The Correlation Between Ventricular Diameter Measured by Transcranial Sonography and Clinical Disability and Cognitive Dysfunction in Patients With Multiple Sclerosis FREE

THE PRESENCE of cerebral atrophy is well described in multiple sclerosis (MS).¹ In contrast to the prevailing concept that brain atrophy occurs only in late stages of MS, a recent study² showed that significant cerebral axonal damage is already present in early stages of the disease. In addition, neuroimaging studies^{3 - 5} showed a clear correlation between cerebral atrophy determined with magnetic resonance imaging (MRI) and disability, although the correlation between the number of focal lesions and disability measures was weak.⁶ Therefore, brain atrophy is a potential surrogate marker of disease progression because it reflects the net effect of tissue-destructive processes, and measuring brain atrophy circumvents some of the problems with lesion measurements.

Magnetic resonance imaging is an advantageous tool for the determination of brain atrophy. Although highly reproducible measurements of brain atrophy were obtained using BPF (brain parenchymal fraction),^{7 - 8} measurements of the ventricular volume³ and analysis of the third ventricular area⁹ or third and lateral ventricular diameter¹⁰ were also used for the determination of brain atrophy. In addition to MRI, the ventricular system can be accurately evaluated using transcranial sonography (TCS), a noninvasive neuroimaging method that is easily applicable and cost-effective.^{11 - 12} In this study, we determined the following: (1) the ventricular diameters in subjects with MS and age- and sex-matched healthy control subjects; (2) the association between the diameter of the ventricular system obtained by MRI and TCS and disability, cognitive performance, and mood in the subjects with MS; and (3) the relationship between MRI and TCS data concerning the enlargement of the ventricular system.

After receiving informed consent, we included 74 patients (48 women and 26 men, with a mean age of 42 years [range, 18-70 years]) with clinically definite MS in the present study. Relapsing-remitting MS was diagnosed in 34 subjects, secondary-progressive MS in 32, and primary-progressive MS in 8. Disability was assessed according to the Expanded Disability Status Scale (EDSS)¹³ by a neurologist unaware of the MRI and TCS data (P.R.). The median EDSS score was 5.5 (range, 1.0-8.0), and the mean duration of the disease was 9.4 years (range, 0.2-35.0 years). All subjects with MS were individually matched with a healthy control subject (48 women and 26 men, with a mean age of 42 years [range, 20-69 years]).

For neuropsychological assessment, we administered the Brief Repeatable Battery of Neuropsychological Tests for MS,¹⁴ which includes the following: Selective Reminding Test (long-term storage, consistent long-term retrieval, and delayed recall), Spatial Recall Test (total and delayed recall), Symbol Digit Modalities Test, Paced Auditory Serial Addition Test (2- and 3-second versions), and the Word List Generation Test. The actual severity of depressive episodes was assessed by the Hamilton Depression Rating Scale and the Montgomery-Åsberg Depression Rating Scale as standardized measures of mood disorders.¹⁵ Furthermore, all subjects were asked to complete the Befindlichkeitsskala test and Beck Depression Inventory (BDI).¹⁶ The neuropsychological and psychiatric assessment was performed by an investigator not aware of the clinical or MRI data (D.B.).

All subjects underwent a standardized TCS protocol. For TCS examination, we used a color-coded, phased-array sonographic system equipped with a 2.5-MHz transducer. Supratentorial and infratentorial brain areas were examined through a preauricular acoustic bone window. The ventricular system can be depicted as anechoic, limited by hyperechoic borders.¹² In subjects with MS, MRI was performed by using a standard head coil on a 1.5-T imager. Proton density T2-weighted (repetition time, 2000 milliseconds; echo time, 30,100 milliseconds) and T1-weighted (repetition time, 460 milliseconds; echo time, 19 milliseconds) sequences were administered covering the whole brain (7-mm slice thickness and 192 × 256 matrix, with an axial orientation and angulation according to the bicommissural line). The T1-weighted images were used for measurements.

On MRI and TCS images, the diameters of the third ventricle and the frontal horns were measured by 2 independent investigators who were blinded to the clinical assessment of the subjects (M.M. and G.B.). The diameter of the third ventricle was determined by the maximum transverse diameter on axial scans. The extension of the frontal horns was ascertained by measuring the maximum distance between the septum pellucidum and the lateral tip of the frontal horn perpendicular to the septum. For TCS, a slight tilting was necessary to obtain an optimal slice of the frontal horns, whereas MRI measurement of the frontal horn was done on axial scans.

Statistical comparison between subjects with MS and control subjects was performed using the t test. Comparisons between clinical data and ventricular diameters were performed using the nonparametric Spearman rank correlation test. According to the guidelines of Fleiss and Shrout,¹⁷ interrater reliability was assessed by calculating intraclass correlation coefficients (random-effects model).

In the results of TCS, the subjects with MS had larger ventricular diameters than the control subjects. The differences between the 2 groups by matched-pair analysis were highly significant (Figure 1), and the mean diameters of the third ventricle and the lateral ventricles of the control group were in accordance with normal values published recently.¹⁸

Table 1 presents the mean ± SD ventricular diameter of subjects with MS determined by MRI and TCS by 2 independent examiners (M.M. and G.B.). The measurements revealed a significant interrater reliability. Furthermore, the ventricular diameters obtained by TCS and MRI were significantly correlated with MS (Spearman rank correlation, r = 0.9 for the third ventricle and r = 0.7 for both frontal horns; P>.01) (Figure 2).

In the group of subjects with MS, we also detected a significant correlation between the diameter of the third ventricle obtained by TCS or MRI and disability measured by the EDSS (Spearman rank correlation, r = 0.4; P<.01). The correlation between the diameter of the frontal horns and disability was substantially lower for both neuroimaging techniques. In addition, the results of a majority of neuropsychological tests correlated significantly with TCS and MRI data, with the strongest correlation observed with the diameter of the third ventricle. No correlation was found between the diameter of the ventricles and depression scales (Table 2).

Our study demonstrated that the ventricular diameter correlates with clinical disability and with the results of neuropsychological assessments in subjects with MS, although no information about the total volume of the ventricles or cortical atrophy could be obtained. Our results agree with those of previous cross-sectional studies,^{19 - 20} which found a positive correlation between cognitive decline and cerebral atrophy in MS, suggesting that these simple ventricular parameters are sufficient for correlation studies. However, a longitudinal study to determine the value of TCS in measuring change in ventricular diameter is still needed. We recently started to monitor prospectively the ventricular diameter in a cohort of subjects with MS. Because of the strong correlation between measurements of ventricular diameter and MRI and TCS results, we also expect to demonstrate changes in atrophy during a short time with TCS.

Quantitative histopathologic examinations revealed that axonal loss is present in active inflammatory lesions early in the course of MS,²¹ and it seems likely that axonal loss is a major contributor to clinical disability and results in cerebral atrophy in subjects with MS. Therefore, it is conceivable that progressive loss of brain and spinal cord tissue starts at disease onset. This possibility was supported by Simon et al,¹⁰ who reported increasing brain ventricular size and decreasing corpus callosum area and brain width during 1- to 2-year intervals in patients with relapsing-remitting MS. By using BPF, Rudick et al⁸ demonstrated that although it was early in the disease course, patients with MS had a mean BPF more than 5 SDs below the mean of healthy age- and sex-matched controls. In addition, it was demonstrated that treatment with interferon beta-1a resulted in a reduction of brain atrophy progression during the second year of the clinical trial. Therefore, measuring brain atrophy with MRI is a surrogate measure of the global pathologic process in patients with MS since it is informative in demonstrating changes over time, shows the relationship with disease progression, and is capable of demonstrating therapeutic effects. However, although sophisticated MRI procedures are scientifically valuable for the determination of brain atrophy, it is doubtful that they will be administered beyond therapeutic trials in the routine follow-up of patients with MS because of their high cost.

The TCS data were strongly correlated with the MRI data of the ventricular system, which is not surprising because modern TCS systems, like most MRI systems, have a 0.7-mm spatial resolution in the focus zone.¹² We found that the strongest correlation was between the level of disability and the diameter of the third ventricle, a finding that also agrees with findings in previous evaluations.^{9 - 10 ,20} It is possible that the third ventricle is a better indicator of cognitive dysfunction since it more closely reflects the periventricular pathologic changes in MS and the disruption of white matter fiber tracts that interconnect prefrontal limbic structures.¹⁸ However, the difference could be due to technical limitations. Increases in the diameter of the third ventricle may be more accurately determined because it is slitlike, whereas the lateral ventricles have more complex irregular anatomy.

According to the recent MRI studies and our results, we believe that measurement of ventricular diameter using TCS could serve as a surrogate marker for brain atrophy in the routine follow-up of patients with MS. Furthermore, evaluation of putative treatments aimed at preventing disability could be followed up more easily using this method. In contrast to other fields of TCS, identification and measurement of the third ventricle is fast and precise and can be delegated to a technician. It is reasonable to incorporate measurement of the diameter of the third ventricle as part of future composite indexes for determining graduating disease severity in addition to the EDSS, which is acknowledged to belimited in its responsiveness to change in neurological status.²²

Accepted for publication December 14, 1999.

We thank Mira Schließer for her excellent technical assistance.

Corresponding author: Peter Rieckmann, MD, Department of Neurology, Bayerische Julius-Maximilians-Universität, Josef-Schneider-Straße 11, 97080 Würzburg, Germany (e-mail: p.rieckmann@mail.uni-wuerzburg.de).