI: A1/A2, BFI: bifrontal index (A1/A3), BCI: bicaudate index (B1/B2), CMI1, 2, 3, 4: cella media index 1 (F1/A2), 2 (F1/F2), 3 (G1/G2), 4 (F1-H/F2), PORI: parieto-occipital ratio index (C1/C2), TRI: temporal ratio index (E1/E2), TWI: temporal width index (J+K), HI: Huckman index (A1+B1), ThVI: third ventricle index (D1/D2) [4, 5, 7].
Symmetry of frontal horns of lateral ventricles
Frontal horns of lateral ventricles were investigated to reveal the amount of symmetry of the ventricles. The maximum width of the right (a) and left (b) frontal horns of the lateral ventricles were measured at the level of minimum BCW. Symmetry, slight, moderate, and extreme asymmetry are concluded when (a/b)2 is ≤1, 1<(a/b)2 ≤2, 2<(a/b)2 ≤4, (a/b)2 >4, respectively (Fig. 2).
Cortical atrophy
Cortical atrophy was investigated by measuring the width of sulci in three regions including the maximum interhemispheric sulcus width (IHSW), average right hemisphere sulci diameter (RHSD), and average left hemisphere sulci diameter (LHSD) (measuring the width of at least five sulci in axial and sagittal planes) (Fig. 3).
Statistical analysis
The normality of the data was investigated by the Kolmogorov–Smirnov test, then the distribution of all of the variables with age was investigated by scatter plot (with linear interpolation) and linear regression (with the line of best fit). After that, the variables which were in a significant relationship with age were selected and underwent detailed statistical analyses. The mean of these significant variables was compared between groups by analysis of variance (post-hoc Tukey) and demonstrated by column diagrams. Multiple linear regression was performed to investigate the correlation of variables with age and each other, and the model of prediction of dependent variables by significant independent variables was provided. The sensitivity of linear regression formulae in predicting the actual dependent variables was also tested and reported by comparing them with the original data. All the data were provided as mean±SD, and values of P≤0.05 were considered significant. Statistical analyses and diagrams were done by SPSS software, version 16 (SPSS Inc.) and Microsoft Office Excel, version 2013 (Microsoft).
Results
Demographic data
Normal brain CT scans of 400 participants (male [n=200], and female [n=200]) were divided into 20 age groups including group 1 (4.41±0.81 years old), group 2 (4.87±1.05 years old), group 3 (15.25±0.75 years old), group 4 (14.83±0.71 years old), group 5 (25.02±1.36 years old), group 6 (23.14±1.07 years old), group 7 (35.15±1.23 years old), group 8 (33.85±1.01 years old), group 9 (44.71±0.91 years old), group 10 (45.85±1.12 years old), group 11 (52.85±1.12 years old), group 12 (54.57±0.71 years old), group 13 (63.15±1.06 years old), group 14 (63.51±1.38 years old), group 15 (74.66±0.84 years old), group 16 (73.62±1.03 years old), group 17 (83.21±0.92 years old), group 18 (83.91±0.67 years old), group 19 (92.22±1.06 years old), group 20 (91.75±0.62 years old) (Table 1).
Major ventricular parameters
Amongst the major ventricular parameters in this study which are all mentioned in each group in Table 2, three parameters show significant correlation with age which are BCW (R2=0.67, P<0.001), third ventricle width (ThVW) (R2=0.64, P<0.001), and the maximum length of lateral ventricle at cella media (MLCM) (R2=0.62, P<0.001) (Table 3). Significant changes are evident in BCW in males at the age of 70–79 to 90–99 compared to group 1, but in females, a remarkable difference is seen at the age of 80–89, and 90–99 (Fig. 4A–C). Also, significant changes are evident in ThVW in males at the age of 60–69 to 90–99 compared to group 1, but in females, a remarkable difference is seen at the age of 70–79 to 90–99 (Fig. 4D–F). Furthermore, remarkable changes are observed in MLCM in males at the age of 70–79 to 90–99 compared to group 1, but in females, a remarkable difference is seen at the age of 80–89, and 90–99 (Fig. 4G–I). Although other major ventricular parameters are not significantly correlated with age, they show a significant increase in higher age groups compared to lower age groups (Table 2).
Frontal horns of lateral ventricle asymmetry
Frontal horns of lateral ventricles demonstrated 47.6% symmetry, 19.7% slight asymmetry, 13.6% moderate asymmetry, and 19% extreme asymmetry (Fig. 5). Frontal horn asymmetry did not show a significant correlation with age.
Ventricular indices
Several ventricular indices were investigated in this study (Table 4), but only six of them, according to linear regression results, were in a significant relationship with age which are EI (R2=0.56, P<0.001), BFI (R2=0.44, P<0.001), BCI (R2=0.71, P<0.001), CMI3 (R2=0.64, P<0.001), CMI4 (R2=0.60, P<0.001), and ThVI (R2=0.65, P<0.001) (Table 3). A significant increase in EI is evident in males and females at the age of 10–19 to 90–99 compared to group 1, and the next significant difference is observed at the age of 70–79 in males, and 80–89 in females (Fig. 6A, B). There is also a significant increase in BFI in males and females at the age of 50–59 to 90–99 compared to group 1, and the next significant difference is observed at the age of 80–89 and 90–99 in male and female (Fig. 6C, D). A remarkable increase in BCI is observed at the age of 60–69 to 90–99 compared to group 1 in males and females (Fig. 6E, F). A significant increase in CMI3 is evident in males and females at the age of 50–59 to 90–99 compared to group 1 (Fig. 6G, H). There is also a significant increase in CMI4 in males and females at the age of 30–39 to 90–99 compared to group 1 (Fig. 6I, J). And ThVI demonstrates a significant increase at the age of 50–59 to 90–99 compared to group 1 (Fig. 6K, L). Although other ventricular indices are not significantly correlated with age, they show a significant increase in higher age groups compared to lower age groups (Table 4).
Cortical atrophy
Cortical atrophy was investigated in three areas including IHSW, RHSD, and LHSD. Maximum IHSW was significantly correlated with age (R2=0.63, P<0.001), and a remarkable increase was seen in age groups of 60–69 to 90–99 compared to group 1 (Fig. 7A, B). Average RHSD was also significantly correlated with age (R2=0.74, P<0.001) and demonstrated a considerable increase in age groups of 60–69 to 90–99 compared to group 1 (Fig. 7C, D). Furthermore, average LHSD was correlated with age significantly (R2=0.73, P<0.001) and showed a meaningful increase in age groups 60–69 to 90–99 compared to group 1 (Fig. 7E, F). Fig. 8 demonstrates how sulci widen through aging in the right hemisphere.
Multiple linear regression studies
Multiple linear regression of the three major ventricular parameters, six derived ventricular indices, and cortical atrophy variables demonstrated higher correlations with age (R2=0.74, 0.76, and 0.78, respectively, P<0.001) providing formulae for age (as a dependent variable) prediction. Variables were also correlated with each other including a significant correlation between cortical atrophy parameters and ventricular variables (R2≥0.73, P<0.001) (Table 5).
Discussion
Brain atrophy is considered a physiologic process in aging, however, infections, nutritional deficiency, metabolic and endocrine causes, trauma, drugs, and neurodegenerative diseases are among pathological causes which should be monitored and prevented [8]. Clinical manifestations of cognitive impairment, dementia, depression, gait, and visual disturbance are among the symptoms which imply brain atrophy [9]. Neuro-imaging (CT scan, and magnetic resonance imaging [MRI]) can be served as an invaluable diagnostic tool to monitor and detect brain atrophy commonly as ventricular dilation and sulci widening [8, 10].
This study investigated the morphologic alterations of a normal brain through aging including ventricular changes, and cortical atrophy to provide reliable criteria for age-related degenerative changes in the brain. Results of this study highlighted important variables which are significantly correlated with age including major ventricular variables (BW, ThVW, and MLCM; R2≥0.64, P<0.001), derived ventricular indices (EI, BFI, BCI, CMI3, CMI4, and ThVI; 0.44≤R2≤0.71, P<0.001), and cortical atrophy (IHSW, RHSD, and LHSD; 0.63≤R2≤0.74, P<0.001).
BW, ThVW, and MLCM and their corresponding derived indices (BCI, ThVI, and CMI3) demonstrated significant changes at the age groups of 70–79 years, 60–69 years, and 50–59 years, respectively, showing MLCM, and CMI3 as sensitive-to-age variables which detect early morphologic alterations of the brain. He et al. [11] reported the same parameter (anteroposterior diameter of the lateral ventricle) as an accurate marker for ventricular volume and claimed values larger than 0.50 can be considered as ventricular enlargement in elderly people. In our study, normal ranges of major ventricular parameters and the derived ventricular indices for each decade of life are provided (Tables 2 and 4), so that clinicians can detect abnormal ventricular enlargement considering our data as standard normal brain variables. Our data also show no values higher than 0.50 for CMI3.
Several studies have referred to the positive relationship between aging and ventricular indices changes [3-5, 11-13]. Currà et al. [4] investigated three ventricular indices in more than 3,000 brain scans, and reported earlier ventricular enlargement in males than females, and also demonstrated significant changes in Evans, bitemporal, and parieto-occipital indices at the ages of 79, 83, and 87 years, respectively. Our study also showed earlier ventricular enlargement in males than females, and the values of the three mentioned indices in our study were in line with Currà et al. [4] study.
Since the literature review suggests the significance of ventricular symmetry [14, 15], it was also investigated in our study. Kiroğlu et al. [15] investigated lateral ventricles symmetry at the level of frontal horns and reported 16.5% severe cases of asymmetry who had more headaches than patients with symmetric ventricles. In our study 47.6, 19.7, 13.6, and 19% of cases had symmetric, slightly, moderately, and extremely asymmetric ventricles, respectively. It is suggested clinicians pay attention to severe asymmetric ventricles or increased ventricular volume, because these conditions may accompany other disorders.
Cortical atrophy was investigated by sulci widening (in three regions of IHSW, RHSD, and LHSD) which was significantly correlated with age and demonstrated a considerable increase in the age group of 60–69 years. Age-related cortical atrophy was reported by Chrzan et al. [5] in age groups of 70–79 to 100–106 years showing a significant increase with age (6 to 8 mm). Neurodegenerative diseases such as Alzheimer’s disease and multiple sclerosis cause brain atrophy [16, 17]. Mechanisms leading to brain atrophy in Alzheimer’s disease include multiple immunological reactions caused by an imbalance between amyloid beta peptides production and clearance [16], in multiple sclerosis brain atrophy is caused by demyelination in an inflammatory process [17]. Therefore disclosing normal ranges of cortical atrophy through aging is of significant importance to differentiate between a physiologic or pathologic process. In our study, RHSD and LHSD were in line with Chrzan et al. [5], reporting normal ranges of sulci widening in every decade of aging.
Correlation studies revealed not only a significant relationship between variables with aging but also a remarkable relationship between the variables with each other. Multiple linear regression results demonstrated a significant correlation between cortical atrophy variables (IHSW, RHSD, and LHSD) and ventricular variables (BW, ThVW, MLCM, EI, BFI, BCI, ThVI, CMI3, and CMI4) (R2≥0.73, P<0.001).
Open access data bases which have investigated the morphology of brain in normal and various medical conditions (Alzheimer’s disease or schizophrenia) are available [18]. These investigations provide data from structural or functional MRIs which are not readily accessible [18]. Several studies have investigated the relationship between structural volume of hippocampus and behavioural measurements of memory performance [19, 20]. One study has investigated cortical changes in normal aging patients and those patients with temporal lobe epilepsy showing several identical and different cortical changes emphasizing on the importance of morphologic investigations in diagnosing neurological conditions [21].
In conclusion, this study is a thorough investigation of cortical and subcortical variables in a normal brain by CT scan, as an easily accessible medical imaging tool, which can be affected by aging disclosing normal ranges of variables including major ventricular variables, derived ventricular indices, lateral ventricles asymmetry, and cortical atrophy, in every decade of life introducing BW, ThVW, MLCM, BCI, ThVI, CMI3 as most significant ventricular parameters, and IHSW, RHSD, LHSD as significant cortical parameters associated with age. These morphometric data can be enforced with serology data of patients and compared with various diseases for future diagnostic methods.
XML Download