SPINE Volume 35, Number 26S, pp S331–S337©2010, Lippincott Williams & Wilkins

A Radiographic Assessment of the Ability of theExtreme Lateral Interbody Fusion Procedure toIndirectly Decompress the Neural Elements

Leonardo Oliveira, BSc,* Luis Marchi, MSc,* Etevaldo Coutinho, MD,*and Luiz Pimenta, MD, PhD†

Study Design. Prospective nonrandomized clinicalstudy on the decompressive effect of the extreme lateralinterbody fusion (XLIF) procedure.

Objective. This study evaluates the results of inter-body distraction from a lateral retroperitoneal approachfor the treatment of lumbar degenerative conditions in-clusive of central and/or lateral stenosis.

Summary of Background Data. Traditional treatmentfor symptomatic lumbar stenosis has been by direct pos-terior decompression (i.e., removal of ligamentum fla-vum, laminotomy/laminectomy, facetectomy, as needed).Stenotic symptoms may also be alleviated indirectly,through correction of intervertebral and foraminal heightand correction of spinal alignment. Anterior-only spinalprocedures rely on this indirect decompression whenused in patients with radicular symptoms.

Methods. Consecutive patients presenting with de-generative conditions that included concomitant lumbarstenosis were consented and treated via stand-aloneXLIF. Pre- and postoperative radiographic measurementswere made from plain lateral radiographs and sagittaland axial magnetic resonance imaging views by an inde-pendent radiologist using medical imaging software.Measurements included disc height, foraminal height, fo-raminal area, and canal diameter.

Results. In all, 7 male and 14 female patients (meanage, 67.6 years; range, 40–83) underwent XLIF at 43 lum-bar levels in an average operative time of 47 minutes andwith an average 23 mL estimated blood loss per level.There were no intraoperative complications. Mean hospi-tal stay was 29.5 hours. Transient postoperative psoasweakness occurred in 3 (14.3%) of the cases. Substantialdimensional improvement was evidenced in all radio-graphic parameters, with increases of 41.9% in averagedisc height, 13.5% in foraminal height, 24.7% in foraminalarea, and 33.1% in central canal diameter. Two patients(9.5%) required a second procedure for additional poste-rior decompression and/or instrumentation.

Conclusion. The XLIF procedure provides the neces-sary decompression for the treatment of central and/or-lateral stenosis in a minimally disruptive way, avoiding, inmost cases, the need for the direct resection of posteriorelements and associated morbidities. Indirect decom-pression may be limited in cases of congenital stenosisand/or locked facets. Its effect may also be reduced bypostoperative subsidence and/or loss of correction.

Key words: XLIF, claudication, MRI, foraminal height,foraminal area. Spine 2010;35:S331–S337

The anatomic changes of hypertrophic articular pro-cesses causing spinal canal stenosis were first describedby Verbiest in 1954.1 At present, stenosis is often definedas occurring in the central, lateral recess, or foraminalareas, and may also be categorized as due to soft tissue orhard tissue encroachment and/or spinal malalignment.When disc height is significantly reduced due to degen-eration or malalignment such as spondylolisthesis, theinterlaminar space and intervertebral foramina are alsoreduced. Bony encroachment through facet hypertrophymay also contribute.

The primary indication and goal of surgery in patientswith symptomatic degenerative lumbar stenosis is neuraldecompression. This can be and often is accomplishedby direct posterior resection of bone and/or soft tissuesuch as in laminectomy/laminotomy, facetectomy, andforaminotomy procedures. Complications associatedwith direct lumbar decompressive surgery include bleed-ing, epidural hematoma, deep venous thrombosis, duraltear, cerebrospinal fluid leak, infection, nerve root injury,epidural fibrosis, iatrogenic instability, and recurrence ofsymptoms.2–5 Wide bony decompression may necessitatestabilization by instrumentation and fusion due to the re-section of elements responsible for natural spinal stability.Interbody fusion may also be indicated when there is a lossof disc height and alignment. Decompression alone in casesof malalignment has been shown to be less effective thananterior interbody fusion at alleviating pain and func-tion,6,7 as interbody fusion provides restoration of discheight and correction of coronal and sagittal alignment.

The extreme lateral interbody fusion (XLIF) proce-dure has been described as an alternative approach toanterior column stabilization in degenerative, deformity,and traumatic conditions of the thoracolumbarspine.8–13 The XLIF procedure has advantages over theearly stand-alone direct anterior lumbar interbody fu-sion techniques, in that the anterior longitudinal liga-

From the *Instituto de Patologia de Coluna, Sao Paulo, Brazil; and†Department of Neurosurgery, University of California San Diego, SanDiego, CA.Acknowledgment date: August 19, 2010. First revision date: Septem-ber 27, 2010. Second revision date: October 14, 2010. Acceptancedate: October 15, 2010.Supported by NuVasive, Inc. financially and materially in the form ofcoverage for nonreimbursable study-related imaging costs and provi-sion of devices.The device(s)/drug(s) is/are FDA-approved or approved by correspond-ing national agency for this indication. Corporate/Industry funds werereceived to support this work. No benefits in any form have been or willbe received from a commercial party related directly or indirectly to thesubject of this manuscript.Address correspondence and reprint requests to Leonardo Oliveira,BSc, Instituto de Patologia da Coluna, Rua Vergueiro, 1421-Sala 305.Sao Paulo, Brazil 04101000; E-mail: leonardo@luizpimenta.com.br

S331

ment is preserved, and the larger XLIF implant spans thedense ring apophysis rather than threading into the cen-tral weaker portions of the endplate. Restoration of discheight and correction of alignment can be betterachieved through the ligamentotaxis created with the an-terior longitudinal ligament and PLL intact, since theseligaments exert great function on spinal alignment andstabilization.14,15 Relief of both back and leg pain hasbeen demonstrated using XLIF technique.9,11,12,16,17

However, the degree of indirect decompression has notyet been formally shown, nor have the indications whenit is feasible to avoid direct posterior decompression beenclearly identified. The goal of this study was to quantifythe indirect decompression achieved in XLIF surgery,and to evaluate the circumstances under which it is likelyor unlikely to achieve resolution of stenotic symptoms.

Patients and Methods

This study was a nonrandomized, single-center, ethics commit-tee-approved evaluation of patients with symptomatic lumbardegenerative conditions with central and/or lateral stenosis.The immediate postoperative radiographic results of these sur-geries were compared with preoperative baseline measures toevaluate the effect of indirect decompression provided by inter-body distraction via XLIF surgery.

The indications for surgery were symptomatic single- or mul-tilevel lumbosacral degenerative disease from L1–L2–L4–L5 withcentral and/or foraminal stenosis of at least 6 months durationthat had not responded to conservative care. Patients with priordirect decompressive or fusion surgery at the operative level(s)were excluded. Additional exclusion criteria included autoim-mune disease, malignancy, and pregnancy.

In all, 21 consecutive patients meeting inclusion/exclusioncriteria consented to participate and enrolled in this prospec-tive study from March 2008 through June 2009. All of thepatients were treated with standalone XLIF using 18-mm widepolyetheretherketone cages (CoRoent XL, NuVasive, Inc., SanDiego, CA) and a tricalcium silicate bone graft substitute (Ac-tifuse, ApaTech, Foxborough, MA), without additional directposterior decompression or internal fixation. Interbody devicesizing was performed intraoperatively for each patient andeach level, and implant height was chosen based on a snugdistractive fit and correction to a disc height similar to adjacentnormal levels. All surgeries were performed by the same sur-geon.

Plain anteroposterior and lateral radiographs as well as T2-weighted sagittal and axial magnetic resonance imaging (MRI)views were obtained at a single radiographic facility (URP Di-agnosticos, Sao Paulo, Brazil) before surgery and immediatelyafter surgery (within 2 weeks of surgery). From the lateral ra-diographs, anterior and posterior disc height, foraminal height,

Figure 1. Measures from plainlateral radiographs. Note thatarea measurements are relativeto the left axis scale (in mm2),while height measurements arerelative to the right axis scale (inmm). All postoperative measureswere statistically significantly in-creased compared with preoper-ative (P � 0.05).

Table 1. Average Radiographic Measures

ResultsPreoperative

Postoperative

Metric Metric % Change From Preoperative P

X-ray measurementsForaminal area (mm2) 243.0 303.1 24.7 0.0020Foraminal height (mm) 20.9 23.7 13.5 0.0027Average disc height (mm) 7.1 10.1 41.9 �0.0001

MRI measurementsSagittal central canal diameter (mm) 7.1 9.5 33.1 �0.0001Axial central canal area (mm2) 147.4 159.8 8.4 0.0092Axial subarticular diameter (R) (mm) 2.1 2.9 37.9 �0.0001Axial subarticular diameter (L) (mm) 1.9 2.8 46.3 0.0004

MRI indicates magnetic resonance imaging.

S332 Spine • Volume 35 • Number 26S • 2010

and foraminal area were measured. In MRI, a single sagittalslice through the anatomic center of the spine (i.e., midsagittal)was used as the comparative measurement location for deter-mining anteroposterior canal diameter. A single axial slicethrough the center of the disc was used as the comparativemeasure location in axial views for midsagittal canal diameter,canal area, and left and right subarticular diameter measure-ments. Custom software (VOXAR 3-dimensional) developedto analyze MRI was used to digitally measure linear dimensionsand areas. All distance measures were made at the narrowestpoint in the view.

T tests and �2 tests (Analyze-It Software, Ltd., Leeds, En-gland) were used to determine statistically significant changesfrom pre- to postoperative, with a level of significance of 0.05.

Results

Forty-three (43) levels in 21 patients were treated withstand-alone XLIF: 14 females, 7 males; average age, 67.6years (range, 40–83 years); average body mass index,25.6. All patients had a primary diagnosis of lumbarstenosis, accompanied by degenerative disc disease withdegenerative spondylolisthesis Grade I or II (n � 8)and/or degenerative scoliosis (n � 19). There were 4single-level, 13 two-level, 3 three-level, and 1 four-levelprocedure including 3 at L1–L2, 6 at L2–L3, 17 at L3–L4, and 17 at L4–L5. The surgeries were performed in anaverage of 86.2 minutes per patient (47.0 minutes perlevel), and with 44.3 mL blood loss per patient (23.4 mLper level); no case needed blood transfusion. In 5 proce-dures, the vertebrae adjacent to the operative disc spacewere prophylactically supplemented with bone cementdue to presumed poor bone quality (not confirmed byDEXA).

There were no intraoperative complications, andthere were no major postoperative changes in sensory ormotor lower extremity function. The average hospitaldischarge was after 29.5 hours, when patients couldmove with few restrictions.

In the immediate postoperative period, 14.3% pre-sented with psoas weakness, a condition that resolved tonormal muscle strength within a few days, with iliop-

soas-directed exercises accelerating the recovery. Onecase (4.8%) involved a transient psoas hematoma, whichresolved without treatment.

Central and foraminal decompression was significant,with an average 41.9% increase in disc height (P �0.0001), 13.5% increase in foraminal height (P �0.0027), 24.7% increase in foraminal area, and 33.1%increase in central canal diameter (P � 0.0001) (Table 1;Figures 1–3).

Two patients (9.5%) needed additional direct poste-rior decompression and/or instrumentation with pediclescrews due to inadequate resolution of stenotic symp-toms. One of the 2 patients showed early postoperativesubsidence of the standalone implant after intraoperativerestoration of disc height and alignment (mentioned incase example 3 described later in the text); in the other,disc and foraminal height were not adequately restored,and so supplemental direct decompression (laminec-tomy) and pedicle screw fixation were performed (men-tioned in case example 4 described later in the text).

Case Examples

Case 1. A 75-year-old man with degenerative scoliosiswith neurogenic claudication, complaining of axial andradicular pain (VAS of the back � 70; VAS of the left

Figure 2. Measures from axialMR images. Note that area mea-surements are relative to the leftaxis scale (in mm2), while dis-tance measurements are relativeto the right axis scale (in mm). Allpostoperative measures werestatistically significantly in-creased compared with preoper-ative (P � 0.05).

Figure 3. Measures from sagittal MR images. Postoperative mea-sures were statistically significantly increased compared withpreoperative (P � 0.05).

S333Extreme Lateral Interbody Fusion • Oliveira et al

leg � 90; VAS of the right leg � 80; ODI � 88) underwenta 2-level L3–L5 procedure (Figure 4). One week after sur-gery, the patient reported healthy improvement, withODI � 29, VAS of the back � 0, and VAS of the legs � 20.

Case 2. A 74-year-old woman presented with degener-ative scoliosis with axial and radicular pain (VAS of theback and right leg � 50; VAS of the left leg � 100;ODI � 67) and underwent an L3–L5 procedure (Figure5). Immediately after operation, the patient reportedpsoas weakness, but relief of preoperative pain, withODI � 54, VAS of the back � 40, and VAS of the legs � 0.

At her 6-week follow-up, psoas weakness was resolvedand ODI had improved to 12, VAS back to 20, VAS legsstill � 0.

Case 3. A 63-year-old woman presented with signifi-cant axial and radicular pain (VAS of the left leg � 100;VAS of the back � 80) due to single-level asymmetricaldisc collapse and L4–L5 Grade II spondylolisthesis. Shehad no prior surgeries and was 1 of the 5 patients in thiscohort who was prophylactically augmented with verte-bral body cement before interbody fusion. She under-went a single-level L4–L5 stand-alone XLIF procedure,

Figure 4. Case example 1. Pre- and postoperative sagittal and axial MRIs revealing indirect decompression of the central canal at L3– 4and L4 –5.

Figure 5. Case example 2. Pre- and postoperative sagittal and axial MRIs revealing indirect decompression of the central canal at L3– 4and L4 –5.

S334 Spine • Volume 35 • Number 26S • 2010

which restored significant disc height and partial reduc-tion of the slip. Early postoperative lateral radiographsshowed evidence of slight subsidence inferiorly with lossof sagittal correction (Figure 6). Revision with supple-mental pedicle screw fixation was attempted due to thelack of improvement in pain and function scores.

Case 4. An 83-year-old man presented with axial andradicular pain (VAS of the back and left leg � 70; VAS ofthe right leg � 100; ODI � 75). Preoperative examina-tions revealed a prior interspinous process spacing deviceat the L4–L5 level and anatomically short pedicles. Afteran L4–L5 XLIF procedure, the patient did not reportclinical improvement, and radiographs revealed persis-

tent central and foraminal stenosis (Figure 7). In thisunique case, disc and foraminal height were not ade-quately restored. This patient’s congenital stenosis andprior interspinous process spacing device, which was notremoved at the time of XLIF surgery, may have limitedthe ability to adequately distract the anterior column andneural foramens. A hemilaminectomy was performed toachieve the necessary decompression, and the constructwas supplemented with bilateral pedicle screws.

Discussion

The published literature supports the practice of fusionwith indirect decompression for spinal pathologies pre-

Figure 6. Preoperative lateral radiograph and sagittal MRI showing degenerative spondylolisthesis, intraoperative lateral fluororadiographshowing restoration of disc height and reduction of slip, and postoperative lateral radiograph and sagittal MRI showing loss of correctionfollowing subsidence.

Figure 7. Case example 4. Preoperative lateral radiograph showing congenitally short pedicles; and pre- and postoperative sagittal andaxial MRIs revealing inadequate indirect decompression at L4 –5.

S335Extreme Lateral Interbody Fusion • Oliveira et al

senting with neurologic symptoms, particularly in casesof instability and malalignment.6,7,18–28 Consistent withthese prior reports, the current study on anterior lumbarinterbody fusion with XLIF showed radiographically itseffectiveness in decompressing stenosis associated withcommon lumbar degenerative disorders.

Unlike direct anterior or posterior approaches, lateralaccess is ligament-sparing and allows for the placementof a large intrinsically stable implant to be placed com-pletely across the interspace, resulting in a very stableinterbody construct. Proper surgical technique includesaggressive disc removal and release of the contralateralanulus. This process not only assures a proper graft bedbut also allows for interbody distraction that restoresdisc height and facilitates reduction of deformity throughligamentotaxis. Moreover, XLIF avoids many of the po-tential complications of traditional surgery, includingthose associated with direct decompression and poste-rior access.

Regarding when indirect decompression will be suffi-cient or not, a staged approach must be recommended,counseling patients about the fact that an additional mi-crodecompression might be required if symptoms per-sist. Severe central spinal stenosis is a relative contrain-dication if the patient is unwilling to accept thepossibility of a second operation (direct decompression)if neurologic symptoms persist after surgery. In the au-thor’s experience, and as was shown in this work, thesecond procedure is unnecessary in the majority of casesand spares the patient undue morbidity and risk of neu-ral injury or scarring from direct posterior surgery.

In patients further along the degeneration disease pro-cess, lateral recess and foraminal stenosis are exacer-bated by osteophyte formation. Low back pain is usuallydiminished or absent, and leg pain is persistent at rest.Interbody distraction and fusion may not adequately de-compress stenosis in this stage because correction ofalignment is blocked by osteophytes on the posteriorsurface of the articular processes, and anterior osteo-phytes still impinge on the neural elements.

In the authors’ experience (although not necessarilystatistically supported by the results in this small cohort),the requirement of direct decompression is determinedbased on the defining preoperative characteristics thatinclude congenital stenosis/congenitally short pedicles;uncontained disc herniation; significant facet arthropa-thy/osteophyte formation (locked facets) coupled withcalcified disc; PLL or osteophytes arising from the pos-terior endplates, with complete or near complete com-promise of the lateral recess; synovial cysts; and/or radic-ular symptoms unimproved with flexion.

Conversely, XLIF appears to provide sufficient indi-rect decompression for disc bulge, such as in recurrentdisc herniation; collapsed disc with loss of foraminalheight and/or soft tissue encroachment (e.g., posterioranulus, PLL, ligamentum flavum) of the canal; lateral,retro-, or spondylolisthesis, with narrowing of the cen-tral and intervertebral foramens due to the malalign-

ment; degenerative scoliosis with unilateral central orforaminal stenosis due to the malalignment.

The limitations of this study include small sample sizeand short radiographic follow-up. Maintenance of cor-rection and decompression was not evaluated. However,1 patient demonstrated subsidence and loss of correctionin the short-term that limited the effect of decompressionand resolution of symptoms. Wider interbody devices(cages) may reduce the incidence of subsidence and com-plications related to its occurrence. Larger cages havelarger contact area and exert less pressure on the verte-bral endplate. Supplemental internal fixation is not man-datory to achieve indirect decompression, but posterioror lateral fixation increases construct stiffness and mayprotect the indirect decompression, particularly incases where bone quality and potential subsidence area concern.

Although subsidence was an incidental finding in thisstudy and its incidence has a clear effect on the mainte-nance of indirect decompression, conclusions relative tothis question were not within the scope of the design ofthis study, which was to quantify the ability of interbodydistraction through XLIF to increase the neural foraminadimensionally through interbody distraction. The ques-tion of subsidence and/or maintenance of correction andprogression to fusion, along with lasting clinical benefit,is dependent on several variables, including grafting andinternal fixation techniques, preferences for which varyamong XLIF surgeons. Until further follow-up data areavailable, the long-term clinical and radiographic effectsof standalone XLIF for stenosis are uncertain. Howeverlimited, the immediate postoperative quantifiable in-creases in central and lateral radiographic measures maybe more generalizable to multiple applications of theXLIF procedure.

Conclusion

Interbody distraction through XLIF is a reproducibleand effective mechanism for achieving indirect decom-pression in stenotic degenerative conditions.

Key Points

● Disc height, foraminal height, and central canalmeasures are significantly increased following in-terbody fusion via stand-alone XLIF in most cases.

● Congenital stenosis and locked facets may limitthe extent of decompression achievable indi-rectly.

● Maintenance of correction, although not studiedin detail here, may be dependent on subsidence,which in turn is dependent on grafting and inter-nal fixation techniques for eventual fusion in thecorrected position.

S336 Spine • Volume 35 • Number 26S • 2010

References

1. Verbiest H. A radicular syndrome from developmental narrowing of thelumbar vertebral canal. J Bone Joint Surg Br 1954;36:230–7.

2. Airaksinen O, Herno A, Turunen V, et al. Surgical outcome of 438 patientstreated surgically for lumbar spinal stenosis. Spine 1997;22:2278–82.

3. Jang IT, Lee SW, Atienza PM, et al. Decompressive surgery alone for lumbarspinal stenosis in elderly patients. Korean J Spine 2008;5:83–8.

4. Kou J, Fischgrund J, Biddinger A, et al. Risk factors for spinal epiduralhematoma after spinal surgery. Spine 2002;27:1670–3.

5. Turner JA, Ersek M, Herron L, et al. Surgery for lumbar spinal stenosis.Attempted meta-analysis of the literature. Spine 1992;17:1–8.

6. Mardjetko SM, Connolly PJ, Shott S. Degenerative lumbar spondylolisthesis.A meta-analysis of literature 1970–1993. Spine 1994;19:2256S–65S.

7. Satomi K, Hirabayashi K, Toyama Y, et al. A clinical study of degenerativespondylolisthesis. Radiographic analysis and choice of treatment. Spine1992;17:1329–36.

8. Billinghurst J, Akbarnia BA. Extreme lateral interbody fusion—XLIF. CurrOrthop Pract 2009;20:238–51.

9. Dakwar E, Cardona RF, Smith DA, et al. Early outcomes and safety of theminimally invasive, lateral retroperitoneal transpsoas approach for adultdegenerative scoliosis. Neurosurg Focus 2010;28:E8.

10. Madhok R, Kanter AS. Extreme-lateral, minimally invasive, transpsoas ap-proach for the treatment of far-lateral lumbar disc herniation. J NeurosurgSpine 2010;12:347–50.

11. Ozgur BM, Aryan HE, Pimenta L, et al. Extreme lateral interbody fusion(XLIF): a novel surgical technique for anterior lumbar interbody fusion.Spine J 2006;6:435–43.

12. Rodgers WB, Cox CS, Gerber EJ. Experience and early results with a mini-mally invasive technique for anterior column support through eXtreme Lat-eral Interbody Fusion (XLIF). US Musculoskelet Rev 2007;1:28–32.

13. Shen FH, Samartzis D, Khanna AJ, et al. Minimally invasive techniques forlumbar interbody fusions. Orthop Clin North Am 2007;38:373–86.

14. Dooris AP, Goel VK, Grosland NM, et al. Load-sharing between anteriorand posterior elements in a lumbar motion segment implanted with an arti-ficial disc. Spine 2001;26:E122–9.

15. White AA, Panjabi MM. Clinical Biomechanics of the Spine. 2 ed. Philadel-phia, PA: Lippincott Williams & Wilkins; 1990:92–111.

16. Oliveira L, Marchi L, Coutinho E, et al. The use of rh-BMP2 in standaloneextreme lateral interbody fusion (XLIF): clinical and radiological resultsafter 24 months follow-up. World Spinal Column J 2010;1:19–25.

17. Rodgers WB, Cox CS, Gerber EJ. Minimally invasive treatment (XLIF) ofadjacent segment disease after prior lumbar fusions. Internet J MinimallyInvasive Spinal Technol 2009;3.

18. Bayley JC, Yoo JU, Kruger DM, et al. The role of distraction in improving thespace available for the cord in cervical spondylosis. Spine 1995;20:771–5.

19. Inoue S, Watanabe T, Hirose A, et al. Anterior discectomy and interbodyfusion for lumbar disc herniation. A review of 350 cases. Clin Orthop RelatRes 1984:22–31.

20. Inoue S, Watanabe T, Goto S, et al. Degenerative spondylolisthesis. Patho-physiology and results of anterior interbody fusion. Clin Orthop Relat Res1988;227:90–8.

21. Kim NH, Kim DJ. Anterior interbody fusion for spondylolisthesis. Ortho-pedics 1991;14:1069–76.

22. Kim NH, Kim HK, Suh JS. A computed tomographic analysis of changes inthe spinal canal after anterior lumbar interbody fusion. Clin Orthop RelatRes 1993:180–91.

23. Kim NH, Lee JW. Anterior interbody fusion versus posterolateral fusionwith transpedicular fixation for isthmic spondylolisthesis in adults. A com-parison of clinical results. Spine 1999;24:812–6.

24. Kozak JA, Heilman AE, O’Brien JP. Anterior lumbar fusion options. Tech-nique and graft materials. Clin Orthop Relat Res 1994:45–51.

25. Kuslich S. The utility of interbody arthrodesis using BAK cages in treatment.In: Gunzburg R, Szpalski M, eds. Lumbar Spinal Stenosis. Philadelphia, PA:LWW; 2000:287–301.

26. Lee SH, Choi WG, Lim SR, et al. Minimally invasive anterior lumbar inter-body fusion followed by percutaneous pedicle screw fixation for isthmicspondylolisthesis. Spine J 2004;4:644–9.

27. Takahashi K, Kitahara H, Yamagata M, et al. Long-term results of anteriorinterbody fusion for treatment of degenerative spondylolisthesis. Spine1990;15:1211–5.

28. Vamvanij V, Ferrara LA, Hai Y, et al. Quantitative changes in spinal canaldimensions using interbody distraction for spondylolisthesis. Spine 2001;26:E13–8.

S337Extreme Lateral Interbody Fusion • Oliveira et al