INTRODUCTION: Most injuries in recent conflicts are due to blast, 70% of which are to the extremities resulting in a large number of lower limb amputations. Functional deficits due to blast induced amputation include difficulty in weight bearing and associated normal gait abnormali-ties. Significant complications following traumatic amputation are pain in the residual limb, and phantom limb pain. Heterotopic Ossification (HO) - ectopic bone formation in the soft tissues - is also highly prevalent (64%) among blast-related military amputations. The existing non-specific treatments include non-steroidal anti-inflammatory drugs (NSAID)s and low-dose radiation therapy which remain unsatisfactory leav-ing surgical bone excision the only possible curative treatment. While the prevention of HO in military amputees is the ultimate choice of treat-ment, it is yet to be identified, as the initial cause of triggering the disease is not understood. For this reason, and because studying amputation complications in humans is difficult, novel in vivo models need to be developed for further understanding of the disease mechanisms. There-fore, we hypothesised that developing a preclinical blast injury model in the hindlimb of rats which better represents the IED detonation in en-closed spaces could answer questions regarding the exact mechanism of HO and phantom limb pain. Current in vivo models exist, but none of these incorporate all blast features, that is, the blast, and the fracture in one insult. This research aims to develop a novel translational blast injury model in rats to better understand the mechanisms of phantom limb pain and HO.

METHODS: This study was performed under institutional and departmental license from the Home Office UK. In line with the 3Rs principle, optimisation of the blast pressure was achieved using 34 male cadaveric Sprague-Dawley rats weighing between 285-481g to refine the experi-ments without using live animals to achieve a trans-tibial fracture at the left hindlimb utilizing different burst pressures (7-13bar). The rats were placed on a special harness which supported and protected the rest of the body exposing only the left hind limb to different blast waves gener-ated by a shock tube. Both blast and blast associated fracture were induced in one experiment. The tibial fracture was evaluated by x-ray and confirmed by dissection.

RESULTS: Different blast pressure parameters showed that the blast waves followed a normal pattern including peak pressure, positive pres-sure duration followed by a negative under-pressure and consequent return to the ambient pressure (Figure 1A). The peak pressure, impulse and the positive duration decreased predictably (Figure 1B,1C, 1D) by reducing the blast pressure. 13 and 12 bar pressures produced multiple fractures in the hip, femur and tibia as well as above and below knee fractures on both limbs (Figure 2) rendering these pressures inappropriate for translation into a survivable model. Experiments at 11 and 10 bar caused consistent above (undesirable) and below (desirable) knee frac-tures on both (undesirable) limbs. Reducing the pressure to 9 bar resulted in 75% reduction in the injuries to the right limb and 62.5% reduction on the left limb, producing highly consistent isolated unilateral tibial fractures in most cases. The anomalies were in the smaller animals (285-300 g) that resulted in 25% and 37.5% undesired injuries in the right and left limbs, respectively. 7 bar pressure induced no fractures.

DISCUSSION: In this study we developed a blast injury model in the left rat hindlimb which can be translated into a survivable model to study conditions following blast-induced amputation. Animal size was a significant factor and thus, small animals should be excluded from in vivo studies. This study recommends 9 bar pressure in medium-sized male Sprague-Dawley rats (320-450g) within a custom shock tube and har-ness to achieve consistent blast-induce isolated unilateral fractures of the tibia. This is the first model that combines the blast and the fracture in a single insult without using a drop weight to separately induce the fracture, therefore, better simulating the battlefield scenario. The translation of this model opens up new avenues to explore the systemic serum biomarkers and to identify novel signalling pathways associated with HO initiation and progression following blast injuries and could potentially be used to analyse other blast-related conditions.

SIGNIFICANCE: Developing a preclinical blast injury model is key for understanding the mechanism of HO suffered by military personnel following blast injuries. This will lead to novel diagnostic tools in the clinic and specific therapeutic design that will prevent HO, thus improving veterans’ quality of life.