Treatment of tibial shaft fractures associated with significant soft tissue injury remains challenging. This study attempted to simulate a high-energy injury resulting in an unstable fracture pattern with significant soft tissue injury. Therefore, a segmental fracture with a standardized muscle contusion was used. Reaming was performed so that its effects could be evaluated within an injured soft tissue envelope. Limited reaming was used because of the well-known detrimental effects of standard reaming [4, 11–23]. Laser doppler flowmetry was employed to measure bone and muscle perfusion. This technique allows instantaneous blood flow measurements in vivo without the sacrifice of the experimental subject [29–31].
There was a profound hyperemic response in muscle perfusion after muscle contusion (Figure 2 and 3). This was pronounced within the zone of injury, although muscle perfusion at proximal and distal sites was also elevated compared to baseline. By the end of the initial procedure, muscle perfusion returned to baseline levels in Contusion and No-Contusion groups.
In the No-Contusion group, muscle perfusion overall and in the zone of injury did not statistically increase with reaming compared to baseline (Figure 2 and 3). This is consistent with previous studies [24, 25]. In the Contusion group, immediately following reaming there was a decrease in muscle perfusion overall and in the zone of injury compared to peak values, but not with respect to baseline (Figure 2 and 3). This could be because muscle perfusion was at its maximal level following contusion, and the natural trend with injury is for muscle perfusion to decrease with time. If perfusion within the zone of injury was maximal following contusion, even reaming could not elevate the perfusion any further.
Bone perfusion decreased to a larger extent in the Contusion group throughout the initial procedure (Figure 4). It is well-known that if the endosteal blood supply is disrupted, as it is in a segmental tibia fracture, the surrounding soft tissues are responsible for the remaining blood supply to the bone [18–23]. Although muscle perfusion was increased in the Contusion group, tibial blood flow was decreased compared to the No-Contusion group. The authors postulate that this could be due to the initiation of inflammation with a resulting diversion of more blood flow for muscle repair, rather than delivering more blood to the injured bone. Moreover, although canal diameters were not statistically different, the low statistical power of the study may not have allowed detection of any real differences present between the two groups. Thus, it may be that the 15% difference in average canal diameter contributed to this finding. In addition, with the muscle being significantly injured, the functional capability of the capillaries is unknown. Therefore, while blood flow is increased, the bone may not be receiving the benefits of increased muscle perfusion. At 11 weeks, the overall bone perfusion in the Contusion group remained significantly lower than the No-Contusion group. This may suggest that soft tissue injury either had a sustained affect on cortical perfusion or had no influence on bone healing. This would need to be addressed in future studies using functional bone healing measurements not done presently.
The mechanical stiffness and strength of the bone-nail repair construct following diaphyseal fracture with simultaneous muscle contusion were not presently considered. A similar prior study assessed the effect of limited versus standard reaming on the 4-point bending stiffness and strength of nails used to repair segmental tibial shaft fractures in a series of canines, but without muscle contusion . The results showed statistically significant decreases in repair construct stiffness (limited reaming, 30%; standard reaming, 46%) and strength (limited reaming, 35%; standard reaming, 22%) compared to intact contralateral tibias. Similar relative decreases in mechanical characteristics might be expected compared to intact tibias for the present specimens, with or without muscle contusion, had biomechanical tests also been performed. In addition, a recent study by Melnyk et al. quantified the revascularization process for diaphyseal fractures with and without surrounding soft tissue injury in a rat model . They report that partly destroyed bone-soft tissue interaction resulted in only temporary reduction of extraosseous blood supply, which might not affect fracture healing. They also tested the mechanical properties of the repair constructs with and without surrounding soft tissue injury at four weeks post-operatively and found no statistical difference in failure load or flexural rigidity. The present authors, thus, suggest that the extent of blood perfusion into surrounding muscle and bone around the fracture site may not have any significant long-term affect on fracture healing and, hence, the mechanical stability of the bone-nail repair construct.
There were limitations to this study. Firstly, the choice of contusing the anterior compartment was arbitrary and may not be representative of all clinical scenarios. Posterior compartment injury is common and may significantly affect perfusion to the tibia. The degree of soft tissue injury will vary in the clinical setting.
Secondly, a small series of 11 animals was used due to funding and sheltering limitations. As such, the post hoc power analysis showed the study was underpowered with values of 24% (overall perfusion) and 38% (intercalary segment). Moreover, although canal diameters between groups were statistically not different, the low statistical power suggests that confounding effects due to canal size may have occurred.
Thirdly, a post hoc, rather than an a priori, statistical power analysis was performed. It is theoretically preferable to perform an a priori power analysis for initial study design to determine how many specimens should be included in an investigation to avoid statistical type II error. However, it is often difficult to do so because of large interspecimen variability and the unpredictability of outcome measures among specimens. Even when it seems possible to perform such a computation confidently, a post hoc power analysis is still necessary to confirm the statistical power of the study using the actual, rather than the predicted, results of the study.
Fourthly, a static 400 N load lasting 20 seconds was applied over a known contact area in order to create a reproducible model of muscle contusion that could be applied in a standardized manner in a laboratory setting. Although 400 N did alter the perfusion profile of bone and muscle, it is difficult to assess how representative this is of most high energy open tibial shaft fractures. Thus, higher load levels applied dynamically for a shorter time period would have more realistically simulated an impact injury. For instance, previous studies showed that a transverse load of about 750 N is required to fracture the diaphyseal region of a dog femur using an impact load applied at 3 m/s , whereas about 5270 N is required to fracture the proximal portion . If an impact injury to the muscle was simulated at present, this may have increased the initial amount of blood loss and subsequently altered the current contusion group blood perfusion results. However, standardizing simulated impact injuries may not always be feasible because it requires recreating the same load level, load application time, and contact area for each animal. Therefore, a standardized static load approach was used in this investigation. The comparative nature of the study may allow the present results to be generalized to higher and dynamic loads.
Fifthly, the authors hypothesize that during the surgical procedure, the small differences in muscle and bone perfusion may have been due to the manipulation and/or injury of muscle bellies and adjacent soft tissues. The effect of this confounding factor, however, would need to be determined more conclusively in future investigations.
Sixthly, No-Contusion and Contusion groups both eventually healed. Thus, the clinical significance of the differences found in blood perfusion into muscle and bone is unknown. However, conditions under which adequate blood flow to surrounding bone and soft tissue can be maintained during trauma surgery and under which significant blood loss can be minimized, may possibly eliminate hemorrhagic or hypovolemic shock, reduce the need for post-operative blood infusion, increase fracture healing rate, and shorten patient recovery time .
Seventhly, the current surgical model simulated a segmental fracture of the tibial shaft, which is the least common type. Of all tibial shaft fractures, about 54% are simple fractures that have a spiral or oblique pattern, about 28% are wedge fractures, and about 18% are comminuted or segmental . However, a segmental fracture was used because it was the easiest to simulate consistently from specimen to specimen in a research setting. Moreover, the current study using a segmental fracture in the presence of muscle contusion could then be compared with prior studies by some of the authors who also used a segmental fracture, but without muscle contusion [24, 25].
Finally, although beyond the scope of the current study, future investigators could consider assessing the effect of standardized muscle contusion on the changes incurred on two other parameters of interest. Specifically, radiographs could be assessed and biomechanical tests could be performed to determine the amount of bone healing (or callus formation) at the fracture site [14, 15]. Moreover, an evaluation could be done to determine whether muscle histology has fully recovered or whether the contusion site has been replaced totally or partially by scar tissue.