Pulmonary Function Tests Correlated With Thoracic Volumes in Adolescent Idiopathic Scoliosis

Transcription

1 Pulmonary Function Tests Correlated With s in Adolescent Idiopathic Scoliosis Charles Gerald T. Ledonio, 1 Benjamin E. Rosenstein, 2 Charles E. Johnston, 3 Warren E. Regelmann, 4 David J. Nuckley, 1,5 David W. Polly Jr. 1 1 Department of Orthopaedic Surgery, University of Minnesota, 2450 Riverside Avenue South, Minneapolis 55454, Minnesota, 2 Medical School, University of Minnesota, Minneapolis, Minnesota, 3 Department of Pediatric Orthopedic Surgery, Texas Scottish Rite Hospital for Children, Dallas, Texas, 4 Department of Pediatrics Pediatric Pulmonology, University of Minnesota, Minneapolis, Minnesota, 5 Zimmer Spine, Minneapolis, Minnesota Received 14 January 2016; accepted 19 May 2016 Published online 20 June 2016 in Wiley Online Library (wileyonlinelibrary.com). DOI /jor ABSTRACT: Scoliosis deformity has been linked with deleterious changes in the thoracic cavity that affect pulmonary function. The causal relationship between spinal deformity and pulmonary function has yet to be fully defined. It has been hypothesized that deformity correction improves pulmonary function by restoring both respiratory muscle efficiency and increasing the space available to the lungs. This research aims to correlate pulmonary function and thoracic volume before and after scoliosis correction. Retrospective correlational analysis between thoracic volume modeling from plain x-rays and pulmonary function tests was conducted. Adolescent idiopathic scoliosis patients enrolled in a multicenter database were sorted by pre-operative Total Lung Capacities (TLC) % predicted values from their Pulmonary Function Tests (PFT). Ten patients with the best and ten patients with the worst TLC values were included. Modeled thoracic volume and TLC values were compared before and 2 years after surgery. Scoliosis correction resulted in an increase in the thoracic volume for patients with the worst initial TLCs (11.7%) and those with the best initial TLCs (12.5%). The adolescents with the most severe pulmonary restriction prior to surgery strongly correlated with post-operative change in total lung capacity and thoracic volume (r 2 ¼ 0.839; p < 0.001). The mean increase in thoracic volume in this group was cm 3 (11.7%) which correlated with a 21.2% improvement in TLC. Scoliosis correction in adolescents was found to increase thoracic volume and is strongly correlated with improved TLC in cases with severe restrictive pulmonary function, but no correlation was found in cases with normal pulmonary function. ß 2016 Orthopaedic Research Society. Published by Wiley Periodicals, Inc. J Orthop Res 35: , Keywords: scoliosis correction; modeling; thoracic volume; pulmonary function test; adolescent idiopathic scoliosis; deformity; total lung capacity Adolescent idiopathic scoliosis (AIS) is diagnosed with a prevalence of approximately 3% in U.S. children between the ages of 10 and 16 years. 1 With increasing curve magnitude, prevalence decreases and a disparity emerges where females are affected with a significantly greater frequency than males. 1 Scoliosis may be classified by its etiology as congenital, idiopathic (both adolescent and adult), or neuromuscular 2 4 with adolescent idiopathic scoliosis being the most common. A comprehensive review of the natural history of AIS was compiled by Weinstein. 1 Much comorbidity accompanies scoliosis such as pain, mental health, and physical inabilities which lead to poor quality of life measures in these patients. 2,5,6 A common medical complaint from AIS patients is a lack of cardiovascular ability or exercise capacity 7 due to the progression of restrictive lung disease and respiratory insufficiency. 2,8,9 Pulmonary function is modulated based upon space available to the lungs, diaphragm, and intercostal muscle function, intrinsic lung health, and airway resistance. Scoliosis may affect the thoracic volume or space available to the lungs, the diaphragm function, and airway resistance. Specifically, scoliosis has been shown to restrict the capacity to breathe by limiting the total lung capacity (TLC) 10 and confining lung Correspondence to: Charles Gerald T. Ledonio (T: þ ; F: þ ; ledon001@umn.edu) # 2016 Orthopaedic Research Society. Published by Wiley Periodicals, Inc. growth 8,11 leading to decreased oxygen saturation. 12 Collapse of the rib cage in scoliosis decreases the space available to the lungs while also progressing with the development of pulmonary compromise of restrictive lung disease. 2,5 Scoliosis has been shown to negatively affect vital capacity, forced vital capacity (FVC), and forced expiratory volume in 1 s (FEV1). 6,12 In those with the most severe deformities, greater reductions in function are seen due to the thoracic cavity s inability to completely empty the lungs. 12 Progression in some scoliosis patients results in atelectasis 13 from the resulting loss of tidal volume and dead space in the lungs increasing the arterial PCO Thus, deformities of the thoracic spine have been shown to have the detrimental effects on pulmonary function. 2,10 Corrective surgery is used primarily to prevent further progression of the deformity, and secondarily to increase the space available for lung expansion. Indications for scoliosis correction surgery are generally based on severity of curve magnitude and risk of curve progression. Unfortunately, these have been recently found to correlate poorly with pulmonary function both pre- and post-operatively. 15 Newton et al. have also shown that clinically significant pulmonary impairment can be out of proportion with the severity of Cobb angle. 16 The poor correlation between Cobb angle and function could be attributed to the difficulty in accounting for curvature of the spine due to vertebral rotation. 14 Surgical correction has been found to successfully correct scoliosis deformity; however, its direct link with pulmonary function 175

2 176 LEDONIO ET AL. improvements is cause for debate Furthermore, it is difficult to determine optimal surgical correction due to confounding factors which are not accounted for by pulmonary function tests, such as fibrosis, asthma, and patient cooperation. 11 Adjustable expansion thoracoplasty surgeries are often performed in order to avoid this problem, but study populations are as yet too small to identify success rates. 6,13 There is also a lack of long-term follow-up data regarding pulmonary function to make comparative assessments. 6 Therefore, this study examined a cohort of patients undergoing spinal correction surgery and measured the relationship between thoracic volume and pulmonary function both pre-operatively and 2 years postoperatively. We aimed to test the hypothesis that pulmonary function improvement is directly related to increased thoracic volume in patients undergoing a scoliosis reconstruction surgery. By choosing those with the best and the worst pulmonary function, we can examine the effects at the extremes. METHODS Model Development and Validation Healthy patient computed tomography (CT) scans were uploaded into a validated image processing software (Mimics 1 v , Materialise 1, 3000 Leuven, Belgium) that calculated the surface 3D thoracic model from the CT stacked image data through image segmentation (Fig. 1A C). The models and radiographic images of scoliosis patients were then imported into a powerful open source 3D animation software (Blender 2.52, blender.org) (Fig. 2A). The original healthy model (Fig. 2B) was then deformed to match patient radiographs with skeletal deformities. Vertebral segments were matched first (Fig. 2C and D) followed by the ribs (Fig. 2E). The final model was checked against the original radiograph using the projected shadow from the model (Fig. 2F and G). Calculation The deformed thorax was calibrated to its correct size based on the radiographic parameters. Using built in functions and algorithms of the animation software, a 3D spherical mesh with a known volume was inserted within the thoracic cavity model then shrinkwrap modifier was applied using the mediastinal space (space available for lungs to expand) as the inner boundaries (Fig. 2H and I). The transformed mesh was exported back to the imaging software (Mimics) and volume was automatically calculated using its built-in algorithm. Validation Validation of this thoracic volume modeling method was performed by comparing the computed volume thoracic model outputs to the volumetric modeling from the gold standard thoracic CT scans of three patients with scoliosis and three patients with pectus deformity using the validated image processing software (Mimics). Percent error was computed as the total thoracic volume from the gold standard thoracic CT scans minus the computed thoracic volume divided by the total thoracic volume from CT then multiplied by 100. Retrospective Cohort A retrospective case cohort correlational analysis of PFT data and modeling reconstructions of thoracic volumes was performed. Twenty patients with AIS were identified based upon a sorting of the Prospective Pediatric Scoliosis Study (PPSS) database to distinguish those 10 patients with the best TLC percent predicted values and those with the worst PFT results (Fig. 3). The basis of this selection was to identify groups wherein modeling and measuring the thoracic volume of patients with AIS might be associated with measurable changes in pulmonary function outcomes. All patients underwent corrective instrumented fusion using pedicle screw contructs. This IRB-approved study included a sample with a mean age of 14.7 years (range years) and sex distribution of 3 males and 17 females. Pre-operative and 2 years post-operative sagittal and coronal plane film images for twenty patients were obtained. The radiographic images were analyzed and calibrated using imaging processing software (ImageJ). Once dimensionalized, the images were modeled to determine the actual volumes of each thorax in a blinded fashion. In the coronal image, the full thoracic area was measured by creating a closed loop from T1 following the perimeter of the lateral portion of the lung, across the perimeter of the base of the lungs, up the lateral perimeter of the other lung. Right and left lung areas were measured in a similar fashion, creating a loop along the perimeter of each lung. The height of each lung was measured from the superior apex of the lung to the inferior apex of the lung. In each reconstruction, the thoracic volume was computed in cubic centimeters. A similar method was used to measure the heights and areas of the lungs from the sagittal image. The lateral area of the lung was found by Figure 1. Imaging software 3D model reconstruction of the thoracic spine and mediastinum: (A) Image import from CT scan. (B) Segmentation. (C) 3D Model (images from Public Domain,

3 THORACIC VOLUME CORRELATED WITH PFT 177 Figure 2. (A) The original thorax in the animation software (shown without cartilage for clarity). (B) Close-up view of the original thorax aligned with the patient radiograph images. (C) A-P view after aligning the original spine with the image s scoliosis curve. (D) Lateral view of spine alignment with the radiograph. (E) Result after manipulating the entire thorax to match the radiographs. (F and G) X-ray views of the sagittal and coronal planes, respectively, after completing the manipulations. (H and I) Posterior and anterior views, respectively, of the fit of the thoracic volume (in black) within the scoliosis thorax model. (J) Comparison showing the original thorax (left) to the fit model (right). creating a closed loop along the lung s perimeter, starting from the top of the lung area. The medial area was found in a similar fashion but only using the area anterior to the spine. The lateral height was found by measuring from the superior apex of the lung to the diaphragm. Thus, the thoracic volume measured was the volume of the lungs and the mediastinum together. Spirometry was obtained at different locations with all efforts complying with the American Society (ATS) criteria for validity. Total lung capacity (TLC) was measured as the volume in liters within the lungs at maximal inflation. Predicted values were calculated using National Health and Nutrition Examination Survey (NHANES) data and equations. The patient s arm width was used as a substitute for height. The thoracic volume data and PFT data for every adolescent was measured both pre- and post-operatively. The surgical change in thoracic volume and PFT percent predicted values were averaged for the best PFT group and the worst PFT group. Paired Student s t-tests were performed to identify if differences existed between the pre- and postoperative values. Finally, the changes in thoracic volume and PFT values were regressed and a statistical F-test performed to identify if a relationship existed. RESULTS Validation The volumes measured for the scoliosis and pectus cases had a mean difference of 172 cm3 between CT thoracic volume and computational thoracic volume model from chest radiographs which translated to a mean error of 4.0% (Table 1). Retrospective Cohort The 10 patients with worst pre-operative PFTs (Group 1) had TLCs of 3.0 L and 55.5% predicted volume, while those with best PFTs (Group 2) had TLCs of 3.9L and 110.4% predicted volume. The average age was 14.3 and 15.4 years, respectively. The mean main thoracic curves (measured using Cobb method) pre-operatively were 68.5 and 35.2 while mean sagittal curves (T2 T12) pre-operatively were 15.8 and 34.2, respectively. The mean difference between the two groups was statistically significant (p < 0.001) using t-test. The mean main thoracic curves at 2 years post-operatively were 26.7 and 11.2, while mean sagittal curves were 26.1 and 30,

4 178 LEDONIO ET AL. Figure 3. Sample radiographs of AIS pre- and post-surgery. Panels A and B represent a patient in the database from the group with the most restrictive PFT values; panels C and D the least restrictive PFT values. respectively. The mean angular difference between the two groups at 2 years follow-up did not meet statistical significance (p ¼ 0.47). There was a strong correlation between the changed in Cobb angle correction and sagittal curve correction with thoracic volume modeled (Pearsons correlation coefficient of r ¼ 0.80) (Table 2). The patients with the lowest pulmonary function values were pre-operatively found to have mean values (liters, [percent predicted]) of 1.89 (59%), 2.20 (59.4%), and 2.97 (55.5%) for FEV1, FVC, and TLC, respectively. Post-operatively at 2 years, the mean values were 2.48 (68.2%), 2.50 (63.7%), and 3.54 (76.7%). This corresponded to average change of (9.2%), (4.3%), and (21.2%) for FEV1, FVC, and TLC, respectively. A paired comparison of these PFT values revealed a significant increase in TLC with surgical correction (p < 0.001). The patients with the highest pre-operative PFT values had mean FEV1, FVC, and TLC values (liters, [percent predicted]) of 3.02 (107.4%), 3.43 (108.4%), and (111%), respectively. Post-operatively, the average values were 3.00 (98.6%), 3.46 (102.4%), and 4.01 (105.8%). The average change in these values was ( 8.83%), ( 6.01%), and ( 5.27%) for FEV1, FVC, and TLC respectively, exhibiting no statistical difference (Fig. 4). Patients with the lowest PFT values had an average pre-operative thoracic volume of 3,194 1,358 cm 3 and average post-operative thoracic volume of 3,466 1,362 cm 3. The mean change in volume for these patients was 373 cm 3 or 11.7% and the difference between pre- and post-surgical volumes exhibited a trend (p ¼ 0.065). Patients with the best PFT scores had an average pre-operative thoracic volume of 3, cm 3 and a post-operative average volume of 3, cm 3. These surgical changes were significant (p ¼ 0.033), demonstrating an increase in thoracic volume of 429 cm 3 or 12.5% (Fig. 5). The adolescents with the most severe pulmonary compromise prior to surgery exhibited a strong positive relationship between post-operative change in TLC and thoracic volume (r 2 ¼ 0.839; p < 0.001) (Fig. 5). In patients with the highest initial PFT values, correlations between pre- and post-operative changes in FEV1 and FVC with pre- and post-operative changes in thoracic volume were not observed. DISCUSSION The test of our central hypothesis revealed that for those with compromised pulmonary function, improvement in PFTs was directly related to increased thoracic volume after AIS surgery (Fig. 6). The null hypothesis was rejected which stated that no relationship or a negative relationship existed between change in PFT values and thoracic volume. Both the highest and lowest PFT values pre-operatively exhibited significant increases in the space available to the lungs (approximately 14%) despite only the lowest initial PFT group demonstrating significantly improved PFTs post-operatively. Table 1. Validation of Modeling Versus Gold Standard CT Scan s Cases CT-Based Calculation (cm 3 ) Computed Modeling (cm 3 ) Difference % Error 1 1,485 1, ,702 5, ,807 8, ,071 9, ,661 6, ,202 6, Mean 6,155 6,

5 THORACIC VOLUME CORRELATED WITH PFT 179 Table 2. Pulmonary Function Test Parameters,, and Curve Magnitude of the 10 Worst and 10 Best Total Lung Capacities (TLC) Preop 2 Yr Postop Difference Preop Versus 2 Yr Postop ID Groups Sex Age at Surgery TLC, Liters TLC, % Predict Main Curve Sagittal Curve (T2 T12) TLC, Liters TLC, % Predict Main Curve Sagittal Curve (T2 T12) Difference % Difference W1 10 M , , W2 Worst F , , W3 TLC F , , W4 M , , W5 F , , W6 F , , W7 M , , W8 F , , W9 F , , W10 F , , Mean , , F , , B2 F , , B1 10 Best TLC B3 F , , B4 F , , B5 F , , , B6 F , , B7 F , , B8 F , , B9 F , ,648 Missing Missing 1, B10 F , , Mean , ,

6 180 LEDONIO ET AL. Figure 4. Estimated total lung capacity (percent) measured by PFTs pre- and post-operatively. Those with the least restrictive PFT values showed an insignificant change after surgery; those with the most restrictive initial PFT scores showed significant improvement in their TLC (p < 0.001). Those AIS patients with the highest PFT values initially did not demonstrate improvement possibly due to a ceiling effect, functional adaptation, or the model is detecting a change in total thoracic but not lung volume. Before surgery, the group with the highest PFT scores had values near or over 110% of the predicted PFT value. Potentially, one lung, on the convex side of the spinal curve, may over-compensate for the restriction of the other lung, causing large PFT scores. After surgery, however, the volumes of the lungs may balance such that one lung s abilities improve while the other lung s abilities normalize and the PFTs show a more normal score, near 100% predicted value. Another possibility, especially for surgeries performed in children and adolescents, is that the corrective surgery may lead to the short straight spine problem as described by Campbell et al. 11 As the body grows, the fused spine and thorax is unable to support development of the lungs leading to adverse pulmonary function. 2,11 In the group with the lowest PFT values, the amount of restriction may have been so great pre-operatively that surgical repair outweighed any spinal fusion issues. The use of pulmonary function as an indication for surgery and a possible predictor of functional outcomes have been suggested by others. Motoyama et al. 13 reported patients with pre-operative FVC of greater than 50% predicted, initially showed post-operative improvements but they returned to baseline nearly 3 years post-operatively. This study also demonstrated the benefits of surgical repair in preventing further deterioration of patients pulmonary function. Yuan et al. 20 showed that FVC and FEV1 results in the immediate post-operative period, relative to preoperative results, can decrease significantly. Similar results had also been seen by Graham et al. 21 at 3- months and 1-year follow-up, though a return to baseline was seen by 2 years post-operatively. Decreases in pulmonary function of a 20 40% can be expected in the acute post-operative stage due to Figure 5. volume measured pre- and post-operatively. Increases in the space available in the thorax were measured for both groups with the least (p ¼ 0.033) and most (p ¼ 0.065) restrictive initial PFTs. surgery s effects on the intercostal muscles. 12 However, these impacts may be limited with less invasive techniques. 22 Our results suggest that pulmonary function can be maintained at 2-year follow-up, as suggested by Bowen, 14 and that it is related to the correction and increase in the thoracic volume. Classically, AIS indications for surgery have been based on Cobb angle measures; though Redding and Mayer showed these do not correlate strongly with PFTs. 15 In contrast, Johnston et al. 9 found that larger spinal curves (T5 T12) were correlated significantly with poor pre-operative PFTs. Nepple and Lenke 23 described a case of improved pulmonary function through FVC and FEV1 after correction of scoliosis curvature. Widmann et al. 6 reported significant inverse correlations with FVC and FEV1 but not with TLC and scoliosis curvatures in patients with osteogenesis imperfecta; however, these values were only reported in the pre-operative period. Kim et al., 24 however, noted no correlation between Cobb angle correction and significant clinical improvement in PFTs of 31 patients. There appears to be conflict on Figure 6. Correlation of change in PFT values with change in thoracic volume in patients with the most restrictive initial PFT values. The percent change in thoracic volume showed a strong correlation to the percent change in the TLC, but only in those patients with the most restrictive pre-operative PFT values.

7 THORACIC VOLUME CORRELATED WITH PFT 181 the use of Cobb angle assessments for predicting pulmonary function 15 and thus measurement of the space available to the lungs (thoracic volume) may enhance the prediction of lung function outcomes and surgical indications for those with AIS, 9 especially in those with moderate to severe restrictive pulmonary function pre-operatively. As with most retrospective patient analyses, a number of limitations exist which contextualize the results. The pulmonary function tests were standardized, but given at different institutions by different individuals. Since patient effort may affect the outcomes and is unknown, the PFT values represent the best effort at a standardized measure of pulmonary function. This is similar to the findings of Gagnon et al. 25 The surgical treatments, while consistent and similar were not performed by the same surgeon in all cases and may affect the post-operative results as described by Newton et al. 26 Even though our statistical comparisons only involved paired analyses, the lack of height-based normalization of all subjects may affect the generalizability of the raw results. Finally, validation of our thoracic volume modeling method is not perfect; however, it is similar to that reported by Koehler and Wischgoll in CONCLUSION The pulmonary function of a cohort of AIS patients was evaluated pre-operatively and 2 years postoperatively. Our data show that pre- to post-operative change in thoracic volumes and total lung capacity are strongly correlated when pre-operative lung capacities are low but not correlated when pre-operative total lung capacities are within normal range. AUTHORS CONTRIBUTIONS All listed authors have contributed substantially to the research design; acquisition, analysis, and interpretation of data; and drafting or critically revising the manuscript of this study. All authors have read and approved the final submitted manuscript. ACKNOWLEDGMENTS None of the authors have received payments or services for this work, and none have financial ties influenced by this work. No funding was received for this study. The authors wish to acknowledge the Prospective Pediatric Scoliosis Study (PPSS) for providing the de-identified images and data which were utilized in this study. REFERENCES 1. Weistein S Adolescent idiopathic scoliosis: prevalence and natural history. In: Weinstein SL, editor. The pediatric spine: principles and practice. New York, NY: Raven Press, Ltd. p Vitale MG, Matsumoto H, Bye MR, et al A retrospective cohort study of pulmonary function, radiographic measures, and quality of life in children with congenital scoliosis: an evaluation of patient outcomes after early spinal fusion. Spine (Phila Pa 1976) 33: Djurasovic M, Glassman SD Correlation of radiographic and clinical findings in spinal deformities. Neurosurg Clin N Am 18: Kuklo TR, Potter BK, Polly DW Jr., et al Monaxial versus multiaxial thoracic pedicle screws in the correction of adolescent idiopathic scoliosis. Spine (Phila Pa 1976) 30: Vitale MG, Matsumoto H, Roye DP Jr., et al Healthrelated quality of life in children with thoracic insufficiency syndrome. J Pediatr Orthop 28: Widmann RF, Bitan FD, Laplaza FJ, et al Spinal deformity, pulmonary compromise, and quality of life in osteogenesis imperfecta. Spine (Phila Pa 1976) 24: Menon B, Aggarwal B Influence of spinal deformity on pulmonary function, arterial blood gas values, and exercise capacity in thoracic kyphoscoliosis. Neurosciences (Riyadh) 12: Mehta HP, Snyder BD, Baldassarri SR, et al Expansion thoracoplasty improves respiratory function in a rabbit model of postnatal pulmonary hypoplasia: a pilot study. Spine (Phila Pa 1976) 35: Johnston CE, Richards BS, Sucato DJ, et al Correlation of preoperative deformity magnitude and pulmonary function tests in adolescent idiopathic scoliosis. Spine (Phila Pa 1976) 36: Hale K, Rasp F Pulmonary function testing. In: Moe JH, Bradford DS, editors. Moe s textbook of scoliosis and other spinal deformities. Philadelphia: W. B. Saunders Co Ltd. p Campbell RM Jr., Smith MD, Mayes TC, et al The characteristics of thoracic insufficiency syndrome associated with fused ribs and congenital scoliosis. J Bone Joint Surg Am 85-A: Lenke LG Pulmonary and chest cage physiology. In: DeWald RL, Arlet V, Carl AL, et al., editors. Spinal deformities. The comprehensive text. New York, NY: Thieme. p Motoyama EK, Deeney VF, Fine GF, et al Effects on lung function of multiple expansion thoracoplasty in children with thoracic insufficiency syndrome: a longitudinal study. Spine (Phila Pa 1976) 31: Bowen R Respiratory management in scoliosis. In: Moe JH, Bradford DS, editors. Moe s textbook of scoliosis and other spinal deformities. Philadelphia: Saunders. p Redding GJ, Mayer OH Structure-respiration function relationships before and after surgical treatment of earlyonset scoliosis. Clin Orthop Relat Res 469: Newton PO, Faro FD, Gollogly S, et al Results of preoperative pulmonary function testing of adolescents with idiopathic scoliosis. A study of six hundred and thirty-one patients. J Bone Joint Surg Am 87: Cotrel Y, Banai M [Scoliosis and lumbarization of the first sacral vertebra with osseous bridging between the transverse apophyses, of L5-L6. Apropos of a case]. Acta Orthop Belg 41: Meister R, Heine J [Pulmonary function in young patients with scoliosis before and after operation according to Harrington (author s transl)]. Z Orthop Ihre Grenzgeb 111: Shneerson JM, Edgar MA Cardiac and respiratory function before and after spinal fusion in adolescent idiopathic scoliosis. Thorax 34: Yuan N, Fraire JA, Margetis MM, et al The effect of scoliosis surgery on lung function in the immediate postoperative period. Spine (Phila Pa 1976) 30: Graham EJ, Lenke LG, Lowe TG, et al Prospective pulmonary function evaluation following open thoracotomy

8 182 LEDONIO ET AL. for anterior spinal fusion in adolescent idiopathic scoliosis. Spine (Phila Pa 1976) 25: Sucato DJ, Erken YH, Davis S, et al Prone thoracoscopic release does not adversely affect pulmonary function when added to a posterior spinal fusion for severe spine deformity. Spine (Phila Pa 1976) 34: Nepple JJ, Lenke LG Severe idiopathic scoliosis with respiratory insufficiency treated with preoperative traction and staged anteroposterior spinal fusion with a 2-level apical vertebrectomy. Spine J 9:e9 e Kim YJ, Lenke LG, Bridwell KH, et al Prospective pulmonary function comparison following posterior segmental spinal instrumentation and fusion of adolescent idiopathic scoliosis: is there a relationship between major thoracic curve correction and pulmonary function test improvement? Spine (Phila Pa 1976) 32: Gagnon S, Jodoin A, Martin R Pulmonary function test study and after spinal fusion in young idiopathic scoliosis. Spine (Phila Pa 1976) 14: Newton PO, Perry A, Bastrom TP, et al Predictors of change in postoperative pulmonary function in adolescent idiopathic scoliosis: a prospective study of 254 patients. Spine (Phila Pa 1976) 32: Koehler C, Wischgoll T Knowledge-assisted reconstruction of the human rib cage and lungs. IEEE Comput Graph Appl 30:17 29.