Sagittal curve and high metal denSity in adoleScent idiopathic ScolioSiS

Objectives: To analyze radiographically the postoperative kyphosis from patients undergoing surgical treatment for AIS with pedicle screws in all vertebrae included in the arthrodesis. Methods: Retrospective study. The following measurements were evaluated: Cobb angle in anteroposterior radiograph of the three curves (proximal thoracic, main thoracic, and lumbar), Cobb angle in the lateral view of the two curves: thoracic kyphosis (T5-T12) and lumbar lordosis (T12-S1). Results: Of the 25 patients evaluated preoperatively, four (16%) were hypokyphotic, 20 patients (80%) were normokyphotic and only one (4%) was hyperkyphotic. For hypokyphotic and hiperkyphotic patients a satisfactory correction of thoracic kyphosis was obtained in 100% of cases, which was preserved in the final result. The same pattern of thoracic kyphosis was observed for all normokyphotic patients throughout the follow-up. Conclusion: Radiographic evaluation of thoracic kyphosis in patients with AIS treated surgically with pedicle screws in all vertebrae showed satisfactory results with respect to the correction of thoracic kyphosis.

(3) to improve appearance, (4) to improve psychosocial health, (5) to reduce the development of low back pain, degenerative changes, and functional disability, and (6) to reduce cardiopulmonary involvement during adulthood. Surgical treatment of AIS is indicated when the primary curve presents a Cobb angle in the coronal plane greater than 40° in skeletally immature patients. 3 A thorough knowledge of the natural history of untreated AIS is essential for a correct evaluation of the outcome of surgical treatment, and to determine whether the benefits outweigh the risks. 8 A third-generation posterior instrumentation was developed by Cotrel-Dubousset, in 1984, using rods combined with hooks. 9 Compared to the other techniques described, its benefits are mainly related to improved correction of the frontal, sagittal, and rotational planes. 10 Several surgical techniques have been performed to correct deformity over the years and, more recently, correction using pedicle screws has become a reference for treatment. [10][11][12][13] The optimal surgical treatment for AIS requires correcting the coronal deformity of the main thoracic curve, improving posture, and achieving balance of the sagittal and coronal axes with the shoulders aligned and parallel with the pelvis. [14][15][16] The alignment of the spine in the sagittal plane is not yet well understood, and many methods have been used to measure this alignment, but to date there is no effective system to evaluate the spine in this plane. 17 Long-term follow-up of patients with surgically-treated AIS has shown that some cases evolve with rectification of both the thoracic kyphosis (TK) and the lumbar lordosis (LL), known as flat back syndrome. This syndrome includes back pain, degenerative disc disease, and functional disability. 14 Patients with primary thoracic scoliosis are typically hypokyphotic compared to patients without scoliosis. 18 Recent studies have evaluated the influence of arthrodesis on TK and general sagittal alignment. Depending on the type of instrumentation and the technique used for correction, postoperative TK may increase or decrease 10,13,18 and can interfere with the radiographic parameters of the profile.
Achieving a trunk that is balanced in both the coronal and sagittal planes is a major challenge in the treatment of AIS. This study is therefore important for increasing knowledge, and for an adequate understanding of the correction of this pathology, particularly in relation to the evaluation of kyphosis during the postoperative period.

MEtHODs
Following project approval by the Research Ethics Committee of the Institution (number 213/11), we selected case histories of patients diagnosed with AIS in the Department of Orthopedics and Traumatology between January 2005 and May 2011 who underwent posterior approach deformity correction surgery and arthrodesis, using third-generation instrumentation of all the vertebrae included in the arthrodesis. The radiographic images were evaluated both pre-and postoperatively, and during follow--up. Anterior-posterior (AP) and profile (P) views of the spine in the orthostatic position were evaluated for magnitude of the curve, using the Cobb method. 19 Angle measurements of the proximal thoracic curve, the main thoracic curve, and the lumbar curve were taken in AP view, while the angle measurements of thoracic kyphosis (T5 to T12) and lumbar lordosis (T12 to S1) were obtained in profile view. The curves were classified according to the criteria of Lenke et al 20 and King et al. 21 As inclusion criteria, patients diagnosed with adolescent idiopathic scoliosis who had undergone surgical treatment with pedicle screws in all the vertebrae included in the arthrodesis with a minimum follow-up time of six months were considered. Patients with scoliosis associated with neuromuscular disease, patients submitted to other surgical procedures or those with another type of implant, and also those who had required anterior access surgery for correction of the deformity were excluded. of 20.0°, and between 0° and 28° in the lumbar/thoracolumbar curve, with an average of 13.7°. The percentage of correction was 44.5% in the proximal thoracic curve, 66.4% in the main thoracic curve, and 59.6% in the lumbar/thoracolumbar curve. (Table 3) In the sagittal plane study (Tables 2 and 4), preoperative kyphosis varied between 5° and 45°, with an average of 22.4°, and preoperative lordosis varied between 22° and 60°, with an average of 41.5°. In the postoperative period, we found kyphosis varying between 12° and 37°, with an average of 23.6°, and lordosis between 12° and 51°, with an average of 37.7°.
There were no cases of infection or neurological complications. In the total group of patients evaluated preoperatively, there were 4 cases (16%) of hypokyphosis, 20 cases (80%) of normokyphosis, and one case (4%) of hyperkyphosis. In the case of the hypokyphotic and hyperkyphotic patients, we obtained satisfactory correction of the thoracic kyphosis in 100% of the cases, with the end result being normokyphosis. All the normokyphotic patients (100%) maintained the same standard of thoracic kyphosis during follow-up. (Table 5) correction of deformities in the coronal, sagittal, and rotational planes, less reduction loss, shorter constructions, and it improves pulmonary function without increasing neurological complications when compared with instrumentation with hooks or hybrid instrumentation (proximal hooks and distal pedicle screws). [10][11][12][13]24 This information agrees with our findings where, in the total case series of 25 cases, all of whom underwent surgery, we performed short constructions (an average of 8.4 levels involved/patient), obtained a high percentage of correction (66.4% in the main curve), and there were no procedurerelated complications.
Kim et al 10,13 analyzed assemblies using screws, hooks, and screws in combination with hooks (hybrid). They reported that assemblies using only screws achieved better coronal reduction than hybrid or hook instrumentations. Lowenstein et al 25 also observed a trend towards better correction of the main thoracic curve in cases where only screws were used.
The factors that determine the viability of posterior approach surgical correction of deformities include the magnitude of the curve, the flexibility of the curve, the extent of liberation of the posterior elements, the number of fixation points, and the correct identification of the stable vertebra. 26 In our study, all the patients underwent the same surgical liberation procedure. The deformity correction technique was performed via a posterior approach, with instrumentation with pedicle screws in all the vertebrae included in the arthrodesis. We achieved a percentage of correction of 44.5% in the proximal thoracic curve, of 66.4% in the main thoracic curve, and of 59.6% in the lumbar/thoracolumbar curve. These findings were similar to the results in recent publications of an average correction of 71.9%. 10,13,24 A case example is shown in Figure 1 in which the patient had the following preoperative angle measurements: main thoracic curve of 60°, lumbar curve of 45°, kyphosis of 18°, and lordosis of 20°. Following The values are presented as averages (minimum-maximum). The values are presented as averages (minimum-maximum). The cases are presented as a percentage (absolute number).

DIscUssION
AIS is a complex, three-dimensional deformity that affects the spine and the rib cage. Cases with curves of great magnitude and the potential for progression require surgical treatment. Currently, surgical techniques are almost always associated with instrumentation, which increases stability and has a higher potential for correction. 22 Different scoliosis classification systems have been described, attempting to categorize models of deformity and guide instrumentation plans. The most popular classification systems are those proposed by King et al 21 and Lenke et al. 20 In the Lenke system, in addition to the deformity in the coronal plane, the thoracic deformity in the sagittal plane (sagittal modifier) and the translation of the apical vertebra of the lumbar curve (lumbar modifier) are also considered. The Lenke system has been shown to be more precise in guiding fusion levels in the treatment of scoliosis.
AIS is more prevalent in females. 23 In our case series, we found a great similarity to the current literature. We had 22 female patients (88%) and 3 male patients (12%) diagnosed with idiopathic scoliosis, data similar to those of Kadoury et al who reported results with 83.9% women and 16.1% men.
The average age of our patients at the time of surgery was 14.2 years, a value very close to that obtained by Kadoury et al 23 , which was 15.3 years of age with a variation of ± 2.3 years.
Modern, multi-segmental instrumentation systems involve numerous options for spinal fixation and enable different techniques for correcting deformity. 15 Segmental instrumentation with pedicle screws permits better A C B D In our case series, we observed that the patients with AIS tended to be hypokyphotic and that by using an instrumentation technique including all the vertebrae in the arthrodesis, our final outcome was an increase in thoracic kyphosis, resulting in all patients being normokyphotic at the end of treatment.
Comparison of the average preoperative kyphosis with the average postoperative kyphosis shows an increase of 2%. This value may seem small, but our sample is made up of patients with general AIS and not only preoperative hypokyphotic patients with AIS. In our sample, if we analyze only the patients with hypokyphosis <20° (10 cases), we get an average correction of the kyphosis of 10.8° to 20.1°, documenting a gain of 9.3°. If we consider the cases with hypokyphosis <10° (four cases) and analyze our outcomes, we report an average kyphosis evolving from 5.5° preoperatively to 20.2° postoperatively, representing a correction of 14.7°.

cONcLUsION
Radiographic evaluation of the sagittal curves of patients with AIS surgically treated with pedicle screws in all the vertebrae included in the arthrodesis showed satisfactory outcomes in relation to the maintenance of thoracic kyphosis.
All authors declare no potential conflict of interest concerning this article.
A better understanding of sagittal spinal balance has been the focus of many recent studies. Patients with inadequate sagittal balance following surgery for scoliosis can develop flat back syndrome. 9 Newton et al 27 demonstrated that there is a direct relationship between thoracic kyphosis and lumbar lordosis. Patients who presented hypokyphosis postoperatively also tended to present rectification of the lumbar lordosis, precipitating the appearance of flat back syndrome. In the analysis of our case series, we were not able to identify this relationship.
Correction of the spinal profile is increasingly reported in publications. However, most authors do not distinguish the outcomes of patients with normal thoracic kyphosis from the preoperative hypokyphotic patients. In general, posterior instrumentation tends to reduce thoracic kyphosis (TK) while anterior instrumentation tends to increase it. 28 Kim et al 10,13 and Lowenstein et al 25 reported a significant reduction in thoracic kyphosis in patients who underwent surgery using only pedicle screws.
Suk et al 28 , in their large series of 203 patients with 5 years of follow-up, observed an average increase in TK of 5 degrees. De Jonge et al 18 observed an increase of 12° in the thoracic kyphosis of hypokyphotic patients (<20°) who underwent surgical treatment using hooks. All these conflicting outcomes demonstrate the difficulty in restoration of TK. 19