5
中華放射醫誌 Chin J Radiol 2004; 29: 217-221 217 Cauda equina syndrome is an unusual complica- tion of ankylosing spondylitis. We present a case of a 68-year-old woman with long-standing ankylosing spondylitis who developed cauda equina syndrome. MRI revealed a characteristic widening of the lum- bosacral thecal sac and numerous dorsal thecal diverticula. MRI is a powerful, noninvasive tool to confirm the diagnosis and exclude a treatable com- pressive lesion and provides a good alternative to myelography, an invasive and potentially dangerous procedure in such patients. Key words: Ankylosing spondylitis; Arachnoiditis; Cauda equina syndrome; Magnetic resonance imaging Cauda equina syndrome is a rare, late neurolog- ical complication of long-standing ankylosing spondylitis. Widening of the thecal sac and numerous dorsal thecal diverticula in the lumbosacral region are usually seen in such patients. It is important to distin- guish this disorder from an intraspinal compressive lesion. The MRI findings are characteristic, and famil- iarity with them is helpful in making a correct diagnosis. CASE REPORT A 68-year-old woman had ankylosing spondylitis for many years. For several months, she complained of low back pain radiating to the left leg, numbness of the right leg, and intermittent claudication. There was also slight weakness in both legs. The patient denied problems with urination or defecation. Neurological examination revealed hyperreflexia in both legs and hypesthesia to pinprick over the saddle region (S3-S5 dermatomes). A KUB and a lateral lumbar spine film showed typical changes of ankylosing spondylitis with fusion of both sacroiliac joints, squaring of the lumbar vertebral bodies, and syndesmophyte formation along the lumbar spine (Fig. 1). MRI examination at 1.5T demonstrated a wide thecal sac from L1 to S2, with extensive scalloping of the pedicles, laminae, and spinous processes of several vertebrae caused by numerous dorsal thecal diverticula (Fig. 2). Axial T2- weighted images (T2WI) showed clumping of the nerve roots of the cauda equina on the side of the ectatic thecal sac in both supine (Fig. 3) and prone positions (not shown). This finding suggested adherence of nerve roots to each other and to the arachnoid membrane. No mass or herniated disc impinging on the cauda equina was noted. The patient was treated with nonsteroidal anti-inflammatory agents for about 3 months without significant improvement. Reprint requests to: Dr. Yang-Kai Fan Department of Radiology, Mackay Memorial Hospital. No. 92, Sec. 2, Chung Shan N. Road, Taipei 104, Taiwan, R.O.C. MRI Appearance of Lumbosacral Spine in a Patient with Ankylosing Spondylitis and Cauda Equina Syndrome YANG-KAI FAN SHO-J EN CHENG J ON-KWAY HUANG Department of Radiology, Mackay Memorial Hospital

MRI Appearance of Lumbosacral Spine in a Patient …...DISCUSSION This patient was thought to have cauda equina syndrome, but the MRI study did not disclose a com-pressive lesion impinging

  • Upload
    others

  • View
    0

  • Download
    0

Embed Size (px)

Citation preview

Page 1: MRI Appearance of Lumbosacral Spine in a Patient …...DISCUSSION This patient was thought to have cauda equina syndrome, but the MRI study did not disclose a com-pressive lesion impinging

中華放射醫誌 Chin J Radiol 2004; 29: 217-221 217

Cauda equina syndrome is an unusual complica-tion of ankylosing spondylitis. We present a case of a68-year-old woman with long-standing ankylosingspondylitis who developed cauda equina syndrome.MRI revealed a characteristic widening of the lum-bosacral thecal sac and numerous dorsal thecaldiverticula. MRI is a powerful, noninvasive tool toconfirm the diagnosis and exclude a treatable com-pressive lesion and provides a good alternative tomyelography, an invasive and potentially dangerousprocedure in such patients.

Key words: Ankylosing spondylitis;Arachnoiditis; Cauda equina syndrome; Magneticresonance imaging

Cauda equina syndrome is a rare, late neurolog-ical complication of long-standing ankylosingspondylitis. Widening of the thecal sac and numerousdorsal thecal diverticula in the lumbosacral region areusually seen in such patients. It is important to distin-guish this disorder from an intraspinal compressivelesion. The MRI findings are characteristic, and famil-iarity with them is helpful in making a correctdiagnosis.

CASE REPORT

A 68-year-old woman had ankylosing spondylitisfor many years. For several months, she complained oflow back pain radiating to the left leg, numbness of theright leg, and intermittent claudication. There was alsoslight weakness in both legs. The patient deniedproblems with urination or defecation. Neurologicalexamination revealed hyperreflexia in both legs andhypesthesia to pinprick over the saddle region (S3-S5dermatomes).

A KUB and a lateral lumbar spine film showedtypical changes of ankylosing spondylitis with fusionof both sacroiliac joints, squaring of the lumbarvertebral bodies, and syndesmophyte formation alongthe lumbar spine (Fig. 1). MRI examination at 1.5Tdemonstrated a wide thecal sac from L1 to S2, withextensive scalloping of the pedicles, laminae, andspinous processes of several vertebrae caused bynumerous dorsal thecal diverticula (Fig. 2). Axial T2-weighted images (T2WI) showed clumping of thenerve roots of the cauda equina on the side of theectatic thecal sac in both supine (Fig. 3) and pronepositions (not shown). This finding suggestedadherence of nerve roots to each other and to thearachnoid membrane. No mass or herniated discimpinging on the cauda equina was noted. The patientwas treated with nonsteroidal anti-inflammatory agentsfor about 3 months without significant improvement.

Reprint requests to: Dr. Yang-Kai FanDepartment of Radiology, Mackay Memorial Hospital. No. 92, Sec. 2, Chung Shan N. Road, Taipei 104, Taiwan,R.O.C.

MRI Appearance of Lumbosacral Spine in aPatient with Ankylosing Spondylitis andCauda Equina SyndromeYANG-KAI FAN SHO-JEN CHENG JON-KWAY HUANG

Department of Radiology, Mackay Memorial Hospital

Page 2: MRI Appearance of Lumbosacral Spine in a Patient …...DISCUSSION This patient was thought to have cauda equina syndrome, but the MRI study did not disclose a com-pressive lesion impinging

DISCUSSION

This patient was thought to have cauda equinasyndrome, but the MRI study did not disclose a com-pressive lesion impinging on the cauda equina.Instead, an ectatic dural sac and several dorsal thecaldiverticula were seen. Clumping of nerve roots of thecauda equina on the side of the ectatic thecal sac was

also found, similar to that seen in arachnoiditis. The exact pathogenesis of cauda equina syn-

drome in long-standing ankylosing spondylitis is notfully understood. Based on previously reported opera-tive and pathological findings in such patients, theprocess is thought to be due to arachnoiditis. Theoperative findings in a patient reported by Hauge wereof a large thecal sac, thinning of the posterior ele-ments, absence of peridural tissues, and atrophic sacralnerve roots adherent to a thickened arachnoid, which

Cauda equina syndrome complicating ankylosing spondylitis218

Figure 1. KUB a. and lateral lumbarspine b. films demonstrating fusionof both sacroiliac joints, squaring oflumbar vertebral bodies, andsyndesmophyte formation along thelumbar spine which is typical forankylosing spondylitis.

1a 1b

Figure 2. Sagittal T2-weighted fast spin-echo MR images(TR/TE = 2500/110) demonstrating a wide thecal sacfrom L1 to S2, numerous dorsal thecal diverticula, andextensive scalloping of the pedicles, laminae, and spinousprocesses of several vertebrae. There is no herniated discor tumor impinging on the cauda equina.

Figure 3. Axial T2-weighted fast spin-echo MR images(TR/TE= 2566/106) in supine position at different levelsshowing clumping of nerve roots of the cauda equina onthe side of the ectatic thecal sac and dorsal thecaldiverticula with erosion of the posterior elements. Theprone images (not shown) had the same appearance. Theconus medullaris seems to be adherent to the posteriorwall of the thecal sac at L1 level.

Page 3: MRI Appearance of Lumbosacral Spine in a Patient …...DISCUSSION This patient was thought to have cauda equina syndrome, but the MRI study did not disclose a com-pressive lesion impinging

was in turn adherent to the dura [1]. Matthews’sreported postmortem findings of numerous arachnoiddiverticula extending posteriorly into erosions of thelaminae and spinous processes. The dura and arach-noid were not inflamed or thickened, but some of thefree-lying roots of the cauda equina had fibrosis andloss of myelin [2]. Bartleson et al. noted similar find-ings at operation, namely atrophy of the peridural tis-sue, adherence of the dura to the surrounding perios-teum and ligaments, small nerve roots exiting throughforamina of ample size, and no active inflammation ofthe arachnoid [3]. Although no active dural or arach-noid inflammation could be seen on the histologicexamination, Matthews suggested that arachnoiditiswas the original insult but subsequently became inac-tive [2]. The imaging findings on myelography [4] andMRI are similar to those seen in arachnoiditis, whichsupports this hypothesis. The initial inflammation inthe ligaments may lead to adjacent meningeal inflam-mation and arachnoiditis, with subsequent nerve rootinflammation, degeneration, fibrosis, adhesion, andtethering, all resulting in cauda equina syndrome. Theneurologic deficits seen in these patients cannot beexplained by the dorsal thecal diverticula, which arelocated posterior to the intervertebral foramina and donot compress the exiting nerve roots [3].

Matthews also suggested that the dorsal thecaldiverticula and bony erosions result from CSF pumpedby arterial pulsation. Under normal conditions, themeninges and especially the thecal sac expandpromptly in response to increased CSF pressure,allowing absorption of CSF and dampening of trans-mitted pressure variations. Ankylosing spondylitiscauses atrophy of the peridural tissues and adherenceof the dura to adjacent structures, thereby reducing theelasticity and compliance of the thecal sac. This inturn would impair the sacs ability to dampen briefCSF pressure fluctuations, chiefly pulse pressure.Over a course of years, excessive pulse pressure couldcause slowly enlarging thecal diverticula andsecondary erosion of the dorsal bony elements of thelower spine.

Although the spine is usually extensivelyinvolved in ankylosing spondylitis, neurologic compli-cations are uncommon. Cauda equina syndrome is arare, late complication of long-standing ankylosingspondylitis, with an average age of onset of 57 years(39 to 70 years). The interval from onset of ankylosingspondylitis to cauda equina syndrome averages 35years (17 to 53 years) [3]. Symptoms of cauda equinasyndrome in patients with ankylosing spondylitis haveprompted evaluation for a possible compressivelesion. Interestingly, there is usually none found.

We suggest that MRI is the investigation ofchoice in these patients because it is noninvasive canclearly demonstrates characteristic findings, and thusexcludes a treatable compressive lesion. Myelographyor computed tomographic (CT) myelography shouldbe avoided if possible. Lumbar puncture in such acondition can be technically difficult and hazardous,owing to bony ankylosis and ossification of the spinalligaments [3, 5]. Dural ectasia is also seen in Marfan’ssyndrome, Ehlers-Danlos syndrome, and neurofibro-matosis, but the bone erosions in ankylosingspondylitis predominantly involve the posteriorelements rather than posterior aspect of the vertebralbodies, as typically seen in these other conditions [6].

The cauda equina syndrome in long-standingankylosing spondylitis is progressive in most reportedcases. In Bartleson’s study, 13 of 14 patients had avery slow but progressive course [3]. No treatment isknown to be effective in this condition. No clinicalimprovement has been reported after treatment withcorticosteroids or nonsteroidal anti-inflammatoryagents [3, 4, 7]. Given the apparent fibrosis andtethering of nerve roots, surgical intervention alsoseems to play no role in the treatment for suchpatients. Although decompressive laminectomy hasbeen reported in a small number of patients, almostnone benefited from the procedure [1, 3, 8, 9, 10, 11,12, 13]. Shaw reported one patient who was success-fully treated in this way. He described some compres-sion of nerve roots, perhaps by arachnoid cysts andattributed the success to early exploration and decom-pression of the arachnoid diverticula [13].

MRI is thus a useful, noninvasive diagnosticmethod for evaluating cauda equina syndrome inpatients with ankylosing spondylitis. If it reveals acompressive lesion, prompt corrective action could beundertaken. However, if the findings are consistentwith most cases reported to date, an appropriatediagnosis and prognosis can be determined withoutsubjecting the patient to invasive diagnostic or thera-peutic procedures. ◆

REFERENCES

1. Hauge T. Chronic rheumatoid polyarthritis and spondy-larthritis associated with neurological symptoms andsigns occasionally simulating an intraspinal expansiveprocess. Acta Chir Scand 1961; 120: 395-401

2. Matthews WB. The neurological complications of anky-losing spondylitis. J Neurol Sci 1968; 6: 561-573

3. Bartleson JD, Cohen MD, Harrington TM, et al. Caudaequina syndrome secondary to long-standing ankylosingspondylitis. Ann Neurol 1983; 14: 662-669

4. Mitchell MJ, Sartoris DJ, Moody D et al. Cauda equinasyndrome complicating ankylosing spondylitis.

Cauda equina syndrome complicating ankylosing spondylitis 219

Page 4: MRI Appearance of Lumbosacral Spine in a Patient …...DISCUSSION This patient was thought to have cauda equina syndrome, but the MRI study did not disclose a com-pressive lesion impinging

Radiology 1990; 175: 521-5255. Young A, Dixon A, Getty J, et al. Cauda equina syn-

drome complicating ankylosing spondylitis: use of elec-tromyography and computerized tomography in diagno-sis (case report). Ann Rheum Dis 1981; 40: 317-322

6. Westmark KD. Weissman BN. Complications of axialarthropathies. Orthop Clin North Am 1990; 21: 423-435

7. Russell ML, Gordon DA, Ogryzlo AM, et al. The caudaequina syndrome of ankylosing spondylitis. Ann InternMed 1973; 78: 551-554

8. Auquier L, Siaud JR, Guiot G, et al. Syndrome de laqueue de cheval au cours de la spondylarthrite anky-losante: dé-couverte opératoire dène fusion entrevertèbre et dure-mére dans la cas rapporte. Rev RhumMal Osteoartic 1974; 41: 733-737

9. Lee MLH, Waters DJ. Neurological complications ofankylosing spondylitis. Br Med J 1982; 1: 798

10. Gordon AL, Yudell A. Cauda equina lesion associatedwith rheumatoid spondylitis. Ann Intern Med 1973; 78:555-557

11. Soeur M. Monseu G. Baleriaux-Waha D. Duchateau M.Williame E. Pasteels JL. Cauda equina syndrome inankylosing spondylitis. Anatomical, diagnostic, andtherapeutic considerations. Acta Neurochir 1981; 55:303-315

12. Byrne E, McNeill P, Gilford E, et al. Intradural cystwith compression of the cauda equina in ankylosingspondylitis. Surg Neurol 1985; 23: 162-164

13. Shaw PJ, Allcutt DA, Bates D, et al. Cauda equina syn-drome associated with multiple lumbar arachnoid cystsin ankylosing spondylitis: improvement following sur-gical therapy. J Neurol Neurosurg Psychiatry 1990; 53:1076-1079

Cauda equina syndrome complicating ankylosing spondylitis220

Page 5: MRI Appearance of Lumbosacral Spine in a Patient …...DISCUSSION This patient was thought to have cauda equina syndrome, but the MRI study did not disclose a com-pressive lesion impinging

Cauda equina syndrome complicating ankylosing spondylitis 221

長期僵直性脊椎炎病患併發馬尾症候群在腰薦椎之磁振造影表現:病例報告

范揚凱 鄭碩仁 黃榮貴

台北馬偕紀念醫院 放射線部

馬尾症候群是僵直性脊椎炎病患一個少見之併發症,我們在此提出一病例報告:一名68歲

女性病患患有長期之僵直性脊椎炎併發馬尾症候群,其腰薦椎之磁振造影顯示具特徵性之硬膜

囊擴張及許多的背側硬膜憩室。磁振造影為一強而有力又不具侵犯性之工具,不僅可以幫助確

立診斷並且可以排除可治療之壓迫性病兆。脊髓腔X光攝影術對於此種病患可能難以施行,並

且可能造成病人的傷害,故應避免。

關鍵詞:僵直性脊椎炎,蜘蛛膜炎,馬尾症候群,磁振造影