CNS small vessel vasculitis: Distinct MRI features and histopathological correlation
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/0028-3886.217929
Source of Support: None, Conflict of Interest: None
Background: Central nervous system (CNS) vasculitis is an uncommon disease, which is a diagnostic and therapeutic challenge for physicians. Large and medium vessel vasculitis is relatively easy to diagnose by angiogram compared to small vessel vasculitis, where angiograms are often normal; imaging features described till date are sensitive but not specific.
Keywords: Magnetic resonance imaging, primary CNS vasculitis, small vessel vasculitis, susceptibility weighted imaging
Central nervous system (CNS) vasculitis is an uncommon condition which is difficult to diagnose on clinical grounds alone. Even with a strong clinical suspicion of vasculitis, only 12% cases have histopathological confirmation during life. The vasculitides are broadly classified according to vessel size into large, medium, small and medium, and small vessel vasculitis. Diagnosis of large and medium vessel vasculitis is possible by angiogram, where histopathological assistance is not required for patient management. However, in small vessel vasculitis, angiogram is not contributory. Magnetic resonance (MR) imaging, though sensitive, lacks specificity.
Susceptibility weighted imaging (SWI) is a gradient echo sequence, which utilizes magnetic susceptibility properties of blood, iron, and calcium. This sequence is used in functional brain imaging, and in detecting vascular diseases such as arteriovenous malformations (AVMs), developmental venous anomalies, microbleeds, hemorrhagic and ischemic infarcts, and nonvascular diseases such as multiple sclerosis (MS), trauma, and degenerative brain diseases. Its ability to detect paramagnetic substances in free radicals of phagocyte cells in pyogenic abscess wall has been recently described.
We hypothesized that the combination of SWI and post-contrast T1-weighted thin MR images with distinct MR imaging features are specific for small vessel vasculitis.
Medical records of clinically suspected vasculitis patients between January 2008 to July 2012 were retrospectively reviewed for patient demographic details such as age, sex, and clinical presentation. The inclusion criteria for the study were (1) a clinical history highly suspicious of CNS vasculitis, (2) availability of MR imaging, in particular SWI and contrast-enhanced T1-weighted, (3) at least one modality of angiogram [MR angiogram (MRA)/computed tomography (CT) angiogram/digital subtraction angiogram (DSA)] interpreted as normal, and (4) a positive histopathology. One hundred and seventeen patients had clinical suspicion of vasculitis. Of these, 17 patients underwent meningocortical biopsy and 7 patients had histopathologically proven vasculitis. Only five patients satisfied the study criteria and were included for final analysis.
MR imaging of the brain was performed using a 12-channel matrix coil on a 1.5T clinical scanner (Avanto, SQ Engine; Siemens, Erlangen, Germany). The following sequences were analyzed: axial T1-, T2-, SWI, three-dimensional time of flight (TOF) MRA of the circle of Willis and post-contrast T1-weighted images. The SWI sequence parameters were as follows: TR (repetition time), 48 ms; TE (echo time), 40 ms; flip angle, 20°; bandwidth, 80 kHz; slice thickness, 2 mm, with 56 slices in a single slab; matrix size, 512 × 256, iPAT factor-2, and acquisition time 2.58 minutes. In this left-handed MRI system, generally, hyperintensity in the phase image is indicative of paramagnetic substances such as blood products, and hypointensity is indicative of diamagnetic substances such as calcification.
Pre and post-contrast conventional spin echo T1-weighted imaging was acquired, with slice thickness, 5 mm; TR (repetition time), 468 ms; TE (echo time), 11 ms; interslice thickness of 6.5 mm. Contrast-enhanced MRI was acquired in axial, sagittal, and coronal planes. The time of flight (TOF) technique was used to obtain the magnetic resonance angiogram (MRA).
Of the 5 patients, 3 were males and 2 females, with a mean age of 34.2 (range: 18–62 years). The total duration of neurological illness varied from 10 months to 12 years. Three patients (cases 1–3) with lymphocytic vasculitis altogether had 43 discrete T2/FLAIR hyperintense brain lesions [Table 1]. All the lesions revealed central areas of SWI blooming in a linear and a lace-like pattern and linear and lace-like contrast enhancement at corresponding locations [Figure 1] and [Figure 2]. Two patients had gangliocapsular region involvement. One-year follow-up imaging revealed no progression of lesions in one (case 3), while in other patients (cases 1 and 2), old lesions disappeared and a few new lesions appeared. The MR imaging patterns of the new lesions were similar to the ones described above.
One patient (case 4) with tuberculous vasculitis had coarse SWI blooming foci in the left parieto-occipital lobe and corresponding hyperintensity in phase images, which was suggestive of foci of bleed [Figure 3]. Contrast-enhanced MRI revealed multiple, intensely enhancing, conglomerated, thick ring, lesions at the corresponding locations. Focal cortical involvement was noted in all 4 positive cases. Case 5 had hypertrophic pachymeningitis with meningocortical biopsy proven small vessel vasculitis; however, the cerebral parenchyma was normal on MRI.
Vasculitis is a pathological term, and unless there is direct involvement of vessel wall demonstrated on imaging, some authors believe radiological diagnosis of this entity is inappropriate. Conventional MR imaging signs of small vessel vasculitis described till date are nonspecific, and direct demonstration of small vessel involvement is seldom investigated.
SWI blooming is seen in conditions with an active inflammatory process (in abscess wall) and is believed to be secondary to the presence of paramagnetic free radicals. The linear and lace-like SWI blooming in cases 1–3 were likely secondary to paramagnetic free radicals from leukocytic infiltrates [Figure 1] and [Figure 2]. The involved vessels were coursing perpendicular to the ventricular margins. Margins of these blooming foci were irregular against the regular pattern with smooth margins of normal transmedullary veins. Few vessel areas showed a beaded appearance [Figure 2]. The third patient had two additional coarse blooming artefacts along the subcortical areas, possibly secondary to bleed. The histopathology of all three patients revealed intense perivascular lymphocytic infiltrate in the Virchow–Robin spaces and adjacent focal parenchymal infiltrate. The corresponding areas showed lace-like contrast enhancement possibly secondary to contrast leak. All the 3 patients on histopathalogical examination had a lymphocytic type of angiitis.
SWI blooming along the course of a parenchymal vessel itself is less specific because it can be seen in MS, white matter ischemic changes, and stroke. In addition, large calibre deep veins can also show a linear enhancement pattern. Hence, to increase specificity for MRI findings of small vessel vasculitis, we examined the following: the lesion should show (1) linear/lace-like SWI blooming with irregular margins, (2) surrounded by FLAIR hyperintensity, and (3) linear/lace/central dot-like enhancement at areas of abnormal SWI blooming. All the 43 lesions were satisfying the abovementioned criteria. To the best of our knowledge, no specific MR findings have been described previously to diagnose CNS small vessel vasculitis.
In case 4, MR imaging revealed multiple coarse granular SWI blooming pattern. Histopathology from these areas revealed hemorrhagic foci with vessel wall destruction. The etiologies of such bleeds are presumed to be secondary to direct vessel damage or endothelial damage secondary to ischemia. The enhancement pattern was also that of conglomerated ring-like appearance. Although hemorrhages are described in 10–12.5% patients with CNS vasculitis,, this finding is less specific for small vessel vasculitis because they can be seen in many nonvasculitic pathologies. Case 5 had hypertrophic pachymeningitis with normal appearing cerebral parenchyma.
Our study has a few limitations. First, the numbers of histopathologically proven vasculitis cases were very less. Second, the site of biopsy was the nondominant temporal lobe in cases 1–3, irrespective of the site of the MR abnormality; hence, the location of imaging abnormalities and biopsy site were different. Under ideal circumstances, biopsy should be obtained from MRI detected abnormal areas, which in clinical practice is not always feasible. In cases 1–3, distinct SWI and contrast-enhanced MRI findings showed the entire cerebral hemisphere involvement, explaining the positive histology, irrespective of the site of biopsy. Third, the typical findings were noted only in lymphocytic type of vasculitis.
The linear/lace-like SWI blooming with irregular margins, surrounded by FLAIR hyperintensity and linear/lace-like/central dot-like enhancement at areas of abnormal SWI blooming are distinct MR imaging features, which are suggestive of lymphocytic CNS small vessel vasculitis. The ability of SWI to detect free radicals and bleed can provide further insights into the pathophysiological changes occurring in various vasculitis. Further large-scale studies with targeted biopsies are required to prove this.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
[Figure 1], [Figure 2], [Figure 3]