CADASIL: MRI may be normal in the fourth decade of life – a case report

Abstract

Background

Cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy (CADASIL) manifests by migraine with aura, cerebral ischemic events, mood disturbances and dementia. Brain MRI lesions typically precede the symptoms from 10 to 15 years and previous evidence showed all CADASIL patients above 35 years old have an abnormal MRI, supporting the clinical diagnosis.

Case results

We present a 37-year-old female patient with migraine without aura, a family history of CADASIL, normal brain 3-Tesla MRI and normal skin biopsy, even though a pathogenic NOTCH3 gene mutation (allele 2, exon 11, c.1672 C\gtT, p.Arg558Cys) was detected.

Conclusions

When CADASIL is strongly suspected, a normal brain MRI, even in the fourth decade of life, does not rule out the diagnosis and should not discourage the genetic test.

Introduction

Cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy (CADASIL) is an autosomal dominant vasculopathy resulting from mutations in the NOTCH3 gene, located on chromosome 19 and involved in the differentiation and maturation of small arteries’ smooth muscle cells (1). The natural history of the disease is classically described as a succession of clinical manifestations: migraine with aura starting at around 30 years old and ischemic events and mood disturbances between 40 and 60 years old, followed by subcortical dementia between 50 and 60 years old.

The brain MRI features typically precede the symptoms by 10–15 years and are characterized by the presence of T2 hyperintense lesions in the anterior temporal poles, external capsules and subcortical frontal white matter, although other areas may be affected (basal ganglia, thalami, corpus callosum and brainstem). These lesions can appear at a variable age, but evidence up until the present led to the assumption that all CADASIL patients above 35 years old present with an abnormal MRI (2), supporting the clinical diagnosis. Over time, there is an increasing extension of white matter lesions, as well as the appearance of lacunar infarcts and microbleeds (3). As CADASIL is a systemic arteriopathy, skin biopsy can also help in the diagnosis. It can show disruption of the media with deposits of granular osmiophilic material (GOM), vessel wall thickening and lumen stenosis (4).

We present a case of a 37–year-old CADASIL patient with normal brain MRI.

Case report

In 2010, a 33-year-old Caucasian female was referred to our neurology outpatient clinic as she had a positive family history of CADASIL: her mother was already demented and bedridden due to multiple strokes since her 50s, with no history of migraine (only episodic tension-type headaches), and her sister had migraine with visual aura and had already experienced two episodes of transient neurological deficits since her 20s, characterized by dysarthria and deviation of the labial commissure. A pathogenic NOTCH3 gene mutation had been identified in both. The genogram is represented in Figure 1. She had no other relevant past medical history.

Figure 1.

Genogram of our patient (arrow), for details please see text. yo: years old; PMDD: psychomotor development delay; TNA: transient neurological accident; ?: genetic test not performed.

Our patient had complained of headaches since she was 12 years old, with menstrual predominance, lasting for several hours, usually being bi-temporal but sometimes more severe with lateralization, pulsatility and photo-phonophobia, but without accompanying nausea/vomiting nor visual/sensitive/motor symptoms. She also reported some memory loss, but without interference in her daily life activities. There was no history of psychiatric disorders, stroke, seizures and other transient neurological symptoms or signs. Her neurological examination was completely normal.

An initial brain 3-Tesla MRI scan was performed and showed no sign or morphological abnormalities of the parenchyma, particularly in the temporal poles and external capsules. She was also submitted to a skin biopsy, revealing normal morphology of the small arterial walls and a regular, well-defined internal elastic lamina; GOM was also unapparent in electronic microscopy (Figure 2). However, a genetic study was conducted searching for mutations by PCR and sequencing of exons 3, 4, 7, 8, 11, 12, 18, 19, 20, 21, 22 and 23 of the NOTCH3 gene, where most of these mutations have been found in Portuguese patients. This study was positive for the same pathogenic mutation of the family in the NOTCH3 gene (allele 2), consisting of a C to T exchange in exon 11 at position 1672, which results in an amino acid change of the protein Arg558Cys (5). This result was confirmed in a second independent blood sample.

Figure 2.

Brain MRI and skin biopsy performed when patient was 37 and 35 years old, respectively. In MRI, notice the absence of white-matter lesions in the external capsules (panel A – axial T1 on the right and axial T2/FLAIR on the left), temporal poles (panel B – sagittal T1 on the right, axial T1 in the middle and axial T2/FLAIR on the left) and superior frontal lobe (panel C – axial T1). In skin biopsy, notice the absence of the characteristic granular osmiophilic material (GOM) in the media of the small arteries (panel D – electronic microscopy 4000X).

Four years later, in 2014, the patient was 37 years old and remained clinically stable. A 3-Tesla MRI scan was repeated and again presented as normal (Figure 2). She is now treated with analgesics in migraine attacks and acetylsalicylic acid 100 mg daily; vascular risk factors are also periodically accessed.

A review of the patient’s mother’s MRI (when she was 53 years old) revealed the presence of multiple cortical and subcortical ischemic lesions, along with petechial hemorrhages; the sister’s brain MRI, performed when she was 25 years old, was normal.

Discussion

Our patient fulfills the International Classification of Headache Disorders, 3rd edition (beta version) criteria for migraine without aura (6). She has a clear family history of CADASIL and she carries a pathogenic NOTCH3 gene mutation. However, her brain MRI (3-Tesla) was completely normal at an age at which some abnormalities should already be present in a CADASIL patient. Searching for similar cases in the literature, we found four relevant patients. One had a 3–year history of migraine with accompanying right hemiparesis and no family history; a NOTCH3 gene mutation was identified, but the brain MRI was unremarkable. However, she was only 14 years old (7). Another reported patient was a 34-year-old woman suffering from migraine with prolonged motor aura, without abnormalities in the brain MRI performed in 1998. At that time, sporadic hemiplegic migraine was suspected, and 10 years later, as the genetic diagnosis of CADASIL had been made in a sister, MRI was repeated and showed characteristic abnormalities. The same NOTCH3 gene mutation was then found in the patient (8). The other two CADASIL patients that were reported had migraine without aura and normal brain MRIs at 27 and 30 years old (9).

Several interesting questions are raised by our case. First, is it possible that the migraine without aura is not part of the disease, but rather a coincidental phenomenon (as it is a quite frequent form of headache in the general population), so that the patient would be still asymptomatic and only a CADASIL mutation carrier? It is known that the penetrance of the disease is virtually 100%, but expression varies in age of onset, severity of the clinical symptoms and progression of the disease (10). As such, we cannot rule out this hypothesis, but the normality of the brain MRI would still be intriguing, as the imaging abnormalities precede symptoms by more than a decade and the patient is already 37 years old. We may think that she would be similar to her mother, who, although without migraine, only developed the stroke and dementia CADASIL manifestations in her 50s, but on the other hand, there is the sister who started with migraine with aura, a clearly different phenotype in the same family.

As we said, skin biopsy was also negative in our patient. It is known that both GOM accumulation and the degeneration of vascular smooth muscle cells may be detectable early in the disease course, even in asymptomatic CADASIL subjects (11). However, a negative result does not absolutely exclude CADASIL since the abnormalities may be too focal or the biopsy might not have been deep enough (12). In fact, the sensitivity of skin biopsy is approximately 50%, being increased with the immunocytochemistry for the detection of the NOTCH3 protein (85–95%) (1), which was not performed in our patient.

Clinical implications

When CADASIL is strongly suspected, a normal brain MRI, even in the fourth decade of life, does not rule out the diagnosis and should not discourage genetic testing. This should always be performed as part of a multidisciplinary approach involving the neurologist, geneticist and psychologist.