Temporal blood flow changes measured by diffuse correlation tomography predict murine femoral graft healing.

Songfeng Han,Regine Choe,Danielle S W Benoit,Ashley R Proctor,Jingxuan Ren,Yufeng Dong

doi:10.1371/journal.pone.0197031

Songfeng Han, Regine Choe + Show 4 more

Open Access

https://doi.org/10.1371/journal.pone.0197031

Copy DOI

Abstract

Blood flow changes during bone graft healing have the potential to provide important information about graft success, as the nutrients, oxygen, circulating cells and growth factors essential for integration are delivered by blood. However, longitudinal monitoring of blood flow changes during graft healing has been a challenge due to limitations in current techniques. To this end, non-invasive diffuse correlation tomography (DCT) was investigated to enable longitudinal monitoring of three-dimensional blood flow changes in deep tissue. Specific to this study, longitudinal blood flow changes were utilized to predict healing outcomes of common interventions for massive bone defects using a common mouse femoral defect model. Weekly blood flow changes were non-invasively measured using a diffuse correlation tomography system for 9 weeks in three types of grafts: autografts (N = 7), allografts (N = 6) and tissue-engineered allografts (N = 6). Healing outcomes were quantified using an established torsion testing method 9 weeks after transplantation. Analysis of the spatial and temporal blood flow reveals that major differences among the three groups were captured in weeks 1–5 after graft transplantation. A multivariate model to predict maximum torque by relative blood flow changes over 5 weeks after graft transplantation was built using partial least squares regression. The results reveal lower bone strength correlates with greater cumulative blood flow over an extended period of time (i.e., 1–5 weeks). The current research demonstrates that DCT-measured blood flow changes after graft transplantation can be utilized to predict long-term healing outcomes in a mouse femoral graft model.

Highlights

More than 2.2 million graft procedures are performed in the clinic every year to treat criticalsized bone defects [1], which will not heal without intervention [2]
With the diffuse correlation tomography (DCT) system, we have revealed that blood flow changes are different among three groups of autografts, allografts, and T.E. allografts
We investigated the spatial and temporal features in blood flow changes to select the most relevant data; used partial least squares (PLS) regression to build a multivariate model to predict maximum torque using blood flow changes from week 1 to 5

Summary

Introduction

More than 2.2 million graft procedures are performed in the clinic every year to treat criticalsized bone defects [1], which will not heal without intervention [2]. The “gold-standard” allograft treatment uses processed bone material from a cadaver and is advantageous in terms of the amount of available bone material. Allografts have a 60% failure rate over 10 years post-implantation [1, 3, 4]. To avoid infection or immunological responses in patients, allografts must be thoroughly cleaned to remove all living tissue. Poor healing typical of allografts is due to loss of the periosteum: a thin layer of tissue covering the bone [3, 5]. Autograft procedures that utilize healthy autologous bone tissue with periosteum from a non-load-bearing skeletal region usually achieve complete healing, but this approach is limited by lack of available tissue volume and donor site morbidity

Methods

Results

Discussion

Conclusion