Journal of Clinical Neuroscience 21 (2014) 841–845

Contents lists available at ScienceDirect

Journal of Clinical Neuroscience

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

Operative Technique

Feasibility of the mini-open vertebral column resection for severethoracic kyphosis

0967-5868/$ - see front matter � 2013 Elsevier Ltd. All rights reserved.http://dx.doi.org/10.1016/j.jocn.2013.10.001

⇑ Corresponding author. Tel.: +1 415 353 9095; fax: +1 415 353 3907.E-mail address: choud@neurosurg.ucsf.edu (D. Chou).

Dean Chou ⇑, Darryl Lau, Esha RoyDepartment of Neurological Surgery, University of California, San Francisco, 505 Parnassus Avenue, Box 0112, San Francisco, CA 94143, USA

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

Article history:Received 27 September 2013Accepted 9 October 2013

Keywords:CorpectomyKyphosisMini-openMinimally invasiveThoracic spineVertebral column resectionVertebrectomy

Severe thoracic kyphosis caused by pathologic fractures often needs to be corrected by resection of thecollapsed vertebral body, reconstruction of the anterior spinal column, and correction of the kyphosiswith long-segment fixation. The resection of this pathologic bone functions essentially as a vertebralcolumn resection. With the advent of minimally invasive technology, the powerful corrective forces affor-ded in open cases can be applied using a less invasive approach. In this article, we describe a mini-openposterior technique for thoracic kyphosis via a vertebrectomy and cantilever technique. Two patientsunderwent kyphosis correction via mini-open vertebrectomy. One patient was corrected from 92 degreesto 65 degrees, and the second patient was corrected from 70 degrees to 53 degrees. Both patients under-went a mini-open approach. Cantilever correction was accomplished over an expandable cage with aminimally invasive pedicle screw system. We describe our technique of mini-open vertebral columnresection and kyphosis correction in the thoracic spine.

� 2013 Elsevier Ltd. All rights reserved.

1. Background and importance

The surgical correction of severe thoracic kyphosis usuallyrequires an osteotomy to mobilize the spine and long segment fix-ation for correction and stabilization. Options include multiplePonte/Smith-Peterson osteotomies, pedicle subtraction osteotomy(PSO), or vertebral column resection (VCR) [1–3]. AlthoughPonte/Smith-Peterson osteotomies may be advantageous forsmooth-gradual kyphoses such as Scheurmann’s kyphosis, thismay not be the case for sharp-angular kyphoses in the thoracicspine with vertebral body compromise. The VCR is a good optionfor many reasons. First, the correction gained can be significant,as much as 60 degrees. Second, there is less posterior columnspinal shortening compared to a PSO, which can be limited becauseof the spinal cord’s inability to withstand severe deformation andkinking. Third, the thoracic VCR is commonly performed for manyother pathologies such as tumor or infection, and thus, it may be amuch more comfortable operation for many surgeons who performposterior vertebrectomies (costo-transversectomies, transpedicu-lar corpectomies, or extracavitary approaches). A vertebrectomywith circumferential bone removal in the face of a severe defor-mity functions essentially as a VCR.

Until recently, the standard posterior thoracic vertebrectomyhas been performed through large, open exposures, with muscleand fascial dissection of seven levels or more [1–4]. However, with

more recent advances in minimally invasive (MI) techniques,vertebrectomies are now being performed in a MI fashion or in amini-open fashion [5,6]. We have previously described the mini-open transpedicular corpectomy as a novel technique to performposteriorly-based corpectomies [6]. As an extension of thispreviously described technique, we describe how we utilized thisapproach to correct kyphosis via a mini-open vertebrectomy(which functions essentially as a VCR) and long segment fixation.

2. Clinical presentation

We describe the clinical presentation of case examples of severekyphosis, indications for surgery, and description of our technique.

2.1. Patient 1

A 61-year-old woman with metastatic breast cancer to T6presented with difficulty walking and gait imbalance. Neurologicalexamination was significant for 4/5 strength and ankle clonus inbilateral lower extremities. Standing lateral radiograph demon-strated 70 degrees of kyphosis in the thoracic region (measuredfrom T2 to T12) (Fig. 1A). MRI revealed a T6 pathological fractureand draping of the spinal cord over the T6 apical kyphosis/defor-mity (Fig. 1B). The patient underwent a mini-open T6 vertebrec-tomy with long segment posterior fusion for kyphosis correction(Supp. Video 1).

Fig. 1. (A) Preoperative 3 foot lateral scoliosis radiograph showing thoracic kyphosis of 70�. (B) Preoperative sagittal T2-weighted MRI demonstrating draping of the spinalcord over the kyphotic apex, with T6 pathological fracture. (C) Postoperative 3 foot lateral scoliosis radiograph showing thoracic kyphosis correction to 53�.

842 D. Chou et al. / Journal of Clinical Neuroscience 21 (2014) 841–845

2.2. Patient 2

An 82-year-old woman who had undergone two prior kypho-plasties for kyphoscoliosis presented with progressively worseningpain. The patient had been relegated to sitting in a lounge chair formost of the day. Standing radiographs demonstrated 92 degrees ofkyphosis (measured from T2 to T12), and compression fractures(two treated with kyphoplasty at T10 and T12) with an apex ofher kyphosis at T11 (Fig. 2A). The patient underwent kyphosis cor-rection via a mini-open vertebrectomy at T11 with polymethyl-methacrylate (PMMA)-augmented posterior instrumentationfrom T5 to L3 (Supp. Video 1).

Fig. 2. (A) Preoperative lateral 3 foot scoliosis radiograph showing thoracickyphosis of 92�. (B) Postoperative lateral 3 foot scoliosis radiograph showingthoracic kyphosis corrected to 65�.

3. Operative techniques

3.1. Pedicle screw fixation

The patient is positioned prone on a radiolucent operating roomtable. The skin is opened midline, but the fascia is preserved. TheMI screws are placed transfascially in the same manner that theyare placed percutaneously. We first placed Jamshidi needles(CareFusion, San Diego, CA, USA) into the pedicles using anterior-posterior and lateral fluoroscopy. Kirschner wires (K-wires) arethen placed into the pedicles. The fascia is then opened over eachK-wire with a #15 blade to accommodate the screws. A tap is thenplaced over the K-wires, the tap is removed, and then the screwsare placed under fluoroscopic guidance. The K-wires are subse-quently removed.

3.2. Mini-open transpedicular vertebrectomy

This technique is described in detail elsewhere [6], but we willbriefly mention our technique here. The fascia over the vertebralbody, superior lamina, and inferior lamina, is opened to facilitatethe corpectomy. The fascia is held open using a self-retainingretractor, and the laminectomy is performed. A temporary rod isplaced, the transpedicular corpectomy is performed, and the discsand posterior longitudinal ligament are removed. We then perform

a trap-door rib head osteotomy to mobilize the rib, and the rib isseparated from the vertebral body [7,8]. An appropriate sizedexpandable cage is placed into the defect. For kyphosis correction,a cage with the largest footplate possible is used, but we do not

Fig. 4. Intraoperative photograph demonstrating a rod contoured to the desiredkyphosis and threaded into proximal screws subfascially. This figure is available incolour at www.sciencedirect.com.

Fig. 5. Intraoperative photograph demonstrating the contour of the spine after thecorrective cantilever maneuver has been performed and the rod placed into caudalscrews. This figure is available in colour at www.sciencedirect.com.

D. Chou et al. / Journal of Clinical Neuroscience 21 (2014) 841–845 843

necessarily expand the cage to its maximal height. This is to allowfor enough motion of the spine to use the cage as a pivot point andto allow us to bring the spine back to correct the kyphosis. Weplace the cage as ventrally as possible to rest on the anteriorapophyseal ring of the vertebral body, which is the strongest partof the vertebral body (Fig. 3).

3.3. Kyphosis correction

Once the cage is in an appropriate position, we choose the tho-racic kyphosis and lumbar lordosis to which we wish to correct thespine, and the rod is then bent into the desired position. The rod isthen threaded subfascially into the proximal screws and lockedinto position with the set screws (Fig. 4). In order to properly per-form the cantilever maneuver, the fascia over the distal screwsmust be cut continuously, not just over each screw, in order to al-low the rod to be seated into all distal screws at once. This is donewith a blade, linking each of the slotted tabs with a single fascialincision. We then gently cantilever the rod down into the distalscrews, always evaluating the spinal cord to ensure no significantbuckling occurs (Fig. 5, Supp. Video 1). The distal set screws areplaced, but not locked tight. Compressing the screws over the cagecan then facilitate further kyphosis correction. Motor evokedpotentials are checked every 5 min for the next 15 min to ensureno changes occur. If motor changes occur, the cord is inspectedto ensure there is no excessive kinking, and the correction can beloosened to check if motor evoked potentials return.

3.4. Fusion and closure

Next, arthrodesis is performed over the vertebrectomy site witheither autologous iliac crest graft or allograft. We use a cross-linkerbecause of the severe instability that ensues after the vertebral col-umn bone removal. The fascia over the vertebrectomy site and overthe screw sites are closed in the usual fashion with interruptedsutures.

4. Outcomes

4.1. Patient 1

There were no intraoperative or postoperative complications.The patient’s postoperative standing 3 foot radiographs demon-strated her thoracic kyphosis improved from 70 degrees to 53 de-grees (Fig. 1C). Her myelopathy and ability to ambulate improved

Fig. 3. Intraoperative photograph of cage reconstruction being performed. Thisfigure is available in colour at www.sciencedirect.com.

as well. At last follow-up, it was found that the patient unfortu-nately died 9 months after surgery from her metastatic disease.

4.2. Patient 2

Although the patient’s kyphosis correction went well, she expe-rienced two complications postoperatively. The patient had a pul-monary embolus secondary to PMMA embolization, which wastreated with anti-coagulation. For 3 weeks, the patient’s pain dra-matically improved, but she subsequently developed new rightleg radiculopathy secondary to an L4 compression fracture. Thiswas treated with an epidural injection, which reduced her pain sig-nificantly. At 6 month follow up, it was found that the patient wasambulating with her walker height raised to accommodate hernew posture, and her Cobb angle was corrected from 92 degreesto 65 degrees (Fig. 2B).

5. Discussion

Posterior based corpectomies and vertebrectomies have beenwell described, and variations on rib head resection have resulted

844 D. Chou et al. / Journal of Clinical Neuroscience 21 (2014) 841–845

in the terms, ‘‘transpedicular corpectomy’’, ‘‘costotransversectory’’,and ‘‘lateral extracavitary approach.’’ When removing the entirevertebral segment for a deformity correction, the term ‘‘vertebralcolumn resection’’ has been used. Regardless of the nomenclatureused, the operations essentially accomplish the same goal: theremoval of the vertebral body from a posterior approach andreconstruction of the anterior spinal column. Add onto this theMI or mini-open approach, and the goal of the operation still re-mains the same.

Traditionally, vertebrectomy is performed through a circumfer-ential approach to the spine, but more recently many surgeonshave conducted alternative spatial approaches to VCR [1–4,9].Currently, posterior-only approaches to vertebrectomy to correctspinal deformity and kyphosis are becoming more common prac-tice [2,10]. The correction gained with VCR can be significant; asmuch as 60 degrees can be corrected. In addition, posterior ap-proaches to multi-level vertebrectomy and corpectomy have beenshown to be feasible and have comparable morbidity to single levelprocedures [11].

The concept of MI surgery continues to evolve. Although a trueMI approach is ideal, there are instances in which a mini-openapproach to the spine provides similar outcomes and still is lessinvasive than open approaches. Kaiser et al. and Chung et al. inde-pendently compared MI (laparoscopic) anterior lumbar interbodyfusion (ALIF) surgeries to mini-open ALIF, and both found thatthere was no advantage to performing a ‘‘true’’ MI ALIF over amini-open one [12,13]. Chi et al. described a mini-open transpedic-ular thoracic discectomy, and revealed that patients who under-went the mini-open discectomy had less blood loss, earlierpostoperative improvement, and decreased postoperative paincompared to the patients who had the open transpedicular discec-tomy [5]. Thus, there is evidence that as the concept of MI surgeryevolves, there may be an increasing role for mini-open proceduresin decreasing morbidity, as well as expanding the ability to per-form traditionally more difficult open procedures in a MI fashion.

We have previously described the mini-open transpedicularcorpectomy, which is essentially a hybrid technique using percuta-neous screws with a small opening for the actual transpedicularcorpectomy [6]. However, many pathological processes, such astumor or infection, are also accompanied by significant kyphosisand spinal deformities. Therefore in addition to the localized spinalpathology, it is important to carefully and correctly address thekyphosis associated with the disease. Given the significant advancesin percutaneous screw technology, kyphosis correction, even insevere deformities, is feasible with a less invasive mini-open approach.

The surgical technique and skills required to perform the mini-open VCR is readily adaptable to surgeons who are comfortablewith performing open posterior vertebrectomies (costo-transver-sectomies, transpedicular corpectomies, or extracavitary ap-proaches) and percutaneous instrumentation and fusion. In themini-open approach, once the screws are placed, the corpectomy,reconstruction, and cantilever correction are very similar to opensurgery. The cantilever procedure does have some nuances as thefascia over the screws that are receiving the rod must be cut in acontinuous fashion. The rod can be technically threaded subfascial-ly over both caudal and rostral receiving screws, but the correctionpower of the rod is minimal this way. The reason for this is becausethe fascia limits the height from which the rod can be cantilevered.We have found it much more effective and ergonomic to correctthe kyphosis using a single fascial opening for the caudal receivingscrews, yet still threading the rod subfascially for the rostralscrews. Another major advantage of the mini-open cantilever pro-cedure is that it avoids an extensive midline incision above and be-low the vertebrectomy site, and therefore, the mini-open approachprotects the musculature and posterior tension band, especiallywhen fusing across spinal junctions.

With regards to the fusion at adjacent non-corpectomy levels,we rely on the data by Dai et al. and Wang et al. [14,15]. In thesetwo studies with 5 and 7 years of follow-up, pedicle screw fixationwithout fusion did not result in implant failure or necessitate theneed for revision surgery (although this was for trauma patients).Our experience of having performed the mini-open transpedicularcorpectomy for the past 4 years is also consistent with these twostudies. If fusion is absolutely necessary, then it is certainly possi-ble to perform arthrodesis at all instrumented levels with the per-cutaneous system. After the K-wire is placed, the facet itself can beexposed either with a series of dilators or a small self-retainingretractor. The facet then can be decorticated and graft materialcan be placed for arthrodesis. However, now that we are correctingkyphoses with percutaneous screw fixation, we may begin to seemanifestations of proximal junction kyphosis and implant failurebecause of the requirement for larger corrective forces placed uponthe spine [16].

For fusion over the corpectomy site, we based our graft materialon the patient’s overall prognosis and baseline factors. For meta-static cancer where survival is short, allograft was used aroundthe cage. If the patient is expected to live a normal life span (suchas an osteomyelitis patient or low grade tumor), iliac crest graftwas harvested and used for fusion. In an elderly patient with akyphosis from either a compression or a burst fracture, we usedlocal vertebrectomy bone for fusion to avoid the morbidity of iliaccrest graft. It remains unclear whether there are significant draw-backs to fusing only at the level of the vertebrectomy and not fus-ing the perivertebrectomy levels. Nonetheless this technique canbe easily modified to address this issue if it arises.

6. Conclusion

Kyphosis correction can be performed with a mini-open verte-brectomy functioning essentially as a VCR, especially with theadvent of more powerful MI instrumentation. However, longerterm studies will be needed to prove the durability of theoperation.

Conflicts of Interest/Disclosures

Dr. Chou has received honoraria from Globus, Depuy and Ortho-fix for teaching. No funding was provided for this study. Otherwise,the authors report no conflict of interest concerning the materialsor methods used in this study or the findings specified in thispaper.

Appendix A. Supplementary material

Supplementary data associated with this article can be found, inthe online version, at http://dx.doi.org/10.1016/j.jocn.2013.10.001.

References

[1] Boachie-Adjei O, Bradford DS. Vertebral column resection and arthrodesis forcomplex spinal deformities. J Spinal Disord 1991;4:193–202.

[2] Lenke LG, Sides BA, Koester LA, et al. Vertebral column resection for thetreatment of severe spinal deformity. Clin Orthop Relat Res 2010;468:687–99.

[3] Hamzaoglu A, Alanay A, Ozturk C, et al. Posterior vertebral column resection insevere spinal deformities: a total of 102 cases. Spine (Phila Pa 1976)2011;36:E340–4.

[4] Wang Y, Zhang Y, Zhang X, et al. Posterior-only multilevel modified vertebralcolumn resection for extremely severe Pott’s kyphotic deformity. Eur Spine J2009;18:1436–41.

[5] Chi JH, Dhall SS, Kanter AS, et al. The Mini-Open transpedicular thoracicdiscectomy: surgical technique and assessment. Neurosurg Focus 2008;25:E5.

[6] Chou D, Lu DC. Mini-open transpedicular corpectomies with expandable cagereconstruction. Technical note. J Neurosurg Spine 2011;14:71–7.

[7] Chou D, Wang VY. Trap-door rib-head osteotomies for posterior placement ofexpandable cages after transpedicular corpectomy: an alternative to lateral

D. Chou et al. / Journal of Clinical Neuroscience 21 (2014) 841–845 845

extracavitary and costotransversectomy approaches. J Neurosurg Spine2009;10:40–5.

[8] Chou D, Eltgroth M, Yang I, et al. Rib head disarticulation for multileveltranspedicular thoracic corpectomies and expandable cage reconstruction.Neurol India 2009;57:469–74.

[9] Helenius I, Serlo J, Pajulo O. The incidence and outcomes of vertebral columnresection in paediatric patients: a population-based, multicentre, follow-upstudy. J Bone Joint Surg Br 2012;94:950–5.

[10] Pappou IP, Papadopoulos EC, Swanson AN, et al. Pott disease in thethoracolumbar spine with marked kyphosis and progressive paraplegianecessitating posterior vertebral column resection and anteriorreconstruction with a cage. Spine (Phila Pa 1976) 2006;31:E123–7.

[11] Lau D, Song Y, Guan Z, et al. Perioperative characteristics, complications, andoutcomes of single-level versus multilevel thoracic corpectomies via modifiedcostotransversectomy approach. Spine (Phila Pa 1976) 2013;38:523–30.

[12] Chung SK, Lee SH, Lim SR, et al. Comparative study of laparoscopic L5–S1fusion versus open mini-ALIF, with a minimum 2-year follow-up. Eur Spine J2003;12:613–7.

[13] Kaiser MG, Haid Jr RW, Subach BR, et al. Comparison of the mini-open versuslaparoscopic approach for anterior lumbar interbody fusion: a retrospectivereview. Neurosurgery 2002;51:97–103 [discussion 103–5].

[14] Dai LY, Jiang LS, Jiang SD. Posterior short-segment fixation with or withoutfusion for thoracolumbar burst fractures. A five to seven-year prospectiverandomized study. J Bone Joint Surg Am 2009;91:1033–41.

[15] Wang ST, Ma HL, Liu CL, et al. Is fusion necessary for surgically treated burstfractures of the thoracolumbar and lumbar spine? A prospective, randomizedstudy. Spine (Phila Pa 1976) 2006;31:2646–52 [discussion 2653].

[16] Marou K, Ha Y, Inoue S, et al. Predictive factors for proximal junctionalkyphosis in long fusions to the sacrum in adult spinal deformity. Spine (PhilaPa 1976) 2013;38:E1469–76.