Biomedical Engineering (BME) - Story Archives: Bone Graft Strength Accurately Predicted

Periodically, the Science, Technology, Engineering, & Mathematics Portal showcases new Biomedical Engineering discoveries. As we post new material to the Biomedical Engineering Portal, we move other Biomedical Engineering stories to the archives.

Return to Biomedical Engineering Story Archives | Return to Biomedical Engineering Portal

Bone Graft Strength Accurately Predicted:
    Reduces Surgery, Enables New Therapy for Bone
    Regeneration

    (Assistant Professor Hani Awad, Biomedical
    Engineering and Orthopaedics)

     by Edward Katich & Lois H. Gresh

October 19, 2007:  A team led by Dr. Hani Awad at the Center for Musculoskeletal Research (CMSR) has developed a method to accurately predict the strength of in-vitro bone grafts in a preclinical model. Clinical researchers may soon use this methodology to aid in conducting clinical trials of new therapies for bone grafts enrolling only dozens, not thousands, of patients. In addition, rather than undergo multiple surgeries for years, a patient can now have scans done of his damaged bone, allowing doctors to choose the best regenerative treatment and to change the course of the treatment based on non-invasive imaging.

Most patients with severe osteosarcoma bone cancer receive allografts, which are sterilized bone transplants from organ donors, in an effort to salvage the limb after tumor resection. These treatments often fail because the allografts aren't viable which makes them susceptible to micro-damage accumulation that decreases their durability and leads to failure. As a result, patients might undergo numerous painful and fruitless revision surgeries to implant metal braces, metal scaffolds, and even artificial joints.

"To study why allograft transplants fail," says Dr. Awad, "we compared allografts to live autografts, which are harvested from the patient's tissue. In current practice, allografts are the standard of care since in most cases autografts are not an option. Allografts are sterilized to remove all traces of the donor's cells, leaving the bone matrix behind. Any donor cells would trigger an immune response in the recipient, causing the graft to be attacked. However, this processing creates other problems in that, unlike autografts, the lack of living cells in processed allografts makes them unable to repair micro-fractures that result from daily activity which leaves them prone to fractures."

The group in the CMSR studies this problem in a preclinical model that was first described in 2004. To simulate an autograft, a critical mass of femur tissue is removed and then re-implanted back into the same femur. To simulate an allograft, a critical mass of femur tissue is harvested from a donor, sterilized, and then implanted it into the femur of a different subject.

What became clear through basic histologic imaging was that the body treats an autograft like body tissue and eventually merges it into the original bone tissue. The allograft, however, is treated like a large splinter; the body builds up a large callus around the graft ends and does not merge it into the original bone tissue.

2-D Bone Graft Images, Dyed
In the histology picture below, the red is soft tissue, the blue is cartilage, and the orange is hard bone tissue. Over time, the autograft is integrated into the old bone, with cartilage forming around the entire graft. The body begins endochondral bone formation, the process by which bone tissue regenerates. In the allograft, however, the old bone never actually envelopes the graft, but rather just builds up large calluses around the junctures.
(Provided courtesy of Xinping Zhang)

What factors in an autograft lead to bone growth and strength that are absent in an allograft?

To determine the strength of femur grafts, Awad applied a common mode of loading that long bones endure: torsion. To test the grafts' torsional strength, the researchers used an EnduraTEC TestBench Torsional System to twist each femur until it snapped. While it was clear that the allografts' strength degraded over time in contrast with the autografts, this fact did not shed light on why allografts are weaker.

Assessment of Graft Torsional Strength
This graph is a plot of the torque vesrus the bone rotation about its axis. The peak of each line graph occurs at the breaking point of the bone. Note how the normal bone resisted the greatest amount of torsion, or torque, with little rotation, while the autograft was significantly weaker. The allograft preformed the worst during these tests, contorting dramatically with little torsional resistance.

Previous work in the CMSR conjectured that perhaps vascularization, or the growth of blood vessels through tissue, could help the process of bone regeneration and affect the strength of the grafted bone. To view the vascularization in the grafts, the group used micro computed tomography (CT) which is simply a three-dimensional x-ray imaging modality. X-rays, however, cannot scan soft tissues, and so the blood vessels were injected with a polymer. After an enzyme dissolved the hard tissues, only the polymerized blood vessels remained to be imaged. The resulting images showed how blood vessels grew throughout the autograft but failed to grow in the allograft.

Graft Vascularization Imaging
This is a computer-generated image of the vascularization in the grafts. The red vessels run through the grafts, while the yellow vessels are part of external vascular network in the surrounding tissues. F and H are the allograft, while G and I are the autograft. Removing the external vessels - shown in H and I - one can see the dramatic difference between the blood supplies to the two grafts.
(Provided courtesy of Xinping Zhang)

Next, using Micro-CT to directly scan the bone grafts, the researchers generated three-dimensional callus volume images. Again, the autografts had larger callus volumes than the allografts. "But," says Dr. Award, "while this is a factor of bone strength, it is not the only one."

He continues: "Another important factor is the Polar Moment of Inertia (PMI) of the bone. The PMI value describes the amount and geometric distribution of the mineralized callus tissue relative to its centroid in any one horizontal slice of the graft." Both the PMI and the callus volume can be determined from a Micro-CT scan.

After statistically correlating the two values with the experimentally determined bone strength, however, these factors could account for only 50% of the variability in the experimentally determined strength of the graft. While this number is significant, it is not high enough to reliably let doctors use the technique clinically with human patients. "We knew," said Dr. Awad, "that there must be yet another factor that seriously affects bone strength. We had an idea and we were determined to find a way to prove it."

Micro-CT Bone Graft Image and Cutaway
As noted by the arrows, the allograft images B and D show how the graft has not been incorporated into the original bone like in autografts A and C. While the Micro-CT scans produce sharp images, those images do not easily lend themselves to statistical measurements.

The missing link turned out to be a new value, the Union Ratio, a novel parameter first measured and coined by Ph.D. candidate David Reynolds. The Union Ratio is a ratio of connected surface area of graft bone tissue with the host and callus bone to the total surface area of the graft. Using algorithms developed by Reynolds, the Union Ratio can be accurately quantified from the Micro-CT scan. For his pioneering work, Reynolds recently won First prize in the PhD student competition of the 2007 meeting of the Tissue Engineering and Regenerative Medicine International Society (TERMIS) meeting in Toronto, Canada.

Union Ratio Visual Representation
The blue areas represent graft surfaces that are unconnected to living bone tissue. The red areas represent connections between the original bone or mineralized callus and the graft. As expected, the autograft's connections are much more evenly spaced along the graft. The Union Ratio is the ratio of red area to blue area.

Reynolds found that accounting for the Union Ratio in the statistical model in addition to the callus volume and the PMI improved the predictive power of the correlative model to 75% of the variability of the experimentally measured bone strength.

This accuracy is high enough to enable researchers to use the technique in clinical trials of bone graft gene therapies.

Currently, Dr. Awad and Mr. Reynolds are working with several associates, including Dr. Edward Schwartz, Professor of Orthopaedics at the University of Rochester's School of Dentistry and Medicine, to translate these findings to clinical applications and to develop growth gene therapies for artificial grafts using the above techniques. They are adapting the Micro-CT scanning methods described above to clinical CT scanners to determine the effectiveness of different growth gene solutions on bone allografts and artificial bone scaffolding implants.

Measured versus Predicted Ultimate Torque of Bone Grafts
The line in the graph is the identity line representing the ideal perfect fit between experimentally measured and predicted values. The scattered dots are individual data of bone grafts. The high estimation accuracy (adjusted R2=0.75) means that Micro-CT scanning can be used in estimating bone strength with high reliability.

Dr. Awad believes that the new technique will aid in the development of effective gene therapies and bone tissue engineering using artificial polymer scaffolds. When implanted, these scaffolds will grow into fully-functioning living bones. "Here's what I envision," says Dr. Awad. "Current 3D printers will construct the Micro-CT scan of a patient's bone into a 3D scaffold using polymer powder. The 3D scaffold will be treated with the most effective gene therapy, and then it will replace the diseased bone. Months after the procedure, the patient will feel no difference between the old pre-disease bone and the new fully functioning bone." Patients no longer will suffer from repeated surgeries and artificial joint replacements.

Additional Details

For further details, see:

Reynolds DG, Hock C, Shaikh S, Jacobson J, Zhang X, Rubery PT, Beck CA, O'Keefe RJ, Lerner AL, Schwarz EM, Awad HA. "Micro-computed tomography prediction of biomechanical strength in murine structural bone grafts," J Biomech. 2007; 40: 3178-3186.

Reynolds DG, Hock C, Shaikh S, Zhang X, Rubery PT, Beck C, O'Keefe RJ, Lerner AL, Schwarz EM, Awad HA. "Novel Measurement of Bone Graft-to-Host Union Using CT Imaging: Implications for Biomechanical Strength," Tissue Engineering and Regenerative Medicine International Society - North America (TERMIS-NA) Conference and Exposition (2007), Toronto, Canada.

Awad HA, Zhang X, Reynolds DG, Guldberg RE, O'Keefe RJ, Schwarz EM. "Recent advances in gene delivery for structural bone allografts," Tissue Eng. 2007 Aug 13(8):1973-85.

For more information, please contact:
   Assistant Professor Hani Awad
   Email: Hani_Awad@URMC.rochester.edu
   Faculty Webpage: http://www.bme.rochester.edu/bmeweb/faculty/awad.html

This article was co-authored by Edward Katich as part of the curriculum of PHY 396, Supervised Science Writing I. Edward is a Microbiology undergraduate student.

Return to Top

Return to Biomedical Engineering Story Archives | Return to Biomedical Engineering Portal

Return to Science, Technology, Engineering, & Math Portal Homepage