increases with bone formation.
In the acute stages, inflammatory cytokines such as interleukins, tumour necrosis factor and prostaglandins are released from platelets and inflammatory cells, stimulating progenitor cells to migrate to the area of damage [15]. Local cells and the forming matrix then release stimulatory factors such as fibroblast growth factor (FGF), platelet derived growth factor, transforming growth factor beta, BMPs and Wnt glycoproteins [15]. These have all been shown to have a stimulatory effect on bone healing, at both cellular and macroscopic levels. Vascularity at the site is also stimulated by release of vascular endothelial growth factor (VEGF) and prostaglandin E2 (PGE2) [25].
In contrast, chronic inflammation (which is most commonly due to infection) results in persistence of inflammatory cytokines, consequential degradation of healing tissues, pain and continuation of an immature matrix that cannot stimulate an osteogenic response. At the cellular level, the controlled release of inflammatory mediators is overwhelmed, leading to dysregulation of the osteoimmunologic response. Osteoblast proliferation is reduced, thus decreasing the immunosuppressive balance they induce. In the presence of infection, bacterial factors such as endotoxic lipopolysaccharide (LPS) produce soluble inflammatory factors leading to osteolysis [26].
Figure 6.2 Relationship of inflammatory and osteogenic cells.
Source: Florence Loi et al. [15]. Reproduced with permission of Elsevier.
Mechanical Influences on Bone Healing
Unlike other tissues, the mechanical environment of bone has a significant influence on the healing environment. Roux initially coined the term ‘developmental mechanics’ in which he hypothesized that the cell type involved in healing is based on the mechanical load [27]. Wolff described skeletal tissue as organized to optimize strength in response to loading [28]. Pauwels took this further to show that progenitor cells differentiate in response to the nature of mechanical load [29]. This classic work explains the influence of the mechanical environment on bone adaptation and demonstrates that the stability of a fracture will dictate the type of cells and tissue matrix that will occupy the site and thus determine the quality of healing.
The mechanical strain that occurs at the time of fracture can lead to significant vascular changes. Increased strain results in continued vascular damage, decreased oxygen tension and consequent stimulation of chondrocyte formation [18, 30, 31]. Local strains of less than 5% lead to intramembranous ossification, strains of less than 15% lead to endochondral ossification and strains greater than 15% result in formation of fibrous tissue and hence lead to non‐union. In contrast, excessive reduction in the local strain environment due to over stabilization by a repair (which has never been documented in horses) can lead to reduced bone healing. Loss of low‐level strain can lead to reduced external callus formation, fracture end osteolysis and adverse remodelling [32]. The vascular response to fracture, since it is dependent on local strain, will cause differences in healing type. With physiologically sound rigid stability and compression of fracture ends, local vascularity is enhanced and bone formation can occur [18, 30, 31]. However, in secondary bone healing, increased strain and a void between the fracture ends will lead to a relatively hypoxic area in which only chondrocytes can thrive. Progenitor cells differentiate into chondrocytes, and as these fill the fracture gap (the soft callus phase of healing), strain decreases due to a relative increase in stability and the environment becomes conducive to the formation of hard callus.
As stability is a major factor in influencing healing, meticulous attention must be paid to adequate reduction, debridement, and application of fixation principles for optimal stabilization. With the advent of minimally invasive procedures, debridement through open reduction is often not necessary. In minimally displaced fractures, the requirements for meticulous reduction to produce stability is overcome by enhanced rigidity of the locking plate system. As experience with minimally invasive techniques increases, the limits of reduction will be tested and further guidelines will evolve.
Discussion of the mechanical environment raises the question of appropriate time for implant removal. At the later stages of bone healing, it is possible that implants can shield stresses that may be necessary to complete healing and restore full bone strength. This continues to be debated in human and veterinary medicine as clinicians constantly question when the mechanical strength of the bone (without the implants) is optimal for removal without reinjuring the fracture site [33]. This is difficult to determine objectively. In most equine repairs, either a staged removal occurs for example if two plates are used or the animal may be exercised with the implants in place in order to apply some stress to the bone before removal. The form of exercise can vary according to individual circumstances. A progressive transition in mechanical environment can follow implant removal to apply gradually increasing loads. Care must be taken from the clinical perspective to be assured that the healing bone is not overloaded. However, at this time there are no objective means of determining bone strength or resilience, and judgement must be made on the basis of clinical signs and results of imaging.
Monitoring Bone Healing
Most derangements in fracture healing require surgery or further surgery and a change in fixation technique. Exogenous therapies may also be of benefit (Section Exogenous Factors That Influence Fracture Healing). The point at which revision must be considered is difficult to determine objectively, but persistent or progressive pain and/or instability are pivotal in decision‐making.
Fractures with impaired healing are usually described as delayed or non‐unions. A delayed union requires increased time, but healing will occur without surgical (or further surgical) intervention [6]. In adult horses, normal cortical bone healing is thought to occur within four months, and in foals within three months [16]. In contrast, a non‐union cannot heal without surgical intervention [6]. Mechanical factors, principally lack of stability, are the most common causes of non‐union; however, biological factors including impaired vascularity (usually due to severe soft tissue damage) and infection can play a role. Non‐unions have been defined by their radiographic appearance and clinical symptoms divided into biological reactive and biological non‐reactive, non‐viable unions [6]. Biological reactive non‐unions are further classified according to radiographic appearance: hypertrophic non‐unions (elephant foot non‐unions) have exuberant callus formation due to instability, while oligotrophic non‐unions lack callus. Horses with non‐unions generally have less callus formation and are a milder form of hypertrophic non‐union compared to humans. Biological non‐reactive, non‐viable non‐unions are defined by lack of activity on nuclear scintigraphic examination. This is typically caused by lack of vascularity at the fracture site. Torsion wedge non‐unions fail to heal due to lack of fragment vitality. Comminuted fracture non‐unions are characterized by a devitalized intermediate fragment; the fracture ends are vascular, but the intervening fragment is avascular. Defect non‐unions occur at sites of bone loss or an intervening infected area. Non‐unions when fibrous tissue alone develops within the defect are described as atrophic [6]. In appropriate cases in humans, vascular grafts and stabilizing techniques can be used to overcome non‐union healing. However, the need for immediate weight‐bearing and cost often restrict use of these techniques in horses.
In clinical practice, objective assessment of fracture healing is difficult. Clinicians generally rely on pain and planar imaging (radiographs) to dictate management. Pain is constantly monitored. In cortical bone, the periosteum contains many nerve endings, and in fractures these are activated creating painful stimuli. The associated inflammatory response also increases nociception, and there is evidence that even with fracture repair, there is an ingrowth of nerve endings into the site [34]. Chronic pain frequently reflects instability, and one of the primary goals of repair is to produce a rapid decrease in pain in order to prevent contralateral limb overload (Chapter 14). In most cases, this can be achieved with rigid internal fixation,