J Neurosurg Spine 16:351358, 2012

Clinical and radiographic evaluation of posterior surgical

correction for the treatment of moderate to severe
post-tuberculosis kyphosis in 36 cases with a minimum
2-year follow-up
Clinical article
Yan Zeng, M.D.,1 Zhongqiang Chen, M.D.,1 Qiang Qi, M.D.,1 Zhaoqing Guo, M.D.,1
Weishi Li, M.D.,1 Chuiguo Sun, M.D.,1 and Andrew P. White, M.D. 2
Orthopedic Department, Peking University 3rd Hospital, Beijing, China; and 2Harvard Medical School, Carl
J. Shapiro Department of Orthopaedic Surgery, Beth Israel Deaconess Medical Center, Boston, Massachusetts

Object. The object of this study was to compare the clinical and radiographic outcomes of 36 patients with posttuberculosis kyphosis who underwent one of two types of osteotomy.
Methods. Each patient underwent single-stage correction via a posterior surgical approach. A modified pedicle
subtraction osteotomy (mPSO) was performed when the kyphotic deformity was less than 70 (7 cases), whereas a
posterior vertebral column resection (VCR) was performed when the kyphotic deformity exceeded 70 (29 cases).
Full-length standing radiographs were obtained before surgery and at follow-up visits. These images were used to
measure the kyphosis angle; sagittal alignment of the lumbar, thoracic, and cervical regions; and sagittal balance of
the spine. Back pain was rated using the visual analog scale (VAS), and neurological function was classified based
on the American Spinal Injury Association (ASIA) grading system. Each patients overall satisfaction with surgical
treatment was measured with the Patient Satisfaction Index. For purposes of comparison, patients were studied in 2
groups based on the region of their kyphotic apex. Half of the cohort had apical kyphosis in the lower thoracic spine
or thoracolumbar junction (TL group). Using both radiographic and clinical assessments, the authors compared this
group with the other half of the patients who had apical kyphosis in the upper to mid thoracic spine (MT group).
Results. The cohort included 15 males and 21 females, with an average age of 34 years at the time of surgery.
The minimum follow-up was 24 months, and the mean follow-up was 31 months. Following surgery, kyphosis across
the treated segments was reduced by an average of 60. Lumbar lordosis also improved by an average of 24, and
thoracic kyphosis improved by an average of 20. Both back pain and neurological function improved after surgical
treatment. There was a 67% improvement in VAS scores, and 13 of the 36 patients had improvement in their ASIA
grade. The 2 surgical procedures used for deformity correction (mPSO and VCR) demonstrated comparable radiographic and clinical results. Note, however, that differences were found in both radiographic and clinical outcomes
in comparing patients who had lower thoracic or thoracolumbar (TL group) versus upper to midthoracic (MT group)
apical kyphosis.
Conclusions. Posterior tubercular kyphosis can be effectively improved through corrective surgery, and deformity correction can be accompanied by improvement in clinical symptoms. When appropriately selected, both
the mPSO and the VCR can be expected to yield satisfactory reduction of post-tuberculosis kyphotic deformities.
Differences in radiographic and clinical outcomes should be anticipated, however, when treating such deformities in
different regions of the spine.
(http://thejns.org/doi/abs/10.3171/2011.12.SPINE11568)

Key Words post-tuberculosis kyphosis moderate to severe kyphosis

kyphosis correction posterior corrective surgery Pott disease
degenerative disease

ost-tuberculosis

kyphosis can result when the

anterior column becomes structurally incompetent after the destruction of infected vertebrae. In

Abbreviations used in this paper: ASIA = American Spinal Injury

Association; mPSO = modified pedicle subtraction osteotomy; PSI
= Patient Satisfaction Index; VAS = visual analog scale; VCR =
vertebral column resection.

J Neurosurg: Spine / Volume 16 / April 2012

children, progressively severe kyphosis can also develop

with ongoing spinal growth, particularly when multiple
vertebrae are involved. The kyphotic deformity can be
significant in many cases.
This article contains some figures that are displayed in color
online but in black and white in the print edition.

351

Y. Zeng et al.
Patients with significant kyphosis typically present
with cosmetic and functional problems related to the biomechanical changes associated with the kyphosis. The influence of spinal balance on function and quality of life
is well established.1,9 Substantial back pain and functional
disability are common in patients with kyphosis.14 Many
patients with post-tuberculosis kyphosis also present with
neurological symptoms. These symptoms may be related
to spinal cord compression or over-distraction of the cord
over the kyphotic deformity. When symptoms related to
significant post-tuberculosis deformities cannot be adequately managed conservatively, surgical correction may
be required.
Since post-tuberculosis kyphosis is one of the most
severe types of spinal kyphosis, surgical correction may
be associated with significant risks and challenges. A
variety of deformity correction techniques have been reported.5,12,13,16,17,19 The majority of these reports focus on
improving deformity correction and reducing surgical
complications. However, surgical decision making, particularly as regards the post-tuberculosis deformities, remains
a challenge.
Because of the uncertainty associated with surgical
decision making, we compared 2 posterior surgical correction procedures used in a cohort of patients with posttuberculosis kyphosis. The surgical procedureeither an
mPSO or a posterior VCRwas selected based on the
severity of the kyphosis. We measured radiographic and
clinical results to compare patient outcomes in an effort to
guide future care.
Clinical Data

Methods

Between July 2005 and September 2009, 36 patients

(15 male and 21 female) with post-tuberculosis kyphosis
underwent posterior surgical correction in a single medical
center in Beijing. All patients had a history of spinal tuberculosis and had previously received chemotherapy during
the active stage of their vertebral tuberculosis infection.
The average patient age at the time of surgery was 34.2
years old (range 1356 years). All patients had a kyphotic
angle > 50. In 18 cases, the apex of the kyphosis was at
T-10 or below, within the lower thoracic spine or thoracolumbar junction (TL group). In the remaining 18 cases, the
kyphotic apex was above T-10, in the upper to midthoracic
spine (MT group). Patients presented with lower extremity
neurological symptoms alone (15 cases), back pain alone (4
cases), both neurological symptoms and back pain (9 cases),
or with deformity-related cosmetic concerns not associated
with pain or neurological symptoms (8 cases). All of the
symptomatic patients had undergone ineffective conservative treatment for at least 6 months. All of the asymptomatic patients had had severe kyphosis (> 90 degrees) for at
least 5 years. To prevent a recurrence of active tuberculous
infection, all patients were prescribed oral medications of
combined isoniazid, ethambutol, and rifampicin from 2
weeks before surgery to 6 months after surgery.

Surgical Procedure

Each patient underwent posterior surgical treatment

352

alone via one of two surgical methods. The method was selected preoperatively based solely on the severity of the kyphosis. For patients with kyphosis less than 70, an mPSO
was used. This procedure (Fig. 1 upper) includes the same
posterior element resection as a traditional PSO but augments the traditional closing wedge osteotomy with opening (lengthening) of the anterior column.16 An interbody
device is used to maintain the anterior column opening in
this case. For patients with kyphosis more than 70, a VCR
was used. In the VCR, a dual-axis rotation correction is
performed to maximize correction (Fig. 1 lower). In our
cohort, the 7 patients with kyphosis less than 70 were assigned to mPSO and the 29 patients with kyphosis exceeding 70 were assigned to VCR.
After inducing general endotracheal anesthesia, the
patient was placed prone. Care was taken to monitor and
maintain normal blood pressure during the operation to
reduce the risk of spinal cord ischemia. Intraoperative neuromonitoring was used for each case, including recording
of somatosensory evoked potentials and transcranial motor
evoked potentials. The posterior elements were exposed,
and bilateral pedicle screws were placed at 2, 3, or 4 vertebrae above and below the apex of the kyphosis. For patients undergoing mPSO, 2 or 3 vertebrae would typically
be instrumented on either side of the apex. For patients
undergoing VCR, 3 or 4 vertebrae would be instrumented above and below the apex of the kyphosis. The PSO
or VCR was then performed. After complete separation of
cephalic and caudal segments, deformity correction was
undertaken. Care was taken to maintain appropriate laxity
of the spinal cord during the correction procedures. For
the mPSO corrections, an interbody cage with bone graft
was placed anteriorly at the level of the osteotomy to augment the correction before closing the posterior elements
(Fig. 1 upper). For the VCR corrections, dual axis rotation
was used, and titanium mesh packed with autologous bone
graft was placed into the osteotomy gap for reconstruction
of the anterior column (Fig. 1 lower).
Radiographic Evaluation

All patients underwent standing full-length spine radiography before surgery and at each postoperative follow-up. The sagittal plane balance was evaluated, and the
kyphosis angle, lumbar lordosis, thoracic kyphosis, and
cervical lordosis were measured in each case.
Overall sagittal plane balance was measured using
the C-7 plumb line on full-length standing lateral spine
radiographs. The horizontal distance from the C-7 plumb
line to the posterior-superior corner of S-1 or the sacral
promontory was measured (Fig. 2). Negative sagittal balance was noted when the C-7 plumb line fell posterior to
the posterior-superior corner of S-1, and positive sagittal
balance was described as the C-7 plumb line falling anterior to the sacral promontory. In our study, normal sagittal balance was designated when the C-7 plumb line fell
between the posterior-superior corner of S-1 and the sacral
promontory.1,9
The kyphosis angle was defined on lateral standing radiographs as the angle between the superior endplate of the
first morphologically normal vertebra above the deformity
J Neurosurg: Spine / Volume 16 / April 2012

Evaluation of surgical treatment for post-tubercular kyphosis

Fig. 1. Images depicting the 2 osteotomies used to treat post-tuberculosis kyphosis, accompanied by deformity correction
and reconstruction techniques. Upper: The mPSO is depicted. After pedicle screw instrumentation is placed above and below
the planned osteotomy segment, a traditional pedicle subtraction is performed. The correction maneuver involves the traditional
closing (shortening) of the posterior column and also introduces an opening (lengthening) of the anterior column. Lengthening of
the anterior column improved the efficiency of the traditional correction and is maintained with an interbody device placed at the
anterior aspect of the osteotomy. Lower: The VCR is depicted. This VCR takes advantage of dual axial rotation correction and
was used for correcting the most severe kyphotic deformities in our cohort. After placing pedicle screw instruments several segments above and below the planned osteotomy, the entirety of the deformed segment is resected. The resection often includes
more than 1 vertebra, and the posterior elements cannot be closed after correction. Printed with the permission of Z. Chen, 2011.

and the inferior endplate of the first morphologically normal vertebra below the deformity (Fig. 2). Lumbar lordosis
was defined as the angle between the superior endplate of
L-1 and the superior endplate of S-1 on lateral standing
radiographs. If the apex of the kyphosis involved an upper
lumbar segment, the angle between the superior endplate of
the first normal vertebra below the deformity and the superior endplate of S-1 was used to measure lumbar lordosis,
in lieu of the angle between L-1 and S-1. Thoracic kyphosis
was defined as the angle between the superior endplate of
T-1 and the inferior endplate of T-12. If the upper thoracic
spine was not clearly visible, the angle between the superior endplate of T-4 and the inferior endplate of T-12 was
measured. If the apex of the kyphosis involved the lower
thoracic segment (caudal to T-10), the first normal vertebra
J Neurosurg: Spine / Volume 16 / April 2012

above the deformity was used instead of T-12. In cases in

which the apex of the kyphosis was above T-10, we could
not measure thoracic kyphosis. In total, 18 cases accommodated the measurement of thoracic kyphosis before and
after surgery. Cervical lordosis was defined as the angle
between the superior endplate of C-3 and the inferior endplate of C-7 on lateral radiographs.
Clinical Evaluation

Back pain was measured using the VAS before surgery

and at follow-up visits. Neurological function was evaluated using the ASIA grading system. The PSI, the patient
satisfaction subscale that was part of the North American
Spine Society Lumbar Spine Outcome Assessment, was
used to evaluate patient satisfaction with treatment.7
353

Y. Zeng et al.
For purposes of comparison, the patients were categorized into 2 groups. One group had a kyphotic apex in the
lower thoracic spine or thoracolumbar junction at or below
T-10 (TL group, 18 cases). The second group had the apex
of the kyphosis in the upper to midthoracic spine, with the
kyphotic apex above T-10 (MT group, 18 cases). Radiographic and clinical evaluations were performed in both
of the groups.

Results

Fig. 2. Drawing depicting the lines used to measure kyphosis angle,

lumbar lordosis, thoracic kyphosis, cervical lordosis, and sagittal balance. Kyphosis angle (green line) is the angle between the superior
endplate of the first normal vertebra above the deformity and the inferior endplate of the first normal vertebra below the deformity. Lumbar
lordosis (red line) is the angle between the superior endplate of L-1 and
the superior endplate of S-1. Thoracic kyphosis (orange line) is the
angle between the superior endplate of T-1 and the inferior endplate of
T-12. Cervical lordosis (purple line) is the angle between the superior
endplate of C-3 and the inferior endplate of C-7. Sagittal balance (blue
line) is the horizontal distance from the C-7 plumb line to the posteriorsuperior corner of S-1. Printed with the permission of Z. Chen, 2011.

354

The average preoperative kyphosis angle was 89.3

(range 54130). Preoperative thoracic kyphosis in our
cohort was less than the normal thoracic kyphosis,15 averaging -8.6 (range -39 to 5). The average lumbar lordosis
was 72 (range 37 to 113), which was greater than normal.11 The average cervical lordosis was 6.5 (range -24
to 29). The average preoperative C-7 plumb line fell 16.7
mm posterior to the posterior-superior corner of S-1 (range
-125 to 65 mm). Negative sagittal balance was demonstrated in 17 cases, with an average posterior shift of 44.6 mm.
Sagittal plane balance was well maintained in 14 cases.
Positive sagittal balance was found in 5 cases, with an average anterior shift of 31.4 mm. The radiographic measurements before surgery are listed in Table 1.
The average VAS score for back pain was 2.0 in our
cohort, including the patients who had no back pain and reported a VAS score of 0 before surgery. Patients with normal sagittal balance reported an average preoperative VAS
score of 1.6; those with negative sagittal balance, a score of
2.6; and those with positive sagittal balance, a score of 1.2.
This difference in preoperative VAS scores based on sagittal balance was not statistically significant (p > 0.1). The
preoperative ASIA grade for neurological functions was E
in 12 cases, D in 16, and C in 8 (Table 1).
After surgery, patients were followed up for a minimum of 2 years, with an average follow-up of 31.3 months
(range 2460 months). The kyphosis angle decreased to
29.3 on average following surgical correction, with a
correction rate of 67.2%. The thoracic kyphosis angle improved postoperatively to 11.7 on average, and the average
lumbar lordosis angle improved to 47.6 (Table 1). These
preoperative to postoperative spinal alignment changes
were statistically significant (p < 0.05; Fig. 3). The average
postoperative cervical lordosis angle was 4.7, however,
which was similar to the preoperative cervical lordosis.
The average postoperative C-7 plumb line fell 8.2 mm
(range -80 to 9.9 mm) posterior to the posterior-superior
corner of S-1 (Table 1). Among the 36 patients, the number
found to have negative sagittal balance following surgery
decreased to 11 cases, the number with well-maintained
sagittal balance increased to 24 cases, and the number with
positive balance decreased to 1. Among the patients who
had negative sagittal balance before surgery, the C-7 plumb
line shifted posteriorly by an average of 13.3 mm following surgery, which was a statistically significant change (p
< 0.05).
The average back pain VAS score was 0.7 after surgical correction. This figure represented a 67.1% improvement compared with preoperative back pain and was statistically significant (p < 0.05). Some patients in the cohort
J Neurosurg: Spine / Volume 16 / April 2012

Evaluation of surgical treatment for post-tubercular kyphosis

TABLE 1: Preoperative and postoperative radiographic measurements and clinical data*

Time

Kyphotic
Angle ()

Thoracic
Kyphosis ()

Lumbar
Lordosis ()

Cervical
Lordosis ()

Sagittal Balance (mm)

Back Pain
VAS Score

BS
at FU

89.3 22.9
29.3 16.2

8.6 11.1
11.7 15.5

72.0 20.4
47.6 12.1

6.5 16.3
4.7 13.6

16.7 36.6
8.2 17.8

2.0 2.9
0.7 1.3

ASIA Grade (no. of cases)

E

12
21

16
11

8
4

* Radiographic measurements and VAS scores are averaged, with all patients included in the average. Abbreviations: BS = before surgery; FU =
follow-up.

were also found to have improved neurological function

following surgical treatment. The postoperative ASIA
grade was E in 21 cases, D in 11, and C in 4 (Table 1).
Patient satisfaction with surgery was generally favorable. The patients in 22 cases were completely satisfied
with surgery and those in 10 cases were partially satisfied. This finding represented an overall satisfaction rate
of 88.9% and a complete satisfaction rate of 61.1%. The
patients who were completely satisfied had an average kyphosis angle of 23.9 after surgery, while the remaining
patients had an average postoperative kyphosis angle of
37.7, which represented a statistically significant difference between the 2 groups (p < 0.05).
The radiological and clinical outcomes comparing
the 2 surgical procedures of mPSO and VCR are listed
in Tables 2 and 3. In the VCR group, both preoperative
kyphosis and lumbar lordosis were significantly greater
than in the mPSO group, whereas thoracic kyphosis was
significantly less in the VCR group (p < 0.05). After surgery, both groups demonstrated good correction of kyphosis, while the lumbar lordosis and thoracic kyphosis were
found to be similar between the 2 groups. With regard to
back pain (VAS score) and neurological function (ASIA
grade), the mPSO and VCR groups had similar improvements following surgery. The PSI was similar following either procedure. Since these 3 clinical outcomes (back pain,
neurological function, self-reported patient satisfaction)
were comparable between the mPSO and VCR groups, it

appears that these 2 surgical procedures were similarly effective, despite the fact that VCR was used to correct more
severe deformities, that is, those exceeding 70 of kyphosis.
Radiographic and clinical outcomes were also compared between the patients with lower thoracic to thoracolumbar (TL group) and upper to midthoracic (MT group)
post-tuberculosis kyphosis. A comparison of radiographic
outcomes is presented in Table 4, and a comparison of clinical outcomes is presented in Table 5.
The severity of the preoperative kyphosis angle was
slightly greater in the TL group than in the MT group. Following surgery, the kyphotic angle significantly improved
in both groups (p < 0.05). The postoperative kyphosis was
less in the TL group, however, signifying that the patients
with lower thoracic to thoracolumbar deformity had a
greater reduction in kyphosis with surgery. Lumbar lordosis was also greater in the TL group before surgery and less
in the TL group after surgery, concordant with the greater
deformity correction achieved in the patients with lower
thoracic to thoracolumbar kyphotic deformities.
Patients with lower thoracic to thoracolumbar deformity reported greater back pain than those in the MT
group, with a significant difference found between the
groups (p < 0.05). The postoperative reduction in back pain
appeared to be greater in the TL group, however. Neurological deficits were recorded in 9 of the patients (50%) in
the TL group and in 15 (83%) of those in the MT group,
representing a significant difference between the 2 groups

Fig. 3. Standing radiographs obtained in a 16-year-old male patient, depicting the correction of his post-tubercular deformity. A: Posteroanterior and lateral radiographs demonstrating the preoperative deformity between T-10 and T-12. Note that
the kyphotic angle was 109, lumbar lordosis was 78, and thoracic kyphosis was 5. The patient underwent posterior surgical
correction via VCR with dual axial rotation. B: Images obtained 9 days after surgery, demonstrating adequate correction of
kyphosis. C: Final radiographic data were collected 36 months after surgery from standing radiographs. The kyphotic angle
was corrected to 32, the lumbar lordosis was 41, and the thoracic kyphosis was 40. D: A CT scan obtained 36 months after
surgery, demonstrating healed fusion, with remodeled bone graft seen within the anterior column space at the site of the VCR
osteotomy.

J Neurosurg: Spine / Volume 16 / April 2012

355

Y. Zeng et al.
TABLE 2: Radiographic measurements made before and after surgical correction*

Proc

No.

Kyphotic Angle ()
BS

Thoracic Kyphosis ()

At FU

mPSO 7 63.3 10.7 21.3 28.1

VCR
29 95.6 20.5 31.2 11.9

BS

Lumbar Lordosis ()

At FU

1.7 2.9 13.7 5.5

10.6 11.0 11.3 16.9

BS

At FU

54.0 15.7 46.5 15.7

76.4 19.2 47.9 11.4

Cervical Lordosis ()
BS

Sagittal Balance (mm)

At FU

BS

17.9 9.4 18.8 6.9

4.5 16.5 4.9 12.9

At FU

1.9 30.8 11.9 22.1

21.2 37.0 7.3 16.9

* Measurements are presented as an average for patients who underwent each type of corrective osteotomy. Abbreviation: Proc = procedure.
Number of cases.

losing spondylitis who had undergone surgical correction

of flat-back deformity. The surgical results were primarily
related to the improvement of sagittal balance. The patients
who had a poor sagittal contour correction were less satisfied with their surgical results.
Our patients with post-tuberculosis deformity had a
more focal and typically more severe kyphotic angle than
the patients described in prior studies evaluating surgical treatment of symptomatic kyphotic deformities. Posttuberculosis kyphosis does also affect the alignment of
morphologically normal (noninfected) segments above
and below the apical kyphosis. Resulting negative sagittal
imbalance can be related to lumbar hyperlordosis and thoracic hypokyphosis.
Patients with post-tuberculosis kyphosis typically do
not have any forward balance tendency other than that produced by their local kyphotic deformity. In an attempt to
preserve sagittal balance, these patients will develop compensatory lordosis in the lumbar and thoracic spine. Significantly increased lumbar lordosis can alter the normal
biomechanics of the functional spinal unit, with increased
loads placed on the facet joints and increased shear forces
placed on the intervertebral discs. This may accelerate degeneration, secondarily resulting in segmental instability
and posterior slippage and contributing to low-back pain.4
The abnormal alignment can also negatively affect the mechanics of the paraspinal musculature, leading to fatigue,
which can also contribute to back pain. Following a successful surgical correction of the focal kyphosis, the compensatory lumbar hyperlordosis and thoracic hypokyphosis can be reduced. This reduction in exaggerated regional
alignment can facilitate the recovery of normal overall spinal balance. The radiographic data in our cohort supported
this assertion, with a significant improvement of negative
sagittal balance.
A posterior closing wedge osteotomy, such as the PSO,
is the most widely used surgical technique to address regional kyphosis.2,3,18,20 When attempting to address severe
post-tuberculosis kyphosis, however, the prominent apex
and overlapping ribs will hinder the closing correction after osteotomy. Furthermore, direct osteotomy closure may

(p < 0.05). The postoperative improvement in neurological

grade, however, was similar in each group.
Surgical complications are reported in Table 6. One
patient who underwent VCR for T58 kyphosis had transient postoperative paralysis. This intraoperative injury
may have been related to prolonged hypotension, contributing to spinal cord ischemia during surgery. Intraoperative
change of somatosensory evoked potential and transcranial
motor evoked potential monitoring predicted the postoperative paralysis. The patient was treated with fluid resuscitation, methylprednisolone, and neurotrophic medications.
The patient exhibited satisfactory recovery 6 months after
surgery. Another patient who underwent VCR for treatment of a T57 kyphosis had a nonunion, exhibiting loss
of fixation and recurrent kyphosis 30 months after the surgery. This patient was treated with revision surgery.

Discussion

Among the various manifestations of kyphosis, posttuberculosis kyphosis is one of the most severe and most
symptomatic. While some patients with this kyphotic type
present with cosmetic symptoms alone, the majority of patients suffer significant back pain and neurological symptoms. Back pain is typically related to mechanical insufficiency of the infected vertebrae, the secondary profile
change of the thoracic and lumbar spine, and sagittal plane
imbalance of the spine. Neurological symptoms may be related not only to direct compression of the spinal cord, but
also to over-tension on the spinal cord across the apex of
the kyphosis.
Several studies have documented the clinical impact
of sagittal imbalance. Glassman and colleauges9 described
352 cases of positive spinal balance in adults. They reported that pain, Oswestry Disability Index, and 12-Item Short
Form Health Survey scores were all negatively influenced
by increasingly severe sagittal imbalance. The majority of
patients in their cohort had kyphoscoliosis and Scheuermann disease, however, with only a few cases of severe
focal kyphosis. Prior to that study, Booth and colleagues1
reported on a series of patients with scoliosis and anky-

TABLE 3: Clinical measurements made before and after surgical correction

Back Pain VAS Score

ASIA Grade (no. of cases)

D

356

Procedure

No. of Cases

BS

At FU

BS

At FU

BS

At FU

BS

At FU

PSI (%)

mPSO
VCR

7
29

3.1 4.0
1.8 2.6

0.9 1.6
0.6 1.3

2
10

4
17

4
12

2
9

1
7

1
3

85.7
89.7

J Neurosurg: Spine / Volume 16 / April 2012

Evaluation of surgical treatment for post-tubercular kyphosis

TABLE 4: Radiographic measurements as an average before and after surgery, categorized by anatomical region of
the post-tuberculosis deformity*
No. of
Group Cases
TL
MT

18
18

Kyphotic Angle ()
BS

Lumbar Lordosis ()

At FU

BS

92.6 25.1 25.6 13.7

85.9 20.7 33.0 18.1

Cervical Lordosis ()

At FU

78.0 22.2 46.9 12.1

66.1 17.1 48.4 12.4

BS

Sagittal Balance (mm)

At FU

0.9 17.7 1.1 10.7

10.9 14.0 9.6 14.2

BS

At FU

8.7 37.0 4.7 10.2

24.7 35.5 12.0 23.1

* MT = kyphosis in upper or midthoracic spine (above T-10); TL = kyphosis in the thoracolumbar region (T-10 and below).

lead to over-shortening of the spinal cord, increasing the

risk of iatrogenic neurological injury. For these reasons,
many authors have suggested that posterior closing osteotomies should be limited to treating kyphotic deformities less than 40.6,8,13 Indeed, prior attempts to treat severe
post-tuberculosis kyphotic deformities by using posterior
closing wedge osteotomies were difficult and inadequate.
Post-tuberculosis kyphosis typically involves multiple
vertebral levels and is frequently associated with a significant disparity between the height of the anterior and posterior columns. For this reason, in part, the anterior column
often must be completely resected to achieve adequate correction. In a previous experience, posterior closing wedge
osteotomy (PSO) was associated with 2.5 of correction
per mm of posterior closing, with a maximum correction
of 45. However, our mPSO, which augments the correction by lengthening the anterior column, was shown to be
more efficient than this, with a mean angle of correction
of 6.2 per mm of posterior closing.10 In prior series, we
also demonstrated a high degree of kyphosis correction via
either mPSO or VCR while preserving or improving spinal
cord function.5,16
Patients in the current series all had post-tuberculosis
kyphosis exceeding 50. Each of these deformities was
adequately corrected with either mPSO or VCR. It is important to note, however, that there is a substantial risk of
complications associated with these procedures. Significant complications were seen in our patients, particularly
associated with VCR. The majority of patients (29 of 36)
were treated with VCR for the more severe kyphotic deformities, all exceeding 70. We reported CSF leakage, nerve
root injury, transient paralysis, and pseudarthrosis in the
patients who underwent correction via VCR. Careful vigilance should be maintained when treating these patients
with severe focal kyphosis, as they do have a high risk of
complications, including neurological injuries, during surgical correction.
Post-tuberculosis kyphosis most frequently affects the
thoracolumbar junction, followed by the middle and lower

thoracic spine. In this study, we categorized patients into

2 groups according to the region of their kyphotic apex.
The patients with apical kyphosis above T-10 (MT group)
reported less back pain than the patients with thoracolumbar kyphosis (TL group). This may be related to the impact
of the deformity on lumbar lordosis. It may also be related
to the stabilizing effect of the thorax in the patients with
kyphosis above T-10. Preoperative neurological function
was worse in the MT group, which may be related to the
sensitivity of the thoracic spinal cord to either direct compression or tension over the kyphotic deformity. Following surgery, neurological function was improved in both
groups, but greater improvements were seen in the patients
with post-tuberculosis kyphosis above T-10.
The radiographic and clinical results were good in this
series of patients treated with surgical correction. The overall satisfaction rate (PSI) was 88.9%, and the complete satisfaction rate was 61.1%. With regard to patient satisfaction
outcomes, we noted that the average postoperative kyphosis of the completely satisfied patients was significantly
better than that of the remaining patients who reported
relatively less satisfaction. This finding supports the prior
claim that improvement in clinical symptoms is related to
the adequacy of deformity correction and the appropriate
restoration of sagittal balance. This finding may also reflect
the patients appreciation for their cosmetic improvement
after correction of this severe deformity.

Conclusions

Posterior tubercular kyphosis can be effectively improved through corrective surgery, and deformity correction can be accompanied by improvement in clinical
symptoms. When appropriately selected, both the mPSO
and the VCR can be expected to yield satisfactory reduction of post-tuberculosis kyphotic deformities. Differences
in radiographic and clinical outcomes should be anticipated, however, when treating such deformities in different
regions of the spine.

TABLE 5: Clinical measurements are presented before and after surgery, categorized by the anatomical region of the
post-tuberculosis deformity
Back Pain VAS Score*

ASIA Grade (no. of cases)

D

Group

No. of Cases

BS

At FU

BS

At FU

BS

At FU

BS

At FU

TL
MT

18
18

3.0 3.2
1.1 2.3

0.9 1.6
0.4 0.9

9
3

11
10

5
11

5
6

4
4

2
2

* The VAS scores are presented as an average of all patients in each category.

J Neurosurg: Spine / Volume 16 / April 2012

357

Y. Zeng et al.
TABLE 6: Complications observed in the surgical correction in 36 patients
Complication

No.*

Duration

Treatment

CSF leakage
nerve root injury
neurological deficit
pseudarthrosis

2
2
1
1

during surgery
during surgery
during surgery
30 mos postop

repair
fluid resuscitation, methylprednisolone, & neurotropic medicines
fluid resuscitation, methylprednisolone, & neurotropic medicines
revision surgery

* A VCR had been performed in all cases.

Disclosure
The authors report no conflict of interest concerning the materials or methods used in this study or the findings specified in this
paper.
Author contributions to the study and manuscript preparation
include the following. Conception and design: Chen. Acquisition
of data: Zeng, Qi, Guo, Li, Sun. Analysis and interpretation of data:
Zeng. Drafting the article: Zeng. Critically revising the article: all
authors. Reviewed submitted version of manuscript: all authors.
Approved the final version of the manuscript on behalf of all authors:
Chen. Statistical analysis: Zeng. Study supervision: Chen.
References
1. Booth KC, Bridwell KH, Lenke LG, Baldus CR, Blanke KM:
Complications and predictive factors for the successful treatment of flatback deformity (fixed sagittal imbalance). Spine
(Phila Pa 1976) 24:17121720, 1999
2. Bridwell KH, Lewis SJ, Lenke LG, Baldus C, Blanke K: Pedicle
subtraction osteotomy for the treatment of fixed sagittal imbalance. J Bone Joint Surg Am 85-A:454463, 2003
3. Chen IH, Chien JT, Yu TC: Transpedicular wedge osteotomy
for correction of thoracolumbar kyphosis in ankylosing spondylitis: experience with 78 patients. Spine (Phila Pa 1976)
26:E354E360, 2001
4. Chen Z, Dang G, Guo Z, Ji L, Qi Q: [Influence of solid thoracolumbar kyphosis on lumbar spine and surgical treatment.]
Zhonghua Wai Ke Za Zhi 38:824826, 2000 (Chinese)
5. Chen Z, Zeng Y, Li W, Guo Z, Qi Q, Sun C: Apical segmental
resection osteotomy with dual axial rotation corrective technique for severe focal kyphosis of the thoracolumbar spine.
Clinical article. J Neurosurg Spine 14:106113, 2011
6. Chen ZQ, Li WS, Guo ZQ, Qi Q, Dang GT: [Surgical correction
of post-traumatic kyphosis of thoracolumbar spine.] Zhonghua
Wai Ke Za Zhi 43:201204, 2005 (Chinese)
7. Daltroy LH, Cats-Baril WL, Katz JN, Fossel AH, Liang MH:
The North American spine society lumbar spine outcome assessment instrument: reliability and validity tests. Spine (Phila
Pa 1976) 21:741749, 1996
8. Gertzbein SD, Harris MB: Wedge osteotomy for the correction
of post-traumatic kyphosis. A new technique and a report of
three cases. Spine (Phila Pa 1976) 17:374379, 1992
9.Glassman SD, Bridwell K, Dimar JR, Horton W, Berven S,
Schwab F: The impact of positive sagittal balance in adult spinal deformity. Spine (Phila Pa 1976) 30:20242029, 2005
10. Hao CK, Li WS, Chen ZQ: The height of the osteotomy and

358

the correction of the kyphotic angle in thoracolumbar kyphosis.

Chin Med J (Engl) 121:19061910, 2008
11.Jackson RP, McManus AC: Radiographic analysis of sagittal
plane alignment and balance in standing volunteers and patients with low back pain matched for age, sex, and size. A prospective controlled clinical study. Spine (Phila Pa 1976) 19:
16111618, 1994
12. Kawahara N, Tomita K, Baba H, Kobayashi T, Fujita T, Murakami H: Closing-opening wedge osteotomy to correct angular kyphotic deformity by a single posterior approach. Spine
(Phila Pa 1976) 26:391402, 2001
13. Lehmer SM, Keppler L, Biscup RS, Enker P, Miller SD, Steffee
AD: Posterior transvertebral osteotomy for adult thoracolumbar kyphosis. Spine (Phila Pa 1976) 19:20602067, 1994
14. Malcolm BW, Bradford DS, Winter RB, Chou SN: Post-traumatic kyphosis. A review of forty-eight surgically treated patients. J Bone Joint Surg Am 63:891899, 1981
15.Propst-Proctor SL, Bleck EE: Radiographic determination of
lordosis and kyphosis in normal and scoliotic children. J Pediatr Orthop 3:344346, 1983
16. Qi Q, Chen ZQ, Guo ZQ, Li WS: [New type spinal osteotomy
with cage inserting anteriorly and closing posteriorly to correct thoracolumbar kyphosis by a single posterior approach.]
Zhonghua Wai Ke Za Zhi 44:551555, 2006 (Chinese)
17. Rajasekaran S, Vijay K, Shetty AP: Single-stage closing-opening wedge osteotomy of spine to correct severe post-tubercular
kyphotic deformities of the spine: a 3-year follow-up of 17 patients. Eur Spine J 19:583592, 2010
18.Thiranont N, Netrawichien P: Transpedicular decancellation
closed wedge vertebral osteotomy for treatment of fixed flexion
deformity of spine in ankylosing spondylitis. Spine (Phila Pa
1976) 18:25172522, 1993
19. Thomasen E: Vertebral osteotomy for correction of kyphosis in
ankylosing spondylitis. Clin Orthop Relat Res 194:142152,
1985
20. van Royen BJ, Slot GH: Closing-wedge posterior osteotomy for
ankylosing spondylitis. Partial corporectomy and transpedicular fixation in 22 cases. J Bone Joint Surg Br 77:117121, 1995
Manuscript submitted June 28, 2011.
Accepted December 14, 2011.
Please include this information when citing this paper: published
online January 20, 2012; DOI: 10.3171/2011.12.SPINE11568.
Address correspondence to: Zhongqiang Chen, M.D., Orthopaedic
Department, Peking University 3rd Hospital, No 49. North Garden
Street, HaiDian District, Beijing, China 100191. email: chenzq5803@
gmail.com.

J Neurosurg: Spine / Volume 16 / April 2012