In the present study, we evaluated DM1 patients using the newly proposed z-Evans Index. Although the original Evans Index has been adopted as a criterion for discriminating the tap-test response[19], the z-Evans Index has been shown to have a higher diagnostic accuracy[12]. We found that the z-Evans Index of DM1 was significantly higher than that of control patients. Furthermore, the z-Evans Index was significantly associated with the patients’ age. These results were consistent with those of previous studies showing that cerebral ventriculomegaly was common in DM1 patients and appeared to be progressive[9].
It is important to note that an increased z-Evans Index (ventriculomegaly) was independently associated with the callosal angle and Sylvian fissure dilation. The callosal angle is an imaging biomarker of NPH[9]. A previous NPH study showed that a small callosal angle on preoperative MRI was a significant predictor of a positive shunt outcome[13]. Another study showed a significantly decreased mean callosal angle in patients with NPH (109 ± 9) compared with patients with Alzheimer’s disease (AD) (135.4 ± 11.3) or dementia with Lewy bodies (DLB) (136.9 ± 8.2)[20]. The mean callosal angle (113 ± 26.4°) in our elderly patients (Group C) (Table 3), which might be close to that in NPH, was smaller than that reported in AD and DLB cases[20]. These results suggest that the morphologic changes accompanied by ventriculomegaly in DM1 patients are similar to those in patients with NPH. Furthermore, we found that DESH was observed in 20.6% of elderly patients with DM1 in our study. Previously, five cases with NPH in association with DM1 have been reported[10]. Three of those five patients received shunt surgery, and their condition markedly improved. These cases may support our results, as all of the cases were more than 50 years old, suggesting that DM1 patients may develop an NPH pathology later in life. Although cognitive deficits were found in a high percentage of DM1 patients at a late stage[6,21,22], the exact mechanism remains unclear. Therefore, further prospective studies with a larger number of DM1 patients are needed to determine whether or not an NPH-like appearance is associated with neuropsychological deficits.
The mechanism underlying the manifestation of NPH may be the obliteration of arachnoid villi (small protrusions of the arachnoid mater that return the cerebrospinal fluid [CSF] to the venous circulation), leading to disrupted CSF reabsorption and subsequent ventriculomegaly[23,24]. However, the etiology and pathophysiology of the NPH-like appearance in DM1 are uncertain. One possible mechanism suspected by previous investigators involves widespread cell membrane defects caused by genetic abnormalities in DM1[10,25], leading to arachnoid granulations[26]. Another possible mechanism may be glymphatic system dysfunction in DM1. The glymphatic system is an effective waste clearance pathway that removes metabolic wastes and neurotoxins from the brain along paravascular channels[27]. The anatomical pathways for the glymphatic system are the perivascular spaces. While the perivascular spaces terminate within the brain parenchyma, paravascular CSF can continue traveling along the basement membranes surrounding arterial vascular smooth muscle[28]. In a previous neuropathological study, DM1 brain showed EPVS in the basal forebrain and deep white matter along with a loss of smooth muscles in small to medium-sized arteries[29]. A previous MRI study also showed that EPVS were significantly more frequent in DM1 patients (67%) than in controls (27%)[7]. Ringstad et al. demonstrated the reduced glymphatic clearance in idiopathic NPH using glymphatic MRI and hypothesized that a reduced glymphatic function is instrumental for dementia in NPH[30]. Based on these previous findings, we suspect that the EPVS and NPH-like appearance in DM1 might reflect a pathology of glymphatic system dysfunction.
Costanzo et al. showed a significant negative correlation between the degree of EPVS and disease duration and suggested that EPVS might progressively disappear during the disease course[7]. Our results regarding the patient age also showed that the presence of CS-EPVS was more common in Group A than in Groups B and C, although there were no significant differences. Furthermore, regarding the frequency of temporo-polar WMLs, we found no significant differences among the age groups. These results suggest that CS-EPVS and/or temporo-polar WMLs may be findings at the early stage of DM1. Most of the elderly patients (Group C) with ventriculomegaly showed Sylvian fissure dilation, whereas only half of the younger patients showed this finding (Group A). In addition, we found that all patients with high-convexity tightness had ventriculomegaly as well as Sylvian fissure dilation. These results suggest that ventriculomegaly might precede the development of Sylvian fissure dilation and high-convexity tightness in DM1.
In DM1, the classic disease range of CTG repeat numbers is 50-4000, with repeat sizes of 50–80 associated with mild clinical phenotypes, while large repeat expansions of up to 4000 are often found in severe, mostly congenital forms of the disorder[31]. Brain abnormalities, including issues with cognitive function and muscle weakness, have been linked to the CTG repeat numbers, cognitive function, and muscle weaknes[32,31,33]. However, the present study showed no correlation between the z-Evans Index and the CTG repeat numbers. Although the reason for this finding remains unclear, one possible reason is the small sample size of elderly patients with large repeat expansions because of their poor prognosis. Furthermore, our negative result may be supported by the findings of previous studies; for example, Minnerop et al. noted no correlations between neuropsychological test results and CTG repeat numbers[31]. With regard to neuroimaging assessments, temporo-polar WMLs have shown no correlation with CTG repeat numbers[7].
This study was limited by its retrospective design. Thus, neuropsychological tests were not performed in many cases, as neuropsychological deficits in cases of myotonic dystrophy can escape commonly applied neuropsychological test batteries for as-yet-unclear reasons. Moreover, no patients undergo lumbar puncture and were clinically diagnosed as having NPH. Second, all subjects were examined by 1.5-T MRI rather than 3-T MRI. However, we believe that the image quality with our MRI protocols was sufficient for the evaluation of the proposed method. We arbitrarily defined ventriculomegaly as a z-Evans index of > 0.3 because an NPH-like appearance in DM1 patients is a relatively new finding.