Книги по МРТ КТ на английском языке / MR Imaging in White Matter Diseases of the Brain and Spinal Cord - K Sartor Massimo Filippi Nicola De Stefano Vincent Dou

336

Spinal Cord

Three variants can be recognized: (1) acute myelitis indistinguishable from acute transverse myelitis (Manabe et al. 2000; de Seze et al. 2001); (2) chronic myelitis similar to that observed in chronic MS (de Seze et al. 2001); and (3) tractopathy reflecting peripheral nervous involvement (Mori et al. 2001). The latter is the consequence of sensory neuronopathy and consists of diffuse hyperintensity of the posterior columns in T2*-weighted images of the spinal cord, better appreciated in the cervical segment.Also, this MRI finding is nonspecific and can be observed in other diseases causing chronic deafferentation such as Friedreich’s ataxia (Mascalchi et al. 1994).

Interestingly, in many cases of primary SS the spinal cord damage is combined with optic nerve involvement (de Seze et al. 2001; Mochizuchi et al. 2000) featuring a neuromyelitis optica (Devic) syndrome as in other systemic vasculitides such as SLE.

22.2.1.3 Polyarteritis Nodosa

Polyarteritis nodosa (PAN) is a rare disease that affects middle-aged subjects and is characterised by focal, segmental necrotising vasculitis of small and mediumsized arteries. The nervous system is usually involved in the form of mononeuritis multiplex, but ischaemic or haemorrhagic strokes and subarachnoid haemorrhages secondary to rupture of aneurysms, which more commonly involve the renal,hepatic and visceral

M. Mascalchi and F. Salvi

arteries, occur in 20–40% of patients, usually after the initial diagnosis is made (Reichhart et al. 2000)

Multifocal white matter lesions were observed in 18% of patients with PAN reviewed by Reichhart et al. (2000) (Fig. 22.6). No case of optic nerve or spinal cord damage has been reported so far.

Brain

Small infarcts in the deep cerebral hemisphere or brainstem related to thrombotic microangiopathy along penetrating arteries, rather than to vasculitis, are the most common MRI features besides white matter damage in PAN (Provenzale and Allen 1996; Reichhart et al. 2000).

Large infarcts and intra-parenchymal haemorrhages are less common (Provenzale and Allen 1996; Reichhart et al. 2000; De Reuck 2003). On conventional MRI, infarcts show the usual appearance of well-defined areas of increased signal in T2weighted images, with a delay of several hours from the clinical onset (Fig. 22.6).Lesion contrast enhancement reflecting damage of the blood–brain barrier can be observed a few days after onset.

22.2.2

Antiphospholipid Antibody Syndrome

Antiphospholipid antibody syndrome (APS) is characterised by arterial or venous thrombosis and pregnant morbidity in the presence of anti-

cardiolipin antibodies and/or lupus anticoagulant. Thrombocytopenia may be part of the syndrome (Harris 1986; Trent et al. 1997; Funauchi et al. 1997). APS can occur as either a primary disorder or be secondary to a connective tissue disease, most frequently SLE (Provenzale et al. 1996; Sanna et al. 2003).

APS is considered among the immunomediated non-MS diseases of the CNS but is not a vasculitis. In fact neuropathological examination indicates that the pathogenesis of the cerebral vasculopathy responsible for the cerebral white matter lesions is non-inflammatory and is associated with reactive endothelial hyperplasia and thrombosis of small arterioles (Westerman et al. 1992). The clinical and laboratory diagnostic criteria were defined by an international committee (Wilson et al. 1999).

Recently, APS has emerged as an important cause of stroke in children and young adults, responsible for either arterial or venous thromboses (Takanashi et al. 1995; Kim et al. 2000; Heller et al. 2003). Moreover,subclinical CNS involvement in the form of multifocal cerebral white matter lesions mimicking MS is demonstrated by MRI in up to about 40% of patients with primary and secondary APS (Provenzale et al. 1996; Kim et al. 2000) (Fig. 22.7).

Brain

Large and small infarcts in children and young adults show the usual appearance on conventional MRI (Takanashi et al. 1995; Kim et al. 2000) (Fig. 22.8).

Also in the context of strokes in children, diffusion MR imaging has an established role for a more rapid detection of cerebral ischaemia (Gadian et al. 2000).

Haemorrhagic infarcts and intra-parenchymal haemorrhages are demonstrated by conventional MRI as areas exhibiting low signal intensity in T2weighted images, due to the paramagnetic properties of deoxyhaemoglobin in the acute phase, and high signal intensity in T1-weighted images, due to extracellular methaemoglobin in the subacute phase. Their identification should promote evaluation of the patency of the arterial or venous vessels with MR angiography, especially to rule out possible concomitant dural sinus thrombosis (Kim et al. 2000).

Cortical atrophy was reported as an additional finding in APS (Kim et al. 2000), but this is nonspecific, representing a possible effect of chronic steroid therapy as in SLE.

Optic Nerve

The MRI features of optic neuropathy in APS that is nonspecific resemble that in other form of “optic neuritis” (Besabs et al. 2001) (Fig. 22.7).

Spinal Cord

Spinal cord damage can be observed in patients with APS as primary clinical manifestation of the disease. In some instances it shows the typical MRI features of spinal cord infarction (Hasegawa et al. 1993).

In other cases, the features are less distinctive and resemble those that can be observed in other systemic vasculitis.

22.2.3

Post-infective Angiitis

Although infective vasculitis can complicate the course of syphilis (meningovascular syphilis), tuber-

Fig. 22.9 Midbrain infarct due to zoster angiitis in a 42-year- old man. Axial T2-weighted image obtained 6 months after right hemiparesis, which came 3 weeks after left zoster ophthalmicus, shows a focal area of signal change and mild thinning of the left cerebral peduncle

culosis (tuberculous meningitis), as well as fungal or bacterial (H. influenzae, staphylococcal, pneumococcal) meningeal infections, these conditions generally do not determine multifocal white matter damage and, as such, they will not be reviewed here. On the other hand, several viral infections–namely, chickenpox, measles, rubella and smallpox–can be followed by usually monophasic, multifocal white matter and grey matter damage (acute disseminated encephalomyelitis) that is addressed elsewhere in this book.

A peculiar case of post-infection vasculitis is the syndrome of herpes zoster ophthalmicus with delayed contralateral hemiparesis (Eidelberg et al. 1986). This association is rare and based upon the usual temporal evolution (first herpes ophthalmicus, second headache and hemiplegia) and the demonstration of viral particles in the smooth muscle cells of the media and varicella-zoster virus antigens in the media of the affected leptomeningeal vessels. Its pathogenesis is assumed to result from direct invasion of arterial walls via viral spread along the intracranial branches of the trigeminal nerve.

MR imaging demonstrates typical infarcted lesions with variable distribution in the anterior, middle and posterior cerebral arteries (Fig. 22.9).

22.2.4 Neurosarcoidosis

Sarcoidosis is a multisystem granulomatous disease, usually presenting with hilar adenopathy, pulmonary infiltration, and skin, eye and CNS involvement (Fig. 22.10).Very rarely,neurosarcoidosis may present

in the absence of systemic involvement, and in such cases it requires biopsy for the diagnosis (Seltzer et al. 1991; Cipri et al. 2000; Bode et al. 2001). Most of the CNS lesions in sarcoidosis are markedly sensitive to steroids (Lexa and Grossman 1994). However, considering this phenomenon, an indirect clue of neurosarcoidosis can delay the correct diagnosis in patients presenting with other steroid-responsive lesions, in particular tumours including lymphoma and germ cell tumour (Mascalchi et al. 1998b).

Brain

Multifocal, cerebral white matter damage indistinguishable from that seen in MS is common in neurosarcoidosis (Miller et al. 1988; Lexa and Grossman 1994) (Fig. 22.10). Although not specific, a key feature indicating sarcoidosis as the possible underlying substrate of the white matter lesions is diffuse focal leptomeningeal or dural enhancement better demonstrated by MRI after intravenous contrast administration (Sherman and Sherry 1990; Khaw et al. 1991; Seltzer et al. 1991). Additional features that should

raise suspicion of neurosarcoidosis include (Seltzer et al. 1991; Lexa & Grossman 1994): extra-axial contrast-enhancing masses mimicking meningioma (Fig. 22.10), intra-axial enhancing masses, periventricular enhancement, enlarged pituitary stalk, and enhancing nerve roots. All lesions, in particular their contrast enhancement, typically regress with steroid therapy (Lexa and Grossman 1994).

Optic Nerve

Optic nerve and chiasma are typical sites of CNS involvement in neurosarcoidosis (Bode et al. 2001; Carmody et al.1994).Enlargement of optic nerve and chiasma, associated with signal changes (Fig. 22.10) and contrast enhancement, is seen in the acute phase, and this regresses with steroid treatment. Atrophic optic nerve is observed in the chronic phase.

Spinal Cord

Spinal cord involvement in neurosarcoidosis is extremely rare (Seltzer et al. 1991; Lexa and

340

Grossman 1994). In the few available descriptions it was associated with extensive intramedullary lesions exhibiting contrast enhancement and was difficult to differentiate from an intramedullary neoplasm.

22.3 Conclusions

In addition to the multifocal white matter damage mimicking MS in secondary vasculitis and APS, the lesions include infarcts, haemorrhages, thrombosis of the intracranial dural sinuses and damage of the optic nerve and spinal cord of uncertain pathophysiology. It is noteworthy that the MRI features of these lesions are specific, and, hence, the differential diagnosis relies completely on the integration of clinical and laboratory findings.

The spectrum of lesions in neurosarcoidosis is wide, but coexistence of multifocal cerebral white matter lesions with meningeal enhancement or intraaxial or extra-axial contrast-enhancing masses should raise suspicions that this condition is present. However, lacking systemic manifestations of the disease, diagnosis of neurosarcoidosis still requires biopsy.

References

Akman-Demir G, Sorderoglu P, Tasci B et al (1999) Clinical patterns of neurological involvement in Behçet’s disease: evaluation of 200 patients. Brain 122:2171–2181

Alexander EL, Beall SS, Gordon B et al (1988) Magnetic resonance imaging of cerebral lesions in patients with Sjögren syndrome. Ann Intern Med 108:815–823

Al Kawi MZ, Bohlega S, Banna M (1991) MRI findings in neuroBehçet disease. Neurology 41:405–408

Angelini L, Rumi V, Nardocci N et al (1993) Hemidystonia symptomatic of primary antiphospholipid syndrome in childhood. Mov Disord 8:383–386

Bekkelund SI, Pierre-Jerome C, Husby G, Mellgren SI (1995) Quantitative cerebral MR in rheumatoid arthritis. AJNR Am J Neuroradiol 16:767–772

Besbas N, Anlar B, Apak A, Kansu T (2001) Optic neuropathy in primary antiphospholipid syndrome in childhood. J Child Neurol 16:690–693

Bode MK, Tikkakoski T, Tuisku S et al (2001) Isolated neuro- sarcoidosis–MR findings and pathologic correlation. Acta Radiol 42:563–567

Carmody RF, Mafee MF, Goodwin JA et al (1994) Orbital and optic pathway sarcoidosis: MR findings. AJNR Am J Neuroradiol 15:775–783

Cellerini M, Bartolucci M, Mortilla M et al (2003) MR imaging and proton MR spectroscopy of the brain in Behçet’s disease. Riv Neuroradiol 16:267–274

M. Mascalchi and F. Salvi

Cipri S, Gambardella G, Campolo C et al (2000) Unusual clinical presentation of cerebral-isolated sarcoidosis. J Neurosurg Sci 44:140–144

Condor SEJ, Jarosz JM (2002) Magnetic resonance imaging of cerebral venous sinus thrombosis. Clin Radiol 57:449–461 De Reuck J (2003) Dorsal thalamic hemorrhage complicating polyarteritis nodosa: a clinico-pathologic case report. Acta

Neurol Belg 103:40–42

De Seze J, Stojkovic T, Hachulla E et al (2001) MyelopathySjogren’s syndrome association: analysis of clinical and radiological findings and clinical course. Rev Neurol 157:669–678

Eidelberg D, Sotrel A, Horoupian DA et al (1986) Thrombotic cerebral vasculopathy associated with herpes zoster. Ann Neurol 19:7–14

Funauchi M, Hamada K, Enomoto H et al (1997) Characteristics of the clinical findings in patients with idiopathic thrombocytopenic purpura who are positive for anti-phos- pholipid antibodies. Intern Med 36:882–885

Gadian DG, Calamante F, Kirkham FJ et al (2000) Diffusion and perfusion magnetic resonance in childhood stroke. J Child Neurol 15:279–283

Galli M, Barbui T (2003) Antiphospolipid antibodies and thrombosis: strength of the association. Hematol J 4:180– 186

Harris EN, Gharavi AE, Hughes GRV (1985) Antiphospholipid antibodies. Clin Rheum Dis 11:591–609

Hasegawa M, Yamashita J, Yamashima T et al (1993) Spinal cord infarction associated with primary antiphospholipid syndrome in a young child. Case report. J Neurosurg 79:447–450

Heller C, Heinecke A, Junker R et al (2003) Cerebral venous thrombosis in children. A multifactorial origin. Circulation 108:1362–1367

Kadota Y, Tokumaru AM, Kamakura K et al (2002) Primary Sjogren syndrome initially manifested by optic neuritis: MRI findings. Neuroradiology 44:338–41

Kermode AG, Plant GT, Mac Manus DG et al (1989) Behçet’s disease with slowly enlarging midbrain mass on MRI: resolution following steroid therapy. Neurology 39:1251– 1252

Khaw KT, Manji H, Britton J, Schon F (1991) Neurosarcoid- osis–demonstration of meningeal disease by gadoliniumenhanced magnetic resonance imaging. J Neurol Neurosurg Psychiatry 54: 499–502

Kim JH, Choi CG, Choi SJ et al (2000) Primary antiphospholipid antibody syndrome: neuroradiologic findings in 11 patients. Korean J Radiol 1:5–10

Koçer N, Islak C, Siva A et al (1999) CNS involvement in neuroBehçet syndrome: an MR study. AJNR Am J Neuroradiol 20:1015–1024

Inaba G (1989) Behçet disease. In: Vynken PJ, Bruyn GW, Klawans HL (eds) Handbook of clinical neurology, vol 12. Elsevier, Amsterdam, pp 593–610

International study group for Behçet’s disease (1990) Lancet 335:1078-1080

Lafitte C (2002) Neuroradiological manifestations of primary Sjogren’s syndrome. Rev Neurol 158:959–965

Lafitte C, Amoura Z, Cacoub P et al (2001) Neurological complications of primary Sjogren’s syndrome. J Neurol 248:577–584

Lexa FJ, Grossman RI (1994) MR of sarcoidosis in the head and spine: spectrum of manifestations and radiographic

response to steroid therapy. AJNR Am J Neuroradiol 15:973–982

Manabe Y, Sasaki H, Marita H et al (2000) Sjogren’s syndrome with acute transverse myelopathy as the initial manifestation. J Neurol Sci 176:158–161

Mascalchi M, Salvi F, Piacentini S, Bartolozzi C (1994) Friedreich’s ataxia: MR findings involving the cervical portion of the spinal cord. AJR Am J Roentgenol 163:187–191

Mascalchi M, Cosottini M, Cellerini M et al (1998a) MRI of spinal cord involvement in Behçet’s disease: case report. Neuroradiology 40:255–257

Mascalchi M, Roncaroli F, Salvi F, Frank G (1998b) Transient regression of intracranial germ cell tumor following steroid intravenous administration. J Neurol Neurosurg Psychiatry 64:670–672

Miller DH, Ormerod IEC, Gibson A et al (1987) MR brain scanning in patients with vasculitis: differentiation from multiple sclerosis. Neuroradiology 29:226–231

Miller DH, Kendall BE, Barter S et al (1988) Magnetic resonance imaging in central nervous system sarcoidosis. Neurology 38:378–383

Mochizuki A, Hayashi A, Hisahara S, Shoji S (2000) Steroidresponsive Devic’s variant in Sjogren’s syndrome. Neurology 54:1391–1392

Mori K, Koike H, Misu K et al (2001) Spinal cord magnetic resonance imaging demonstrates sensory neuronal involvement and clinical severity in neuronopathy associated with Sjogren’s syndrome. J Neurol Neurosurg Psychiatry 71:488–492

Pierot L, Sauve C, Leger JM et al (1993) Asymptomatic cerebral involvement in Sjogren’s syndrome: MRI findings in 15 cases. Neuroradiology 35:378–380

Provenzale JM, Allen NB (1996) Neuroradiologic findings in polyarteritis nodosa. AJNR Am J Neuroradiol 17:1119– 1126

Provenzale JM,Barboriak DP,Allen NB,Ortel TL (1996) Patients with antiphospholipid antibodies: CT and MR findings of the brain. AJR Am J Roentgenol 167:1573–1578

Reichhart MD, Bogousslavsky J, Janzer RC (2000) Early lacunar strokes complicating polyarteritis nodosa. Thrombotic microangiopathy. Neurology 54:883–889

Reshef E, Greenberg SB, Jankovic J (1985) Herpes zoster ophthalmicus followed by contralateral hemiparesis: report of

two cases and review of the literature. J Neurol Neurosurg Psychiatry 48:122–127

Salvi F, Mascalchi M, Malatesta R et al (1999) Optic neuropathy in Behçet’s disease. Report of two cases. Ital J Neurol Sci 20:181–186

Sanna G, Bertolaccini ML, Cuadrado MJ et al (2003) Central nervous system involvement in the antiphospholipid (Hughes) syndrome. Rheumatology 42:200–213

Seltzer S, Mark AS, Atlas SW (1991) CNS sarcoidosis: evaluation with contrast-enhanced MR imaging. AJNR Am J Neuroradiol 12:1227–1233

Shakir RA, Sulaiman K, Kahn RA, Rudwan M (1990) Neurological presentation of Neuro-Behçet’s syndrome: clinical categories. Eur Neurol 30:249–253

Sherman JL, Stern BJ (1990) Sarcoidosis of the CNS: comparison of unenhanced and enhanced MR images. AJNR Am J Neuroradiol 11:915–923

Sklar AM, Schatz NJ, Glaser JS et al (1996) MR of vasculitisinduced optic neuropathy. AJNR Am J Neuroradiol 17:121– 128

Takanashi J, Sugita K, Miyazato S et al (1995) Antiphospholipid antibody syndrome in childhood strokes. Pediatr Neurol 13:323–326

Tesar JT, McMillan V, Molina R, Armstrong J (1992) Optic neuropathy and central nervous system disease associated with primary Sjogren’s syndrome. Am J Med 92:686–692

Trent K, Neustater BR, Lottenberg R (1997) Chronic relapsing thrombotic thrombocytopenic purpura and antiphospholipid antibodies: a report of two cases. Am J Hematol 54:155–159

Wechsler B, Dell’Isola B, Vidailhet M et al (1993) MRI in 31 patients with Behçet’s disease and neurological involvement: prospective study with clinical correlation. J Neurol Neurosurg Psychiatry 56:793–798

Westerman EM, Miles JM, Backonja M, Sundstrom WR (1992) Neuropathologic findings in multi-infarct dementia associated with anticardiolipin antibody. Evidence for endothelial injury as the primary event. Arthritis Rheum 35:1038–1042

Wilson WA, Gharavi AE, Koike T et al (1999) International consensus statement on preliminary classification criteria for definite antiphospholipid syndrome. Arthritis Rheum 12:1309–1311

CONTENTS

23.1Introduction 343

23.2Conventional MRI 343

23.4Diffusion-Weighted MRI 348

23.6Conclusions 350 References 350

23.1 Introduction

Systemic immune-mediated diseases (SID) can affect the central nervous system (CNS), either as the onset clinical manifestation or as a late complication in the context of a multiple organ involvement (Calabrese et al. 1997; Fieschi et al. 1998). Among SID, systemic lupus erythematosus (SLE), Behçet disease (BD), small-vessel vasculitides (SVV) and primary antiphospholipid antibody syndrome (APS) may often have a flare-like clinical course which closely resembles that of multiple sclerosis (MS) and, therefore, be considered in the differential diagnosis of this latter condition.

Numerous studies reported that conventional, T2weighted brain magnetic resonance imaging (MRI) is sensitive for detecting CNS lesions in patients with SID (Miller et al. 1987; Coban et al. 1996; Liem et al. 1996; Provenzale and Allen 1996; Gumà et al. 1998; Hachulla et al. 1998). A wide spectrum of non-specific MRI abnormalities has been described in these patients, including cerebral infarctions, brain atrophy, dural sinus thrombosis and, more rarely, hemorrhages or meningeal involvement. In a significant proportion of patients with SID, however, multiple brain white matter lesions mimicking those due to MS can be the only MRI-visible abnormali-

M. Rovaris, MD; M. Filippi, MD

Neuroimaging Research Unit, Department of Neurology, Scientific Institute and University Ospedale San Raffaele, Milan, Italy

ties (Miller et al. 1987; Nadeau 1997; Triulzi and Scotti 1998) (Fig. 23.1). Unfortunately, the limited pathological specificity of the latter findings hampers their diagnostic accuracy, as well as their prognostic value and relevance in the work-up of SID patients.

Quantitative MR-based techniques, such as magnetization transfer (MT) MRI, diffusion-weighted (DW) MRI and magnetic resonance spectroscopy (MRS), make it possible to obtain information about white matter damage with increased pathological specificity over T2-weighted sequences (Miller et al. 2003). In addition, all these techniques have the ability to assess and quantify the extent of white matter pathology beyond the resolution of conventional MRI, e.g. in the so-called normal-appearing white matter (NAWM). NAWM pathology seems to be a relevant contributor to the neurological impairment in patients with MS (Miller et al. 2003) and its accurate assessment might also be useful for the differential diagnosis between SID and other CNS diseases, as well as for a better understanding of the mechanisms leading to CNS dysfunction in the former conditions.

This chapter reviews the contributions of conventional and modern MRI techniques to the in vivo investigation of white matter pathology in SID.

23.2 Conventional MRI

Between 30% and 70% of patients with SLE develop neuropsychiatric complications (NSLE) during the course of the disease. In NSLE patients, brain MRI is abnormal in 40%–60% of cases (Csépàny et al. 2003; Jennings et al. 2004), but the observed findings are not disease-specific, the commonest being cerebral infarcts, brain atrophy and multiple, MS-like white matter lesions. The prevalence of brain MRI abnormalities in patients with BD and CNS involvement (NBD) varies between 30% and 86% (Akman-Demir et al. 1999; Coban et al. 1996; Gerber et al. 1996; Lee

et al. 2001). Most frequently, these abnormalities consist of brainstem and basal ganglia lesions that may shrink or disappear at follow-up (Lee et al. 2001). In 20%–40% of NBD cerebral white matter is diffusely involved (Akman-Demir et al. 1999; Lee et al. 2001). Among patients with SVV, only a minority of those with Wegener granulomatosis (WG) may show some degree of brain MRI abnormalities (Provenzale and Allen 1996), although the prevalence of clinical CNS involvement ranges between 22% and 54% of cases (Fieschi et al. 1998). Focal infarcts are the commonest MRI abnormalities described in APS, but brain atrophy and diffuse, T2-hyperintense white matter lesions can also be found in a high percentage of cases (Hachulla et al. 1998; Ijdo et al. 1999; Weingarten

et al. 1997). Moreover, recent reports found that up to 20% of MS patients with brain MRI findings highly suggestive for this disease (Ijdo et al. 1999; Karussis et al. 1998) can be consistently positive for antiphospholipid antibodies,thus suggesting the possibility of a concomitant diagnosis of MS and APS.

It is worthy noting that, in SID patients, the use of fast fluid-attenuated inversion recovery (fFLAIR) sequences, which are known to be more sensitive than conventional T2-weighted MRI for detecting lesions in the brain of patients with inflammatory or demyelinating diseases of the CNS (Maubon et al. 1998), does not seem to increase the sensitivity of brain scanning for the detection of white matter abnormalities (Rovaris et al. 2000a). Among the po-

tential explanations for this finding, the pathological heterogeneity of white matter damage in SID also has to be considered. Brain pathology in SID may range from inflammation to acute and chronic ischaemia, which can be secondary to vessel thrombosis or vasculitis. Small chronic infarcts can appear isointense to normal tissue on fFLAIR scans, and, therefore, go undetected when using this sequence.

White matter abnormalities detected on T2weighted and post-contrast T1-weighted MRI of the brain have a limited diagnostic specificity for SID. Several studies (Boumpas et al. 1990; Deodhar et al. 1999; Mascalchi et al. 1998; Mok et al. 1998; Morrissey et al. 1993; Provenzale et al. 1994; Salmaggi et al. 1994; Yoshioka et al. 1996) have suggested that MRI abnormalities in the cord of patients with SID are well correlated with ongoing symptoms of myelopathy and that they can completely disappear after steroid or immunosuppressive treatment (Boumpas et al. 1990; Lee et al. 2001; Mascalchi et al. 1998; Rovaris et al. 2000c; Salmaggi et al. 1994; Yoshioka et al. 1996). On the other hand, spinal cord MRI abnormalities can be detected in up to 90% of MS patients (Lycklama à Nijeholt et al. 1998; Miller et al. 1998; Rocca et al. 1999), often without a concomitant clinical involvement, and the presence of such lesions may increase the confidence when diagnosing MS at its clinical onset (Fazekas et al.1999). Against this background, the detection of MRI-visi- ble white matter pathology in the cord might provide useful information for the work-up of patients with CNS disturbances,especially when a differential diagnosis between SID and MS has to be made. In a crosssectional study of 44 patients with SID (Rovaris et al. 2002), of whom 48% had had clinical manifestations of CNS involvement, cervical cord MRI scans were always found to be normal, whereas brain MRI revealed white matter lesions in 52% of the cases. The application of standardized criteria for brain MR image interpretation (Barkhof et al. 1997), which were originally developed for predicting the evolution to established MS in patients at presentation with clinically isolated neurological syndromes suggestive of MS, yielded an accuracy of about 85% in differentiating SID from age-matched MS patients and, by using the presence or absence of cervical cord MRI lesions as a dichotomized criterion, a correct re-classifica- tion of 77% of MS patients and two SLE patients who were misclassified based upon their patterns of brain abnormalities was possible (Rovaris et al. 2002). The additional value of cord MRI in the differentiation of MS from SID and other neurological diseases of ischaemic etiology has been emphasized by another

study (Bot et al. 2002), where asymptomatic cord lesions were found in 92% of MS and 6% of non-MS patients and the concomitant evaluation of both brain and cord MRI findings achieved a 95% diagnostic accuracy (Fig. 23.2). Both these studies (Bot et al. 2002; Rovaris et al. 2002) confirm that conventional MRI patterns of brain white matter damage related to SID do not have a clear-cut diagnostic value and they also underpin the importance of a careful interpretation

a

b

Fig. 23.2a,b. Brain and spinal cord MR images in a patient with MS (a) and in a patient with Sjögren disease associated with clinical CNS dysfunction (b). Brain images fulfill diagnostic criteria for MS in both cases. However, sagittal cord images reveal the presence of diffuse abnormalities in the cervical tract and one focal lesion in the thoracic tract of the patient with MS, whereas no abnormalities can be seen in the patient with Sjögren disease. [Reproduced with permission from Bot et al. (2002)]

of MRI findings based on a comprehensive image evaluation, which should consider, on the one hand, the radiologist’s opinion and, on the other, the application of standardized criteria.

Several studies have investigated whether the conventional MRI patterns of white matter pathology in SID patients may correlate with their clinical or immunological status and, therefore, have a value as paraclinical outcomes to monitor the disease evolution.

The severity of neuropsychological dysfunction in SLE patients seems to be associated with the presence of cerebral infarcts (Waterloo et al. 2001) rather than with the overall burden of white matter lesions (Kozora et al. 1998; Sailer et al. 1997). The limited functional relevance of T2-visible white matter abnormalities is also consistent with their weak correlation with neuroimaging correlates of brain metabolisms (Sailer et al. 1997; Waterloo et al. 2001). On the other hand, however, it has been found (Bell et al. 1991; West et al. 1995) that NSLE patients with non-focal psychiatric symptoms more frequently show an MRI pattern of multiple white matter abnormalities and have higher titers of antineurofilament antibodies than those with focal CNS disturbances. Two recent studies of NSLE patients also reported a significant relationship of MRI findings with the presence of antiphospholipid antibodies (Csèpàny et al. 2003) and the levels of biochemical markers of axonal damage in the cerebrospinal fluid (Trysberg et al. 2003). These findings suggest that, in NSLE, different pathogenic mechanisms may be responsible for brain tissue damage, including both vasculopathic and im- mune-mediated neuronal injury, and underpin the need of MRI-derived measures with higher pathological specificity to monitor the disease course and its response to treatment. In patients with NBD, conventional MRI patterns of CNS involvement do not seem to have a clear-cut prognostic value (Akman-Demir et al. 1999; Lee et al. 2001), even though the presence of brainstem lesions, together with other clinical and laboratory characteristics, make it possible to identify those patients with a poorer clinical outcome (Akman-Demir et al. 1999). In a study of patients with SVV (Mattioli et al. 2002), the presence of multiple white matter MRI abnormalities was frequently associated with subclinical cognitive impairment.Admittedly,caution must be exercised when interpreting MRI data from the latter study (Mattioli et al. 2002), which were obtained from a small sample of patients. However, the observation that less than 15% of cognitively impaired SVV patients in this series had a normal brain MRI (Mattioli et al. 2002)

may indicate that the presence of MRI-visible white matter lesions significantly increases the possibility for an SVV patient to have neuropsychological impairment.

23.3

Magnetization Transfer MRI

MT MRI has several advantages over conventional T2and T1-weighted MRI for the in vivo structural investigation of CNS disorders. First, it provides information with a high pathological specificity. Low magnetization transfer ratio (MTR) indicates a reduced capacity of the molecules in the brain tissue matrix to exchange magnetization with the surrounding (MRI-visible) water molecules and is associated with severe demyelination and axonal loss (Brochet and Dousset 1999). Secondly, MT MRI enables us to assess the “invisible” disease burden in the normal-appearing brain tissue (NABT). Thirdly, MT MRI can provide, from a single procedure, multiple parameters influenced by both the MRI-visible and MRI-undetectable disease burden. They are obtained after the automated creation of MTR maps, where the signal intensity of each pixel represents its MTR value. The analysis can then be done on a region-of-interest (ROI) basis or more globally using MTR histograms (van Buchem et al. 1996).

An ROI-based study comparing MT MRI findings from 21 patients with MS and nine with SLE showed that the average MTR values of T2-visible white matter lesions and NAWM in the brain were significantly lower for the former group (Campi et al. 1996). In the same study, MTR values from the NAWM of SLE patients were found not to differ from those of healthy controls.Using histogram-based analysis,an MT MRI study of 44 patients with SID, of whom 15 with SLE, nine with NSLE, five with BD, nine with WG and six with APS, was also conducted (Rovaris et al. 2000b). Ten patients with MS served as a control group. T2weighted brain MRI abnormalities were found in all MS cases and in 52% of patients with SID, for whom a white matter lesion pattern indistinguishable from that of MS was observed in 40% of cases. MS patients had significantly lower lesion MTR than SLE and WG patients; NSLE had significantly lower lesion MTR than SLE patients. MS patients had significantly lower average MTR values in the NABT than all SID but NSLE patients, who, in turn, had significantly lower average NABT MTR than SLE patients (Fig. 23.3). Both T2-weighted lesion volume and av-