Orthopedics

Feature Articles 

Micro-CT Density Analysis of the Medial Wall of the Human Medial Cuneiform

Christopher E. Pelt, MD; Chad M. Turner, MD; Kent N. Bachus, PhD; K. Bo Foreman, PT, PhD; Timothy C. Beals, MD

Abstract

The human medial cuneiform is incompletely characterized with regard to anatomical morphology, including mineral density and bone quality. Clinically, we have observed failures of fixation by pull-through of devices through relatively soft medial bone. Defining patterns of relative density may provide valuable information regarding implant placement as higher cortical density bone may offer better resistance to such failures. We sought to identify an area of greatest density along the medial wall of the medial cuneiform.

Ten fresh-frozen human cadaveric medial cuneiforms underwent micro-computed tomography imaging. Images were analyzed to obtain densities in 4 quadrants along the medial wall of the medial cuneiform. Seven of 10 specimens revealed a maximum density in the plantar distal quadrant of the medial wall of the medial cuneiform. Chi-square goodness-of-fit testing indicated that the density of this quadrant was significantly different from 3 other quadrants (P<.009). Using the Principle of Standard Residuals, the density of the plantar distal quadrant was significantly different than the other 3.

We conclude that the plantar distal quadrant of the medial cuneiform contains bone of maximal density when compared to 3 other quadrants. Surgeons who place implants in this region should be aware that this area might offer better resistance to fixation failure.

Although the human medial cuneiform is an incompletely characterized bone, multiple surgical procedures use its medial wall as a point of fixation during the internal fixation of midfoot injuries and reconstructive procedures. By virtue of its anatomical morphology, the medial wall of the medial cuneiform provides the surgeon with a choice for the placement of an internal fixation device. It is desirable to place fixation devices at a location on or through this bone with the least likelihood of failure. By quantifying the density of bone along the medial wall of the medial cuneiform, we hope to offer information that will allow surgeons to make an informed decision regarding placement of internal fixation devices, particularly suture-button devices that have gained popularity.1-3

Previous studies have looked at the density of bone near the first tarsometatarsal joint, but to our knowledge, none have investigated the density along the medial wall of the medial cuneiform.4,5 Unlike other long bones in the body, the human medial cuneiform has limited cortical thickness.6,7 The medial cuneiform displays an architecture of cortical and trabecular bone such that they are similar in quality, thickness, and appearance. Other human bones, such as the vertebrae and carpal bones, display similar characteristics. In both cortical and trabecular bone, material and mechanical properties (strength, modulus, energy absorption, ductility, brittleness) are proportional to the density of the bone as well as the microarchitecture.8 Notably, the relative contribution of cortical and trabecular bone in the human vertebrae is related to peak strength such that the cortex accounts for 45% to 75% of the peak strength and the trabecular bone varies based on the ash content, with higher trabecular bone ash content affording greater force transmission.9 Given this information, analyzing the density of cortical and trabecular bone together should provide valuable information about the ability of bone to resist failure of fixation.

Our goal was to identify an area along the medial wall of the medial cuneiform of greatest combined cortical and trabecular bone mineral density. Placement of an internal fixation device on or through bone of greater density along the medial wall potentially offers the greatest resistance to failure. The null hypothesis for this study was that there is no significant difference in bone mineral density among 4 proportional quadrants along the medial wall of the medial cuneiform (ie, that the probability of 1 quadrant having the maximum bone mineral density…

Abstract

The human medial cuneiform is incompletely characterized with regard to anatomical morphology, including mineral density and bone quality. Clinically, we have observed failures of fixation by pull-through of devices through relatively soft medial bone. Defining patterns of relative density may provide valuable information regarding implant placement as higher cortical density bone may offer better resistance to such failures. We sought to identify an area of greatest density along the medial wall of the medial cuneiform.

Ten fresh-frozen human cadaveric medial cuneiforms underwent micro-computed tomography imaging. Images were analyzed to obtain densities in 4 quadrants along the medial wall of the medial cuneiform. Seven of 10 specimens revealed a maximum density in the plantar distal quadrant of the medial wall of the medial cuneiform. Chi-square goodness-of-fit testing indicated that the density of this quadrant was significantly different from 3 other quadrants (P<.009). Using the Principle of Standard Residuals, the density of the plantar distal quadrant was significantly different than the other 3.

We conclude that the plantar distal quadrant of the medial cuneiform contains bone of maximal density when compared to 3 other quadrants. Surgeons who place implants in this region should be aware that this area might offer better resistance to fixation failure.

Although the human medial cuneiform is an incompletely characterized bone, multiple surgical procedures use its medial wall as a point of fixation during the internal fixation of midfoot injuries and reconstructive procedures. By virtue of its anatomical morphology, the medial wall of the medial cuneiform provides the surgeon with a choice for the placement of an internal fixation device. It is desirable to place fixation devices at a location on or through this bone with the least likelihood of failure. By quantifying the density of bone along the medial wall of the medial cuneiform, we hope to offer information that will allow surgeons to make an informed decision regarding placement of internal fixation devices, particularly suture-button devices that have gained popularity.1-3

Previous studies have looked at the density of bone near the first tarsometatarsal joint, but to our knowledge, none have investigated the density along the medial wall of the medial cuneiform.4,5 Unlike other long bones in the body, the human medial cuneiform has limited cortical thickness.6,7 The medial cuneiform displays an architecture of cortical and trabecular bone such that they are similar in quality, thickness, and appearance. Other human bones, such as the vertebrae and carpal bones, display similar characteristics. In both cortical and trabecular bone, material and mechanical properties (strength, modulus, energy absorption, ductility, brittleness) are proportional to the density of the bone as well as the microarchitecture.8 Notably, the relative contribution of cortical and trabecular bone in the human vertebrae is related to peak strength such that the cortex accounts for 45% to 75% of the peak strength and the trabecular bone varies based on the ash content, with higher trabecular bone ash content affording greater force transmission.9 Given this information, analyzing the density of cortical and trabecular bone together should provide valuable information about the ability of bone to resist failure of fixation.

Our goal was to identify an area along the medial wall of the medial cuneiform of greatest combined cortical and trabecular bone mineral density. Placement of an internal fixation device on or through bone of greater density along the medial wall potentially offers the greatest resistance to failure. The null hypothesis for this study was that there is no significant difference in bone mineral density among 4 proportional quadrants along the medial wall of the medial cuneiform (ie, that the probability of 1 quadrant having the maximum bone mineral density is 25% in each of the 4 regions).

Materials and Methods

Ten fresh-frozen human medial cuneiform specimens were obtained from 7 donated cadavers. Use of this de-identified cadaveric research tissue was approved by our Institutional Review Board. Specimens were allowed to thaw to room temperature (approximately 20°C), and the medial cuneiforms were isolated, carefully removing all surrounding soft tissues and maintaining articular cartilage.

Micro-Computed Tomography Scanning

Specimens underwent volumetric cone-beam micro-computed tomography (CT) imaging using an eXplore Locus EVS-RS9 scanner (General Electric Healthcare, London, Ontario, Canada). Specimens were scanned with a resolution of 46 µm3. Other scanning parameters were as follows: tube voltage=80 kV; tube current=450 mA; number of views=720; exposure time=500 ms; number of frames to average=7. The device was calibrated with the SB3 phantom (General Electric Healthcare) per the protocol designed and published by the manufacturer to obtain the appropriate measurements calibrated in Hounsfield units. Our calibration values of both water and the SB3 were within reasonable error limits (<0.5%), with measured densities of 1 mg/cc and 1068 mg/cc, respectively (compared to known values of 0 mg/cc and 1073 mg/cc, respectively). With appropriate calibration established, a professional, experienced technician scanned each specimen, and reconstructed images were created. Images were stored on disk for later analysis.

Image Analysis

Images were analyzed using MicroView 3D Volume Viewer and Analysis Software (General Electric Healthcare). Using the 3-D analysis system, the images were oriented such that the medial wall was easily visualized. Four quadrants of the medial wall of the medial cuneiform were defined by measuring the proximal-to-distal distance and the dorsal-to-plantar distance, with the center of these 2 distances defined as the center point of the medial wall. Using this as the reference point, the medial wall was divided into 4 quadrants, with center points of each quadrant then located in a similar manner. A region of interest was defined centered at these points. The size of the region of interest was defined as 10 mm in the proximal-distal plane, 3.5 mm in the dorsal-plantar plane, and 3.5 mm in the medial-lateral plane, to match a footprint that equals the size of an implant known to be used in the fixation of midfoot injuries. Additionally, this size of region of interest contains enough volume to allow for obtaining an accurate statistical mean, avoiding sampling errors, which per the manufacturer requires a few hundred voxels.

The region of interest was positioned in the medial to lateral plane to allow for complete inclusion of cortex and the underlying corresponding cancellous bone. In each region of interest, bone mineral density analysis was performed. To include analysis of structures comprising normal bone, densities less than water (ie, air) were excluded by defining a lower exclusion arbitrary density unit of –400. Bone mineral densities were then obtained from the analysis software and recorded.

Statistical Analysis

Data were recorded and analyzed using Microsoft Excel (Microsoft, Redmond, Washington). Further analysis was performed using Internet-based software from VassarStats10 by performing a chi-square goodness-of-fit test to detect differences in values of density among the 4 quadrants. A significance level was defined as P<.05. Using the Principles of Standard Residuals, cutoff values of –1.96 to +1.96 were defined, with values falling in this range being no different than the 25% frequency expected with the null hypothesis.

Results

The plantar distal quadrant contains bone of the maximal density when compared to 3 other quadrants of bone along the medial wall of the medial cuneiform. Seven of 10 specimens revealed a maximum density in the plantar distal quadrant of the medial wall of the medial cuneiform (Table, Figure 1). Variability in the ranges of absolute values of bone mineral densities of the specimens were noticed (Figure 2); however, 1 quadrant, the plantar distal quadrant, was found to have the relative maximum density in 7 of 10 specimens (Figure 3).

Table


Figure 1
Figure 1: Composite of 6 micro-CT images taken from a left medial cuneiform. Three-dimensional rendering of a medial cuneiform showing location of 5 sections labeled A through E (top). Section taken near the proximal end (A). Section taken near the proximal fourth (B). Section taken at the midpoint (C). Section taken near the distal fourth (D). Section taken near the distal end. The star indicates the region of greatest density. Notice the thick cortex and trabeculae at this location (E).

Figure 2
Figure 2: Distribution of the 10 specimen bone mineral density values. Each specimen’s range of density is represented here, with the maximum density value represented by the dot.

Figure 3
Figure 3: Depiction of the combined distribution of maximum density of the 10 specimens. Note that 7 of 10 specimens have a maximum density in the plantar distal quadrant.

Statistical analysis with the chi-square test, assessing the null hypothesis that the frequency of maximum density in each quadrant was 25%, yielded a P value of .0089, indicating we should reject the null hypothesis. Further insight is revealed with the standard residuals yielding values of –1.58 for the dorsal distal quadrant, –0.95 for the dorsal proximal quadrant, +2.85 for the plantar distal quadrant, and –0.32 for the plantar proximal quadrant. As only 1 value, +2.85 for the plantar distal quadrant, falls outside of the range of –1.96 to 1.96, only this quadrant is statistically different than each of the other 3 quadrants (Figure 4).

Figure 4
Figure 4: Using the Principle of Standard Residuals from chi-square analysis, the plantar distal quadrant is shown to be statistically different than the other 3 quadrants. The plantar distal quadrant contains bone of the greatest density

Discussion

Surgeons must choose where to place implants based on the injury pattern, soft tissue considerations, and bone characteristics to achieve the highest likelihood of clinical success. When an injury pattern suggests the need for implant placement through the medical cuneiform, we have noted clinically that this bone presents varying areas of density with a tendency for implants to pull-through the bone. Based on this finding, we sought to determine if 1 area or quadrant along the medial wall of the medial cuneiform was of greater density and thus less likely to lead to such failures. This investigation is clinically important, particularly as it relates to the use of suture-button devices, such as the TightRope (Arthrex, Naples, Florida), which is US Food and Drug Administration approved for the treatment of midfoot injuries and has been described for use at this location elsewhere in the literature.1-3

Characterization of the medial cuneiform has been pursued by Coskun et al,4 who performed a densitometric analysis of the first tarsometatarsal complex. The aim of their study was to investigate the relationship between the density of the subchondral bone of the distal medial cuneiform and proximal first metatarsal at the tarsometatarsal articulation and define how it relates to the biomechanical properties and forces experienced in this area. As predicted by the principles described by Wolff’s law, they found that the dorsal and lateral regions of the distal medial cuneiform bone were more dense. These data only summarize density of the subchondral bone near the tarsometatarsal articulation and reveal nothing about the density of the medial wall of the medial cuneiform. It is this area that we were interested in characterizing further.

Muehleman et al5 performed a similar densitometric analysis, but used dual-energy x-ray absorptiometry of the first metatarsal and found this bone to be more dense laterally and dorsally. Additionally, they found the metatarsal head was more dense than the base. The patterns of density were related to areas of the bone subjected to greatest stress, again consistent with the principles of Wolff’s law.

Our findings indicate that the plantar distal quadrant of the medial wall of the medial cuneiform possess greater total bone mineral density than the other 3 quadrants. Clinically, this location correlates to that of the dense tubercle of bone beneath the course of the tibialis anterior tendon. These data offer valuable information to a surgeon when faced with an option of placing an internal fixation device on or through the medial wall of the medial cuneiform. Cortical thickness and cortical bone mineral density have been shown to affect the stability and failure rate of various fixation techniques in bone.6,7 The plantar distal quadrant of the medial cuneiform may offer more resistance to failure of fixation devices due to its greater density. These data indicate that if one were to place a suture-button device on the medial cuneiform, such as in the treatment of Lisfranc injuries, application of the button on that most dense quadrant would be potentially important to resist mechanical failure.

Strengths of this study include the use of highly accurate micro-CT technology and the use of fresh-frozen (non-embalmed) specimens. Micro-CT has been shown to be a valid tool for studying bone mineral density. The bone mineral density calculation is sensitive to changes in the size of the region of interest, and an accurate measurement requires just a few hundred voxels to sufficiently avoid sampling errors. One study using regions of interest containing single mouse vertebrae correlated well to the ash fraction method of bone mineral density determination.11 In addition, micro-CT offers a technology that can analyze a bone for multiple characteristics such as cortical thickness, bone mineral density, and dimensions, unlike ash fractioning. It also offers the advantage of archiving the raw images and data for possible future analysis of the human medial cuneiform. Embalmed specimens have been shown to have altered bone properties.12-14 We attempted to avoid potential confounding variables and sources of error in density analysis by using only fresh-frozen cadaveric specimens.

Potential weaknesses of this study include the use of an elderly specimen population and a lack of complete demographic information for all specimens. Although complete demographic information was not available, the age of each specimen was approximately 80 years. One concern is that this may allow for a greater presentation of osteoporosis as a potential confounding variable. However, because our study was concerned with density patterns as opposed to absolute density, we feel that this is not a significant limitation. It would be ideal to biomechanically test the hypothesis that the increased density corresponds to increased resistance to failure of fixation in this bone, but that was beyond the scope of this investigation.

The purpose of this study was to detect differences in the relative density patterns along the medial wall of the medial cuneiform. We have identified the plantar distal quadrant along the medial wall of the medial cuneiform as possessing greater bone mineral density than the other 3 regions. This finding may support placement of internal fixation devices in this location so as to minimize risk of failure of fixation.

References

  1. Panchbhavi VK, Vallurupalli S, Yang J, Andersen CR. Screw fixation compared with suture-button fixation of isolated Lisfranc ligament injuries. J Bone Joint Surg Am. 2009; 91(5):1143-1148.
  2. Cottom JM, Hyer CF, Berlet GC. Treatment of Lisfranc fracture dislocations with an interosseous suture button technique: a review of 3 cases [published online ahead of print March 19, 2008]. J Foot Ankle Surg. 2008; 47(3):250-258.
  3. Pelt C, Bachus K, Vance R, Beals T. A biomechanical analysis of a tensioned suture device in the fixation of the ligamentous Lisfranc injury. Foot Ankle Int. 2011; 32(4):422-431.
  4. Coskun N, Deniz Akman-Mutluay S, Erkilic M, Koebke J. Densitometic analysis of the human first tarsometatarsal joint [published online ahead of print December 23, 2005]. Surg Radiol Anat. 2006; 28(2):135-141.
  5. Muehleman C, Bareither D, Manion BL. A densitometric analysis of the human first metatarsal bone. J Anat. 1999; 196(Pt 2):191-197.
  6. Thiele OC, Eckhardt C, Linke B, Schneider E, Lill CA. Factors affecting the stability of screws in human cortical osteoporotic bone: a cadaver study. J Bone Joint Surg Br. 2007; 89(5):701-705.
  7. Tingart MJ, Apreleva M, Lehtinen J, Zurakowski D, Warner JJ. Anchor design and bone mineral density affect the pull out strength of suture anchors in rotator cuff repair: which anchors are best to use in patients with low bone quality [published online ahead of print July 20, 2004]? Am J Sports Med. 2004; 32(6):1466-1473.
  8. Hall BK. Bone. Caldwell, NJ: Telford Press; 1990.
  9. Rockoff SD, Sweet E, Bleustein J. The relative contribution of trabecular and cortical bone to the strength of human lumbar vertebrae. Calcif Tissue Res. 1969; 3(2):163-175.
  10. Chi-Square “Goodness of Fit” Test. VassarStats Web site. http://faculty.vassar.edu/lowry/csfit.html. Accessed April 18, 2009.
  11. Kozloff K, Thornton M, Goldstein SA. Validation of a micro-CT system for quantitative densitomety. Paper presented at: 5th Combined Meeting of the Orthopaedic Research Societies of the USA, Canada, Japan and Europe; October 10-13, 2004; Banff, Alberta, Canada.
  12. Boskey AL, Cohen ML, Bullough PG. Hard tissue biochemistry: a comparison of fresh-frozen and formalin-fixed tissue samples. Calcif  Tissue Int. 1982; 34(4):328-331.
  13. Koval KJ, Blair B, Takei R, Kummer FJ, Zuckerman JD. Surgical neck fractures of the proximal humerus: a laboratory evaluation of ten fixation techniques. J Trauma. 1996; 40(5):778-783.
  14. McElhaney J, Fogle J, Byars E, Weaver G. Effect of embalming on the mechanical properties of beef bone. J Appl Physiol. 1964; (19):1234-1236.

Authors

Drs Pelt, Bachus, Foreman, and Beals are from the Department of Orthopedics, University of Utah, Salt Lake City, Utah; and Dr Turner is from the Department of Orthopedics, Indiana University, Bloomington, Indiana.

Drs Pelt, Turner, Bachus, Foreman, and Beals have no relevant financial relationships to disclose.

This study was funded in part by a portion of a $10,000 grant from the American Orthopaedic Foot & Ankle Society with funding from the Orthopaedic Foot and Ankle Outreach & Education Fund, and by a portion of a $5000 grant from the AO North America Resident Research Support Program.

Correspondence should be addressed to: Christopher E. Pelt, MD, Department of Orthopedics, University of Utah, 590 Wakara Way, Salt Lake City, UT 84108 (chris.pelt@hsc.utah.edu).

doi: 10.3928/01477447-20110317-07

10.3928/01477447-20110317-07

Sign up to receive

Journal E-contents
click me