J Neurosurg Spine 8:376–380, 2008

376 J. Neurosurg.: Spine / Volume 8 / April 2008

UVENILE RA can lead to early epiphysial closure andimpairment in growth including micrognathia and ashort stiff neck.12,17,18 Cranial settling and basilar in-

vagination can also develop and may cause progressivecompression of the CMJ and death. In this situation surgi-cal intervention is warranted. Although traction-inducedreduction and posterior fixation are the treatment of choice,anterior decompression may be indicated in cases involv-ing an irreducible cervicomedullary abnormality.5,15

The minimal requirement for transoral odontoid resec-tion is that the mouth be capable of opening . 25 mm.4Inadequate access may require an extensive midline soft-palate splitting, hard-palate resection, or extended max-illotomy, particularly if a microscope is being used for visualization.3,10 As an alternative, several endoscopic ap-proaches to the CMJ have been developed including ananterior cervical,19 transoral,7,8 and transnasal11 route. Thelatter approach, although described in several cadavericdissections,1,6,16 has only been reported in a single operative

case.11 We add a second case to the literature, reporting ona patient in whom a microscope-assisted transoral approachwas not possible due to the patient’s micrognathia. Wewished to validate and reproduce this transnasal endoscop-ic approach to the CMJ and contribute several operativenuances to the literature.

Case Report

Examination. This 25-year-old man with a history of non-achondroplastic dwarfism and juvenile RA presented withprogressive occipital headache, neck pain, and intermittentparesthesias in bilateral upper extremities. Chronic steroidtherapy for his arthritis caused secondary osteoporosis. Inthe past he had undergone bilateral hip arthroplasties andwas being evaluated for knee arthroplasty. On examinationhe had full motor strength and no evidence of myelopathy.Cervical magnetic resonance imaging revealed pronouncedbasilar invagination with compression of the CMJ (Fig.1A).

First Operation. Initially, the patient was placed in trac-tion and experienced significant reduction of the invagina-

Endonasal endoscopic resection of the odontoid process in a nonachondroplastic dwarf with juvenile rheumatoidarthritis: feasibility of the approach and utility of theintraoperative Iso-C three-dimensional navigation

Case report

ILYA LAUFER, M.D.,1 JEFFREY P. GREENFIELD, M.D., PH.D.,1 VIJAY K. ANAND, M.D.,2ROGER HÄRTL, M.D.,1 AND THEODORE H. SCHWARTZ, M.D.1

Departments of 1Neurological Surgery and 2Otolaryngology, Weill Cornell Medical College, NewYork Presbyterian Hospital, New York, New York

PPThe authors report a case of a nonachnodroplastic dwarf with severe basilar invagination and compression of thecervicomedullary junction (CMJ) due to juvenile rheumatoid arthritis. Initially excellent reduction of the invaginationand decompression of the CMJ was achieved using posterior fixation. However, 1 month postoperatively symptomsrecurred and the authors found imaging evidence of recurrence as well. The patient subsequently underwent an endo-scopic transnasal resection of the dens with assistance of Iso-C navigation. He recovered well and tolerated regular dieton postoperative Day 2. (DOI: 10.3171/SPI/2008/8/4/376)

KEY WORDS • basilar invagination • endoscopy • odontoidectomy •rheumatoid arthritis

J

Abbreviations used in this paper: CMJ = cervicomedullary junc-tion; RA = rheumatoid arthritis.

tion (Fig. 1B and C). He subsequently underwent occipito-cervical fixation and arthrodesis. Intraoperatively, his bonewas noted to be severely osteoporotic and was judged to be too brittle for stable screw purchase. Thus, sublaminarwires at C-2 through C-6 were attached to bilateral titani-um rods, connecting the construct to an occipital plate heldin place with bicortical screws. The fusion was enhancedwith allograft bone and bone morphogenetic protein. Thepatient was placed in halo fixation. Postoperative imagingrevealed significant reduction of the invagination (Fig. 2A).

Second Operation. At the 1-month follow-up the patientcomplained of recurrent neck pain, and imaging revealedincreased angulation of the cranial base in relationship tothe neck with recurrent basilar invagination, but stable pos-terior instrumentation (Fig. 2B). The halo was readjustedand he was taken for an endoscopic transnasal odontoidec-tomy. The Iso-C system (Siremobil Iso-C Three-Dimen-sional, Siemens Medical Solutions) with 3D fluoroscopicreconstruction was used for frameless stereotactic naviga-tion (BrainLAB) and for postresection intraoperative eval-uation of the anterior decompression.

Postoperative Course. The patient was extubated themorning after surgery and was eating a regular diet within

2 days. Postoperative computed tomography scanning con-firmed complete resection of the odontoid process anddecompression of the CMJ (Fig. 2C).

Surgical Procedure

The details of the anatomical dissection and surgery havebeen reported elsewhere.1,6,11,16 We have incorporated smallmodifications to this previously described technique, whichwill be highlighted. In brief, following oropharyngeal in-tubation with the halo in place, antibiotics and dexameth-asone were administered. A submucosal resection of theseptal cartilage was performed to facilitate the lateral-to-medial endonasal movement of the instruments. The inferi-or turbinates were outfractured bilaterally and the posterior2 cm of the nasal septum was removed using a tissue shaverand a high-speed drill to enlarge the choana for a widerexposure. The sphenopalatine artery and the posterior sep-tal artery were cauterized bilaterally. We did not remove themiddle turbinate, and the sphenoid sinus was only openedslightly as a landmark. A 30-cm, 4-mm rigid 0° endoscope(Karl Storz) was held in the left nostril with an endoscopeholder and surgery was performed using a bimanual tech-nique either via the right nostril or both nostrils as required.

J. Neurosurg.: Spine / Volume 8 / April 2008

Endonasal endoscopic resection of the odontoid process

377

FIG. 1. Imaging studies. Preoperative T2-weighted sagittal magnetic resonance image (A) showing significant basilarinvagination and compression of the CMJ. Lateral radiographs obtained before (B) and after (C) the application of trac-tion.

FIG. 2. Sagittal computed tomography reconstructions. A: Initial image acquired after posterior fixation, showingsignificant reduction of invagination. B: One-month follow-up scan revealing recurrent settling. C: Postoperativeimage obtained after the endoscopic operation, demonstrating complete odontoid resection and decompression of theCMJ.

A red rubber catheter was placed though the nasal cavityinto the oral cavity for downward retraction of the softpalate to facilitate exposure (Fig. 3). An inverted U-shapedincision was made in the basopharyngeal fascia with its su-perior extent just below the sphenoid sinus, passing downthe clivus with its lateral limits just medial to the eustachi-an tube and its base at the level of the C-2 vertebral body(Fig. 4A). The flap was pushed inferiorly into the orophar-ynx and the longus colli and capitis muscles were pushedlaterally with the bovie to expose the ring of C-1. The archof C-1 was removed with a high-speed drill (Fig. 4B) toexpose the lateral margins of the dens, which was transect-ed across its base. Because the bone was demineralizedfrom chronic steroid use, the odontoid was resected in apiecemeal fashion using an angled curette and a pituitaryrongeur. The base of C-2 was then removed with the aid ofa 30° 18-cm, 4-mm endoscope (Karl Storz) angled down-ward to facilitate inferior visualization. The alar and apicalligaments were transected. The soft-tissue pannus, trans-verse ligament, and tectorial membrane were then removedwith a cavitron until dural pulsations were seen. The Iso-Cwas brought into the field to create a triplanar tomograph ofthe CMJ. The residual odontoid was identified laterally andit still compressed the thecal sac; the images were used fornavigation to remove the remaining odontoid bone. A finalimage with the Iso-C demonstrated a complete resection.The basopharyngeal fascial flap was reapproximated usinga fibrin sealant (Tisseel VH, Baxter).

Discussion

Basilar invagination is found in up to 11% of patientswith RA and often manifests as occipital or neck pain, fre-

quently progressing to disabling myelopathy. Once conser-vative management fails, occipitocervical fixation with orwithout anterior decompression is a viable treatment op-tion. Posterior fixation can often be complicated in the RApopulation by poor bone quality, long-standing steroid use,and extensive ligamentous laxity.5,9,13

The transoral approach provides the most direct route tothe ventral craniovertebral junction and is currently thestandard of care for anterior decompression.5,14 Transmax-illary and transmandibular extensions of this approach maybe used to provide wider rostral or caudal working fields.4However, these approaches often require a postoperativetracheostomy or gastrostomy and cause significant tongueand tracheal swelling. Hypernasal speech and nasal regur-gitation can occur. Mandibular splitting is cosmeticallyunappealing, and prolonged recovery and hospitalizationare often needed. The wound is constantly bathed in salivaand oropharyngeal flora, which increases the risk of in-fection. Use of the operative microscope limits the field ofview, and in patients with micrognathia and/or dwarfismthe transoral microscope-assisted resection of the odontoidis contraindicated.

Endoscopic illumination brings the lens and light sourcecloser to the operative field, thereby reducing the need fora wide operative approach.2 Endoscopic anterior cervical,19transoral,7,8 and transnasal11 approaches have been previ-ously described. The advantage of the transnasal approachis that the incision lies above the oropharynx. Thus, thetransnasal approach should decrease the risk of tongue andtracheal swelling, hypernasal and nasal speech and woundinfection. Our patient was extubated within 12 hours ofcompletion of the operation and his diet was returned tonormal within 2 days, certainly quicker than would have

378 J. Neurosurg.: Spine / Volume 8 / April 2008

I. Laufer et al.

FIG. 3. Illustration of the approach (left), and schematic drawing of the operative field (right).

been possible after a transoral procedure. This led to anexpeditious recovery and hospital discharge. In addition,patients with dwarfism or micrognathia and children whosenatural orifices are too small for the microscope-based tran-soral approach can be treated using a transnasal endoscopicapproach. The main limitation of the approach is the inferi-or extent, which can only reach the bottom of C-2 whereasthe transoral approach can reach C-3.

We wished to acknowledge the excellent work in cadav-ers and the prior case in the literature in which the authorsinitially described the endoscopic transnasal approach tothe CMJ.1,6,11,16 The purpose of our report was to validate theapproach and demonstrate its reproducibility and success atanother institution. In addition we wished to describe sev-eral operative nuances that differentiate our approach fromprior reports. 1) Although the surgery benefits from collab-oration between an otolaryngologist and a neurosurgeon,during the resection of the odontoid process, the operativefield is relatively stable, allowing us to place the endoscopeon a scope holder and obviating the need for 2 surgeons atall times. 2) We did not remove the middle turbinate andfound the space adequate for surgery. We believe that pres-ervation of both middle turbinates is helpful in restoringnormal laminar flow of air through the nose and that it min-imizes postoperative crusting. 3) We did not widely openthe sphenoid sinus and drill down the clivus, as previouslydescribed.11 The top of the inverted U-shaped incision cansit just below the sphenoid sinus, which makes the opera-tion quicker and even less invasive. 4) We performed a sub-mucosal resection of the septal cartilage. This maneuverpermits the scope and instruments to be moved mediallywithout resistance. 5) The surgery in our case was per-formed in a dwarf with micrognathia, indicating that thesize of the nasal passages is not a limitation, further validat-ing the minimal-access philosophy of this approach. 6) Be-cause the operative access for the transnasal endoscope ap-proach requires exposure only of the nostrils, we were ableto perform the approach while the patient was still fixed inthe halo vest. This would not be possible using the transcer-

vical endoscope approach.19 7) We used the Iso-C systemboth for navigation and for intraoperative quality assurancethat the odontoid had been adequately decompressed. Theaddition of triplanar fluoroscopy and real-time imaging ofsurgical results is extremely helpful to guarantee reliableresults.

Conclusions

We report a case where the transnasal endoscopic ap-proach to the CMJ provided a safe and effective minimalaccess alternative to the transoral approach. Patients withsmall oral and nasal cavities are still candidates for surgeryand the morbidity and hospital length of stay should theo-retically be decreased compared with those features inpatients undergoing the traditional transoral surgery. Theadvent of this approach is particularly significant for pa-tients whose anatomy precludes the standard transoral ap-proach and requires the more morbidity-prone alternativeapproaches. Intraoperative Iso-C–based 3D navigation sig-nificantly contributed to a complete, accurate, and safe re-section of the osseous mass. Larger case series will be re-quired to further validate this approach.

References

1. Alfieri A, Jho HD, Tschabitscher M: Endoscopic endonasal ap-proach to the ventral cranio-cervical junction: anatomical study.Acta Neurochir (Wien) 144:219–225, 2002

2. Anand VK, Schwartz TH: Practical Endoscopic Skull Base Sur-gery. San Diego: Pleural Publishing, 2007

3. Bhangoo RS, Crockard HA: Transmaxillary anterior decompres-sions in patients with severe basilar impression. Clin Orthop Re-lat Res 359:115–125, 1999

4. Casey AT, Crockard HA: Rheumatoid arthritis, in Dickman CA,Spetzler RF, Sonntag VKH (eds): Surgery of the Craniover-tebral Junction. New York: Thieme, 1998, 151–174

5. Casey AT, Crockard HA, Stevens J: Vertical translocation. Part II.Outcomes after surgical treatment of rheumatoid cervical myelop-athy. J Neurosurg 87:863–869, 1997

6. Cavallo LM, Cappabianca P, Messina A, Esposito F, Stella L, de

J. Neurosurg.: Spine / Volume 8 / April 2008

Endonasal endoscopic resection of the odontoid process

379

FIG. 4. Intraoperative images. A: Below the sphenoid sinus (SS), one can see the mucosa overlying the back of thenasopharynx (NP) between the eustachian tube (ET) where an inverted U-shaped incision was cauterized. B: The ba-sopharyngeal fascia was flapped downward to expose the clivus (CL) and ring of C-1, which was drilled off.

380

Divitiis E, et al: The extended endoscopic endonasal approach tothe clivus and cranio-vertebral junction: anatomical study. ChildsNerv Syst 23:665–671, 2007

7. Fraser JF, Anand VK, Schwartz TH: Endoscopic biopsy samplingof tophaceous gout of the odontoid process. Case report and re-view of the literature. J Neurosurg Spine 7:61–64, 2007

8. Frempong-Boadu AK, Faunce WA, Fessler RG: Endoscopicallyassisted transoral-transpharyngeal approach to the craniovertebraljunction. Neurosurgery 51 (5 Suppl):S60–S66, 2002

9. Grob D, Schütz U, Plötz G: Occipitocervical fusion in patientswith rheumatoid arthritis. Clin Orthop Relat Res 366:46–53,1999

10. Kanamori Y, Miyamoto K, Hosoe H, Fujitsuka H, Tatematsu N,Shimizu K: Transoral approach using the mandibular osteotomyfor atlantoaxial vertical subluxation in juvenile rheumatoid arth-ritis associated with mandibular micrognathia. J Spinal DisordTech 16:221–224, 2003

11. Kassam AB, Snyderman C, Gardner P, Carrau R, Spiro R: Theexpanded endonasal approach: a fully endoscopic transnasal ap-proach and resection of the odontoid process: technical case re-port. Neurosurgery 57:E213, 2005

12. Kjellberg H: Craniofacial growth in juvenile chronic arthritis.Acta Odontol Scand 56:360–365, 1998

13. Matsunaga S, Ijiri K, Koga H: Results of a longer than 10-year fol-low-up of patients with rheumatoid arthritis treated by occipito-cervical fusion. Spine 25:1749–1753, 2000

14. Menezes AH, VanGilder JC: Transoral-transpharyngeal approach

to the anterior craniocervical junction. Ten-year experience with72 patients. J Neurosurg 69:895–903, 1988

15. Menezes AH, VanGilder JC, Clark CR, el-Khoury G: Odontoidupward migration in rheumatoid arthritis. An analysis of 45 pa-tients with “cranial settling”. J Neurosurg 63:500–509, 1985

16. Messina A, Bruno MC, Decq P, Coste A, Cavallo LM, de DivittisE: Pure endoscopic endonasal odontoidectomy: anatomical study.Neurosurg Rev 30:189–194, 2007

17. Narayanan K, Rajendran CP, Porkodi R, Shanmuganandan K: Afollow-up study of juvenile rheumatoid arthritis into adulthood. JAssoc Physicians India 50:1039–1041, 2002

18. Stabrun AE: Impaired mandibular growth and micrognathic devel-opment in children with juvenile rheumatoid arthritis. A longitu-dinal study of lateral cephalographs. Eur J Orthod 13:423–434,1991

19. Wolinsky JP, Sciubba DM, Suk I, Gokaslan ZL: Endoscopic im-age-guided odontoidectomy for decompression of basilar invagi-nation via a standard anterior cervical approach. Technical note. JNeurosurg Spine 6:184–191, 2007

Manuscript submitted August 23, 2007.Accepted December 4, 2007.Address correspondence to: Theodore H. Schwartz, M.D., De-

partment of Neurological Surgery, Weill Cornell Medical College,New York Presbyterian Hospital, 525 East 68th Street, Box #99,New York, New York 10021. email: schwarh@med.cornell.edu.

J. Neurosurg.: Spine / Volume 8 / April 2008

I. Laufer et al.