Posterior Selective Thoracic Fusion in Adolescent Idiopathic Scoliosis Patients: a Comparison of All Pedicle Screws versus Hybrid Instrumentation

Chin Med Sci J March 2009

Vol. 24, No. 1 P. 30-35

CHINESE MEDICAL SCIENCES JOURNAL ORIGINAL ARTICLE

Posterior Selective Thoracic Fusion in Adolescent Idiopathic Scoliosis Patients: a Comparison of All Pedicle Screws versus Hybrid Instrumentation Bin Yu, Jian-guo Zhang, Gui-xing Qiu*, Yi-peng Wang, Yu Zhao, Jian-xiong Shen, Hong Zhao, and Xin-yu Yang Department of Orthopedic Surgery, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences & Peking Union Medical College, Beijing 100730, China

Key words: adolescent idiopathic scoliosis; selective thoracic fusion; internal fixator; pedicle screw; decompensation Objective To analyze the influence of segmental pedicle screws versus hybrid instrumentation on the correction results in adolescent idiopathic scoliosis patients undergoing posterior selective thoracic fusion. Methods By reviewing the medical records and roentgenograms of adolescent idiopathic scoliosis patients who underwent selective thoracic fusion from February 2000 to January 2007 in our hospital, the patients were divided into 2 groups according to different instrumentation fashions: Group A was hook-screw-rod (hybrid) internal fixation type, Group B was screw-rod (all pedicle screws) internal fixation type, and the screws were used in every segment on the concave side of the thoracic curve. The parameters of the scoliosis were measured and the correction results were analyzed. Results Totally, 48 patients (7 males, 41 females) were included, with an average age of 14.4 years old and a mean follow-up time of 12.3 months. Thirty and 18 patients were assigned to group A and group B, respectively. The mean preoperative coronal Cobb angles of the thoracic curve were 48.8° and 47.4°, respectively. After surgery, they were corrected to 13.7° and 6.8°, respectively. At final follow-up, they were 17.0° and 9.5°, with an average correction rate of 64.6% and 79.0%, respectively, and the correction rate of group B was significantly higher than that of group A (P = 0.003). The mean preoperative coronal Cobb angles of the lumbar curve were 32.6° and 35.2°, respectively. After surgery, they were corrected to 8.6° and 8.3°, respectively. At final follow-up, they were 10.3° and 11.1°, with an average correction rate of 66.8% and 69.9%, respectively, and the correction rate of group B was significantly higher than that of group A (P = 0.003). The correction loss of the thoracic curve and lumbar curve in the 2 groups were 3.1° and 1.8°, 2.4° and 2.4°, respectively. No significant difference was noted (both P > 0.05). The decompensation rate at final follow-up in these 2 groups were 4% (1/25) and 7.1% (1/14) respectively, with no significant difference (P > 0.05). Received for publication April 17, 2008. *Corresponding author Tel: 86-10-65296080, E-mail: [email protected]

Vol. 24, No.1

CHINESE MEDICAL SCIENCES JOURNAL

31

Conclusions Both all pedicle screws and hybrid instrumentation can offer good correction results of the thoracic curve and lumbar curve in posterior selective thoracic fusion in patients with adolescent idiopathic scoliosis. While with all pedicle screws, the correction results of the thoracic curve and lumbar curve are both better than those with hybrid instrumentation without increased decompensation rate.

F

OR King type II scoliosis, selective thoracic fu-

Statistical analysis

sion can get good correction results of both the

A repeated-measures analysis of variance (ANOVA) was

thoracic curve and the lumbar curve.1,2 With the

used to compare changes in the preoperative, postopera-

wide application of pedicle screws, the fixation

tive, and follow-up data. All statistical tests were per-

fashion has changed from hooks to hybrid (hook and screw)

formed using P value of 0.05 as the level of significance.

3-5

system, and more recently to segmental pedicle screws.

Some doctors reported that segmental pedicle screws

RESULTS

system could get a better correction results compared with hybrid system,6 but the influence of segmental screws on

General data

the correction results of posterior selective fusion has been

Forty-eight patients (7 males and 41 females) were in-

rarely reported. The aim of the current study was to

cluded, with an average age of 14.4 years (range, 10-18

compare the correction results of segmental screws with

years). The mean follow-up time was 12.3 months (range,

hybrid systems in posterior selective thoracic fusion in

6-36 months). Thirty cases were included in group A and

patients with adolescent idiopathic scoliosis (AIS).

18 were in group B. Figures 1 and 2 show X-ray films of typical cases in groups A and B, respectively.

PATIENTS AND METHODS

Preoperative coronal Cobb angle, convex bending

Patients

film Cobb angle, flexibility, postoperative coronal

The medical records and radiographs of the AIS patients

Cobb angle, and correction rate

that underwent posterior selective thoracic fusion in our

For either the thoracic curve or the lumbar curve, there

hospital from February 2000 to January 2007 were ret-

were no significant differences in preoperative coronal

rospectively reviewed. The types of scoliosis were all King

Cobb angle, convex bending film Cobb angle, or flexibility

type II,1 PUMC classification type IIb1 or IIc3 (1 case with

between the 2 groups (all P > 0.05). Both the thoracic and

7

a 50° lumbar curve, PUMC classification type IIb2), and

lumbar curves of the 2 groups were matched. For both the

Lenke classification type 1.8 According to different fixation

thoracic curve and lumbar curve, the coronal Cobb angles

fashion, the patients were divided into 2 groups: group A

were significantly improved after operation (both P =

was hook-screw-rod (hybrid) internal fixation type, group

0.000), and group B had a better correction rate than

B was screw-rod (all pedicle screws) internal fixation type,

group A (both P = 0.000, Table 1).

and the screws were used in every segment on the concave side of the thoracic curve.

Coronal Cobb angle at final follow-up and correction rate

Radiological evaluation

At final follow-up, the coronal Cobb angles of the thoracic

The standing anteroposterior (AP) film, lateral film, supine

curve of the 2 groups were 17.0° and 9.5°, respectively.

bending film of the full spine preoperatively and the

There were no significant differences of the thoracic Cobb

standing AP and lateral films of the full spine postopera-

angle at final follow-up compared with that after operation

tively and at final follow-up were measured. The coronal

in either group (both P>0.05). The correction rates of the 2

and sagittal Cobb angles, flexibility of the curves, correc-

groups were 64.6％ and 79.0％, respectively, and group B

tion rate, and coronal trunk balance were recorded. Ac-

had a better correction rate than group A (P=0.003).

cording to the definition of Scoliosis Research Society, the

The coronal Cobb angles of the lumbar curve of the 2

coronal trunk balance was defined as the horizontal distance

groups at final follow-up were 10.3° and 11.1°, respectively.

of the midpoint of the C7 from center sacral vertical line

There were no significant differences of the lumbar Cobb

(CSVL) on standing AP film.9 Postoperative decompensation

angle at final follow-up compared with that after operation in

was defined as coronal trunk balance over 20 mm or lumbar

either group (both P>0.05). The spontaneous correction rates

curve deterioration in the coronal plane, or junctional kypho-

of the 2 groups were 66.8％ and 69.9％, respectively, and

sis (T12-L1) larger than 10° in the saggital plane.10-12

group B had a better correction rate than group A (P=0.003).

32

CHINESE MEDICAL SCIENCES JOURNAL

Figure 1.

March 2009

A

B

C

D

E

F

G

H

A 13-year-old female, preoperative films show a 65º thoracic curve and a 45º lumbar curve with a normal sagittal plane, and the lumbar curve is much flexible (A-D). So selective thoracic fusion (T4-L1) using hybrid system was performed according to the principle of PUMCIIb1. Postoperative X-ray shows a good coronal and sagittal balance, with a coronal correction rate of 83.1% of the thoracic curve and 75.6% of the lumbar curve (E, F). At two-year follow-up, the corrections on both planes are well-maintained (G, H).

Figure 2.

A

B

C

D

E

F

G

H

A 12-year-old female, preoperative films show a 47º thoracic curve and a 31º lumbar curve with a normal sagittal plane, and the lumbar curve is much flexible (A-D). So selective thoracic fusion (T4-T12) using segmental pedicle screws was performed according to the principle of PUMCIIb1. Postoperative X-ray shows a good coronal and sagittal balance, with a coronal correction rate of 85.1% of the thoracic curve and 77.4% of the lumbar curve (E, F). At one-year follow-up, the corrections on both planes are well-maintained (G, H).

Table 1.

Coronal Cobb angles, convex bending film Cobb angle, flexibility, and postoperative correction rates of the thoracic curve and lumbar curve between the 2 groups

Group

Preoperation

Bending

Flexibility

Postoperation

Correction rate

Group A

48.8°

24.9°

49.5%

13.7°*

Group B

47.4°

21.6°

54.9%

6.8°*

85.8%#

Group A

32.6°

-2.5°

94.9%

8.6°*

73.7%

Group B

35.2°

0°

93.4%

8.3°*

76.6%#

Thoracic curve 72.0%

Lumbar curve

*P=0.000 compared with preoperation; #P=0.000 compared with the correction rate of group A.

Vol. 24, No.1

CHINESE MEDICAL SCIENCES JOURNAL

33

Loss of Cobb angle

results showed that using pedicle screws in the lumbar

The loss of Cobb angles of the thoracic curve and lumbar

spine could get better correction results of the lumbar

curve in groups A and B were 3.1° and 1.8°, 2.4° and 2.4°,

curve and maintain the correction results. With the im-

respectively, and there were no significant differences of

provement of surgical technique, pedicle screws have been

the Cobb angle loss between the 2 groups (both P > 0.05).

used in the thoracic spine. Using pedicle screw in every segment on the concave side of the curve can get a more

Coronal trunk balance and decompensation rate

rigid fixation, reduce the stress concentration on certain

The coronal trunk balances in both groups after operation

screw during correction procedure, and get a better cor-

were greater than those before operation (group A: 11.7 mm

rection result. Many authors performed selective thoracic

vs. 7.3 mm, P=0.044; group B: 14.5 mm vs. 5.7 mm, P=

fusion in suitable patients with three-dimensional seg-

0.011), and they were both greatly improved at final

mental fixation and got good correction results.2,5,6

follow-up compared with those after operation (group A:

Kim et al6 compared the correction results of AIS pa-

6.2 mm vs. 11.7 mm, P=0.006; group B: 7.0 mm vs. 14.5

tients treated with all pedicle screws or hook-screw hybrid

mm, P=0.044). The decompensation rates of the 2 groups

fixation system, with 29 patients in each group and the 2

were 23.3% (7/30) and 16.7% (3/18) after operation, and

groups were matched. The correction results showed that

4% (1/25) and 7.1% (1/14) at final follow-up, respectively.

the all pedicle screws group had a better correction result

There were no significant differences of the decompensa-

of the thoracic curve than that of the hook-screw hybrid

tion rates after operation or at final follow-up between the

fixation group (65% vs. 46%, P<0.001), while the dif-

2 groups (both P > 0.05).

ference of the correction rate of the lumbar curve was not

Junctional kyphosis

authors may not agree. Lowenstein et al16 reported the

In the sagittal plane, no thoracolumbar kyphosis was noted

correction results of AIS patients treated with all pedicle

before operation, after operation, or at final follow-up.

screws or hook-screw hybrid fixation system, with 17 pa-

significant (64% vs. 53%, P=0.06). While some other

tients in each group. After operation, the coronal Cobb

DISCUSSION In 1983, King et al1 classified idiopathic thoracic scoliosis into 5 types and introduced the concept of selective

angles of the main thoracic curve and the lumbar curve were all significantly improved (all P<0.05), while the differences between the 2 groups were not significant (P=0.169 and P=0.760).

thoracic fusion in King type II patients. In his report, 111

In the current study, the correction result of the main

patients underwent selective fusion with Harrington in-

thoracic curve at final follow-up was better in the all pedicle

strumentation, and only 2 patients suffered from lumbar

screws fixation system group than that in the hybrid fixa-

curve progression and needed further treatment. So they

tion system group, which is similar to the report of Kim et

suggested that for King type II curve, selective thoracic

al’s,6 while not to that of Lowenstein et al’s.16 As for the

fusion could be safely performed down to the stable ver-

correction rate of the lumbar curve, most authors reported

tebra as the thoracic curve less than 80°, while leaving the

no significant difference between different fixation fash-

lumbar curve spontaneously corrected to obtain a new

ions.6,16-18 However, in the current study, the correction

trunk balance. Long-term follow-up study showed that

rates of the lumbar curve after operation and at final fol-

fusion to the middle or lower lumbar might increase the

low-up were both better in the all pedicle screws fixation

occurrences of low back pain, so selective thoracic fusion

system group than those of the hybrid fixation system

has significant importance in preventing low back pain.

group. This was similar to that of Ritzman et al’s report.19

Harrington et al13 reported a correction rate of 28%-

Postoperative decompensation was also a concern in

30% with Harrington instrumentation. With the application

selective thoracic fusion. In Kim et al’s report,6 the coronal

of three-dimensional segmental correction implant, the

trunk balances were 13 mm and 21 mm before operation in

correction results got much improved. In Fitch et al’s re-

the all pedicle screws fixation system and hybrid fixation

port,14 the correction rate was improved to 46%-57% with

system respectively. At 2 years follow-up, they were 15

Cotrel-Dubousset instrumentation. With the application of

mm and 11 mm, respectively, with no significant difference

pedicle screws, the correction results got further improved.

between the 2 groups (P=0.18). Lowenstein et al16 re-

reported 39 patients with double curves, among

ported that the coronal trunk balances were 10 mm and 11

which 20 cases were treated with hybrid hook-screw fixa-

mm before operation in the all pedicle screws fixation

tion system and 19 cases with all hook fixation system. The

system and hybrid fixation system, respectively. After

Barr et al

15

34

CHINESE MEDICAL SCIENCES JOURNAL

operation, they were 6.06 mm and 6.69 mm, respectively,

fusion in adolescent idiopathic scoliosis. Chin Med Sci J

with no significant differences compared with those before operation (P=0.416, P=0.238) and no significant differ-

2004;19:216-20. 3.

ence between the 2 groups (P=0.951).

Suk SI, Lee CK, Kim W, et al. Segmental pedicle screw fixation in the treatment of thoracic idiopathic scoliosis.

In the current study, the coronal trunk balances after operation in the 2 groups were both greater than those

March 2009

Spine 1995; 20:1399-405. 4.

Hamill CL, Lenke LG, Bridwell KH, et al. The use of pedicle

before operation, while during follow-up, the coronal trunk

screw fixation to improve correction in the lumbar spine

balances got much improved. This might imply that after

of patients with idiopathic scoliosis. Spine 1996; 21:

operation, the lumbar curve could not spontaneously com-

1241-9.

pensate the better correction results of the thoracic curve.

5.

Suk SI, Lee SM, Chung ER, et al. Selective thoracic fusion

While during follow-up, the lumbar curve spontaneously

with segmental pedicle screw fixation in the treatment of

compensated the correction result of the thoracic curve,

thoracic idiopathic scoliosis: more than 5-year follow-up.

thus a much improved coronal trunk balance was created. The differences between the 2 groups after operation and

Spine 2005; 30:1602-9. 6.

Kim YJ, Lenke LG, Kim JH, et al. Comparative analysis of

at final follow-up were neither significant, so the better

pedicle screw versus hybrid instrumentation in posterior

correction results did not cause an increased decom-

spinal fusion of adolescent idiopathic scoliosis. Spine

pensation rate.

2006;31:291-8.

In group B, segmental pedicle screws were used,

7.

among which most were located in the thoracic spine. The

cation of idiopathic scoliosis: a Peking Union Medical

insertion of thoracic pedicle screw is a technique with potential risks compared with hooks. In the upper thoracic

Qiu G, Zhang J, Wang Y, et al. A new operative classifiCollege Method. Spine 2005; 30:1419-26.

8.

Lenke LG, Betz RR, Harms J, et al. Adolescent idiopathic

spine, the pedicles are small and it’s difficult to correctly

scoliosis: a new classification to determine extent of

insert the pedicle screws. And even if the screws are

spinal

properly inserted, there is high risk of pedicle cutting of the

83:1169-81.

screws. In addition, as the pedicles are rotated due to the

9.

arthrodesis.

J Bone

Joint

Surg

Am

2001;

The Working Group on 3-D Classification (Chair Larry

scoliosis deformity, many pedicle screws are not located in

Lenke, MD), and the Terminology Committee. SRS Ter-

the center of the pedicle. In Belmont et al’s report,20 279

minology Committee and Working Group on Spinal

thoracic pedicle screws were evaluated with CT scan. The

Classification: Revised Glossary of Terms. [cited 2007

results showed that 159 screws (57%) were fully contained

Jun 1]. Available from: http://www.srs.org/ profession-

within the cortical boundaries of the pedicle, 81 screws

als/glossary/glossary.asp

(29%) penetrated the lateral cortex, and 39 screws (14%)

10.

Richards BS, Birch JG, Herring JA, et al. Frontal plane and

penetrated the medial cortex. Thus, there are high risks of

sagittal plane balance following Cotrel-Dubousset instru-

vascular and neurologic structure injury with more thoracic

mentation for idiopathic scoliosis. Spine 1989; 14:733-7.

pedicle screws. And with more screws used, the cost is also

11.

a problem for some family.

pensation after Cotrel-Dubousset instrumentation of

Although there were some biases in the current study: the operations were performed by different surgeons and

Thompson JP, Transfeldt EE, Bradford DS, et al. Decomidiopathic scoliosis. Spine 1990;15:927-31.

12.

Bridwell KH, McAllister JW, Betz RR, et al. Coronal de-

the all pedicle screws group had more fixation anchors than

compensation produced by Cotrel-Dubousset derotation

that of hybrid fixation type. The results showed that

maneuver for idiopathic right thoracic scoliosis. Spine

segmental pedicle screws fixation fashion could get a better correction rate in both the thoracic curve and the

1991; 16:769-77. 13.

Harrington PR, Dickson JH. An 11-year clinical investi-

lumbar curve, with no increased decompensation rate in

gation of Harrington instrumentation: a preliminary re-

posterior selective thoracic fusion in AIS.

port on 578 cases. Clin Orthop 1973; 93:113-8. 14.

Cotrel-Dubousset and Harrington rod instrumentations in

REFERENCES 1.

2.

King HA, Moe JH, Bradford DS, et al. The selection of

Fitch RD, Turi M, Bowman BE, et al. Comparison of idiopathic scoliosis. J Pediatr Orthop 1990; 10:44-7.

15.

Barr S, Schuette A, Emans J. Lumbar pedicle screws

fusion levels in thoracic idiopathic scoliosis. J Bone Joint

versus hooks: results in double major curves in adoles-

Surg A 1983; 65:1302-13.

cent idiopathic scoliosis. Spine 1997; 22:1369-79.

Yu B, Zhang JG, Qiu GX, et al. Posterior selective thoracic

16.

Lowenstein JE, Matsumoto H, Vitale MG, et al. Coronal

Vol. 24, No.1

CHINESE MEDICAL SCIENCES JOURNAL

screw fixation versus hook constructs. J Pediatr Orthop

and sagittal plane correction in adolescent idiopathic

2005; 25:415-9.

scoliosis: a comparison between all pedicle screw versus hybrid thoracic hook lumbar screw constructs. Spine 17.

19.

Ritzman TF, Upasani VV, Bastrom TP, et al. Comparison

2007; 32:448-52.

of compensatory curve spontaneous derotation after

Dobbs MB, Lenke LG, Kim YJ, et al. Selective posterior

selective thoracic or lumbar fusions in adolescent idio-

thoracic fusions for adolescent idiopathic scoliosis:

pathic scoliosis. Spine 2008; 33:2643-7.

comparison of hooks versus pedicle screws. Spine 2006; 18.

35

20. Belmont PJ, Klemme WR, Dhawan A, et al. In vivo ac-

31:2400-4.

curacy of thoracic pedicle screws. Spine 2001; 26:2340-

Storer SK, Vitale MG, Hyman JE, et al. Correction of

6.

adolescent idiopathic scoliosis using thoracic pedicle