Offloading treatment is linked to activation of proinflammatory cytokines and start of bone repair and remodeling in Charcot arthropathy patients

The Charcot foot is often characterized by local inflammation, diastases, joint dislocations, and fractures. A great number of experimental and clinical studies have shown that most fractures heal by overlapping phases characterized by inflammation, bone regeneration and remodeling and that these phases normally follow a preset schedule with a rather controlled set of events when fracture healing is allowed to proceed undisturbed (Fig. 1) [1, 2]. In the primary phase of fracture healing, proinflammatory cytokines, such as IL-6, IL-8, IL-1β, IL-17 and TNF-α, are expressed and released within the first hours or days after tissue injury. This initial cytokine boost from the damaged bone and soft tissue is transitory in nature and only lasts a few days but yet is responsible for a number of key events such as removal of necrotic tissues and recruitment of neutrophils and mesenchymal stem cells [1, 3]. Stem cells are stimulated to undergo differentiation into chondrocytes leading to formation of cartilaginous callus that stabilizes the fracture and, when absent or disturbed, will have detrimental effects on mesenchymal cell maturation and callus formation [3]. An intermediate phase of low activity by proinflammatory cytokines follows and is characterized by cartilaginous callus being calcified and replaced by woven bone. A secondary or late phase of expression and release of proinflammatory cytokines will ensue and stimulate the transformation of stem cells into osteoblasts and osteoclasts allowing for the coupled remodeling phase characterized by woven bone of the callus being replaced with lamellar [1, 3].

We currently know little of the mechanisms triggering the pathological changes characteristic of Charcot arthropathy although trauma to the foot and the ensuing inflammatory reaction has been proposed as an important driving force [4]. Although the acute Charcot foot presents itself with the classical features of a local inflammatory reaction, i.e. a red, hot and swollen foot [5], the pathogenic role of inflammation can be multifaceted as it may both play a primary role in triggering the onset of the destructive processes leading to the anatomical changes of foot, or be secondary to these changes [6]. Inflammation, being a general response by the body to all types of trauma, serves as an important part of the repair and defense mechanisms, before subsiding when the cause of inflammation has been removed or the healing process has been allowed to proceed uninterrupted. A critical factor for the Charcot patient in this context is their loss of pain as protective sensory input and thus inability to identify the severity of trauma which will often delay the onset of medical treatment by weeks or months. This, in turn, will lead to continued weight-bearing on the diseased foot with motion at fracture site causing delayed bone union which has led clinicians globally to adopt a treatment strategy of prolonged non-weight-bearing to create better healing conditions [7, 8]. As proinflammatory cytokines have been shown to play a pivotal role in bone repair and there are no previous longitudinal studies investigating their role in Charcot patients we set out to monitor systemic levels from presentation until a clinically and radiographically documented cold/chronic state has been reached.

This observational prospective study was approved 2007 by the Review Board of Västra Götalands Regionen (EPN D-number 499–07) with additional approval 2010 (T 762–10 ad 499–07). All the subjects provided informed consent before participating in the study. The study complies with the STROBE-statement for observational studies [9].

Clinical and demographic data are presented in Table 1. Radiographic data are presented in Table 2. Tables 1 and 2 have in part been presented in a previous study [12]. One patient had no radiographic data after 18 months into the study as the patient chose to interrupt participation.

Although radiographs showed significant bone pathology in a majority of Charcot patients at presentation (Table 2), we were only able to identify four Charcot patients with visible callus on radiographs despite that 12 weeks had elapsed since debut of symptoms (8 weeks for the Charcot group as a whole). The time for bony callus to appear on radiographs may depend on the location of a fracture but will in most cases be visible within 3 weeks of the initial trauma. It is thus conceivable to assume that bony healing had failed to materialize in most of our Charcot patients with little proof of progress beyond the endochondral phase. Several factors could account for this lack of progress. One is the diabetic disease per se which, aside from being associated with higher incidence of fractures than the population as a whole [15, 16], is prone to develop complications in the sequel of a fracture by as much as 87 % and often heal with impaired bone quality [17‐19]. Accumulation of highly reactive agents in diabetic patients, such as advanced glycation end products (AGE), reactive oxygen species (ROS) and proinflammatory cytokines (particularly TNF-α), have been linked to these complications. Besides prolonging the survival of osteoclasts and inducing cell death (apoptosis) of chondrocytes resulting in premature destruction of the cartilage callus, these agents also inhibit the differentiation of osteoblasts thereby hampering the subsequent mineralization and remodeling of callus [20]. Another co-morbidity of Charcot with a potential of negatively affecting bone healing on multiple levels is the mandatory neuropathy of the lower limbs [18, 21]. Neuropathic diabetic patients may suffer vitamin D deficiency causing impaired mineralization of the bone, have abnormal blood flow pattern in the lower extremities, higher incidence of atherosclerotic changes in peripheral arteries, and reduced numbers of periostal mechanoreceptors of importance for fracture union [18, 21]. In addition, recent studies have revealed that neuropeptides from autonomic and sensory neurons are able to influence bone formation and fracture healing by direct actions on osteoblastic and osteoclastic cells [21]. It is however the impaired sensitivity in the neuropathic foot which is considered to be of paramount importance for for the development of Charcot arthropathy [22] as the patients, unable to recognize the severity of ongoing pathology, continue to weight-bear on the diseased foot [18, 21], thereby complicating the already complex dynamics of inflammation and fracture healing. A significant and rapidly expanding knowledge base of signaling pathways and the complex cross-talk between these pathways has continuously added to our understanding of the role played by the innate and adaptive immune systems, enzymes and hormones in bone repair and remodeling [3, 23]. This has led to steadily increasing numbers of biological agents targeting specific components of the bone repair cascade with the aim of improving the outcome [24]. Nonetheless, our current understanding of the biological mechanisms regulating the pathology and recovery of Charcot arthropathy is almost entirely circumstantial in nature. In two recent studies to be discussed below, we presented evidence of the important role played by proinflammatory cytokines belonging to the Th17 subset of T helper cells [12] and by mediators of two major bone-regulating pathways [10] for bone healing in Charcot arthropathy patients. In the current study, we extended our investigation to the THh1/Th2 subset of proinflammatory cytokines.

A comparison of diabetic neuropathic control patients and with healthy individuals revealed no significant differences as for IL-6, IL-8 and IL-1β, whereas TNF-α was significantly higher in diabetic controls. Although previous studies have linked diabetes to high systemic levels of proinflammatory cytokines [25‐27], therapeutic regulation of hyperglycemia has proven a rather effective measure of normalizing cytokine levels [25, 26] which could suggest that the reports showing systemic proinflammatory cytokines in diabetic patients to range from lower to equal or higher than healthy individuals are linked to the duration and state of the patient diabetic disease at the time of sampling, although differences in assay type and sensitivity could have played in [28]. In two previous studies investigating monocytes from acute Charcot patients, results showed increased levels of proinflammatory immune phenotype [29] and increased numbers of CD-14 positive monocytes with a potential of transforming into osteoclasts [30]. They also showed slightly but significantly elevated serum IL-6, TNF-α and IL-1β in Charcot versus healthy [29] and higher TNF-α but not IL-1β in Charcot versus diabetic controls and healthy [30]. Our results showing all proinflammatory cytokines in Charcot patients at presentation to be below or at the level of diabetic controls are thus in opposition to the previous studies. We cannot offer a plausible explanation to these differences but emphasize that our study had a larger number of patients (66 in our study vs. 27 by Mabilleau and 26 by Uccioli) and used a significantly more sensitive technique to analyze biomarkers having with a dynamic range of 10⁶ as compared to 10³ for colorimetric ELISA [31, 32]. It may appear paradoxical that proinflammatory cytokines in Charcot patients at presentation are below or at the level of diabetic controls when considering that the acute Charcot foot presents itself with soft tissue, cartilage and bone pathology and the classical features of a local inflammatory reaction, i.e. a red, hot and swollen foot [5]. A plausible explanation could be that proinflammatory cytokines such as TNF-α and IL-1β, although important for the initiation of tissue inflammation following trauma [1, 3], only last a few days and constitute a small portion of the great basin of mediators (e.g. bradykinin, histamine, prostaglandins, substance P) involved in the local regulation of blood flow (flare and heat reaction) and vascular permeability (swelling) [33] with the ability to entertain the complex neuro-inflammatory cascade until optimal healing conditions are met. Another explanation for the low levels of proinflammatory cytokines at Charcot presentation is the counter-regulatory activation of anti-inflammatory cytokines, such as IL-4, IL-10 and IL-13, in order to down-regulate potentially harmful levels of proinflammatory cytokines [34, 35]. The latter mechanism is supported by the current results showing high levels of IL-1RA, the circulating natural antagonist of the potent proinflammatory cytokine IL-1β. IL-RA acts as a counter-regulatory positive feed-back loop with direct inhibitory effect on IL-1β thereby protecting the bone from excessive osteoclastic activity [36]. However, as we also showed high IL-1RA level in diabetic control patients, it could be inferred that its activity is linked to the diabetic disease per se as high inflammasome activity and excessive release of ROS, typical of diabetes [37], have been shown to induce high levels of IL-1RA [38]. Diabetes as the sole explanation is however contradicted by the IL-1RA/IL-1β ratio being significantly higher in Charcot patients due to lower IL-1β being lower than in diabetic controls. A high ratio at presentation and further increase during offloading suggest that IL-1β had reached a harmful level in the diseased foot thus prompting a counter-regulatory inhibition and a shift of the ratio in an anti-inflammatory direction. This inhibition of IL-1β may also explain the contained levels of IL-6 and TNF-α, as most proinflammatory cytokines which are present at high levels in the blood of diabetic patients are IL-1-driven and reduced by blocking its activity [39]. This inhibition appear however to extend beyond the Th1/Th2 subset of proinflammatory cytokines as we were recently able to show a similar containment of IL-17 cytokines belonging to the Th17 subset in Charcot patients exerting full weight-bearing at presentation [12].

Although minor movements in a fracture have been proven beneficial for bone healing, instability and excessive motion has been shown to cause predominantly catabolic activity, suppress ingrowth of new blood vessels, trigger excessive formation of cartilage and inhibit its subsequent replacement with bone thus hindering bone from bridging the fracture gap [3]. Stabilization and offloading of the Charcot foot in order to create a better bio-environment for recovery has therefore emerged as the most important strategy to prevent further deterioration and reverse the pathological processes [5]. Such a mechanism could explain the current results showing that IL-6 and TNF-α had begun to increase shortly after TCC treatment when unfavorable healing conditions caused by continued weight-bearing on the diseased foot had ceased. A similar increase by IL-17 family cytokines shortly after TCC in Charcot patients [12] lend further support to offloading being a critical factor which interrupts the vicious cycle caused by continued weight-bearing and sets the wheels of bone repair in motion. Although it may seem paradoxical that increased activity by inflammatory cytokines is beneficial to the healing process, there is ample evidence to show that a secondary or late phase of proinflammatory mediators is in fact mandatory for bone healing to proceed normally [3, 40]. Proinflammatory cytokines, such as TNF-α, IL-1β, IL-6 and IL-17, have been shown to play a pivotal role in the initiation of bone mineralization and remodeling as they, in addition to inducing osteoclast differentiation, also stimulate angiogenesis and promote osteoblast differentiation and when lacking cause fracture healing to be significantly delayed [1, 3]. The importance of offloading is further emphasized by recent results showing that TCC treatment coincided with a significant increase by dickkopf-1 (dkk-1) and Wnt ligand-1 (Wnt-1) [10], two mediators of the bone anabolic Wnt/β-catenin pathway implicated in bone remodeling [41, 42] and that an inter-regulatory link exists between proinflammatory cytokines and the Wnt system [43]. Our results showing that IL-8 did not change in the aftermath of offloading treatment could suggest that it had outplayed its most important role as chemotactic agent, i.e. to attract leukocytes, in the early post-injury phase. An interesting observation of the study was that although radiographs and MRI showed mineralized bone and no visible edema after 2 years, IL-1RA/IL-1β ratio remained high thus signifying a need for continued anti-inflammatory activity. A plausible reason for such a defensive measure in the chronic phase of Charcot could be that renewed weight-bearing on the Charcot foot, often characterized by bone deformities, will lead to unnatural high pressure on areas of the foot leading to exacerbated inflammation and microdamage [44]. The chronic phase being highly active from a biological point of view was further supported by the high bone remodeling activity of sclerostin and Wnt inhibitory factor-1 (Wif-1), mediators of the bone anabolic Wnt/β-catenin pathway, two years after Charcot presentation [10].

A limitation of the current study is the small sample size. We believe that the use of a high-sensitivity platform for analysis of biomarkers compensated this limitation by showing that the number of patients in the study was sufficient to achieve a power of 0.8 in an ANOVA comparison of three groups. The current protocol was further strengthened by repetitive measurements of biomarkers yielding a trajectory in support of the individual time points as well as by the well-defined groups of the study.

In conclusion, proinflammatory cytokines play a critical role in the early and late phase of bone fracture repair. Our results showing that proinflammatory cytokines were not above the level of diabetic controls at Charcot presentation, when patients exerted full weight-bearing on the diseased foot, and increased significantly after offloading, support TCC as a key factor setting off the secondary cascade of inflammatory cytokines responsible for the coupled bone remodeling phase. High IL-1RA/IL-1β ratio in the chronic phase of Charcot is probably a protective measure to reduce further tissue damage induced by renewed full weight-bearing on a deformed foot with impaired bone quality. Additional work is needed to more clearly define the complex cross-talk between pathways involved in the healing of the Charcot foot.