- Email: [email protected]
Radiological Outcomes in Adolescent Idiopathic Scoliosis Patients More Than 22 Years After Treatment
Spine Deformity 3 (2015) 436e439 www.spine-deformity.org
Long-Term Follow-Up Case Series
Radiological Outcomes in Adolescent Idiopathic Scoliosis Patients More Than 22 Years After Treatment Ane Simony, MDa,*, Steen Bach Christensen, DrMeda,b, Leah Y. Carreon, MD, MSca, Mikkel Osterheden Andersen, MDa b
a Center for Spine Research and Surgery, Lillebaelt Hospital, Ostre Hougvej 55, DK-5500 Middelfart, Denmark Department of Orthopaedic Surgery, University Hospital e Rigshospitalet, Blegdamsvej 9, DK-2100 Copenhagen, Denmark Received 27 November 2014; revised 25 February 2015; accepted 2 March 2015
Abstract Study Design: Longitudinal cohort. Objectives: To evaluate the long-term radiologic outcomes in adolescent idiopathic scoliosis (AIS) patients more than 22 years after treatment. Summary of Background Data: Although treatment for AIS is prophylactic and is aimed at preventing curve progression, very few studies report long-term outcomes of treatment. Methods: AIS patients treated with Boston brace or posterior spinal fusion (PSF) with Harringtonedorso-transverse traction (DTT) instrumentation from 1983 to 1990 were requested to return to clinic. Subsequently, 36-inch standing radiographs were obtained after patient consent. Cobb angles were compared with pretreatment and immediate posttreatment radiographs. Any evidence of adjacent-level disease or local kyphosis was also noted. Results: One hundred fifty-nine (78%) of 219 patients were available for follow-up, 66 braced and 93 surgical. There were 85 females with an average age at surgery of 14.3 years and an average age at follow-up of 37.6 years. The mean length of follow-up was 24.5 years (range, 22e30 years). There was a statistically significant curve progression of 2.9 in the PSF group. There was a greater degree of curve progression in the braced group (5.5 ), but this was not statistically significant. Proximal segment degeneration was seen in 8 (5%), 2 in the brace cohort and 6 in the PSF cohort. Distal segment degeneration was seen in 26 (16%) patients, 4 treated with brace and 22 treated with PSF. No patient developed proximal junction kyphosis. Three patients in the PSF cohort required additional surgery for distal adding-on. Four patients had a noncontiguous L5eS1 fusion, three from the PSF cohort and one from the braced cohort. Conclusion: In this cohort with an average follow-up of 24.5 years, with 78% available for follow-up, both the braced and surgically treated patients had a very small degree of curve progression, with a small incidence of distal segment degeneration and reoperation. Level of Evidence: III. Ó 2015 Scoliosis Research Society. Keywords: Long-term follow-up; Adolescent idiopathic scoliosis; Brace treatment; Surgical fusion; Harrington-DTT instrumentation
Introduction Scoliosis affects 1% to 2% of the population, with adolescent idiopathic scoliosis (AIS) the most common subgroup [1,2]. It affects teenagers, mostly females, and is diagnosed from age 11 to 14 years,
where growth in the spine is rapid. Untreated, AIS can progress to severe deformation of the spine, pulmonary impairment, back pain, and poor selfimage [3].
This study was approved by the Ethical Committee of Southern Denmark, Datatilsynet and was conducted at the Section for Spine Surgery and Research, Lillebaelt Hospital, Ostre Hougvej 55, 5000 Middelfart, Denmark. This study was funded by the Danish Rheumatism Association, Fonden af 17-12-1981, Erik Birger Christensen Fund and AP Møllers Fund. *Corresponding author. Sector for Spine Research and Surgery, Sygehus Lillebaelt, Ostre Hougvej 55, DK-5500 Middelfart, Denmark. Tel.: þ45 24 45 37 42; fax: þ 45 63 48 42 81. E-mail address: [email protected] (A. Simony). 2212-134X/$ - see front matter Ó 2015 Scoliosis Research Society. http://dx.doi.org/10.1016/j.jspd.2015.03.003
A. Simony et al. / Spine Deformity 3 (2015) 436e439
Brace treatment [4] and surgical correction [5] are the only two treatments that have been shown to alter the natural history and progression of AIS. At Copenhagen University Hospital, brace treatment with Boston brace was initiated in curves within 30 to 50 with documented curve progression and remaining skeletal growth. The patients were instructed to wear the brace for 23 hours a day and were examined twice a year including posteroanterior x-ray by an orthopedic surgeon. If the curve had not progressed on 2 successive visits and the patient was judged skeletally mature based on the Risser sign [6,7], brace weaning over a 6-month period was initiated. In patients where the Cobb angle exceeded 50 , surgical correction, fusion with Harringtonedorso-transverse traction (DTT) instrumentation was recommended, which was the standard of care at that time. Postoperatively, the patients were placed in a cast for a 12-month period. To our knowledge, only a few studies have been published with more than 20 years of follow-up after AIS treatment [3,5,8-12]. The purpose of this study was to evaluate the long-term radiologic outcomes, specifically curve progression and adjacent level degeneration, in AIS patients at a mean 25 years after treatment. Methods Patients diagnosed and treated with either a Boston brace or posterior spinal fusion (PSF) with Harrington-DTT instrumentation for AIS at Copenhagen University Hospital, Denmark, from 1983 to 1990 were identified. These patients were contacted and requested to return to the clinic for an interview and clinical examination, including assessment with health-related quality of life questionnaires. Two hundred nineteen patients were identified, among whom 104 were braced and 115 had a PSF. Of the 115 who had surgery, 36 were braced initially, but went to surgery because of progression during brace treatment. Although the original radiographs from pretreatment, postoperation, and termination of treatment were not available, medical charts including radiographic reports from before treatment to termination of treatment were available for review. Cobb measurements and end vertebrae were clearly documented in the medical chart by both the radiologist and the spine surgeon, and the original data set with curve classification, and Cobb measurement including identification of the measured levels, performed by the authors SBC and MOAwas available. For the current study, full-length 36-inch anterioposterior and lateral standing radiographs were obtained after patient consent. Cobb angles were measured and compared with the patient’s pretreatment and immediate posttreatment radiographs. Radiographs were also evaluated for any evidence of adjacent level disease or local kyphosis. All statistical analysis was done using Stata, version 13 (Metrika Consulting AB, Stockholm, Sweden) by an independent statistician. Comparisons between pretreatment, immediate posttreatment, and latest Cobb measurements
437
were done using Wilcoxon signed rank test as the data were not normally distributed. Results One hundred fifty-nine patients participated in the radiologic follow-up (78%), 66 treated with Boston brace and 93 treated with PSF. Eleven patients refused x-ray examination because they were pregnant at the time of examination or because they had a history of breast or other cancers, and were thereby excluded from this study. Eight patients had emigrated out of Denmark and were not available for follow-up, two are deceased because of cardiac arrest and breast cancer and four patients were diagnosed with a neurologic disease after the time of AIS treatment and were excluded. There were 35 nonresponders. Among the PSF patients, there were 85 females and 8 males, with an average age at surgery of 14.3 years and an average age at follow-up of 37.6 years. The mean length of follow-up was 24.5 years, with a range of 22e30 years (Table 1). Among the braced patients, there were 62 females and 4 males. The mean length of follow-up was 26.5 years with a range of 22 to 31 years. The mean Cobb angles in the braced and PSF patients prior to treatment, immediately after treatment, and at long-term follow-up are presented in Table 2. There was no clinically relevant curve progression in the PSF group. There was a small, but not clinically relevant, curve progression of 5.5 in the braced group, p value .739, but this was not statistically significant. In the brace-treated group, nine patients had a curve progression O5 during the 25-year follow-up period. Seven patients classified with Curve Type 2 or 3, with apex of the deformity at the seventh, or eighth, thoracic vertebra. The remaining two patients progressed severely during the brace treatment, but the parents refused to allow surgery, and the deformity at brace termination was 54 and 63 . The patient terminated at 54 had a Curve Type 2, and after 23 years she has progressed to 67 . The patient terminated at 63 , with a Curve Type 1, has remained stable after 24 years. None of the patients are complaining of back pain or taking analgesics, and both are working full-time. Other radiographic findings include proximal segment degeneration in 8 (5%), 2 in the brace cohort, and 6 in the PSF group. Distal segment degeneration was seen in 26 (16%) patients, 4 treated with brace and 22 treated with PSF. Radiographic evidence of degeneration was defined as Table 1 Summary of demographic characteristics.
N Female Age at end of treatment (years) Age at follow-up (years)
Harrington-DTT instrumentation
Boston brace
Total
93 85 14.3 (12e18)
66 62 15.8 (14e19)
165 147
37.6 (33e47)
41.4 (34e47)
DTT, dorso-transverse traction.
438
A. Simony et al. / Spine Deformity 3 (2015) 436e439
Table 2 Mean Cobb angle of the major curve at each time period.
N Pretreatment End of treatment Follow-up Progressiona p valueb
Tabel 5 Kyphosis and lordosis, measured in Cobb angles.
Harrington-DTT instrumentation
Boston brace
93 54.5 (50.4 e58.8 ) 29.5 (25.7 e33.9 ) 32.4 (27.9 e39.5 ) 2.9 (0 e9.8 ) !.010
66 37.5 34.7 40.2 5.5 .739
(35.1 e40.0 ) (31.9 e37.5 ) (36.7 e43.6 ) (0 e16 )a
DTT, dorso-transverse traction. Progression represents the difference between the Cobb angle at the end of treatment and the Cobb angle at the latest follow-up. b p value is from the Wilcoxon signed rank test comparing the Cobb angle at the end of treatment and the Cobb angle at the latest follow-up. a
decreased disc space height and/or presence of osteophytes in the distal adjacent level. No patient had developed proximal junctional kyphosis (PJK), defined as proximal sagittal Cobb angle O10 or proximal sagittal Cobb angle of at least 10 greater than the preoperative measurement as defined by Glattes [13] (Table 3). More than 70% of the surgeries in the PSF group were performed by a single surgeon (SBC). The lowest instrumented vertebra (LIV) was primarily at L1 in 36 patients and L2 in 44 patients (see Table 4). Kyphosis and lordosis were measured and compared with the preoperative measurements from the original medical charts (see Table 5). The sagittal profile of the patients remained unchanged, from pretreatment to the 25-year follow-up. Three patients in the PSF cohort required additional surgery as a result of pseudoarthrosis, and distal adding with posterior fusion was performed within the first year after index procedure. Two patients had an LIV at L1 and one patient had LIV at L2. Table 3 Summary of radiographic findings and additional surgical intervention. Harrington-DTT Boston Total instrumentation brace N Proximal segment degenerationa Proximal junctional kyphosis Distal segment degenerationa Required fusion L5-S1 noncontiguous fusion
93 6 (6%) 0 22 (24%) 4 (4%) 3 (3%)
66 2 (3%) 0 4 (6%) 0 1 (2%)
8 (5%) 0 26 (16%) 4 (4%) 4 (4%)
DTT, dorso-transverse traction. Radiographic degeneration defined as decreased disc space height and/or presence of osteophytes. a
Table 4 Level of instrumented vertebrae (LIV) and Curve type.a Type 1 Th 11 Th 12 L1 L2 L3 a
Type 2
Type 3
Type 4
Type 5
11 3
8
1
2
3 9 11 3
26 13 1
Preoperative classification according to King-Moe.
Pre-operative kyphosis Kyphosis at 25-year follow-up Progressiona Pre-operative lordosis Lordosis at 25-year follow-up Progressiona
20.9o (18.2oe23.6o) 23.5o (21.5oe 25.6o) 2.6o, p 5 .028 42.4o (39.9oe44.9o) 38.9 (36.8e40.9) 3.5o, p 5 .063
a
Progression represents the change from pre-operative x-rays to latest follow-up.
Four patients in the PSF group progressed O5 in the 25year follow-up period, two had implant failures, with rod breakage and dislocation of the distal hook, one was reoperated 1 year after index surgery because of pseudoarthrosis, and one patient had degeneration distal to the instrumentation, causing a lateral shift of the third lumbar vertebra. Four patients had a noncontiguous L5eS1 fusion, all performed 7 to 10 years after the index procedure, three from the PSF cohort and one from the braced cohort.
Discussion Although treatment for AIS is prophylactic, aimed at the prevention of curve progression and its associated consequences such as impaired pulmonary function and lumbar degeneration, there are very few studies looking at the longterm outcomes of AIS treatment [3,5,8-12]. In our cohort of AIS patients treated with a Boston brace, there was no clinically relevant progression of the curve from prebracing to completion of bracing. This is to be expected, as the goal of bracing is prevention of curve progression and not correction. There was a small progression of 5.5 from the completion of treatment to the latest follow-up of mean 25 years, which was not statistically significant or clinically relevant. Our results are comparable to the results reported by Danielsson and Nachemson [5]. Nine patients had progressed more than 5 , all with apex of the deformity above the thoracolumbar junction. The Boston brace is well known for its ability to correct and maintain curves, but because of the design of the brace, it is more effective in correcting curves with apex at the thoracolumbar junction. The incidence of distal segment degeneration in our study was 6%. These data show that AIS patients treated in a brace maintain their correction and are not at a higher risk of disc degeneration. Similarly, patients treated surgically maintained their correction at long-term follow-up. Although there was a small statistical increase in curve magnitude, the increase of 2.9 is not clinically significant because of the inaccuracy in Cobb measurements [14]. Even though 24% of the surgically treated patients have evidence of distal degeneration on radiographs, only 3% required distal spinal fusion. Four patients progressed in the PSF group during the follow-up period, two due to loss of fixation, one due to
A. Simony et al. / Spine Deformity 3 (2015) 436e439
pseudoarthrosis, and one due to distal degeneration. None of the patients had back pain or progression of the deformity, and all the curves were less than 40 after 25 years. The reoperation rate in this series is less than previously reported [8,10-12]. This may be explained by the surgeon’s conscious practice to attempt to contour the rods, fixate the apex with DTT instrumentation, and not to instrument below L3. In the handful of patients fused to L3, none have had distal segment degeneration requiring surgical intervention. In addition, the primary goal of surgery was to prevent curve progression and achieve a solid fusion, rather than achieving a maximal correction. The mean difference between the preoperative and postoperative Cobb angle was 25 , representing 45% correction [15]. This percentage of correction is less than that reported for modern surgical techniques. This then opens the discussion of whether maximal correction is necessary to prevent problems associated with spine deformity in the long term. In contrast to the radiographic outcomes in patients with adult spinal deformity, where PJK is a substantial problem, no patient in this cohort developed PJK at long-term follow-up. This could be explained by the fact that the patients only had a small correction of their hypokyphotic thoracic spine after surgery and have maintained their original sagittal profile. As with any retrospective study design, there are limitations to this study. The pretreatment films were not available. However, medical and radiographic reports with clear descriptions of the coronal curve, and the original data set with curve classification, apex, kyphosis, and lordosis measurements were available. Unfortunately, a thorough description of the deformity in the sagittal plane including the pelvic parameters was not available. The importance of the sagittal plane in the surgical decision making and treatment of AIS was not well elucidated at that time. Most importantly, the surgical technique of Harrington rod-DTT Instrumentation with correction only in the coronal plane is no longer the standard of care. However, even with this less sophisticated technique, the loss of correction, incidence of PJK, distal segment degeneration, and reoperations were few. It would be difficult to determine if the availability of hooks, pedicle screws, and rods to allow correction of the deformity in three planes will have the same outcomes. Conclusion In this cohort with an average follow-up of 25 years, with 78% available for follow-up, both the braced and
439
surgically treated patients had a very small degree of curve progression, with a small incidence of PJK, distal segment degeneration, and reoperation. Although the criterion standard for AIS treatment has evolved in the past 25 years, it is our experience that AIS patients treated with either Boston brace or Harrington-DTT instrumentation will preserve the correction achieved for at least 25 years. References [1] Lonstein JE, Bjorklund S, Wanninger MH, Nelson RP. Voluntary school screening for scoliosis in Minnesota. J Bone Joint Surg Am 1982;64:481e8. [2] Andersen MO, Kyvik K, Thomsen K. Adolescent idiopathic scoliosis in twins, a population based survey. Spine 2010;35:1571e4. [3] Danielsson AJ. Natural history of adolescent idiopathic scoliosis: a tool for guidance in decision of surgery of curves above 50 . J Child Orthop 2013;7:37e41. [4] Weinstein SL, Dolan LA, Wright JG, et al. Effect of bracing in adolescent with idiopathic scoliosis. N Engl J Med 2013;369: 1512e21. [5] Danielsson AJ, Nachemson AL. Back pain and function 22 years after brace treatment for adolescent idiopathic scoliosis: a casecontrol studydpart I. Spine (Phila Pa 1976) 2003;28:2078e85. [6] Risser JC. The iliac apophysis; an invaluable sign in the management of scoliosis. Clin Orthop 1958;11:111e9. [7] Reem J, Carney J, Stanley M, Cassidy J. Risser sign inter-rater and intra-rater agreement: is the Risser sign reliable? Skeletal Radiol 2009;38:371e5. [8] Mariconda M, Galasso O, Barca P, et al. Minimum 20-year follow-up results of Harrington rod fusion for idiopathic scoliosis. Eur Spine J 2005;14:854e61. [9] Niemeyer T, B€ovingloh AS, Grieb S, et al. Low back pain after spinal fusion and Harrington instrumentation for idiopathic scoliosis. Int J Orthop 2005;29:47e50. [10] Helenius I, Remes V, Yrj€onen T, et al. Comparison of long-term function and radiological outcomes after Harrington instrumentation and spondylodesis in adolescent idiopathic scoliosis: a review of 78 patients. Spine 2002;15(27):176e80. [11] Paonessa KJ, Engler GL. Back pain and disability after Harrington rod fusion to the lumbar spine for scoliosis. Spine (Phila Pa 1976) 1992;17(8 suppl):S249e53. [12] Danielsson AJ, Nachemson AL. Back pain and function 23 years after fusion for adolescent idiopathic scoliosis: a case-control studye part II. Spine (Phila Pa 1976) 2003;28:E373e83. [13] Glattes RC, Bridwell KH, Lenke LG, et al. Proximal junctional kyphosis in adult spinal deformity following long instrumented posterior spinal fusion: incidence, outcomes, and risk factor analysis. Spine (Phila Pa 1976) 2005;30:1643e9. [14] Kuklo TR, Potter BK, Schroeder TM, et al. Comparison of manual and digital measurements in adolescent idiopathic scoliosis. Spine (Phila Pa 1976) 2006;31:1240e6. [15] Andersen GR, Andersen MO, Christensen SB. Selection of fusion levels in idiopathic scoliosis treated by Harrington-DTT instrumentation: a short term radiologic study. J Pediatr Orthop B 1995;4:86e90.
Long-Term Follow-Up Case Series
Radiological Outcomes in Adolescent Idiopathic Scoliosis Patients More Than 22 Years After Treatment Ane Simony, MDa,*, Steen Bach Christensen, DrMeda,b, Leah Y. Carreon, MD, MSca, Mikkel Osterheden Andersen, MDa b
a Center for Spine Research and Surgery, Lillebaelt Hospital, Ostre Hougvej 55, DK-5500 Middelfart, Denmark Department of Orthopaedic Surgery, University Hospital e Rigshospitalet, Blegdamsvej 9, DK-2100 Copenhagen, Denmark Received 27 November 2014; revised 25 February 2015; accepted 2 March 2015
Abstract Study Design: Longitudinal cohort. Objectives: To evaluate the long-term radiologic outcomes in adolescent idiopathic scoliosis (AIS) patients more than 22 years after treatment. Summary of Background Data: Although treatment for AIS is prophylactic and is aimed at preventing curve progression, very few studies report long-term outcomes of treatment. Methods: AIS patients treated with Boston brace or posterior spinal fusion (PSF) with Harringtonedorso-transverse traction (DTT) instrumentation from 1983 to 1990 were requested to return to clinic. Subsequently, 36-inch standing radiographs were obtained after patient consent. Cobb angles were compared with pretreatment and immediate posttreatment radiographs. Any evidence of adjacent-level disease or local kyphosis was also noted. Results: One hundred fifty-nine (78%) of 219 patients were available for follow-up, 66 braced and 93 surgical. There were 85 females with an average age at surgery of 14.3 years and an average age at follow-up of 37.6 years. The mean length of follow-up was 24.5 years (range, 22e30 years). There was a statistically significant curve progression of 2.9 in the PSF group. There was a greater degree of curve progression in the braced group (5.5 ), but this was not statistically significant. Proximal segment degeneration was seen in 8 (5%), 2 in the brace cohort and 6 in the PSF cohort. Distal segment degeneration was seen in 26 (16%) patients, 4 treated with brace and 22 treated with PSF. No patient developed proximal junction kyphosis. Three patients in the PSF cohort required additional surgery for distal adding-on. Four patients had a noncontiguous L5eS1 fusion, three from the PSF cohort and one from the braced cohort. Conclusion: In this cohort with an average follow-up of 24.5 years, with 78% available for follow-up, both the braced and surgically treated patients had a very small degree of curve progression, with a small incidence of distal segment degeneration and reoperation. Level of Evidence: III. Ó 2015 Scoliosis Research Society. Keywords: Long-term follow-up; Adolescent idiopathic scoliosis; Brace treatment; Surgical fusion; Harrington-DTT instrumentation
Introduction Scoliosis affects 1% to 2% of the population, with adolescent idiopathic scoliosis (AIS) the most common subgroup [1,2]. It affects teenagers, mostly females, and is diagnosed from age 11 to 14 years,
where growth in the spine is rapid. Untreated, AIS can progress to severe deformation of the spine, pulmonary impairment, back pain, and poor selfimage [3].
This study was approved by the Ethical Committee of Southern Denmark, Datatilsynet and was conducted at the Section for Spine Surgery and Research, Lillebaelt Hospital, Ostre Hougvej 55, 5000 Middelfart, Denmark. This study was funded by the Danish Rheumatism Association, Fonden af 17-12-1981, Erik Birger Christensen Fund and AP Møllers Fund. *Corresponding author. Sector for Spine Research and Surgery, Sygehus Lillebaelt, Ostre Hougvej 55, DK-5500 Middelfart, Denmark. Tel.: þ45 24 45 37 42; fax: þ 45 63 48 42 81. E-mail address: [email protected] (A. Simony). 2212-134X/$ - see front matter Ó 2015 Scoliosis Research Society. http://dx.doi.org/10.1016/j.jspd.2015.03.003
A. Simony et al. / Spine Deformity 3 (2015) 436e439
Brace treatment [4] and surgical correction [5] are the only two treatments that have been shown to alter the natural history and progression of AIS. At Copenhagen University Hospital, brace treatment with Boston brace was initiated in curves within 30 to 50 with documented curve progression and remaining skeletal growth. The patients were instructed to wear the brace for 23 hours a day and were examined twice a year including posteroanterior x-ray by an orthopedic surgeon. If the curve had not progressed on 2 successive visits and the patient was judged skeletally mature based on the Risser sign [6,7], brace weaning over a 6-month period was initiated. In patients where the Cobb angle exceeded 50 , surgical correction, fusion with Harringtonedorso-transverse traction (DTT) instrumentation was recommended, which was the standard of care at that time. Postoperatively, the patients were placed in a cast for a 12-month period. To our knowledge, only a few studies have been published with more than 20 years of follow-up after AIS treatment [3,5,8-12]. The purpose of this study was to evaluate the long-term radiologic outcomes, specifically curve progression and adjacent level degeneration, in AIS patients at a mean 25 years after treatment. Methods Patients diagnosed and treated with either a Boston brace or posterior spinal fusion (PSF) with Harrington-DTT instrumentation for AIS at Copenhagen University Hospital, Denmark, from 1983 to 1990 were identified. These patients were contacted and requested to return to the clinic for an interview and clinical examination, including assessment with health-related quality of life questionnaires. Two hundred nineteen patients were identified, among whom 104 were braced and 115 had a PSF. Of the 115 who had surgery, 36 were braced initially, but went to surgery because of progression during brace treatment. Although the original radiographs from pretreatment, postoperation, and termination of treatment were not available, medical charts including radiographic reports from before treatment to termination of treatment were available for review. Cobb measurements and end vertebrae were clearly documented in the medical chart by both the radiologist and the spine surgeon, and the original data set with curve classification, and Cobb measurement including identification of the measured levels, performed by the authors SBC and MOAwas available. For the current study, full-length 36-inch anterioposterior and lateral standing radiographs were obtained after patient consent. Cobb angles were measured and compared with the patient’s pretreatment and immediate posttreatment radiographs. Radiographs were also evaluated for any evidence of adjacent level disease or local kyphosis. All statistical analysis was done using Stata, version 13 (Metrika Consulting AB, Stockholm, Sweden) by an independent statistician. Comparisons between pretreatment, immediate posttreatment, and latest Cobb measurements
437
were done using Wilcoxon signed rank test as the data were not normally distributed. Results One hundred fifty-nine patients participated in the radiologic follow-up (78%), 66 treated with Boston brace and 93 treated with PSF. Eleven patients refused x-ray examination because they were pregnant at the time of examination or because they had a history of breast or other cancers, and were thereby excluded from this study. Eight patients had emigrated out of Denmark and were not available for follow-up, two are deceased because of cardiac arrest and breast cancer and four patients were diagnosed with a neurologic disease after the time of AIS treatment and were excluded. There were 35 nonresponders. Among the PSF patients, there were 85 females and 8 males, with an average age at surgery of 14.3 years and an average age at follow-up of 37.6 years. The mean length of follow-up was 24.5 years, with a range of 22e30 years (Table 1). Among the braced patients, there were 62 females and 4 males. The mean length of follow-up was 26.5 years with a range of 22 to 31 years. The mean Cobb angles in the braced and PSF patients prior to treatment, immediately after treatment, and at long-term follow-up are presented in Table 2. There was no clinically relevant curve progression in the PSF group. There was a small, but not clinically relevant, curve progression of 5.5 in the braced group, p value .739, but this was not statistically significant. In the brace-treated group, nine patients had a curve progression O5 during the 25-year follow-up period. Seven patients classified with Curve Type 2 or 3, with apex of the deformity at the seventh, or eighth, thoracic vertebra. The remaining two patients progressed severely during the brace treatment, but the parents refused to allow surgery, and the deformity at brace termination was 54 and 63 . The patient terminated at 54 had a Curve Type 2, and after 23 years she has progressed to 67 . The patient terminated at 63 , with a Curve Type 1, has remained stable after 24 years. None of the patients are complaining of back pain or taking analgesics, and both are working full-time. Other radiographic findings include proximal segment degeneration in 8 (5%), 2 in the brace cohort, and 6 in the PSF group. Distal segment degeneration was seen in 26 (16%) patients, 4 treated with brace and 22 treated with PSF. Radiographic evidence of degeneration was defined as Table 1 Summary of demographic characteristics.
N Female Age at end of treatment (years) Age at follow-up (years)
Harrington-DTT instrumentation
Boston brace
Total
93 85 14.3 (12e18)
66 62 15.8 (14e19)
165 147
37.6 (33e47)
41.4 (34e47)
DTT, dorso-transverse traction.
438
A. Simony et al. / Spine Deformity 3 (2015) 436e439
Table 2 Mean Cobb angle of the major curve at each time period.
N Pretreatment End of treatment Follow-up Progressiona p valueb
Tabel 5 Kyphosis and lordosis, measured in Cobb angles.
Harrington-DTT instrumentation
Boston brace
93 54.5 (50.4 e58.8 ) 29.5 (25.7 e33.9 ) 32.4 (27.9 e39.5 ) 2.9 (0 e9.8 ) !.010
66 37.5 34.7 40.2 5.5 .739
(35.1 e40.0 ) (31.9 e37.5 ) (36.7 e43.6 ) (0 e16 )a
DTT, dorso-transverse traction. Progression represents the difference between the Cobb angle at the end of treatment and the Cobb angle at the latest follow-up. b p value is from the Wilcoxon signed rank test comparing the Cobb angle at the end of treatment and the Cobb angle at the latest follow-up. a
decreased disc space height and/or presence of osteophytes in the distal adjacent level. No patient had developed proximal junctional kyphosis (PJK), defined as proximal sagittal Cobb angle O10 or proximal sagittal Cobb angle of at least 10 greater than the preoperative measurement as defined by Glattes [13] (Table 3). More than 70% of the surgeries in the PSF group were performed by a single surgeon (SBC). The lowest instrumented vertebra (LIV) was primarily at L1 in 36 patients and L2 in 44 patients (see Table 4). Kyphosis and lordosis were measured and compared with the preoperative measurements from the original medical charts (see Table 5). The sagittal profile of the patients remained unchanged, from pretreatment to the 25-year follow-up. Three patients in the PSF cohort required additional surgery as a result of pseudoarthrosis, and distal adding with posterior fusion was performed within the first year after index procedure. Two patients had an LIV at L1 and one patient had LIV at L2. Table 3 Summary of radiographic findings and additional surgical intervention. Harrington-DTT Boston Total instrumentation brace N Proximal segment degenerationa Proximal junctional kyphosis Distal segment degenerationa Required fusion L5-S1 noncontiguous fusion
93 6 (6%) 0 22 (24%) 4 (4%) 3 (3%)
66 2 (3%) 0 4 (6%) 0 1 (2%)
8 (5%) 0 26 (16%) 4 (4%) 4 (4%)
DTT, dorso-transverse traction. Radiographic degeneration defined as decreased disc space height and/or presence of osteophytes. a
Table 4 Level of instrumented vertebrae (LIV) and Curve type.a Type 1 Th 11 Th 12 L1 L2 L3 a
Type 2
Type 3
Type 4
Type 5
11 3
8
1
2
3 9 11 3
26 13 1
Preoperative classification according to King-Moe.
Pre-operative kyphosis Kyphosis at 25-year follow-up Progressiona Pre-operative lordosis Lordosis at 25-year follow-up Progressiona
20.9o (18.2oe23.6o) 23.5o (21.5oe 25.6o) 2.6o, p 5 .028 42.4o (39.9oe44.9o) 38.9 (36.8e40.9) 3.5o, p 5 .063
a
Progression represents the change from pre-operative x-rays to latest follow-up.
Four patients in the PSF group progressed O5 in the 25year follow-up period, two had implant failures, with rod breakage and dislocation of the distal hook, one was reoperated 1 year after index surgery because of pseudoarthrosis, and one patient had degeneration distal to the instrumentation, causing a lateral shift of the third lumbar vertebra. Four patients had a noncontiguous L5eS1 fusion, all performed 7 to 10 years after the index procedure, three from the PSF cohort and one from the braced cohort.
Discussion Although treatment for AIS is prophylactic, aimed at the prevention of curve progression and its associated consequences such as impaired pulmonary function and lumbar degeneration, there are very few studies looking at the longterm outcomes of AIS treatment [3,5,8-12]. In our cohort of AIS patients treated with a Boston brace, there was no clinically relevant progression of the curve from prebracing to completion of bracing. This is to be expected, as the goal of bracing is prevention of curve progression and not correction. There was a small progression of 5.5 from the completion of treatment to the latest follow-up of mean 25 years, which was not statistically significant or clinically relevant. Our results are comparable to the results reported by Danielsson and Nachemson [5]. Nine patients had progressed more than 5 , all with apex of the deformity above the thoracolumbar junction. The Boston brace is well known for its ability to correct and maintain curves, but because of the design of the brace, it is more effective in correcting curves with apex at the thoracolumbar junction. The incidence of distal segment degeneration in our study was 6%. These data show that AIS patients treated in a brace maintain their correction and are not at a higher risk of disc degeneration. Similarly, patients treated surgically maintained their correction at long-term follow-up. Although there was a small statistical increase in curve magnitude, the increase of 2.9 is not clinically significant because of the inaccuracy in Cobb measurements [14]. Even though 24% of the surgically treated patients have evidence of distal degeneration on radiographs, only 3% required distal spinal fusion. Four patients progressed in the PSF group during the follow-up period, two due to loss of fixation, one due to
A. Simony et al. / Spine Deformity 3 (2015) 436e439
pseudoarthrosis, and one due to distal degeneration. None of the patients had back pain or progression of the deformity, and all the curves were less than 40 after 25 years. The reoperation rate in this series is less than previously reported [8,10-12]. This may be explained by the surgeon’s conscious practice to attempt to contour the rods, fixate the apex with DTT instrumentation, and not to instrument below L3. In the handful of patients fused to L3, none have had distal segment degeneration requiring surgical intervention. In addition, the primary goal of surgery was to prevent curve progression and achieve a solid fusion, rather than achieving a maximal correction. The mean difference between the preoperative and postoperative Cobb angle was 25 , representing 45% correction [15]. This percentage of correction is less than that reported for modern surgical techniques. This then opens the discussion of whether maximal correction is necessary to prevent problems associated with spine deformity in the long term. In contrast to the radiographic outcomes in patients with adult spinal deformity, where PJK is a substantial problem, no patient in this cohort developed PJK at long-term follow-up. This could be explained by the fact that the patients only had a small correction of their hypokyphotic thoracic spine after surgery and have maintained their original sagittal profile. As with any retrospective study design, there are limitations to this study. The pretreatment films were not available. However, medical and radiographic reports with clear descriptions of the coronal curve, and the original data set with curve classification, apex, kyphosis, and lordosis measurements were available. Unfortunately, a thorough description of the deformity in the sagittal plane including the pelvic parameters was not available. The importance of the sagittal plane in the surgical decision making and treatment of AIS was not well elucidated at that time. Most importantly, the surgical technique of Harrington rod-DTT Instrumentation with correction only in the coronal plane is no longer the standard of care. However, even with this less sophisticated technique, the loss of correction, incidence of PJK, distal segment degeneration, and reoperations were few. It would be difficult to determine if the availability of hooks, pedicle screws, and rods to allow correction of the deformity in three planes will have the same outcomes. Conclusion In this cohort with an average follow-up of 25 years, with 78% available for follow-up, both the braced and
439
surgically treated patients had a very small degree of curve progression, with a small incidence of PJK, distal segment degeneration, and reoperation. Although the criterion standard for AIS treatment has evolved in the past 25 years, it is our experience that AIS patients treated with either Boston brace or Harrington-DTT instrumentation will preserve the correction achieved for at least 25 years. References [1] Lonstein JE, Bjorklund S, Wanninger MH, Nelson RP. Voluntary school screening for scoliosis in Minnesota. J Bone Joint Surg Am 1982;64:481e8. [2] Andersen MO, Kyvik K, Thomsen K. Adolescent idiopathic scoliosis in twins, a population based survey. Spine 2010;35:1571e4. [3] Danielsson AJ. Natural history of adolescent idiopathic scoliosis: a tool for guidance in decision of surgery of curves above 50 . J Child Orthop 2013;7:37e41. [4] Weinstein SL, Dolan LA, Wright JG, et al. Effect of bracing in adolescent with idiopathic scoliosis. N Engl J Med 2013;369: 1512e21. [5] Danielsson AJ, Nachemson AL. Back pain and function 22 years after brace treatment for adolescent idiopathic scoliosis: a casecontrol studydpart I. Spine (Phila Pa 1976) 2003;28:2078e85. [6] Risser JC. The iliac apophysis; an invaluable sign in the management of scoliosis. Clin Orthop 1958;11:111e9. [7] Reem J, Carney J, Stanley M, Cassidy J. Risser sign inter-rater and intra-rater agreement: is the Risser sign reliable? Skeletal Radiol 2009;38:371e5. [8] Mariconda M, Galasso O, Barca P, et al. Minimum 20-year follow-up results of Harrington rod fusion for idiopathic scoliosis. Eur Spine J 2005;14:854e61. [9] Niemeyer T, B€ovingloh AS, Grieb S, et al. Low back pain after spinal fusion and Harrington instrumentation for idiopathic scoliosis. Int J Orthop 2005;29:47e50. [10] Helenius I, Remes V, Yrj€onen T, et al. Comparison of long-term function and radiological outcomes after Harrington instrumentation and spondylodesis in adolescent idiopathic scoliosis: a review of 78 patients. Spine 2002;15(27):176e80. [11] Paonessa KJ, Engler GL. Back pain and disability after Harrington rod fusion to the lumbar spine for scoliosis. Spine (Phila Pa 1976) 1992;17(8 suppl):S249e53. [12] Danielsson AJ, Nachemson AL. Back pain and function 23 years after fusion for adolescent idiopathic scoliosis: a case-control studye part II. Spine (Phila Pa 1976) 2003;28:E373e83. [13] Glattes RC, Bridwell KH, Lenke LG, et al. Proximal junctional kyphosis in adult spinal deformity following long instrumented posterior spinal fusion: incidence, outcomes, and risk factor analysis. Spine (Phila Pa 1976) 2005;30:1643e9. [14] Kuklo TR, Potter BK, Schroeder TM, et al. Comparison of manual and digital measurements in adolescent idiopathic scoliosis. Spine (Phila Pa 1976) 2006;31:1240e6. [15] Andersen GR, Andersen MO, Christensen SB. Selection of fusion levels in idiopathic scoliosis treated by Harrington-DTT instrumentation: a short term radiologic study. J Pediatr Orthop B 1995;4:86e90.