Department of Neurosurgery, University of Virginia, Charlottesville, Virginia

J Neurosurg Spine 9:000 000, 9:326 331, 2008 Neurological symptoms and deficits in adults with scoliosis who present to a surgical clinic: incidence and association with the choice of operative versus nonoperative management Clinical article Ju s t i n S. Sm i t h, M.D., Ph.D., Ka i-mi n g Fu, M.D., Ph.D., Pe t e r Ur b a n, R.N., a n d Ch r i s t o p h e r I. Sh a f f r ey, M.D. Department of Neurosurgery, University of Virginia, Charlottesville, Virginia Object. Adults with scoliosis often present with neurological symptoms and deficits. However, the incidence of these findings and how they may affect treatment decisions have not been clearly defined. The purpose of this study was to quantify the prevalence of neurological symptoms and deficits in adults with scoliosis presenting to a surgical clinic, and to assess for an association between these factors and the decision to pursue operative treatment. Methods. In this study, the authors document the Oswestry Disability Index (ODI), radiographic findings, and the incidences of back pain, neurological symptoms (radiculopathy and claudication), and neurological deficits (weakness, myelopathy, and bowel/bladder dysfunction) and correlate these with operative versus nonoperative management. Pain was assessed using the visual analog scale (VAS) score. Of 207 patients, 25% underwent surgery. Results. Incidences of back pain (VAS score > 0 points) and radiculopathy (VAS score > 0) were 99 and 85%, respectively. The incidences of severe (VAS score > 5) back pain and radiculopathy were 66 and 47%, respectively. Neurological symptoms and deficits included weakness in 8% of patients, claudication in 9%, myelopathy in 1%, and bowel/bladder dysfunction in 3%. Patients with severe radiculopathy had greater mean ODI scores (p < 0.001) and reduced lumbar lordosis (p = 0.04) and were more likely to have de novo scoliosis (p = 0.009). Patients who underwent surgery had higher ODI scores (p < 0.001) and a greater incidence of severe radiculopathy (p = 0.006), weakness (p < 0.001), and neurogenic claudication (p = 0.003). Factors associated with operative management on multivariate analysis included weakness (p < 0.001), severe radiculopathy (p = 0.02), and sagittal imbalance (p = 0.03). Conclusions. Neurological symptoms and deficits are common among adults with scoliosis. Development of neurological symptoms and/or deficits is strongly associated with the decision to pursue operative treatment. (DOI: 10.3171.SPI.2008.9.10.326) Ke y Wo r d s adult scoliosis claudication neurological symptoms nonoperative treatment radiculopathy surgery In contrast to adolescents with idiopathic scoliosis, adults with scoliosis often present with neurological symptoms and/or deficits. However the incidence of these neurological findings and how they may affect management decisions have not been clearly defined. The results of a recent study suggest that the prevalence of scoliosis among healthy individuals older than 60 years of age may be as high as 68%. 19 As demographic shifts favoring a significant expansion of the elderly population Abbreviations used in this paper: AIS = adolescent idiopathic sco liosis; ODI = Os westry Disability Index; VAS = visual analog scale. are coupled with medical advances that continue to extend the average life expectancy, 8 it will become increasingly more important to understand the factors that lead adult patients with scoliosis to seek medical and surgical treatment. There are several ways in which adults with scoliosis come to medical attention. They may present without symptoms based on incidental radiographic findings and require only education and follow-up. Alternatively, they may present with progressive deformity or for treatment of severe and disabling back or neurological symptoms. Initial management of the symptomatic patient without progressive neurological deficits typically involves nonoperative treatment in an effort to avoid the potential 326 J. Neurosurg.: Spine / Volume 9 / October 2008

Neurological symptoms and deficits in adult scoliosis complications of an extensive surgical intervention, especially in elderly patients. The frequent use of nonoperative resources in adults with scoliosis has been well-documented by Glassman et al. 10 These resources, including physical therapy, steroid injections, nonsteroidal antiinflammatory drugs, and narcotics, may temporize, but a subset of patients will ultimately reach a point at which the risks of surgical intervention are offset by the impact of the deformity. The factors that govern the transition from nonoperative to operative therapy are not well understood. Despite the substantial expense and potential for complications or death with surgical intervention, the benefits have been documented in selected populations. 2,4 7,9,14 The authors of several studies have reported on the incidence of pain in adult scoliosis patients without distinguishing between incidences of back and leg pain 2,3,6,9,20 22 or without separation of adult scoliosis from other deformities such as primary sagittal imbalance. 9 11 The purpose of the present study is to quantify the prevalence of specific neurological symptoms and findings in a large population of adult patients with scoliosis and assess for a potential association between these factors and the decision to pursue operative treatment. Patient Population Methods This study is a retrospective review of an ongoing pro - spectively collected, consecutive series of patients from a single-surgeon database of adult spinal deformity. Enrollment criteria for this database include age older than 21 years and scoliosis 20 or sagittal imbalance > 6 cm. Patients with a history of spinal surgery are not excluded provided that the residual deformity meets the enrollment criteria. At the time of entry into the database, patients undergo a detailed history and physical examination, including a thorough neurological examination. All patients at time of database enrollment complete a series of objective functional and pain assessment tools including the ODI questionnaire and VAS for back and leg pain, in which the extreme scores of 0 and 10 reflect states of no pain and unbearable pain, respectively. At the time of enrollment into the adult spinal deformity database, patients are assigned to either the operative or nonoperative treatment groups based on the initial management approach. The decision whether to pursue operative or nonoperative management is complex and is based on patient input and physician counseling. With rare exceptions, patients seen in our surgical clinic have already received nonoperative therapies and are being referred for a surgical evaluation. If patients present with a progressive neurological deficit, severe myelopathy, or frank incontinence, they are typically strongly counseled to pursue operative treatment. The remaining patients are counseled extensively about the details of possible surgical intervention including such details as recovery, potential complications, and the range of expected outcomes. The type and results of prior nonoperative treatments are thoroughly investigated to see if other treatment options J. Neurosurg.: Spine / Volume 9 / October 2008 exist. Attempts at physical therapy, epidural and selective nerve-root steroid injections, and pharmacological therapy are evaluated. Many patients have had inadequate or inconsistent pain management, and referral for enrollment in a formal pain management program is frequently performed. Many patients are severely deconditioned and a reconditioning program through aquatics therapy is often recommended. Patients with predominantly back pain in whom nonoperative treatment has failed are typically counseled that on average surgical treatment achieves a 50% improvement in pain and function. 12 Patients with radiculopathy are typically counseled that on average surgical treatment gives > 75% improvement in neurological symptoms and a 50% improvement in back pain and overall function. Patients with stable weakness, myelopathy, or bowel/ bladder dysfunction are counseled that surgery may offer some improvement, although the degree of improvement is difficult to predict and is based in part on the magnitude of the deficit and the length of time the deficit has been present. Patients with significant sagittal imbalance are advised that surgical correction could potentially improve overall function, but that sagittal imbalance alone does not necessitate surgical intervention. After counseling and within the confines of appropriate practice, the ultimate management decision is typically guided by patient choice. At the time of data extraction for the present study, 319 patients were enrolled in the full adult spinal deformity database (74 in the operative and 245 in the nonoperative group). Inclusion criteria for the present analysis included a primary diagnosis of adult untreated AIS or degenerative (de novo) scoliosis without a history of prior scoliosis surgery. Specifically excluded were patients with neuromuscular or congenital scoliosis and patients with isolated sagittal plane deformities. Patients with a progressive neurological deficit or frank incontinence that would require the surgeon to counsel strongly for surgical management were excluded. Basic clinical data extracted for the present study included age at enrollment, sex, history of AIS, and history of spine surgery. Data related to the presence of back pain and neurological symptoms and findings were also extracted, including baseline individual VAS scores for back and leg pain. A VAS score > 5 was used to classify either back pain or radiculopathy as severe. Radicular weakness was defined as one or more motor strength scores 4 on a 5-point scale on physical examination. Neurogenic claudication was strictly defined as the onset of bilateral lower extremity pain that develops with ambulation and is relieved by rest. The diagnosis of myelopathy was based on the presence of hyperreflexia, abnormal clonus, and/or the Babinski sign. Bladder or bowel dysfunction clearly resulting from a diagnosis unrelated to spinal disease was not included in the present study. This study was approved by the University of Virginia Institutional Review Board. Imaging Parameters Full-length scoliosis radiographs (anteroposterior and lateral) are obtained in all patients at the time of enrollment. Curve classification as thoracic, thoracolumbar, or 327

J. S. Smith et al. lumbar is based on the largest curve measuring 20. If 1 or 2 additional curves measuring 25 are present, then the deformity is classified as either a double or triple major curve, respectively. Coronal balance, maximum Cobb angle, sagittal balance, thoracic kyphosis (T2 T12), and lumbar lordosis (T12 S1) were assessed using the standard techniques. 15 Statistical Analysis Frequency distributions and summary statistics were calculated for all clinical and radiographic variables. For categorical variables, cross-tabulations were generated and the Fisher exact test was used to compare their distributions. For continuous variables, the Student t-test was used to investigate differences in the distributions between subsets of patients classified by categorical data (such as sex or operative versus nonoperative management). Forward stepwise binary logistic regression analysis was used to assess differences between patients in the operative and nonoperative arms of the study using the following component factors: age at presentation, sex, scoliosis etiology (adult untreated AIS versus adult de novo), history of prior spine surgery, presence of severe back pain, severe radiculopathy, weakness, neurogenic claudication, myelopathy, and bladder/bowel dysfunction, the magnitude of coronal imbalance (absolute value), maximum Cobb angle on coronal radiographs, coronal curve classification (see above), sagittal balance, thoracic kyphosis (T2 T12), and lumbar lordosis (T12 S1). All statistical analyses were 2-sided and were performed using commercially available software (SPSS software, version 11.5; SPSS Inc.). Probability values < 0.05 were considered statistically significant. Results Patient Population and Imaging Parameters The 207 patients who met the inclusion criteria were 76% women and had a mean age of 64 years. Degenerative scoliosis was present in 87% of patients, and AIS without prior surgical correction was present in 13% of cases. Patients with untreated AIS were significantly younger at presentation (mean age 50 years), that those patients with adult de novo scoliosis (mean age 66 years, p < 0.001). Approximately one-third of patients had a history of prior spine surgery, generally decompressive or short-segment fusion procedures. All patients with fusion at > 2 levels were specifically excluded from the series. Of the 207 total patients, 51 patients (25%) underwent surgery, and 156 (75%) received nonoperative therapy. Patients who underwent surgery had a significantly higher mean disability score as assessed with the ODI (52 versus 40, p < 0.001). Treatment groups did not differ significantly with regard to sex distribution, patient age at presentation, scoliosis origin, or history of prior spine surgery (Table 1). The mean coronal imbalance (absolute value) and mean maximum Cobb angle were 23 mm and 38, respectively, and the mean thoracic kyphosis (T2 T12) and mean lumbar lordosis (T12 S1) were 39 and 34, respectively. The 2 most frequent coronal curve classifications were thoracolumbar major (43%) and double major (32%). Of the radiographic parameters assessed, only sagittal balance differed significantly between the operatively treated (mean 71 mm) and nonoperatively treated (mean 48 mm) patients (p = 0.02; Table 1). Neurological Symptoms and Findings The frequencies of clinical findings are summarized TABLE 1 Patient demographics and radiological descriptives of 207 adults with scoliosis* Parameter Treated Operatively (51 patients) Treated Nonoperatively (156 patients) female sex (%) 36 (71) 121 (78) mean age in yrs, (median, range) 63 (68, 24 85) 64 (65, 30 86) mean ODI score (median, range) 52 (54, 8 88) 40 (40, 10 82) scoliosis origin no. w/ adult de novo 44 (86) 136 (87) no. w/ adult untreated AIS 7 (14) 20 (13) history of prior spine op (%) 24 (47) 49 (31) coronal radiographic results mean coronal balance (median, range) 23 mm (20, 0 84) 23 mm (18, 0 294) mean max Cobb angle (median, range) 37 (32, 20 78) 38 (35, 20 80) no. w/ thoracic major curve (%) 2 (4) 9 (6) no. w/ thoracolumbar major curve (%) 24 (47) 64 (41) no. w/ lumbar major curve (%) 9 (18) 24 (15) no. w/ double major curve (%) 15 (29) 52 (33) no. w/ triple major curve (%) 1 (2) 7 (5) sagittal radiographic results mean sagittal balance (median, range) 71 mm (65, 60 to 320) 48 mm (40, 74 236) mean T2 T12 kyphosis (median, range) 37 (40, 12 to 69) 40 (39, 3 95) mean T12 S1 lordosis (median, range) 28 ( 30, 80 to 43) 36 ( 38, 110 to 56) * Coronal balance reflects the absolute value of coronal imbalance. Coronal curves were classified based on the largest curve 20. If 1 or 2 additional curves were 25, then the deformity was classified as either double or triple major, respectively. Of the parameters presented, 2 demonstrated statistically significant differences between operatively and nonoperatively treated patients, ODI (p 0.001), and sagittal balance (p = 0.02). A greater proportion of patients who underwent surgery had a history of prior surgery but this finding did not reach statistical significance (p = 0.06). 328 J. Neurosurg.: Spine / Volume 9 / October 2008

Neurological symptoms and deficits in adult scoliosis TABLE 2 Clinical findings in 207 adults with scoliosis* Treated Treated Parameter Operatively Nonoperatively p Value no. of patients (%) 51 (25) 156 (75) back pain present (%) 50 (98) 154 (99) 0.6 mean VAS score (median) 7.2 (8) 6.1 (7) 0.002 severe (%) 40 (82) 92 (61) 0.009 radicular pain present (%) 44 (86) 132 (85) 0.5 mean VAS score (median) 5.9 (7) 4.6 (5) 0.007 severe (%) 33 (65) 63 (40) 0.006 radicular weakness (%) 13 (25) 3 (2) 0.001 unilateral (%) 7 (54) 2 (67) bilateral (%) 6 (46) 1 (33) neurogenic claudication (%) 10 (20) 8 (5) 0.003 myelopathy (%) 2 (4) 0 (0) 0.06 bladder/bowel dysfunction (%) 4 (8) 3 (2) 0.06 * Back and radicular pain were considered present if the patient had a VAS score 0 or if there was documentation of radicular or back pain in the medical records, but VAS was not recorded. Severe pain was defined as a VAS score 5. A VAS score was available in 201 patients. A VAS score was available in 204 patients. in Table 2. Back pain was present in the vast majority (99%) of patients studied. The mean VAS score for back pain was significantly higher, and the presence of severe back pain (VAS score > 5) was significantly more frequent in patients who underwent surgery (p = 0.002 and p = 0.009, respectively). Radiculopathy was present in 85% of patients, and the mean VAS score for radiculopathy and prevalence of severe radiculopathy were significantly higher in patients who received operative treatment (p = 0.007 and p = 0.006, respectively). Radicular weakness was present in 8% of patients, and approximately half of these patients had bilateral weakness. Neurogenic claudication, myelopathy, and bladder/bowel dysfunction were present in 9, 1, and 3% of patients, respectively. Both radicular weakness and neurogenic claudication were significantly more frequent in patients who underwent surgery than in those who received nonoperative treatment (p < 0.001 and p = 0.003, respectively). Clinical and radiographic factors associated with severe radiculopathy are summarized in Table 3. Patients with severe radiculopathy had significantly higher ODI scores (p < 0.001) and had a lesser magnitude of lumbar lordosis (p = 0.04). In addition, patients with a history of AIS had significantly lower mean VAS scores for radiculopathy compared to patients with de novo scoliosis (p = 0.009). The following parameters did not have a significant association with the presence of severe radiculopathy: age (p = 0.07), sex (p = 0.4), history of prior spine surgery (p = 0.9), coronal curve classification (p = 0.4), magnitude of coronal balance (p = 0.1), maximum Cobb angle (p = 0.2), sagittal balance (p = 0.9), and thoracic kyphosis (p = 0.8). Analysis of Crossover Between Treatment Groups All patients in the operative treatment group underwent surgery. During the follow-up period, 23 patients (15%) who had initially been classified into the nonoperative group crossed over into the operative group at a mean interval of 11 months after enrollment. The mean follow-up period in the remaining patients in the nonoperative group was 25 months. Patients from the nonoperative group who crossed over reported significantly higher VAS scores for leg and back pain at enrollment compared with those who did not cross over (VAS scores for leg pain: 6.0 versus 4.3, p = 0.019; VAS scores for back pain: 7.1 versus 6.0, p = 0.016). Among patients who crossed over, the principal symptom prompting surgery was radicular pain in 14 patients, back pain in 5 patients, and equivalent radicular and back pain in 4 patients. Notably, progressive neurological deficits had also developed in 3 patients. Patients who crossed over did not significantly differ from those who did not cross over with regard to baseline ODI score (p = 0.4), age (p = 0.7), baseline maximum Cobb angle (p = 0.1), baseline sagittal balance (p = 0.4), or baseline lumbar lordosis (p = 0.3). Logistic Regression Analysis Multivariate analyses were performed to assess the relative contributions of clinical and radiographic parameters to the decision between operative and nonoperative management among the 207 adult patients with scoliosis TABLE 3 Clinical and radiographic parameters that correlate with presence of severe lower extremity radicular pain in adults with scoliosis* Severe Lower Extremity Radicular Pain Parameter No Yes Mean VAS Score (median) p Value no. of patients (%) 108 (53) 96 (47) mean ODI score (median, range) 37 (38, 8 82) 50 (49, 22 88) 0.001 mean lumbar lordosis (T12 S1) (median, range) 37 ( 35, 110 to 32) 30 ( 35, 90 to 56) 0.04 scoliosis origin 0.009 adult idiopathic 3.4 (2) adult de novo 5.1 (5) * Assessed parameters without statistically significant difference included: age (p = 0.07), sex (p = 0.4), history of prior spine surgery (p = 0.9), coronal curve classification (p = 0.4), magnitude of coronal balance (p = 0.1), maximum Cobb angle (p = 0.2), sagittal balance (p = 0.9), and thoracic kyphosis (p = 0.8). J. Neurosurg.: Spine / Volume 9 / October 2008 329

J. S. Smith et al. studied. The best fit model (Table 4) indicates that patients who elected operative management were significantly more likely to have radicular weakness (p < 0.001), severe radicular pain (p = 0.02), and a more positive sagittal balance (p = 0.03). Notably, the odds ratio for sagittal balance is calculated per millimeter. Discussion Significant progress has been made in the recognition, classification, and management of scoliosis in adolescents. 4,17,18,23 25 Although such progress has lagged behind for adults, several factors are driving rapid advancements including improved safety of anesthesia, advancements in spinal instrumentation, a longer average life expectancy, a demographic shift toward an older population, and increasing expectations of the elderly to remain functionally active later in life. 1,16 The purpose of the present study was to define the incidence of neurological symptoms and deficits in adults with scoliosis at presentation and assess whether these factors were associated with the decision to pursue operative over nonoperative treatment. Neurological symptoms and deficits were frequent at presentation in the 207 patients with scoliosis in the present study. Few prior reports have described the incidence of these findings. In the present series, nearly one-half of patients (47%) had severe radiculopathy, comparable to the 41% incidence recently reported by Glassman et al. 13 The modestly higher incidence in our series may be related to the proportion of patients with a history of AIS, a factor we found to be associated with a lower incidence of severe radicular pain. Although Glassman and colleagues did not specifically report what proportion of their patients had a history of AIS, it is likely that this proportion is substantially greater than that in the present study, given our considerably older patient age (mean 64 years) compared with the mean patient age of 44 years in the study by Glassman et al. 5,13 In addition, the present study is a consecutive series of patients, which may reflect a more accurate assessment of the incidence of symptoms at presentation with scoliosis compared with the casecontrol matched series of Glassman et al. In addition to characterizing the incidence and severity of radicular pain, our data suggest factors that are significantly associated with its development including scoliosis etiology (adult untreated AIS versus adult de novo) and lesser magnitude of lumbar lordosis. The significant association of adult untreated AIS with a lower incidence of leg pain may reflect a different pathobiological process (fewer degenerative changes such as canal or foraminalstenosis) or may simply reflect the significantly younger mean age of these patients in the present series. Consistent with our finding that the only radiographic parameter correlating with the presence of leg pain was a lesser magnitude of lumbar lordosis, Schwab and colleagues 21 recently reported that the only radiographic features correlated with self-reported pain were lumbar parameters in 98 adults with scoliosis. The incidence of radicular weakness in adults with scoliosis at presentation has not been well-characterized. In the present series, we identified 8% of patients with TABLE 4 Multivariate predictors of operative versus nonoperative management in adults with scoliosis* Parameter OR 95% CI p Value radicular weakness (4 5) 15.0 4.0 56.4 0.001 severe radicular pain (VAS score 5) 2.4 1.2 5.0 0.02 sagittal balance (mm) 1.006 1.001 1.011 0.03 * Forward stepwise binary logistic regression. See text for component factors included in the analysis; results of best-fit model are presented. Odds ratio for sagittal balance is per millimeter. Abbreviations: CI = confidence interval; OR = odds ratio. objective radicular weakness ( 4 on a 5-point scale). Remarkably, the only clinical or radiographic factor associated with the presence of radicular weakness was the loss of lumbar lordosis. The choice of operative versus nonoperative scoliosis management in adults is complex and multifactorial. Consideration must be given to the overall health of the patient, the degree to which the deformity impacts function and quality of life, and the nature of the surgery that would be required and whether it would be expected to provide adequate benefits to warrant the risks. In the absence of progressive neurological deficits, the ultimate decision typically rests with the individual patient. A better understanding of the factors that contribute to this decision would benefit both the patient and the surgeon. In the present study, a multivariate model was created using several clinical and radiographic component factors. The best-fit model included only 3 of these: severe radiculopathy, radicular weakness, and greater sagittal imbalance. Notably, the presence of severe back pain was not part of the best-fit model, suggesting that the presence of severe radiculopathy has a stronger association with elective operative treatment. It is not surprising that radicular weakness and positive sagittal imbalance would be predictive of a patient's choice to undergo surgical treatment because the former would obviously impair function and the importance of the latter to quality of life has been previously demonstrated. 9 The primary strengths of this study are the relatively large sample size, the relative homogeneity of the population (all adults with scoliosis), the quantitative assessment of pain using the VAS, and the objective assessment of disability using the ODI. It is also important to recognize the limitations of this study. First, it should be recognized that this is not a cross-sectional study. Patients referred to our clinic have had reason to come to medical attention, have typically undergone some type of prior nonoperative treatment, and have been referred to us for surgical evaluation. The incidence of neurological symptoms and other findings might differ if patients managed by primary care physicians were included, since these patients may not yet have findings warranting referral to a surgeon. Second, although the data were extracted from a prospectively collected database on adult deformity, the present study is retrospective in design. Third, the present study does not include specific questioning of patients and surgeons as to the factors most important in the decision-making 330 J. Neurosurg.: Spine / Volume 9 / October 2008

Neurological symptoms and deficits in adult scoliosis process between operative and nonoperative treatment. Thus this study does not provide a complete model of potential factors involved in the choice of operative versus nonoperative treatment. Other factors not accounted for in the present study are undoubtedly involved in the decision process. Fourth, our emphasis is on findings at presentation and on initial management decisions. We did not analyze the surgical outcome or the long-term clinical outcome of conservative therapy in these patients; this will need to be evaluated in subsequent studies. Conclusions Neurological symptoms and deficits are common among adults with scoliosis and demonstrate a strong association with the decision to pursue operative treatment. Careful assessment and consideration of neurological symptoms and findings should be part of the decision making process in adults with scoliosis. Disclaimer The authors report no conflict of interest concerning the materials or methods used in this study or the findings specified in this paper. References 1. Aebi M: The adult scoliosis. Eur Spine J 14:925 948, 2005 2. Albert TJ, Purtill J, Mesa J, McIntosh T, Balderston RA: Health outcome assessment before and after adult deformity surgery. A prospective study. Spine 20:2005, 1995 3. Berven S, Deviren V, Demir-Deviren S, Hu SS, Bradford DS: Studies in the modified Scoliosis Research Society Outcomes Instrument in adults: validation, reliability, and discriminatory capacity. Spine 28:2164 2169, 2003 4. Bradford DS, Tay BK, Hu SS: Adult scoliosis: surgical indications, operative management, complications, and outcomes. Spine 24:2617 2629, 1999 5. Bridwell KH, Berven S, Edwards C II, Glassman S, Hamill C, Schwab F: The problems and limitations of applying evidence-based medicine to primary surgical treatment of adult spinal deformity. Spine 32:S135 S139, 2007 6. Bridwell KH, Berven S, Glassman S, Hamill C, Horton WC III, Lenke LG, et al: Is the SRS-22 instrument responsive to change in adult scoliosis patients having primary spinal deformity surgery? Spine 32:2220 2225, 2007 7. Daubs MD, Lenke LG, Cheh G, Stobbs G, Bridwell KH: Adult spinal deformity surgery: complications and outcomes in patients over age 60. Spine 32:2238 2244, 2007 8. Fleming KC, Evans JM, Chutka DS: Caregiver and clinician shortages in an aging nation. Mayo Clin Proc 78:1026 1040, 2003 9. Glassman SD, Berven S, Bridwell K, Horton W, Dimar JR: Correlation of radiographic parameters and clinical symptoms in adult scoliosis. Spine 30:682 688, 2005 10. Glassman SD, Berven S, Kostuik J, Dimar JR, Horton WC, Bridwell K: Nonsurgical resource utilization in adult spinal deformity. Spine 31:941 947, 2006 11. Glassman SD, Bridwell K, Dimar JR, Horton W, Berven S, Schwab F: The impact of positive sagittal balance in adult spinal deformity. Spine 30:2024 2029, 2005 12. Glassman SD, Hamill CL, Bridwell KH, Schwab FJ, Dimar JR, Lowe TG: The impact of perioperative complications on clinical outcome in adult deformity surgery. Spine 32: 2764 2770, 2007 13. Glassman SD, Schwab FJ, Bridwell KH, Ondra SL, Berven S, Lenke LG: The selection of operative versus nonoperative treatment in patients with adult scoliosis. Spine 32:93 97, 2007 14. Grubb SA, Lipscomb HJ, Suh PB: Results of surgical treatment of painful adult scoliosis. Spine 19:1619 1627, 1994 15. Heary RF, Albert TJ: Overview of spinal deformity, in Heary RF, Albert TJ (eds): Spinal Deformity: The Essentials. New York: Thieme, 2007, pp 3 11 16. Kanter AS, Asthagiri AR, Shaffrey CI: Aging spine: challenges and emerging techniques. Clin Neurosurg 54:10 18, 2007 17. Lenke LG, Betz RR, Harms J, Bridwell KH, Clements DH, Lowe TG, et al: Adolescent idiopathic scoliosis: a new classification to determine extent of spinal arthrodesis. J Bone Joint Surg Am 83:1169 1181, 2001 18. Puno RM, An KC, Puno RL, Jacob A, Chung SS: Treatment recommendations for idiopathic scoliosis: an assessment of the Lenke classification. Spine 28:2102 2105, 2003 19. Schwab F, Dubey A, Gamez L, El Fegoun AB, Hwang K, Pagala M, et al: Adult scoliosis: prevalence, SF-36, and nutritional parameters in an elderly volunteer population. Spine 30: 1082 1085, 2005 20. Schwab F, Dubey A, Pagala M, Gamez L, Farcy JP: Adult scoliosis: a health assessment analysis by SF-36. Spine 28: 602 606, 2003 21. Schwab F, el-fegoun AB, Gamez L, Goodman H, Farcy JP: A lumbar classification of scoliosis in the adult patient: preliminary approach. Spine 30:1670 1673, 2005 22. Schwab FJ, Smith VA, Biserni M, Gamez L, Farcy JP, Pagala M: Adult scoliosis: a quantitative radiographic and clinical analysis. Spine 27:387 392, 2002 23. Smith JS, Abel MF, Shaffrey CI, Ames CP: Decision making in pediatric spinal deformity. Neurosurgery [in press], 2008 24. Smith JS, Shaffrey CI, Kuntz CI, Mummaneni PV: Classification systems for adolescent and adult scoliosis. Neurosurgery [in press], 2008 25. Wiggins GC, Shaffrey CI, Abel MF, Menezes AH: Pediatric spinal deformities. Neurosurg Focus 14(1):E3, 2003 Manuscript submitted January 24, 2008. Accepted July 21, 2008. Address correspondence to: Christopher I. Shaffrey, M.D., De part ment of Neurosurgery, University of Virginia, P.O. Box 800212, Charlottesville, Virginia 22908. email: CIS8Z@hscmail. mcc.virginia.edu. J. Neurosurg.: Spine / Volume 9 / October 2008 331