- Email: [email protected]
Evaluation of pelvic morphology in the sagittal plane
The Spine Journal 13 (2013) 1500–1509
Clinical Study
Evaluation of pelvic morphology in the sagittal plane Tomaz Vrtovec, PhDa,*, Michiel M.A. Janssen, MD, PhDb, Bostjan Likar, PhDa, Rene M. Castelein, MD, PhDb, Max A. Viergever, PhDc, Franjo Pernus, PhDa a
University of Ljubljana, Faculty of Electrical Engineering, Laboratory of Imaging Technologies, Trzaska cesta 25, SI-1000 Ljubljana, Slovenia b University Medical Center Utrecht, Department of Orthopaedics, Heidelberglaan 100, 3584 CX, Utrecht, The Netherlands c University Medical Center Utrecht, Image Sciences Institute, Heidelberglaan 100, 3584 CX, Utrecht, The Netherlands Received 26 September 2012; revised 26 March 2013; accepted 15 June 2013
Abstract
BACKGROUND CONTEXT: It is generally accepted that for normal subjects the angle of pelvic incidence (PI) increases during childhood and then remains unchanged throughout adolescence and adulthood. However, recent findings show that PI increases linearly throughout the lifespan due to morphological changes of the pelvis. PURPOSE: A retrospective study aiming to determine the extent of morphological changes of the pelvis related to the age of the subjects. STUDY DESIGN: Pelvic morphology was evaluated in a normal adult population by measuring the anatomical parameters of sagittal pelvic alignment. PATIENT SAMPLE: The final study cohort consisted of 330 subjects (mean age, 45.3 years; standard deviation, 18.1 years; range, 18–87 years; 164 male and 166 female subjects). OUTCOME MEASURES: Physiologic measures, obtained as measurements of PI, sacral end plate width (S1W), and pelvic thickness (PTH). METHODS: Parameters of PI, S1W, and PTH were evaluated from computed tomography images of the subjects. The measured PTH was normalized according to S1W and age of the subjects, allowing the comparison among anatomies of different sizes. The normalized components of PTH in anteroposterior and cephalocaudal directions were computed to determine the configuration and extent of changes in pelvic morphology related to subject age. RESULTS: Statistically significant correlation with both age and PI was obtained for all normalized parameters (except for the anteroposterior component of PTH for male subjects), and no statistically significant differences were observed between the sexes. With increasing PI that occurs due to the aging process, a decrease of PTH can be observed that is manifested not only as an increase of the distance between the sacrum and the hip axis in the anterior direction but considerably more as a decrease of the distance between the sacrum and the hip axis in the cephalic direction. By considering these morphological changes in the pelvis simultaneously, the hip axis can move only within a narrow area. CONCLUSIONS: The changes in pelvic morphology due to the aging process occur in the anterior direction, which may be due to the remodeling process affecting the coxal bone that results in an anterior drift of the acetabulum relative to the sacrum. More importantly, the changes are considerably more evident in the cephalic direction, which may be the result of the weight-bearing loads and consequent wear of acetabular cartilage. Ó 2013 Elsevier Inc. All rights reserved.
Keywords:
Pelvic morphology; Sagittal alignment; Pelvic incidence; Pelvic thickness
FDA device/drug status: Not applicable. Author disclosures: TV: Nothing to disclose. MMAJ: Nothing to disclose. BL: Nothing to disclose. RMC: Speaking/Teaching Arrangements: Medtronic (A); Research Support (Investigator Salary): Synthes (D, Paid directly to institution/employer). MAV: Grant: Philips Healthcare (E, Paid directly to institution/employer). FP: Nothing to disclose. 1529-9430/$ - see front matter Ó 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.spinee.2013.06.034
The disclosure key can be found on the Table of Contents and at www. TheSpineJournalOnline.com. * Corresponding author. University of Ljubljana, Faculty of Electrical Engineering, Laboratory of Imaging Technologies, Trzaska cesta 25, SI1000 Ljubljana, Slovenia. Tel.: þ386 1-4768-327; fax: þ386 1-4768-279. E-mail address: [email protected] (T. Vrtovec)
T. Vrtovec et al. / The Spine Journal 13 (2013) 1500–1509
Introduction Pelvic morphology has a significant role in the spinopelvic relationship, as it is considered to influence the spinal balance and postural equilibrium, and has been recognized as an essential component in the regulation of sagittal plane alignment [1–6]. In contrast to positional parameters, such as the sacral slope or pelvic tilt, pelvic morphology is best evaluated by anatomical parameters of pelvic alignment in the sagittal plane, which are constant and unchanged for each individual subject and describe the position of the sacrum in relation to the pelvis [7,8]. Since its introduction by Duval-Beaupere et al. [9], the angle of pelvic incidence (PI) has gained large attention and acceptance as one of the key parameters in the complex framework of sagittal spinal alignment and related deformities [10–13]. By being an anatomical parameter, PI describes structural characteristics of the sacrum and pelvis that are not modified by posture [1,14] and can be therefore observed and compared among subjects in either standing, sitting, or supine position. For normal subjects, it is generally accepted that there is no difference in PI between male and female subjects, that PI increases during childhood and then remains unchanged throughout adolescence and adulthood [2,15], and that PI is not significantly influenced by normal degenerative changes of the hips, sacrum, and sacroiliac joints [16]. However, Mendoza-Lattes et al. [17] recently suggested that PI continues to increase linearly even after skeletal maturity and throughout the lifespan as a result of the increasing distance between the sacrum and the femoral heads, suggesting a morphological change of the pelvis. If the femoral heads are in a relatively more anterior position, then PI is consequently going to present larger angular values. The purpose of this study was to observe whether there are morphological changes of the pelvis related to the age of the subjects as described by Mendoza-Lattes et al. [17]. By measuring the anatomical parameters of sagittal pelvic alignment in a normal adult population, the configuration of pelvic morphology and the extent of changes that occur due to the aging process are evaluated.
1501
body parameters were not recorded for scan acquisition, and the original DICOM files were further anonymized by removing all confidential information and retaining only subject sex and age. The following exclusion criteria were applied to the initial cohort: age !18 years (skeletal maturity not reached) and age O90 years (small population ratio, inadequate image quality due to bone aging); prior hip or spine surgery; any clinical or radiological history or evidence for pathology or trauma of the spine, pelvis, or femur; evidence for anatomical anomalies (eg, abnormal number of lumbar vertebrae); syndromes associated with disorder of growth; any form of psychomotor retardation; poor quality of the scan (eg, pelvis not fully included, artifacts); pregnancy. As a result, 100 subjects were excluded from the initial cohort and the remaining 330 subjects represented the final study cohort. Measurement of pelvic parameters To measure pelvic parameters that allow the evaluation of pelvic morphology, the following anatomical references were determined in the CT scan of each subject: the center of the left femoral head, the center of the right femoral head, the center of the sacral end plate, and the inclination of the sacral end plate (Fig. 1). Multiplanar 3D image reformation was then performed to obtain the superposition of
Materials and methods Subjects The initial cohort consisted of 430 subjects who received a computed tomography (CT) scan of the pelvis or abdomen in the emergency room of the (University Medical Center Utrecht, the Netherlands) during a 5-year period (2005–2010) for reasons such as trauma or acute abdominal pathology. Computed tomography scans (Brilliance 16 and 64 scanners, Philips Healthcare, The Netherlands) were axially reconstructed three-dimensional (3D) images (pixel size, 0.4–1.0 mm; slice thickness, 3.0–4.0 mm), showing the femoral heads, the whole pelvis, and at least the lower lumbar region of the spine. Subject height, weight, or other
Fig. 1. Pelvic parameters in the sagittal plane are used to evaluate pelvic morphology.
1502
T. Vrtovec et al. / The Spine Journal 13 (2013) 1500–1509
the femoral heads in the sagittal view, so that all anatomical structures were completely in line with the hip axis. In such perfect sagittal view, the following anatomical pelvic parameters were measured: Pelvic incidence: The angle between the line orthogonal to the sacral end plate inclination and the line connecting the center of the sacral end plate with the hip axis [9]. Sacral end plate width (S1W): The length of the sacral end plate in the direction of its inclination. Pelvic thickness (PTH): The distance between the center of the sacral end plate and the hip axis [9]. By considering the measured PI and PTH within an orthogonal triangle, the anteroposterior (PTHA-P) and cephalocaudal (PTHC-C) components of PTH were in addition calculated as (Fig. 1): PTHAP 5PTH sinðPIÞ; PTHCC 5PTH cosðPIÞ:
ð1Þ
Evaluation of pelvic morphology Although the measured pelvic parameters are constant for each individual subject, it is not appropriate to directly compare distance parameters (ie, S1W and PTH) among subjects because they depend on individual body size. As a result, for two subjects with equal anatomy shape but different anatomy size (ie, equal PI but different S1W and PTH), distance parameters may be different and therefore have to be normalized before they are compared. As we are interested in changes of PTH and its components, the normalization is performed according to S1W, for which we assume that it reflects anatomy size. The normalized pelvic thickness (nPTH) and its normalized anteroposterior (nPTHA-P) and cephalocaudal (nPTHC-C) components are therefore defined for each subject as: mS1WðageÞ 100% ; S1W aPTH
nPTHAP ð%Þ5PTHAP nPTHCC ð%Þ5PTHCC
After dividing by aPTH, aPTHA-P, or aPTHC-C and multiplying by 100%, the resulting value represents the percentage of variation of the observed parameter around its average in the studied population. After dividing by S1W and multiplying by mS1W(age), the influence of different anatomy size is removed. The average S1W is expected to increase with subject age because the sacral end plate adapts to the weight-bearing loads. Normalization of PTH is therefore not performed according to the average S1W but the modeled expected average S1W for the observed age of the subject (ie, mS1W(age)). Statistical analysis
The measured pelvic parameters are constant for the observed subject, do not change with subject position and/or orientation, and can be therefore measured from CT images acquired in patient supine position. To determine the exact location of the aforementioned anatomical references in 3D images and therefore measure the previously described pelvic parameters in CT scans, we used a computerized method based on image processing techniques that was introduced and validated in a recently published study [18].
nPTHð%Þ5PTH
where mS1W(age) is the modeled expected average S1W for the observed age of the subject and aPTH, aPTHA-P, and aPTHC-C are the average PTH, average anteroposterior PTH component, and average cephalocaudal PTH component across population, respectively. The effects of such normalization are twofold:
100% mS1WðageÞ ; S1W aPTHAP mS1WðageÞ 100% ; S1W aPTHCC ð2Þ
The obtained measurements were ordered according to age and sex of the subjects and analyzed in terms of mean and standard deviation (SD). Regression lines between any two observed variables were determined by the random sample consensus method (assumed ratio of outliers, 25%; number of iterations, 1,000). Pearson correlation analysis was performed to search for statistically significant correlations between a selected measurement and age or PI. The independent samples t test was used to evaluate statistical differences between the sexes (level of significance a50.05).
Results Subjects The final study cohort of 330 subjects (mean age, 45.3 years; SD, 18.1 years; range, 18–87 years) was divided according to sex into the group of 164 male subjects (mean age, 44.3 years; SD, 18.0 years; range, 18–87 years) and the group of 166 female subjects (mean age, 46.2 years; SD, 18.1 years; range, 18–86 years). No statistically significant differences in age were found between the sexes (pO.05). Measurement of pelvic parameters Pelvic parameters were measured by a computerized method that was introduced and validated in a recently published study [18] and proved to be highly consistent with manual measurements (SD of up to 1.8 mm for distances and 3.5 for angles). Fig. 2 shows examples of measurements, whereas Fig. 3 and Table 1 show the obtained measurements for PI, S1W, and PTH. The correlation of each parameter with age was relatively low but statistically
T. Vrtovec et al. / The Spine Journal 13 (2013) 1500–1509
1503
Fig. 2. Measurement of pelvic parameters in three dimensions is performed by applying a computerized method to computed tomography images. The results for (Left) a 48-year-old male and (Right) a 31-year-old female subject are shown in the perfect sagittal view for the maximal intensity projection (top) and midsagittal cross-section (bottom).
significant (p!.05). The distribution of S1W measurements shows a distinctive difference between male and female subjects, which was expected because of anatomy size, and also a distinctive increasing tendency of the average S1W with age (Fig. 3, Middle), which was expected because of the adaptation of the sacral end plate to the weight-bearing loads. Because a statistically significant difference between the sexes was found for S1W (p!.0001), the expected average S1W for the observed age of the subject (ie, mS1W(age)) was modeled separately for male and female subjects. The obtained linear regression model using random sample consensus was mS1W(age)50.08ageþ33.25 for male subjects and mS1W(age)50.11ageþ28.19 for female subjects. The modeled expected average S1W for the two examples shown in Fig. 2 were therefore mS1W(48)5
0.0848þ33.25537.1 mm for the 48-year-old male subject (Fig. 2, Left) and mS1W(31)50.1131þ28.19531.6 mm for the 31-year-old female subject (Fig. 2, Right). On the contrary, no significant differences were found between the sexes for PTH and its components PTHA-P and PTHC-C (pO.05); therefore, the average values aPTH5104.9, aPTHA-P575.9, and aPTHC-C570.1 mm were computed together for male and female subjects. Evaluation of pelvic morphology The measured PTH and its components PTHA-P and PTHC-C were normalized according to their average values (ie, aPTH, aPTHA-P, and aPTHC-C, respectively) and according to the modeled expected average S1W for the
1504
T. Vrtovec et al. / The Spine Journal 13 (2013) 1500–1509 Table 1 Pelvic incidence, S1W, and PTH, as measured for all subjects (N5330) and separately for male (N5164) and female (N5166) subjects Vs. age* Pelvic parameter
Mean
SD
r
p
py
PI ( ) Male Female S1W (mm) Male Female PTH (mm) Male Female
47.6 47.3 47.9 35.2 37.0 33.4 104.9 104.5 105.4
10.2 9.5 10.8 4.4 4.1 4.0 8.6 8.3 8.9
þ0.25 þ0.16 þ0.32 þ0.38 þ0.37 þ0.51 0.26 0.27 0.25
!.0001 !.05 !.0001 !.0001 !.0001 !.0001 !.0001 !.05 !.05
O.05 !.0001 O.05
PI, pelvic incidence; S1W, sacral end plate width; PTH, pelvic thickness; SD, standard deviation. * Correlation coefficient r and significance level p in relation to age. y Significance level p between male and female subjects.
Discussion
Fig. 3. (Top) PI, (Middle) S1W, and (Bottom) PTH in relation to age for male and female subjects with superimposed regression lines. PI, pelvic incidence; S1W, sacral end plate width; PTH, pelvic thickness.
observed age of the subject (ie, mS1W(age)). Fig. 4 and Table 2 show the obtained results in relation to the age of the subjects. Statistically significant correlation with age was obtained for all parameters, except for nPTHA-P for male subjects. Fig. 5 and Table 2 show the obtained results in relation to PI of the subjects. Statistically significant correlation with PI was obtained for all parameters. By comparing the resulting values between the sexes, no statistically significant differences were observed for all three parameters (pO.05).
For normal subjects, it is generally accepted that PI increases during the first months of life, becomes stable at around 10 years of age, and then remains unchanged throughout adolescence and adulthood. However, it may be altered by pathologic processes that modify the shape of the sacrum or the position of the acetabulae within the pelvis [15], such as spondylolisthesis or spondylolisis [2,19–23], scoliosis [1,24–26], and other spinopelvic pathologies [5,27–32]. Studies were conducted to determine the difference in PI between male and female subjects, reporting in general no relationship between PI and sex [6,9,14–16,20,33–35] but in some cases higher but not statistically significant values for male [1,17,36] or female [33,37] subjects. On the contrary, Vialle et al. [4] reported statistically significant differences between the sexes, with higher PI for female subjects. In examining the relationship between PI and age of the subjects, studies reported no relationship in fetuses [15], adults [4,15,20], elderly population [36], or various age groups [38], but PI was found to be higher in normal elderly [36,38] than normal adult population [1,2,4,6,9,14,35], and a relationship between PI and age was found for children [15,19,33,37] and adolescents [33,37]. Recent studies also reported that aging or degenerative processes, manifested as a progressive loss in the height of intervertebral discs, induce compensation in the sagittal balance by a backward rotation (ie, retroversion) of the pelvis [39], which is for an individual limited by PI that remains constant [40]. When considering the size of the pelvis, a statistically significant difference was reported between the sexes for PTH [41] and a significant negative correlation between PTH and PI [42], meaning that an increase in PI is coupled with a decrease in PTH. In contrast to the generally accepted views, by comparing the results of different studies Mendoza-Lattes et al. [17] recently concluded that PI increases linearly throughout the lifespan of normal subjects due to a remodeling process affecting the coxal bone (ie, the hip bone consisting of
T. Vrtovec et al. / The Spine Journal 13 (2013) 1500–1509
1505
Table 2 Normalized pelvic thickness, nPTHA-P, and nPTHC-C, as evaluated for all subjects (N5330) and separately for male (N5164) and female (N5166) subjects Vs. PIy
Vs. age* Pelvic parameter
SD
r
p
r
p
pz
nPTH (%) Male Female nPTHA-P (%) Male Female nPTHC-C (%) Male Female
13.2 11.7 11.7 17.8 17.2 16.6 27.1 23.9 28.0
0.18 0.19 0.19 þ0.11 þ0.01 þ0.22 0.27 0.21 0.33
!.05 !.05 !.05 !.05 O.05 !.05 !.0001 !.05 !.0001
0.27 0.20 0.39 þ0.70 þ0.75 þ0.74 0.87 0.86 0.91
!.0001 !.05 !.0001 !.0001 !.0001 !.0001 !.0001 !.0001 !.0001
O.05 O.05 O.05
nPTH, normalized pelvic thickness; nPTHA-P, normalized anteroposterior component of pelvic thickness; nPTHC-C, normalized cephalocaudal component of pelvic thickness; PI, pelvic incidence; SD, standard deviation around the 100% mean. * Correlation coefficient r and significance level p in relation to age. y Correlation coefficient r and significance level p in relation to PI. z Significance level p between male and female subjects.
Fig. 4. (Top) nPTH, (Middle) nPTHA-P, and (Bottom) nPTHC-C in relation to age for male and female subjects with superimposed regression lines. nPTH, normalized pelvic thickness; nPTHA-P, normalized anteroposterior component of pelvic thickness; nPTHC-C, normalized cephalocaudal component of pelvic thickness.
the ilium, ischium, and pubis), which results in an anterior drift of the acetabulum relative to the sacroiliac joint. They explained the gradual increase of PI throughout the lifespan by an increase of the sacropelvic translation [43], defined as the horizontal distance between the reference vertical through the center of the femoral head and the reference vertical through the posterior corner of the sacral end plate.
The horizontal distance between the sacrum and the hip axis can also be described by the pelvic overhang [9], if the distance is measured against the reference vertical through the center of the sacral end plate, or by the sacrofemoral distance [44], if the distance is measured against the reference vertical through the anterior corner of the sacral end plate. As a result, if the femoral heads move to a relatively more anterior position, PI is consequently going to present larger angular values. The results of our study confirm the conclusions of Mendoza-Lattes et al. [17] and also bring new insight into the evaluation of pelvic morphology. The increase of the horizontal distance between the sacrum and the hip axis, as suggested by Mendoza-Lattes et al. [17], implies an increase of the straight distance between the sacrum and the hip axis, represented by PTH. However, our study shows that with increasing PI that occurs due to the aging process (Fig. 3, Top), PTH decreases (Fig. 4, Top, and Fig. 5, Top). The decrease of PTH is manifested not only as an increase of the distance between the sacrum and the hip axis in the anterior direction (ie, PTHA-P; Fig. 4, Middle, and Fig. 5, Middle) but considerably more as a decrease of the distance between the sacrum and the hip axis in the cephalic direction (ie, PTHC-C; Fig. 4, Bottom, and Fig. 5, Bottom). The horizontal distance between the sacrum and the hip axis can be, to some extent, compared with PTHA-P in our study. Our results show that PTHA-P in general increases with age for female subjects but slightly decreases with age for male subjects. The same tendency can be observed for PTHC-C, which manifests a stronger decrease for female than for male subjects. Nevertheless, by considering these morphological changes in the pelvis simultaneously (ie, increase of PI, decrease of PTH, increase of PTHA-P, and decrease of PTHC-C), the hip axis can move only within a narrow area (Fig. 6). These changes therefore
1506
T. Vrtovec et al. / The Spine Journal 13 (2013) 1500–1509
Fig. 5. (Top) nPTH, (Middle) nPTHA-P, and (Bottom) nPTHC-C in relation to PI for male and female subjects with superimposed regression lines. nPTH, normalized pelvic thickness; nPTHA-P, normalized anteroposterior component of pelvic thickness; nPTHC-C, normalized cephalocaudal component of pelvic thickness; PI, pelvic incidence.
occur in the anterior direction, which may be due to the remodeling process affecting the coxal bone that results in an anterior drift of the acetabulum relative to the sacrum. More importantly, the changes are considerably more evident in the cephalic direction, which may be the result of the weight-bearing loads and consequent wear of acetabular cartilage.
The main shortcoming of the performed study is that it is cross-sectional and therefore the differentiation between cause and effect is difficult. However, performing a longitudinal study and therefore radiographically screening a large number of normal individuals throughout their lifespan is not feasible and would be questionable in terms of costs and medical ethics. Regarding cause and effect, there is no doubt that the increasing PI is the effect, while the cause is represented by the morphological changes of the pelvis, observed through pelvic parameters. It is, however, important to note that assuming the size of anatomy is reflected in the size of the sacrum, and therefore S1W, can be treated as adequate only for cases of normal anatomy, or anatomy that is considered normal. In the case of severe abnormalities or pathologies, for example, dome-shaped sacral end plate in the case of spondylolysis [45], sacral dysplasia in the case of spondylolisthesis [46] or even sacral insufficiency fractures [47], the corresponding expected values mS1W(age) would have to be modeled according to the studied population. Although the observed cohort of subjects was assumed to represent a normal population, several subjects were excluded from the original cohort due to several criteria (ie, age, clinical, or radiological history or evidence for pathology or trauma, evidence for anatomical anomalies, and so on), and imposing of additional criteria may result in further exclusion of subjects. Unfortunately, criteria such as activity (eg, sedentary vs. mobile lifestyle, practicing sports, and so on), body mass index, or other physiological characteristics were not recorded along with other subject information but may nevertheless influence the sagittal pelvic alignment [19,34]. However, the natural biological variability of the human anatomy is relatively large, which is also reflected in the large variability of the measured parameters that describe pelvic morphology. Ascertaining a truly normal and asymptomatic population therefore remains poorly defined and represents a limitation of the present as well as of the aforementioned studies. In contrast to current routine clinical protocols, the measurements in this study were obtained in 3D from volumetric CT images of the pelvis. This does not imply that pelvic morphology should not be evaluated in two dimensions from plain sagittal radiographic images; however, computerized measurements in 3D proved to be highly consistent and reliable, especially because they were obtained from sagittal views after multiplanar reformation according to the hip axis. In such perfect sagittal views, all anatomical structures were observed completely in line with the hip axis and were therefore not biased by the projective nature of radiographic image acquisition [18]. Because CT images are acquired in subject supine position, the obtained findings cannot be transferred to positional pelvic parameters (eg, sacral slope, pelvic tilt, sacropelvic translation, pelvic overhang, sacrofemoral distance) but need to be contained within anatomical pelvic parameters (eg, PI, PTH) that are invariant against arbitrary patient position and/or orientation.
T. Vrtovec et al. / The Spine Journal 13 (2013) 1500–1509
1507
Fig. 6. (A) If PI increases, the hip axis must move into the shown area. (B) If PTH decreases, the hip axis must move into the shown area. (C) If PTHA-P increases and PTHC-C decreases, the hip axis must move into the shown area. (D) If all changes occur simultaneously, the hip axis must be located within the intersection of these areas, that is, within the shown area. PI, pelvic incidence; PTH, pelvic thickness; PTHA-P, anteroposterior component of pelvic thickness; PTHC-C, cephalocaudal component of pelvic thickness.
Conclusion Despite the previously described shortcomings, this study revealed that there are indications for morphological changes in the pelvis that continue to occur throughout the lifespan and reflect the effort of the human body to maintain adequate sagittal balance. The pelvis is also subject to a posteriorly inferiorly directed load that leads to bony remodeling changes according to Wolff’s law [48,49], which asserts that bone in a healthy person will adapt to the loads it is placed under to maximize mechanical efficiency. According to the Utah paradigm of skeletal physiology [50–52], which is a refinement of Wolff’s law, bone growth and bone loss are stimulated by the local mechanical elastic deformation of bone. The adaptation (feedback control loop) of bone according to the maximum forces is considered to be a lifelong process; hence, bone adapts
its mechanical properties according to the needed mechanical function—bone mass, geometry, and strength are therefore adapted according to the everyday usage and needs.
Acknowledgments This work has been supported by the Ministry of Education, Science, Culture and Sport, Slovenia, under grants P2-0232, J7-2264, L2-7381, and L2-2023, and partly by Biomet and the Anna Foundation. References [1] Legaye J, Duval-Beaupere G, Hecquet J, Marty C. Pelvic incidence: a fundamental pelvic parameter for three-dimensional regulation of spinal sagittal curves. Eur Spine J 1998;7:99–103.
1508
T. Vrtovec et al. / The Spine Journal 13 (2013) 1500–1509
[2] Labelle H, Roussouly P, Berthonnaud E, et al. Spondylolisthesis, pelvic incidence, and spinopelvic balance: a correlation study. Spine 2004;29:2049–54. [3] Roussouly P, Gollogly S, Berthonnaud E, Dimnet J. Classification of the normal variation in the sagittal alignment of the human lumbar spine and pelvis in the standing position. Spine 2005;30: 346–53. [4] 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 2005;87:260–7. [5] Lafage V, Schwab F, Skalli W, et al. Standing balance and sagittal plane spinal deformity: analysis of spinopelvic and gravity line parameters. Spine 2008;33:1572–8. [6] Mac-Thiong JM, Roussouly P, Berthonnaud E, Guigui P. Sagittal parameters of global spinal balance: normative values from a prospective cohort of seven hundred nine Caucasian asymptomatic adults. Spine 2010;35:E1193–8. [7] Vrtovec T, Janssen M, Likar B, et al. A review of methods for evaluating the quantitative parameters of sagittal pelvic alignment. Spine J 2012;12:433–46. [8] Cheng I. Commentary: spinopelvic parameters-how far have we come? Spine J 2012;12:447–8. [9] Duval-Beaupere G, Schmidt C, Cosson P. A barycentremetric study of the sagittal shape of spine and pelvis: the conditions required for an economic standing position. Ann Biomed Eng 1992;20:451–62. [10] Roussouly P, Nnadi C. Sagittal plane deformity: an overview of interpretation and management. Eur Spine J 2010;19:1824–36. [11] Schwab F, Patel A, Ungar B, et al. Adult spinal deformity—postoperative standing imbalance: how much can you tolerate? An overview of key parameters in assessing alignment and planning corrective surgery. Spine 2010;35:2224–31. [12] Lafage V, Schwab F, Vira S, et al. Spino-pelvic parameters after surgery can be predicted: a preliminary formula and validation of standing alignment. Spine 2011;36:1037–45. [13] Mac-Thiong JM, Duong L, Parent S, et al. Reliability of the Spinal Deformity Study Group classification of lumbosacral spondylolisthesis. Spine 2012;37:E95–102. [14] Vaz G, Roussouly P, Berthonnaud E, Dimnet J. Sagittal morphology and equilibrium of pelvis and spine. Eur Spine J 2002;11:80–7. [15] Mangione P, Gomez D, Senegas J. Study of the course of the incidence angle during growth. Eur Spine J 1997;6:163–7. [16] Mac-Thiong JM, Roussouly P, Berthonnaud E, Guigui P. Age- and sex-related variations in sagittal sacropelvic morphology and balance in asymptomatic adults. Eur Spine J 2011;20:S572–7. [17] Mendoza-Lattes S, Ries Z, Gao Y, Weinstein S. Natural history of spinopelvic alignment differs from symptomatic deformity of the spine. Spine 2010;35:E792–8. [18] Vrtovec T, Janssen M, Pernus F, et al. Analysis of pelvic incidence from 3-dimensional images of a normal population. Spine 2012;37: E479–85. [19] Hanson D, Bridwell K, Rhee J, Lenke L. Correlation of pelvic incidence with low- and high-grade isthmic spondylolisthesis. Spine 2002;27:2026–9. [20] Marty C, Boisaubert B, Descamps H, et al. The sagittal anatomy of the sacrum among young adults, infants, and spondylolisthesis patients. Eur Spine J 2002;11:119–25. [21] Huang R, Bohlman H, Thompson G, Poe-Kochert C. Predictive value of pelvic incidence in progression of spondylolisthesis. Spine 2003;28:2381–5. [22] Labelle H, Roussouly P, Berthonnaud E, et al. The importance of spino-pelvic balance in L5-S1 developmental spondylolisthesis: a review of pertinent radiologic measurements. Spine 2005;30:S27–34. [23] Roussouly P, Gollogly S, Berthonnaud E, et al. Sagittal alignment of the spine and pelvis in the presence of L5-S1 isthmic lysis and lowgrade spondylolisthesis. Spine 2006;31:2484–90.
[24] Mac-Thiong JM, Labelle H, Charlebois M, et al. Sagittal plane analysis of the spine and pelvis in adolescent idiopathic scoliosis according to the coronal curve type. Spine 2003;28:1404–9. [25] Hong JY, Suh SW, Modi H, et al. Correlation of pelvic orientation with adult scoliosis. J Spinal Disord Tech 2010;23:461–6. [26] Lonner B, Auerbach J, Sponseller P, et al. Variations in pelvic and other sagittal spinal parameters as a function of race in adolescent idiopathic scoliosis. Spine 2010;35:E374–7. [27] Barrey C, Jund J, Noseda O, Roussouly P. Sagittal balance of the pelvis-spine complex and lumbar degenerative diseases: a comparative study about 85 cases. Eur Spine J 2007;16:1459–67. [28] Hart R, Badra M, Madala A, Yoo J. Use of pelvic incidence as a guide to reduction of H-type spino-pelvic dissociation injuries. J Orthop Trauma 2007;21:369–74. [29] Abelin K, Vialle R, Lenoir T, et al. The sagittal balance of the spine in children and adolescents with osteogenesis imperfecta. Eur Spine J 2008;17:1697–704. [30] Rose P, Bridwell K, Lenke L, et al. Role of pelvic incidence, thoracic kyphosis, and patient factors on sagittal plane correction following pedicle subtraction osteotomy. Spine 2009;34:785–91. [31] Hong JY, Suh SW, Modi H, et al. Analysis of sagittal spinopelvic parameters in achondroplasia. Spine 2011;36:E1233–9. [32] Videbaek T, B€unger C, Henriksen M, et al. Sagittal spinal balance after lumbar spinal fusion: the impact of anterior column support: results from a randomized clinical trial with an eight- to thirteenyear radiographic follow-up. Spine 2011;36:183–91. [33] Mac-Thiong JM, Berthonnaud E, Dimar J, et al. Sagittal alignment of the spine and pelvis during growth. Spine 2004;29:1642–7. [34] Boulay C, Tardieu C, Hecquet J, et al. Sagittal alignment of spine and pelvis regulated by pelvic incidence: standard values and prediction of lordosis. Eur Spine J 2006;15:415–22. [35] Janssen M, Drevelle X, Humbert L, et al. Differences in male and female spino-pelvic alignment in asymptomatic young adults: a three-dimensional analysis using upright low-dose digital biplanar X-rays. Spine 2009;34:E826–32. [36] Hammerberg E, Wood K. Sagittal profile of the elderly. J Spinal Disord Tech 2003;16:44–50. [37] Mac-Thiong JM, Labelle H, Berthonnaud E, et al. Sagittal spinopelvic balance in normal children and adolescents. Eur Spine J 2007;16:227–34. [38] Schwab F, Lafage V, Boyce R, et al. Gravity line analysis in adult volunteers: age-related correlation with spinal parameters, pelvic parameters, and foot position. Spine 2006;31:E959–67. [39] Le Huec J, Saddiki R, Franke J, et al. Equilibrium of the human body and the gravity line: the basics. Eur Spine J 2011;20: S558–63. [40] Roussouly P, Pinheiro-Franco J. Biomechanical analysis of the spino-pelvic organization and adaptation in pathology. Eur Spine J 2011;20:S609–18. [41] Rajnics P, Pomero V, Templier A, et al. Computer-assisted assessment of spinal sagittal plane radiographs. J Spinal Disord 2001;14:135–42. [42] Boulay C, Tardieu C, Hecquet J, et al. Anatomical reliability of two fundamental radiological and clinical pelvic parameters: incidence and thickness. Eur J Orthop Surg Traumatol 2005;15:197–204. [43] Jackson R, Peterson M, McManus A, Hales C. Compensatory spinopelvic balance over the hip axis and better reliability in measuring lordosis to the pelvic radius on standing lateral radiographs of adult volunteers and patients. Spine 1998;23:1750–67. [44] Takemitsu Y, Harada Y, Iwahara T, et al. Lumbar degenerative kyphosis: clinical, radiological and epidemiological studies. Spine 1988;13:1317–26. [45] Legaye J. The femoro-sacral posterior angle: an anatomical sagittal pelvic parameter usable with dome-shaped sacrum. Eur Spine J 2007;16:219–25.
T. Vrtovec et al. / The Spine Journal 13 (2013) 1500–1509 [46] Vialle R, Dauzac C, Khouri N, et al. Sacral and lumbar-pelvic morphology in highgrade spondylolisthesis. Orthopedics 2007;30:642–9. [47] Shah R. Sacral kyphoplasty for the treatment of painful sacral insufficiency fractures and metastases. Spine J 2012;12:113–20. [48] Wolff J. On the inner architecture of bones and its importance for bone growth (Ueber die innere Architectur der Knochen und ihre Bedeutung f€ ur die Frage vom Knochenwachsthum). Virchows Arch 1870;50:389–450. [49] Wolff J. The classic: on the inner architecture of bones and its importance for bone growth. (reprinted from Wolff J.: Ueber die innere Architectur der Knochen und ihre Bedeutung f€ur die Frage vom
1509
Knochenwachsthum. Virchows Archiv, 50:389–450, 1870). Clin Orthop Relat Res 2010;468:1056–65. [50] Frost H. Skeletal structural adaptations to mechanical usage (SATMU): 1. Redefining Wolff’s law: the bone modeling problem. Anat Rec 1990;226:403–13. [51] Frost H. Skeletal structural adaptations to mechanical usage (SATMU): 2. Redefining Wolff’s law: the remodeling problem. Anat Rec 1990;226:414–22. [52] Frost H. The Utah paradigm of skeletal physiology: an overview of its insights for bone, cartilage and collagenous tissue organs. J Bone Miner Metab 2000;18:305–16.
Clinical Study
Evaluation of pelvic morphology in the sagittal plane Tomaz Vrtovec, PhDa,*, Michiel M.A. Janssen, MD, PhDb, Bostjan Likar, PhDa, Rene M. Castelein, MD, PhDb, Max A. Viergever, PhDc, Franjo Pernus, PhDa a
University of Ljubljana, Faculty of Electrical Engineering, Laboratory of Imaging Technologies, Trzaska cesta 25, SI-1000 Ljubljana, Slovenia b University Medical Center Utrecht, Department of Orthopaedics, Heidelberglaan 100, 3584 CX, Utrecht, The Netherlands c University Medical Center Utrecht, Image Sciences Institute, Heidelberglaan 100, 3584 CX, Utrecht, The Netherlands Received 26 September 2012; revised 26 March 2013; accepted 15 June 2013
Abstract
BACKGROUND CONTEXT: It is generally accepted that for normal subjects the angle of pelvic incidence (PI) increases during childhood and then remains unchanged throughout adolescence and adulthood. However, recent findings show that PI increases linearly throughout the lifespan due to morphological changes of the pelvis. PURPOSE: A retrospective study aiming to determine the extent of morphological changes of the pelvis related to the age of the subjects. STUDY DESIGN: Pelvic morphology was evaluated in a normal adult population by measuring the anatomical parameters of sagittal pelvic alignment. PATIENT SAMPLE: The final study cohort consisted of 330 subjects (mean age, 45.3 years; standard deviation, 18.1 years; range, 18–87 years; 164 male and 166 female subjects). OUTCOME MEASURES: Physiologic measures, obtained as measurements of PI, sacral end plate width (S1W), and pelvic thickness (PTH). METHODS: Parameters of PI, S1W, and PTH were evaluated from computed tomography images of the subjects. The measured PTH was normalized according to S1W and age of the subjects, allowing the comparison among anatomies of different sizes. The normalized components of PTH in anteroposterior and cephalocaudal directions were computed to determine the configuration and extent of changes in pelvic morphology related to subject age. RESULTS: Statistically significant correlation with both age and PI was obtained for all normalized parameters (except for the anteroposterior component of PTH for male subjects), and no statistically significant differences were observed between the sexes. With increasing PI that occurs due to the aging process, a decrease of PTH can be observed that is manifested not only as an increase of the distance between the sacrum and the hip axis in the anterior direction but considerably more as a decrease of the distance between the sacrum and the hip axis in the cephalic direction. By considering these morphological changes in the pelvis simultaneously, the hip axis can move only within a narrow area. CONCLUSIONS: The changes in pelvic morphology due to the aging process occur in the anterior direction, which may be due to the remodeling process affecting the coxal bone that results in an anterior drift of the acetabulum relative to the sacrum. More importantly, the changes are considerably more evident in the cephalic direction, which may be the result of the weight-bearing loads and consequent wear of acetabular cartilage. Ó 2013 Elsevier Inc. All rights reserved.
Keywords:
Pelvic morphology; Sagittal alignment; Pelvic incidence; Pelvic thickness
FDA device/drug status: Not applicable. Author disclosures: TV: Nothing to disclose. MMAJ: Nothing to disclose. BL: Nothing to disclose. RMC: Speaking/Teaching Arrangements: Medtronic (A); Research Support (Investigator Salary): Synthes (D, Paid directly to institution/employer). MAV: Grant: Philips Healthcare (E, Paid directly to institution/employer). FP: Nothing to disclose. 1529-9430/$ - see front matter Ó 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.spinee.2013.06.034
The disclosure key can be found on the Table of Contents and at www. TheSpineJournalOnline.com. * Corresponding author. University of Ljubljana, Faculty of Electrical Engineering, Laboratory of Imaging Technologies, Trzaska cesta 25, SI1000 Ljubljana, Slovenia. Tel.: þ386 1-4768-327; fax: þ386 1-4768-279. E-mail address: [email protected] (T. Vrtovec)
T. Vrtovec et al. / The Spine Journal 13 (2013) 1500–1509
Introduction Pelvic morphology has a significant role in the spinopelvic relationship, as it is considered to influence the spinal balance and postural equilibrium, and has been recognized as an essential component in the regulation of sagittal plane alignment [1–6]. In contrast to positional parameters, such as the sacral slope or pelvic tilt, pelvic morphology is best evaluated by anatomical parameters of pelvic alignment in the sagittal plane, which are constant and unchanged for each individual subject and describe the position of the sacrum in relation to the pelvis [7,8]. Since its introduction by Duval-Beaupere et al. [9], the angle of pelvic incidence (PI) has gained large attention and acceptance as one of the key parameters in the complex framework of sagittal spinal alignment and related deformities [10–13]. By being an anatomical parameter, PI describes structural characteristics of the sacrum and pelvis that are not modified by posture [1,14] and can be therefore observed and compared among subjects in either standing, sitting, or supine position. For normal subjects, it is generally accepted that there is no difference in PI between male and female subjects, that PI increases during childhood and then remains unchanged throughout adolescence and adulthood [2,15], and that PI is not significantly influenced by normal degenerative changes of the hips, sacrum, and sacroiliac joints [16]. However, Mendoza-Lattes et al. [17] recently suggested that PI continues to increase linearly even after skeletal maturity and throughout the lifespan as a result of the increasing distance between the sacrum and the femoral heads, suggesting a morphological change of the pelvis. If the femoral heads are in a relatively more anterior position, then PI is consequently going to present larger angular values. The purpose of this study was to observe whether there are morphological changes of the pelvis related to the age of the subjects as described by Mendoza-Lattes et al. [17]. By measuring the anatomical parameters of sagittal pelvic alignment in a normal adult population, the configuration of pelvic morphology and the extent of changes that occur due to the aging process are evaluated.
1501
body parameters were not recorded for scan acquisition, and the original DICOM files were further anonymized by removing all confidential information and retaining only subject sex and age. The following exclusion criteria were applied to the initial cohort: age !18 years (skeletal maturity not reached) and age O90 years (small population ratio, inadequate image quality due to bone aging); prior hip or spine surgery; any clinical or radiological history or evidence for pathology or trauma of the spine, pelvis, or femur; evidence for anatomical anomalies (eg, abnormal number of lumbar vertebrae); syndromes associated with disorder of growth; any form of psychomotor retardation; poor quality of the scan (eg, pelvis not fully included, artifacts); pregnancy. As a result, 100 subjects were excluded from the initial cohort and the remaining 330 subjects represented the final study cohort. Measurement of pelvic parameters To measure pelvic parameters that allow the evaluation of pelvic morphology, the following anatomical references were determined in the CT scan of each subject: the center of the left femoral head, the center of the right femoral head, the center of the sacral end plate, and the inclination of the sacral end plate (Fig. 1). Multiplanar 3D image reformation was then performed to obtain the superposition of
Materials and methods Subjects The initial cohort consisted of 430 subjects who received a computed tomography (CT) scan of the pelvis or abdomen in the emergency room of the (University Medical Center Utrecht, the Netherlands) during a 5-year period (2005–2010) for reasons such as trauma or acute abdominal pathology. Computed tomography scans (Brilliance 16 and 64 scanners, Philips Healthcare, The Netherlands) were axially reconstructed three-dimensional (3D) images (pixel size, 0.4–1.0 mm; slice thickness, 3.0–4.0 mm), showing the femoral heads, the whole pelvis, and at least the lower lumbar region of the spine. Subject height, weight, or other
Fig. 1. Pelvic parameters in the sagittal plane are used to evaluate pelvic morphology.
1502
T. Vrtovec et al. / The Spine Journal 13 (2013) 1500–1509
the femoral heads in the sagittal view, so that all anatomical structures were completely in line with the hip axis. In such perfect sagittal view, the following anatomical pelvic parameters were measured: Pelvic incidence: The angle between the line orthogonal to the sacral end plate inclination and the line connecting the center of the sacral end plate with the hip axis [9]. Sacral end plate width (S1W): The length of the sacral end plate in the direction of its inclination. Pelvic thickness (PTH): The distance between the center of the sacral end plate and the hip axis [9]. By considering the measured PI and PTH within an orthogonal triangle, the anteroposterior (PTHA-P) and cephalocaudal (PTHC-C) components of PTH were in addition calculated as (Fig. 1): PTHAP 5PTH sinðPIÞ; PTHCC 5PTH cosðPIÞ:
ð1Þ
Evaluation of pelvic morphology Although the measured pelvic parameters are constant for each individual subject, it is not appropriate to directly compare distance parameters (ie, S1W and PTH) among subjects because they depend on individual body size. As a result, for two subjects with equal anatomy shape but different anatomy size (ie, equal PI but different S1W and PTH), distance parameters may be different and therefore have to be normalized before they are compared. As we are interested in changes of PTH and its components, the normalization is performed according to S1W, for which we assume that it reflects anatomy size. The normalized pelvic thickness (nPTH) and its normalized anteroposterior (nPTHA-P) and cephalocaudal (nPTHC-C) components are therefore defined for each subject as: mS1WðageÞ 100% ; S1W aPTH
nPTHAP ð%Þ5PTHAP nPTHCC ð%Þ5PTHCC
After dividing by aPTH, aPTHA-P, or aPTHC-C and multiplying by 100%, the resulting value represents the percentage of variation of the observed parameter around its average in the studied population. After dividing by S1W and multiplying by mS1W(age), the influence of different anatomy size is removed. The average S1W is expected to increase with subject age because the sacral end plate adapts to the weight-bearing loads. Normalization of PTH is therefore not performed according to the average S1W but the modeled expected average S1W for the observed age of the subject (ie, mS1W(age)). Statistical analysis
The measured pelvic parameters are constant for the observed subject, do not change with subject position and/or orientation, and can be therefore measured from CT images acquired in patient supine position. To determine the exact location of the aforementioned anatomical references in 3D images and therefore measure the previously described pelvic parameters in CT scans, we used a computerized method based on image processing techniques that was introduced and validated in a recently published study [18].
nPTHð%Þ5PTH
where mS1W(age) is the modeled expected average S1W for the observed age of the subject and aPTH, aPTHA-P, and aPTHC-C are the average PTH, average anteroposterior PTH component, and average cephalocaudal PTH component across population, respectively. The effects of such normalization are twofold:
100% mS1WðageÞ ; S1W aPTHAP mS1WðageÞ 100% ; S1W aPTHCC ð2Þ
The obtained measurements were ordered according to age and sex of the subjects and analyzed in terms of mean and standard deviation (SD). Regression lines between any two observed variables were determined by the random sample consensus method (assumed ratio of outliers, 25%; number of iterations, 1,000). Pearson correlation analysis was performed to search for statistically significant correlations between a selected measurement and age or PI. The independent samples t test was used to evaluate statistical differences between the sexes (level of significance a50.05).
Results Subjects The final study cohort of 330 subjects (mean age, 45.3 years; SD, 18.1 years; range, 18–87 years) was divided according to sex into the group of 164 male subjects (mean age, 44.3 years; SD, 18.0 years; range, 18–87 years) and the group of 166 female subjects (mean age, 46.2 years; SD, 18.1 years; range, 18–86 years). No statistically significant differences in age were found between the sexes (pO.05). Measurement of pelvic parameters Pelvic parameters were measured by a computerized method that was introduced and validated in a recently published study [18] and proved to be highly consistent with manual measurements (SD of up to 1.8 mm for distances and 3.5 for angles). Fig. 2 shows examples of measurements, whereas Fig. 3 and Table 1 show the obtained measurements for PI, S1W, and PTH. The correlation of each parameter with age was relatively low but statistically
T. Vrtovec et al. / The Spine Journal 13 (2013) 1500–1509
1503
Fig. 2. Measurement of pelvic parameters in three dimensions is performed by applying a computerized method to computed tomography images. The results for (Left) a 48-year-old male and (Right) a 31-year-old female subject are shown in the perfect sagittal view for the maximal intensity projection (top) and midsagittal cross-section (bottom).
significant (p!.05). The distribution of S1W measurements shows a distinctive difference between male and female subjects, which was expected because of anatomy size, and also a distinctive increasing tendency of the average S1W with age (Fig. 3, Middle), which was expected because of the adaptation of the sacral end plate to the weight-bearing loads. Because a statistically significant difference between the sexes was found for S1W (p!.0001), the expected average S1W for the observed age of the subject (ie, mS1W(age)) was modeled separately for male and female subjects. The obtained linear regression model using random sample consensus was mS1W(age)50.08ageþ33.25 for male subjects and mS1W(age)50.11ageþ28.19 for female subjects. The modeled expected average S1W for the two examples shown in Fig. 2 were therefore mS1W(48)5
0.0848þ33.25537.1 mm for the 48-year-old male subject (Fig. 2, Left) and mS1W(31)50.1131þ28.19531.6 mm for the 31-year-old female subject (Fig. 2, Right). On the contrary, no significant differences were found between the sexes for PTH and its components PTHA-P and PTHC-C (pO.05); therefore, the average values aPTH5104.9, aPTHA-P575.9, and aPTHC-C570.1 mm were computed together for male and female subjects. Evaluation of pelvic morphology The measured PTH and its components PTHA-P and PTHC-C were normalized according to their average values (ie, aPTH, aPTHA-P, and aPTHC-C, respectively) and according to the modeled expected average S1W for the
1504
T. Vrtovec et al. / The Spine Journal 13 (2013) 1500–1509 Table 1 Pelvic incidence, S1W, and PTH, as measured for all subjects (N5330) and separately for male (N5164) and female (N5166) subjects Vs. age* Pelvic parameter
Mean
SD
r
p
py
PI ( ) Male Female S1W (mm) Male Female PTH (mm) Male Female
47.6 47.3 47.9 35.2 37.0 33.4 104.9 104.5 105.4
10.2 9.5 10.8 4.4 4.1 4.0 8.6 8.3 8.9
þ0.25 þ0.16 þ0.32 þ0.38 þ0.37 þ0.51 0.26 0.27 0.25
!.0001 !.05 !.0001 !.0001 !.0001 !.0001 !.0001 !.05 !.05
O.05 !.0001 O.05
PI, pelvic incidence; S1W, sacral end plate width; PTH, pelvic thickness; SD, standard deviation. * Correlation coefficient r and significance level p in relation to age. y Significance level p between male and female subjects.
Discussion
Fig. 3. (Top) PI, (Middle) S1W, and (Bottom) PTH in relation to age for male and female subjects with superimposed regression lines. PI, pelvic incidence; S1W, sacral end plate width; PTH, pelvic thickness.
observed age of the subject (ie, mS1W(age)). Fig. 4 and Table 2 show the obtained results in relation to the age of the subjects. Statistically significant correlation with age was obtained for all parameters, except for nPTHA-P for male subjects. Fig. 5 and Table 2 show the obtained results in relation to PI of the subjects. Statistically significant correlation with PI was obtained for all parameters. By comparing the resulting values between the sexes, no statistically significant differences were observed for all three parameters (pO.05).
For normal subjects, it is generally accepted that PI increases during the first months of life, becomes stable at around 10 years of age, and then remains unchanged throughout adolescence and adulthood. However, it may be altered by pathologic processes that modify the shape of the sacrum or the position of the acetabulae within the pelvis [15], such as spondylolisthesis or spondylolisis [2,19–23], scoliosis [1,24–26], and other spinopelvic pathologies [5,27–32]. Studies were conducted to determine the difference in PI between male and female subjects, reporting in general no relationship between PI and sex [6,9,14–16,20,33–35] but in some cases higher but not statistically significant values for male [1,17,36] or female [33,37] subjects. On the contrary, Vialle et al. [4] reported statistically significant differences between the sexes, with higher PI for female subjects. In examining the relationship between PI and age of the subjects, studies reported no relationship in fetuses [15], adults [4,15,20], elderly population [36], or various age groups [38], but PI was found to be higher in normal elderly [36,38] than normal adult population [1,2,4,6,9,14,35], and a relationship between PI and age was found for children [15,19,33,37] and adolescents [33,37]. Recent studies also reported that aging or degenerative processes, manifested as a progressive loss in the height of intervertebral discs, induce compensation in the sagittal balance by a backward rotation (ie, retroversion) of the pelvis [39], which is for an individual limited by PI that remains constant [40]. When considering the size of the pelvis, a statistically significant difference was reported between the sexes for PTH [41] and a significant negative correlation between PTH and PI [42], meaning that an increase in PI is coupled with a decrease in PTH. In contrast to the generally accepted views, by comparing the results of different studies Mendoza-Lattes et al. [17] recently concluded that PI increases linearly throughout the lifespan of normal subjects due to a remodeling process affecting the coxal bone (ie, the hip bone consisting of
T. Vrtovec et al. / The Spine Journal 13 (2013) 1500–1509
1505
Table 2 Normalized pelvic thickness, nPTHA-P, and nPTHC-C, as evaluated for all subjects (N5330) and separately for male (N5164) and female (N5166) subjects Vs. PIy
Vs. age* Pelvic parameter
SD
r
p
r
p
pz
nPTH (%) Male Female nPTHA-P (%) Male Female nPTHC-C (%) Male Female
13.2 11.7 11.7 17.8 17.2 16.6 27.1 23.9 28.0
0.18 0.19 0.19 þ0.11 þ0.01 þ0.22 0.27 0.21 0.33
!.05 !.05 !.05 !.05 O.05 !.05 !.0001 !.05 !.0001
0.27 0.20 0.39 þ0.70 þ0.75 þ0.74 0.87 0.86 0.91
!.0001 !.05 !.0001 !.0001 !.0001 !.0001 !.0001 !.0001 !.0001
O.05 O.05 O.05
nPTH, normalized pelvic thickness; nPTHA-P, normalized anteroposterior component of pelvic thickness; nPTHC-C, normalized cephalocaudal component of pelvic thickness; PI, pelvic incidence; SD, standard deviation around the 100% mean. * Correlation coefficient r and significance level p in relation to age. y Correlation coefficient r and significance level p in relation to PI. z Significance level p between male and female subjects.
Fig. 4. (Top) nPTH, (Middle) nPTHA-P, and (Bottom) nPTHC-C in relation to age for male and female subjects with superimposed regression lines. nPTH, normalized pelvic thickness; nPTHA-P, normalized anteroposterior component of pelvic thickness; nPTHC-C, normalized cephalocaudal component of pelvic thickness.
the ilium, ischium, and pubis), which results in an anterior drift of the acetabulum relative to the sacroiliac joint. They explained the gradual increase of PI throughout the lifespan by an increase of the sacropelvic translation [43], defined as the horizontal distance between the reference vertical through the center of the femoral head and the reference vertical through the posterior corner of the sacral end plate.
The horizontal distance between the sacrum and the hip axis can also be described by the pelvic overhang [9], if the distance is measured against the reference vertical through the center of the sacral end plate, or by the sacrofemoral distance [44], if the distance is measured against the reference vertical through the anterior corner of the sacral end plate. As a result, if the femoral heads move to a relatively more anterior position, PI is consequently going to present larger angular values. The results of our study confirm the conclusions of Mendoza-Lattes et al. [17] and also bring new insight into the evaluation of pelvic morphology. The increase of the horizontal distance between the sacrum and the hip axis, as suggested by Mendoza-Lattes et al. [17], implies an increase of the straight distance between the sacrum and the hip axis, represented by PTH. However, our study shows that with increasing PI that occurs due to the aging process (Fig. 3, Top), PTH decreases (Fig. 4, Top, and Fig. 5, Top). The decrease of PTH is manifested not only as an increase of the distance between the sacrum and the hip axis in the anterior direction (ie, PTHA-P; Fig. 4, Middle, and Fig. 5, Middle) but considerably more as a decrease of the distance between the sacrum and the hip axis in the cephalic direction (ie, PTHC-C; Fig. 4, Bottom, and Fig. 5, Bottom). The horizontal distance between the sacrum and the hip axis can be, to some extent, compared with PTHA-P in our study. Our results show that PTHA-P in general increases with age for female subjects but slightly decreases with age for male subjects. The same tendency can be observed for PTHC-C, which manifests a stronger decrease for female than for male subjects. Nevertheless, by considering these morphological changes in the pelvis simultaneously (ie, increase of PI, decrease of PTH, increase of PTHA-P, and decrease of PTHC-C), the hip axis can move only within a narrow area (Fig. 6). These changes therefore
1506
T. Vrtovec et al. / The Spine Journal 13 (2013) 1500–1509
Fig. 5. (Top) nPTH, (Middle) nPTHA-P, and (Bottom) nPTHC-C in relation to PI for male and female subjects with superimposed regression lines. nPTH, normalized pelvic thickness; nPTHA-P, normalized anteroposterior component of pelvic thickness; nPTHC-C, normalized cephalocaudal component of pelvic thickness; PI, pelvic incidence.
occur in the anterior direction, which may be due to the remodeling process affecting the coxal bone that results in an anterior drift of the acetabulum relative to the sacrum. More importantly, the changes are considerably more evident in the cephalic direction, which may be the result of the weight-bearing loads and consequent wear of acetabular cartilage.
The main shortcoming of the performed study is that it is cross-sectional and therefore the differentiation between cause and effect is difficult. However, performing a longitudinal study and therefore radiographically screening a large number of normal individuals throughout their lifespan is not feasible and would be questionable in terms of costs and medical ethics. Regarding cause and effect, there is no doubt that the increasing PI is the effect, while the cause is represented by the morphological changes of the pelvis, observed through pelvic parameters. It is, however, important to note that assuming the size of anatomy is reflected in the size of the sacrum, and therefore S1W, can be treated as adequate only for cases of normal anatomy, or anatomy that is considered normal. In the case of severe abnormalities or pathologies, for example, dome-shaped sacral end plate in the case of spondylolysis [45], sacral dysplasia in the case of spondylolisthesis [46] or even sacral insufficiency fractures [47], the corresponding expected values mS1W(age) would have to be modeled according to the studied population. Although the observed cohort of subjects was assumed to represent a normal population, several subjects were excluded from the original cohort due to several criteria (ie, age, clinical, or radiological history or evidence for pathology or trauma, evidence for anatomical anomalies, and so on), and imposing of additional criteria may result in further exclusion of subjects. Unfortunately, criteria such as activity (eg, sedentary vs. mobile lifestyle, practicing sports, and so on), body mass index, or other physiological characteristics were not recorded along with other subject information but may nevertheless influence the sagittal pelvic alignment [19,34]. However, the natural biological variability of the human anatomy is relatively large, which is also reflected in the large variability of the measured parameters that describe pelvic morphology. Ascertaining a truly normal and asymptomatic population therefore remains poorly defined and represents a limitation of the present as well as of the aforementioned studies. In contrast to current routine clinical protocols, the measurements in this study were obtained in 3D from volumetric CT images of the pelvis. This does not imply that pelvic morphology should not be evaluated in two dimensions from plain sagittal radiographic images; however, computerized measurements in 3D proved to be highly consistent and reliable, especially because they were obtained from sagittal views after multiplanar reformation according to the hip axis. In such perfect sagittal views, all anatomical structures were observed completely in line with the hip axis and were therefore not biased by the projective nature of radiographic image acquisition [18]. Because CT images are acquired in subject supine position, the obtained findings cannot be transferred to positional pelvic parameters (eg, sacral slope, pelvic tilt, sacropelvic translation, pelvic overhang, sacrofemoral distance) but need to be contained within anatomical pelvic parameters (eg, PI, PTH) that are invariant against arbitrary patient position and/or orientation.
T. Vrtovec et al. / The Spine Journal 13 (2013) 1500–1509
1507
Fig. 6. (A) If PI increases, the hip axis must move into the shown area. (B) If PTH decreases, the hip axis must move into the shown area. (C) If PTHA-P increases and PTHC-C decreases, the hip axis must move into the shown area. (D) If all changes occur simultaneously, the hip axis must be located within the intersection of these areas, that is, within the shown area. PI, pelvic incidence; PTH, pelvic thickness; PTHA-P, anteroposterior component of pelvic thickness; PTHC-C, cephalocaudal component of pelvic thickness.
Conclusion Despite the previously described shortcomings, this study revealed that there are indications for morphological changes in the pelvis that continue to occur throughout the lifespan and reflect the effort of the human body to maintain adequate sagittal balance. The pelvis is also subject to a posteriorly inferiorly directed load that leads to bony remodeling changes according to Wolff’s law [48,49], which asserts that bone in a healthy person will adapt to the loads it is placed under to maximize mechanical efficiency. According to the Utah paradigm of skeletal physiology [50–52], which is a refinement of Wolff’s law, bone growth and bone loss are stimulated by the local mechanical elastic deformation of bone. The adaptation (feedback control loop) of bone according to the maximum forces is considered to be a lifelong process; hence, bone adapts
its mechanical properties according to the needed mechanical function—bone mass, geometry, and strength are therefore adapted according to the everyday usage and needs.
Acknowledgments This work has been supported by the Ministry of Education, Science, Culture and Sport, Slovenia, under grants P2-0232, J7-2264, L2-7381, and L2-2023, and partly by Biomet and the Anna Foundation. References [1] Legaye J, Duval-Beaupere G, Hecquet J, Marty C. Pelvic incidence: a fundamental pelvic parameter for three-dimensional regulation of spinal sagittal curves. Eur Spine J 1998;7:99–103.
1508
T. Vrtovec et al. / The Spine Journal 13 (2013) 1500–1509
[2] Labelle H, Roussouly P, Berthonnaud E, et al. Spondylolisthesis, pelvic incidence, and spinopelvic balance: a correlation study. Spine 2004;29:2049–54. [3] Roussouly P, Gollogly S, Berthonnaud E, Dimnet J. Classification of the normal variation in the sagittal alignment of the human lumbar spine and pelvis in the standing position. Spine 2005;30: 346–53. [4] 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 2005;87:260–7. [5] Lafage V, Schwab F, Skalli W, et al. Standing balance and sagittal plane spinal deformity: analysis of spinopelvic and gravity line parameters. Spine 2008;33:1572–8. [6] Mac-Thiong JM, Roussouly P, Berthonnaud E, Guigui P. Sagittal parameters of global spinal balance: normative values from a prospective cohort of seven hundred nine Caucasian asymptomatic adults. Spine 2010;35:E1193–8. [7] Vrtovec T, Janssen M, Likar B, et al. A review of methods for evaluating the quantitative parameters of sagittal pelvic alignment. Spine J 2012;12:433–46. [8] Cheng I. Commentary: spinopelvic parameters-how far have we come? Spine J 2012;12:447–8. [9] Duval-Beaupere G, Schmidt C, Cosson P. A barycentremetric study of the sagittal shape of spine and pelvis: the conditions required for an economic standing position. Ann Biomed Eng 1992;20:451–62. [10] Roussouly P, Nnadi C. Sagittal plane deformity: an overview of interpretation and management. Eur Spine J 2010;19:1824–36. [11] Schwab F, Patel A, Ungar B, et al. Adult spinal deformity—postoperative standing imbalance: how much can you tolerate? An overview of key parameters in assessing alignment and planning corrective surgery. Spine 2010;35:2224–31. [12] Lafage V, Schwab F, Vira S, et al. Spino-pelvic parameters after surgery can be predicted: a preliminary formula and validation of standing alignment. Spine 2011;36:1037–45. [13] Mac-Thiong JM, Duong L, Parent S, et al. Reliability of the Spinal Deformity Study Group classification of lumbosacral spondylolisthesis. Spine 2012;37:E95–102. [14] Vaz G, Roussouly P, Berthonnaud E, Dimnet J. Sagittal morphology and equilibrium of pelvis and spine. Eur Spine J 2002;11:80–7. [15] Mangione P, Gomez D, Senegas J. Study of the course of the incidence angle during growth. Eur Spine J 1997;6:163–7. [16] Mac-Thiong JM, Roussouly P, Berthonnaud E, Guigui P. Age- and sex-related variations in sagittal sacropelvic morphology and balance in asymptomatic adults. Eur Spine J 2011;20:S572–7. [17] Mendoza-Lattes S, Ries Z, Gao Y, Weinstein S. Natural history of spinopelvic alignment differs from symptomatic deformity of the spine. Spine 2010;35:E792–8. [18] Vrtovec T, Janssen M, Pernus F, et al. Analysis of pelvic incidence from 3-dimensional images of a normal population. Spine 2012;37: E479–85. [19] Hanson D, Bridwell K, Rhee J, Lenke L. Correlation of pelvic incidence with low- and high-grade isthmic spondylolisthesis. Spine 2002;27:2026–9. [20] Marty C, Boisaubert B, Descamps H, et al. The sagittal anatomy of the sacrum among young adults, infants, and spondylolisthesis patients. Eur Spine J 2002;11:119–25. [21] Huang R, Bohlman H, Thompson G, Poe-Kochert C. Predictive value of pelvic incidence in progression of spondylolisthesis. Spine 2003;28:2381–5. [22] Labelle H, Roussouly P, Berthonnaud E, et al. The importance of spino-pelvic balance in L5-S1 developmental spondylolisthesis: a review of pertinent radiologic measurements. Spine 2005;30:S27–34. [23] Roussouly P, Gollogly S, Berthonnaud E, et al. Sagittal alignment of the spine and pelvis in the presence of L5-S1 isthmic lysis and lowgrade spondylolisthesis. Spine 2006;31:2484–90.
[24] Mac-Thiong JM, Labelle H, Charlebois M, et al. Sagittal plane analysis of the spine and pelvis in adolescent idiopathic scoliosis according to the coronal curve type. Spine 2003;28:1404–9. [25] Hong JY, Suh SW, Modi H, et al. Correlation of pelvic orientation with adult scoliosis. J Spinal Disord Tech 2010;23:461–6. [26] Lonner B, Auerbach J, Sponseller P, et al. Variations in pelvic and other sagittal spinal parameters as a function of race in adolescent idiopathic scoliosis. Spine 2010;35:E374–7. [27] Barrey C, Jund J, Noseda O, Roussouly P. Sagittal balance of the pelvis-spine complex and lumbar degenerative diseases: a comparative study about 85 cases. Eur Spine J 2007;16:1459–67. [28] Hart R, Badra M, Madala A, Yoo J. Use of pelvic incidence as a guide to reduction of H-type spino-pelvic dissociation injuries. J Orthop Trauma 2007;21:369–74. [29] Abelin K, Vialle R, Lenoir T, et al. The sagittal balance of the spine in children and adolescents with osteogenesis imperfecta. Eur Spine J 2008;17:1697–704. [30] Rose P, Bridwell K, Lenke L, et al. Role of pelvic incidence, thoracic kyphosis, and patient factors on sagittal plane correction following pedicle subtraction osteotomy. Spine 2009;34:785–91. [31] Hong JY, Suh SW, Modi H, et al. Analysis of sagittal spinopelvic parameters in achondroplasia. Spine 2011;36:E1233–9. [32] Videbaek T, B€unger C, Henriksen M, et al. Sagittal spinal balance after lumbar spinal fusion: the impact of anterior column support: results from a randomized clinical trial with an eight- to thirteenyear radiographic follow-up. Spine 2011;36:183–91. [33] Mac-Thiong JM, Berthonnaud E, Dimar J, et al. Sagittal alignment of the spine and pelvis during growth. Spine 2004;29:1642–7. [34] Boulay C, Tardieu C, Hecquet J, et al. Sagittal alignment of spine and pelvis regulated by pelvic incidence: standard values and prediction of lordosis. Eur Spine J 2006;15:415–22. [35] Janssen M, Drevelle X, Humbert L, et al. Differences in male and female spino-pelvic alignment in asymptomatic young adults: a three-dimensional analysis using upright low-dose digital biplanar X-rays. Spine 2009;34:E826–32. [36] Hammerberg E, Wood K. Sagittal profile of the elderly. J Spinal Disord Tech 2003;16:44–50. [37] Mac-Thiong JM, Labelle H, Berthonnaud E, et al. Sagittal spinopelvic balance in normal children and adolescents. Eur Spine J 2007;16:227–34. [38] Schwab F, Lafage V, Boyce R, et al. Gravity line analysis in adult volunteers: age-related correlation with spinal parameters, pelvic parameters, and foot position. Spine 2006;31:E959–67. [39] Le Huec J, Saddiki R, Franke J, et al. Equilibrium of the human body and the gravity line: the basics. Eur Spine J 2011;20: S558–63. [40] Roussouly P, Pinheiro-Franco J. Biomechanical analysis of the spino-pelvic organization and adaptation in pathology. Eur Spine J 2011;20:S609–18. [41] Rajnics P, Pomero V, Templier A, et al. Computer-assisted assessment of spinal sagittal plane radiographs. J Spinal Disord 2001;14:135–42. [42] Boulay C, Tardieu C, Hecquet J, et al. Anatomical reliability of two fundamental radiological and clinical pelvic parameters: incidence and thickness. Eur J Orthop Surg Traumatol 2005;15:197–204. [43] Jackson R, Peterson M, McManus A, Hales C. Compensatory spinopelvic balance over the hip axis and better reliability in measuring lordosis to the pelvic radius on standing lateral radiographs of adult volunteers and patients. Spine 1998;23:1750–67. [44] Takemitsu Y, Harada Y, Iwahara T, et al. Lumbar degenerative kyphosis: clinical, radiological and epidemiological studies. Spine 1988;13:1317–26. [45] Legaye J. The femoro-sacral posterior angle: an anatomical sagittal pelvic parameter usable with dome-shaped sacrum. Eur Spine J 2007;16:219–25.
T. Vrtovec et al. / The Spine Journal 13 (2013) 1500–1509 [46] Vialle R, Dauzac C, Khouri N, et al. Sacral and lumbar-pelvic morphology in highgrade spondylolisthesis. Orthopedics 2007;30:642–9. [47] Shah R. Sacral kyphoplasty for the treatment of painful sacral insufficiency fractures and metastases. Spine J 2012;12:113–20. [48] Wolff J. On the inner architecture of bones and its importance for bone growth (Ueber die innere Architectur der Knochen und ihre Bedeutung f€ ur die Frage vom Knochenwachsthum). Virchows Arch 1870;50:389–450. [49] Wolff J. The classic: on the inner architecture of bones and its importance for bone growth. (reprinted from Wolff J.: Ueber die innere Architectur der Knochen und ihre Bedeutung f€ur die Frage vom
1509
Knochenwachsthum. Virchows Archiv, 50:389–450, 1870). Clin Orthop Relat Res 2010;468:1056–65. [50] Frost H. Skeletal structural adaptations to mechanical usage (SATMU): 1. Redefining Wolff’s law: the bone modeling problem. Anat Rec 1990;226:403–13. [51] Frost H. Skeletal structural adaptations to mechanical usage (SATMU): 2. Redefining Wolff’s law: the remodeling problem. Anat Rec 1990;226:414–22. [52] Frost H. The Utah paradigm of skeletal physiology: an overview of its insights for bone, cartilage and collagenous tissue organs. J Bone Miner Metab 2000;18:305–16.