Keywords

1 Background

In patients with diabetes mellitus, it is estimated that up to 25% will develop a foot ulcer within their lifetime [1]. Of those with diabetic foot ulcers, up to 60% can go on to develop diabetic foot infection, which complicates ulcer treatment while increasing risk of developing osteomyelitis and amputation [2, 3]. There has been an increase in the incidence of osteomyelitis of the foot and ankle, partly due to the increase in predisposing conditions such as diabetes mellitus and peripheral vascular disease. The increase in availability and use of sensitive imaging tests , such as magnetic resonance imaging (MRI) and bone scintigraphy, has also improved its diagnostic accuracy. Osteomyelitis in the diabetic foot behaves very differently from osteomyelitis in the foot of patients without diabetes. The diabetic foot is different because three great pathologies come together in this disease process. The combination of neuropathy, ischemia, and immunopathy can present significant challenges. It makes the natural history of osteomyelitis in the diabetic foot rapidly progressive as well as chronic. The aim of this chapter is to give an overview of the pathophysiology, diagnosis and management.

2 Pathophysiology

Osteomyelitis is an infection of the bone that, when progressive, can cause destruction of the bone [4]. In patients with diabetic foot infections, osteomyelitis is generally caused by contiguous spread from infected skin and soft tissue in the setting of an ulceration, infected wound following surgery, or a contaminated open fracture (Fig. 34.1). Hematogenous spread of bacterial organisms resulting in osteomyelitis of the diabetic foot is relatively uncommon.

Fig. 34.1
figure 1

Pathophysiology of osteomyelitis in the diabetic foot

Full size image

In contiguous osteomyelitis , bacteria can gain access to bone by direct inoculation or by extension to bone from contaminated soft tissue that is adjacent to the bone [5]. Foreign bodies, trauma, deep pressure ulcers and ischemia may all cause osteomyelitis in the diabetic foot. Figure 34.2 depicts a patient who presented with a right heel diabetic foot ulcer. In patients with diabetes mellitus or peripheral vascular disease, osteomyelitis due to a contiguous focus of infection is often associated with vascular insufficiency and may go under-recognized in the setting of peripheral neuropathy [5]. Acute osteomyelitis , particularly if inadequately treated, may progress to a chronic infection. Sequestrum, or dead bone , is common in the setting of chronic osteomyelitis [4]. Patients with chronic osteomyelitis also have chronic bone loss and involucrum (reactive bony encasement of the sequestrum).

Fig. 34.2
figure 2

Right heel diabetic foot ulcer

Full size image

There are several components involved in the pathogenesis of osteomyelitis in the diabetic foot. Important factors to consider are the causative organism, location and vascular status of the bone and whether or not the host is immunocompromised [6]. The most common cause of osteomyelitis is Staphylococcus aureus. This organism is a cause of contiguous and hematogenous osteomyelitis and produces many cell-associated and extracellular virulence factors that promote bone destruction . This destruction is achieved through proteolytic activity, resistance to host defense mechanisms and by promoting bacterial adherence [6]. Other common bacteria include anaerobic bacteria, Gram-negative enteric organisms, and streptococci [7].

Since normal, healthy bone is highly resistant to infection, a large amount of bacterial burden is found in cases of osteomyelitis [8]. Adhesins are proteins that facilitate bacterial attachment to the bone and formation of biofilm , which is the layer that protects bacteria from antimicrobial agents [4, 6]. The immune response of the host can also result in bone destruction. Cytokines have osteolytic properties, and phagocytes also produce proteolytic enzymes and toxic oxygen radicals that can destroy host cells. This inflammation causes an increase in intraosseous pressure, which limits blood flow and can lead to necrosis of the bone. This necrosis is known as sequestrum , which is susceptible to attachment of biofilm [9]. Specifically, IL-1β[beta] is a proinflammatory cytokine that has a role in bone destruction in osteomyelitis [10]. Additionally, chronically poor blood flow as seen in peripheral arterial disease also makes it more difficult for antimicrobial agents to be effective [9].

Osteomyelitis is very common in the foot, and the risk of its development increases when patients have ulcers that are 2 cm or greater or have exposed bone or joint in the ulcer [11]. The forefoot is the most common site, with up to 90% of cases involving the weight bearing bones of the foot (first metatarsal head, calcaneum, and fifth metatarsal head). The midfoot and hindfoot comprise about 10% of infections [12].

Immunocompromised patients are at higher risk for osteomyelitis, likely related to the inadequate immunological response to infection at a more superficial level.

3 Histological Appearance

Histopathological signs that are seen in acute osteomyelitis include acute inflammatory cells with edema, small vessel thrombosis, and vascular congestion [6]. Initially, in early disease, the infection extends to the soft tissue surrounding the bone, and therefore the vascular supply to the bone is reduced [6]. If sequestrum is formed, surrounded by ischemic and necrotic tissue, bacteria can be difficult to eradicate by antimicrobial therapy alone. Acute and chronic osteomyelitis have similar histologic pictures [6]. Chronic osteomyelitis includes necrotic bone, the formation of new bone, and exudates with large numbers of lymphocytes, histiocytes, and some plasma cells. Osteomyelitis is also characterized by tissue necrosis . Granulation tissue at the surface of dead bone is absorbed and granulation tissue can completely destroy the bone and cause a cavity in the area [6]. Trabecular bone in localized osteomyelitis is usually absorbed, and often parts of the dead cortical bone are detached from the healthy residual bone to form sequestra.

In osteomyelitis, there is also new bone formation , which can be formed around the dead bone, though it may be of poor quality. This is known as involucrum . It is irregular and often osteoporotic, and may have areas of perforation where there may be purulence [6]. Even after removal of the sequestrum , a cavity may still be present which can fill with fibrous tissue and connect with the skin through a sinus tract.

4 Diagnostic Testing

Establishing a diagnosis of osteomyelitis in the diabetic foot can be difficult. Physicians should have a high clinical suspicion for underlying osteomyelitis in the setting of a chronic non-healing ulcer with poor vascular supply overlying a bony prominence [13]. There is no one test that aids in diagnosis, rather a combination of laboratory tests, imaging and bone biopsy with culture can be used in diagnosing osteomyelitis in the appropriate history and clinical examination.

4.1 Laboratory Testing

Among currently available laboratory tests , the erythrocyte sedimentation rate (ESR) appears to be most useful in diagnosing osteomyelitis [13,14,15]. In patients with suspected diabetic foot infections, an ESR ≥ 70 mm/h has a sensitivity of 83–89% and specificity of 77–100% in diagnosing osteomyelitis [14, 16]. Combining the use of tests like ulcer depth and C-reactive protein (CRP) or ESR has been shown to improve the sensitivity of the diagnosis of osteomyelitis to 100% [17]. The use of procalcitonin (PCT) level has also been studied in the diagnosis of osteomyelitis and one study reported a sensitivity and specificity of 0.81 and 0.71, respectively [18]. The role of PCT level is still unclear as another study found no statistical difference in levels to help distinguish osteomyelitis from soft tissue infection [19]. Another common test obtained is a serum white blood cell count ; however this is usually not helpful as it can be normal in almost half of patients with bone infections [15]. Blood glucose monitoring is important as hyperglycemia can increase risk of complications and mortality in patients with diabetes undergoing surgery [20]. One study found that among patients admitted with diabetic foot osteomyelitis, a predictive factor for amputation was perioperative hyperglycemia [21].

4.2 Probe to Bone Test

A probe-to-bone (PTB) test is a commonly used clinical test for diagnosing osteomyelitis, especially in the setting of diabetic foot ulcer and suspicion of infection. Given that the etiology of diabetic foot osteomyelitis is usually via contiguous spread from surrounding tissue, bacteria can easily access bone. Therefore, a PTB test suggests that if a probe can reach bone, bacteria can as well [22]. A sterile, blunt metal surgical probe is used when performing a PTB test and is considered positive if a hard surface is palpated with a grinding sensation when moving the probe over the surface. Although there have been prior small literature reviews of the PTB test, its applicability has been questioned, especially in settings with low pretest probability of osteomyelitis [23, 24]. A recent systematic review was performed to help delineate the diagnostic accuracy of the PTB test in detecting osteomyelitis in the diabetic foot. This review found that the pooled sensitivity and specificity for the PTB test was 87% and 83%, respectively [22]. In high-risk patients with a high pretest probability, the PTB test can support a diagnosis of diabetic foot osteomyelitis, while ruling out osteomyelitis in low-risk patients.

4.3 Bone Biopsy

A combination of microbiological culture and histopathological bone examination is considered the gold standard in the diagnosis of osteomyelitis [25,26,27]. Identifying responsible pathogens and antibiotic susceptibility testing of organisms cultured from surgically obtained bone is helpful in guiding antibiotic therapy . Superficial cultures are often not useful as they often grow numerous microorganisms that may not correlate with deeper bone culture [28]. Less than 50% concordance has been seen when comparing bone culture to superficial culture of a wound [29]. Culture of surgically obtained bone may not be performed in all patients. The use of an anticoagulant, severe ischemia, or very small bone involvement are some examples of possible clinical reasons why a surgical bone specimen may not be obtained. Receipt of antibiotics prior to obtaining a bone specimen for microbiological culture may decrease the ability to grow organisms from the specimen. Prior to obtaining a surgical bone specimen for microbiological culture, antibiotics should be held or discontinued, if feasible, to help maximize the culture yield [30, 31]. A bone biopsy may be less useful in patients who will undergo extensive debridement or amputation. However, when surgical resection and/or amputation is performed, the proximal margin should be obtained for culture and to assess for residual infected bone based on histopathological evaluation .

5 Imaging

5.1 Plain Radiography

Radiographic imaging is of high importance in establishing the diagnosis of osteomyelitis of the diabetic foot. Both the Infectious Diseases Society of America (IDSA) and the National Institute for Health and Clinical Excellence (NICE (UK), recommend plain radiographs as initial evaluation for all diabetic foot ulcers [26, 32]. Plain radiographs are helpful in identifying the presence of foreign bodies, arterial calcifications and bony deformities [33]. Some key characteristic features of osteomyelitis on plain radiography include periosteal elevation, bony erosion, marrow radiolucency and new bone formation, which is often surrounded by soft tissue swelling [13]. A significant loss of bone mineral content of 30–50% is necessary to produce visible changes on plain radiographs [34]. There has been a broad range of the sensitivity and specificity of radiography reported in the literature. One study among 27 diabetic patients with suspected foot infection reported a sensitivity and specificity of 22% and 94%, respectively in diagnosing osteomyelitis [35]. The NICE (UK) guidelines performed a significant literature review and found that plain radiographs had a sensitivity ranging from 22% to 75% and specificity of 17–94% among published studies [32]. Sequential imaging of the foot over time may be more likely to predict the presence of osteomyelitis than a single image. Poor specificity of radiographs is likely due to difficulty differentiating patients with bony destruction secondary to Charcot neuropathic osteoarthropathy [13]. Figure 34.3 shows a right foot X-ray in a diabetic patient with suspected osteomyelitis of the second toe. Table 34.1 illustrates the sensitivity and specificity of various imaging techniques in diagnosing osteomyelitis in diabetic foot infections from selected studies.

Fig. 34.3
figure 3

Right foot X-ray revealing destruction of the second toe middle phalanx (unbroken arrow) and possible cortical erosion of the distal third toe (broken arrow)

Full size image
Table 34.1 Sensitivity and specificity among various imaging modalities from selected studies in diagnosing osteomyelitis in diabetic foot infections
Full size table

Of the current imaging techniques available to aid in diagnosing osteomyelitis, magnetic resonance imaging (MRI) with gadolinium contrast is usually considered the most optimal. One study among 29 diabetic patients with suspected foot infections found that MRI was 100% sensitive and 63% specific in diagnosing osteomyelitis [36]. In a more recent meta-analysis evaluating MRI for diagnosing foot osteomyelitis in which the vast majority of patients were diabetic, the pooled sensitivity was reported as 77–100% and the pooled specificity ranged from 40–100% [33]. Some characteristic features of osteomyelitis on MRI include low focal signal intensity on T1-weighted images and high focal signal on T2-weighted images [13]. MRI scans are able to accurately outline the extent of inflammation and more precisely define the anatomic location in the foot, which is a significant advantage compared to radionuclide bone scans. Unfortunately, MRI is not feasible in all circumstances due to availability or patient contraindications. When MRI is unavailable, an alternative approach based on IDSA guidelines consists of combined radionuclide bone scan and a labeled white blood cell scan [26]. The IDSA does not recommend any other type of nuclear medicine imaging. Alternatively, the NICE (UK) guidelines recommend the use of a labeled white blood scan alone when MRI is unavailable or contraindicated and advises against the use of other nuclear medicine scans [32]. Figure 34.4 is a right foot MRI showing evidence of calcaneal osteomyelitis in a patient with an underlying diabetic foot ulcer.

Fig. 34.4
figure 4

Right foot MRI with contrast revealing loss of the normal dark T1 cortex with effacement of the normal T1 bright marrow fat along the posterior and lateral aspects of the calcaneus (arrow)

Full size image

5.2 Nuclear Medicine

There are several nuclear medicine techniques available for diagnosing diabetic foot infections. One such technique includes bone scans, which are commonly performed using 99mTc-methylene diphosphate and findings of abnormally increased intensity localized to bone are suggestive of osteomyelitis [23]. One study found that technetium bone scanning alone had sensitivity and specificity each of 50% [35]. A more recent study combined the use of labeled leukocyte scanning and technetium bone scans with a reported sensitivity and specificity of 91% and 67%, respectively [37]. Therefore, bone scans are more useful when negative as it can reliably rule out osteomyelitis; however there appears to be increased sensitivity and specificity when used in combination with a labeled leukocyte scan.

6 Management

There is a widely varied approach to the management of diabetic foot osteomyelitis. A combined surgical and medical approach is most frequently utilized, though in some cases surgery, often amputation, may be curative while select cases may be treated with medical therapy alone. Little data exists to help support clinical decision making with respect to the optimal route of antibiotic delivery or duration of therapy in either soft tissue infection or osteomyelitis in the diabetic foot [13, 26, 27, 38]. Initial antibiotic regimens usually consist of empiric, broad-spectrum parenteral therapy , especially in severe infections. Once microbiological data is available, the goal in most scenarios is to use the most narrow-spectrum antibiotics based on culture and sensitivity , and switch to oral therapy, when appropriate and feasible [26].

Antibiotic penetration to the site of infection in the diabetic foot is an important aspect of antibiotic selection. Beta-lactam antibiotics (penicillins, cephalosporins, and carbapenems) have been shown to penetrate bone at levels up to 20% of those in serum [39]. When given parenterally, these antibiotics reach high serum levels and therefore absolute bone levels likely surpass minimum inhibitory concentrations (MICs) of most organisms. Oral dosing of beta-lactam agents, however, is unlikely to achieve necessary bone concentration due to very low serum concentration [39].

Some antibiotics with high oral bioavailability have been shown to achieve adequate bone penetration. High oral bioavailability and bone concentrations at about 50% of serum has been reported when using fluoroquinolones, linezolid, and trimethoprim [13, 38,39,40,41]. Clindamycin also reliably penetrates into bone and necrotic tissue [13, 42, 43]. An oral treatment option in anaerobic osteomyelitis is metronidazole, as this reaches similar concentrations in bone as in serum [39, 44].

Decisions regarding initial empiric antibiotic therapy are patient specific, depending on suspected organisms of concern and the severity of infection. Some examples of initial broad-spectrum empiric therapy include ertapenem, levofloxacin, ceftriaxone or ampicillin-sulbactam [26, 32, 39]. It is important to treat Gram-positive cocci, specifically streptococci and staphylococci, as these are the most common pathogens in diabetic foot infections, and in severe infections, to consider empiric therapy treating Gram-negative organisms [26, 32]. In certain individuals, such as those with a prior history of methicillin-resistant Staphylococcus aureus (MRSA) or where the local prevalence of MRSA is high, it is reasonable to initiate therapy active against MRSA. Some examples of agents with activity against MRSA include vancomycin, linezolid, and daptomycin [26, 39]. Antipseudomonal therapy must be considered in special circumstances but this is often unnecessary in many cases. In areas of high local prevalence or frequent exposure of the foot to water, should prompt consideration for antipseudomonal therapy [26]. Multidrug-resistant Gram-negative organisms , with resistance to beta-lactams and fluoroquinolones are a growing concern, and should be considered when selecting empiric therapy for a patient with risk factors for these organisms, including prior treatment with broad-spectrum antibiotics.

6.1 Duration of Treatment

The optimal duration of antimicrobial therapy in diabetic foot osteomyelitis is still unclear [27, 38, 39, 45]. In a systematic review, the mean antibiotic duration ranged from 6 to 28 days [45]. The degree of debridement or resection performed does affect the duration of treatment. Aggressive surgical debridement and proximal amputation to the site of osteomyelitis are usually considered sufficient to consider shortening treatment to 2–5 days [26]. When infected bone remains or surgical resection is not possible, treatment should consist of at least 4 weeks of targeted intravenous therapy or high bioavailability oral antibiotics with good bone penetration [13, 26, 39]. To date there are no tests proven to correlate with long-term resolution of osteomyelitis. The consensus guidelines concluded that the following were suggestive of a response: a decrease in inflammatory markers (especially the ESR); healing of any wound; resolution of superimposed soft tissue infection; and radiographic changes that suggest healing [26]. PCT levels are likely less helpful during long-term follow-up as one study found that PCT values return to near-normal within approximately 1 week [18].

7 Summary

A high clinical suspicion is necessary in diagnosing osteomyelitis in the diabetic foot, which can be supported by histopathology, microbiologic culture and radiographic imaging . Combining exam findings with results from imaging studies and inflammatory markers will increase the accuracy and reliability of diagnosing osteomyelitis. An individualized approach to treatment is necessary, preferably utilizing a multidisciplinary approach.