Adult spinal deformity (ASD) is associated with pain and disability including difficulties with standing, walking, performing personal hygiene, and pursuing social activities.1–7 In the past decade, spinal alignment has drawn considerable focus.1–5,8,9 Targets for ASD correction have been proposed in line with studies that have reported correlations between alignment parameters and health-related qualify of life measures (HRQOL).3,5,7 However, investigators are beginning to realize that a “one size fits all” approach to alignment in ASD corrections is an oversimplification in spite of the fact that the field of spine surgery and orthopaedics in general are amenable to measurements and guidelines.10,11 Age is one variable in particular that has gained recent attention as a factor that has to be considered when defining optimal alignment in ASD patients.10,11 Elderly patients, essentially those who are more frail, do not require as rigorous spinal alignments when undergoing ASD corrections as younger more robust patients.10 Overcorrecting the elderly can lead to mechanical complications such as proximal junctional kyphosis without any incremental improvement in health outcomes.11
Many authors have recognized the importance of pelvic morphology in determining spinal curvatures.4,5,9,12 The pelvic incidence (PI) parameter is the most widely utilized description of pelvic shape and it sets the foundation for spinal morphology, determining the spinal curvatures of adjacent spinal regions.4,5 While lumbar lordosis (LL) has been correlated to pelvic incidence and pelvic incidence minus lumbar lordosis (PI-LL) mismatch has been utilized to determine the gap in lordosis in lumbar flatback deformity, no study has determined if sagittal spinal alignment targets in ASD correction should depend on pelvic morphology.
The purpose of this study is to investigate sagittal spinal alignment among asymptomatic subjects with varying degrees of pelvic incidence and to determine if targets for deformity correction in ASD patients should vary by pelvic incidence as well as age.
MATERIALS AND METHODS
This study is a multicenter, prospective data analysis of patients with adult spinal deformity conducted through the International Spine Study Group, a collaboration of spine surgeons from 11 sites across the United States. Institutional Review Board approval was obtained at each site for the patient enrollment and data collection protocols. Inclusion criteria for enrollment were age > 18 years and a radiographic diagnosis of ASD defined as at least one of the following: coronal Cobb angle ≥ 20°, sagittal vertical axis (SVA) ≥ 5 cm, pelvic tilt (PT) ≥ 25°, or thoracic kyphosis ≥ 60°. For the present study, all subjects had 36 inch standing scoliosis x-rays and HRQOL data available at baseline.
Radiographic Analysis and Data Collection
Data collection at baseline included standardized HRQOL questionnaires, as well as clinical, demographic, and radiographic information. Basic demographic and clinical data included patient age, sex, body mass index, and Charlson comorbidity index.13
All subjects had 36 inch standing scoliosis x-rays for which the patients were free of any external support such as walkers or hanging bars. All radiographic measurements were performed at a central location using standard techniques for established parameters. Radiographic analysis was performed on baseline x-rays using a dedicated and validated software (Spineview, ENSAM, Paris).14,15 Radiographic parameters recorded included T1 pelvic angle (TPA) (Figure 1), SVA, PT, and PI − LL.
Health-related quality of life (HRQOL) assessment tools included the Oswestry Disability Index (ODI), Scoliosis Research Society (SRS)-22, and Short Form (SF)-36 questionnaires. Two standard summary scores were calculated based on the Short Form 36 Survey (SF-36), the Physical Component Score (PCS), and the Mental Component Score. The SRS-22 provides a summary score and multiple subdomains, including activity, pain, appearance, mental, and satisfaction.
Age-specific Alignment Targets
In the ASD cohort, correlations were assessed within the radiographic parameters, and between radiographic parameters and HRQOL scores. Linear regression analysis was performed on the baseline alignment and HRQOL data. Published United States SF-36 age-specific normative data was used to establish optimal alignment. Linear regression analyses were performed with PI, age, and HRQL as independent parameters and the radiographic alignment measurements as dependent parameters. Using these regressions with SF-36 physical component score (SF-36 PCS) data, targets for optimal correction for different PI magnitudes and age groups were determined from the radiographic alignments. Additionally, using the ODI data, the values of the alignment variables corresponding to moderate disability (ODI = 20) and severe disability (ODI = 40) were calculated.
Asymptomatic Subject Analysis
A cohort of asymptomatic subjects was analyzed. These subjects had no back pain, neck pain, or complaints related to postural deformity. Correlations were assessed between alignment, age, and pelvic incidence. Linear regression analysis was performed on their TPA, age, and pelvic incidence to determine normative TPA values for different pelvic incident and age categories.
Adult Spinal Deformity Patient Population
A total of 903 ASD patients (mean age 59.8 yrs) were included (Table 1). Patients were subanalyzed by PI: low, medium, high (<40°, 40°–75°, >75°); and age: elderly (>65 yr, n = 375), middle age (MA 45–65 yr n = 387), and young (18–45 yr, n = 141).
TPA correlated with age and PI in ASD and normative subjects (r>0.42, P < 0.0001). ODI correlated with SF36 PCS (r = 0.71, P < 0.0001). TPA was correlated with SVA (r = 0.837), PT (r = 0.933), PI – LL (r = 0.889), and T1SPi (r = 0.589). In terms of HRQOL, TPA correlated with SF-36 physical component (r = −0.445), ODI (r = 0.435), and SRS-22 total (r = −0.358) scores. These associations were similar to those for SVA, PT, PI-LL, and T1SPI. All the reported correlations were significant at the P < 0.01 level.
The linear relationship between TPA, PI, and SF-36 PCS was used to generate a linear regression model (r = 0.53, r2 = 0.28, P < 0.001) leading to the following equation:
This linear regression was used to generate TPA values corresponding to US normative values of SF-36 PCS for a full range of ages and pelvic incidence values (Table 2).
Using the aforementioned linear regression and the ODI data, the TPA values corresponding to low disability (ODI>20) and severe disability (ODI > 40) were determined for each PI category (Table 3). An ODI of 20 corresponded to a TPA of 9.2° for low PI patients, 18.0° for medium PI patients, and 29.1° for high PI patients (Table 3).
The linear relationship between PI, SF-36 physical component scores and the other SRS-Schwab parameters for ASD classification was used to generate linear regression models for SVA, PT, and PI-LL leading to the following three equations:
The linear regression model for SVA was used to create SVA targets for different PI and age categories using the normative age-specific SF-36 physical component scores (Table 4). Additionally, alignments corresponding to low and severe disability were calculated for the different PI categories (Table 3).
Asymptomatic Subject Analysis
Among the 111 asymptomatic subjects (mean age 50.7 yr), TPA correlated with PI (r = 0.60, P < 0.001) and age (r = 0.49, P < 0.001). The linear relationship between TPA, PI, and age was used to generate a linear regression model leading to the following equation:
This linear regression was used to calculate normative alignments for different age and PI categories (Table 5).
Preoperative planning of adult spinal deformity has been shown to improve postoperative radiographic results and health-related quality of life outcomes.16 It is convenient to have a simplified, easy-to-remember, set of rules when planning ASD surgery; however, in practice, every patient presents a distinct challenge and the goals of spinal deformity surgery may not always fit into a simplified set of realignment targets.3,5,10,11 For example, there are patients who have significant clinical improvement following ASD surgery despite not meeting “optimal” alignment criteria; and conversely there are patients who may seem adequately aligned but may have residual functional limitations. The results of this study demonstrate that the discrepancy in the surgical outcomes of such patients may lie in the fact that pelvic incidence was not considered when assessing the adequacy of spinal alignment.
A similar discrepancy has been recognized once studies identified that realignment targets should depend on age.10,11 In a multicenter study on ASD surgery, Schwab et al17 reported that 23% of patients undergoing deformity correction were not optimally aligned; however, that investigation predated our understanding of age-specific alignment. The failure to bring patients into acceptable alignment could be attributable to the clinical acumen of the surgeons in recognizing that elderly patients do not need to be brought into the same alignment as their younger counterparts.
Other authors have recognized the importance of pelvic morphology in determining spinal curvatures and shapes.4,5,12 Dubousset favored the term, “pelvic vertebra” when describing its relation to the spine and lower limbs.18 Roussouly published a classification of spinal curvatures and one essential feature in distinguishing the different types was the magnitude of the pelvic incidence.19 However, no study to date has reported how pelvic incidence can affect specific sagittal spinal alignment measurements corresponding to normative age-specific health status.
This study utilized published age-specific normative values for the SF-36 PCS and data from a large multicenter database of ASD patients to determine the alignments that would correspond to the expected function status of age-matched peers. The SF-36 is health outcome measure that has been utilized to quantify health status in many different diseases. Since it is not specific to any particular pathology, it can serve as a comparative measure of functional status between various diseases.6 Furthermore, since normative data is available for subjects of different ages, the SF-36 can be used to compare patients with a specific pathology, in this case ASD, to those of their age-matched asymptomatic peers. This study affirms prior studies that demonstrated that alignment parameters in ASD patients correlate to functional status (SF-36 PCS) and age.10 However, the novel aspect of this investigation is that it demonstrates that there is also a strong correlation with a third factor, pelvic incidence. As can be seen in Tables 2–4, when pelvic incidence is factored into our calculation of optimal alignments that correspond to normative functional status, the values vary by more than 50% within each age category. For example, the optimal TPA alignment for an elderly patient with a low pelvic incidence is 11.9° whereas an elderly patient with a high pelvic incidence has an optimal TPA of 33.6° (Table 2). The fact that sagittal spinal alignment varies by age and pelvic incidence was also demonstrated in the asymptomatic subject analysis. The model correlating TPA alignment with age and PI among asymptomatic subjects had an r = 0.72, demonstrating a strong association with these parameters. Comparing Tables 2–5, optimal TPA alignments for ASD patients are sagittally more forward than those of asymptomatic subjects. In using this data to guide surgical planning, alignment targets for TPA, SVA, PI-LL, and PT should match the optimal alignments from the ASD analysis since the asymptomatic patient data would lead to overcorrection without any improvement in functional outcome but with a greater risk of mechanical complications such as proximal junctional kyphosis as demonstrated by Lafage et al11 (Figures 2A–C, 3A–C).
Other authors have reported that patients often do not correct their pelvic tilt following ASD surgery.20 This too may result from the fact that the magnitude of pelvic incidence was not considered when assessing optimal alignment. This study shows that among asymptomatic normative subjects, those with high pelvic incidence will stand with a higher pelvic tilt. Moreover, this was confirmed in the analysis of ASD patients which demonstrated that optimal PT varied by pelvic incidence and age.
Similarly, for sagittal spinal alignment, an ASD patient with high pelvic incidence does not require as rigorous a postoperative alignment as one with a low PI (Figures 2 and 3). In spinal deformity planning this is where T1 pelvic angle can be particularly helpful.3 While surgeons can simulate how a particular osteotomy can affect SVA and PT, it is difficult to determine definitively how a patient will stand postoperatively in terms of their pelvic retroversion (PT) and their upper thoracic inclination (SVA).21 However, TPA describes the geometry of the underlying global spinopelvic alignment.3 This more directly measures the geometry that is modified during spinal realignment surgery.3 TPA and its component angles have been shown to be effective intraoperative tools to confirm the adequacy of sagittal spinal deformity correction.22,23
Historically, thresholds for various spinal alignment parameters have been based on average values observed in a heterogeneous population. With the technological advancements in imaging and surgical planning software, it is becoming increasingly well recognized that a “one-size-fits-all” approach to adult spinal deformity surgery is not appropriate. The current study evaluates the effects of PI and age on alignment. These results from this study reaffirm the conclusions of previous studies, finding that older adults require less aggressive corrections than younger patients. This study also demonstrated that PI is significantly associated with global sagittal alignment as well as spinopelvic parameters and as such, alignment targets in ASD surgery should not only account for age, but also PI. Similar to older patients, a patient with a larger PI requires a less rigorous alignment than a patient with a smaller PI. While the formulas presented in this study are useful for calculating precise alignment targets in ASD surgical planning, it is unrealistic and often unnecessary to perform these calculations for every patient in a clinical setting. Therefore, targets for various alignment parameters are provided for different age and PI categories that can be more easily utilized when evaluating a patient's spinal alignment.
- Optimal sagittal alignment targets differ for each patient. Previous research has shown that alignment targets for adult spinal deformity correction are age specific. The current study demonstrates that alignment targets should account for pelvic incidence as well as age.
- Using age-normalized SF-36 physical component scores and pelvic incidence, optimal alignment targets can be defined for T1-pelvic angle pelvic tilt, pelvic incidence-lumbar lordosis mismatch, and sagittal vertical axis.
- Elderly patients and patients with high pelvic incidence require less rigorous sagittal alignment to attain age-specific normative levels of health status.
1. Lafage V, Schwab FJ, Patel A, et al. Pelvic tilt and truncal inclination: two key radiographic parameters in the setting of adults with spinal deformity. Spine (Phila Pa 1976)
2. Glassman SD, Bridwell KM, Dimar JR, et al. The impact of positive sagittal balance in adult spinal deformity
. Spine (Phila Pa 1976)
3. Protopsaltis T, Schwab F, Bronsard N, et al. TheT1 pelvic angle, a novel radiographic measure of global sagittal deformity, accounts for both spinal inclination and pelvic tilt and correlates with health-related quality of life. J Bone Joint Surg Am
4. Legaye J, Duval-Beaupère G, Hecquet J, et al. Pelvic incidence: a fundamental pelvic parameter for three-dimensional regulation of spinal sagittal curves. Eur Spine J
5. Terran J, Schwab FJ, Shaffrey CI, et al. The SRS-schwab adult spinal deformity
classification: assessment and clinical correlations based on a prospective operative and nonoperative cohort. Neurosurgery
6. Fu K-MG, Bess RS, Schwab FJ, et al. Health impact comparison of different disease states and population norms to adult spinal deformity
(ASD): a call for medical attention. Spine J
7. Schwab FJ, Dubey A, Gamez L, et al. Adult scoliosis: prevalence, SF-36, and nutritional parameters in an elderly volunteer population. Spine (Phila Pa 1976)
8. Smith JS, Singh M, Klineberg E, et al. Surgical treatment of pathological loss of lumbar lordosis (flatback) in patients with normal sagittal vertical axis achieves similar clinical improvement as surgical treatment of elevated sagittal vertical axis. J Neurosurg Spine
9. Vialle R, Levassor N, Rillardon L, et al. Radiographic analysis of the sagittal alignment and balance of the spine in asymptomatic subjects. J Bone Joint Surg Am
10. Lafage R, Schwab F, Challier V, et al. Defining spino-pelvic alignment thresholds: should operative goals in adult spinal deformity surgery
account for age? Spine (Phila Pa 1976)
11. Lafage R, Schwab F, Glassman S, et al. Age-adjusted alignment goals have the potential to reduce PJK. Spine (Phila Pa 1976)
12. Roussouly P, Nnadi C. Sagittal plane deformity: an overview of interpretation and management. Eur Spine J
13. Charlson ME, Pompei P, Ales KL, et al. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis
14. Champain S, Benchikh K, Nogier A, et al. Validation of new clinical quantitative analysis software applicable in spine orthopaedic studies. Eur Spine J
15. Rillardon L, Levassor N, Guigui P, et al. [Validation of a tool to measure pelvic and spinal parameters of sagittal balance]. Rev Chir Orthop Reparatrice Appar Mot
16. Gupta M, Henry JK, Schwab F, et al. Dedicated spine measurement software quantifies key spino-pelvic parameters more reliably than traditional picture archiving and communication systems tools. Spine (Phila Pa 1976)
17. Schwab FJ, Patel A, Shaffrey CI, et al. Sagittal realignment failures following pedicle subtraction osteotomy surgery
: are we doing enough?: clinical article. J Neurosurg Spine
18. Skalli W, Zeller RD, Miladi L, et al. Importance of pelvic compensation in posture and motion after posterior spinal fusion using CD instrumentation for idiopathic scoliosis. Spine (Phila Pa 1976)
19. Laouissat F, Sebaaly A, Gehrchen M, et al. Classification of normal sagittal spine alignment: refounding the Roussouly classification. Eur Spine J
20. Lafage V, Blondel B, Smith JS, et al. Preoperative planning for pedicle subtraction osteotomy: does pelvic tilt matter? Spine Deform
21. Diebo BG, Lafage R, Ames CP, et al. Ratio of lumbar 3-column osteotomy closure: patient-specific deformity characteristics and level of resection impact correction of truncal versus pelvic compensation. Eur Spine J
22. Oren JH, Tishelman JC, Day LM, et al. Measurement of spinopelvic angles on prone intraoperative long-cassette lateral radiographs predicts postoperative standing global alignment in adult spinal deformity surgery
. Spine Deform
23. Vaynrub M, Hirsch BP, Tishelman J, et al. Validation of prone intraoperative measurements of global spinal alignment. J Neurosurg Spine