Eur J Orthop Surg Traumatol DOI 10.1007/s00590-014-1425-0
ORIGINAL ARTICLE
Does curve convexity affect the surgical outcomes of thoracic adolescent idiopathic scoliosis? Wei-Jun Wang • Ai-Bing Huang • Ze-Zhang Zhu Feng Zhu • Xu Sun • Yong Qiu
•
Received: 20 December 2013 / Accepted: 2 February 2014 Ó Springer-Verlag France 2014
Abstract Background Major left thoracic (LT) curve is an atypical type in adolescent idiopathic scoliosis (AIS) and showed independent clinical characteristics and natural history compared to major right thoracic (RT) curve. However, it’s unclear whether the convexity of major thoracic curve would affect the surgical outcomes and risk of complications. A retrospective follow-up study was conducted to investigate whether the convexity of major thoracic curve would affect the surgical outcomes of patients with main thoracic AIS. Methods Twelve LT-AIS patients underwent corrective spinal instrumentation and fusion were retrieved, and twelve patients with main RT-AIS matched for gender, chronological age, curve type, magnitude and surgical strategy were selected as control. All patients underwent at least 2-year follow-up. The pre- and post-operative radiographic parameters, intraoperative data and functional outcome assessed by Scoliosis Research Society questionnaire 22 (SRS-22) were analyzed and compared between two groups. Results Patients with LT- and RT-AIS presented with similar magnitudes of thoracic curves, flexibility, fusion level and correction rate. Compared with RT-AIS, patients with LT-AIS showed longer operation time (average, 364 vs. 348 min) and larger amount intraoperative estimated blood loss (2,060 vs. 1,720 ml) although the differences were not statistically significant (p [ 0.05). With at least
W.-J. Wang A.-B. Huang Z.-Z. Zhu F. Zhu X. Sun Y. Qiu (&) Spine Surgery, The Affiliated Drum Tower Hospital of Nanjing University Medical School, Nanjing 210008, China e-mail: [email protected]
2-year follow-up, patients in two groups showed comparable loss of correction, coronal and sagittal balance, and the sagittal profiles. With regard to functional outcome, the scores of five categories of SRS-22 questionnaire were similar between two groups. No neurological or vascular complication was observed in these patients. Conclusions The radiographic and functional outcomes of LT-AIS patients underwent operation were comparable to those with RT-AIS. Longer operation time and more intraoperative blood loss may be expected in instrumentation and fusion for patients with LT-AIS, which might be because of the inconvenience on the surgical procedure. Keywords Idiopathic scoliosis Adolescent Thoracic Direction Correction
Introduction Scoliosis is a curvature of the spine on the coronal plane, along with deformity on the sagittal and horizontal planes. According to the location of the main curve on the coronal plane, the deformity may be classified as single major thoracic, single thoracolumbar, single lumbar, double major thoracic, double major curves, or as a multiple curve pattern [1, 2]. Within these curve types, the major thoracic curve has been found to be the most common type [3–5]. Most patients have right convex major thoracic (RT) curve and show a normal neural axis [1, 6–8]. In contrast, patients with major left convex thoracic (LT) scoliosis are rare, with the prevalence varying from 1.3 to 9.1 % [1, 9–13]. Furthermore, functional or radiological neural abnormalities have been reported in 20.4–54 % of ‘‘idiopathic’’ scoliosis patients with major LT curve [1, 6, 7, 14].
123
Eur J Orthop Surg Traumatol
Therefore, the major left thoracic curve has been defined as an atypical curve pattern and a strong indicator for associated intraspinal neuroanatomic abnormalities of scoliosis [14–18]. Scoliosis with left convex thoracic curve has a lower morbidity but higher risk of associated neural axis abnormalities compared with RT scoliosis; moreover, this condition typically presents with a smaller curve magnitude at diagnosis, a faster growth rate, but a lower likelihood of progression [19–22]. These findings indicate that patients with LT scoliosis might have an independent behavior from those with RT scoliosis, leading to a different surgical outcome. The surgical strategies used and the short-term and long-term outcomes are different between patients with LT scoliosis associated with neural axis abnormalities, such as Chiari malformation and syringomyelia, and patients with right thoracic scoliosis with normal neural axis [23–25]. However, no study has reported whether the surgical outcomes of LT scoliotic patients with normal neural axis were different from those with RT scoliosis. Adolescent idiopathic scoliosis (AIS) is the most common type of scoliosis [3, 26]. In the present study, the radiological and functional outcomes of surgical treatment of AIS patients with main left and right thoracic curves were compared to determine any affects due to curve lateralization (Figs. 1, 2).
Materials and methods
Fig. 1 A female AIS patient with a left convex major thoracic curve at 15 years old. The thoracic curve was corrected from 46° preoperatively (a) to 17° at 2 years postoperatively (c), with a
correction rate of 63 %. Thoracic kyphosis was 7° preoperatively (b) and 13° at the latest follow-up (d)
123
Subjects To identify patients with LT-AIS, AIS patients receiving surgical treatment in our center between January 2003 and December 2010 were retrospectively reviewed in the database. Patients with a main left convex thoracic curve were recruited into this study if they met the following criteria: (1) a radiological and clinical diagnoses of AIS with a normal spinal cord and a cerebral tonsil position confirmed by magnetic resonance (MR) scanning [27, 28]; (2) a single main thoracic curve with the apex located between T7 and T9; (3) age ranging from 12 to 20 years; (4) a posterior-only selective thoracic fusion and instrumentation with all pedicle screws; (5) a minimum of 2 years of follow-up after the operation. Patients meeting the following criteria were excluded: (1) any form of prior surgical treatment for scoliosis; (2) neural abnormalities, i.e., low-lying cerebellar tonsil or conus medullaris and syringomyelia. We identified only 12 patients out of a total database of 1,534 surgically treated AIS. Among the 1,534 AIS patients, those with a typical main RT curve meeting the criteria for LT-AIS were identified. Patients who matched in gender, chronological age, Risser stages, curve patterns and magnitude, and curve flexibility were considered as adequate controls for LT-AIS. From
Eur J Orthop Surg Traumatol
these patients, a random sample of 12 RT-AIS patients having a major curve within 5° of the LT-AIS patients were selected as controls [29]. The gender distribution, chronological age, and skeletal maturity at the time of operation are shown in Table 1. According to the Lenke classification [4], eight patients were defined as Lenke type 1A, two as Lenke type 1B, and two as Lenke type 1C in each group. All patients were treated with posterior correction and instrumentation by the same surgical team. Comparable instrumentation strategies, including implant density, were used between the two groups [4]. The implants used included TSRH, CDH, or CDH legacy (Medtronic, Memphis, USA) in different periods. Surgeons always stood on the concave side of patients while performing the surgical procedures. Thoracoplasty was carried out in patients with significant rib hump after curve correction. Iliac crest bone grafts were harvested as autografts for all posterior procedures. Somatosensory-evoked potentials, motor-evoked potentials, and Stagnara wake-up tests were used for intraoperative neurological monitoring. Intraoperative parameters were recorded, including the operation time, estimated blood loss, volume of blood transfusion, number of vertebrae fused, and the number of patients underwent thoracoplasty. Short-term and longterm complications were also recorded.
Fig. 2 A female AIS patient with a right convex major thoracic curve at 15 years of age. The preoperative thoracic curve was 42° and kyphosis was 15°. The patient was instrumented with all pedicle
Radiographic evaluation Standing long-cassette posteroanterior and lateral radiographs of the entire spine, which were obtained preoperatively and postoperatively at 3 months, 1 year, 2 years, and at the time of the final follow-up, as well as preoperative right- and left-side bending radiographs, were reviewed and analyzed by an independent surgeon. The Cobb angle of the main thoracic curve, lower instrumented vertebra (LIV) tilt, and global coronal balance (the horizontal distance between the C7 plumb line and the central sacral vertical line, C7CSVL) were measured on the posteroanterior radiographs. The implant density of the pedicle screws was calculated as follows: [(number of screws/number of pedicles in fusion levels)] 9 100 %. The coronal spinal decompensation was defined as a deviation of C7 greater than 2 cm from the CSVL [30]. On lateral radiographs, the thoracic kyphosis (T5–T12), thoracolumbar kyphosis (T10–L2), lumbar lordosis (T12–S1), and global sagittal balance (the horizontal distance from a vertical line extended from the center of the C7 vertebral body to the posterior superior cortex of S1, SVA) were quantified [30, 31]. Furthermore, the apical vertebral rotation was presented by the apical rib hump prominence (RH), which is the linear distance on a lateral radiograph, between the left and right ribs at the rib deformity apex or the most prominent rib [32]. Curve flexibility
screws. The thoracic curve was 12° and thoracic kyphosis was 18° at 2.6 years following curve correction
123
Eur J Orthop Surg Traumatol Table 1 Perioperative data for left and right thoracic AIS patients
Gender
LT-AIS (n = 12)
RT-AIS (n = 12)
7 Females, 5 males
7 Females, 5 males
p Value
Age at operation (years)
16.4 ± 2.2
16.0 ± 1.8
0.65
Risser sign Operative time (min)a
3.4 ± 1.1 364 ± 88
3.9 ± 1.2 348 ± 85
0.14 0.54
Estimated blood loss (ml)a
2,060 ± 830
1,720 ± 442
0.17
Blood transfusion (ml)a
1,190 ± 482
1,040 ± 493
0.39
No. of vertebrae in fusion
11.4 ± 1.9
11.4 ± 2.5
1.00
Follow-up (years)
5.8 ± 3.2
5.4 ± 2.9
0.48
AIS adolescent idiopathic scoliosis, LT left thoracic, RT right thoracic a
Comparisons were performed by nonparametric analysis
was determined on the preoperative supine side bending films. The flexibility index was acquired by subtracting the magnitude of the coronal Cobb angle on the supine bending radiograph from that on the preoperative standing radiograph, then dividing by the preoperative magnitude of the coronal Cobb angle on the standing radiograph. The curve correction rates, immediately after surgery and at the followup, and loss of correction were calculated. SRS-22 questionnaire To determine the functional outcome of surgery, the Scoliosis Research Society questionnaire 22 (SRS-22, Chinese version) [33] was completed by patients at the last follow-up and was reviewed. The SRS-22 contains 22 questions covering five domains: function/activity, pain, self-perceived image, mental health, and satisfaction with treatment [34]. Statistical analysis The results were presented as mean and standard division. Statistical analysis was performed with Statistical Package for the Social Sciences (SPSS, version 13.0, Chicago, IL, USA). Clinical data were compared using Student’s t tests or the Wilcoxon’s ranks test between the two groups. The preoperative and postoperative radiological parameters in each group, as well as differences in changes in these parameters between the two groups, were compared by repeated-measures analysis. Significance was defined as p \ 0.05.
grade, and gender distribution (Table 1). The average fusion levels were 11.7 ± 1.9 (8–14) and 11.1 ± 2.5 (8–15), while the implant densities were 46.5 ± 7.5 and 50.2 ± 9.2 % in patients with LT-AIS and RT-AIS, respectively (p [ 0.05). In each group, 7 of 12 patients underwent thoracoplasty to correct the rib hump. Despite the fact that differences in the perioperative variables were not statistically significant, patients with LT-AIS tended to have a longer operation time (average, 363.5 ± 87.7 vs. 347.5 ± 85.4 min), greater intraoperative blood loss (average, 2,060.0 ± 830.3 vs. 1,720.0 ± 441.7 ml), and larger volume of blood transfusion (average, 1,190.0 ± 481.8 vs. 1,040.0 ± 492.6 ml) than those with RT-AIS (Table 1); however, the difference was not statistically significant (p [ 0.05). The patients with LTand RT-AIS were followed for an average of 5.8 and 5.4 years, respectively. Radiographic results Correction of major thoracic curve The patients in both groups were well matched in terms of the magnitude and flexibility of the primary thoracic curve. In patients with LT-AIS, the major thoracic curve was corrected from 59.3° to 28.9°, immediately following the operation, with an overall correction rate of 54.2 %. The mean LT curve was 30.7°, 29.9°, and 30.4° at 1, 2, and 5.8 years postoperatively, respectively, with a mean 1.5° loss of correction during follow-up. In the matched patients with RT-AIS, the preoperative thoracic curve of 60.5° was corrected to 27.4°, immediately following surgery, with an overall correction rate of 56.7 %. During follow-up, the mean RT curve was 26.8°, 28.0°, and 29.3° at 1, 2, and 5.4 years following surgery, respectively, with a mean 1.9° loss of correction at the final follow-up. Repeated-measures analysis revealed significant differences between pre- and postoperative measures, but with no difference during different points of follow-up, the curve correction rate and the mean loss of correction of the primary thoracic curve between the two groups showed no significant difference. The tilt of LIV also improved significantly from 13.8° to 3.4° in LT-AIS and 14.0° to 2.9° in RT-AIS following surgery, respectively, (p \ 0.05). A slight deterioration of LIV was noted in patients with either LT-AIS or RT-AIS during follow-up (1.7° vs. 1.6°); however, there was no significant difference between the two groups (Table 2).
Results Sagittal profile and rib hump correction Clinical results Due to the recruitment criteria, the AIS patients with LT and RT curves were similar in chronological age, Risser
123
Measurements from preoperative sagittal X-ray films showed significantly larger thoracic kyphosis in LT-AIS patients, compared to that in RT-AIS. The difference was
Eur J Orthop Surg Traumatol Table 2 Comparison of radiological data between left and right thoracic AIS patients LT-AIS (n = 12)
in patients from both groups; however, no significant difference was recorded between the two groups (Table 2).
RT-AIS (n = 12)
Global balance Main thoracic curve (°) Pre-op
59.3 ± 19.6 (43–99)
60.5 ± 18.6 (45–95)
Flexibility (%)
43.9 ± 18.4
42.9 ± 14.8
Immediate post-op
28.9 ± 18.4
27.4 ± 16.6
Correction rate (%)
54.2 ± 20.0
56.7 ± 13.0
1 year post-op
30.7 ± 20.1
26.8 ± 17.2
2 year post-op
29.9 ± 19.7
28.0 ± 18.3
Latest follow-up
30.4 ± 18.3
29.3 ± 17.1
1.5 ± 2.4
1.9 ± 1.2
13.8 ± 5.9
14.0 ± 9.5
Loss of correction LIV tilt angle (°)* Pre-op Immediate post-op
3.4 ± 3.8^^
2.9 ± 3.2^^
Latest follow-up
5.1 ± 4.5^^
4.5 ± 3.6^^
Complications
T5–T12 (°) Preoperativea
30.9 ± 22.9
14.2 ± 10.8*
Immediatea
28.2 ± 13.2
18.8 ± 7.1*
Final follow-upa
29.5 ± 13.6
17.9 ± 8.4*
Preoperativea
3.4 ± 4.8
0.2 ± 11.7
Immediatea
1.2 ± 6.2
2.0 ± 4.8
Final follow-upa
2.4 ± 7.7
1.9 ± 4.1
Preoperative
48.1 ± 13.4
40.3 ± 23.1
Immediate
46.2 ± 11.2
40.7 ± 11.1
Final follow-up
47.8 ± 11.0
41.7 ± 10.0
Preoperative
2.3 ± 1.2
4.5 ± 1.3
Immediate
1.0 ± 0.8
1.2 ± 0.7
Final follow-up
1.1 ± 0.9
1.2 ± 1.0
T10–L2 (°)
L1–S1 (°)
Rib hump (cm)
C7-CSVL (cm) a
The coronal balance was similar between the two groups before the operation and at final follow-up, although LT-AIS patients showed significantly smaller coronal balance than RT-AIS patients, immediately following the operation (Table 2). None of the patients showed decompensation on coronal planes. Patients with LT-AIS showed a significantly larger SVA, compared to RT-AIS patients, before the operation, which indicated that patients with LTAIS had more imbalances on the sagittal plane. Following corrective surgery, however, patients in both groups showed comparable sagittal balance.
Pre-op
0.7 ± 0.7
0.6 ± 0.6
Immediate post-opa
0.6 ± 0.4
0.9 ± 0.3*
Latest follow-upa
0.4 ± 0.5
0.5 ± 0.5
Preoperativea
1.2 ± 0.8
0.6 ± 0.5*
Immediatea
0.9 ± 0.5
1.4 ± 0.7
Final follow-upa
1.0 ± 0.8
0.5 ± 0.7
C7PL (cm)
LIV low instrument vertebrae, CSVL central sacral vertical line, C7PL C7 plumb line, op operation
Two patients with LT-AIS had hydrothorax following thoracoplasty, but recovered 1 week after surgery. One patient with LT-AIS suffered a superficial wound infection following the operation. The wound healed, uneventfully, with debridement and delayed closure. In the patients with RT-AIS, two pedicle screws were found purchased slightly into the spinal canal, but no neurological symptoms were reported. For all other cases, no neurological, vascular, visceral complications, or venous thromboembolism was observed. No pseudarthrosis, implant breakage, or loosening was observed in patients in either group, during followup. There was no statistical difference in the incidence of complications in LT-AIS and RT-AIS patients. SRS-22 scores All patients completed the SRS-22 questionnaire at the last follow-up. The score for the function/activity domain was 4.1 and 4.2, while the score for the satisfaction with treatment domain was 4.2 and 4.4 in patients with LT- and RT-AIS, respectively. No significant differences were observed in the SRS-22 scores between the two groups, with respect to the pain, mental health, function/activity, self-image/appearance, or satisfaction of management domains (Table 3).
Significant change compared to pre-op data, ^ p \ 0.05, ^^ p \ 0.01 Significant difference between the two groups, * p \ 0.05 a
Comparisons were done by Mann–Whitney nonparametric analysis
also significant following surgery and at the last follow-up (p \ 0.05). In contrast, thoracolumbar kyphosis and lumbar lordosis were similar between the two groups before the operation, immediately after the operation, and at final follow-up. A significant correction of rib hump was found
Discussion Although differences between scoliosis with LT and RT have been identified in terms of the incidence of neural axis abnormalities and curve progression [19–22], the surgical outcomes of LT-AIS have not been well documented, and it has not been reported whether the surgical outcomes of
123
Eur J Orthop Surg Traumatol Table 3 Comparison scores of SRS-22 at the last follow-up between left and right thoracic AIS patients Domain
Group A (n = 12)
Group B (n = 12)
p Value
Function/activity
4.1 ± 0.5
4.2 ± 0.3
0.63
Pain
4.3 ± 0.5
4.0 ± 0.6
0.39
Self-image Mental health
4.0 ± 0.4 4.2 ± 0.4
4.2 ± 0.4 4.1 ± 0.5
0.62 0.80
Satisfaction with treatment
4.2 ± 0.4
4.4 ± 0.5
0.47
thoracic AIS are affected by curve convexity. Abnormal anatomical relationships between the spine and the perispinal organs have been well described in scoliotic patients, due to spinal deformity [35–37]. However, the pattern of these abnormalities is different in patients who suffer from RT and LT-AIS due to the different direction of curvature [38]. Spine surgeons, largely familiar with corrective procedures for RT-AIS, might be uncomfortable with the procedures for correcting LT-AIS. It has been of concern whether patients with LT-AIS could achieve the same surgical outcome and comparable incidence of complications, as those with RT-AIS. The rarity of the left thoracic curve pattern in idiopathic scoliosis was described by McCarver et al. [10] at the Scoliosis Research Society annual meeting in Toronto in 1,971. It was found that 1.3 % of 550 patients with idiopathic scoliosis had primary left thoracic curves. However, the diagnosis of idiopathic scoliosis in these patients was not confirmed by MR, and it is unknown as to whether some of them had neural axis abnormalities [39]. In addition, Coonrad et al. [40] found that 27 of 1,622 patients with noncongenital scoliosis presented with a major left convex thoracic curve. However, only 14 patients received a final diagnosis of idiopathic scoliosis, and the incidence of LT scoliosis was 0.86 %. This was inconsistent with other studies that found the incidences of neural axis abnormalities in patients with left thoracic scoliosis to range from 43 to 54 % [14, 17, 40]. In our center, routine preoperative MR scanning was performed for each surgical case, and AIS was diagnosed by confirming no neural axis abnormalities. From January 2003 to December 2010, 1,534 patients with AIS received corrective spinal surgery, and 12 of them had major left thoracic curves, without any form of neural axis abnormality, confirmed by MRI. Hence, the incidence of LT-AIS was 0.78 % in the surgically corrected AIS patients in our center. According to our knowledge, the present study may give the most accurate incidence of LT-AIS in the population of AIS patients requiring surgical intervention. Although left thoracic scoliosis are commonly associated with neural axis abnormalities [14–18], the surgical
123
outcomes of these patients were not compared with patients with RT-AIS in the present study. Patients with neural axis abnormalities such as Chiari malformation or syringomyelia would present with hyperkyphosis in thoracic spine, younger at the time of needing surgical correction, longer fusion levels, higher implant density, and higher risk of neurological complications [41]. With the aim to study whether the direction of major thoracic curve would affect the surgical outcome of AIS, only scoliosis patients with normal neural axis confirmed by MR scanning were recruited in the present study. By comparing the radiographic profiles of LT-AIS patients to those with RT-AIS, a significantly higher incidence of thoracic kyphosis was identified in patients with LT-AIS. This finding indicated that hyperkyphosis might be a coexistent phenotype of left thoracic scoliosis instead of an indicator of neural axis abnormality in scoliosis patients. Careful preoperative planning for surgical strategies and procedures were made for each LT-AIS patient, considering the different anatomical relationship between the spine and peri-spinal organs, as well as inconveniences of surgical procedures. In patients with RT-AIS, the thoracic vertebrae are in right-side torsion, the thoracic aorta is in a more posterolateral position related to vertebrae, and the spinal cord is more close to pedicle in left side when compare to normal subjects [36, 42]. In contrast, in patients with left thoracic curve, the thoracic aorta locate in a more anterolateral position and the spinal cord is more close to pedicle in right side compared to normal subjects [38]. Hence, the incidences of aortic and neurological complications by pedicle screws insertion were higher in the leftside pedicles in RT-AIS, but in the right side in LT-AIS when compare to the other side. Since spinal surgeons are accustomed to the surgical procedures of RT-AIS correction and are more commonly right handed, they may feel uncomfortable when performing surgery for LT-AIS, and the procedures might not be performed in a very fluent manner. Theoretically, the different pattern of abnormalities in LT-AIS from RT-AIS would result in higher incidence of aortic and neurological complications. By comparing the surgical treatment outcomes between AIS patients with either left or right thoracic curves, matched for gender, chronological age, curve type, major curve magnitude, and instrumentation, our results indicate that patients with LT-AIS have a tendency to require a longer duration of surgery, have a greater intraoperative blood loss, and require larger volumes of blood transfusions; however, the differences were not statistically significant. Moreover, the radiological outcomes in both the coronal planes and sagittal planes were comparable between patients with LT- and RT-AIS. In addition, the risk of complications is not higher for surgical correction in LTAIS patients.
Eur J Orthop Surg Traumatol
Despite the radiological outcome, the functional outcome is another important issue in surgical treatment for AIS. The simplified Chinese version of the Scoliosis Research Society 22 was recently made available for clinical evaluation of Chinese adolescents and young adults with scoliosis treated with surgery [33]. All patients enrolled in our study completed the SRS-22 questionnaire at their last follow-up. Patients in both groups showed similar scores for each of the five domains: function/ activity, pain, self-image, mental health, and satisfaction with treatment. These results indicate that patients with LT-AIS and RT-AIS have comparable functional outcomes when using the same surgical strategies.
Conclusion The present study compared surgical outcomes between left and right thoracic AIS patients who were matched for gender, age, curve type, major curve magnitude, and instrumentation. Our results showed that LT-AIS can be corrected as effectively as RT-AIS in terms of risk of complications and radiological and functional outcomes. Longer operation times and greater intraoperative blood loss can be expected for LT-AIS. Acknowledgments This work was supported by National Natural Science Foundation of China (81101335), National Post-doctoral Foundation of China (2012M52101062), National Key Clinical Specialty Construction Project in Orthopaedics, and Jiangsu Province’s Key Medical Talents Project (RC2011149). Conflict of interest
None.
References 1. Coonrad RW, Murrell GA, Motley G et al (1998) A logical coronal pattern classification of 2,000 consecutive idiopathic scoliosis cases based on the scoliosis research society-defined apical vertebra. Spine (Phila Pa 1976) 23:1380–1391 2. King HA, Moe JH, Bradford DS et al (1983) The selection of fusion levels in thoracic idiopathic scoliosis. J Bone Jt Surg Am 65:1302–1313 3. Weinstein SL, Dolan LA, Cheng JC et al (2008) Adolescent idiopathic scoliosis. Lancet 371:1527–1537 4. Lenke LG, Betz RR, Harms J et al (2001) Adolescent idiopathic scoliosis: a new classification to determine extent of spinal arthrodesis. J Bone Jt Surg Am 83-A:1169–1181 5. Qiu G, Zhang J, Wang Y et al (2005) A new operative classification of idiopathic scoliosis: a Peking Union Medical College method. Spine (Phila Pa 1976) 30:1419–1426 6. Hausmann ON, Boni T, Pfirrmann CW et al (2003) Preoperative radiological and electrophysiological evaluation in 100 adolescent idiopathic scoliosis patients. Eur Spine J 12:501–506 7. Inoue M, Minami S, Nakata Y et al (2005) Preoperative MRI analysis of patients with idiopathic scoliosis: a prospective study. Spine (Phila Pa 1976) 30:108–114
8. Richards BS, Sucato DJ, Johnston CE et al (2010) Right thoracic curves in presumed adolescent idiopathic scoliosis: which clinical and radiographic findings correlate with a preoperative abnormal magnetic resonance image? Spine (Phila Pa 1976) 35:1855–1860 9. James JI (1954) Idiopathic scoliosis; the prognosis, diagnosis, and operative indications related to curve patterns and the age at onset. J Bone Jt Surg Br 36-B:36–49 10. McCarver C, Levine D, Velaskis K (1971) Left thoracic curve patterns in idiopathic scoliosis. J Bone Jt Surg Am 53:196 11. Ylikoski M (2005) Growth and progression of adolescent idiopathic scoliosis in girls. J Pediatr Orthop B 14:320–324 12. Qiu Y, Zhu F, Wang B et al (2009) Clinical etiological classification of scoliosis: report of 1,289 cases. Orthop Surg 1:12–16 13. Elsig JPJ, Kaech DL (2007) Dynamic imaging of the spine with an open upright MRI: present results and future perspectives of fmri. Eur J Orthop Surg Traumatol 17:119–124 14. Wu L, Qiu Y, Wang B et al (2010) The left thoracic curve pattern: a strong predictor for neural axis abnormalities in patients with ‘‘idiopathic’’ scoliosis. Spine (Phila Pa 1976) 35:182–185 15. Spiegel DA, Flynn JM, Stasikelis PJ et al (2003) Scoliotic curve patterns in patients with Chiari I malformation and/or syringomyelia. Spine (Phila Pa 1976) 28:2139–2146 16. Qiu Y, Zhu ZZ, Wang B et al (2008) Radiological presentations in relation to curve severity in scoliosis associated with syringomyelia. J Pediatr Orthop 28:128–133 17. Goldberg CJ, Moore DP, Fogarty EE et al (1999) Left thoracic curve patterns and their association with disease. Spine (Phila Pa 1976) 24:1228–1233 18. Wang W, Zhu Z, Zhu F et al (2012) Different curve pattern and other radiographical characteristics in male and female patients with adolescent idiopathic scoliosis. Spine (Phila Pa 1976) 37:1586–1592 19. Goldberg CJ, Dowling FE, Fogarty EE (1994) Left thoracic scoliosis configurations. Why so different? Spine (Phila Pa 1976) 19:1385–1389 20. Soucacos PN, Zacharis K, Gelalis J et al (1998) Assessment of curve progression in idiopathic scoliosis. Eur Spine J 7:270–277 21. Chiu YL, Huang TJ, Hsu RW (1998) Curve patterns and etiologies of scoliosis: analysis in a university hospital clinic in Taiwan. Changgeng Yi Xue Za Zhi 21:421–428 22. Gupta R, Sharma R, Vashisht S et al (1999) Magnetic resonance evaluation of idiopathic scoliosis: a prospective study. Australas Radiol 43:461–465 23. Bradley LJ, Ratahi ED, Crawford HA et al (2007) The outcomes of scoliosis surgery in patients with syringomyelia. Spine (Phila Pa 1976) 32:2327–2333 24. Ferguson RL, DeVine J, Stasikelis P et al (2002) Outcomes in surgical treatment of ‘‘idiopathic-like’’ scoliosis associated with syringomyelia. J Spinal Disord Tech 15:301–306 25. Sponseller PD, Young AT, Sarwark JF et al (1999) Anterior only fusion for scoliosis in patients with myelomeningocele. Clin Orthop Relat Res 364:117–124 26. Wang WJ, Yeung HY, Chu WC et al (2011) Top theories for the etiopathogenesis of adolescent idiopathic scoliosis. J Pediatr Orthop 31:S14–S27 27. Lao LF, Chen ZG, Qiu GX et al (2013) Whole-spine magnetic resonance imaging study in healthy Chinese adolescents. Orthop Surg 5:164–170 28. Sun X, Qiu Y, Zhu Z et al (2007) Variations of the position of the cerebellar tonsil in idiopathic scoliotic adolescents with a Cobb angle[40°: a magnetic resonance imaging study. Spine (Phila Pa 1976) 32:1680–1686 29. Cobb J (1948) Outline for the study of scoliosis. In: Instr Course Lect. American Academy of Orthopaedic Surgeons, Ann Arbor, MI, pp 61–75
123
Eur J Orthop Surg Traumatol 30. McCance SE, Denis F, Lonstein JE et al (1998) Coronal and sagittal balance in surgically treated adolescent idiopathic scoliosis with the King II curve pattern. A review of 67 consecutive cases having selective thoracic arthrodesis. Spine (Phila Pa 1976) 23:2063–2073 31. Gelb DE, Lenke LG, Bridwell KH et al (1995) An analysis of sagittal spinal alignment in 100 asymptomatic middle and older aged volunteers. Spine (Phila Pa 1976) 20:1351–1358 32. Kuklo TR, Potter BK, Polly DW Jr et al (2005) Monaxial versus multiaxial thoracic pedicle screws in the correction of adolescent idiopathic scoliosis. Spine (Phila Pa 1976) 30:2113–2120 33. Qiu G, Qiu Y, Zhu Z et al (2011) Re-evaluation of reliability and validity of simplified Chinese version of SRS-22 patient questionnaire: a multicenter study of 333 cases. Spine (Phila Pa 1976) 36:E545–E550 34. Asher M, Lai SM, Burton D et al (2003) The reliability and concurrent validity of the Scoliosis Research Society-22 patient questionnaire for idiopathic scoliosis. Spine (Phila Pa 1976) 28:63–69 35. Qiu Y, He YX, Wang B et al (2007) The anatomical relationship between the aorta and the thoracic vertebral bodies and its importance in the placement of the screw in thoracoscopic correction of scoliosis. Eur Spine J 16:1367–1372 36. Sucato DJ, Duchene C (2003) The position of the aorta relative to the spine: a comparison of patients with and without idiopathic scoliosis. J Bone Jt Surg Am 85-A:1461–1469
123
37. Sarlak AY, Buluc L, Sarisoy HT et al (2008) Placement of pedicle screws in thoracic idiopathic scoliosis: a magnetic resonance imaging analysis of screw placement relative to structures at risk. Eur Spine J 17:657–662 38. Milbrandt TA, Sucato DJ (2007) The position of the aorta relative to the spine in patients with left thoracic scoliosis: a comparison with normal patients. Spine (Phila Pa 1976) 32:E348–E353 39. Richards BS, Sucato DJ, Johnston CE et al. (2010) Right thoracic curves in presumed adolescent idiopathic scoliosis: Which clinical and radiographic findings correlate with a preoperative abnormal magnetic resonance image? Spine (Phila Pa 1976) 35: 1855–1860 40. Coonrad R, Richardson W, Oakes W (1985) Left thoracic curves can be different. Orthop Trans 9:126–129 41. Ozerdemoglu RA, Denis F, Transfeldt EE (2003) Scoliosis associated with syringomyelia: clinical and radiologic correlation. Spine (Phila Pa 1976) 28:1410–1417 42. Wang W, Zhu Z, Zhu F et al (2008) The changes of relative position of the thoracic aorta after anterior or posterior instrumentation of type I Lenke curve in adolescent idiopathic thoracic scoliosis. Eur Spine J 17:1019–1026
ORIGINAL ARTICLE
Does curve convexity affect the surgical outcomes of thoracic adolescent idiopathic scoliosis? Wei-Jun Wang • Ai-Bing Huang • Ze-Zhang Zhu Feng Zhu • Xu Sun • Yong Qiu
•
Received: 20 December 2013 / Accepted: 2 February 2014 Ó Springer-Verlag France 2014
Abstract Background Major left thoracic (LT) curve is an atypical type in adolescent idiopathic scoliosis (AIS) and showed independent clinical characteristics and natural history compared to major right thoracic (RT) curve. However, it’s unclear whether the convexity of major thoracic curve would affect the surgical outcomes and risk of complications. A retrospective follow-up study was conducted to investigate whether the convexity of major thoracic curve would affect the surgical outcomes of patients with main thoracic AIS. Methods Twelve LT-AIS patients underwent corrective spinal instrumentation and fusion were retrieved, and twelve patients with main RT-AIS matched for gender, chronological age, curve type, magnitude and surgical strategy were selected as control. All patients underwent at least 2-year follow-up. The pre- and post-operative radiographic parameters, intraoperative data and functional outcome assessed by Scoliosis Research Society questionnaire 22 (SRS-22) were analyzed and compared between two groups. Results Patients with LT- and RT-AIS presented with similar magnitudes of thoracic curves, flexibility, fusion level and correction rate. Compared with RT-AIS, patients with LT-AIS showed longer operation time (average, 364 vs. 348 min) and larger amount intraoperative estimated blood loss (2,060 vs. 1,720 ml) although the differences were not statistically significant (p [ 0.05). With at least
W.-J. Wang A.-B. Huang Z.-Z. Zhu F. Zhu X. Sun Y. Qiu (&) Spine Surgery, The Affiliated Drum Tower Hospital of Nanjing University Medical School, Nanjing 210008, China e-mail: [email protected]
2-year follow-up, patients in two groups showed comparable loss of correction, coronal and sagittal balance, and the sagittal profiles. With regard to functional outcome, the scores of five categories of SRS-22 questionnaire were similar between two groups. No neurological or vascular complication was observed in these patients. Conclusions The radiographic and functional outcomes of LT-AIS patients underwent operation were comparable to those with RT-AIS. Longer operation time and more intraoperative blood loss may be expected in instrumentation and fusion for patients with LT-AIS, which might be because of the inconvenience on the surgical procedure. Keywords Idiopathic scoliosis Adolescent Thoracic Direction Correction
Introduction Scoliosis is a curvature of the spine on the coronal plane, along with deformity on the sagittal and horizontal planes. According to the location of the main curve on the coronal plane, the deformity may be classified as single major thoracic, single thoracolumbar, single lumbar, double major thoracic, double major curves, or as a multiple curve pattern [1, 2]. Within these curve types, the major thoracic curve has been found to be the most common type [3–5]. Most patients have right convex major thoracic (RT) curve and show a normal neural axis [1, 6–8]. In contrast, patients with major left convex thoracic (LT) scoliosis are rare, with the prevalence varying from 1.3 to 9.1 % [1, 9–13]. Furthermore, functional or radiological neural abnormalities have been reported in 20.4–54 % of ‘‘idiopathic’’ scoliosis patients with major LT curve [1, 6, 7, 14].
123
Eur J Orthop Surg Traumatol
Therefore, the major left thoracic curve has been defined as an atypical curve pattern and a strong indicator for associated intraspinal neuroanatomic abnormalities of scoliosis [14–18]. Scoliosis with left convex thoracic curve has a lower morbidity but higher risk of associated neural axis abnormalities compared with RT scoliosis; moreover, this condition typically presents with a smaller curve magnitude at diagnosis, a faster growth rate, but a lower likelihood of progression [19–22]. These findings indicate that patients with LT scoliosis might have an independent behavior from those with RT scoliosis, leading to a different surgical outcome. The surgical strategies used and the short-term and long-term outcomes are different between patients with LT scoliosis associated with neural axis abnormalities, such as Chiari malformation and syringomyelia, and patients with right thoracic scoliosis with normal neural axis [23–25]. However, no study has reported whether the surgical outcomes of LT scoliotic patients with normal neural axis were different from those with RT scoliosis. Adolescent idiopathic scoliosis (AIS) is the most common type of scoliosis [3, 26]. In the present study, the radiological and functional outcomes of surgical treatment of AIS patients with main left and right thoracic curves were compared to determine any affects due to curve lateralization (Figs. 1, 2).
Materials and methods
Fig. 1 A female AIS patient with a left convex major thoracic curve at 15 years old. The thoracic curve was corrected from 46° preoperatively (a) to 17° at 2 years postoperatively (c), with a
correction rate of 63 %. Thoracic kyphosis was 7° preoperatively (b) and 13° at the latest follow-up (d)
123
Subjects To identify patients with LT-AIS, AIS patients receiving surgical treatment in our center between January 2003 and December 2010 were retrospectively reviewed in the database. Patients with a main left convex thoracic curve were recruited into this study if they met the following criteria: (1) a radiological and clinical diagnoses of AIS with a normal spinal cord and a cerebral tonsil position confirmed by magnetic resonance (MR) scanning [27, 28]; (2) a single main thoracic curve with the apex located between T7 and T9; (3) age ranging from 12 to 20 years; (4) a posterior-only selective thoracic fusion and instrumentation with all pedicle screws; (5) a minimum of 2 years of follow-up after the operation. Patients meeting the following criteria were excluded: (1) any form of prior surgical treatment for scoliosis; (2) neural abnormalities, i.e., low-lying cerebellar tonsil or conus medullaris and syringomyelia. We identified only 12 patients out of a total database of 1,534 surgically treated AIS. Among the 1,534 AIS patients, those with a typical main RT curve meeting the criteria for LT-AIS were identified. Patients who matched in gender, chronological age, Risser stages, curve patterns and magnitude, and curve flexibility were considered as adequate controls for LT-AIS. From
Eur J Orthop Surg Traumatol
these patients, a random sample of 12 RT-AIS patients having a major curve within 5° of the LT-AIS patients were selected as controls [29]. The gender distribution, chronological age, and skeletal maturity at the time of operation are shown in Table 1. According to the Lenke classification [4], eight patients were defined as Lenke type 1A, two as Lenke type 1B, and two as Lenke type 1C in each group. All patients were treated with posterior correction and instrumentation by the same surgical team. Comparable instrumentation strategies, including implant density, were used between the two groups [4]. The implants used included TSRH, CDH, or CDH legacy (Medtronic, Memphis, USA) in different periods. Surgeons always stood on the concave side of patients while performing the surgical procedures. Thoracoplasty was carried out in patients with significant rib hump after curve correction. Iliac crest bone grafts were harvested as autografts for all posterior procedures. Somatosensory-evoked potentials, motor-evoked potentials, and Stagnara wake-up tests were used for intraoperative neurological monitoring. Intraoperative parameters were recorded, including the operation time, estimated blood loss, volume of blood transfusion, number of vertebrae fused, and the number of patients underwent thoracoplasty. Short-term and longterm complications were also recorded.
Fig. 2 A female AIS patient with a right convex major thoracic curve at 15 years of age. The preoperative thoracic curve was 42° and kyphosis was 15°. The patient was instrumented with all pedicle
Radiographic evaluation Standing long-cassette posteroanterior and lateral radiographs of the entire spine, which were obtained preoperatively and postoperatively at 3 months, 1 year, 2 years, and at the time of the final follow-up, as well as preoperative right- and left-side bending radiographs, were reviewed and analyzed by an independent surgeon. The Cobb angle of the main thoracic curve, lower instrumented vertebra (LIV) tilt, and global coronal balance (the horizontal distance between the C7 plumb line and the central sacral vertical line, C7CSVL) were measured on the posteroanterior radiographs. The implant density of the pedicle screws was calculated as follows: [(number of screws/number of pedicles in fusion levels)] 9 100 %. The coronal spinal decompensation was defined as a deviation of C7 greater than 2 cm from the CSVL [30]. On lateral radiographs, the thoracic kyphosis (T5–T12), thoracolumbar kyphosis (T10–L2), lumbar lordosis (T12–S1), and global sagittal balance (the horizontal distance from a vertical line extended from the center of the C7 vertebral body to the posterior superior cortex of S1, SVA) were quantified [30, 31]. Furthermore, the apical vertebral rotation was presented by the apical rib hump prominence (RH), which is the linear distance on a lateral radiograph, between the left and right ribs at the rib deformity apex or the most prominent rib [32]. Curve flexibility
screws. The thoracic curve was 12° and thoracic kyphosis was 18° at 2.6 years following curve correction
123
Eur J Orthop Surg Traumatol Table 1 Perioperative data for left and right thoracic AIS patients
Gender
LT-AIS (n = 12)
RT-AIS (n = 12)
7 Females, 5 males
7 Females, 5 males
p Value
Age at operation (years)
16.4 ± 2.2
16.0 ± 1.8
0.65
Risser sign Operative time (min)a
3.4 ± 1.1 364 ± 88
3.9 ± 1.2 348 ± 85
0.14 0.54
Estimated blood loss (ml)a
2,060 ± 830
1,720 ± 442
0.17
Blood transfusion (ml)a
1,190 ± 482
1,040 ± 493
0.39
No. of vertebrae in fusion
11.4 ± 1.9
11.4 ± 2.5
1.00
Follow-up (years)
5.8 ± 3.2
5.4 ± 2.9
0.48
AIS adolescent idiopathic scoliosis, LT left thoracic, RT right thoracic a
Comparisons were performed by nonparametric analysis
was determined on the preoperative supine side bending films. The flexibility index was acquired by subtracting the magnitude of the coronal Cobb angle on the supine bending radiograph from that on the preoperative standing radiograph, then dividing by the preoperative magnitude of the coronal Cobb angle on the standing radiograph. The curve correction rates, immediately after surgery and at the followup, and loss of correction were calculated. SRS-22 questionnaire To determine the functional outcome of surgery, the Scoliosis Research Society questionnaire 22 (SRS-22, Chinese version) [33] was completed by patients at the last follow-up and was reviewed. The SRS-22 contains 22 questions covering five domains: function/activity, pain, self-perceived image, mental health, and satisfaction with treatment [34]. Statistical analysis The results were presented as mean and standard division. Statistical analysis was performed with Statistical Package for the Social Sciences (SPSS, version 13.0, Chicago, IL, USA). Clinical data were compared using Student’s t tests or the Wilcoxon’s ranks test between the two groups. The preoperative and postoperative radiological parameters in each group, as well as differences in changes in these parameters between the two groups, were compared by repeated-measures analysis. Significance was defined as p \ 0.05.
grade, and gender distribution (Table 1). The average fusion levels were 11.7 ± 1.9 (8–14) and 11.1 ± 2.5 (8–15), while the implant densities were 46.5 ± 7.5 and 50.2 ± 9.2 % in patients with LT-AIS and RT-AIS, respectively (p [ 0.05). In each group, 7 of 12 patients underwent thoracoplasty to correct the rib hump. Despite the fact that differences in the perioperative variables were not statistically significant, patients with LT-AIS tended to have a longer operation time (average, 363.5 ± 87.7 vs. 347.5 ± 85.4 min), greater intraoperative blood loss (average, 2,060.0 ± 830.3 vs. 1,720.0 ± 441.7 ml), and larger volume of blood transfusion (average, 1,190.0 ± 481.8 vs. 1,040.0 ± 492.6 ml) than those with RT-AIS (Table 1); however, the difference was not statistically significant (p [ 0.05). The patients with LTand RT-AIS were followed for an average of 5.8 and 5.4 years, respectively. Radiographic results Correction of major thoracic curve The patients in both groups were well matched in terms of the magnitude and flexibility of the primary thoracic curve. In patients with LT-AIS, the major thoracic curve was corrected from 59.3° to 28.9°, immediately following the operation, with an overall correction rate of 54.2 %. The mean LT curve was 30.7°, 29.9°, and 30.4° at 1, 2, and 5.8 years postoperatively, respectively, with a mean 1.5° loss of correction during follow-up. In the matched patients with RT-AIS, the preoperative thoracic curve of 60.5° was corrected to 27.4°, immediately following surgery, with an overall correction rate of 56.7 %. During follow-up, the mean RT curve was 26.8°, 28.0°, and 29.3° at 1, 2, and 5.4 years following surgery, respectively, with a mean 1.9° loss of correction at the final follow-up. Repeated-measures analysis revealed significant differences between pre- and postoperative measures, but with no difference during different points of follow-up, the curve correction rate and the mean loss of correction of the primary thoracic curve between the two groups showed no significant difference. The tilt of LIV also improved significantly from 13.8° to 3.4° in LT-AIS and 14.0° to 2.9° in RT-AIS following surgery, respectively, (p \ 0.05). A slight deterioration of LIV was noted in patients with either LT-AIS or RT-AIS during follow-up (1.7° vs. 1.6°); however, there was no significant difference between the two groups (Table 2).
Results Sagittal profile and rib hump correction Clinical results Due to the recruitment criteria, the AIS patients with LT and RT curves were similar in chronological age, Risser
123
Measurements from preoperative sagittal X-ray films showed significantly larger thoracic kyphosis in LT-AIS patients, compared to that in RT-AIS. The difference was
Eur J Orthop Surg Traumatol Table 2 Comparison of radiological data between left and right thoracic AIS patients LT-AIS (n = 12)
in patients from both groups; however, no significant difference was recorded between the two groups (Table 2).
RT-AIS (n = 12)
Global balance Main thoracic curve (°) Pre-op
59.3 ± 19.6 (43–99)
60.5 ± 18.6 (45–95)
Flexibility (%)
43.9 ± 18.4
42.9 ± 14.8
Immediate post-op
28.9 ± 18.4
27.4 ± 16.6
Correction rate (%)
54.2 ± 20.0
56.7 ± 13.0
1 year post-op
30.7 ± 20.1
26.8 ± 17.2
2 year post-op
29.9 ± 19.7
28.0 ± 18.3
Latest follow-up
30.4 ± 18.3
29.3 ± 17.1
1.5 ± 2.4
1.9 ± 1.2
13.8 ± 5.9
14.0 ± 9.5
Loss of correction LIV tilt angle (°)* Pre-op Immediate post-op
3.4 ± 3.8^^
2.9 ± 3.2^^
Latest follow-up
5.1 ± 4.5^^
4.5 ± 3.6^^
Complications
T5–T12 (°) Preoperativea
30.9 ± 22.9
14.2 ± 10.8*
Immediatea
28.2 ± 13.2
18.8 ± 7.1*
Final follow-upa
29.5 ± 13.6
17.9 ± 8.4*
Preoperativea
3.4 ± 4.8
0.2 ± 11.7
Immediatea
1.2 ± 6.2
2.0 ± 4.8
Final follow-upa
2.4 ± 7.7
1.9 ± 4.1
Preoperative
48.1 ± 13.4
40.3 ± 23.1
Immediate
46.2 ± 11.2
40.7 ± 11.1
Final follow-up
47.8 ± 11.0
41.7 ± 10.0
Preoperative
2.3 ± 1.2
4.5 ± 1.3
Immediate
1.0 ± 0.8
1.2 ± 0.7
Final follow-up
1.1 ± 0.9
1.2 ± 1.0
T10–L2 (°)
L1–S1 (°)
Rib hump (cm)
C7-CSVL (cm) a
The coronal balance was similar between the two groups before the operation and at final follow-up, although LT-AIS patients showed significantly smaller coronal balance than RT-AIS patients, immediately following the operation (Table 2). None of the patients showed decompensation on coronal planes. Patients with LT-AIS showed a significantly larger SVA, compared to RT-AIS patients, before the operation, which indicated that patients with LTAIS had more imbalances on the sagittal plane. Following corrective surgery, however, patients in both groups showed comparable sagittal balance.
Pre-op
0.7 ± 0.7
0.6 ± 0.6
Immediate post-opa
0.6 ± 0.4
0.9 ± 0.3*
Latest follow-upa
0.4 ± 0.5
0.5 ± 0.5
Preoperativea
1.2 ± 0.8
0.6 ± 0.5*
Immediatea
0.9 ± 0.5
1.4 ± 0.7
Final follow-upa
1.0 ± 0.8
0.5 ± 0.7
C7PL (cm)
LIV low instrument vertebrae, CSVL central sacral vertical line, C7PL C7 plumb line, op operation
Two patients with LT-AIS had hydrothorax following thoracoplasty, but recovered 1 week after surgery. One patient with LT-AIS suffered a superficial wound infection following the operation. The wound healed, uneventfully, with debridement and delayed closure. In the patients with RT-AIS, two pedicle screws were found purchased slightly into the spinal canal, but no neurological symptoms were reported. For all other cases, no neurological, vascular, visceral complications, or venous thromboembolism was observed. No pseudarthrosis, implant breakage, or loosening was observed in patients in either group, during followup. There was no statistical difference in the incidence of complications in LT-AIS and RT-AIS patients. SRS-22 scores All patients completed the SRS-22 questionnaire at the last follow-up. The score for the function/activity domain was 4.1 and 4.2, while the score for the satisfaction with treatment domain was 4.2 and 4.4 in patients with LT- and RT-AIS, respectively. No significant differences were observed in the SRS-22 scores between the two groups, with respect to the pain, mental health, function/activity, self-image/appearance, or satisfaction of management domains (Table 3).
Significant change compared to pre-op data, ^ p \ 0.05, ^^ p \ 0.01 Significant difference between the two groups, * p \ 0.05 a
Comparisons were done by Mann–Whitney nonparametric analysis
also significant following surgery and at the last follow-up (p \ 0.05). In contrast, thoracolumbar kyphosis and lumbar lordosis were similar between the two groups before the operation, immediately after the operation, and at final follow-up. A significant correction of rib hump was found
Discussion Although differences between scoliosis with LT and RT have been identified in terms of the incidence of neural axis abnormalities and curve progression [19–22], the surgical outcomes of LT-AIS have not been well documented, and it has not been reported whether the surgical outcomes of
123
Eur J Orthop Surg Traumatol Table 3 Comparison scores of SRS-22 at the last follow-up between left and right thoracic AIS patients Domain
Group A (n = 12)
Group B (n = 12)
p Value
Function/activity
4.1 ± 0.5
4.2 ± 0.3
0.63
Pain
4.3 ± 0.5
4.0 ± 0.6
0.39
Self-image Mental health
4.0 ± 0.4 4.2 ± 0.4
4.2 ± 0.4 4.1 ± 0.5
0.62 0.80
Satisfaction with treatment
4.2 ± 0.4
4.4 ± 0.5
0.47
thoracic AIS are affected by curve convexity. Abnormal anatomical relationships between the spine and the perispinal organs have been well described in scoliotic patients, due to spinal deformity [35–37]. However, the pattern of these abnormalities is different in patients who suffer from RT and LT-AIS due to the different direction of curvature [38]. Spine surgeons, largely familiar with corrective procedures for RT-AIS, might be uncomfortable with the procedures for correcting LT-AIS. It has been of concern whether patients with LT-AIS could achieve the same surgical outcome and comparable incidence of complications, as those with RT-AIS. The rarity of the left thoracic curve pattern in idiopathic scoliosis was described by McCarver et al. [10] at the Scoliosis Research Society annual meeting in Toronto in 1,971. It was found that 1.3 % of 550 patients with idiopathic scoliosis had primary left thoracic curves. However, the diagnosis of idiopathic scoliosis in these patients was not confirmed by MR, and it is unknown as to whether some of them had neural axis abnormalities [39]. In addition, Coonrad et al. [40] found that 27 of 1,622 patients with noncongenital scoliosis presented with a major left convex thoracic curve. However, only 14 patients received a final diagnosis of idiopathic scoliosis, and the incidence of LT scoliosis was 0.86 %. This was inconsistent with other studies that found the incidences of neural axis abnormalities in patients with left thoracic scoliosis to range from 43 to 54 % [14, 17, 40]. In our center, routine preoperative MR scanning was performed for each surgical case, and AIS was diagnosed by confirming no neural axis abnormalities. From January 2003 to December 2010, 1,534 patients with AIS received corrective spinal surgery, and 12 of them had major left thoracic curves, without any form of neural axis abnormality, confirmed by MRI. Hence, the incidence of LT-AIS was 0.78 % in the surgically corrected AIS patients in our center. According to our knowledge, the present study may give the most accurate incidence of LT-AIS in the population of AIS patients requiring surgical intervention. Although left thoracic scoliosis are commonly associated with neural axis abnormalities [14–18], the surgical
123
outcomes of these patients were not compared with patients with RT-AIS in the present study. Patients with neural axis abnormalities such as Chiari malformation or syringomyelia would present with hyperkyphosis in thoracic spine, younger at the time of needing surgical correction, longer fusion levels, higher implant density, and higher risk of neurological complications [41]. With the aim to study whether the direction of major thoracic curve would affect the surgical outcome of AIS, only scoliosis patients with normal neural axis confirmed by MR scanning were recruited in the present study. By comparing the radiographic profiles of LT-AIS patients to those with RT-AIS, a significantly higher incidence of thoracic kyphosis was identified in patients with LT-AIS. This finding indicated that hyperkyphosis might be a coexistent phenotype of left thoracic scoliosis instead of an indicator of neural axis abnormality in scoliosis patients. Careful preoperative planning for surgical strategies and procedures were made for each LT-AIS patient, considering the different anatomical relationship between the spine and peri-spinal organs, as well as inconveniences of surgical procedures. In patients with RT-AIS, the thoracic vertebrae are in right-side torsion, the thoracic aorta is in a more posterolateral position related to vertebrae, and the spinal cord is more close to pedicle in left side when compare to normal subjects [36, 42]. In contrast, in patients with left thoracic curve, the thoracic aorta locate in a more anterolateral position and the spinal cord is more close to pedicle in right side compared to normal subjects [38]. Hence, the incidences of aortic and neurological complications by pedicle screws insertion were higher in the leftside pedicles in RT-AIS, but in the right side in LT-AIS when compare to the other side. Since spinal surgeons are accustomed to the surgical procedures of RT-AIS correction and are more commonly right handed, they may feel uncomfortable when performing surgery for LT-AIS, and the procedures might not be performed in a very fluent manner. Theoretically, the different pattern of abnormalities in LT-AIS from RT-AIS would result in higher incidence of aortic and neurological complications. By comparing the surgical treatment outcomes between AIS patients with either left or right thoracic curves, matched for gender, chronological age, curve type, major curve magnitude, and instrumentation, our results indicate that patients with LT-AIS have a tendency to require a longer duration of surgery, have a greater intraoperative blood loss, and require larger volumes of blood transfusions; however, the differences were not statistically significant. Moreover, the radiological outcomes in both the coronal planes and sagittal planes were comparable between patients with LT- and RT-AIS. In addition, the risk of complications is not higher for surgical correction in LTAIS patients.
Eur J Orthop Surg Traumatol
Despite the radiological outcome, the functional outcome is another important issue in surgical treatment for AIS. The simplified Chinese version of the Scoliosis Research Society 22 was recently made available for clinical evaluation of Chinese adolescents and young adults with scoliosis treated with surgery [33]. All patients enrolled in our study completed the SRS-22 questionnaire at their last follow-up. Patients in both groups showed similar scores for each of the five domains: function/ activity, pain, self-image, mental health, and satisfaction with treatment. These results indicate that patients with LT-AIS and RT-AIS have comparable functional outcomes when using the same surgical strategies.
Conclusion The present study compared surgical outcomes between left and right thoracic AIS patients who were matched for gender, age, curve type, major curve magnitude, and instrumentation. Our results showed that LT-AIS can be corrected as effectively as RT-AIS in terms of risk of complications and radiological and functional outcomes. Longer operation times and greater intraoperative blood loss can be expected for LT-AIS. Acknowledgments This work was supported by National Natural Science Foundation of China (81101335), National Post-doctoral Foundation of China (2012M52101062), National Key Clinical Specialty Construction Project in Orthopaedics, and Jiangsu Province’s Key Medical Talents Project (RC2011149). Conflict of interest
None.
References 1. Coonrad RW, Murrell GA, Motley G et al (1998) A logical coronal pattern classification of 2,000 consecutive idiopathic scoliosis cases based on the scoliosis research society-defined apical vertebra. Spine (Phila Pa 1976) 23:1380–1391 2. King HA, Moe JH, Bradford DS et al (1983) The selection of fusion levels in thoracic idiopathic scoliosis. J Bone Jt Surg Am 65:1302–1313 3. Weinstein SL, Dolan LA, Cheng JC et al (2008) Adolescent idiopathic scoliosis. Lancet 371:1527–1537 4. Lenke LG, Betz RR, Harms J et al (2001) Adolescent idiopathic scoliosis: a new classification to determine extent of spinal arthrodesis. J Bone Jt Surg Am 83-A:1169–1181 5. Qiu G, Zhang J, Wang Y et al (2005) A new operative classification of idiopathic scoliosis: a Peking Union Medical College method. Spine (Phila Pa 1976) 30:1419–1426 6. Hausmann ON, Boni T, Pfirrmann CW et al (2003) Preoperative radiological and electrophysiological evaluation in 100 adolescent idiopathic scoliosis patients. Eur Spine J 12:501–506 7. Inoue M, Minami S, Nakata Y et al (2005) Preoperative MRI analysis of patients with idiopathic scoliosis: a prospective study. Spine (Phila Pa 1976) 30:108–114
8. Richards BS, Sucato DJ, Johnston CE et al (2010) Right thoracic curves in presumed adolescent idiopathic scoliosis: which clinical and radiographic findings correlate with a preoperative abnormal magnetic resonance image? Spine (Phila Pa 1976) 35:1855–1860 9. James JI (1954) Idiopathic scoliosis; the prognosis, diagnosis, and operative indications related to curve patterns and the age at onset. J Bone Jt Surg Br 36-B:36–49 10. McCarver C, Levine D, Velaskis K (1971) Left thoracic curve patterns in idiopathic scoliosis. J Bone Jt Surg Am 53:196 11. Ylikoski M (2005) Growth and progression of adolescent idiopathic scoliosis in girls. J Pediatr Orthop B 14:320–324 12. Qiu Y, Zhu F, Wang B et al (2009) Clinical etiological classification of scoliosis: report of 1,289 cases. Orthop Surg 1:12–16 13. Elsig JPJ, Kaech DL (2007) Dynamic imaging of the spine with an open upright MRI: present results and future perspectives of fmri. Eur J Orthop Surg Traumatol 17:119–124 14. Wu L, Qiu Y, Wang B et al (2010) The left thoracic curve pattern: a strong predictor for neural axis abnormalities in patients with ‘‘idiopathic’’ scoliosis. Spine (Phila Pa 1976) 35:182–185 15. Spiegel DA, Flynn JM, Stasikelis PJ et al (2003) Scoliotic curve patterns in patients with Chiari I malformation and/or syringomyelia. Spine (Phila Pa 1976) 28:2139–2146 16. Qiu Y, Zhu ZZ, Wang B et al (2008) Radiological presentations in relation to curve severity in scoliosis associated with syringomyelia. J Pediatr Orthop 28:128–133 17. Goldberg CJ, Moore DP, Fogarty EE et al (1999) Left thoracic curve patterns and their association with disease. Spine (Phila Pa 1976) 24:1228–1233 18. Wang W, Zhu Z, Zhu F et al (2012) Different curve pattern and other radiographical characteristics in male and female patients with adolescent idiopathic scoliosis. Spine (Phila Pa 1976) 37:1586–1592 19. Goldberg CJ, Dowling FE, Fogarty EE (1994) Left thoracic scoliosis configurations. Why so different? Spine (Phila Pa 1976) 19:1385–1389 20. Soucacos PN, Zacharis K, Gelalis J et al (1998) Assessment of curve progression in idiopathic scoliosis. Eur Spine J 7:270–277 21. Chiu YL, Huang TJ, Hsu RW (1998) Curve patterns and etiologies of scoliosis: analysis in a university hospital clinic in Taiwan. Changgeng Yi Xue Za Zhi 21:421–428 22. Gupta R, Sharma R, Vashisht S et al (1999) Magnetic resonance evaluation of idiopathic scoliosis: a prospective study. Australas Radiol 43:461–465 23. Bradley LJ, Ratahi ED, Crawford HA et al (2007) The outcomes of scoliosis surgery in patients with syringomyelia. Spine (Phila Pa 1976) 32:2327–2333 24. Ferguson RL, DeVine J, Stasikelis P et al (2002) Outcomes in surgical treatment of ‘‘idiopathic-like’’ scoliosis associated with syringomyelia. J Spinal Disord Tech 15:301–306 25. Sponseller PD, Young AT, Sarwark JF et al (1999) Anterior only fusion for scoliosis in patients with myelomeningocele. Clin Orthop Relat Res 364:117–124 26. Wang WJ, Yeung HY, Chu WC et al (2011) Top theories for the etiopathogenesis of adolescent idiopathic scoliosis. J Pediatr Orthop 31:S14–S27 27. Lao LF, Chen ZG, Qiu GX et al (2013) Whole-spine magnetic resonance imaging study in healthy Chinese adolescents. Orthop Surg 5:164–170 28. Sun X, Qiu Y, Zhu Z et al (2007) Variations of the position of the cerebellar tonsil in idiopathic scoliotic adolescents with a Cobb angle[40°: a magnetic resonance imaging study. Spine (Phila Pa 1976) 32:1680–1686 29. Cobb J (1948) Outline for the study of scoliosis. In: Instr Course Lect. American Academy of Orthopaedic Surgeons, Ann Arbor, MI, pp 61–75
123
Eur J Orthop Surg Traumatol 30. McCance SE, Denis F, Lonstein JE et al (1998) Coronal and sagittal balance in surgically treated adolescent idiopathic scoliosis with the King II curve pattern. A review of 67 consecutive cases having selective thoracic arthrodesis. Spine (Phila Pa 1976) 23:2063–2073 31. Gelb DE, Lenke LG, Bridwell KH et al (1995) An analysis of sagittal spinal alignment in 100 asymptomatic middle and older aged volunteers. Spine (Phila Pa 1976) 20:1351–1358 32. Kuklo TR, Potter BK, Polly DW Jr et al (2005) Monaxial versus multiaxial thoracic pedicle screws in the correction of adolescent idiopathic scoliosis. Spine (Phila Pa 1976) 30:2113–2120 33. Qiu G, Qiu Y, Zhu Z et al (2011) Re-evaluation of reliability and validity of simplified Chinese version of SRS-22 patient questionnaire: a multicenter study of 333 cases. Spine (Phila Pa 1976) 36:E545–E550 34. Asher M, Lai SM, Burton D et al (2003) The reliability and concurrent validity of the Scoliosis Research Society-22 patient questionnaire for idiopathic scoliosis. Spine (Phila Pa 1976) 28:63–69 35. Qiu Y, He YX, Wang B et al (2007) The anatomical relationship between the aorta and the thoracic vertebral bodies and its importance in the placement of the screw in thoracoscopic correction of scoliosis. Eur Spine J 16:1367–1372 36. Sucato DJ, Duchene C (2003) The position of the aorta relative to the spine: a comparison of patients with and without idiopathic scoliosis. J Bone Jt Surg Am 85-A:1461–1469
123
37. Sarlak AY, Buluc L, Sarisoy HT et al (2008) Placement of pedicle screws in thoracic idiopathic scoliosis: a magnetic resonance imaging analysis of screw placement relative to structures at risk. Eur Spine J 17:657–662 38. Milbrandt TA, Sucato DJ (2007) The position of the aorta relative to the spine in patients with left thoracic scoliosis: a comparison with normal patients. Spine (Phila Pa 1976) 32:E348–E353 39. Richards BS, Sucato DJ, Johnston CE et al. (2010) Right thoracic curves in presumed adolescent idiopathic scoliosis: Which clinical and radiographic findings correlate with a preoperative abnormal magnetic resonance image? Spine (Phila Pa 1976) 35: 1855–1860 40. Coonrad R, Richardson W, Oakes W (1985) Left thoracic curves can be different. Orthop Trans 9:126–129 41. Ozerdemoglu RA, Denis F, Transfeldt EE (2003) Scoliosis associated with syringomyelia: clinical and radiologic correlation. Spine (Phila Pa 1976) 28:1410–1417 42. Wang W, Zhu Z, Zhu F et al (2008) The changes of relative position of the thoracic aorta after anterior or posterior instrumentation of type I Lenke curve in adolescent idiopathic thoracic scoliosis. Eur Spine J 17:1019–1026
