BMJ 1999;318:1604-1605 ( 12 June )

Clinical review

Lesson of the week

Cavernous haemangioma mimicking multiple sclerosis

M Zameel Cader, house officer J B Winer, consultant neurologist

Department of Neurology, Queen Elizabeth Hospital, Birmingham B15 2TT

Correspondence to: Dr Winer j.b.winer{at}bham.ac.uk

Patients diagnosed with multiple sclerosis before the advent of magnetic resonance imaging and whose symptoms could be attributable cavernous haemangioma should be reviewed with magnetic resonance imaging

Cavernous haemangiomas are vascular malformations that rarely affect the brain, but their clinical presentation can simulate multiple sclerosis. The likelihood of mistaking cavernous haemangioma for multiple sclerosis is increased further by the fact that cavernous haemangiomas are poorly identified by angiography, and even computed tomography has a relatively low sensitivity and specificity for these lesions.1 However, high field magnetic resonance imaging is able to distinguish cavernous haemangiomas, and since its introduction increasing numbers have been reported.1

We describe two patients who had been diagnosed as having multiple sclerosis many years before the widespread use of computed tomography or the advent of magnetic resonance imaging. Magnetic resonance imaging subsequently showed that they had cavernous haemangioma.

    Case reports
Top
Case reports
Discussion
References

Case 1
A 17 year old boy presented in 1975 with diplopia, left sided facial weakness, and dizziness. He was right handed. Physical examination showed that he had left VI nerve palsy and mild weakness of the lower left side of his face. The patient was admitted to hospital for further investigation. A lumbar puncture showed clear cerebrospinal fluid under normal pressure but a slightly high gamma  globulin fraction (0.04 g/l; 12.1% of total protein), a red cell count of 4×109/l, and a white cell count of 1×109/l.

The patient's facial weakness resolved over several months, but the left VI nerve palsy persisted. About eight months after his initial presentation, he had a further episode of neurological dysfunction with numbness of the right hand. He had reduced sensations of light touch and pain, and a lumbar puncture again showed an increased gamma  globulin fraction. One month later, the numbness in the right hand had resolved, but the palsy in the left VI nerve remained unchanged. A diagnosis of probable multiple sclerosis was made.

The patient had no further problems until he was 34 years old, when he presented with numbness of the right hand and left side of the face and impaired balance. Physical examination showed a subjective sensory disturbance in the right hand, continuing palsy of the left VI nerve, and considerably reduced visual acuity in the left eye.

Electrophysiological studies, computed tomography, and magnetic resonance imaging were arranged. Visual stimuli either failed to evoke recordable cortical responses or produced abnormal waveforms on both sides. Stimulation of the left median nerve produced well formed responses at the cervical cord, but a poor cortical response. Stimulation of the right median nerve produced poor responses at the cervical cord and the cerebral cortex. Computed tomography showed hyperdense, ill defined lesions with irregular calcification in the posterior aspect of the pons, the right frontal lobe, and the periventricular region of the right parietal lobe. There was no enhancement after contrast. With magnetic resonance imaging (figure), the lesions returned mixed signals, there was no surrounding cerebral oedema and no mass effect. It was concluded that this patient had multiple cavernous haemangioma.

Case 2
A 36 year old man presented to a neurologist in 1964. He was right handed. He had a five year history of laughing without reason (which had led to considerable embarrassment) and increasing weakness of the right arm and leg. Physical examination showed that he had a right spastic hemiparesis and a pseudobulbar palsy. No abnormalities were detected by a radioisotope scan or angiography of the brain, and multiple sclerosis was considered the most likely diagnosis.

Despite the neurological impairment, his disability over the next 15 years was slight, and he continued to work as a manager for an advertising company. However, at the age of 51, the right sided weakness and spasticity began to progress. His inappropriate laughter was also becoming more troublesome. He presented again, aged 56, when he developed a squint in his left eye. At this time physical examination showed palsy of the left VI nerve, right spastic hemiparesis, and pseudobulbar palsy. Visual evoked responses and electroencephalographic findings were normal.



View larger version (109K):
[in this window]
[in a new window]
 
Magnetic resonance imaging scan (T2 weighted sagittal section) of a large cavernous haemangioma in the upper brain stem of case 1 

The patient was not seen again until he was 68, when he asked about the possibility of treatment with interferon beta. At this stage he was very disabled---he could walk only 15 yards with tripod support. Magnetic resonance imaging was arranged to confirm the diagnosis of multiple sclerosis. However, this showed a large, well circumscribed mass in the lower pons and upper medulla that returned mixed signals. There was no surrounding oedema and no mass effect. It was concluded that he had a large cavernous haemangioma.

    Discussion
Top
Case reports
Discussion
References

Cavernous haemangiomas are an irregular honeycomb of thin-walled vascular spaces without intervening brain tissue, and usually without any increase in the vascularity of adjacent tissues. The lesions generally occur singly. Only a few are infratentorial, and the pons is the commonest site in these cases. Cavernous haemangiomas usually manifest with the effects of a space occupying lesion or from spontaneous haemorrhage.2

However, there have been several published reports of vascular malformations of the brain stem that run a long, insidious course and simulate multiple sclerosis.3-5 Cavernous haemangiomas are not usually visualised by angiography because the blood flow through them is slow.6 Nor is computed tomography able to identify reliably all cavernous haemangiomas. A misdiagnosis of multiple sclerosis was therefore a particular hazard before magnetic resonance imaging became widely available.

It is interesting that patient 1 had a high gamma  globulin fraction---a finding that would be consistent with multiple sclerosis. A recent case report describes a patient who was thought to have multiple sclerosis on the basis of an insidious history, negative computed tomographic findings, cerebrospinal fluid with oligoclonal bands, and (later) a raised gamma  globulin fraction.3 These authors noted that oligoclonal bands and raised gamma  globulin values are sensitive, but not specific, for multiple sclerosis.

The clinical distinction between cavernous haemangiomas and multiple sclerosis affecting the brain stem is sometimes difficult. In neither of our patients was there an obvious episode of optic neuritis or any disordered eye movement other than the VI nerve palsy---findings that would have been unusual for multiple sclerosis. In both our patients, the initial clinical picture could have been attributed to a single brain stem lesion. In these cases there should be a high index of suspicion for a structural lesion. However, cavernous haemangioma can occasionally be multiple. This adds to the difficulty in making a diagnosis and supports the value of magnetic resonance imaging in confirming a diagnosis of multiple sclerosis.

Abnormal findings in cerebrospinal fluid and results of electrophysiological studies are clearly non-specific. However, with the increasing use of magnetic resonance imaging for detecting white matter plaques when multiple sclerosis is suspected, differentiating between the two abnormalities should not pose a problem. For patients diagnosed as having multiple sclerosis before the advent of modern imaging technologies, particularly those whose symptoms could be attributable to single lesions, the diagnosis should be reviewed carefully.

    Acknowledgments

We thank Dr R Hughes and Professor A C Williams.

    References
Top
Case reports
Discussion
References

1. Curling OD, Kelly DL, Elster AD, Craven TE. An analysis of the natural history of cavernous angioma. J Neurosurg 1991; 75: 702-708[Medline].
2. Russell DS, Rubinstein LJ. Pathology of tumours of the nervous system. 5th ed. London: Edward Arnold , 1989.
3. Vrethem M, Thomas KA, Hillman J. Cavernous angioma of the brain stem mimicking multiple sclerosis [letter]. N Engl J Med 1997; 336: 875-876[Free Full Text].
4. Honczarenko K, Fryze C, Nowacki P, Osuch Z, Grzelec H, Fabian A. Cavernous angioma of brain stem mimicking multiple sclerosis. Folia Neuropatholigica 1995; 33: 251-254.
5. Stahl SM, Johnson KP, Malamud N. The clinical and pathological spectrum of brain stem vascular malformations. Long term course stimulates multiple sclerosis. Arch Neurol 1980; 37: 25-29[Abstract].
6. Requena I, Arias M, Lopez-Ibor L, Pereiro I, Barba A, Alonso A, et al. Cavernomas of the central nervous system: clinical and neuroimaging manifestations in 47 patients. J Neurol Neurosurg Psychiatry 1991; 54: 590-595[Abstract].

(Accepted 9 November 1998)


© BMJ 1999

Add to CiteULike CiteULike   Add to Complore Complore   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us   Add to Digg Digg   Add to Reddit Reddit   Add to Technorati Technorati    What's this?

Relevant Article

Vascular anomalies must be properly classified
Patrick Mallucci
BMJ 1999 319: 919. [Extract] [Full Text]

This article has been cited by other articles:

  • Mallucci, P. (1999). Vascular anomalies must be properly classified. BMJ 319: 919-919 [Full text]  

Rapid Responses:

Read all Rapid Responses

Neuropsychiatric presentation of cavernous hemangiomas
Mina Fazel
bmj.com, 4 Aug 1999 [Full text]
Two problems
David Clements
bmj.com, 9 Nov 1999 [Full text]
The term cavernous haemangioma is a confusing one and should ideally be avoided
Patrick Mallucci
bmj.com, 19 Jul 1999 [Full text]
BMJ admits error
Richard Smith
bmj.com, 9 Nov 1999 [Full text]



Access all current jobs at BMJ Group
Whats new online at Student 

BMJ
Listen to the latest 

BMJ Interview