The Spine Journal 15 (2015) 1000–1020

Basic Science

Age- and gender-related changes in pediatric thoracic vertebral morphology James R. Peters, BSa, Charanya Chandrasekaran, BSa, Lucy F. Robinson, PhDb, Sabah E. Servaes, MDc, Robert M. Campbell, Jr., MDd, Sriram Balasubramanian, PhDa,* b

a School of Biomedical Engineering, Science and Health Systems, Drexel University, 3141 Chestnut St, Bossone 718, Philadelphia, PA 19104, USA Department of Epidemiology and Biostatistics, School of Public Health, Drexel University, Nesbitt Hall, 3215 Market St. Philadelphia, PA 19104, USA c Department of Radiology, The Children’s Hospital of Philadelphia, 34th St and Civic Center Boulevard, Philadelphia, PA 19104, USA d Division of Orthopaedic Surgery, The Children’s Hospital of Philadelphia, 34th St and Civic Center Boulevard, Philadelphia, PA 19104, USA

Received 9 October 2014; revised 8 December 2014; accepted 10 January 2015

Abstract

BACKGROUND CONTEXT: Although it is well known that the growth of thoracic spine changes significantly with age, gender, and vertebral level in the skeletally normal pediatric population, there have been very few studies attempting to comprehensively quantify such variations. Biomechanical and computational models of the growing thoracic spine have provided insight into safety and efficacy of surgical and noninvasive treatments for spinal deformity. However, many of these models only consider growth of the vertebral body and pedicles and assume a consistent growth rate for these structures across thoracic levels. PURPOSE: To enhance the understanding of age-, gender-, and level-related growth dynamics of the pediatric thoracic spine by comprehensively quantifying the thoracic vertebral morphology for subjects between 1 and 19 years. STUDY DESIGN: A retrospective computed tomography (CT) image analysis study. METHODS: Retrospectively obtained chest CT scans from 100 skeletally normal pediatric subjects (45 males and 55 females between the ages 1 and 19 years) were digitally reconstructed using medical imaging software. Surface point clouds of thoracic vertebrae were extracted and 26 vertebral geometry parameters were measured using 25 semiautomatically identified surface landmarks and anatomical slices from each thoracic vertebra (T1–T12). Data were assessed for normality, symmetry, and age-, gender-, and level-related differences in geometric measures and growth. Linear regression was performed to estimate of the rates of variation with age for each measurement. RESULTS: Asymmetries (bilateral, superior-inferior, and anteroposterior) were observed in vertebral body heights, end plate widths and depths, and interfacet widths. Within genders, significant interlevel differences were observed for all geometric measures, and significant differences in the rates of growth were found across thoracic levels for most parameters. Significant differences were observed between genders for pedicle, spinous process, and facet measurements. Growth rates of the pedicles and vertebral bodies were also found to vary significantly between genders. CONCLUSIONS: The rates of growth for most thoracic vertebral structures varied between genders and across vertebral levels. These growth rates followed trends similar to those of their associated vertebral dimensions and this indicates that, across levels and between genders, larger vertebral structures grow at faster rates, whereas smaller structures grow at a slower rate. Such

FDA device/drug status: Not applicable. Author disclosures: JRP: Nothing to disclose. CC: Nothing to disclose. LFR: Nothing to disclose. SES: Nothing to disclose. RMC: Nothing to disclose. SB: Nothing to disclose. All authors were fully involved in the study and preparation of this manuscript. The material within has not been and will not be submitted for publication elsewhere. This study was partly funded by the Scoliosis http://dx.doi.org/10.1016/j.spinee.2015.01.016 1529-9430/Ó 2015 Elsevier Inc. All rights reserved.

Research Society-New Investigator research grant. The authors do not have anything to disclose. * Corresponding author. School of Biomedical Engineering, Science and Health Systems, Drexel University, 3141 Chestnut St, Bossone 718, Philadelphia, PA 19104, USA. Tel.: (1) 215-571-3738; fax: (1) 215-8954983. E-mail address: [email protected] (S. Balasubramanian)

J.R. Peters et al. / The Spine Journal 15 (2015) 1000–1020

1001

level- and gender-specific information could be used to inform clinical decisions about spinal deformity treatment and adapted for use in biomechanical and computational modeling of thoracic growth and growth modulation. Ó 2015 Elsevier Inc. All rights reserved. Keywords:

Pediatric; Spine; Morphology; Vertebra; Growth; Thoracic

Introduction For the skeletally normal pediatric population, it is known that growth of the spine varies with age and gender and that such variations in spinal growth may influence the progression of spinal deformities, confounding treatment outcomes [1–8]. Growth sparing spine instrumentation, in contrast to fusion-based techniques, guides the growth of the spine through periodic distraction and is the preferred treatment method for correcting thoracic spinal deformity in pediatric subjects. This preference can be attributed to a combination of effeteness and the instrumentation’s preservation of subjects’ mobility, the natural anatomy [9–12].

Current trends indicate that the future direction for spine deformity correction will rely on growth modification. This will require the study and characterization of normal growth to inform more effective growth modification strategies and enable better device design. Because of the lack of pediatric cadaveric material, current growth sparing spine instrumentations are commonly evaluated using biomechanical and computational models, which often make the assumptions of scaled material properties and a nongrowing thoracic geometry [13–29]. Thorough quantitative evaluations of adult thoracic vertebral morphology report detailed geometric measures of the end plates, vertebral bodies, pedicles, spinal canals,

Fig. 1. All LMPs displayed in (A) x-y (transverse) plane, (B) x-z (sagittal) plane, and (C) y-z (coronal) plane. (D) Vertebral body LMPs 1 to 9 and anatomical slices (Slice 1–2). (E) Spinal canal (LMP10), facet (LMPs 19–22), and transverse and spinous process LMPs (23–25). (F) Right pedicle LMPs (11–14) and anatomical slice (Slice 3). (G) Left pedicle LMPs (15–18) and anatomical slice (Slice 4). (H) Spinal canal anatomical slice (Slice 5). LMP, landmark point.

1002

J.R. Peters et al. / The Spine Journal 15 (2015) 1000–1020

Fig. 2. Thoracic vertebrae geometric measures—VBs, EPs, PDs, SC, intervertebral discs, SP and IT process, and IF dimensions. Height (H), width (W), depth (D), length (L), area (A), angle (q), anterior (a), posterior (p), superior (s), inferior (i), right (r), and left (l). VB, vertebral body; EP, end plate; PD, pedicle; SC, spinal canal; SP, spinous process; IT, intertransverse; IF, interfacet.

processes, and facets in skeletally normal adult subjects [30–35]. However, the existing literature on pediatric thoracic spine morphology is either restricted to geometric characterization of a small subset of vertebral structures or extensive quantification of pedicle morphology alone [36–43]. Although the morphologic studies by Taylor [39], Lord et al. [36], and Masharawi et al. [37] collectively report measurements from the vertebral bodies, spinous and transverse processes, and facets of pediatric subjects, possible gender- and level-related differences and rates of dimensional variation with age (growth) were not quantitatively assessed. In addition, there is little overlap between the age ranges considered in each study. The thoracic pedicle morphology studies by Taylor [40], Rajwani et al. [38], Zindrick et al. [43], Zheng et al. [42], and Tian et al. [41] provide a very thorough description of the pedicle and spinal canal dimensions in pediatric subjects and report differences with age and between genders; however, no attempts were made to describe the interlevel differences or quantify growth of the vertebral structures. Computational models that include growth and modulation were introduced by Stokes and Laible [44] to study the progression of spinal deformity. These models work on the Hueter-Volkmann principle which states that the cycle of growth and deformation (progression), in bone, can be partly attributed to an asymmetric distribution of forces [44–46]. Subsequent applications of these methods to study deformity progression in scoliosis and the effects of fusionless spine instrumentation on pediatric vertebral morphology have been limited by a narrow focus on vertebral body and pedicle growth. In addition, there is an underlying assumption that

the basis growth strain rate (growth rate) is consistent across thoracic levels with no gender differentiation [44,47–53]. Because the current literature lacks a comprehensive data set on pediatric vertebral morphology and vertebral growth, an indepth analysis is required to enhance our understanding of the dynamics of thoracic spinal growth [26]. Hence, the objectives of this study were to comprehensively quantify the three-dimensional (3D) surface morphology of the vertebral bodies, pedicles, spinal canal, processes, and facets of the thoracic vertebrae in pediatric male and female subjects between 1 and 19 years, to assess symmetry (bilateral, anteroposterior, and superior-inferior), to provide estimates of rates of variation with age for these measurements at all thoracic vertebral levels and, to evaluate age-, gender-, and level-related differences in geometric parameter values and their rates of growth.

Materials and methods Data collection This study protocol was reviewed and of the respective institutions. Retrospectively obtained chest computed tomography (CT) scans from 100 skeletally normal subjects (45 males, average age: 9.866 years, range: 1.1–18.62 years and 55 females, average age: 9.8465.91 years, range: 1.2–18.78 years) were obtained from the Department of Radiology at The Children’s Hospital of Philadelphia. All subjects were chosen to be within 5th and 95th percentiles in height, weight, and body mass index, as determined by Centers for Disease Control and Prevention (CDC) growth

J.R. Peters et al. / The Spine Journal 15 (2015) 1000–1020

1003

Table 1 Dimensions for the VBs, EPs, PDs, SC, SP and IT process, F, and IF of the thoracic spine Structure

Abbreviation

Dimension

VB and EP

VBH a VBH p EPW s EPiW EPD s EPD i PDH r PDH l PDW r PDW l PDA r PDA l PDqr PDql SCW SCD SCA SPL SPq ITW

VB anterior height VB posterior height EP superior width EP inferior width EP superior depth EP inferior depth Right PD height Left PD height Right PD width Left PD width Right PD area Left PD area Right PD angle Left PD angle SC width SC depth SC area SP length SP angle Transverse process width Right IF height Left IF height Superior IF width Inferior IF width Right facet angle Left facet angle

PD

SC

Processes

Articular facet

IFH r IFH l IFW s IFiW Fqr Fql

Measurement points LMP3, LMP4 LMP2, LMP5 LMP6, LMP7 LMP8, LMP9 LMP2, LMP5 LMP4, LMP5 LMP15, LMP17 LMP11, LMP13 LMP16, LMP18 LMP12, LMP14 Slice 4 Slice 3 Slice 4 Slice 3 Slice 5 Slice 5 Slice 5 LMP1, LMP25 LMP1, LMP25 LMP23, LMP24 LMP19, LMP20, LMP19, LMP21, LMP10, LMP10,

LMP22 LMP20 LMP20 LMP22 LMP22 LMP21

VB, vertebral body; EP, end plate; PD, pedicle; SC, spinal canal; SP, spinous process; IT, intertransverse; F, facet; IF, interfacet; LMP, landmark point. Note: Superscripts: height (H), width (W), depth (D), length (L), area (A), and angle (q). Subscripts: anterior (a), posterior (p), superior (s), inferior (i), right (r), and left (l). Graphical depiction of measurements shown in the table (placement of LMPs and anatomical slices shown in Fig. 1).

charts for children and by CDC National Health and Nutrition Examination Survey (CDC NHANES) data for subjects 18 years and older [54,55]. Computed tomography scans with complete thoracic spines (T1–T12) and axial slice thickness of up to 5 mm with an in-plane resolution of 0.658
0.658 mm were considered for analysis. Subject CTs were digitally reconstructed using the medical image processing software MIMICS (Materialise, Inc., Belgium), surface point clouds of thoracic vertebral geometries were extracted, and the 3D morphology of all vertebrae were analyzed using custom MATLAB (The MathWorks, Inc, Natick, MA, USA) codes. A total of 25 surface landmark points (LMPs) and 5 anatomical slices were semiautomatically identified and used to measure 26 vertebral geometry dimensions from each thoracic vertebra (Figs. 1–2). All measured vertebral geometry dimensions, their abbreviations, associated LMPs, and slices are provided in Table 1. The LMPs used in this study are comparable with those previously reported in the literature for geometric quantification, statistical modeling, and 3D reconstruction of the spine [30–32,37,56–59].

Fig. 3. Scatter plot showing male and female T5 anterior vertebral body heights as a function of age and their respective linear regressions (black, male and gray, female). The slopes of these regression lines represent the rates of growth for the parameter (mm/y).

Statistical analysis Based on non-normality of the data determined by Kolmogorov-Smirnov tests, nonparametric statistics were used. Levene tests of parameter ranks were used to determine the equality of variance within and between gender groups. For each gender, Friedman tests were used to compare the geometric measures across levels, and Wilcoxon ranked-sign tests were used to compare parameters between levels in addition to assessing symmetry in all subjects. Mann-Whitney U and median tests were conducted to compare all parameters between genders, and Spearman rho correlations were used to assess the correlation of parameters with age at all vertebral levels for both genders. Linear regression models were created using age as the independent variable to estimate parameter growth rates. For each parameter, the growth rate was assessed as the slope of the linear regression model at each thoracic level individually. An example scatter plot showing the T5 anterior vertebral body height for both genders and their associated linear regression lines is shown in Fig. 3. Analyses of variance performed on multiple regression models were used to evaluate the differences in growth rates across vertebral levels and between genders. All statistics were calculated using SPSS (IBM Corp, Armonk, NY, USA) with a significance level of p!.05, with an appropriate Bonferroni correction factor for each test hypothesis (eg, for symmetry p!.05/[12 levels
20 measurements
2 genders] 5p!.000417).

Results Mean and standard deviations for all thoracic vertebral geometry parameters at each thoracic level for both genders

1004

J.R. Peters et al. / The Spine Journal 15 (2015) 1000–1020

across age were calculated (Figs. 4–8). Interlevel differences were assessed for both genders using Wilcoxon ranked-sign tests (results shown as checkerboard plots, Figs. 9–10). The results of the regression analyses, that is, slopes (growth rates), confidence intervals, intercepts, and proportion of variance explained (R2 values) are shown in Table 2. The following paragraphs describe the withinand between-gender differences for each geometric measure and associated growth rates.

significantly correlated with age (p!.00417) for both genders at all vertebral levels. Significant correlations with age (p!.00417) were observed for right pedicle angles in males from T6–T8 and in females at T6 and T7 and for left pedicle angles from T5–T9 in males and at T6 in females. For both genders, significant growth rate differences (p!.0015) were observed across thoracic vertebral levels for pedicle height, width, and area. No age-related comparisons were performed for bilateral pedicle angles because of the lack of consistent correlations with age across thoracic levels.

Vertebral body and end plate Within gender Significant differences (p!.000417) were observed in anterior versus posterior vertebral body height in male subjects at T2, T5–T7, T9, and T11–T12 and in female subjects at T1–T9 and T11–T12. Significant differences in superior versus inferior end plate width (p!.000417) were also observed for all thoracic levels for male and female subjects. Additionally, significant differences (p!.000417) were observed in superior versus inferior end plate depth in male subjects at T1–T3 and T5 and in female subjects at T1–T6. For all vertebral body and end plate geometry measures, significant correlations with age (p!.00417) were observed for both genders at all vertebral levels. Significant anterior vertebral body height growth rate differences (p!.0015) were seen across thoracic vertebral levels for females and in posterior vertebral body height growth rates for both males and females across all vertebral levels. Significant growth rate differences (p!.0015) were also observed in superior and inferior end plate width and depth measurements across thoracic vertebral levels for both genders. Between genders Vertebral body height and associated growth rates did not significantly vary between male and female subjects. Although no differences were observed in the end plate width and depth between genders, the superior end plate width and depth growth rates significantly varied (p!.00417) between males and females at T9 and significant differences (p!.00417) were also observed in the inferior end plate width and depth growth rates at levels T6 and T9, respectively. Figure 5 shows the vertebral body and end plate dimensions (median values and interquartile distributions) at each thoracic vertebral level for both genders.

Between genders No differences were observed in bilateral pedicle height and associated growth rates between male and female subjects. For all ages, significant differences (p!.00417) were observed between genders in right (T4–T10, T12) and left (T3–T9) pedicle widths. Significant growth rate differences (p!.00417) were also observed in right (T9, T10, and T12) and left pedicle widths (T9). Although bilateral pedicle area and pedicle angle did not vary between male and female subjects, significant differences (p!.00417) in growth rates were found for right pedicle area (T6–T9, T11, and T12) and for left pedicle area at T9. Figure 6 shows the pedicle dimensions (median values and interquartile distributions) at each thoracic vertebral level for all subjects.

Spinal canal Within gender Although spinal canal width significantly correlated with age (p!.00417) for both males and females at all vertebral levels, no differences were observed in growth rates for spinal canal width across vertebral levels. Spinal canal area also significantly correlated with age in males (T1, T3–T6, T8, and T11–T12) and females (T1–T12), with no differences observed in associated growth rates across vertebral levels. Since spinal canal depth was not found to be significantly correlated with age, no age-related comparisons were performed. Between genders No significant differences were observed in spinal canal width, depth, and area or their associated growth rates between males and females. Figure 7 shows the spinal canal dimensions (median values and interquartile distributions) at each thoracic vertebral level for both genders.

Pedicle Within gender No significant differences were observed in bilateral (left vs. right) pedicle height, width, area, or angle in male and female subjects. Pedicle height, width, and area

Processes Within gender Although significant correlations with age (p!.00417) were observed for spinous process length and

J.R. Peters et al. / The Spine Journal 15 (2015) 1000–1020

1005

Fig. 4. Median and interquartile ranges (indicated by box), minimum and maximum values (indicated by whiskers) for thoracic vertebral body, and end plate geometry for male and female pediatric subjects (from birth to 19 years). (A) Anterior vertebral body height, (B) posterior vertebral body height, (C) superior end plate width, (D) inferior end plate width, (E) superior end plate depth, and (F) inferior end plate depth.

intertransverse process width at all thoracic vertebral levels for both genders, significant growth rate differences (p!.0015) were only seen in females. Also, while spinous process angles significantly correlated with age for males (T2–T10) and females (T1–T10 and T12), the growth rates only significantly varied (p!.0015) across the vertebral levels in females.

transverse process dimensions (median values and interquartile distributions) at each thoracic vertebral level for all subjects.

Articular facet Within gender

Between genders No significant differences were observed in spinous process length, intertransverse process width, or their respective growth rates between males and females. Significant differences (p!.00417) were observed in spinous process angles between genders at levels T3 and T12. However, there were no significant differences in spinous process angle growth rates. Figure 8 shows the spinous and

Although no significant bilateral differences (left vs. right) were observed in interfacet height, significant correlations with age (p!.00417) were observed for both males and females at all vertebral levels. Significant differences (p!.000417) were observed between superior and inferior interfacet widths in males (T1 and T4–T9) and females (T1, T3–T9, and T11). Superior and inferior interfacet widths significantly correlated correlations with age (p!.00417) for both genders at all vertebral levels. No

1006

J.R. Peters et al. / The Spine Journal 15 (2015) 1000–1020

Fig. 5. Median and interquartile ranges (indicated by box) and minimum and maximum values (indicated by whiskers) for thoracic vertebral pedicle geometry for male and female pediatric subjects (from birth to 19 years). (A) Left pedicle height, (B) right pedicle height, (C) left pedicle width, (D) right pedicle width, (E) left pedicle area, (F) right pedicle area, (G) left pedicle angle, and (H) right pedicle angle. An * above the thoracic level indicates a significant difference between males and females (p!.00417).

bilateral asymmetries were observed in the facet angle at all thoracic levels for both genders. Significant correlations with age (p!.00417) were observed for right facet angles in males in T9 and T12 and females in T1 and T3, and for left facet angles in males in T9 and T12 and females in T2, respectively.

Significant differences (p!.0015) in bilateral growth rates across thoracic vertebral levels were observed for interfacet height (both males and females) and width (females only). No age-related comparisons were performed for facet angle because of the lack of consistent correlation with age across thoracic levels.

J.R. Peters et al. / The Spine Journal 15 (2015) 1000–1020

1007

Fig. 6. Median and interquartile ranges (indicated by box) and minimum and maximum values (indicated by whiskers) for thoracic spinal canal geometry for male and female pediatric subjects (from birth to 19 years). (Top Left) Spinal canal width, (Top Right) spinal canal depth, and (Bottom) spinal canal area.

Fig. 7. Median and interquartile ranges (indicated by box), minimum and maximum values (indicated by whiskers) for thoracic spinous process, and transverse process geometry for male and female pediatric subjects (from birth to 19 years). (Top Left) Spinous process length, (Top Right) spinous process angle, and (Bottom) intertransverse process width. An * above the thoracic level indicates a significant difference between males and females (p!.00417).

1008

J.R. Peters et al. / The Spine Journal 15 (2015) 1000–1020

Fig. 8. Median and interquartile ranges (indicated by box), minimum and maximum values (indicated by whiskers) for thoracic vertebral facet geometry for male and female pediatric subjects (from birth to 19 years). (A) Left interfacet height, (B) right interfacet height, (C) superior interfacet width, (D) inferior interfacet width, (E) left facet angle, and (F) right facet angle. An * above the thoracic level indicates a significant difference between males and females (p!.00417).

Between genders No significant differences were observed in interfacet height, width, or their respective growth rates between males and females. Significant differences (p!.00417) were observed in right facet angles between males and females at thoracic vertebral levels T2, T4, T6, and T10–T11. For left facet angles, significant differences (p!.00417) between males and females were observed at T6 and T9. Figure 9 shows the facet dimensions (median values and interquartile distributions) at each thoracic vertebral level for both genders.

Discussion Although there is much literature available on the detailed quantitative evaluations of the thoracic vertebrae in the adult

population, only limited data on select vertebral features exist for the pediatric thoracic spine. In addition, few attempts have been made to quantify the variations in thoracic vertebral geometry with age. To the authors’ knowledge, this study is the first to systematically assess the thoracic vertebral morphology and the age-, gender-, and level-related changes in the size of thoracic vertebral features in the pediatric population. These data can be translated to guide clinical decision-making for corrective spinal surgery, improve the biomechanical and computational models of growth and modulation for the thoracic spine, and advice the design of growth sparing and modulating devices. Although symmetry was generally observed in most thoracic vertebral geometry measures for both genders, asymmetries were found in vertebral body heights, end plate widths and depths, and interfacet widths. Asymmetry

J.R. Peters et al. / The Spine Journal 15 (2015) 1000–1020

1009

Fig. 9. Checker plots (based on Wilcoxon ranked-sign tests) showing interlevel differences in thoracic vertebral geometry parameters for all male pediatric subjects. Because of limitations in space for data presentation, two parameters are contained in each plot. In sub-figures A-M, the results pertaining to the top parameter label are shown in the top-right triangular section of the plot and those pertaining to the left parameter label are shown in the bottom-left triangular section. A black cell indicates that no comparison was made for that vertebral level pair. A gray cell indicates a significant difference (p!.000013) in the geometric measure between any two thoracic vertebral levels. For example: In Fig. 9 (M), significant differences were observed for the right facet angle between levels T1 and T7, T9, T10 and between T2 and T4–T10, T12. Significant differences were also observed for the left facet angle between levels T2 and T5–T10, T12.

of the vertebral body and end plate dimensions indicate anteriorly directed wedging and proximally directed tapering of the vertebral bodies and are consistent with similar morphologic data reported in the adult spine morphology literature [30,32,35,57]. Previous studies on pedicle morphology have reported asymmetry in normative pediatric subjects; however, these studies may have been limited

by small sample sizes and lack of rigorous statistical analyses [38,40]. The present study, supported by thorough analyses of a larger data set, found no asymmetry in pedicle dimensions for the normative pediatric population. Although the symmetry of interfacet width dimensions is previously unreported in the pediatric literature, the asymmetries observed in the present study may be because of a

1010

J.R. Peters et al. / The Spine Journal 15 (2015) 1000–1020

Fig. 9. (continued).

J.R. Peters et al. / The Spine Journal 15 (2015) 1000–1020

1011

Fig. 10. Checker plots (based on Wilcoxon ranked-sign tests) showing interlevel differences in thoracic vertebral geometry parameters for all female pediatric subjects. Because of limitations in space for data presentation, two parameters are contained in each plot. In sub-figures A-M, results pertaining to the top parameter label are shown in the top-right triangular section of the plot and those pertaining to the left parameter label are shown in the bottom-left triangular section. A black cell indicates that no comparison was made for that vertebral level pair. A gray cell indicates a significant difference (p!.000013) in the geometric measure between any two thoracic vertebral levels. For example: in Fig. 10 (M), significant differences were observed for the right facet angle between levels T3 and T1, T6–T8. Significant differences were also observed for left facet angle between levels T6 and T1, T2, T4.

widening of the inferior facets to accommodate the superior facets of distal vertebrae. The rates of growth followed trends similar to those of their associated vertebral dimensions and such concomitant trends suggest the growth of vertebral features to be proportional to their relative sizes across thoracic levels. This indicates that, across levels and between genders, larger

vertebral structures grow at faster rates, whereas smaller structures grow at a slower rate. Although there was an effect of age on most thoracic vertebral features, certain structures such as the spinal canal displayed little size variation with age. This is consistent with the findings by Dimeglio and Canavese [3] that the spinal canal reaches approximately 95% of its size at skeletal maturity by 5

1012

J.R. Peters et al. / The Spine Journal 15 (2015) 1000–1020

Fig. 10. (continued).

Table 2 Coefficients of linear regression models used to estimate thoracic vertebral geometry measures at any thoracic vertebral level in male and female subjects aged 1 to 19 years Gender

T1

T2

T3

T4

T5

T6

T7

T8

T9

T10

T11

T12

VBH a (mm)

M

0.728 [0.616, 0.839] (3.807) !0.855O 0.731 [0.653, 0.809] (3.784) !0.885O 0.724 [0.639, 0.81] (4.886) !0.909O 0.671 [0.596, 0.745] (5.607) !0.881O 0.787 [0.637, 0.938] (14.66) !0.797O 0.504 [0.396, 0.612] (16.36) !0.653O 0.81 [0.69, 0.931] (15.705) !0.867O 0.461 [0.352, 0.569] (17.773) !0.607O 0.314 [0.205, 0.423] (8.335) !0.537O 0.307 [0.243, 0.371] (7.937) !0.671O 0.396 [0.301, 0.492] (8.825) !0.705O 0.318 [0.257, 0.379] (8.552) !0.706O

0.82 [0.715, 0.926] (4) !0.857O 0.843 [0.759, 0.927] (4.003) !0.886O 0.764 [0.68, 0.848] (5.375) !0.891O 0.742 [0.66, 0.824] (5.884) !0.863O 0.578 [0.474, 0.682] (15.829) !0.754O 0.458 [0.385, 0.531] (15.493) !0.755O 0.636 [0.545, 0.726] (17.589) !0.831O 0.41 [0.318, 0.502] (17.574) !0.61O 0.367 [0.279, 0.456] (9.233) !0.633O 0.357 [0.289, 0.426] (8.611) !0.678O 0.385 [0.297, 0.472] (9.945) !0.659O 0.404 [0.348, 0.459] (9.178) !0.803O

0.775 [0.685, 0.865] (5.065) !0.881O 0.832 [0.748, 0.916] (4.812) !0.881O 0.788 [0.714, 0.862] (5.896) !0.918O 0.79 [0.705, 0.876] (6.225) !0.866O 0.65 [0.556, 0.745] (13.321) !0.825O 0.477 [0.409, 0.546] (13.584) !0.787O 0.618 [0.524, 0.713] (14.912) !0.81O 0.374 [0.297, 0.451] (15.087) !0.642O 0.464 [0.36, 0.568] (9.603) !0.665O 0.47 [0.411, 0.529] (8.723) !0.828O 0.532 [0.442, 0.622] (9.915) !0.776O 0.506 [0.448, 0.563] (9.471) !0.855O

0.817 [0.729, 0.906] (4.986) !0.891O 0.766 [0.687, 0.846] (5.644) !0.878O 0.776 [0.689, 0.864] (6.269) !0.884O 0.786 [0.689, 0.883] (6.761) !0.835O 0.62 [0.53, 0.711] (13.055) !0.821O 0.485 [0.418, 0.551] (12.764) !0.802O 0.577 [0.49, 0.664] (14.435) !0.81O 0.388 [0.324, 0.453] (14.367) !0.734O 0.582 [0.485, 0.679] (9.847) !0.778O 0.534 [0.474, 0.594] (9.381) !0.858O 0.549 [0.466, 0.632] (10.566) !0.811O 0.553 [0.496, 0.609] (9.947) !0.878O

0.821 [0.718, 0.925] (5.037) !0.862O 0.86 [0.774, 0.946] (5.406) !0.883O 0.817 [0.738, 0.895] (6.079) !0.915O 0.826 [0.753, 0.899] (6.678) !0.907O 0.691 [0.613, 0.769] (12.767) !0.887O 0.527 [0.457, 0.598] (12.748) !0.809O 0.619 [0.533, 0.704] (14.329) !0.84O 0.433 [0.364, 0.503] (14.401) !0.748O 0.653 [0.58, 0.727] (10.289) !0.887O 0.592 [0.518, 0.667] (9.889) !0.827O 0.658 [0.585, 0.732] (10.858) !0.89O 0.59 [0.519, 0.661] (10.432) !0.84O

0.796 [0.718, 0.874] (5.34) !0.912O 0.855 [0.752, 0.958] (5.559) !0.838O 0.864 [0.789, 0.939] (6.029) !0.929O 0.873 [0.776, 0.971] (6.488) !0.86O 0.756 [0.661, 0.851] (12.883) !0.864O 0.556 [0.479, 0.633] (13.164) !0.798O 0.717 [0.611, 0.824] (14.442) !0.82O 0.447 [0.372, 0.523] (15.14) !0.728O 0.706 [0.612, 0.8] (10.799) !0.849O 0.643 [0.572, 0.715] (10.561) !0.86O 0.726 [0.645, 0.807] (11.146) !0.888O 0.638 [0.578, 0.697] (11.098) !0.897O

0.828 [0.732, 0.923] (5.014) !0.882O 0.897 [0.789, 1.005] (5.416) !0.843O 0.864 [0.781, 0.948] (6.007) !0.914O 0.9 [0.803, 0.998] (6.553) !0.869O 0.849 [0.733, 0.965] (13.344) !0.842O 0.63 [0.552, 0.708] (13.499) !0.835O 0.772 [0.646, 0.898] (15.096) !0.789O 0.507 [0.422, 0.593] (15.485) !0.734O 0.8 [0.716, 0.884] (11.406) !0.9O 0.69 [0.613, 0.767] (10.885) !0.861O 0.839 [0.747, 0.93] (11.346) !0.892O 0.677 [0.598, 0.755] (11.555) !0.852O

0.803 [0.697, 0.91] (5.886) !0.85O 0.967 [0.849, 1.085] (5.368) !0.838O 0.854 [0.763, 0.944] (6.268) !0.899O 0.912 [0.803, 1.021] (6.857) !0.844O 0.935 [0.818, 1.053] (13.584) !0.866O 0.693 [0.616, 0.769] (13.907) !0.865O 0.829 [0.718, 0.939] (15.592) !0.852O 0.569 [0.492, 0.646] (15.972) !0.808O 0.889 [0.781, 0.996] (11.156) !0.872O 0.751 [0.674, 0.827] (11.183) !0.882O 0.937 [0.835, 1.04] (10.889) !0.893O 0.741 [0.664, 0.818] (11.43) !0.878O

0.926 [0.823, 1.03] (5.295) !0.886O 0.981 [0.869, 1.092] (6.078) !0.857O 0.945 [0.874, 1.016] (6.022) !0.945O 0.972 [0.871, 1.072] (6.664) !0.878O 0.981 [0.869, 1.093] (13.834) !0.881O 0.748 [0.664, 0.831] (14.306) !0.86O 0.908 [0.791, 1.024] (15.607) !0.855O 0.617 [0.539, 0.696] (16.59) !0.828O 0.893 [0.785, 1.001] (11.742) !0.869O 0.818 [0.746, 0.89] (10.983) !0.909O 0.911 [0.819, 1.002] (11.559) !0.906O 0.809 [0.73, 0.888] (11.235) !0.89O

0.998 [0.89, 1.106] (5.828) !0.892O 1.015 [0.907, 1.122] (6.602) !0.871O 0.99 [0.903, 1.078] (6.164) !0.925O 0.993 [0.887, 1.099] (7.284) !0.869O 1.07 [0.956, 1.183] (14.047) !0.896O 0.772 [0.686, 0.859] (15.072) !0.859O 0.978 [0.863, 1.094] (16.427) !0.874O 0.669 [0.573, 0.765] (17.327) !0.787O 0.971 [0.87, 1.073] (11.126) !0.899O 0.82 [0.739, 0.9] (11.054) !0.888O 0.953 [0.863, 1.044] (11.309) !0.915O 0.806 [0.72, 0.893] (11.406) !0.869O

1.027 [0.896, 1.158] (5.974) !0.863O 1.061 [0.949, 1.173] (6.664) !0.875O 1.075 [0.97, 1.18] (6.773) !0.914O 1.12 [1.005, 1.235] (7.388) !0.88O 1.171 [1.06, 1.282] (14.522) !0.919O 0.9 [0.802, 0.998] (15.04) !0.866O 1.081 [0.958, 1.205] (16.942) !0.886O 0.754 [0.647, 0.862] (17.848) !0.792O 0.955 [0.864, 1.047] (11.193) !0.918O 0.843 [0.758, 0.929] (10.924) !0.884O 0.961 [0.864, 1.057] (11.401) !0.91O 0.799 [0.717, 0.881] (11.706) !0.88O

1.08 [0.965, 1.194] (6.464) !0.898O 1.106 [1.01, 1.202] (7.35) !0.914O 1.174 [1.076, 1.273] (6.551) !0.934O 1.198 [1.101, 1.296] (7.723) !0.923O 1.214 [1.097, 1.332] (15.485) !0.914O 0.903 [0.796, 1.009] (16.407) !0.85O 1.09 [0.961, 1.218] (17.947) !0.877O 0.792 [0.691, 0.893] (18.406) !0.83O 0.936 [0.849, 1.023] (11.875) !0.92O 0.82 [0.733, 0.906] (11.624) !0.876O 0.902 [0.798, 1.006] (12.816) !0.882O 0.796 [0.71, 0.881] (11.827) !0.873O

F

VBH p (mm)

M

F

EPW s (mm)

M

F

EPiW (mm)

M

F

EPD s (mm)

M

F

EPD i (mm)

M

F

1013

(Continued)

J.R. Peters et al. / The Spine Journal 15 (2015) 1000–1020

Measure

1014

Table 2 (Continued ) Gender

T1

T2

T3

T4

T5

T6

T7

T8

T9

T10

T11

T12

PDH r (mm)

M

0.587 [0.467, 0.707] (5.152) !0.796O 0.551 [0.466, 0.636] (5.295) !0.832O 0.523 [0.412, 0.635] (5.532) !0.782O 0.584 [0.479, 0.688] (5.188) !0.803O 0.322 [0.249, 0.394] (5.643) !0.748O 0.201 [0.158, 0.244] (5.614) !0.652O 0.343 [0.283, 0.404] (5.455) !0.822O 0.197 [0.159, 0.236] (5.689) !0.687O 5.038 [3.935, 6.141] (20.808) !0.803O 3.588 [2.949, 4.227] (20.967) !0.788O 5.033 [3.665, 6.4] (19.297) !0.697O 3.83 [3.191, 4.469] (20.257) !0.828O

0.528 [0.45, 0.606] (5.489) !0.824O 0.47 [0.411, 0.528] (5.76) !0.838O 0.505 [0.434, 0.576] (5.623) !0.833O 0.474 [0.414, 0.534] (5.814) !0.834O 0.271 [0.228, 0.314] (4.945) !0.795O 0.197 [0.154, 0.239] (4.702) !0.628O 0.26 [0.219, 0.301] (5.021) !0.795O 0.187 [0.147, 0.228] (4.87) !0.627O 4.03 [3.1, 4.961] (21.933) !0.657O 3.292 [2.812, 3.773] (19.647) !0.791O 4.325 [3.521, 5.129] (20.197) !0.742O 3.531 [2.936, 4.127] (20.248) !0.736O

0.466 [0.418, 0.514] (6.123) !0.904O 0.463 [0.41, 0.517] (5.761) !0.854O 0.474 [0.422, 0.525] (6.056) !0.897O 0.453 [0.405, 0.5] (5.76) !0.874O 0.202 [0.148, 0.257] (4.692) !0.577O 0.152 [0.111, 0.192] (4.217) !0.517O 0.221 [0.165, 0.277] (4.603) !0.608O 0.141 [0.102, 0.18] (4.296) !0.501O 3.865 [3.22, 4.511] (20.297) !0.79O 2.68 [2.274, 3.087] (17.697) !0.782O 4.067 [3.28, 4.854] (18.969) !0.732O 2.805 [2.368, 3.243] (17.758) !0.768O

0.471 [0.417, 0.526] (5.882) !0.879O 0.418 [0.372, 0.465] (5.83) !0.863O 0.475 [0.429, 0.522] (5.765) !0.912O 0.43 [0.376, 0.484] (5.718) !0.832O 0.179 [0.133, 0.225] (4.325) !0.594O 0.108 [0.074, 0.142] (4.032) !0.44O 0.191 [0.134, 0.249] (4.215) !0.518O 0.119 [0.088, 0.15] (3.89) !0.524O 3.473 [2.907, 4.04] (18.43) !0.793O 2.289 [1.862, 2.717] (17.155) !0.69O 3.729 [3.092, 4.366] (16.099) !0.773O 2.452 [2.061, 2.843] (15.798) !0.753O

0.472 [0.426, 0.517] (5.671) !0.913O 0.404 [0.366, 0.443] (5.817) !0.893O 0.444 [0.399, 0.49] (5.679) !0.905O 0.407 [0.362, 0.452] (5.66) !0.866O 0.18 [0.135, 0.226] (4.168) !0.608O 0.102 [0.069, 0.134] (3.987) !0.418O 0.172 [0.115, 0.23] (4.257) !0.47O 0.115 [0.079, 0.15] (3.839) !0.445O 3.477 [2.972, 3.982] (16.311) !0.825O 2.213 [1.843, 2.584] (16.92) !0.734O 3.236 [2.53, 3.942] (17.187) !0.677O 2.319 [1.941, 2.698] (16.024) !0.752O

0.496 [0.444, 0.548] (5.478) !0.902O 0.422 [0.381, 0.463] (5.632) !0.89O 0.462 [0.408, 0.517] (5.635) !0.876O 0.409 [0.362, 0.457] (5.673) !0.85O 0.201 [0.154, 0.248] (4.11) !0.645O 0.104 [0.069, 0.139] (3.946) !0.4O 0.186 [0.123, 0.249] (4.087) !0.462O 0.105 [0.071, 0.14] (3.825) !0.415O 3.848 [3.243, 4.453] (14.594) !0.801O 2.343 [1.972, 2.715] (16.255) !0.751O 3.602 [2.841, 4.363] (15.069) !0.69O 2.272 [1.912, 2.632] (15.45) !0.751O

0.496 [0.439, 0.553] (5.415) !0.883O 0.431 [0.382, 0.481] (5.595) !0.854O 0.485 [0.438, 0.531] (5.472) !0.915O 0.438 [0.396, 0.481] (5.505) !0.891O 0.203 [0.155, 0.251] (4.273) !0.641O 0.09 [0.051, 0.13] (4.144) !0.291O 0.213 [0.155, 0.27] (4.156) !0.581O 0.113 [0.076, 0.15] (3.932) !0.422O 4.188 [3.281, 5.095] (14.169) !0.685O 2.353 [1.992, 2.714] (16.987) !0.767O 3.915 [3.194, 4.637] (15.294) !0.75O 2.443 [2.12, 2.766] (15.792) !0.816O

0.5 [0.45, 0.55] (5.511) !0.911O 0.445 [0.395, 0.496] (5.773) !0.858O 0.507 [0.448, 0.565] (5.665) !0.881O 0.45 [0.399, 0.5] (5.803) !0.861O 0.225 [0.164, 0.286] (4.311) !0.575O 0.093 [0.057, 0.13] (4.426) !0.335O 0.189 [0.145, 0.233] (4.558) !0.656O 0.093 [0.056, 0.131] (4.352) !0.323O 4.277 [3.612, 4.942] (15.216) !0.813O 2.454 [2.067, 2.842] (19.196) !0.76O 4.048 [3.383, 4.714] (17.776) !0.795O 2.567 [2.143, 2.991] (18.569) !0.74O

0.508 [0.457, 0.559] (5.928) !0.905O 0.464 [0.412, 0.515] (6.36) !0.865O 0.493 [0.439, 0.547] (6.138) !0.889O 0.462 [0.398, 0.525] (6.51) !0.807O 0.216 [0.17, 0.262] (4.659) !0.682O 0.088 [0.05, 0.126] (4.839) !0.291O 0.213 [0.167, 0.258] (4.776) !0.681O 0.101 [0.061, 0.142] (4.683) !0.331O 4.557 [3.696, 5.418] (18.237) !0.751O 2.817 [2.358, 3.276] (22.079) !0.745O 4.465 [3.717, 5.213] (19.749) !0.78O 2.819 [2.279, 3.36] (23.711) !0.683O

0.582 [0.512, 0.652] (6.848) !0.87O 0.534 [0.472, 0.596] (7.32) !0.85O 0.577 [0.519, 0.634] (6.84) !0.907O 0.522 [0.464, 0.58] (7.371) !0.859O 0.276 [0.225, 0.326] (4.914) !0.744O 0.145 [0.102, 0.187] (5.086) !0.465O 0.241 [0.195, 0.287] (5.258) !0.729O 0.153 [0.113, 0.193] (5.186) !0.528O 5.947 [4.865, 7.028] (23.989) !0.751O 4.022 [3.396, 4.649] (26.766) !0.762O 5.619 [4.863, 6.374] (26.151) !0.846O 4.123 [3.507, 4.739] (27.124) !0.776O

0.702 [0.634, 0.769] (7.063) !0.916O 0.596 [0.527, 0.665] (7.727) !0.854O 0.645 [0.589, 0.701] (7.392) !0.93O 0.61 [0.543, 0.677] (7.783) !0.866O 0.334 [0.275, 0.392] (5.164) !0.77O 0.229 [0.177, 0.28] (4.961) !0.614O 0.343 [0.285, 0.401] (5.358) !0.785O 0.253 [0.193, 0.313] (5.053) !0.578O 7.793 [6.692, 8.893] (21.639) !0.837O 5.396 [4.576, 6.216] (25.812) !0.781O 7.544 [6.594, 8.494] (24.81) !0.878O 5.578 [4.756, 6.4] (27.309) !0.788O

0.676 [0.612, 0.739] (7.407) !0.919O 0.602 [0.525, 0.679] (7.912) !0.834O 0.686 [0.624, 0.747] (7.439) !0.926O 0.596 [0.53, 0.662] (8.064) !0.876O 0.387 [0.317, 0.458] (5.113) !0.751O 0.217 [0.15, 0.284] (5.357) !0.492O 0.388 [0.292, 0.483] (5.454) !0.622O 0.311 [0.225, 0.396] (5.142) !0.562O 8.666 [7.32, 10.012] (19.57) !0.809O 5.493 [4.701, 6.285] (29.016) !0.816O 8.566 [7.31, 9.821] (23.615) !0.83O 6.516 [5.684, 7.347] (25.436) !0.872O

F

PDH l (mm)

M

F

PDW r (mm)

M

F

PDW l (mm)

M

F

2 PDA r (mm )

M

F

2 PDA l (mm )

M

F

J.R. Peters et al. / The Spine Journal 15 (2015) 1000–1020

Measure

PDqr (  )

M

F

PDql (  )

M

F

M

F

SCD (mm)

M

F

SCA (mm2)

M

F

0.191 [0.507, 0.125] (21.618) !0.034O 0.473 [0.695, 0.251] (19.124) !0.261O 0.198 [0.43, 0.034] (20.448) !0.066O

0.051 [0.013, 0.114] (13.886) !0.05O 3.115 [1.459, 4.771] (157.176) !0.33O

0.002 [0.068, 0.072] (14.56) !0O 1.395 [0.037, 2.827] (155.599) !0.084O 1.441 [0.248, 2.634] (147.386) !0.101O

2.311 [1.052, 3.571] (158.877) !0.221O

0.489 [0.712, 0.266] (18.743) !0.272O 0.181 [0.097, 0.264] (14.752) !0.313O 0.129 [0.069, 0.19] (14.908) !0.261O 0.017 [0.09, 0.056] (15.331) !0.005O

0.085 [0.4, 0.231] (16.583) !0.007O 0.222 [0.452, 0.009] (16.534) !0.066O 0.349 [0.056, 0.643] (21.031) !0.123O

0.115 [0.108, 0.338] (15.425) !0.025O 0.024 [0.227, 0.275] (12.972) !0.001O 0.305 [0.005, 0.606] (16.541) !0.091O

0.262 [0.053, 0.577] (14.914) !0.064O 0.207 [0.032, 0.445] (13.684) !0.054O 0.521 [0.232, 0.81] (16.118) !0.244O

0.429 [0.179, 0.68] (13.904) !0.226O

0.44 [0.121, 0.759] (14.892) !0.163O

0.57 [0.28, 0.861] (14.863) !0.282O

0.644 [0.221, 1.068] (18.016) !0.183O

0.369 [0.019, 0.72] (14.703) !0.097O

0.359 [0.128, 0.591] (13.173) !0.154O

0.525 [0.28, 0.769] (14.089) !0.267O

0.317 [0.076, 0.559] (11.538) !0.118O

0.196 [0.09, 0.482] (12.453) !0.035O

0.261 [0.018, 0.505] (11.792) !0.081O

0.449 [0.173, 0.725] (14.777) !0.209O

0.591 [0.315, 0.867] (15.068) !0.319O

0.574 [0.332, 0.816] (15.006) !0.365O

0.767 [0.435, 1.099] (18.217) !0.341O

0.652 [0.237, 1.066] (15.821) !0.193O

0.118 [0.161, 0.397] (6.882) !0.018O 0.14 [0.081, 0.361] (5.101) !0.031O 0.27 [0.033, 0.508] (7.905) !0.129O

0.327 [0.101, 0.553] (11.069) !0.137O

0.008 [0.195, 0.21] (4.272) !0O

0.165 [0.088, 0.242] (14.181) !0.307O 0.164 [0.093, 0.235] (14.055) !0.288O 0.036 [0.043, 0.114] (14.45) !0.02O 0.009 [0.079, 0.097] (14.874) !0.001O 1.517 [0.25, 2.784] (144.616) !0.125O

0.181 [0.092, 0.269] (15.133) !0.299O 0.219 [0.14, 0.298] (14.72) !0.374O 0.06 [0.016, 0.135] (15.443) !0.06O 0.033 [0.049, 0.115] (15.59) !0.012O 2.525 [1.081, 3.97] (160.454) !0.238O

1.608 [0.232, 2.984] (144.154) !0.096O

2.727 [1.181, 4.273] (156.752) !0.194O

0.107 [0.175, 0.388] (17.84) !0.011O 0.143 [0.064, 0.222] (13.969) !0.248O 0.133 [0.068, 0.199] (13.812) !0.238O 0.004 [0.076, 0.085] (14.941) !0O 0.025 [0.043, 0.093] (14.213) !0.01O 1.523 [0.155, 2.892] (142.226) !0.11O

0.203 [0.002, 0.403] (12.859) !0.072O

0.379 [0.121, 0.636] (12.985) !0.141O

0.332 [0.103, 0.561] (13.086) !0.137O

0.373 [0.138, 0.608] (12.347) !0.163O

0.482 [0.255, 0.708] (13.849) !0.26O

0.183 [0.103, 0.263] (12.97) !0.337O 0.153 [0.088, 0.218] (13.033) !0.296O 0.077 [0.005, 0.158] (14.419) !0.079O 0.051 [0.031, 0.133] (14.145) !0.029O 2.312 [0.926, 3.698] (132.964) !0.213O

0.16 [0.084, 0.236] (12.859) !0.308O 0.126 [0.058, 0.194] (13.134) !0.207O 0.057 [0.014, 0.128] (14.298) !0.06O 0.026 [0.048, 0.1] (14.5) !0.009O

0.136 [0.063, 0.21] (13.501) !0.253O 0.147 [0.082, 0.212] (13.195) !0.282O 0.024 [0.042, 0.089] (13.994) !0.013O 0.017 [0.052, 0.086] (14.249) !0.005O 1.624 [0.419, 2.828] (135) !0.153O

0.135 [0.066, 0.203] (13.797) !0.278O 0.157 [0.094, 0.22] (13.386) !0.322O 0.021 [0.048, 0.089] (14.196) !0.009O 0.008 [0.064, 0.08] (14.515) !0.001O

2.008 [0.742, 3.275] (131.936) !0.2O

0.158 [0.084, 0.232] (12.959) !0.312O 0.15 [0.084, 0.216] (12.927) !0.281O 0.041 [0.034, 0.116] (14.188) !0.029O 0.036 [0.043, 0.115] (14.317) !0.016O 1.927 [0.747, 3.107] (131.654) !0.21O

1.505 [0.405, 2.604] (138.449) !0.157O

0.242 [0.072, 0.555] (14.54) !0.044O 0.161 [0.089, 0.232] (13.688) !0.329O 0.182 [0.121, 0.244] (13.652) !0.402O 0.011 [0.054, 0.076] (14.436) !0.003O 0.002 [0.072, 0.068] (14.491) !0O 1.57 [0.43, 2.711] (139.035) !0.155O

1.779 [0.574, 2.983] (136.863) !0.142O

2.094 [0.826, 3.363] (131.772) !0.171O

1.991 [0.714, 3.267] (133.155) !0.156O

2.233 [0.938, 3.529] (130.612) !0.184O

2.205 [0.949, 3.461] (132.26) !0.193O

1.945 [0.723, 3.166] (135.992) !0.164O

1.887 [0.656, 3.118] (137.588) !0.154O

0.165 [0.384, 0.053] (0.86) !0.055O 0.008 [0.219, 0.204] (2.2) !0O 0.054 [0.117, 0.225] (2.868) !0.011O 0.154 [0.474, 0.165] (1.366) !0.025O 0.172 [0.09, 0.254] (16.754) !0.303O 0.23 [0.142, 0.319] (16.276) !0.349O 0.075 [0.006, 0.143] (16.317) !0.107O 0.094 [0.007, 0.181] (16.407) !0.084O 2.788 [1.132, 4.444] (193.079) !0.22O

J.R. Peters et al. / The Spine Journal 15 (2015) 1000–1020

SCW (mm)

0.051 [0.263, 0.365] (31.982) !0.004O 0.057 [0.164, 0.278] (32.042) !0.006O 0.207 [0.232, 0.646] (34.262) !0.031O 0.179 [0.406, 0.047] (31.302) !0.05O 0.207 [0.108, 0.306] (16.61) !0.377O 0.188 [0.112, 0.263] (16.821) !0.343O 0.106 [0.023, 0.189] (13.887) !0.184O

3.81 [1.873, 5.747] (186.295) !0.234O (Continued)

1015

1016

Table 2 (Continued ) Gender

T1

T2

T3

T4

T5

T6

T7

T8

T9

T10

T11

T12

SPL (mm)

M

1.885 [1.614, 2.157] (26.192) !0.9O 1.578 [1.424, 1.733] (28.534) !0.898O 0.524 [0.051, 0.997] (22.53) !0.186O

1.964 [1.725, 2.202] (26.926) !0.895O 1.631 [1.476, 1.785] (29.263) !0.903O 0.398 [0.104, 0.692] (27.168) !0.187O

1.891 [1.667, 2.116] (28.131) !0.902O 1.609 [1.456, 1.761] (29.858) !0.902O 0.556 [0.238, 0.874] (26.082) !0.284O

1.98 [1.778, 2.182] (27.85) !0.923O 1.652 [1.494, 1.81] (30.356) !0.906O 0.608 [0.34, 0.876] (28.434) !0.393O

1.914 [1.661, 2.166] (28.594) !0.882O 1.778 [1.63, 1.926] (30.12) !0.927O 0.573 [0.176, 0.969] (31.626) !0.213O

1.91 [1.655, 2.166] (29.22) !0.89O 1.824 [1.668, 1.98] (30.861) !0.925O 0.761 [0.51, 1.011] (31.291) !0.571O

2.041 [1.732, 2.35] (28.178) !0.867O 1.925 [1.754, 2.096] (30.846) !0.919O 0.574 [0.238, 0.911] (32.22) !0.304O

2.039 [1.753, 2.324] (28.978) !0.885O 1.901 [1.75, 2.051] (31.464) !0.935O 0.591 [0.313, 0.868] (29.657) !0.404O

2.022 [1.779, 2.265] (28.961) !0.906O 1.908 [1.761, 2.055] (30.927) !0.935O 0.454 [0.174, 0.734] (29.834) !0.268O

1.95 [1.726, 2.174] (29.536) !0.913O 1.705 [1.556, 1.854] (32.478) !0.917O 0.344 [0.089, 0.6] (29.085) !0.202O

1.863 [1.653, 2.073] (31.45) !0.922O 1.662 [1.525, 1.8] (33.249) !0.93O 0.149 [0.103, 0.4] (27.877) !0.05O

0.321 [0.186, 0.456] (23.615) !0.321O 2.008 [1.707, 2.31] (39.71) !0.861O 1.893 [1.689, 2.096] (40.059) !0.879O 0.908 [0.766, 1.05] (14.325) !0.864O 0.715 [0.602, 0.827] (14.338) !0.817O 0.891 [0.751, 1.031] (14.385) !0.863O 0.731 [0.621, 0.841] (14.456) !0.831O 0.613 [0.351, 0.875] (26.591) !0.481O 0.661 [0.462, 0.859] (25.411) !0.551O

0.377 [0.269, 0.484] (25.563) !0.507O 1.921 [1.704, 2.137] (36.269) !0.884O 1.715 [1.522, 1.908] (36.399) !0.859O 0.884 [0.76, 1.007] (14.928) !0.843O 0.876 [0.77, 0.982] (14.69) !0.841O 0.86 [0.733, 0.988] (15.433) !0.815O 0.903 [0.792, 1.013] (14.353) !0.844O 0.571 [0.347, 0.795] (21.54) !0.399O 0.551 [0.413, 0.689] (21.154) !0.558O

0.337 [0.2, 0.474] (25.463) !0.333O 1.66 [1.463, 1.857] (33.827) !0.879O 1.524 [1.346, 1.701] (34.201) !0.848O 0.854 [0.731, 0.978] (15.786) !0.838O 0.922 [0.822, 1.022] (14.221) !0.866O 0.835 [0.747, 0.923] (15.719) !0.9O 0.924 [0.826, 1.022] (14.419) !0.87O 0.525 [0.36, 0.689] (18.642) !0.503O 0.358 [0.255, 0.462] (18.546) !0.476O

0.38 [0.225, 0.536] (29.603) !0.345O 1.609 [1.412, 1.807] (32.875) !0.868O 1.502 [1.317, 1.686] (33.116) !0.837O 0.959 [0.832, 1.085] (14.587) !0.848O 0.909 [0.81, 1.008] (15.214) !0.865O 0.928 [0.817, 1.039] (14.637) !0.871O 0.901 [0.799, 1.004] (15.1) !0.854O 0.466 [0.311, 0.621] (17.282) !0.467O 0.301 [0.193, 0.409] (17.51) !0.371O

0.481 [0.334, 0.628] (31.802) !0.485O 1.827 [1.627, 2.027] (32.306) !0.892O 1.584 [1.411, 1.758] (32.972) !0.866O 0.969 [0.846, 1.091] (14.405) !0.865O 0.944 [0.853, 1.035] (15.157) !0.892O 0.827 [0.718, 0.935] (15.435) !0.855O 0.938 [0.843, 1.032] (14.763) !0.882O 0.331 [0.188, 0.474] (17.739) !0.347O 0.303 [0.181, 0.425] (16.995) !0.319O

0.606 [0.476, 0.736] (34.266) !0.663O 1.837 [1.631, 2.044] (32.976) !0.887O 1.593 [1.419, 1.767] (33.204) !0.867O 0.996 [0.876, 1.116] (14.32) !0.875O 0.942 [0.829, 1.055] (16.168) !0.841O 0.967 [0.848, 1.087] (14.267) !0.872O 0.909 [0.811, 1.008] (16.162) !0.868O 0.359 [0.233, 0.485] (17.485) !0.446O 0.291 [0.188, 0.394] (16.725) !0.377O

0.507 [0.359, 0.655] (34.557) !0.514O 1.827 [1.641, 2.013] (33.099) !0.906O 1.571 [1.406, 1.737] (33.137) !0.877O 1 [0.849, 1.151] (14.845) !0.814O 0.943 [0.833, 1.052] (16.633) !0.852O 0.996 [0.853, 1.139] (14.959) !0.832O 0.982 [0.867, 1.097] (16.178) !0.849O 0.389 [0.249, 0.529] (17.508) !0.435O 0.244 [0.12, 0.368] (17.288) !0.23O

0.421 [0.255, 0.587] (33.932) !0.367O 1.814 [1.617, 2.012] (32.377) !0.896O 1.481 [1.339, 1.623] (33.236) !0.896O 0.983 [0.812, 1.153] (15.524) !0.768O 1.015 [0.869, 1.16] (16.3) !0.79O 0.996 [0.857, 1.135] (15.562) !0.843O 0.969 [0.817, 1.12] (16.628) !0.764O 0.338 [0.202, 0.474] (18.112) !0.38O 0.416 [0.301, 0.53] (16.41) !0.505O

0.313 [0.175, 0.451] (33.451) !0.308O 1.806 [1.618, 1.993] (31.507) !0.9O 1.468 [1.333, 1.604] (32.761) !0.903O 1.085 [0.96, 1.211] (15.647) !0.882O 1.032 [0.888, 1.176] (16.934) !0.801O 1.066 [0.943, 1.188] (15.631) !0.88O 1.043 [0.902, 1.184] (16.752) !0.813O 0.561 [0.428, 0.694] (16.831) !0.634O 0.59 [0.481, 0.698] (16.242) !0.697O

0.229 [0.095, 0.363] (32.763) !0.197O 1.702 [1.528, 1.876] (31.07) !0.903O 1.253 [1.06, 1.447] (32.369) !0.769O 1.24 [1.122, 1.359] (15.74) !0.916O 1.166 [1.036, 1.295] (17.424) !0.86O 1.221 [1.079, 1.364] (15.982) !0.88O 1.172 [1.024, 1.32] (17.199) !0.826O 0.563 [0.42, 0.707] (17.331) !0.598O 0.64 [0.532, 0.747] (16.486) !0.728O

1.734 [1.486, 1.981] (31.211) !0.884O 1.617 [1.477, 1.757] (32.228) !0.92O 0.148 [0.209, 0.505] (28.977) !0.026O 0.183 [0.014, 0.351] (30.767) !0.092O 1.506 [1.302, 1.711] (30.643) !0.847O 0.985 [0.774, 1.195] (31.436) !0.643O 1.343 [1.2, 1.487] (16.416) !0.9O 1.273 [1.122, 1.424] (18.466) !0.849O 1.297 [1.128, 1.465] (16.994) !0.858O 1.303 [1.173, 1.434] (18.399) !0.887O 0.719 [0.538, 0.899] (16.677) !0.618O 0.703 [0.594, 0.812] (15.475) !0.763O

F

SPq (  )

M

F

ITW (mm)

M

F

IFH r (mm)

M

F

IFH l (mm)

M

F

IFW s (mm)

M

F

0.27 [0.104, 0.436] (24.435) !0.192O 1.143 [0.916, 1.37] (29.985) !0.717O 0.729 [0.537, 0.922] (30.004) !0.547O 1.515 [1.352, 1.677] (17.899) !0.898O 1.494 [1.336, 1.652] (18.895) !0.883O 1.479 [1.314, 1.644] (18.192) !0.891O 1.58 [1.434, 1.727] (18.511) !0.904O 0.657 [0.497, 0.817] (17.832) !0.626O 0.628 [0.504, 0.753] (16.789) !0.673O

J.R. Peters et al. / The Spine Journal 15 (2015) 1000–1020

Measure

IFiW (mm)

M

F

Fqr (  )

M

F

M

F

0.333 [0.321, 0.988] (58.716) !0.039O 0.705 [0.187, 1.223] (52.073) !0.152O

0.927 [0.759, 1.096] (16.23) !0.75O 0.624 [0.517, 0.731] (19.096) !0.725O 0.226 [0.705, 0.252] (64.345) !0.023O 0.378 [0.065, 0.822] (65.101) !0.055O 0.001 [0.539, 0.536] (63.49) !0O 0.614 [0.168, 1.06] (62.657) !0.133O

0.938 [0.776, 1.1] (15.817) !0.778O 0.578 [0.482, 0.674] (18.996) !0.733O 0.065 [0.577, 0.447] (69.925) !0.002O 0.574 [0.188, 0.96] (67.69) !0.149O

0.791 [0.664, 0.918] (17.321) !0.79O 0.573 [0.47, 0.677] (17.483) !0.7O 0.248 [0.791, 0.294] (75.004) !0.022O 0.709 [0.359, 1.059] (71.949) !0.245O

0.272 [0.203, 0.746] (67.673) !0.032O 0.431 [0.004, 0.866] (70.263) !0.071O

0.146 [0.429, 0.72] (71.505) !0.007O 0.37 [0.046, 0.786] (72.028) !0.058O

0.846 [0.691, 1] (16.378) !0.754O 0.575 [0.47, 0.679] (17.726) !0.701O 0.047 [0.477, 0.57] (73.604) !0.001O 0.335 [0.021, 0.691] (76.196) !0.064O 0.244 [0.227, 0.715] (71.781) !0.027O 0.717 [0.281, 1.152] (72.16) !0.171O

0.92 [0.792, 1.047] (16.707) !0.849O 0.69 [0.568, 0.813] (16.032) !0.708O 0.158 [0.366, 0.681] (72.825) !0.01O 0.089 [0.219, 0.396] (80.886) !0.006O 0.141 [0.405, 0.687] (73.676) !0.008O 0.368 [0.01, 0.746] (78.778) !0.067O

0.969 [0.801, 1.138] (16.294) !0.771O 0.698 [0.587, 0.809] (16.701) !0.754O 0.116 [0.326, 0.559] (76.601) !0.007O 0.303 [0.012, 0.594] (78.49) !0.077O 0.323 [0.19, 0.837] (74.802) !0.047O 0.128 [0.165, 0.422] (80.585) !0.015O

0.95 [0.798, 1.102] (16.114) !0.804O 0.657 [0.522, 0.792] (17.075) !0.651O 0.099 [0.41, 0.609] (76.061) !0.004O

0.972 [0.789, 1.154] (15.509) !0.734O 0.774 [0.638, 0.91] (16.922) !0.72O 0.664 [0.213, 1.114] (67.92) !0.186O

0.365 [0.087, 0.642] (76.132) !0.118O

0.304 [0.006, 0.601] (75.504) !0.075O

1.018 [0.765, 1.27] (15.361) !0.618O 0.814 [0.673, 0.955] (16.203) !0.718O 0.023 [0.399, 0.353] (75.214) !0O 0.268 [0.02, 0.516] (76.394) !0.081O

0.185 [0.314, 0.684] (74.11) !0.014O 0.261 [0.124, 0.645] (77.606) !0.034O

0.701 [0.325, 1.077] (67.365) !0.267O

0.203 [0.173, 0.58] (74.552) !0.031O

0.177 [0.121, 0.475] (78.284) !0.027O

0.025 [0.257, 0.307] (78.942) !0.001O

0.859 [0.673, 1.046] (17.159) !0.684O 0.68 [0.541, 0.82] (17.602) !0.649O 0.43 [0.008, 0.869] (69.035) !0.094O 0.208 [0.135, 0.551] (75.907) !0.028O 0.471 [0.075, 1.017] (69.179) !0.072O 0.244 [0.044, 0.532] (76.932) !0.053O

0.526 [0.353, 0.699] (17.932) !0.478O 0.299 [0.169, 0.429] (19.158) !0.303O 0.87 [0.545, 1.195] (67.049) !0.458O 0.519 [0.141, 0.896] (71.9) !0.132O 0.556 [0.194, 0.918] (70.143) !0.203O 0.578 [0.195, 0.961] (70.2) !0.155O

VB, vertebral body; EP, end plate; PD, pedicle; SC, spinal canal; SP, spinous process; IT, intertransverse; F, facet; IF, interfacet; M, male; F, female. Note: Superscripts: height (H), width (W), depth (D), length (L), area (A), and angle (q). Subscripts: anterior (a), posterior (p), superior (s), inferior (i), right (r), and left (l). Note: The linear regression model is given by the equation: geometric measure5b1* (age in years)þb0. Coefficient b1 is the increase in parameter value per year, shown in the first row; lower and upper bounds of the 95% confidence interval for b1 are the values in the second row ([]); coefficient b0, the approximate parameter value at birth is the value in the third row(()); and R2 (percentage of variance explained) is the value in the fourth row (!O). Thoracic levels at which significant differences (p!.00417) were observed between genders are shaded gray.

J.R. Peters et al. / The Spine Journal 15 (2015) 1000–1020

Fql (  )

0.623 [0.381, 0.864] (19.497) !0.49O 0.747 [0.587, 0.907] (20.803) !0.647O 0.438 [0.233, 1.108] (54.14) !0.058O 0.808 [0.294, 1.322] (52.249) !0.201O

1017

1018

J.R. Peters et al. / The Spine Journal 15 (2015) 1000–1020

years. The aforementioned study also reported the height of the entire thoracic column (T1–T12) to increase by 7 to 13 mm per year in children from birth to adolescence [3]. These data can be translated to approximately 0.58 to 1.08 mm/y for each thoracic vertebra, which are comparable with the growth rates for vertebral body height (0.72– 1.08 mm/y) reported in the present study (b1 values in Table 2). Such information regarding the normative growth of vertebral features, especially the pedicles and vertebral bodies, combined with the knowledge of interlevel and between-gender differences may be used to optimize the design of orthopedic devices that could be further used to aid the growth modulation of instrumented levels. Not only are the effects of growth important considerations during the design of deformity correcting spine devices, but also during the determination of when such devices should be implemented. Particularly, for posterior spinal fusion, a common treatment method for adolescent idiopathic scoliosis, there is a considerable risk for the deformity to progress because of unequal relative growths of the anterior and posterior elements resulting from posterior arthrodesis. This progression, called the ‘‘crankshaft phenomenon,’’ could render the treatment ineffective or put the patient at risk of multiple follow-up surgeries [3,60–62]. The knowledge on the growth remaining in the spine, as well as the associated changes with level and gender provided in the present study, could better inform the method and timing of treatment for correcting spinal deformity. In addition to their design and clinical uses, these growth data can also be applied to biomechanical modeling. Previously, computational models that incorporate vertebral growth, do so by converting the vertebral growth rates or growth strains to thermal loads, which were then applied to the model [44,47–53]. The vertebral growth strain data reported in the literature are derived from partial thoracic spinal segments (T5–L5) and do not account for thoracic level- or gender-specific differences [46]. A more complete data set for the age-, thoracic level-, and gender-dependent growth strains of the various vertebral structures can be computed easily using our results. We performed an exemplar measurement of growth strain for the T5 vertebral body height for male subjects using the linear regression model presented in Table 2. The growth strain rate for vertebral body height was calculated using the equation below: Growth strain rate ðat age t yearsÞ5

b1 ðb1
tÞ þ b0

ð1Þ

Where b15the growth rate (eg mm/y), b05the approximate value of a parameter at birth, and t5subject’s age (in years). Figure 11 shows the variation in growth strain rate with age in pediatric male subjects for the T5 vertebral body height. Equation 1 predicts decreasing vertebral body height growth strain rate values with increasing age (eg, 0.0719, 0.0661, and 0.0620 per year for 8-, 9-, and 10year-old males, respectively). These data concur closely

Fig. 11. Male T5 anterior vertebral body height growth strain rate as calculated using Equation 1 and the data obtained from the linear regressions in Table 2.

with the vertebral body height growth strains used by Fok et al. [50] (0.0803, 0.0699, and 0.0603 per year for the same ages) to simulate growth of the lumbar vertebrae. Previous quantitative investigations of vertebral geometry measures have used physical/digitized 3D vertebral specimens [30–37,39,40,43] or planar medical images [39,41,63–67]. The reliability and accuracy of such methods are inherently limited by interobserver variability. Such limitations are allayed in the present study by the use of semiautomated LMP-based computation of geometric measures from CT scan reconstructions. Although the CT-based reconstruction of anatomical structures is not novel, such semiautomated methodology has the potential be integrated into popular low-dose radiographic imaging systems such as EOS (EOS imaging, Inc., Cambridge, MA, USA), which uses biplanar X-ray images for 3D reconstruction of the spine, pelvis, and lower extremities. The identification and use of corresponding sets of anatomical landmarks to record vertebral geometric measures would also help facilitate the creation of a data repository for future parametric and statistical modeling of the spine. There are some limitations of the present study. The 3D reconstruction process required extensive, time-consuming manual segmentation; however, this issue may be eschewed with the use of semiautomatic radiograph-based methods (such as EOS), which would allow for faster reconstruction and reduced radiation exposure. The highly conservative Bonferroni correction factor used in the present study may have overshadowed some statistically significant differences, otherwise observable with a less conservative approach. However, the significant age- and gender-related differences reported here hold a high degree of certainty. Additionally, the thoracic vertebral dimensions and the associated age- and gender-related variations described here cannot be extrapolated beyond the age range of the subjects included in this study (1–19 years). Data detailing age- and gender-related changes in adult thoracic spine morphology are available in the literature [4,68,69]. Although age is generally considered to be a nonlinear continuous variable, inspection of the data revealed linear trends with age for

J.R. Peters et al. / The Spine Journal 15 (2015) 1000–1020 2

most of the geometric measures. The R values may be marginally improved with the addition of nonlinear terms, however, the physical meaning of such parameters may be undetermined. In conclusion, the data from the present study would serve as a reference data set for the clinical assessment of normative growth patterns of the pediatric thoracic vertebrae, aid in the biomechanical and computational modeling of the growing spine, and also help bridge gaps in the datadriven approaches needed for the design of pediatric medical devices. Further analysis of these data could focus on identifying the principal factors influencing the size and growth of thoracic vertebral features, such as age, gender, height, weight, vertebral level, and so on. Future studies could investigate the effects of pathologic conditions on the morphology of the spine and associated structures in the pediatric population.

Acknowledgments This study was partly funded by the Scoliosis Research Society-New Investigator research grant. The authors would also like to thank Reese Juel for creating the digital artwork. References [1] Anderson M, Hwang S-C, Green WT. Growth of the normal trunk in boys and girls during the second decade of life related to age, maturity, and ossification of the iliac epiphyses. J Bone Joint Surg Am 1965;47:1554–64. [2] Dimeglio A. Growth of the spine before age 5 years. J Pediatr Orthop B 1993;1:102–7. [3] Dimeglio A, Canavese F. The growing spine: how spinal deformities influence normal spine and thoracic cage growth. Eur Spine J 2012;21:64–70. [4] Goh S, Price R, Song S, Davis S, Singer K. Magnetic resonancebased vertebral morphometry of the thoracic spine: age, gender and level-specific influences. Clin Biomech 2000;15:417–25. [5] Gold M, Dombek M, Miller PE, Emans JB, Glotzbecker MP. Prediction of thoracic dimensions and spine length on the basis of individual pelvic dimensions: validation of the use of pelvic inlet width obtained by radiographs compared with computed tomography. Spine 2014;39:74–80. [6] Masharawi YM, Kjaer P, Bendix T, Manniche C, May H, Mirovsky Y, et al. Lumbar facet and interfacet shape variation during growth in children from the general population: a three-year follow-up MRI study. Spine 2009;34:408–12. [7] Schultz AB, Sorensen S-E, Andersson GB. Measurements of spine morphology in children, ages 10-16. Spine 1984;9:70–3. [8] Veldhuizen A, Baas P, Webb P. Observations on the growth of the adolescent spine. J Bone Joint Surg Br Vol 1986;68:724–8. [9] Dede O, Demirkiran G, Yazici M. 2014 Update on the ‘‘growing spine surgery’’ for young children with scoliosis. Curr Opin Pediatr 2014;26:57–63. [10] Jain V, Lykissas M, Trobisch P, Wall E, Newton P, Sturm P, et al. Surgical aspects of spinal growth modulation in scoliosis correction. Instr Course Lect 2013;63:335–44. [11] Karatas AF, Dede O, Rogers K, Ditro CP, Holmes L, Bober M, et al. Growth-Sparing Spinal Instrumentation in Skeletal Dysplasia. Spine 2013;38:E1517–26.

1019

[12] Skaggs DL, Akbarnia BA, Flynn JM, Myung KS, Sponseller PD, Vitale MG. A Classification of growth friendly spine implants. J Pediatr Orthop 2014;34:260–74. [13] Clin J, Aubin C-E, Parent S, Sangole A, Labelle H. Comparison of the biomechanical 3D efficiency of different brace designs for the treatment of scoliosis using a finite element model. Eur Spine J 2010;19:1169–78. [14] Desroches G, Aubin C-E, Sucato DJ, Rivard C-H. Simulation of an anterior spine instrumentation in adolescent idiopathic scoliosis using a flexible multi-body model. Med Biol Eng Comput 2007;45:759–68. [15] Kim YE, Lee C, Chae SW. Biomechanical analysis of pedicle screw instrumentation in a thoracic scoliosis spine finite element model. Open Spine J 2010;2:1–7. [16] Lafon Y, Steib J-P, Skalli W. Intraoperative three dimensional correction during in situ contouring surgery by using a numerical model. Spine 2010;35:453–9. [17] Little JP, Adam C. Patient-specific computational biomechanics for simulating adolescent scoliosis surgery: predicted vs clinical correction for a preliminary series of six patients. Int J Numer Method Biomed Eng 2011;27:347–56. [18] Lopez-Valdes FJ, Lau S, Riley P, Lamp J, Kent R. The biomechanics of the pediatric and adult human thoracic spine. Paris, France: Paper presented at: Annals of Advances in Automotive Medicine/Annual Scientific Conference, 2011. [19] Majdouline Y, Aubin C-E, Sangole A, Labelle H. Computer simulation for the optimization of instrumentation strategies in adolescent idiopathic scoliosis. Med Biol Eng Comput 2009;47:1143–54. [20] P
eri
e D, Aubin C-E, Petit Y, Beaus
ejour M, Dansereau J, Labelle H. Boston brace correction in idiopathic scoliosis: a biomechanical study. Spine 2003;28:1672–7. [21] Perie D, De Gauzy JS, Hobatho M. Biomechanical evaluation of Cheneau-Toulouse-Munster brace in the treatment of scoliosis using optimisation approach and finite element method. Med Biol Eng Comput 2002;40:296–301. [22] Rohlmann A, Richter M, Zander T, Kl€ockner C, Bergmann G. Effect of different surgical strategies on screw forces after correction of scoliosis with a VDS implant. Eur Spine J 2006;15:457–64. [23] Rohlmann A, Zander T, Burra N, Bergmann G. Flexible non-fusion scoliosis correction systems reduce intervertebral rotation less than rigid implants and allow growth of the spine: a finite element analysis of different features of orthobiomÔ. Eur Spine J 2008;17:217–23. [24] Salmingo R, Tadano S, Fujisaki K, Abe Y, Ito M. Corrective force analysis for scoliosis from implant rod deformation. Clin Biomech 2012;27:545–50. [25] Salmingo RA, Tadano S, Abe Y, Ito M. Influence of implant rod curvature on sagittal correction of scoliosis deformity. Spine J 2014;14: 1432–9.
Growth considerations of the immature spine. [26] Sarwark J, Aubin CE. J Bone Joint Surg Am 2007;89(1 Suppl):8–13. [27] Wang W, Baran GR, Betz RR, Samdani AF, Pahys JM, Cahill PJ. The use of finite element models to assist understanding and treatment for scoliosis: a review paper. Spine Deformity 2014;2:10–27. [28] Wang X, Aubin C-E, Crandall D, Labelle H. Biomechanical modeling and analysis of a direct incremental segmental translation system for the instrumentation of scoliotic deformities. Clin Biomech 2011;26:548–55. [29] Wang X, Aubin C-E, Crandall D, Labelle H. Biomechanical comparison of force levels in spinal instrumentation using monoaxial versus multi degree of freedom postloading pedicle screws. Spine 2011;36:E95–104. [30] Laporte S, Mitton D, Ismael B, de Fouchecour M, Lasseau J, Lavaste F, et al. Quantitative morphometric study of thoracic spine a preliminary parameters statistical analysis. Eur J Orthop Surg Traumatol 2000;10:85–91. [31] Panjabi MM, Oxland T, Takata K, Goel V, Duranceau J, Krag M. Articular facets of the human spine quantitative three-dimensional anatomy. Spine 1993;18:1298–310.

1020

J.R. Peters et al. / The Spine Journal 15 (2015) 1000–1020

[32] Panjabi MM, Takata K, Goel V, Federico D, Oxland T, Duranceau J, et al. Thoracic human vertebrae quantitative three-dimensional anatomy. Spine 1991;16:888–901. [33] Scoles PV, Linton AE, Latimer B, Levy ME, Digiovanni BF. Vertebral body and posterior element morphology: the normal spine in middle life. Spine 1988;13:1082–6. [34] Singh R, Srivastva SK, Prasath CSV, Rohilla RK, Siwach R, Magu NK. Morphometric measurements of cadaveric thoracic spine in Indian population and its clinical applications. Asian Spine J 2011;5:20–34. [35] Tan S, Teo E, Chua H. Quantitative three-dimensional anatomy of cervical, thoracic and lumbar vertebrae of Chinese Singaporeans. Eur Spine J 2004;13:137–46. [36] Lord MJ, Ogden JA, Ganey TM. Postnatal development of the thoracic spine. Spine 1995;20:1692–7. [37] Masharawi YM, Peleg S, Albert HB, Dar G, Steingberg N, Medlej B, et al. Facet asymmetry in normal vertebral growth: characterization and etiologic theory of scoliosis. Spine 2008;33:898–902. [38] Rajwani T, Bagnall K, Lambert R, Videman T, Kautz J, Moreau M, et al. Using magnetic resonance imaging to characterize pedicle asymmetry in both normal patients and patients with adolescent idiopathic scoliosis. Spine 2004;29:E145–52. [39] Taylor J. Growth of human intervertebral discs and vertebral bodies. J Anat 1975;120(Pt 1):49. [40] Taylor J. Scoliosis and growth: patterns of asymmetry in normal vertebral growth. Acta Orthop Scand 1983;54:596–602. [41] Tian NF, Xu HZ, Wang XY, Chen QJ, Zheng LC. Morphometric comparisons between the pedicle and the pedicle rib unit in the immature Chinese thoracic spine: a computed tomographic assessment. Spine 2010;35:1514–9. [42] Zheng C, Huang Q, Hu Y, Wang X, Chen W. Computed tomographic morphometry of thoracic pedicles: safety pedicle parameter measurement of the Chinese immature thoracic spine. Int Orthop 2009;33:1663–8. [43] Zindrick MR, Knight GW, Sartori MJ, Carnevale TJ, Patwardhan AG, Lorenz MA. Pedicle morphology of the immature thoracolumbar spine. Spine 2000;25:2726–35. [44] Stokes IA, Laible JP. Three-dimensional osseo-ligamentous model of the thorax representing initiation of scoliosis by asymmetric growth. J Biomech 1990;23:589–95. [45] Stokes I. Mechanical effects on skeletal growth. J Musculoskelet Neuronal Interact 2002;2:277–80. [46] Stokes I, Burwell RG, Dangerfield PH. Biomechanical spinal growth modulation and progressive adolescent scoliosis—a test of the ‘‘vicious cycle’’ pathogenetic hypothesis: summary of an electronic focus group debate of the IBSE. Scoliosis 2006;1:7161. [47] Abolaeha O, Weber J, Ross L. Finite element simulation of a scoliotic spine with periodic adjustments of an attached growing rod. Paper presented at: Engineering in Medicine and Biology Society (EMBC), Annual International Conference of the IEEE 2012. San Diego, USA. [48] Agarwal A, Agarwal AK, Jayaswal A, Goel VK. Effect of distraction force on growth and biomechanics of the spine: a finite element study on normal juvenile spine with dual growth rod instrumentation. Spine Deformity 2014;2:260–9. [49] Driscoll M, Aubin C-E, Moreau A, Parent S. Biomechanical comparison of fusionless growth modulation corrective techniques in pediatric scoliosis. Med Biol Eng Comput 2011;49:1437–45. [50] Fok J, Adeeb S, Carey J. FEM simulation of non-progressive growth from asymmetric loading and vicious cycle theory: scoliosis study proof of concept. Open Biomed Eng J 2010;4:162–9.

[51] Huynh A-M, Aubin C-E, Rajwani T, Bagnall KM, Villemure I. Pedicle growth asymmetry as a cause of adolescent idiopathic scoliosis: a biomechanical study. Eur Spine J 2007;16:523–9. [52] Shi L, Wang D, Driscoll M, Villemure I, Chu W, Cheng J, et al. Biomechanical analysis and modeling of different vertebral growth patterns in adolescent idiopathic scoliosis and healthy subjects. Scoliosis 2011;6. [53] Villemure I, Aubin CE, Labelle H, Dansereau J. Simulation of progressive deformities in adolescent idiopathic scoliosis using a biomechanical model integrating vertebral growth modulation. J Biomech Eng 2002;124:784–90. [54] NHANES Data Center for Disease Control and Prevention. Anthropometric reference data (2011-2012) 2012. Available at: http:// www.cdc.gov/nchs/nhanes/nhanes2011-2012/BMX_G.htm. Accessed February 1, 2014. [55] Center for Disease Control and Prevention CDC. Growth charts. Washington, DC: Department of Health and Human Services, National Center for Health Statistics, 2000. [56] Aubin C-E, Dansereau J, Parent F, Labelle H, De Guise J. Morphometric evaluations of personalised 3D reconstructions and geometric models of the human spine. Med Biol Eng Comput 1997;35:611–8. [57] Masharawi Y, Salame K, Mirovsky Y, Peleg S, Dar G, Steinberg N, et al. Vertebral body shape variation in the thoracic and lumbar spine: characterization of its asymmetry and wedging. Clin Anat 2008;21:46–54. [58] Parent S, Labelle H, Skalli W, de Guise J. Thoracic pedicle morphometry in vertebrae from scoliotic spines. Spine 2004;29:239–48. [59] Pomero V, Mitton D, Laporte S, de Guise JA, Skalli W. Fast accurate stereoradiographic 3D-reconstruction of the spine using a combined geometric and statistic model. Clin Biomech 2004;19:240–7. [60] Canavese F, Dimeglio A. Normal and abnormal spine and thoracic cage development. World J Orthop 2013;4:167–74. [61] Dimeglio A. Growth in pediatric orthopaedics. J Pediatr Orthop 2001;21:549–55. [62] Dubousset J, Herring JA, Shufflebarger H. The crankshaft phenomenon. J Pediatr Orthop 1989;9:541–50. [63] C¸atan H, Buluc¸ L, Anık Y, Ayyıldız E, S¸arlak AY. Pedicle morphology of the thoracic spine in preadolescent idiopathic scoliosis: magnetic resonance supported analysis. Eur Spine J 2007;16:1203–8. [64] Cui G, Watanabe K, Hosogane N, Tsuji T, Ishii K, Nakamura M, et al. Morphologic evaluation of the thoracic vertebrae for safe free-hand pedicle screw placement in adolescent idiopathic scoliosis: a CTbased anatomical study. Surg Radiol Anat 2012;34:209–16. [65] Liljenqvist UR, Allkemper T, Hackenberg L, Link TM, Steinbeck J, Halm HF. Analysis of vertebral morphology in idiopathic scoliosis with use of magnetic resonance imaging and multiplanar reconstruction. J Bone Joint Surg Am 2002;84:359–68. [66] Liljenqvist UR, Link TM, Halm HF. Morphometric analysis of thoracic and lumbar vertebrae in idiopathic scoliosis. Spine 2000;25:1247–53. [67] O’Brien MF, Lenke LG, Mardjetko S, Lowe TG, Kong Y, Eck K, et al. Pedicle morphology in thoracic adolescent idiopathic scoliosis: is pedicle fixation an anatomically viable technique? Spine 2000;25:2285–93. [68] Morales-Avalos R, Leyva-Villegas J, S
anchez-Mejorada G,
C
ardenas-Serna M, V
ılchez-Cavazos F, Le
on AMP, et al. Age-and gender-related variations in morphometric characteristics of thoracic spine pedicle. Clin Anat 2014;27:441–50. [69] R€ uhli F, M€untener M, Henneberg M. Age-dependent changes of the normal human spine during adulthood. Am J Hum Biol 2005;17: 460–9.