Quantitative Assessment of Osseous Healing of the Reconstructed Mandible Utilizing High Resolution CT

Thursday, October 10, 2013
Eric D. Hoverstad DDS, Oral and Maxillofacial Surgery, Boston University Goldman School of Dental Medicine / Boston Medical Center, Boston, MA
Andrew R. Salama DDS, MD, Boston University Goldman School of Dental Medicine / Boston Medical Center, Boston, MA
Peter Hoang BS, Boston University School of Medicine, Boston, MA
Hernan Jara PhD, Radiology, Boston University School of Medicine, Boston, MA
Osamu Sakai MD, PhD, Radiology, Boston University School of Medicine, Boston, MA
The microvascular free fibula flap is a commonly used reconstructive modality in mandibular reconstruction following extirpative surgery for benign and malignant tumors. Bone healing in cortico-cancellous bone grafting of the mandible has been elegantly described and is widely accepted1.  However, the histologic and radiographic evidence of bone healing in bone containing free flaps has not been well documented or described.  A quantitative serial retrospective analysis was undertaken in patients who underwent mandibular reconstruction with free fibula flaps to elucidate radiographic evidence of bone healing using high resolution computed tomography (CT).  

A matched pair analysis of pre-operative and post-operative CT scans was performed on patients who underwent mandibular reconstruction with free fibula flaps at Boston Medical Center from 2008-2012.  Inclusion criteria included a pair of CT scans; a 30-day peri-operative scan and a subsequent study no earlier than 6 months post-operatively.  Nine (9) patients with complete records were included in the final analysis. The CT scan DICOM data was processed with commercially available software (OsiriX, Pixmeo; Geneva, Switzerland).  Two types of osteotomies were analyzed; native mandible and the fibula graft (nm-g) and within the reconstructed fibula graft (g-g).  A Region of Interest (ROI) was created using serial axial images around each osteotomy site using a range of 500-2800 Hounsfield Units (HU) to indicate bone formation.   A 3-dimensional volumetric analysis was performed comparing peri-operative to post-operative volumes. 

Nine patients with 17 reconstructive junctions, 14 nm-g and 3 g-g, met our inclusion criteria and were analyzed. Volumetric calculations of each junction were analyzed comparing peri-operative volumes to post-operative volumes. Of the nm-g junctions analyzed, 43% (6/14) demonstrated and increase in bone volume.  The remaining 57% (8/14) demonstrated a decrease in bone volume.  Among the three graft-graft junctions analyzed, 100% (3/3) demonstrated an increase in bone volume. 

Analysis of the volumetric data indicates that the nm-g junctions may heal in a similar manner as compared to long bone fractures.  Studies in long bone fractures have clearly demonstrated local and regional bone loss during healing2.  In g-g junctions, an increase in bone volume was discovered over time.  This may be due to the intact fibula periosteum that is maintained throughout the procedure, providing a constant source of blood supply to the osteotomy sites.  In the nm-g junctions, extensive periosteal stripping occurs which may decrease overall blood supply at these junctions. These preliminary results would benefit from an increased study population where further conclusions and potential clinical correlates could be discovered.

1. Marx, Robert E. Bone and Bone Graft Healing. Oral Maxillofacial Surg Clin N Am. 19 (2007): 455–466.

2. Veitch, S.W. et al. Changes in Bone Mass and Bone Turnover Following Tibial Shaft Fracture. Osteoporos Int.  17 (2006): 364-372.