Customize your JAMA Network experience by selecting one or more topics from the list below.
Cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy (CADASIL) and primary angiitis of the central nervous system (PACNS) share several clinical and radiological features. However, digital subtraction angiogram (DSA) is generally reported as normal in CADASIL, whereas lumen irregularities in distal cerebral arteries indicate PACNS.
To describe a potential pitfall of DSA interpretation, which led to the tentative diagnosis of PACNS in a CADASIL patient.
Patient and Methods
Single case observation.
A 47-year-old man sustained recurrent subcortical infarcts. He had mild hypercholesterolemia and migraine. His family history was unremarkable. The underlying cause of stroke could not be elucidated. Transcranial Doppler sonography revealed decreased intracranial blood flow velocities compatible with CADASIL. Lumen irregularities of several peripheral intracranial arteries were seen on DSA, which suggested PACNS. CADASIL was confirmed by results from skin biopsy and genetic testing.
First, in patients with CADASIL, DSA can show segmental lumen irregularities in distal cerebral arteries suggestive of PACNS. Second, the potential role of transcranial Doppler sonography to distinguish CADASIL from PACNS deserves further testing.
CEREBRAL AUTOSOMAL dominant arteriopathy with subcortical infarcts and leukoencephalopathy (CADASIL) is clinically characterized by recurrent subcortical infarcts (70%), migraine (38%) mostly with aura, psychiatric disturbances (30%), cognitive impairment (28%), and epilepsy (10%) in middle-aged patients.1 Despite autosomal dominant inheritance, family history may be unremarkable because of mild symptoms in other family members or because of spontaneous mutations.2 CADASIL shares several clinical symptoms and magnetic resonance imaging (MRI) features with primary angiitis of the central nervous system (PACNS).3 We present a patient with CADASIL whose clinical presentation, unremarkable family history, and results of cerebral digital subtraction angiogram (DSA) first suggested PACNS, whereas the results of transcranial Doppler sonography (TCD) were more indicative of CADASIL. Skin biopsy findings of granular osmiophilic material within the basal lamina surrounding vascular smooth muscle cells and genetic testing (presence of notch 3 mutation) eventually confirmed CADASIL.
Report of a case
A 47-year-old man had migraine attacks with prolonged aura since age 23 years. At the age of 32 years, he was hospitalized due to transient left-sided hemiparesis, episodes of depersonalization, and clumsiness of the left hand. These symptoms were interpreted as prolonged migraine aura. Ten years later, he developed right-sided sensorimotor hemiparesis. He recovered completely within 3 months. In the past 2 years, he developed progressive emotional instability with aggressive outbursts that were successfully treated with carbamazepine. Three weeks prior to hospital admission, the patient had sudden onset of vertigo and severe disturbance of balance.
Neurological examination findings revealed dysarthria and severe brachiofacial spastic hemiparesis with slight hemihypesthesia on the left side. Extensor plantar response was present on both sides. Perioral reflexes were brisk.
Findings from computed tomography (not shown) and MRI (Figure 1) revealed multiple chronic subcortical lesions bilaterally and an acute ischemic lesion in the right internal capsule. Findings from Doppler and duplex sonography of the extracranial arteries were normal; no atherosclerosis was recorded. Transesophageal echocardiography and 24-hour electrocardiograph monitoring revealed no abnormalities. There was no patent foramen ovale. Cerebrospinal fluid examination results showed a slightly elevated protein level (0.069 g/dL [reference, <0.048 g/dL]) but were otherwise normal. Laboratory testing did not indicate systemic vasculitis, coagulopathy, or sarcoidosis. Hematocrit levels were normal (41% [reference, 38%-52%]). The only vascular risk factor present was moderate hypercholesterolemia (266 mg/dL [6.9 mmol/L]). Cerebral DSA revealed segmental narrowing of several peripheral arterial branches within the territories of the anterior cerebral artery (ACA), middle cerebral artery (MCA), and posterior cerebral artery bilaterally and a mild dilatation of proximal portions of major cerebral arteries (Figure 2). Findings on DSA (most notably the peripheral irregularities) were considered indicative of PACNS. Accordingly, the patient was treated with prednisone (1 mg/kg body weight) and oral cyclophosphamide (150 mg/d). Findings from TCD revealed low mean flow velocities in the major intracranial arteries (Table 1), as recently reported in CADASIL patients.5 A skin biopsy was performed, which showed granular osmiophilic material within the basal lamina of vascular smooth muscle cells in small subcutaneous arterioles. Genetic testing (Laboratoire Cytogénétique, E. Tournier-Lasserve, MD, Hôpital Lariboisiere, Paris, France) revealed a mutation of the notch 3 gene with a nucleotide substitution (CGC→TGC) at position 475 on exon 4. CADASIL was diagnosed, and the immunosuppressive therapy was stopped. Within the next 3 months, the patient recovered only slowly with persistent severe left-sided hemiparesis.
Magnetic resonance imaging of subcortical infarcts and leukoencephalopathy in a patient with CADASIL (cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy). A representative T2-weighted magnetic resonance image demonstrates multiple areas of high-signal intensity in the periventricular white matter bilaterally.
Digital subtraction angiogram (DSA) of a patient with CADASIL (cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy). Selective DSA (oblique view, magnified) of the left internal carotid artery shows multifocal vessel narrowing in the peripheral anterior cerebral artery territories (small arrows). In addition, mild segmental arterial dilatation can be seen in the proximal middle cerebral artery (large arrow).
From his mid-thirties on, our patient had recurrent subcortical brain infarcts without evidence of an underlying atherothrombotic or embolic disease, of a coagulopathy, or of systemic vasculitis. Investigation results were negative until DSA showed focal lumen irregularities with multiple stenosis of distal intracranial arteries. In the clinical context of an unremarkable family history, as yet unexplained recurrent subcortical strokes, and (slightly) abnormal cerebrospinal fluid findings (increased protein), these DSA findings suggested PACNS.6 However, clinical signs and symptoms of PACNS are unspecific. In addition, angiographic findings in PACNS have a sensitivity of 80% and a specificity of only 30%,7 indicating a moderate diagnostic yield. Of interest, brain biopsy has a sensitivity of only 53% and a specificity of 87%.7 Despite the limited diagnostic utility of both methods, the diagnosis of PACNS in most cases essentially relies on angiographic or biopsy findings.7 Differential diagnosis of PACNS includes systemic vasculitis, central nervous system infection (eg, human immunodeficiency virus, syphilis), lymphoma, demyelinating disease, drug use (eg, cocaine or amphetamine), sarcoidosis, and rare vasculopathies such as fibromuscular dysplasia and moyamoya disease.6,7 However, none of these diseases was confirmed in our patient.
A recent case report3 illustrates the difficulties in distinguishing CADASIL from PACNS owing to similar clinical presentation and MRI appearance. Although it has been reported that most CADASIL patients have normal DSA results,8,9 this patient, like ours, showed multifocal segmental intracranial stenosis, which first suggested PACNS. A positive history of stroke at mid-adulthood in several first-degree relatives provided the clue to CADASIL in this patient,3 whereas our patient lacks a positive family history of stroke or dementia.
Concerning CADASIL and DSA findings in the literature, at an international CADASIL workshop it was reported8(pp705-706) that
cerebral angiography was performed in at least one affected member in all studied [ie, eight] families. It was essentially normal except in the family from . . . with dolichomega-arteries, and in one . . . with angiographic aspect suggestive of fibromuscular dysplasia. . . . they most likely were associated conditions.
In their article about the clinical spectrum of CADASIL, Chabriat et al9(p936) reported that 14 of 45 symptomatic family members underwent DSA. Findings were normal in 13 patients, while 1 patient had "a narrowing of some small branches of the anterior, middle, and posterior cerebral arteries." No further clinical data (eg, presence of migraine) on the patients with abnormal DSA findings were given.8,9
In our patient, TCD findings revealed decreased mean flow velocities in both MCA and ACA. These findings are in accordance with the results of a recent TCD study in CADASIL patients that focused on carbon dioxide reactivity but also reported in detail on mean flow velocities; the 29 CADASIL patients in the study showed significantly lower MCA mean flow velocities than the 29 age-matched controls.5 Low MCA mean flow velocities are unspecific, and primary causes include increasing age and high blood viscosity. Both factors were not applicable in our patient because (1) his mean blood flow velocities were low (ie, <2 SDs of the mean) compared with age-matched normal values, and (2) his hematocrit level was normal. Low mean flow velocities can be present in other forms of subcortical vascular encephalopathy, most notably in Binswanger disease, which, however, was not present in our patient. Of interest, Mohr10(p715) stated that "Binswanger's famous case, no less, may have been CADASIL," and that "CADASIL may join granulomatous arteritis of the central nervous system [PACNS] to suggest that the brain vasculature has biological properties somewhat different from other vascular beds."10(p715)
In contrast to CADASIL patients (with low MCA mean flow velocities), in all PACNS case studies that include TCD investigations, all 5 patients had increased MCA mean flow velocities.11-13 These observational data support the assumption that TCD might be helpful in choosing which further invasive diagnostic tests should be applied (ie, skin biopsy vs DSA) in patients with signs of a subcortical leukoencephalopathy of unclear origin. An algorithm for the distinction of both entities would have clinical implications. First, DSA encompasses an increased complication rate in CADASIL.14,15 Second, immunosuppressive therapy is indicated only in PACNS. However, functional TCD variables such as carbon dioxide reactivity were assessed neither in our study nor in the mentioned TCD reports in PACNS. Furthermore, this is only a single case observation. Therefore, a detailed discussion about a possible future role of TCD in CADASIL and PACNS is beyond the limitations of this case report and has to await further studies.
Our finding of low MCA and ACA blood flow velocities in CADASIL is in agreement with hemodynamic studies in CADASIL using MRI,16 positron emission tomography,17 or single-photon emission computed tomography.18 All these methods showed reduced cerebral blood flow and cerebral blood volume, especially in white matter. In addition, carbon dioxide–induced vasoreactivity was reduced in CADASIL patients.5 Low mean flow velocity of MCA and ACA as well as low cerebral blood flow, low cerebral blood volume, and impaired vasoreactivity all point to a dilatation of the major intracranial arteries in CADASIL. Vasodilatation of the proximal arterial segments might be interpreted as an attempt to compensate for the lumen reduction in distal arterial branches as visualized by DSA in our patient. Distal lumen reduction of small perforating arteries may reflect vascular smooth muscle cell dysfunction19 or vessel destruction and replacement by extracellular matrix with superimposed subendothelial fibrous proliferation.19,20 The fact that histological alterations in CADASIL are predominantly found in smaller arteries and arterioles19 might explain our DSA result of stenosis involving the distal arterial branches rather than the major cerebral arteries. However, it remains unclear why, in general, DSA findings in CADASIL have been reported to be normal.
Accepted for publication February 13, 2002.
Author contributions:Study concept and design (Drs Engelter, Steck, and Lyrer); acquisition of data (Drs Engelter, Rueegg, Kirsch, Fluri, and Lyrer); analysis and interpretation of data (Drs Engelter, Rueegg, Kirsch, Probst, and Lyrer); drafting of the manuscript (Drs Engelter and Rueegg); critical revision of the manuscript for important intellectual content (Drs Rueegg, Kirsch, Fluri, Probst, Steck, and Lyrer); obtained funding (Drs Engelter and Lyrer); administrative, technical, and material support (Drs Kirsch and Probst); study supervision (Drs Steck and Lyrer).
The study was supported by the Basel Hirnschlag [Stroke]-Fonds, Basel.
We thank Susanna Papa for helpful language assistance.
There are no conflicts of interest involved in the study.
Corresponding author and reprints: Stefan T. Engelter, MD, Neurological Clinic and Stroke Unit, University Hospital Basle, Petersgraben 4, CH–4031 Basel, Switzerland (e-mail: email@example.com).
Engelter ST, Rueegg S, Kirsch EC, et al. CADASIL Mimicking Primary Angiitis of the Central Nervous System. Arch Neurol. 2002;59(9):1480–1483. doi:10.1001/archneur.59.9.1480
Monkeypox Resource Center