Journal of Clinical Neuroscience 19 (2012) 557–560

Contents lists available at SciVerse ScienceDirect

Journal of Clinical Neuroscience

journal homepage: www.elsevier .com/ locate/ jocn

Clinical Study

Surgical results of arachnoid-preserving posterior fossa decompressionfor Chiari I malformation with associated syringomyelia

Hyun Seok Lee, Sun-Ho Lee ⇑, Eun Sang Kim, Jong-Soo Kim, Jung-Il Lee, Hyung Jin Shin, Whan EohDepartment of Neurosurgery, Samsung Medical Center, Sungkyunkwan University, School of Medicine, 50 Ilwon-dong, Gangnam-gu, Seoul 135-710, South Korea

a r t i c l e i n f o a b s t r a c t

Article history:Received 2 May 2011Accepted 26 June 2011

Keywords:ArachnoidChiari I malformationPosterior fossa decompressionSyringomyelia

0967-5868/$ - see front matter � 2011 Elsevier Ltd. Adoi:10.1016/j.jocn.2011.06.034

⇑ Corresponding author. Tel.: +82 2 3410 3491; faxE-mail address: sobotta72@hotmail.com (S.-H. Lee

We analyzed the outcome of posterior fossa decompression accompanied by widening of the cisternamagna, without disturbing the arachnoid, in patients with Chiari I malformation (CMI) associated withsyringomyelia. Twenty-five adult patients with CMI and syringomyelia, who underwent surgery betweenOctober 2000 and December 2008, were enrolled in this study. All patients underwent foramen magnumdecompression with C1 decompression, with or without C2 decompression. Three surgeons performed adura opening with duraplasty in 20 patients, and another surgeon excised the outer layer of the durawithout duraplasty in five patients. Clinical and radiological assessments were performed preoperativelyand during the follow-up period. After surgery, 20 (80%) patients achieved a significant improvement intheir clinical symptoms. However, four patients (16%) achieved only a stable state, and one patient’ssymptoms worsened. Radiological analysis showed that 17 patients (68%) had a favorable result; thatis, a total collapse, or a marked reduction, of the syrinx. Seven patients (28%) were stable in terms of syr-inx size. However, the syrinx enlarged in one patient who had undergone excision of the outer dura.Twenty-four patients achieved a widened cisterna magna with ascent of the cerebellar tonsils into theposterior fossa and acquisition of a more rounded shape. Postoperative complications included a tran-sient headache and vomiting in three patients and transient motor weakness in one patient. Two patientsdeveloped a superficial wound infection. This study shows that arachnoid-preserving posterior fossadecompression is a safe and effective treatment for patients with CMI with associated syringomyelia.

� 2011 Elsevier Ltd. All rights reserved.

1. Introduction

Chiari I malformation (CMI) is a cerebellar tonsillar herniationthrough the foramen magnum associated with an abnormal pos-terior fossa constitution.1–4 This condition may result in syringo-myelia due to an obstruction of the flow of cerebrospinal fluid(CSF) at the craniocervical junction.2,5,6 Although a number of sur-gical techniques have been described to restore the CSF circulationand decompress the neuraxis, no consensus has been reached as tothe optimal technique.5,6 Usually, surgery consists of a posteriordecompressive craniectomy of the cervico-occipital junction asso-ciated with duraplasty, arachnoid opening and, sometimes, tonsil-lectomy. However, a number of surgical adjuvants to standardbony decompression remain controversial. Furthermore, some dis-advantages, such as a high rate of CSF fistulas, aseptic meningitis,and long hospitalization periods, have been reported in patientswho have undergone tonsillectomy and arachnoid dissection.7–9

To minimise the risks related to handling of the arachnoidmembrane and tonsillectomy, we performed decompressive

ll rights reserved.

: +82 2 3410 0048.).

craniectomy and arachnoid-preserving duraplasty consecutivelyin adult patients with CMI and syringomyelia. This procedure re-sulted in decompression and shrinkage of the syringomyelia andascension of the cerebellar tonsils. We report the functional resultsachieved by adult patients with symptomatic CMI and syringomy-elia using arachnoid-preserving posterior fossa decompression andshow its efficiency using radiography.

2. Materials and methods

2.1. Patient characteristics

Forty-three consecutive patients with CMI with associatedsyringomyelia underwent arachnoid-preserving foramen magnumdecompression with excision of the outer dura or duraplasty at ourinstitution between October 2000 and December 2008. Medicalcharts, radiographical features and outcomes determined by pa-tient interviews were retrospectively assessed. Inclusion criteriawere: (i) a diagnosis of CMI with syringomyelia; (ii) an age>18 years at the time of surgery; (iii) a minimum of two years ofpostoperative follow-up; and (iv) at least one preoperative andpostoperative MRI. Patients with a tumor, a craniocervical bony

558 H.S. Lee et al. / Journal of Clinical Neuroscience 19 (2012) 557–560

anomaly, hydrocephalus, a history of meningitis, or history of pre-vious posterior fossa decompression or shunt surgery were ex-cluded. Twenty-five patients who met the inclusion criteria wereenrolled in this study.

2.2. Surgical technique

Patients underwent surgery in the prone position with the headfixed in a Mayfield head holder. A midline linear skin incision wasmade from the inion to C2 or C3, depending on the extent of cere-bellar herniation. The suboccipital region below the inferior nuchalline was carefully decompressed by removing the posterior lip ofthe foramen magnum. After laminectomy of C1 and C2 (if neces-sary), the dural surface of the craniocervical junction was exam-ined under an operative microscope and bands of thickenedtissue on the outer dura were carefully removed. A constrictingband of thickened tissue on the outer surface of the dura was re-moved at the craniospinal level. Three surgeons (ESK, JSK, HJS)opened the dura and performed duraplasty (Supplementary video).Duraplasty was performed using an autologous periosteum patchor a Lyoplant (B. Braun, Melsungen AG, Melsungen, Germany). Carewas taken to ensure that the underlying arachnoid was not com-promised. Osteodural decompression was considered appropriatewhen we could observe, under a transparent arachnoid, good pul-sation of the cerebellar tonsils and a free flow of CSF at the crani-ospinal level. Another surgeon (JIL) opened the outer layer of thedura, without duraplasty, to widen the cisterna magna.

2.3. Clinical and radiological assessment

Clinical and radiological assessments were performed preoper-atively and during follow-up by two neurosurgeons not involvedwith the surgery. Preoperative assessments considered symptomduration, neurological features, and syrinx levels and diameters.Patients underwent a cerebrospinal MRI and a tonsilar herniationevaluation. After surgery, neurological findings and complicationswere assessed. Direct examinations were performed one, three,and six months after surgery, and subsequently at six-monthlyintervals. Patients underwent craniospinal MRI three months aftersurgery, and this was repeated depending on the clinical evolutionand MRI findings. MRI were used to determine the postoperativesize of the syrinx, widened cisterna magna volumes in the mid-sagittal plane, and the size of the CSF space surrounding themedulla. Syrinx improvement was defined as a decrease in themaximal syrinx diameter on MRI.

Table 1Demographic data of 25 patients who underwent extra-arachnoidal cranio-cervicaldecompression for Chiari I malformation associated with syringomyelia

No. patients

GenderMale 8Female 17

Mean age (years) ± SD (range) 36.6 ± 14.2 (18–67)Clinical symptoms

Sensory disturbance 23Motor weakness 9Neck pain 9Cranial nerve symptoms 4

Syrinx levelCervical 7Cervicothoracic 16Entire cord 2

Type of operationFMD + DO 5FMD + DO + DP 20

DO = opening of the outer layer of the dura, DP = duroplasty without arachnoidopening, FMD = foramen magnum decompression, SD = standard deviation.

3. Results

3.1. Clinical and radiological findings

The cohort was composed of 17 female and eight male patients(mean age = 36.6 years, range = 18–67 years). Follow-ups wereconducted over 24 to 76 months, and the duration of symptomsranged from two to 120 months. Sensory disturbances were thepresenting symptom in 23 patients (92%); eight patients (32%)exhibited motor weakness, and eight patients (32%) had a suboc-cipital headache. Suboccipital headaches were exacerbated by val-salva maneuvers and sudden changes in posture. Lower cranialnerve deficits were observed in four patients (16%). One patienthad a basilar impression and Klippel–Feil syndrome, which didnot interfere with the CMI assessment. The cerebellar tonsils werepositioned just below the foramen magnum in two patients, at theC1 level in six, and at the C2 level in three. Preoperative clinical andradiological features are summarized in Table 1.

3.2. Surgery and complications

Twenty patients (80%) underwent posterior cranial fossadecompression, C1 laminectomy, and duraplasty. An additionalC2 laminectomy was required in five patients (21%) who hadincomplete tonsillar decompression after C1 laminectomy. No se-vere adhesive arachnoiditis or subarachnoid scarring associatedwith CMI was observed. Accidental pinholing of the arachnoidmembrane occurred during dural opening in three patients whounderwent duraplasty. In addition, a large accidental arachnoidalrent occurred in one patient during a difficult opening of the duramater. Postoperative complications included a transient headacheand vomiting in three patients who underwent dural openingand transient motor weakness occurred in one patient. Two pa-tients experienced a superficial wound infection, but respondedto antibiotics. No complications occurred in the five patients whodid not undergo dura opening.

3.3. Follow-up outcomes

After surgery, 20 of 25 patients (80%) showed significantimprovement of their clinical symptoms. However, four patients(16%) achieved only a stabilized state of clinical symptoms andone patient’s symptoms worsened. Improvements were observedin 18 (78.3%) of 23 patients with sensory disturbances, after sur-gery (Table 2). A suboccipital headache, when present, improvedin eight (88.9%) of nine patients, motor weakness improved in four(44%) of nine patients and lower cranial nerve deficits in two offour patients (Fig. 1). Postoperative MRI showed that 17 of 25 pa-tients (68%) achieved a favorable outcome; that is, syrinxes weretotally collapsed, or syrinxes or segments were remarkably re-duced in diameter. Syrinx size stabilized in seven patients (28%).Of the 20 patients who underwent duraplasty, 14 showed a de-crease in syrinx volume and clinical improvement. Six patientsshowed symptomatic improvement without syringomyeliaimprovement. Of the five patients who did not undergo duraplasty,three had clinical improvement with a decrease in cavity size. Onepatient had symptomatic improvement with no syringomyeliaimprovement, and in one patient, who had no clinical improve-ment, the cavity size increased. A wider cisterna magna was de-tected in 24 patients and ascent of the cerebellar tonsils into theposterior fossa with acquisition of a more rounded shape occurredin 24 of 25 patients (Fig. 2).

Table 2Clinical and radiological outcomes of 25 patients who underwent extra-arachnoidalcraniocervical decompression for Chiari I malformation associated withsyringomyelia

Parameter No. patients

Clinical outcomesImproved 20Stable 4Worse 1

Radiological outcomesImproved 17Stable 7Worse 1

ComplicationsTransient headache 3Transient weakness 1Superficial wound infection 2

Fig. 1. Graph showing the presenting signs and symptoms and follow-up outcomesof 25 patients operated for Chiari I malformation with associated syringomyelia.

H.S. Lee et al. / Journal of Clinical Neuroscience 19 (2012) 557–560 559

4. Discussion

The association between CMI and syringomyelia has been welldocumented.2,7,10–14 Recent studies have attempted to explainCMI-associated syringomyelia on the basis of CSF flow obstructionat the craniocervical junction.1,4,9,11,14,15 It is generally acceptedthat the genesis of spinal cord cavitations can be attributed to a dy-namic anomaly of CSF flow.2,4,15 The signs and symptoms observedin patients with this condition are related to compression of neural

Fig. 2. (a) Preoperative sagittal T1-weighted MRI of a 21-year-old women showing acquone month after posterior fossa decompression and duraplasty showing marked shrinkagshowing a reduction in the size of the cavity and a normal intracranial cerebellar tonsil

structures by the herniated tonsils and/or by the presence of anassociated syrinx. In patients with CMI associated with syringomy-elia, surgical intervention is indicated in symptomatic patientswith neuroradiological abnormalities.12,15

Posterior fossa decompression directly relieves bony compres-sion at the craniocervical junction. However, the usefulness andsafety of additional procedures such as duraplasty, tonsilar resec-tion, syringosubarachnoid shunting, or obex plugging is debat-able.5,6,16,17 Several previous studies have suggested that foramenmagnum decompression alone is not sufficiently effective withoutduraplasty.6,16 In the most recent survey conducted, the majorityof respondents favored a treatment combination of posterior fossacraniectomy, C1 laminectomy and duraplasty, but no consensuswas reached regarding whether intradural exploration or tonsillarmanipulation was warranted. Some authors have reported thatopening the arachnoid does not guarantee improved effectiveness,and that plugging the obex in addition to the arachnoid openingdoes not have a positive effect.4,7,18–20 It has also been reportedthat complementary tonsilar resection does not offer substantialbenefits compared to foramen magnum decompression with dura-plasty; in addition, only complete opening of the dura achievedslightly better results than a simple incision in the outer dura layeralone.7,9,17,21 Some authors advocate that duraplasty is essential forthe prevention of scar formation and symptom recurrence.6,9,17

Debate exists regarding whether arachnoid dissection should beperformed once the dura is opened.5,17 Arachnoid dissection allowsdirect observation of CSF pulsatile flow as it exits the fourth ventri-cle. Furthermore, the tonsils can be directly inspected and decisionsmade regarding whether resection is required and if subarachnoidadhesions need to be released. Conversely, opening the arachnoidmembrane probably incurs risks of CSF leakage and exposes the pa-tient to the risk of symptomatic sterile meningitis, which can bevery debilitating and difficult to treat.8,9,14 Interestingly, Imaeet al., in a comparative study, described four patient groups whounderwent tonsillectomy with duraplasty, intra-arachnoidal lysiswith duraplasty, duraplasty alone, or delamination alone.8 No sig-nificant differences were found between these groups with respectto degree of syrinx reduction, but duraplasty was clearly less inva-sive than tonsillectomy or arachnoidal lysis.8,17,22

Thus, avoiding arachnoid opening achieves the same outcomeswith fewer complications. Furthermore, if duraplasty must be per-

ired tonsillar herniation and the syringomyelic cavity. (b) Sagittal T1-weighted MRIe of the intramedullary cavity. (c) Sagittal T1-weighted MRI 12 months after surgeryposition.

560 H.S. Lee et al. / Journal of Clinical Neuroscience 19 (2012) 557–560

formed, it is strongly advisable to leave the arachnoid intact. In thisstudy, opening the CSF pathways did not require either resection orcoagulation of the cerebellar tonsils. Additionally, comparison ofpre- and postoperative T1-weighted MRI showed that expansionof the CSF pathways and posterior fossa resulted in a normal cere-bellar morphology with ascent of the cerebellar tonsils and acqui-sition of a rounded, rather than pointed, shape. The tendency of thecerebellum to return to its normal morphology after surgicalenlargement of the posterior fossa supports the contention thatCMI is a secondary deformation that arises from a small posteriorfossa and not a primary abnormality of the cerebellum.

Our experience indicates that preservation of the arachnoidallayer reduces the risk of postoperative complications, such as CSFcollection in the operative wound, hydrocephalus, aseptic menin-gitis, arachnoiditis, and pseudomeningocele. Recently, there havebeen a small number of reports of infratentorial subdural extra-arachnoid fluid collection associated with attempted arachnoid-preserving duraplasty in patients with a pinhole arachnoidal in-jury.16,19 These studies suggest that arachnoidal injury may beassociated with non-communicating hydrocephalus, which ispotentially life threatening. Suggested management options in-clude closing the pinhole in the arachnoid or opening the arach-noid widely and suturing it to the dura. The authors of thesereports also advocate routine thorough microscopic examinationof the arachnoid plane to identify any breaches or CSF leakage, inpatients for whom an arachnoid-sparing procedure is attempted.

The arachnoid-preserving procedure is challenging in patientswith CMI and extensive subarachnoid scarring at the craniospinallevel, although preservation of the arachnoid membrane, wheneverthere is no evidence of obstruction of the foramen of Magendieand/or arachnoiditis, decreases complications. In other words, forpatients with CMI with syringomyelia and/or hydrocephalus, thearachnoid should be dissected whenever simple decompressiondoes not appear to be sufficient to re-establish CSF flow. It is diffi-cult to draw conclusions based on a series of limited size, but thisstudy suggests that the use of duraplasty to treat CMI leads to agreater decrease in concurrent syringomyelia. Further studies areneeded to better characterize patients with this condition and todetermine which patients with CMI are better treated with anarachnoid-sparing procedure and those who require arachnoid dis-section for syringomyelia resolution.

5. Conclusion

Foramen magnum decompression and arachnoid preservationwith duraplasty is a safe and effective treatment for patients withCMI and associated syringomyelia. The procedure achieves a reduc-tion in size, or collapse, of the syrinx in the majority of patients andis associated with improvements in many neurological distur-bances. Although our results are encouraging, a larger number ofpatients and a longer follow-up are required to determine the effi-cacy of arachnoid-preserving foramen magnum decompression.

Appendix A. Supplementary data

Supplementary video data (surgical technique of posterior fossadecompression using duraplasty for Chiari I malformation) associ-ated with this article can be found, in the online version, atdoi:10.1016/j.jocn.2011.06.034.

References

1. Caldarelli M, Di Rocco C. Diagnosis of Chiari I malformation and relatedsyringomyelia: radiological and neurophysiological studies. Childs Nerv Syst2004;20:332–5.

2. Heiss JD, Patronas N, DeVroom HL, et al. Elucidating the pathophysiology ofsyringomyelia. J Neurosurg 1999;91:553–62.

3. Marin-Padilla M, Marin-Padilla TM. Morphogenesis of experimentally inducedArnold-Chiari malformation. J Neurol Sci 1981;50:29–55.

4. Oldfield EH, Muraszko K, Shawker TH, et al. Pathophysiology of syringomyeliaassociated with Chiari 1 malformation of the cerebellar tonsils. Implications fordiagnosis and treatment. J Neurosurg 1994;80:3–15.

5. Isu T, Sasaki H, Takamura H, et al. Foramen magnum decompression withremoval of the outer layer of the dura as treatment for syringomyelia occurringwith Chiari I malformation. Neurosurgery 1993;33:844–50.

6. Schijman E, Steinbok P. International survey on the management of Chiari Imalformation and syringomyelia. Childs Nerv Syst 2004;20:341–8.

7. Guyotat J, Bret P, Jouanneau E, et al. Syringomyelia associated with type I Chiarimalformation. A 21-year retrospective study on 75 cases treated by foramenmagnum decompression with a special emphasis on the value of tonsilsresection. Acta Neurochir 1998;140:745–54.

8. Imae S. Clinical evaluation on etiology and surgical outcome in syringomyeliaassociated with Chiari type I malformation. No To Shinkei 1997;49:1131–8.

9. Munshi I, Frim D, Stine-Reyes R, et al. Effects of posterior fossa decompressionwith and without duraplasty on Chiari malformation associated withhydromyelia. Neurosurgery 2000;46:1384–90.

10. Aboulker J. Syringomyelia and intra-rachidian fluids. XII. Actual surgerysyringomyelia. Neurochirurgie 1979;25(Suppl.1):111–31.

11. Alzate JC, Kothbauer KF, Jallo GI, et al. Treatment of Chiari I malformation inpatients with and without syringomyelia: a consecutive series of 66 cases.Neurosurg Focus 2001;11:1–9.

12. Klekamp J, Iaconetta G, Samii M. Spontaneous resolution of Chiari Imalformation and syringomyelia: case report and review of the literature.Neurosurgery 2001;48:664–7.

13. Matsumoto T, Syrnon L. Surgical management of syringomyelia-current results.Surg neurol 1989;32:258–65.

14. Van den Bergh R, Hoorens R, Van Calenbergh R. Syringomyelia: a retrospectivestudy. Part I: clinical features. Acta Neurol Belg 1990;90:93–9.

15. Batzdorf U. Chiari I malformation with syringomyelia. Evaluation of surgicaltherapy by magnetic resonance imaging. J Neurosurg 1988;68:726–30.

16. Caldarelli M, Novegno F, Massimi L, et al. The role of limited posterior fossacraniectomy in the surgical treatment of Chiari malformation Type I:experience with a pediatric series. J Neurosurg 2007;106:187–95.

17. Sindou M, Chavez-Machuca J, Hashish H. Cranio-cervical decompression forChiari type I-malformation, adding extreme lateral foramen magnum openingand expansile duroplasty with arachnoid preservation. Technique and long-term functional results in 44 consecutive adult cases–comparison withliterature data. Acta Neurochir 2002;144:1005–19.

18. Klekamp J, Batzdorf U, Samii M, et al. The surgical treatment of Chiari Imalformation. Acta Neurochir 1996;138:788–801.

19. Navarro R, Olavarria G, Seshadri R, et al. Surgical results of posterior fossadecompression for patients with Chiari I malformation. Childs Nerv Syst2004;20:349–56.

20. Paul KS, Lye RH, Strang FA, et al. Arnold-Chiari malformation. Review of 71cases. J Neurosurg 1983;58:183–7.

21. Tognetti F, Calbucci F. Syringomyelia: syringo-subarachnoid shunt versusposterior fossa decompression. Acta Neurochir 1993;123:196–7.

22. Mobbs R, Teo C. Endoscopic assisted posterior fossa decompression. J ClinNeurosci 2001;8:343–4.