Keywords: Asymmetry, callosal angle, Evans index, hydrocephalus, SOV
The right and left lateral ventricles are the largest parts of the ventricular system of the brain. After it is produced in the choroid plexus of the lateral ventricles, the cerebrospinal fluid (CSF) follows a pathway from the lateral ventricles into the third ventricle through the interventricular foramen, then into the fourth ventricle through the cerebral aqueduct, and at last, into the subarachnoid space through the median and lateral apertures. Finally, the CSF is absorbed by the arachnoid granulations and drained into the venous and lymp LESS absorbtion hatic circulation. CSF is produced at a rate of 0.2 to 0.7 mL/minute and 600 to 700 mL/day. Turnover of the entire CSF is three to four times/day. This dynamic process causes variable intraventricular CSF volume and pressure. When a problem with this production and absorption equilibrium exists, CSF abnormally accumulates in the ventricles, Causing ventricular enlargement known as “hydrocephalus.” Primarily, there are two types of hydrocephalus—communicating hydrocephalus (CH) and noncommunicating hydrocephalus (NCH). CH occurs when there is a problem either with the production or absorption of the CSF. Overproduction Due to choroid plexus tumors or underabsorbtion because of arachnoid granulation pathologies is some of the causes of CH. In the elderly population, we see another type of CH that is more common. It is called the normal pressure hydrocephalus (NPH). NPH occurs when intraventricular pressure increases because of excess CSF, and ventricles become enlarged.,,,,, It may be idiopathic or secondary to brain trauma, cranial surgery, or subarachnoid bleeding. The increase in pressure might be temporary, but the enlargement of the ventricles is usually permanent unless the fluid is drained by a shunt., In noncommunicating or so-called obstructive hydrocephalus, there is a stenosis along the passage of CSF. The stenosis is mostly at the level of cerebral aqueduct when it is congenitally caused by aqueductal stenosis.When it is congenital. Other congenital causes are neural tube defects, Dandy–Walker syndrome (DWS), and Chiari malformation More Detailss. The acquired causes are pineal gland tumors (mostly pineal cysts), tectal gliomas, and cerebral-cerebellar and intraventricular tumors. The mass effect leads to NCH by blocking CSF passage. Any intraparenchymal, and intraventricular pathology, such as hemorrhage, infection, and inflammation, can cause NCH by blocking passage.
The lateral ventricles are vertically separated by a thin midline septum called the septum pellicidum. On each side of the septum, lateral ventricles are like a mirror image of each other and, most of the time, are identical and symmetrical. However, asymmetrical lateral ventricles (ALV) or unilateral hydrocephalus are not rare. The asymmetry can be caused by a unilateral intra or extraventricular space-occupying lesion that obstructs the ipsilateral ventricle or diseases, such as Parkinson's disease, which causes unilateral volume loss and ventricular enlargement because of atrophy., There are some studies showing ALV correlation with developmental disorders, mental retardation, autism, schizophrenia, and anorexia nervosa in children and adults.,, Some studies also have linked asymmetry to right- or left-handedness.,
However, we encounter ALV more frequently in the normal population., Unless a lesion or A disorder accompanies, the degree of asymmetry is not measured routinely and is not even noticed most of the time. However, if the asymmetry is noticeable and obvious in an otherwise normal MRI, it is usually reported as normal variant.
During routine practice, we use some indices and measurement methods for quantifying ventricular enlargement/hydrocephalus. The most common ones are the Evans index,,, and callosal angle,, which have been in use since the early 1940s. Despite some limitations and controversies against them, they are easy to use, practical, and reproducible. The Evans index values greater than 0.30, are compatible with hydrocephalus and lower values are regarded as normal. The callosal angle is 100° to 120° in the normal population, and studies have shown that, values less than 90° are suggestive of NPH.
The ventricular system and CSF are closely connected to the venous system of the brain. The enlargement of the ventricles affects venous structures by increasing the intracranial pressure, blocking venous backflow, and causing venous dilatation. The SOV is the largest and most easily identified orbital venous structure. There is a single SOV in each orbit/hemisphere, and similar to lateral ventricles, they are assumed to be identical. The SOVs drain into the CSF through the cavernous sinuses, which are extensions of the cranial venous drainage and are easily affected by the CSF hemodynamics. It has been shown in some studies that SOV diameter is positively correlated with intracranial pressure and bilateral enlargement should be a warning sign for hydrocephalus., On the contrary,, Bright bill et al. found no relationship between the size of SOVs and hydrocephalus.
In our study, we aimed to investigate the lateral ventricular asymmetry in ALV and hydrocephalus by measuring the septum pellicidum shift, Evans index, right and left semi-Evans indexes, callosal angle and right and left semi-callosal angles, and SOV diameters. We aimed to compare the results with the normal population and discuss the literature in the light of the results.
Patient population and study design
This retrospective study included 256 patients who received a brain magnetic resonance imaging (MRI) scan because of various symptoms between the years 2011 to 2016 in our university hospital's department of radiology. Ninety-six (37.5%) male and 160 (62.5%) female patients (mean age 60.95 ± 15,55 years; range 18–94 years) were enrolled. Of those, 93 patients had hydrocephalus (idiopathic normal pressure hydrocephalus [iNPH] and obstructing hydrocephalus), 80 patients had ALV, and 83 patients with normal findings on MRI were included as a control group. We grouped these patients as group 1 (hydrocephalus), group 2 (ALV), and group 3 (normal findings). Only adult patients (≥18 years old) were included in the study. Patients with hydrocephalus were identified in our database by a search method. We used the terms “hydrocephalus,” “dilatation,” “normal pressure hydrocephalus,” “idiopathic normal pressure hydrocephalus (iNPH),” “NPH,” and “iNPH” for collecting the hydrocephalus group, and we used the terms “asymmetrical,” “variant,” and “psychological” for the ALV group. The CH cases due to corpus callosum metastasis and over-production of CSF because of tumors or radiotherapy were excluded and the rest were defined as iNPH according to the clinical status (triad of symptoms of gait disturbance, dementia, and urinary incontinence), MRI findings, and follow-up data. The iNPH cases with associated pathologies, such as volume loss due to cute-chronic ischemic changes, and tumoral lesions were excluded, and only 80 iNPH cases with no other pathologies were included. For the noncommunicating group, 13 cases were collected. For the ALV group, only cases with no other pathological findings were selected.
A senior radiologist with 15 years of neuroradiology experience, who was blinded to the patient history, evaluated the images. We randomly selected 45 patients (15 from each group) to be evaluated by another radiologist to check interobserver reliability. Our local ethics committee approved the study, and signed consent was obtained from the patients.
The MRI scans were performed by a 1.5 Tesla (T) MRI scanner (Intera, Philips Medical Systems, Best, The Netherlands) with a dedicated four-channel head coil when the patient was in the supine position. The T2-weighted turbo spin-echo sequence in axial plane (TE/TR 110/5533 ms, flip angle 90°, FOV 230 Õ 185 Õ 136 mm3, acquired voxel size 0.8 Õ 0.8 mm2, reconstructed voxel size 0.449 mm, and total scanning time 138 s) and T1-weighted spin-echo sequence in coronal plane (TE/TR 13/627 ms, flip angle 69°, FOV 230 Õ 183 Õ 136 mm3, acquired voxel size 0.9 Õ 1.1 mm2, reconstructed voxel size 0.9 mm, and total scanning time 104s) were used for the measurements. A contrast agent of 0.1 mmol/kg gadolinium diethylenetriaminepentaacetic acid (Gd-DTPA) had been used for various reasons (e.g., tumors and postoperative) for 44 cases.
The Evans index and callosal angle were measured for each case. The Evans index was defined as a/b; a: the maximal transverse diameter of the frontal horn of lateral ventricles on the axial plane; b: the maximal transverse diameter (from one inner table to opposite outer table) of the skull on the axial plane (measured from the same slice as a). We measured the indexes of the right and left hemisphere separately and called it the semi-Evans index. The right semi-Evans index is defined as c/d; c: the maximal transverse diameter of the frontal horn of the right lateral ventricle on the axial plane; d: the maximal transverse diameter measured from the right inner table of the skull to the midline on the axial plane (measured from the same slice as a). The left semi-Evans index was defined as e/f; e: the maximal transverse diameter of the frontal horn of the left lateral ventricle on axial plane; f: the maximal transverse diameter measured from the left inner table of the skull to the midline on the axial plane (measured from the same slice as a). [Figure 1]. An axial turbo spin echo T2-weighted sequence (TSE T2WS) was used for the Evans indexes. The callosal angle was defined as the anterior angle between the lateral ventricle horns on the coronal plane.It was measured through a perpendicular line to the anterior commissure-posterior commissure on TSE T2WS on the sagittal plane. Values above 40° and below 90° are diagnostic for NPH (24). We divided the angle anteriorly into two parts, perpendicular to the midline on the same slice that the callosal angle was measured and called these two angles the right and left semi-callosal angles [Figure 2].
We also investigated the septum pellicidum shift in the ALV and control groups. Septum shift was defined as the transverse distance from the septum pellicidum to the vertical line passing through midline of the cranium (line between anterior and posterior falx attachments) at the level of the frontal horns. It was measured only in the second and third groups and not measured in the first group because of the angular and “S”- shaped appearance of the septum caused by increased pressure in most of the cases. Septum variations, such as cavum septum pellicidum, were excluded.
The largest dimension of the SOVs were measured, perpendicular to the long axis on the nonfat saturated spin-echo T1WS coronal plane [Figure 3]. The measurements were made only on precontrast fatsat T1WS because it was noticed that the diameters were overestimated on contrast-enhanced sequences. Asymmetry of the right and left SOV was defined as a 0.3 mm or greater difference between the SOV of the right and left globe.
Inclusion and exclusion criteria
From the search methods, we found 294 adults (≥18 years of age) patients. Thirty-eight patients were excludedand a final number of 256 patients were left for the study.
The exclusion criteria were septum pellicidum variations, such as cavum septum pellicidum and septum pellicidum et vergae (n = 12); “S”-shaped septum pellicidum in the ALV and control groups (n = 11); a history of globe operation (n = 1); unilateral SOV enlargement due to other pathologies (n = 3); history of shunt surgery (n = 8); and history of intracranial surgery (n = 3).
The NumberCruncher Statistical System 2007 (Kaysville, Utah, USA) software was used for statistical analysis. When evaluating the study data, in addition to descriptive statistical methods (mean, standard deviation [SD], median, frequency, minimum, and maximum), for quantitative data, the Mann–Whitney U test was used in the comparison of two groups not showing normal distribution and Student's t-test was used in the comparison of two groups showing normal distribution. In the groups (three or more) showing normal distribution, a one-way analysis of variance (ANOVA) test was used and the Tukey test was used to detect the group causing the differences. In the comparison of three or more groups not showing normal distribution, the Kruskal–Wallis test was used, and in the determination of the group showing the difference, the Mann-–Whitney U test was used. A paired samples test was used in the comparison of intergroups showing normal distribution. Spearman's correlation analysis was used to measure the degree of association between parameters, and the intraclass correlation coefficient (ICC) was used for assessing inter-rater reliability. The Pearson's Chi-square test and Fisher's exact test were used for comparing independent variables. Significance was considered at a level of P < 0.05.
The characteristics of the patients are summarized in [Table 1].A statistically significant difference was found between the age distributions of the patients. (P = 0.001; P < 0.01). According to the Mann–Whitney U test, the age of patients in group 1 was significantly higher than the age of patients in groups 2 and 3 (P = 0.001; P = 0.001). The age of patients in group 2 was significantly higher than that of group 3 (P = 0.014; P < 0.05).
No statistically significant difference was found between gender distribution of the groups (P = 0.409; P > 0.05).
Evans indexes of groups were A statistically significantly different as expected (P = 0.001; P < 0.01) [Table 2]. According to the Mann–Whitney U test results, the Evans indexes of group 3 were significantly lower than those of other groups (P = 0.001; P = 0.001; P = 0.001), and the Evans indexes of group 2 were significantly lower than those of group 1 (P = 0.001; P = 0.001).
Also, statistically significant difference was found between the right and left semi-Evans indexes of groups (P = 0,001; P < 0.01) [Table 2]. According to the Mann–Whitney U test results, the right and left semi-Evans indexes of group 3 were significantly lower than those of other groups (P = 0.005; P = 0.001; P = 0.001), (P = 0.001; P = 0.001; P = 0.001), respectively. And the right and left semi-Evans indexes of group 2 were significantly lower than those of group 1 (P = 0.001; P = 0.001), (P = 0.001; P = 0.001), respectively.
The cut-off point for Evans index for differentiating groups 1 and 2 was calculated as 28.68% and higher. This cut-off point for diagnosing hydrocephalus had a sensitivity of 98.9% and a specificity of 87.5%, a positive predictive value of 90.2, and a negative predictive value of 98.6. The area under the curve was 98.1% with a standard error of 0.8%.
The cut-off point for right and left semi-Evans indexes for differentiating groups 1 and 2 were calculated as 30.77% and 30.88% and higher, respectively. These cut-off points for diagnosing hydrocephalus had sensitivity of 98.9% and 79.5%, respectively, and specificity of 93.7% and 83.7%, respectively. The positive predictive value for differentiating groups 1 and 2 were 94.0 and 85.1, respectively, and the negative predictive values were 83.3 and 77.9, respectively. The area under the curve was 94.9% and 90% with a standard error of 1.6% and 2.2%, all respectively.
A statistically significant difference was found in the values of the right and left semi-callosal angle between the hydrocephalus group and the other groups (P = 0.001, P < 0.01; P = 0.001, P < 0.01). According to the Mann–Whitney U test results, the right and left semi-callosal angle of group 1 were significantly lower than those of other groups (P = 0.001, P = 0.001, P < 0.01; P = 0.001, P = 0,001, P < 0.01, respectively). No statistically significant difference was found between the callosal angle and the right-left semi-callosal angle of the ALV and control groups. The mean callosal angle of groups 1, 2, and 3 (iNPH and obstructive) were 135.60 ± 6.85, 134.75 ± 7, 84, 112.64 ± 21.60, 117, and 54 ± 18.83, respectively.
No statistically significant difference was found in the values of right and left SOV between gender (P > 0.05) and age (P > 0.05) and between the dimensions of right and left SOVs of the three groups (P = 0.64, P > 0,05) [Table 3].
Asymmetrical SOVs (≥ 0.3-mm difference between diameters of right and left SOV) were found in 16.9% of the normal group, 17.5% of the ALV group, and 18.8% of the hydrocephalus group. A larger right SOV was found in 43.4% of group 3, 52.5% of group 2, and 51.2% of group 1. The c, d, e, f, right semi-Evans, and left semi-Evans parameters were not correlated with SOV dimensions (P = 0.955; P = 0.954; P = 0.335; P = 0.993; P = 0.330; P = 0.913; P > 0.05). No statistically significant difference was found between right SOV and right semi-Evans index (P > 0.05), left SOV and left semi-Evans index (P > 0.05) and between the right and left semi-Evans indexes of the control group (P = 0.797; P > 0.05).
Septum shift degree was statistically significantly higher in the ALV group compared to the control group (P = 0.010; P < 0.05). In the ALD group, 60 of 80 (75%) cases had septum shift compared to 77 of 83 (7.2%) cases in the control group. No statistically significant difference was found between the direction of the shift (right/left) (P = 0.681; P > 0.05).
The left semi-Evans index was higher in the cases with right septum shift (P = 0.001; P < 0.01), and the right semi-Evans index was higher in the cases with left septum shift (P = 0.001; P < 0.01).
Between the two readers, excellent agreement was observed for assessing the values of a, b, c, d, e, f, callosal angles, and SOV diameters (ICC varied between 0.853 and 0.997).
ALV is a common finding in the daily practice. It is reported as a normal variant unless a mass-occupying lesion obstructs one of the lateral ventricles or an ipsilateral neural parenchymal volume loss because of atrophy causes the one-sided/asymmetrical enlargement of the ventricle, which we call unilateral hydrocephalus. Besides being regarded as a normal variant, ALV has been linked to many diseases. There are some studies that link ALV and Parkinson's disease., Lewis et al. showed that the lateral ventricle volume was larger on the ipsilateral side with the Parkinsonism More Details symptoms. It is thought that periventricular parenchymal volume loss and atrophy on the disease side was responsible for the ventricular enlargement. ALV cases due to intraventricular lesions have been shared., These cases are in the minority, and most of the time ALV is reported as a normal variant. Kiroglu et al. showed that left lateral ventricle enlargement was twice as frequent as right ventricle enlargement. In our study, we found that in 40% of the cases, the right lateral ventricle was larger compared to 60% of cases with enlarged left lateral ventricle. We showed that ALV without other pathological findings on MRI is mostly caused by the off midline septum pellicidum. Seventy-five percent of the ALV cases had an off-midline septum pellicidum. We do not know whether the ventricular enlargement retracts the septum and causes the shift or the congenital midline shift causes the asymmetrical appearance. Toth et al. showed that ALV was a precursor of midline shift in trauma patients and noted that ventricles respond to unilateral pressure gradient before the midline shift. We had 20 ALV cases without midline shift. Half of them (50%, n = 10) had a right/left semi-Evans or Evans index higher than 30. On the contrary, only 26.25% of cases (n = 21) of the group with midline shift had right/left semi-Evans or Evans indexes higher than 30. We believe that ALV cases without septum shift but with Evans indexes greater than 30, should be further investigated or followed up. However, if the ALV is primarily caused by midline shift and not accompanied by a lesion, we believe it could be reported as a normal variant. Jamous et al. reported that indices such as Evans index and front occipital horn ratio are not useful in the presence of ALV. Some new indexes to assess the relation of hydrocephalus with ALV could be developed.
In the hydrocephalus group, only one of the cases had right semi-Evans index less than 30 compared to ten cases of left semi-Evans index. Right lateral ventricular enlargement seems to be more contributing to Evans index than left ventricle. Intracranial pressure increase might be affecting the right lateral ventricle initially.
Only two cases in the control group had right semi-Evans greater than 30, and one case had left semi-Evans greater than 30. These three patients were aged 72, 88, and 83 years, respectively. In those cases, the ventricular enlargement may be attributed to the atrophy of the aging brain.
The callosal angle has been used to differentiate iNPH from ventricular dilatation caused by atrophy. Ishii et al. showed that angles less than 90° are suggestive of iNPH. We found higher mean values (112.64 ± 21.60) of callosal angle for iNPH cases.
We found no correlation between SOVs and age/gender. As SOVs were previously assumed to be closely effected by hydrocephalus and ventricular enlargement, we investigated the relation between ALV/hydrocephalus and SOV. Brightbill et al. examined 176 SOVs and showed that the diameter ranged in size from 1.4 to 3.6 mm; the mean diameter was 2.16 mm. We investigated 256 SOVs and found lower mean diameters [Table 3]. In the study of Lirng et al., 69 patients were included and it was shown that the diameters of SOVs on both sides were the same in 68% of the patients, and the Pearson correlation coefficient was found to be 0.85 between the diameters of both sides. Brightbill et al. showed that there was no significant difference in the range of the SOV sizes between the right and left orbits, and only 18.2% of the SOVs were asymmetrical according to their asymmetry criteria, which was defined as 5 mm or greater difference in the diameters. In our study,the asymmetry was defined as 3 mm or greater difference in the diameters. We have also found no difference in the SOV diameter of the right and left orbit. Lirng et al. showed that patients with larger SOVs were more likely to have hydrocephalus, and the risk markedly increased for values higher than 2.5 mm. Whereas Brightbill et al. found no relationship in their relatively small study group consisting of 10 participants. In our study, we showed that SOV dimensions do not necessarily increase ipsilateral-to-ventricular enlargement and dimensions are not associated with a, b, c, d, e, f, callosal angles, and Evans index. We showed that SOV dimensions are not correlated with hydrocephalus or ALV.
Our study does have limitations. Hydrocephalus cases were not diagnosed by lumbar puncture, and and the diagnosis was mostly based on clinical findings, MRI findings, and follow-up data. Atrophy in the elderly population might have overestimated the Evans index; however, this limitation is a general limitation of the Evans index and hence a common problem. We only selected hydrocephalus cases from the database, and we do not know if there were cases that were assumed to be normal with right or left semi Evans index greater than 30 but with Evans index less than 30.
In conclusion, mild ALV is mostly associated with an off-midline septum. We believe no further follow-up is required for these patients. Although some overlaps in the ALV and hydrocephalus groups exist, the Evans index is still the most practical index for diagnosing hydrocephalus in the clinical setting. Bilateral SOV diameters should not be used to diagnose, grade, or follow up hydrocephalus.
Ethics approval and consent to participate
Ethics approval was obtained from the Maltepe University Ethics Committee. Signed consent was obtained from all participants.
Consent for publication
Images are entirely unidentifiable and there are no details on individuals reported within the manuscript.
Availability of data and materials
The datasets used and/or analyzed during the current study available from the corresponding author on reasonable request.
The authors declare that they have no competing interest.
Nuri Tasali revised the paper with permission.
GA designed the study, evaluated the MR images, wrote the manuscript, NBU collected the data, wrote the results.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
[Figure 1], [Figure 2], [Figure 3]
[Table 1], [Table 2], [Table 3]