Available online at www.sciencedirect.com
Surgical Neurology 71 (2009) 357 – 361 www.surgicalneurology-online.com
Spine
Delayed pseudomyelomeningocele: a rare complication after foramen magnum decompression for Chiari malformation Deniz Belen, MD⁎, Uygur Er, MD, Levent Gurses, MD, Kazim Yigitkanli, MD Ministry of Health, Diskapi Educational and Research Hospital Neurosurgery Department, Ankara 06110, Turkey Received 26 February 2007; accepted 9 August 2007
Abstract
Background: A variety of complications after FMD with or without duraplasty for Chiari malformation have been described. Although cerebellar ptosis through the dural opening and pseudomeningocele are well-known rare complications of this procedure, spinal cord herniation manifesting as pseudomyelomeningocele formation has not previously been cited in the literature. In this report, we present a case of delayed pseudomyleomeningocele after FMD. Case Description: A 22-year-old man presented with progressive spinal cord compression symptoms 7 years after undergoing FMD for Chiari malformation. Craniocervical MRI of the patient revealed pseudomyelomeningocele at C1 level together with cord distortion. The patient underwent untethering surgery for spinal cord adhesions; neural tissue was freed microsurgically from the adjacent structures and duraplasty was performed with synthetic graft. In the early postoperative period, his symptoms resolved moderately. Conclusion: Neural tissue shift including the spinal cord through the dural opening may occur after a large posterior fossa decompression without duraplasty for Chiari malformation. Postoperative inadequate CSF circulation due to arachnoid scarring around the decompression area may facilitate this type condition. Early diagnosis of such a complication is of vital importance to prevent serious neurologic deficits, and release surgery with proper duraplasty and also restoration of CSF circulation is the choice of treatment. © 2009 Elsevier Inc. All rights reserved.
Keywords:
Arachnoiditis; Chiari malformation; Foramen magnum; Pseudomyelomeningocele; Spinal cord; Syringomyelia
1. Introduction Chiari malformation is a rare abnormality encountered predominantly in young people. Patients may present with symptoms secondary to compression of brain stem at the level of the foramen magnum, hydrocephalus, spinal cord compression due to syringomyelic cavity, or transient elevations of intracranial pressure [1]. The natural history
Abbreviations: CSF, cerebrospinal fluid; FMD, foramen magnum decompression; MRI, magnetic resonance imaging. ⁎ Corresponding author. 60 Sokak 16-2, Emek 06510, Ankara 06510, Turkey. Fax: +90 312 223 6975. E-mail address:
[email protected] (D. Belen). 0090-3019/$ – see front matter © 2009 Elsevier Inc. All rights reserved. doi:10.1016/j.surneu.2007.08.031
of the disease is not known with certainty. Minority of patients may remain stable for years with intermittent periods of deterioration; however, the remaining greater part becomes symptomatic at any age [10]. Based on the fact that patients respond best when operated on within 2 years of the onset of symptoms, early surgery is recommended for symptomatic patients. The most frequently performed operation is posterior fossa decompression, with or without other procedures. Some surgeons prefer to open the dura to allow a more complete and immediate enlargement of the subarachnoid space; however, opinions differ regarding watertight dural closure after decompression [2,4,8,11]. Some authors advocate leaving the dura open with the arachnoid either intact or open [16]. Various complications after posterior fossa decompression for Chiari malformation
358
D. Belen et al. / Surgical Neurology 71 (2009) 357–361
malformation. To our knowledge, this is the first case that exhibited such a delayed postoperative high cervical spinal cord herniation. 2. Case report 2.1. Presentation and history
Fig. 1. A: Sagittal T1-weighted MRI scan of the craniocervical region a year later after FMD shows a large pseudomeningocele formation (asterisk) and spinal cord tethering at C2 level (arrow) in addition to increased anterior subarachnoid space. Syringomyelic cavity in the cord persisted. B: Axial T2weighted MR image of cervical spine obtained at the same time demonstrating the large pseudomeningocele at C1 level.
have been described previously; among them, cerebellar ptosis with intact arachnoid and pseudomeningocele through the dural opening are well known [1,3,9,13]. We present in this report an extremely rare complication that manifested as delayed pseudomyelomeningocele after FMD for Chiari
A 22-year-old man initially presented at the age of 16 years with a 10-month history of bilateral progressive diminished hand functions and pain in the suboccipital region. At the time of presentation, his neurologic examination revealed distal upper extremity weakness (muscle strengths were 3/5 in the right and 4/5 in the left), intrinsic muscle atrophy in both hands, and incomplete sensory deficit in the trunk and hands. Magnetic resonance imaging studies demonstrated 24 mm of cerebellar tonsillar herniation below the foramen magnum and a large syringomyelic cavity extending from C1 to mid-thoracic level. The diagnosis of Chiari malformation associated with syringomyelia was made, and the patient underwent FMD in addition to C1 laminectomy. Suboccipital craniectomy was performed 3 cm in width and 3 cm in length from the edge of the foramen magnum. After opening the dura, cerebellar tonsils were reduced in size by bipolar coagulation and arachnoid bands were removed microsurgically. The edge of the retracted arachnoid was stitched to the dura to prevent arachnoid adhesion and the dura was left open. There was no patency of the central canal at the obex. After meticulous hemostasis, the anatomic layers were closed in watertight fashion. On the third postoperative day, CSF leak was observed at the caudal margin of skin incision and lumbar drainage was used with antibiotic regimen. Cerebrospinal fluid analysis for evaluating meningitis showed no abnormality. The leakage was successfully treated with this protocol and the patient was discharged from the hospital on 10th postoperative day in pain-free condition. During the first year of follow-up period, his sensorial deficit improved to a nondisturbing state; however, the patient's motor deficit and muscle atrophy remained unchanged. An early radiologic examination of craniocervical region was not available. Magnetic resonance imaging of the cervical region obtained 1 year later showed a large suboccipital subcutaneous CSF collection that was concomitant with artificial cisterna magna (pseudomeningocele formation) and also a local cervical cord adhesion to the posterior surface of the spinal canal at C2 level (Fig. 1A and B). The upper cervical cord was retracted posteriorly so that anterior subarachnoid space at this level was increased and cervical syringomyelic cavity persisted. The patient refused reoperation for a cord-releasing procedure and he was lost to follow-up for more than 4 years. 2.2. Examination In 2006, 7 years after the initial diagnosis and operation, the patient was admitted to the hospital with the complaints of worsened hand function in both upper extremities within
D. Belen et al. / Surgical Neurology 71 (2009) 357–361
359
phalangeal joints and flexion contracture at the interphalyngeal joints, which was particularly progressive in the right hand. Marked extensor and flexor weakness of the both forearms was observed. Spasticity and hyperreflexia were present in all extremities. His Romberg test result was positive and cerebellar test results were normal. Magnetic resonance imaging of the craniocervical region revealed posterior herniation of the spinal cord at C1 level, which obliterated the artificial cisterna magna inferiorly and caused pseudomyelomeningocele formation. Marked cervical syringomyelic cavity was evident, and partial adhesion of posterior surface of the herniated neural tissue to the adjacent structures was noticed (Fig. 2A and B). In view of rapid progression of symptoms, the presence of new neurologic deficits, and the evolution of the spinal cord herniation, a decision was made to offer surgical treatment with the aim of releasing the neural tissue from surrounding structures in addition to duraplasty.
Fig. 2. Magnetic resonance imaging scans of the same patient shown in Fig. 1, obtained 7 years after FMD procedure. These T1-weighted images in the sagittal (A) and T2-weighted axial (B) planes show the formation of pseudomyelomeningocele in the artificial cisterna magna. The cervical cord is distorted in posterior direction and attached to the muscular layer broadly.
3 months. He reported moderate dissociated sensorial deficit and burning dysesthesia in the arms and in the trunk. Loss of temperature sensation was found in the hands. The patient exhibited bilateral extension contracture at the metacarpo-
Fig. 3. T1-weighted MRI scan of the craniocervical region obtained 7 months after the untethering procedure demonstrates normal position of the cervical spinal cord. The dorsal CSF flow void over restoration area was observable (arrowheads). Syringomyelic cavity was reduced in size. Note the metal artifact in the epidural space that caused distortion of the image at C3 level.
360
D. Belen et al. / Surgical Neurology 71 (2009) 357–361
2.3. Operation The patient underwent exploration of the posterior fossa in prone position. The dissection of the subcutaneous and muscular layers was initiated from the cephalic corner of artificial cistern and advanced down to the herniated neural tissue. The spinal cord was firmly attached to the muscular layer and the dura edges. It was freed microsurgically from the surrounding tissues using sharp dissection. Bipolar coagulation was performed minimally. After removing the arachnoid and pial adhesions completely, the spinal cord moved freely in the anterior direction and eventually its pulsation started. Subsequently, duraplasty was performed with polytetrafluoroethylene (Gore-Tex; Flagstaff, AZ, USA) sheet. 2.4. Postoperative course The patient showed partial improvement in symptoms postoperatively. In the early postoperative period, the patient exhibited myoclonus in the right hand that spontaneously resolved within a week. On the 15th postoperative day, the patient was readmitted to the hospital with severe headache and deterioration of consciousness. On examination, bilateral papilledema was found and his Glasgow coma scale score was 14. Cranial computed tomography revealed enlarged lateral ventricles with hydrocephalus index of 40%; the fourth ventricle was in normal size. Occurrence of early postoperative hydrocephalus after untethering surgery, however, was considered noteworthy for an inadequate CSF circulation in the cervical spinal canal; normal-sized fourth ventricle indicated an obstruction at the aqueductus sylvii level probably as a result of blood products. The patient underwent ventriculoperitoneal shunt procedure; the CSF pressure was assessed intraoperatively as 240 mm H2O. After surgery, patient's awareness returned to normal level and his headache disappeared. Follow-up MRI obtained 7 months postoperatively showed normal position of the cord at C1 through C2 level and CSF flow over dorsal C1 level spinal cord with reduced syringomyelic cavity (Fig. 3). 3. Discussion Recurrence of the symptoms of Chiari malformation after initial surgery, especially after FMD, is a well-known condition that may be induced by craniocervical instability, cerebellar ptosis, arachnoid scarring, and pseudomeningocele formation [13]. Most of these complications may induce cervical cord tethering, a significant cause for failed surgery. The symptoms may manifest in a range of 1 to 10 years and the pseudomeningocele formation appears typically several years after an oversized craniectomy [6,14]. Although the literature regarding these types of complications after FMD is scarce, some authors have meticulously studied underlying mechanisms and have recommended preventive measures [1,6,12,15,18].
Posterior fossa decompression procedure for the management of Chiari malformation still remains as a nonstandardized method regarding particularly dura opening and the craniectomy size. While some authors prefer dura opening with or without duraplasty, others advise leaving the dura intact. In the last decade, the so-called simple FMD was described, which means removing the tight dural epidural band and the outer layer of the dura of the posterior fossa instead of opening the dura mater [4,5,17]. The results of FMD with removal of the outer layer of the dura were as good as the results of conventional FMD with duraplasty [5]. More recently, Gamberdella et al [2] described transverse microincisions of the outer layer of the dura mater with similar results. Foramen magnum decompression with removal of the outer layer of the dura seems to be one method of preventing delayed postoperative tethering and also pseudomeningocele. Oversized craniectomy is a further crucial concern in the FMD procedure. Klekamp et al [6] recommended a craniectomy limited to the width of the spinal canal and not extended further more than 2 cm from the rim of the foramen magnum to prevent cerebellar herniation. Currently, we prefer the portrayed smaller craniectomy with removal of the outer layer of the dura mater. If dura opening with duraplasty is aimed, we favor devitalized or artificial graft material. Spinal cord tethering after FMD may also occur secondary to insufficient hemostasis, obex plugging with muscle, leaving the arachnoid edges unstitched in the subarachnoid space, autologous graft material that show a greater tendency for arachnoid scarring, suture materials, or a large pseudomeningocele [1,6]. Multiple factors may have played a role in the development of pseudomyelomeningocele formation in the present case. An oversized craniectomy + C1 laminectomy without duraplasty facilitated cervical cord herniation through the dural opening. Besides, relatively early growth of a large pseudomeningocele and development of arachnoid scarring that obliterated free CSF flow over the dorsal surface of the spinal cord introduced intensified systolic CSF pressure waves on the ventral cervical cord. With time, the pathology was progressed by the continuous CSF pressure that forced the cervical cord against the posterior wall and eventually neural tissue came in contact with muscular layer making the condition more complicated. Radiologic examinations and the dense fibrosis between the cord and muscular layer evidently proved the progressive nature of this pseudomyelomeningocele formation. Although development of spinal cord tethering after FMD causes significant neurologic deficit, early diagnosis is crucial. An epidural CSF collection needs to be followed with further MR scans [6]. It has been reported that the absence of CSF signals dorsal to the cord was diagnostic of tethered cord [14]. Careful examination of MR images, myelography, or postmyelography CT studies will reveal an abrupt change in spinal cord diameter, an irregularly defined
D. Belen et al. / Surgical Neurology 71 (2009) 357–361
spinal cord, a spinal cord adhesion, impaired passage of contrast material, septations in the subarachnoid space, or abnormal flow-void signals at a level adjacent to or along the syrinx, and thus indicate the underlying arachnoid pathology [7]. Patients with arachnoid scarring after FMD are candidates for surgery once neurologic symptoms start to progress [1,6,7]. For arachnoid scarring, a sharp dissection should be performed with microsurgical techniques aiming at free CSF flow and untethering of the spinal cord. Care should be taken especially for cases that developed arachnoid scarring after spinal meningitis. In these patients, severe arachnoid changes over several spinal segments result, which are difficult to dissect without causing morbidity or postoperative re-tethering [7]. 4. Conclusions Spinal cord tethering after FMD procedure for Chiari malformation is a serious complication. It is a progressive pathology that can lead to severe neurologic deficits. When dura is left open, although extremely rare, the lesion may progress to a pseudomyelomeningocele formation making the treatment more complicated. The aim in FMD should be prevention of such a condition by using a limited suboccipital craniectomy and leaving the dura mater intact. If the dura mater is opened, devitalized or synthetic graft material should be used for duraplasty. The realistic goal of surgical treatment of this pathology is stabilization of the patient's neurologic status by meticulous untethering of the spinal cord and restoration of CSF flow. References [1] Batzdorf U. Complications and deterioration following surgery. In: Anson JA, Benzel EC, Awad IA, editors. Syringomyelia and the Chiari Malformations. Park Ridge: AANS; 1997. p. 159-65. [2] Gambardella G, Caruso G, Caffo M, et al. Transverse microincisions of the outer layer of the dura mater combined with foramen magnum decompression as treatment for syringomyelia with Chiari I malformation. Acta Neurochir (Wien) 1998;140:134-9. [3] Guyotat J, Bret P, Jouanneau E, et al. Syringomyelia associated with type I Chiari malformation. A 21-year retrospective study on 75 cases treated by foramen magnum decompression with a special emphasis on the value of tonsils resection. Acta Neurochir (Wien) 1998;140:745-54. [4] Hida K, Iwasaki Y, Koyangi I, et al. Surgical indication and results of foramen magnum decompression versus syringosubarachnoid shunting for syringomyelia associated with Chiari I malformation. Neurosurgery 1995;37:673-9. [5] Isu T, Sasaki H, Takamura H, et al. Foramen magnum decompression with removal of the outer layer of the dura as treatment for syringomyelia occurring with Chiari malformation (type I). Neurosurgery 1993;33:845-50. [6] Klekamp J, Batzdorf U, Samii M, et al. The surgical treatment of Chiari I malformation. Acta Neurochir (Wien) 1996;138:788-801. [7] Klekamp J, Iaconetta G, Batzdorf U, et al. Syringomyelia associated with foramen magnum arachnoiditis. J Neurosurg 2002;97(3 Suppl): 317-22.
361
[8] Krieger MD, McComb JG, Levy ML. Toward a simpler surgical management of Chiari I malformation in a pediatric population. Pediatr Neurosurg 1999;30:113-21. [9] Lazereff JA, Galarza M, Gravori T, et al. Tonsillectomy without craniectomy for the management of infantile Chiari I malformation. J Neurosurg 2002;97:1018-22. [10] Moriwaka F, Tashiro K, Tachibana S, et al. Epidemiology of syringomyelia in Japan: the nationwide survey [in Japanese]. Rinsho Shinkeigaku 1995;35:1395-7. [11] Munshi I, Frim D, Stine-Reyes R, et al. Effects of posterior fossa decompression with and without duraplasty on Chiari malformation– associated hydromyelia. Neurosurgery 2000;46:1384-90. [12] Pare LS, Batzdorf U. Syringomyelia persistence after Chiari decompression as a result of pseudomeningocele formation: implications for syrinx pathogenesis: report of three cases. Neurosurgery 1998;43: 945-8. [13] Pařízek J, Měřička P, Němeček S, et al. Posterior cranial fossa surgery in 454 children: comparison of results obtained in pre-CT and CT era and after various types of management of dura mater. Child's Nerv Syst 1998;14:426-9. [14] Smith KA, Rekate HL. Delayed postoperative tethering of the cervical spinal cord. J Neurosurg 1994;81:196-201. [15] Takahashi Y, Tajima Y, Ueno S, et al. Syringobulbia caused by delayed postoperative tethering of the cervical spinal cord: delayed complication of foramen magnum decompression for Chiari malformation. Acta Neurochir (Wien) 1999;141:969-73. [16] Williams B. Surgery for hindbrain related syringomyelia. Adv Tech Stand Neurosurg 1993;20:107-64. [17] Yundt KD, Park TS, Tantsumaya VS, et al. Posterior fossa decompression without duraplasty in infants and young children for treatment of Chiari malformation and achondroplasia. Pediatr Neurosurg 1996;22:221-6. [18] Zide B, Constantini S, Epstein FJ. Prevention of recurrent tethered spinal cord. Pediatr Neurosurg 1995;22:111-4.
Commentary I am among those surgeons who prefer to open the dura mater and meticulously patch it after obsessively preventing any spillage of blood, usually from the dural sinuses, into the subdural space. I believe that an extensive craniectomy is responsible for more postoperative complications than duraplasty. Nonetheless, I understand how tempting it is to just release the dura band. I would recommend to confirm, before and after releasing the dural band, with operative ultrasound that the cerebellar tonsils are pulsating before you decide that the objectives of the procedure have been achieved. I commend the authors for this well-written article in which they ably avoided pontificating over what we now know is a surgical error. This article should generate an extensive correspondence between Surgical Neurology readers. Jorge A. Lazareff, MD Department of Neurosurgery University of California Los Angeles (UCLA) Los Angeles, CA 90095, USA