Myelographic findings in adhesive spinal arachnoiditis (with a brief surgical note)

Patterns of chronic adhesive arachnoiditis following Myodil myelography: the significance of spinal canal stenosis and previous surgery

109 patients who had undergone Myodil myelography on at least one occasion were identified. The patterns of lumbar nerve root distribution in this group were examined using magnetic resonance imaging. The relationship between these patterns and the presence of spinal stenosis or previous surgery was investigated. Chronic adhesive arachnoiditic nerve root patterns were seen in 68 patients and were classified into three groups according to Delemarter et al. Central clumping of nerve roots (type 1) and complete opacification of the thecal sac (type 3), extending over at least one vertebral level, were significantly related to spinal stenosis at an adjacent level (p < 0.0001). Peripheral adhesion of nerve roots to the theca (type 2) was significantly related to previous surgery at the level of abnormality (p < 0.00005). Only a single case of arachnoiditic nerve root patterns was seen in the absence of stenosis or previous surgery. We conclude that chronic adhesive arachnoiditis is significantly related to previous Myodil myelography in the presence of spinal stenosis or previous surgery but that Myodil alone rarely produces these changes.

Does degenerative disease of the lumbar spine cause arachnoiditis? A magnetic resonance study and review of the literature

The magnetic resonance appearances in 165 patients with symptoms suggestive of degenerative lumbar spine disease were reviewed. The aim of the study was to evaluate the relationship between abnormalities of nerve root distribution and degenerative disease of the lumbar spine in the absence of other known risk factors for arachnoiditis. Central clumping of nerve roots was present in 16 patients (9.7%) and was associated with spinal stenosis at one of the affected levels in all (p < 0.001). Spinal stenosis was present in 44 patients giving an incidence of abnormal nerve root distribution of 36% in this group. Nerve root clumping occurred in association with pure spinal stenosis (10 cases), stenosis secondary to disc prolapse (four cases) and degenerative spondylolisthesis (two cases). Nerve root clumping was confined to one vertebral level in nine cases and extended over two to four levels in seven. In five of the latter spinal stenosis was present at multiple levels. The appearance of nerve root clumping described here may result entirely from mechanical apposition of nerve roots but is indistinguishable from the central pattern of nerve root adhesions which occurs in adhesive lumbar arachnoiditis. No abnormalities of nerve root distribution were seen in association with any indicator of degenerative disk disease in the absence of stenosis. We have been unable to demonstrate the previously reported relationship between lumbar disk degeneration and arachnoiditis and discuss this with a critical review of the literature. Abnormal central clumping of nerve roots as described in arachnoiditis may occur in association with spinal stenosis in the absence of other risk factors although the cause for this appearance remains unexplained. Arachnoiditis-like changes extending over more than one vertebral level are rare (7%) except in the presence of spinal stenosis at multiple levels (29%). Awareness of this appearance may avoid a possibly incorrect diagnosis of arachnoiditis in the presence of a treatable stenosis.

MRI in intraspinal tuberculosis

We studied 20 patients with intraspinal tuberculosis (TB), to characterise the MRI features of tuberculous meningitis and myelitis. MRI leptomeningitis and intramedullary involvement in 11 patients, intramedullary lesions alone in 5, leptomeningitis alone in 2, and isolated extradural disease in 2. TB leptomeningitis was characterised by loculation of the cerebrospinal fluid (CSF), nerve root thickening and clumping (seen only in the lumbar region) or complete obliteration of the subarachnoid space on unenhanced images. Gd-DTPA-enhanced images proved useful in 6 cases, revealing linear enhancement of the surface of the spinal cord and nerve roots or plaque-like enhancement of the dura-arachnoid mater complex. Intramedullary lesions included tuberculomas (8), cord oedema (5) and cavitation (3). In seven cases of intramedullary tuberculoma multiple lesions with skip areas were seen, without significant cord swelling. One patient had an isolated lesion in the conus medullaris. The lesions were iso- or hypointense on T1-weighted images, iso-, hypo- or hyperintense on T2-weighted images and showed rim or nodular enhancement with contrast medium.

Intrathecal hyaluronidase treatment of chronic spinal arachnoiditis of noninfective etiology

Fifteen cases of spinal arachnoiditis of noninfective etiology were treated with intrathecal hyaluronidase. In 10 patients, no obvious cause was evident; in four patients, arachnoiditis was the result of disk prolapse; and in one it followed spinal surgery. Fifteen hundred international units of hyaluronidase was injected intrathecally at intervals of 7-14 days. Four to nine injections (mean, 5.9) were given to each patient, and the patients were observed for 3-30 months (mean, 9.7). The effect of the treatment was assessed by the change in clinical grading from the pretreatment phase to the last examination. Satisfactory improvement was observed in 11 patients (73.3%) and "mild" recovery was observed in three patients. Because of this therapeutic efficacy and lack of serious toxic effects, intrathecally administered hyaluronidase appears to be of value in the management of spinal arachnoiditis.

Spinal arachnoiditis

A review of the literature points to the many causes of arachnoiditis and the failure of treatment to arrest or reverse its effects. The true incidence cannot be determined, although it is probably lower than might at first appear from the published articles. In the radiological literature the diagnosis seems to derive from an examination of the films alone, often without reference to the clinical findings or appearance at operation. While attempts at treatment are usually unsuccessful, some iatrogenic cases can be prevented by the avoidance of intrathecal steroid injections or unduly rough or repeated surgical exploration of the lumbar vertebral canal.

Chronic mycotic meningitis with spinal involvement (arachnoiditis): a report of five cases

Five patients developed mycotic spinal arachnoiditis-meningitis causing signs and symptoms of spinal cord neoplasm. Four had cryptococcal infection, the fifth had aspergillosis. In three patients, diagnosis was made at surgery; all three developed acute fungal meningitis postoperatively and two died. The diagnosis was made nonsurgically in two patients and was followed by medical cure. These five and twelve other reported patients with mycotic spinal arachnoiditis shared features that suggested the diagnosis. In contrast to most patients with spinal tumors, those reported here tended to be young (mean age, 32 years), to lack evidence for a primary tumor, and to have a fluctuating history of spinal symptoms for several months. Frequent associated findings were recent pregnancy; the abuse of alcohol, narcotics, or both; and the presence of headache and fever. Plain roentgenograms of the spine were normal. No single finding was diagnostic, but the combination of several would be rare with spinal tumor.

Non-traumatic adhesive arachnoiditis as a cause of spinal cord syndromes. Investigation of 507 patients

Spinal cord syndromes with a mainly syringomyelic pattern of sensory diorders, radiculopathies, mixed paresis of varying degree (without any history of trauma), have been found in 507 out of 1305 new patients referred to out Clinic from January 1976 till 31 October 1977. In 105 randomised and unselected cases with these syndromes, myelographies have disclosed findings compatible with an adhesive spinal and/or cisternal arachnoiditis. A prospective study of the syndromes for evidence of infectious aetiology has been performed, in which tuberculosis, syphilis and other infections appear to be causative agents. A randomised therapeutic trial on a limited number of cases has been evaluated, as well as the results of specific therapy in a larger number of cases. Results of treatment have not been satisfactory. Operations were performed on only five patients and in no case was an autopsy obtained. Spinal cord syndromes due to non-traumatic adhesive arachnoiditis are discussed. The possible pathogenetic mechanisms the predominantly syringomyelic sensory deficits in those syndromes are briefly mentioned.

Spinal arachnoiditis: disease or coincidence?

In 86% of 63 patients with spinal arachnoiditis the localization was lumbosacral, and in 14% it was cervical or thoracic. The most important aetiological factor was the combination of one or more myelographies with one or more operations. It is remarkable that in most patients with lumbosacral arachnoiditis the clinical picture did not show new or typical signs of arachnoiditis, but was the same as before the arachnoiditis had developed. In only 20% of patients with lumbosacral arachnoiditis, and in all patients with cervical or thoracic arachnoiditis, did new signs and symptoms occur concomitantly with and due to the development of the arachnoiditis. The results of this retrospective study suggest that lumbosacral arachnoiditis is a coincidental finding in the majority of cases, without clinical consequences.

Late meningeal effects of myelographic contrast media with special reference to metrizamide

Two hundred and fifty-six myelograms in 192 patients were following by repeat myelography using a water-soluble contrast medium. The films of these re-examinations were analysed for signs of meningeal changes. Adhesive arachnoiditis had developed in five of 46 cases after methiodal sodium, in 24 of 96 cases after iothalamate meglumine, in seven of 12 cases after iocarmate meglumine, in seven of ten cases after iodophendylate, but in none of 90 cases after metrizamide. Dose of contrast medium, observation time and protein concentration of the CSF were found to be factors of importance for the development of adhesive arachnoiditis.

Obliterative arachnoiditis complicating lumbar spinal stenosis

The authors report five patients with spinal stenosis who had a total myelographic block at the level of the obliterated subarachnoid space. Arachnoiditis had not been considered as a primary diagnosis until laminectomy revealed a non-pulsating, thickened dural sac that conformed to the internal configuration of the involved spinal canal. Two patients had stenosis complicated by spondyloarthrosis over multiple lumbar levels, one had a previous spinal fusion, another had degenerative spondylolisthesis, and the fifth had a large midline extruded disc at L2-3 that completely blocked the spinal canal. The dura was opened in two patients, confirming the lesion. Despite obliteration of the subarachnoid space, significant relief for approximately 1 year followed decompressive laminectomy, foraminotomy, and discectomy, with disappearance of neurogenic claudication in three patients. Postoperative erect films showed no caudad passage of contrast. While further observations are required, an awareness of this complication of spinal stenosis is important in the diagnosis and management of such patients and in evaluating their ultimate prognosis.

A myelographic variant in lumbar arachnoiditis

✓ A variant of lumbar arachnoiditis is described in patients undergoing rhizotomy for intractable radicular pain following multiple lumbar epidural operative procedures. The peculiar adhesions of the arachnoid, nerve roots, and dura are described and pictured. The slight myelographic abnormality seen in the presence of this kind of arachnoiditis, namely, lack of root sleeve visualization, is correlated with the gross operative findings and contrasted with the commonly recognized myelographic picture of lumbar arachnoiditis. Diagnostic and therapeutic implications are discussed.