SpineUniverse Case Study Library

Kyphosis Deformity Correction in Ankylosing Spondylitis


A 63-year-old male with known history of ankylosing spondylitis initially presented to the emergency department after a ground-level fall during which he struck his head. He complained of neck pain but was neurologically intact upon examination. 

Initial imaging revealed minor anterior compression fractures at C7 and T1 without significant loss of vertebral body height (Figures 1-3).  The patient was initially treated nonoperatively with a cervical-thoraco-lumbar orthosis (CTLO). 

lateral cervico thoraco x-rayFigure 1. Image courtesy of Jason M. Cuéllar, MD, and SpineUniverse.com.

(Above):Immediately post-trauma cervical CT mid-sagittal image demonstrates mild C7 and T1 compression fracture without loss of height.  Stigmata of ankylosing spondylitis with ossification along the anterior and posterior longitudinal ligaments with disc space sparing. Midline ankylosis of C2-C6 is apparent on this image.

anterior compression fractures at C7 and T1Figure 2. Image courtesy of Jason M. Cuéllar, MD, and SpineUniverse.com.

 (Above): Paramedian sagittal CT image demonstrates complete ankylosis of the cervico-thoracic facet joints.

anterior compression fractures at C7 and T1Figure 3. Image courtesy of Jason M. Cuéllar, MD, and SpineUniverse.com.

(Above):  Immediately post-trauma cervical MRI demonstrates edema in the C7 and T1 vertebral bodies without spinal cord compression or canal compromise.

Two months later, his post-trauma pain had resolved. However, he complained that although he has had difficulty with his alignment for the past 40 years, since the fall, his line of sight is now worse and no longer tolerable. He could not look people in the eyes, could not stand or walk more than a few minutes and could not sit normally in a chair.


The patient’s examination demonstrated significant sagittal imbalance with visible thoracic kyphosis and loss of lumbar lordosis.  He stands and walks with his knees flexed, and his hips extended in an attempt to improve his downward gaze. He is unable to look upward as he has no cervical motion.  His measured chin-brow-vertebral angle (CBVA) is 52-degrees (Figure 4).

clinical photo of the patient standing, fixed sagittal imbalanceFigure 4. Image courtesy of Jason M. Cuéllar, MD, and SpineUniverse.com.

(Above):  Standing photo of the patient with measurement of the chin-brow-vertebral angle at 52-degrees. The patient has been instructed to straighten his knees as much as possible.

lateral scoliosis x-ray; severe kyphosisFigure 5. Image courtesy of Jason M. Cuéllar, MD, and SpineUniverse.com.(Above): Lateral scoliosis x-ray demonstrates severe kyphosis. Mid-sagittal CT scan provided for easier alignment measurement. Since the patient was completely rigidly fused, angular alignment standing versus supine are similar. Thoracic kyphosis measured at 71-degrees while lumbar kyphosis was measured at 3-degrees (no lordosis).

pre-operative planning; SVA measurementsFigure 6. Image courtesy of Jason M. Cuéllar, MD, and SpineUniverse.com.(Above): Pre-operative planning with Surgimap® demonstrated sagittal vertical axis (SVA) of positive 18.9cm, pelvic incidence of 45-degrees, pelvic tilt of 26-degrees and pelvic incidence lumbar lordosis (PI-LL) mismatch of 24-degrees.  Two planned wedge corrections of 33-degrees and 26-degrees are predicted to provide a corrected SVA of 6.5cm.

wedge corrections; pedicle length measurementsFigure 7. Image courtesy of Jason M. Cuéllar, MD, and SpineUniverse.com.(Above): Measured planned wedge corrections and pedicle length measurements on pre-operative CT scan.

Prior Treatment

The patient was treated in a CTLO, which he found intolerable and without benefit.  He manages his upper back pain with NSAIDs. He makes positional modifications to complete his daily activities and work as he finds it difficult to sit in a normal chair.


Ankylosing spondylitis with severe kyphotic deformity

Suggest Treatment

Indicate how you would treat this patient by completing the following brief survey. Your response will be added to our survey results below.

Selected Treatment

Double pedicle subtraction osteotomy at L1 and L3; pedicle screw instrumentation at T11, T12, L1, L2, L4 and L5; laminectomy at L1 and L3.

intraoperative fluoroscopic images; kyphosis correctionFigure 8. Image courtesy of Jason M. Cuéllar, MD, and SpineUniverse.com.(Above): Intra-operative fluoroscopic images demonstrate creation of 31-degree kyphosis correction at the L3 level by closure of the pedicle subtraction osteotomy (PSO).  Angular change can be seen between the L2 and L4 pedicle screws before and after the closure of the wedge osteotomy.

provisional rodFigure 9. Image courtesy of Jason M. Cuéllar, MD, and SpineUniverse.com.(Above): The use of a provisional rod can be seen to stabilize the spine while the L1 PSO is performed. 

closure of the L1 wedgeFigure 10. Image courtesy of Jason M. Cuéllar, MD, and SpineUniverse.com.(Above): The closure of the L1 wedge is performed. 

intra-operative verification of correctionFigure 11. Image courtesy of Jason M. Cuéllar, MD, and SpineUniverse.com.(Above): Intra-operative verification of correction was performed with the NuvaMap® program (NuVasive) and can also be estimated by measurement of angulation between the pedicle screws on lateral fluoroscopic image.

Intraoperative Complications
The patient’s dura was ossified and fused to the lamina in 3 regions at the L3 and L1 levels. Duraplasty was performed with 3 large AlloDerm patches of 3 x 3cm, 3 x 2 cm and 1 x 2 cm, sutured in place with running locking 4-0 Nurolon.


Postoperatively, the patient remained neurologically intact after the deformity correction. He demonstrated no signs of cerebrospinal fluid leak and was mobilized in a regular room after spending one night in the intensive care unit for large blood loss and fluid shifts.

He was fitted with a thoraco-lumbo-sacral (TLSO) brace.  He mobilized well and was discharged to an inpatient rehab unit on postop day 5. 

At 2.5 months postop, his incisions are healed, and he is off of all narcotic pain medications.  He is very satisfied with his new alignment, now being able to maintain a normal horizontal gaze, walk more than pre-op without rapid fatigue and sit upright in a standard chair.

standing lateral and AP lumbar spine x-raysFigure 12. Image courtesy of Jason M. Cuéllar, MD, and SpineUniverse.com.(Above): Standing lateral and anteroposterior (AP) lumbar spine radiographs and standing lateral scoliosis radiograph 2.5-month postop demonstrating deformity correction.

pre-op and 2/5-month postop comparison of scoliosis x-rayFigure 13. Image courtesy of Jason M. Cuéllar, MD, and SpineUniverse.com.(Above): Pre-op and 2.5-month postop comparison of scoliosis x-ray and clinical image.  Chin brow vertical angle improved from 52-degrees to 18-degrees.

Authors' Discussion

The choice of osteotomy type and the level(s) to perform osteotomy is critical for the success of this case. Xu et al (2015) reported outcomes after one- and two-level PSO for deformity correction in ankylosing spondylitis patients, concluding that correction of 60-degrees and above requires two-level PSOs.1 Yao et al (2016) studied the optimal lowest instrumented vertebrae and optimal PSO levels, reporting best outcomes with instrumentation to L5 with a L3 PSO (+ L1 or T12 PSO for severe deformity).2 Patients with instrumentation to the sacrum reported more long-term pain without any additional benefit.

Why not perform multiple Smith-Peterson osteotomies (SPOs) at the apex of the kyphotic deformity? Liu et al (2015) reported on their results of a systematic review of studies comparing PSO to multiple SPOs for the correction of ankylosing spondylitis.3 While both methods are effective, SPO can cause aortic rupture and death by the fracture of ossified anterior longitudinal ligaments.

How much correction is optimal? Another difficult pre-operative planning consideration is determining how much to correct the chin brow vertical angle (CBVA). Song et al (2016) reported that the optimal postoperative CBVA is between 10- and 20-degrees.4 Patients are far less satisfied if over-correction occurs, as this makes it difficult for them to see their feet and the ground during ambulation. Whether or not the patient has cranio-cervical motion is also critical when determining CBVA correction.

1. Liu H, Yang C, Zheng Z, Ding W, Wang J, Wang H, Li S. Comparison of Smith-Petersen osteotomy and pedicle subtraction osteotomy for the correction of thoracolumbar kyphotic deformity in ankylosing spondylitis: a systematic review and meta-analysis. Spine (Phila Pa 1976). 2015 Apr 15;40(8):570-9. PMID: 25868095.
2. Song K, Su X, Zhang Y, Liu C, Tang X, Zhang G, Zheng G, Cui G, Zhang X, Mao K, Wang Z, Wang Y. Optimal chin-brow vertical angle for sagittal visual fields in ankylosing spondylitis kyphosis. Eur Spine J. 2016 Aug;25(8):2596-604. PMID: 27146808.
3. Xu H, Zhang Y, Zhao Y, Zhang X, Xiao S, Wang Y. Radiologic and clinical outcomes comparison between single- and two-level pedicle subtraction osteotomies in correcting ankylosing spondylitis kyphosis. Spine J. 2015 Feb 1;15(2):290-7. PubMed PMID: 25264319.
4. Yao Z, Zheng G, Zhang Y, Wang Z, Zhang X, Cui G, Wang Y. Selection of lowest instrumented vertebra for thoracolumbar kyphosis in ankylosing spondylitis. Spine (Phila Pa 1976). 2016 Apr;41(7):591-7. PMID: 27018899.

Peer Case Discussion

This is a very interesting case with many learning points contained within it.

• Even minor injuries in patients with ankylosing spondylitis (AS) or diffuse idiopathic skeletal hyperostosis (DISH) can have big consequences. When these patients present to the emergency room or office with pain after minor trauma, they have a fracture until proven otherwise. Often the fractures are extension distraction fractures, which are highly unstable and can cause neurologic deficit if not stabilized.

• Rapid loss of alignment and gaze (chin-brow vertical angle) is also not uncommon.

• Choice of where to do the correction is not always straightforward, and these authors did a great job using modern technology to help guide in decision making.

• When correcting a patient with AS, ideally you want to leave their head slightly down (10-15 degrees) so they can read, walk stairs, etcetera.

• Dural adhesion/absence is not uncommon in these patients and should be planned for.

Congratulations to the authors on an excellent job!!

Community Case Discussion (0 comments)

SpineUniverse invites spine professionals to share their thoughts on this case.


Get new patient cases delivered to your inbox

Sign up for our healthcare professional eNewsletter, SpineMonitor.
Sign Up!