Upload
others
View
1
Download
0
Embed Size (px)
Citation preview
QUT Digital Repository: http://eprints.qut.edu.au/
Hay, Douglas and Izatt, Maree T. and Adam, Clayton J. and Labrom, Robert D. and Askin, Geoffrey N. (2009) Radiographic outcomes over time after endoscopic anterior scoliosis correction. Spine, 34(11). pp. 1176-1184.
© Copyright 2009 Lippincott Williams & Wilkins
Radiographic outcomes over time after endoscopic scoliosis correction
1
1 2
Radiographic outcomes over time following endoscopic anterior scoliosis correction: A 3 prospective series of 106 patients 4
5 6
7 Douglas Hay (FRCS), Maree T Izatt (BPhty), Clayton J Adam (PhD), Robert D Labrom (MSc, 8
FRACS), Geoffrey N Askin (FRACS) 9
10
Paediatric Spine Research Group, Queensland University of Technology and Mater Health Services 11
Brisbane Ltd, Queensland, Australia 12
13
14
Name and Address for Correspondence 15
Associate Professor Clayton Adam 16
Paediatric Spine Research Group 17
Level 2, Mater Children’s Hospital 18
Raymond Terrace, South Brisbane, Qld, 4101, Australia 19
Phone: +61 7 3163 6162 20
Mobile: 0405 148 686 21
Fax: +61 7 3163 1744 22
Email: [email protected] 23
24
25
Sources of Support 26
No financial support was received for this study. 27
28
29
Radiographic outcomes over time after endoscopic scoliosis correction
2
Keywords 30
anterior scoliosis surgery, endoscopic anterior scoliosis correction, anterior thoracoscopic 31
instrumentation, major Cobb angle, instrumented Cobb angle, curve decompensation 32
33
Introduction 34
Since the use of a thoracoscope to access the spine anteriorly was first reported in 1993,1 35
minimally invasive approaches for scoliosis correction have become an alternative to open 36
surgical techniques for selected cases.2-14 Endoscopic anterior scoliosis correction has been used 37
for the past decade, with reported benefits of less fused levels, sagittal profile restoration, 38
reduced pain and chest wall morbidity, shorter hospital stay, faster recovery of lung function, 39
better cosmesis, and lower infection rates, blood loss and lower incidence of neurological 40
complications.6-16 41
42
However, the technique is technically demanding with a substantial surgical learning 43
curve.12,17,18 Due to the level-sparing approach, any potential for decompensation following 44
endoscopic anterior surgery is of considerable clinical interest. Several studies have reported on 45
Cobb angle correction at a single time point after surgery,6,12,14,17-19 but to our knowledge only 46
one paper has reported correction at more than one time point post-operatively.20 However, this 47
paper did not perform statistical analysis of loss of correction versus time after surgery. In 48
addition, the two most commonly reported instrumentation related complications following 49
endoscopic scoliosis surgery are rod breakage and screw loosening, and it is important to qualify 50
whether these two occurrences are associated with greater decompensation following surgery. 51
52
Accordingly, the aim of this study was to prospectively analyse how key radiological parameters 53
and rib hump change during the two years following endoscopic anterior scoliosis correction 54
Radiographic outcomes over time after endoscopic scoliosis correction
3
surgery. The effects of instrumentation related complications on these changes are also 55
investigated. 56
57
Materials and Methods 58
Study Cohort. Between April 2000 and June 2006, a total of 106 patients underwent endoscopic 59
anterior instrumented fusion using a single rod technique to correct progressive scoliosis. The 60
study data was gathered prospectively for all cases. The option to undergo an endoscopic 61
procedure was presented to each patient after clinical and radiologic assessment by the senior 62
authors to assess suitability. Patients and/or their parents were given the option of either open 63
posterior or endoscopic anterior surgery, and the benefits, risks, and potential complications 64
associated with each approach were presented. 65
66
Deformity Details. The 99 idiopathic curves were classified according to the Lenke classification 67
with the majority (88 of 99 = 89%) Lenke Type 1 curves, and eight Type 2, one Type 3 and two 68
Type 5 curves. The remaining seven cases had a small associated syrinx or dilated central canal 69
found by magnetic resonance imaging, despite having presented with an idiopathic type curve. 70
104 patients had right sided thoracic major curves and two had left sided thoracolumbar curves. 71
All patients had either a normal thoracic kyphosis or hypokyphosis (three patients had a thoracic 72
lordosis) with a mean (± SD) of 18.4±9.8° (range -13 to 40). Mean preoperative rib hump 73
measurement was 16.5±4.4° (range 10 to 30). 74
75
Surgical Technique. The procedures were performed by the two senior authors (GNA and 76
RDL) at the Mater Children's Hospital in Brisbane, Australia. The surgical technique has been 77
reported previously.11,12 Briefly, the disc spaces of the levels to be instrumented are cleared and 78
the intervertebral spaces packed with either femoral head allograft (62 cases) or mulched 79
Radiographic outcomes over time after endoscopic scoliosis correction
4
autograft (rib heads for 38 cases, iliac crest for 6 cases). Allograft is supplied through the 80
Queensland Bone Bank (either cadaveric donation or femoral head donation at time of hip 81
replacement). Bone banking in Australia is well established & tightly regulated by health 82
authorities. Donors are screened using stringent protocols employed for organ donation. The 83
bone is irradiated & stored at temperatures in the vicinity of -70 Celsius. Autograft was used 84
early in the series but due to donor site pain and the inadequate volume of bone available, an 85
alternative was sought to achieve what our surgeons consider to be optimum conditions for bony 86
fusion. If instrumentation extended beyond T12, an interbody spacer cage packed with graft 87
material was placed between T12-L1 to assist the spine’s transition into lordosis. A single 88
4.5mm diameter rod was used for the first 78 procedures and a 5.5mm diameter rod was used for 89
all subsequent cases. The Eclipse (98 cases) or Legacy (8 cases) Instrumentation systems 90
(Medtronic Sofamor Danek, Memphis, TN) were used to achieve curve correction using a 91
standard compression technique with x-ray monitoring. The Legacy system assists the surgeon to 92
avoid cross threading of the screw head which may result in loosening of the nut and is now 93
standard practice, differing only in the design of the screw-rod interface. 94
95
Radiographic Evaluation. All patients had a standardised postero-anterior and lateral standing 96
radiograph using a long 36 inch plate with a grid, prior to surgery, and at 2, 6, 12 and 24 months 97
postoperatively. The use of postero-anterior radiographs has been shown to reduce breast 98
irradiation by 92%, and by >99% when combined with shielding and filtration21. Radiographic 99
parameters were measured using the Cobb method at all review appointments by experienced 100
spinal orthopaedic surgeons, according to Scoliosis Research Society definitions (SRS Revised 101
Glossary of Terms). 102
103
Radiographic outcomes over time after endoscopic scoliosis correction
5
The following radiological parameters were investigated; the Cobb angles of the major curve, the 104
distal compensatory curve, the instrumented curve and T5-T12 sagittal kyphosis, shoulder 105
balance, and coronal spinal balance (distance of midpoint of C7 body from the central sacral line 106
on a posteroanterior radiograph). Sagittal spinal balance (distance of midpoint of the C7 body to 107
a vertical line through the posterior superior corner of the sacrum on lateral radiograph) was not 108
analysed as erratic results due to variations in patient trunk and arm positioning raised doubts as 109
to the reliability of the results. Proximal compensatory curves were present and measured in too 110
few patients to analyse, as the patient group that suits the surgical technique typically do not 111
exhibit this curve to any degree. 112
113
After surgery, there is a distinction made between the major Cobb angle and the instrumented 114
Cobb angle. The major Cobb angle is a true measure according to the definition of Cobb,22 i.e., 115
between the most inclined endplates at the proximal and distal ends of the postoperative major 116
curve. The instrumented Cobb angle is measured only for the instrumented vertebral levels, and 117
therefore does not always encapsulate the full extent of the postoperative major curve. This 118
distinction is illustrated in Figure 1. The curve correction or correction rate is defined as the 119
difference in Cobb angle after surgery divided by the preoperative Cobb angle and is expressed 120
as a percentage of the preoperative major Cobb angle. 121
122
Rib Hump Correction. The rib hump or rotational distortion of the torso was assessed at each 123
medical review using an inclinometer (Scoliometer, Scoliosis Research Society, Milwaukee, WI) 124
with the standing patient in the forward flexed position, with knees locked, arms hanging and 125
palms opposed. 126
127
Radiographic outcomes over time after endoscopic scoliosis correction
6
Statistical Analysis. The mean and standard deviation of each radiographic parameter as well as 128
the rib hump were calculated preoperatively, and at 2, 6, 12 and 24 months after surgery. To 129
determine whether postoperative changes in radiographic parameters and rib hump were 130
statistically significant for the entire patient group, paired t-tests were used to compare values at 131
each review with the values at the subsequent review (ie preoperative vs 2 months, 2 vs 6 132
months, 6 vs 12 months, and 12 vs 24 months). The entire group was then divided into three 133
subgroups; subgroup 1 contained those patients with no mechanical complications over the 24 134
month postoperative period, subgroup 2 contained those patients in which a rod fracture occurred 135
in the 24 month postoperative period, and subgroup 3 contained those patients who experienced 136
screw-related complications (screw pullout or loosening) in the 24 month postoperative period. 137
Due to the smaller patient numbers in subgroups 2 and 3, non-parametric Wilcoxon signed ranks 138
tests were used to assess the significance of any changes between the review intervals. 139
140
Results 141
Patient Cohort. 106 patients (95 females, 11 males) were included in the study and all 142
underwent endoscopic anterior scoliosis correction. Data for all patients was available from the 143
two month review, 105 patients at 6 months, 103 patients at 1 year and 99 patients at two years. 144
Of the seven patients with incomplete follow-up at 24 months, two were from the screw-related 145
complication sub-group, and the remaining five were from the ‘no mechanical complications’ 146
subgroup. All seven patients with incomplete datasets are considered to be either lost to follow-147
up or unable to return for review due to geographical isolation. The mean age at surgery was 148
16.1 years (range 10-46), with 11 patients aged over 18 years who demonstrated sufficient major 149
and compensatory curve flexibility to be considered suitable for this selective anterior fusion 150
procedure. Figure 2 shows preoperative coronal, lateral and fulcrum bending radiographs and 151
postoperative coronal and lateral views for two representative patients from the series. 152
Radiographic outcomes over time after endoscopic scoliosis correction
7
153
Intrumentation Levels. The mean number of levels instrumented was 6.8 (range 5 to 9). The 154
most common proximal extension of the instrumentation was to T5 (n=41) and T6 (n=49), 155
though in one case T4 was instrumented. The most common distal instrumented levels were T11 156
(n=33) and T12 (n=51). Ten patients were instrumented to L1 and three to L2. 157
158
Mechanical Complications 159
There were 12 cases where the rod fractured (11.3%) and all occurred after the 12 month review, 160
being found on the 24 month radiographs despite the patients being asymptomatic. In another 12 161
cases there were screw related complications including eight with top screw partial pullout, two 162
with the bottom nut separated from the screw head, one with top screw plough, and one where 163
the top screw moved partially off the end of the rod. The screw related complications all 164
occurred early in the postoperative recovery period and were either found prior to discharge from 165
the hospital or on the two month radiograph. 166
167
Radiographic Results 168
The radiographic results for the entire patient series at all follow-up intervals are presented in 169
Table 1, including the correction rates for the major Cobb angle, the instrumented Cobb angle 170
and rib hump, and the distal compensatory curve. Table 2 gives the minimum, maximum, and 171
10th, 25th, 50th, 75th, 90th percentiles for progression of each radiographic parameter between 2 172
and 24 months after surgery. 173
174
Figure 3 displays the change in Cobb angle with time following surgery for the (i) major, (ii) 175
instrumented, (iii) distal compensatory, and (iv) T5-12 kyphosis curves as well as the changes in 176
rib hump over the same intervals. For the entire patient cohort, there were statistically significant 177
Radiographic outcomes over time after endoscopic scoliosis correction
8
increases in major and instrumented Cobb angle over each follow-up interval. There was a 178
statistically significant increase in rib hump (from 6.4 to 7.3 degrees) between 2 and 6 months 179
post-surgery. There were no other statistically significant changes between 2 and 24 months at 180
the P=0.05 level. Figure 4 illustrates the variation in major Cobb angle over time with the patient 181
cohort divided into the three subgroups described earlier (no mechanical complications, rod 182
fractures and screw related complications). Subgroup 1 (no complications) showed statistically 183
significant increases in major Cobb angle between 2 and 6 months, 6 and 12 months, and 12 and 184
24 months. Subgroup 2 (rod fractures) showed a significant increase in major Cobb angle 185
between 6 and 12 months (ie before the rod fractures occurred) but did not show a statistically 186
significant increase between 12 and 24 months despite this interval being where the rod fractures 187
occurred. Subgroup 3 (screw-related complications) only showed a statistically significant 188
increase between the 2 and 6 month review. Figures 5 - 8 present the results (using the same 189
three subgroups as in Figure 4) for instrumented Cobb angle, distal compensatory curve, T5-T12 190
sagittal kyphosis and rib hump values over time for the same three subgroups. In Figure 5, 191
subgroup 1 (no complications) showed statistically significant increases in instrumented Cobb 192
angle between 2 and 6 months, and between 12 and 24 months. Subgroup 2 (rod fractures) 193
showed a statistically significant increase between 2 and 6 months. In Figure 6, there were no 194
statistically significant changes in distal compensatory curve magnitude from 2 to 24 months. In 195
Figure 7, the only statistically significant increase in T5-T12 sagittal kyphosis was for subgroup 196
1 (no complications) between 2 and 6 months post-surgery. In Figure 8, there was a statistically 197
significant increase in rib hump for subgroups 1 and 2 from 6 to 12 months, and for subgroup 2 198
(rod fractures) from 12 to 24 months. 199
200
Examination of shoulder balance revealed that the mean shoulder height deviation for subgroup 201
1 (no mechanical complications) and subgroup 2 (rod fracture) was almost zero (
Radiographic outcomes over time after endoscopic scoliosis correction
9
surgery. At the 2 month radiograph, subgroups 1 and 2 had mean shoulder height differences of 203
7 and 10mm respectively, and by 24 months these had reduced to between 2.5 and 4.1mm. 204
Subgroup 3 (screw-related complications) had a mean preoperative shoulder height difference of 205
8.6±13mm, which reduced to almost level by the 2 month review (2.5± 11mm) and thereafter 206
remained stable. Table 1 gives mean values for shoulder height balance at all reviews for the 207
entire cohort. 208
209
Coronal spinal balance measurements revealed the mean deviation of the C7 vertebral body from 210
the central sacral vertical line before surgery to be between 0 and 3.2mm for subgroups 1 and 3, 211
and 5.4 ± 13mm for subgroup 2, the rod fracture group. Interestingly, all groups were deviated to 212
the left of the central sacral vertical line at all time intervals after surgery with a trend over time 213
decreasing the distance away from the central sacral vertical line. At the 12 month follow-up all 214
subgroup means were in the range of 4.2 to 5.5mm, and by 24 months remained deviated to the 215
left between 2.1 to 5.4 mm. Table 1 gives mean (±SD) values for coronal spinal balance at all 216
review time points for the entire cohort. 217
218
Discussion 219
The objectives of surgery in adolescent idiopathic scoliosis are to halt progression, permanently 220
correct the deformity in three dimensions, improve trunk appearance and keep short and long-221
term complications to a minimum.23 Excellent results have been published for clinical outcome, 222
coronal correction, and in some cases sagittal correction using multi-segmental dual rod 223
posterior systems.24-26 Against this benchmark, similar coronal correction has been reported for 224
anterior approaches with a reduced number of fused levels, and anterior approaches have also 225
been credited with superior kyphotic restoration.13,27-29 The endoscopic anterior approach is our 226
method of choice for selected single thoracic curves, and we have previously reported on 227
Radiographic outcomes over time after endoscopic scoliosis correction
10
pulmonary function restoration11, perioperative surgical aspects12, quality of life questionnaire 228
outcomes30, and the predictive ability of fulcrum bending radiographs31 for this technique. 229
Herein we report on the behaviour of the deformity correction with time after surgery, in order to 230
provide clinicians with quantitative data on the level of decompensation to expect in the two 231
years following endoscopic anterior scoliosis surgery. 232
233
The correction rate of the major curve in this study was in keeping with previously published 234
results for both open anterior and posterior techniques.8,13,14,19,24,28 During the two years 235
following surgery, there was a small (4°) loss of correction in the major Cobb angle, which is the 236
same as the value in the one existing longitudinal study mentioned earlier.20 Although this loss of 237
correction was statistically significant, it is less than the universally accepted 5° measurement 238
error for the Cobb technique.32-39 As a result, any loss of correction reported is not of sufficient 239
magnitude to be clinically relevant to the patient or surgeon. This also applies to the 2.8° loss of 240
correction for the instrumented curve between 2 and 24 months. The technique also 241
demonstrated significant improvement in the distal compensatory curve which was maintained 242
over time. The stable correction of the distal compensatory curve provides evidence to support 243
its exclusion from the fusion in this type of procedure. 244
245
With reference to Table 1, the overall difference between major and instrumented Cobb angles 246
increases slightly from 2.2° at 2 months, to 3.5° at two years. For the subgroup of screw related 247
complications the difference changes from 1° at 2 months to 2.2° at two years. The rod fracture 248
group shows a larger overall difference but less creep over time with 4.2° difference at 2 months 249
and 4.7° at 2 years. We suggest that it is useful to analyse the major and instrumented curves 250
separately, as it provides information on where any loss of correction may have occurred. 251
Wedging of the discs adjacent to the fused segment at either end of the construct contributes to 252
Radiographic outcomes over time after endoscopic scoliosis correction
11
the major Cobb angle. However, it is this commonly occurring wedging that allows for better 253
shoulder balance when the patient adjusts their posture after anterior fusion. Similarly, the rib 254
hump correction achieved during surgery is maintained over time. The rib hump demonstrated an 255
overall increase of only 1.5° degrees after surgery across the study period which is not of 256
sufficient size to be clinically relevant. 257
258
Mean T5-T12 kyphosis increased by 48% from 18 preoperatively to 27 at 2 months, and 259
further to 31 at 24 months, a 67% increase from the preoperative mean value, adding support to 260
the claim of sagittal profile improvement following anterior scoliosis correction surgery. 261
Although the rod fracture group had a significant increase in thoracic kyphosis between 12 and 262
24 months, the increase in kyphosis for the whole group was not just due to rod fractures, as 263
evidenced by the fact that the 24 month increase in kyphosis for subgroup 1 (no mechanical 264
complications), was also 67% relative to the mean preoperative value, the same as for the overall 265
group. 266
267
Anterior scoliosis correction using flexible rods has been associated with a higher complication 268
rate when compared with posterior segmental instrumented fusions.27, 40 Betz et al28 found a 31% 269
rod breakage rate compared with 1% for posterior instrumentation, but used smaller threaded 270
rods. Newton et al20 reported a 6% rod fracture rate for a group of 50 patients who had similar 271
procedures to the current study (11% incidence of rod fractures). In our series, only one patient 272
has required a revision procedure at 2 year follow-up. With reference to Figure 4, the increase in 273
mean major Cobb angle from 2 to 24 months for Subgroup 2 (rod fractures) was 21.4° to 26.9° 274
(5.5°), whereas for subgroup 1 (no mechanical complications) it was 20.5° to 24.2° (ie. 3.7°), 275
suggesting that rod fractures are associated with slightly more decompensation after surgery, 276
however we note that differences of a few degrees are neither clinically measurable nor 277
Radiographic outcomes over time after endoscopic scoliosis correction
12
significant in individual patients. When the rod fractures were identified early in the series, the 278
titanium rod diameter was increased from 4.5 to 5.5mm and the bone graft changed from 279
autograft (either rib head or iliac crest) to mulched femoral head allograft (typically 2 heads) 280
densely packed into well prepared disc spaces. Since these changes, there have been no further 281
rod fractures in the last sixty cases. Examination of Figures 4 to 7 shows that there were no 282
statistically significant increases in rib hump or in major, instrumented or compensatory Cobb 283
angles after the rod fractures occurred (ie between 12 and 24 months after surgery). 284
285
The other mechanical complication in our series was partial proximal screw pullout. These cases 286
were spread throughout the entire series and in most cases were noted in the first postoperative 287
(2 month) radiograph. Figures 4 to 6 suggest that although screw-related complications reduce 288
postoperative correction by a few degrees relative to patients without complications, the 289
subsequent decompensation from 2 to 24 months after surgery is no greater for screw-related 290
complications than for patients with no complications. None of the screw-related complications 291
have required revision surgery to date. 292
293
Endoscopic anterior instrumentation for adolescent idiopathic scoliosis is a safe and viable 294
surgical option. After the initial correction provided by the procedure there is a small loss of 295
coronal plane correction over time, but it is of subclinical magnitude and falls within the 296
accepted measurement error (5) for the Cobb technique. Fractured rods and partial proximal 297
screw pullout reduce correction by a few degrees relative to cases without complications, but do 298
not lead to clinically significant progression after the complication has occurred. In our patient 299
series, the incidence of rod fractures has been reduced to zero since the adoption of a pure 300
titanium 5.5mm rod and meticulous bone grafting technique utilizing allograft. 301
302
Radiographic outcomes over time after endoscopic scoliosis correction
13
Table 1. Radiographic and rib hump measures for the entire patient group. Shoulder balance is 303 displayed as a negative figure if the left shoulder is higher than the right, and a positive figure if 304 right shoulder is higher than the left. Coronal spinal balance is the distance of the midpoint of C7 305 verterbal body from the central sacral line and is a negative value if C7 is deviated to the left and 306 a positive figure if it is deviated to the right. All numbers are mean ± standard deviation. 307 308 309
310 311 312 313 314 Table 2. Changes between the 2 and 24 month reviews for all radiographic parameters (degrees) 315 and rib hump (degrees) displayed according to their percentile rankings for the entire cohort. A 316 negative value indicates a decrease and a positive value indicates an increase between the 2 and 317 24 month reviews. 318 319 320
Percentile Major Cobb Angle Instrumented Cobb Angle
Minor Cobb
T5-T12 Kyphosis
Rib Hump
Maximum 12 11 12 23 12 90th 10 8.5 5 12 5 75th 7 5 2 7.5 2
50th (median) 4 2 0 3 1 25th 1 0 -4 0 0 10th -1 -2 -6.7 -4 -1
Minimum -5 -7 -13 -11 -3 321 322
Variable Preop. 2 months 6 months 12 months 24 months Major Cobb (°) 51.8 ±8.7 21.1 ±7.2 22.9 ±7.5 24.1 ±8.1 25.1 ±8.4 Instrumented Cobb (°) as above 18.8 ±7.6 20.3 ±7.6 21.1 ±7.8 21.6 ±8.4 Distal compensatory Cobb (°) 31.9 ±10.6 18.3 ±8.9 18.3 ±9.1 17.8 ±9.6 18.1 ±10.4 T5-T12 kyphosis (°) 18.3 ±9.8 27.0 ±7.8 27.5 ±8.5 29.4 ±8.9 30.6 ±9.3 Rib hump(°) 16.5 ±4.4 6.4 ±3.1 7.3 ±3.2 7.5 ±3.5 7.8 ±3.4 Shoulder balance (cm) 0.1 ±1.3 -0.9 ±1.2 -0.6 ±0.9 -0.4 ±1.0 -0.2 ±0.9 Coronal spinal balance (cm) 0.05 ±1.4 -0.3 ±1.3 -0.4 ±1.3 -0.5 ±1.1 -0.4 ±0.9
Correction Rates (%) Preop. 2 months 6 months 12 months 24 months Major curve 59.5 55.8 53.5 51.8 Instrumented curve 63.9 61.0 59.6 58.8 Distal compensatory curve 42.7 43.1 44.7 45.4 Rib hump 60.9 54.4 53.9 51.7
Radiographic outcomes over time after endoscopic scoliosis correction
14
Figure Captions 323
324
Figure 1. Coronal plane radiograph of single rod anterior endoscopic construct showing the 325
difference between major and instrumented Cobb angles 326
Figure 2. Preoperative coronal, lateral and fulcrum bending radiographs and postoperative 327
coronal and lateral views for two representative patients from the series (patients 81 and 93) 328
Figure 3. Changes in radiograph parameters over time following surgery (preoperative, 2 329
months, 6 months, 12 months, 24 months). The radiographic parameters included are major 330
Cobb angle, instrumented Cobb angle, distal compensatory Cobb angle, rib hump, and T5-T12 331
kyphosis angle. Note that a bar on the graph has been included for preoperative instrumented 332
Cobb angle. The value of this bar is a repeat of the major Cobb angle. Error bars represent ±1 333
standard deviation. 334
Figure 4. Major Cobb angle versus time for (a) subgroup 1 - no mechanical complications, (b) 335
subgroup 2 – rod fractures, (c) subgroup 3 – screw-related complications. Error bars represent ±1 336
standard deviation. 337
Figure 5. Instrumented Cobb angle versus time for (a) subgroup 1 - no mechanical 338
complications, (b) subgroup 2 – rod fractures, (c) subgroup 3 – screw-related complications. 339
Error bars represent ±1 standard deviation. 340
Figure 6. Distal compensatory Cobb angle versus time for (a) subgroup 1 - no mechanical 341
complications, (b) subgroup 2 – rod fractures, (c) subgroup 3 – screw-related complications. 342
Error bars represent ±1 standard deviation. 343
Figure 7. Rib hump versus time for (a) subgroup 1 - no mechanical complications, (b) subgroup 344
2 – rod fractures, (c) subgroup 3 – screw-related complications. Error bars represent ±1 standard 345
deviation. 346
Radiographic outcomes over time after endoscopic scoliosis correction
15
Figure 8. T5-T12 kyphosis angle versus time for (a) subgroup 1 - no mechanical complications, 347
(b) subgroup 2 – rod fractures, (c) subgroup 3 – screw-related complications. Error bars 348
represent ±1 standard deviation. 349
350
351
352
References 353
1. Mack MJ, Regan JJ, Bobechko WP, et al. Application of thoracoscopy for diseases of the 354
spine. Ann Thorac Surg 1993; 56:736-8. 355
356
2. Rosenthal D, Rosenthal R, deSimone A. Removal of a protruded thoracic disk using 357
microsurgical endoscopy: a new technique. Spine 1994; 19:1087-91. 358
359
3. McAfee PC, Regan JR, Zdeblick T, et al. The incidence of complications in endoscopic 360
anterior thoracolumbar spinal reconstructive surgery: a prospective multicenter study comprising 361
the first 100 consecutive cases. Spine 1995; 20:1624-32. 362
363
4. Regan JJ, Mack MJ, Picette GD III. A technical report on video-assisted thoracoscopy in 364
thoracic spinal surgery. Preliminary description. Spine 1995; 20: 831-7. 365
366
5. Picetti GD III. Video-assisted thoracoscopy (VATS) in the treatment of congenital 367
hemivertebrae. Spine Society of Australia Annual Scientific Meeting, Cairns, Australia, 368
September 1996. 369
370
Radiographic outcomes over time after endoscopic scoliosis correction
16
6. Picetti GD III, Ertl JP, Bueff HU. Endoscopic instrumentation, correction, and fusion of 371
idiopathic scoliosis. Spine J 2001; 1:190-7. 372
373
7. Arlet V. Anterior thoracoscopic spine release in deformity surgery: a meta-analysis and 374
review. Eur Spine J 2000; 9:S17-23. 375
376
8. Newton PO, Marks M, Faro F, et al. Use of video assisted thoracoscopic surgery to reduce 377
perioperative morbidity in scoliosis surgery. Spine 2003; 28:S249-S254. 378
9. Lenke L. Anterior endoscopic discectomy and fusion for adolescent idiopathic scoliosis. 379
Spine 2003; 28:S36-43. 380
381
10. Al-Sayyad MJ, Crawford AH, Wolf RK. Early experiences with video-assisted 382
thoracoscopic surgery: our first 70 cases. Spine 2004; 29:1945-51. 383
384
11. Izatt MT, Harvey JR, Adam CJ, et al. Recovery in pulmonary function following endoscopic 385
anterior scoliosis correction: Evaluation at 3, 6, 12 and 24 months after surgery. Spine 2006; 386
31:2469-77. 387
388
12. Gatehouse SC, Izatt MT, Adam CJ, et al. Perioperative aspects of endoscopic anterior 389
scoliosis surgery: The learning curve for a consecutive series of 100 patients. J Spinal Disord 390
Tech 2007; 20:317-23. 391
392
13. Lonner BS, Kondrachov D, Siddiqi F, et al. Thoracoscopic spinal fusion compared with 393
posterior spinal fusion for the treatment of thoracic adolescent idiopathic scoliosis. J Bone Joint 394
Surg Am 2006; 88:1022-1034 395
Radiographic outcomes over time after endoscopic scoliosis correction
17
396
14. Newton PO, Marks MJ, Faro F, et al. Use of video-assisted thoracoscopic surgery to reduce 397
perioperative morbidity in scoliosis surgery. Spine 2003; 28:S249-54 398
399
15. Kishan S, Bastrom T, Betz RR, et al. Thoracoscopic scoliosis surgery affects pulmonary 400
function less than thoracotomy at 2 years post surgery. Spine 2007; 32:453-8. 401
402
16. Faro FD, Marks MC, Newton PO, et al. Perioperative changes in pulmonary function after 403
anterior scoliosis instrumentation: Thoracoscopic versus open approaches. Spine 2005; 30:1058-404
63. 405
406
17. Newton PO, Shea KG, Granlund KF. Defining the pediatric spinal thoracoscopy learning 407
curve: sixty five consecutive cases. Spine 2000; 25:1028-35. 408
409
18. Lonner BS, Scharf C, Antonacci D, et al. The learning curve associated with thoracoscopic 410
spinal instrumentation. Spine 2005; 30:2835-40. 411
412
19. Grewel H, Betz RR, D’Andrea LP, et al. A prospective comparison of thoracoscopic vs open 413
anterior instrumentation and spinal fusion for idiopathic thoracic scoliosis in children. J Pediatr 414
Surg. 2005; 40:153-156; discussion 156-157. 415
416
20. Newton PO, Parent S, Marks M, et al. Prospective evaluation of 50 consecutive scoliosis 417
patients surgically treated with thoracoscopic anterior instrumentation. Spine 2005; 30:S100-418
S109. 419
420
Radiographic outcomes over time after endoscopic scoliosis correction
18
21. Lescrève JP, Van Tiggelen RP, Lamoureux J. Reducing the radiation dosage in patients with 421
a scoliosis. Int Orthop 1989; 13:47-50. 422
423
22. Cobb JR. Outline for the study of scoliosis. In Edwards JW, ed. Instructional Course lectures. 424
Ann Arbor, MI: Am Acad Orthop Surg Lect 1948; 5:261-275. 425
426
23. Weinstein SL, Dolan LA, Cheng JCY, et al. Adolescent idiopathic scoliosis. Lancet 2008; 427
371:1527-37. 428
429
24. Lenke LG, Bridwell KH, Baldus C, et al. Cotrel-Debousset instrumentation for adolescent 430
idiopathic scoliosis. J Bone Joint Surg Am 1992; 74A:1056-1067. 431
432
25. Lenke LG, Bridwell KH, Blanke K, et al. Radiographic results of arthrodesis with Cotrel-433
Dubousset instrumentation for the treatment of adolescent idiopathic scoliosis. A five to ten-year 434
follow-up study. J Bone Joint Surg Am 1998; 80A:807-14. 435
436
26. Shufflebarger HL, King WF. Composite measurement of scoliosis: A new method of 437
anlaysis of the deformity. Spine 1987; 12:228-232. 438
439
27. Lowe T, Betz R, Lenke L, et al. Anterior single-rod instrumentation of the thoracic and 440
lumbar spine: saving levels. Spine 2003; 28:S208-S216. 441
442
28. Betz RR, Harms J, Clements DH III, et al. Comparison of anterior and posterior 443
instrumentation for correction of adolescent thoracic idiopathic scoliosis. Spine 1999; 24:225-444
239. 445
Radiographic outcomes over time after endoscopic scoliosis correction
19
446
29. Betz RR, Shufflebarger H. Controversies in spine. Anterior versus posterior instrumentation 447
for the correction of thoracic idiopathic scoliosis. Spine 2001; 26:1095-1100. 448
449
30. Crawford JR, Izatt MT, Adam CJ, Labrom RD, Askin GN. A prospective assessment of 450
SRS-24 scores after endoscopic anterior instrumentation for scoliosis. Spine 2006; 31:E817-22. 451
452
31. Hay D, Izatt MT, Adam CJ, Labrom RD, Askin GN. The use of fulcrum bending radiographs 453
in anterior thoracic scoliosis correction. Spine 2008; 33:999-1005. 454
455
32. Adam CJ, Izatt MT, Harvey JR et al. Variability in cobb angle measurements using 456
reformatted computerized tomography scans. Spine 2005; 30:1664-1669. 457
458
33. Beekman CE, Hall V. Variability of scoliosis measurement from spinal roentgenograms. 459
Phys Ther 1979; 59:764-5. 460
461
34. Oda M, Rauh S, Gregory P, et al. The significance of roentgenographic measurement in 462
scoliosis. J Pediatr Orthop 1982; 2:378–82. 463
464
35. Goldsmith G, Morrissy R, Hall CE, et al. Accuracy of the Cobb technique in 465
scoliosis measurement. Orthop Trans 1987; 11:32. 466
467
36 Carman DL, Browne RH, Birch JG. Measurement of scoliosis and kyphosis radiographs. J 468
Bone Joint Surg Am 1990; 72:328–33. 469
470
Radiographic outcomes over time after endoscopic scoliosis correction
20
37. Morrissy RT, Goldsmith GS, Hall EC, et al. Measurement of the Cobb Angle on radiographs 471
of patients who have scoliosis. J Bone Joint Surg Am 1990; 72:320–7. 472
473
38. Ylikoski M, Tallroth K. Measurement variations in scoliotic angle, vertebral rotation, 474
vertebral body height, and intervertebral disc space height. J Spinal Disord 1990; 3:387–91. 475
476
39. Dutton KE, Jones TJ, Slinger BS, et al. Reliability of the Cobb-angle index derived by 477
traditional and computer-assisted methods. Australas Phys Eng Sci Med 1989; 12:16–23. 478
479
40. Segel MH, Betz RR, Bridwell KH, et al. A critical analysis comparing Zielke 480
instrumentation alone, Cotrel-Dubousset instrumentation alone and a combination of the two in 481
the treatment of idiopathic thoracolumbar and lumbar idiopathic scoliosis. Orthop Trans 1992; 482
16:153-4. 483
484
485
486
Instrumented Cobb
MajorCobb
Figure 1.
#93
#81
Figure 2.
0
10
20
30
40
50
60
70
Pre-op 2 months 6 months 12 months 24 months
Ang
le (d
egre
es)
Major Cobb angleInstrumented Cobb angleDistal compensatory Cobb angleRib humpT5-T12 kyphosis angle
**
* *
*
*
* Denotes statistically significant difference (P
0
10
20
30
40
50
60
70
Pre-op 2 months 6 months 12 months 24 months
Ang
le (d
egre
es)
No mechanical complications
Screw related complications
Rod fractures* Denotes statistically significant difference (P
0
10
20
30
40
50
60
70
Pre-op 2 months 6 months 12 months 24 months
Ang
le (d
egre
es)
No mechanical complications
Screw related complications
Rod Fractures
**
***
†
* Denotes statistically significant difference (P
0
5
10
15
20
25
30
35
40
45
50
Pre-op 2 months 6 months 12 months 24 months
Ang
le (d
egre
es)
No mechanical complications
Screw related complications
Rod Fractures* Denotes statistically significant difference (P
0
5
10
15
20
25
Pre-op 2 months 6 months 12 months 24 months
Ang
le (d
egre
es)
No mechanical complications
Screw related complications
Rod fractures
** †*
*
* Denotes statistically significant difference (P
0
5
10
15
20
25
30
35
40
45
Pre-op 2 months 6 months 12 months 24 months
Ang
le (d
egre
es)
No mechanical complications
Screw related complications
Rod Fractures
* *
*
*
* Denotes statistically significant difference (P