Elsevier

The Spine Journal

Volume 11, Issue 7, July 2011, Pages 659-667
The Spine Journal

Basic Science
Cervical spine bone mineral density as a function of vertebral level and anatomic location

https://doi.org/10.1016/j.spinee.2011.05.007Get rights and content

Abstract

Background context

Bone mineral density (BMD) measurements acquired from quantitative computed tomography scans have been shown to correlate with bone mechanical properties such as strength, stiffness, and yield load. There are currently no reports of BMD as a function of anatomic location within each vertebra.

Purpose

The overall objective of this study was to characterize BMD in the cervical spine as a function of level and anatomic location.

Study design

Cervical spine BMD was evaluated in vivo using a clinically relevant age group.

Patient sample

Twenty-two subjects (13 women and 9 men) were included with an average age of 48±7 years (range, 35–61 years). Ten subjects were recently diagnosed with cervical radiculopathy (age 49±8 years; six women and four men; and two smokers and eight nonsmokers), and 12 subjects were asymptomatic controls (age 46±6 years; seven women and five men; and three smokers, three quit smoking, and six nonsmokers).

Outcome measures

Physiologic measures included overall BMD for C3–C7, average BMD within 11 anatomically defined regions of interest for each vertebra, and density distribution (by volume) within each anatomic region and vertebral level.

Methods

Subject-specific three-dimensional bone models were created from high-resolution computed tomography scans of the subaxial cervical spine (C3–C7). Custom software calculated the average BMD within 11 anatomically defined regions of interest for each three-dimensional bone model. Bone mineral density values for each voxel of bone tissue were binned into 50 mg/cc ranges to determine the density distribution by volume. Repeated-measures analysis of variance was used to test for differences within subjects by level (C3–C7) and anatomic location. The correlation between BMD in the central vertebral body and the pedicle and lateral mass regions was tested using Pearson correlation.

Results

Average BMDs by level were 476, 503, 507, 473, and 414 mg/cm3 for C3–C7, respectively. C3 and C6 BMDs were significantly less than those of C4 and C5 (p<.007). C7 BMD was significantly less than those of all other levels (all p<.001). Control and female subjects showed a trend toward higher BMD than radiculopathy and male subjects across all levels (p value: .06–.17). Wide variation in BMD was observed over anatomical regions, with the pedicles having significantly higher BMD than all other anatomic locations and the anterior portion of the central vertebral body having significantly lower BMD than all other anatomic locations. There was a significant positive correlation between central vertebral body BMD and lateral mass BMD at each level. Bone mineral density distribution by volume plots revealed women had a higher volume of very high-density bone than men but only in the posterior elements.

Conclusions

This study has characterized BMD in the cervical spine according to vertebral level and anatomic location within each vertebral level using live subjects from a clinically relevant age group. The results indicate significant differences in BMD according to vertebral level and among anatomical regions within each vertebra. The results suggest to the surgeon and device manufacturer that surgical procedures involving instrumentation attached to C7 may require a modification in instrumentation or in surgical technique to attain results equivalent to more superior levels. The results suggest to the basic scientist that computational models may be improved by taking into account the wide variation in BMD over different anatomical regions.

Introduction

The complex geometry of the cervical spine provides multiple locations where load may be applied to each vertebra either through articulating joint surfaces (eg, vertebral body and facets) or through muscle and ligament attachments (eg, lamina and spinous processes). The asymptomatic neck is capable of a wide variety of movements, ranging from simple single-plane motion (eg, flexion-extension) to complex multiplane motions. Thus, it is clear that the anatomical structures that comprise the cervical vertebrae are routinely subjected to diverse combinations of load. Furthermore, the multiple joint articulations and ligament and tendon attachment sites on each bone lead to variation in the type and magnitude of loading applied to different anatomic locations within each vertebra (eg, pedicles are loaded differently than vertebral bodies). Finally, because of the natural curvature of the cervical spine, it is probable that vertebral loading is different among cervical levels. The bone tissue, following Wolff's Law, adapts to these various loading patterns both within an individual vertebra and among vertebra. Therefore, it follows that bone mechanical properties are likely to differ among anatomic locations within an individual vertebra and among vertebrae depending on level.

Bone mechanical properties may be estimated by bone mineral density (BMD) measurements acquired from dual-energy X-ray absorptiometry (DXA or DEXA) or quantitative computed tomography (QCT) scans. For example, DEXA scans of the cervical spine have shown a correlation between BMD and pullout strength [1], [2]. Similarly, BMD measured using QCT has been shown to be related to mechanical properties, such as elastic modulus, strength, stiffness, and yield load [3], [4], [5], [6], [7]. From the clinical perspective, a clear relationship has been established between BMD and surgical instrumentation performance, such as screw pullout strength [8], [9], [10] and disc replacement subsidence [3]. Previously, BMD measurements at specific anatomic locations within the cervical spine have been performed almost exclusively on the central vertebral body, by measuring either trabecular BMD in a single computed tomogaphy (CT) slice [5], [11], [12] or a large volume of trabecular bone in multiple CT scan slices [13], [14]. In addition to the central vertebral body, the only other BMD measurements of specific anatomic regions within the cervical spine were performed on the end plates [3], [11]. The relationship between site-specific bone characteristics and bone mechanical properties has been investigated to a limited extent in the thoracic and lumbar spine. In the thoracic spine, bone mass and thickness at specific anatomic locations within the posterior elements have been related to the strength at these sites [15], whereas in the lumbar spine, regional measures of bone density have been correlated to whole-bone fracture load [16].

In the case of the cervical spine, it would be beneficial to know the extent to which BMD, and therefore bone mechanical properties, vary in relation to vertebral level and anatomic location. However, this information does not currently exist. Knowledge of BMD variation according to vertebral level and anatomic location would be beneficial to surgeons who must secure instrumentation to the vertebrae during surgery, to those in the medical device industry who must design instrumentation such as screws and disc replacements, and to basic scientists who assign mechanical properties to bones in computational models of the spine.

The overall objective of this study was to characterize BMD in the cervical spine as a function of vertebral level and anatomic location. The first specific aim was to determine the overall BMD of cervical vertebrae C3–C7. The second specific aim was to determine the BMD in distinct anatomic regions within C3–C7, and the third specific aim was to correlate BMD in the central vertebral body to BMD in anatomic regions commonly used for screw fixation. The three null hypotheses tested were the following: BMD does not change with vertebral level; BMD is not significantly different across anatomic regions; and BMD in the central vertebral body is not correlated with pedicle and lateral mass BMD.

Section snippets

Methods

High-resolution CT scans (0.29×0.29×1.25 mm voxels) of the subaxial cervical spine (C3–C7) were collected from 22 subjects after institutional review board approval (overall average age 48±7 years; range 35–61 years; and 13 women, and 9 men). Ten subjects were diagnosed with cervical radiculopathy within 30 days of data collection (age 49±8 years; six women and four men; and two smokers and eight nonsmokers), and 12 subjects were asymptomatic controls (age 46±6 years; seven women and five men;

Results

Average BMD varied significantly by level. C3 and C6 BMDs were significantly less than those of C4 and C5 (all p≤.007). C7 BMD was significantly less than those of all other levels (all p≤.001) (Table 1, Fig. 2). Control subject average BMD was higher than radiculopathy subject BMD at each vertebral level; however, this difference did not reach significance at any level (p value range: .06–.17). Female average BMD was higher than male BMD at each vertebral level; however, this difference

Discussion

Differences in BMD by level for the entire C3–C7 vertebrae were consistent with previous reports based on single-CT slices [5], [11], [20] and volumetric measurements [13], [14] of trabecular vertebral body bone density. These previous reports indicated that cervical vertebral body BMD was highest at C5 and decreased in the direction of C3 and C7 [5], [11], [13], [14], [20], [21]. Significantly lower BMD of C7 in comparison to C4–C6 has also been reported in cadaver specimens (mean age 71, 9

Acknowledgment

The study was funded by NIH/NIAMS, Grant R03-AR056265, and the Cervical Spine Research Society 21st Century Development Grant.

References (34)

  • N.R. Ordway et al.

    Correlation of cervical endplate strength with CT measured subchondral bone density

    Eur Spine J

    (2007)
  • T.R. Oxland et al.

    The relative importance of vertebral bone density and disc degeneration in spinal flexibility and interbody implant performance. An in vitro study

    Spine

    (1996)
  • X.M. Zhuang et al.

    Effect of the degree of osteoporosis on the biomechanical anchoring strength of the sacral pedicle screws: an in vitro comparison between unaugmented bicortical screws and polymethylmethacrylate augmented unicortical screws

    Spine

    (2010)
  • O.C. Thiele et al.

    Factors affecting the stability of screws in human cortical osteoporotic bone: a cadaver study

    J Bone Joint Surg Br

    (2007)
  • T.C. Ryken et al.

    Biomechanical analysis of bone mineral density, insertion technique, screw torque, and holding strength of anterior cervical plate screws

    J Neurosurg

    (1995)
  • F. Kandziora et al.

    Comparison between sheep and human cervical spines: an anatomic, radiographic, bone mineral density, and biomechanical study

    Spine

    (2001)
  • D. Weishaupt et al.

    Relationships of cervical, thoracic, and lumbar bone mineral density by quantitative CT

    J Comput Assist Tomogr

    (2001)
  • Cited by (41)

    • Finite Element Analysis of the Effect of Dynamic Plating on Two-Level Anterior Cervical Discectomy Fusion Biomechanics

      2022, World Neurosurgery
      Citation Excerpt :

      First, in the present study, the cancellous bone elements were assumed to be a constant value, and the related uniform Young's modulus from the literature was used.34 Anderst et al.64 detailed the differing bone mineral densities by level and location, indicating that differences would occur in the results. However, we found no study that had used the bone mineral density to define the material properties in an FE model, which requires separation of the cancellous bone into separate geometries.

    • Does screw length for primary two-level ACDF influence pseudarthrosis risk?

      2020, Spine Journal
      Citation Excerpt :

      This strong association between screw VB% and successful fusion may, in part, be due to variations in bone mineral density within the VB. The anterior portion of the central VB has a significantly lower BMD than all other anatomical regions within the VB (p<.001) [44]. This is consistent with our finding that screw length alone may not be a significant factor since variations in VB depth exist between patients.

    • Regional bone mineral density differences measured by quantitative computed tomography in patients undergoing anterior cervical spine surgery

      2020, Spine Journal
      Citation Excerpt :

      The mean BMD of C4 (283.5 mg/cm3) was approximately 1.5 times the mean BMD of T1 (184.4 mg/cm3). The observed trend of a gradual BMD decrease from mid-cervical levels to T1 is in general agreement with several previous smaller studies of nonsurgical subjects [5,15–17,33,35,36]. The wide BMD variation over the cervical spine with significantly lower BMD in the caudal part might partially contribute to several clinically relevant phenomena.

    View all citing articles on Scopus

    FDA device/drug status: not applicable.

    Author disclosures: WJA: Nothing to disclose. EDT: Nothing to disclose. JYL: Nothing to disclose. WFD: Nothing to disclose. JDK: Research Support (Staff/Materials): Stryker (E, Paid directly to institution/employer); Grants: J & J (C, Paid directly to institution/employer).

    The disclosure key can be found on the Table of Contents and at www.TheSpineJournalOnline.com.

    The authors WJA and EDT contributed equally to this article.

    View full text