When the history and physical findings are suspicious for infection, it is helpful to order laboratory studies for further evaluation, including a complete blood count with differential and inflammatory markers such as erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP) . This data can be helpful in demonstrating a heightened inflammatory state and can be trended over time to gauge responsiveness to the selected treatment. However, these are nonspecific markers of inflammation which can occur in the setting of infection or as a natural response to trauma and surgery itself. In the absence of obvious signs of infection such as a draining sinus tract, an increased white blood cell (WBC) count combined with an increased ESR and CRP has shown 100% sensitivity but 24% specificity for detecting the presence of infection [43]. While an abnormal WBC count (outside the range of 4–10 × 10³ leucocytes/ul) indicates an ongoing acute inflammatory response in the body, an increased immature granulocyte percentage can be more indicative of bacterial infection in particular. ESR represents greater migration of erythrocytes in the blood and changes in fibrinogen in response to inflammation. When it is elevated above its normal baseline level of less than 20 mm/hr, it has been shown to be highly sensitive for osteomyelitis but much less sensitive for infection in general [44, 45]. Therefore, a single increased postoperative WBC count or ESR value has poor diagnostic and prognostic value by itself and should be clinically correlated with history and physical findings, as well as trends over time [46].

CRP is produced by the liver during times of acute inflammation and does not normally occur in the body. It is indicative of some acute inflammatory change at any increase above 0.6 mg/dl. Under normal circumstances, CRP levels can be expected to rise and equilibrate at 24–72 h after surgical intervention, followed by a gradual but consistent decrease with resolution to normal levels after 7 days. When correlated with other signs and symptoms, infection should be suspected when CRP continues to increase after this time frame or undergoes a “second rise” to levels of 100 mg/mL or above, after the initial decrease [47]. The location of surgical intervention can also have an impact on the postsurgical CRP value. For example, Neumaier showed that the average CRP levels at 2 days after open reduction and internal fixation (ORIF) of femoral fractures was 136 mg/L, compared to 46 mg/L at 2 days after ORIF of ankle fractures [48]. While the initial increase was proportional to the amount of injury and amount of subsequent dissection performed, levels which remained higher than 100 mg/L after 4 days were associated with infection in both groups [49]. Proper utilization of CRP as a correlation to postoperative infection has shown a sensitivity of 85–92% and specificity of 86–93% [48, 50]. In pediatric patients, Laporta et al. showed that CRP levels over 11 mg/DL at 48 h had a sensitivity of 87% and specificity of 89% for postoperative infection [51].

Other infectious markers , such as CD64, TNF-α, and IL-6, have been proposed and are being studied, but require further research before recommendations can be made regarding their diagnostic or prognostic clinical use [52–55]. Procalcitonin has shown some promise as a diagnostic marker for sepsis and severe sepsis in critically ill postsurgical ICU patients who are unable to communicate their history and symptoms [53]. However, there is not enough evidence to support its utilization in foot and ankle postoperative infections.

Other laboratory values such a metabolic panel, albumin, and pre-albumin, while not helpful in the diagnosis of infection, provide an overall look at the patient’s general health and nutritional status and, therefore, may be indirectly prognostic of a patient’s ability to ward off and recover from infection. Finally, studies such as blood lactate levels [56] are important prognostic markers in the management of sepsis.

Imaging techniques can be very helpful in evaluating infection. There are many imaging techniques available including plain radiographs, ultrasonography, various types of scintigraphy, CT scans, and magnetic resonance imaging (MRI) . The important thing to remember is that the imaging study alone is just an aid and should not be relied upon to make the diagnosis of infection. The physician must utilize a combination of imaging studies, laboratory values, and clinical evaluation to make an accurate diagnosis and devise a proper treatment plan.

At the first suspicion of infection, plain radiographs should be ordered to look for early signs of hardware failure, migration of hardware or prostheses, changes in bone density and joint space margins, or other alarm signs such as gas in the tissues, which can result from aggressive infectious agents including Clostridia species [57] (Fig. 5.3). Although their specificity and sensitivity are low during the acute phase of infection versus normal postoperative changes, they should be the first imaging study obtained to evaluate possible infection. They can act as a baseline for comparison and are useful for serial evaluations of changes over time. The classic radiographic findings of osteomyelitis are well known and include soft tissue swelling, periosteal reaction, cortical erosion, and osteolysis and destruction (Fig. 5.4). The radiographic manifestations of osteomyelitis, however, appear after the destruction of bone and may take up to 4 weeks for the osseous changes to actually be visualized on plain films. Because of this delay, plain film radiographs should not be relied upon for the diagnosis of osteomyelitis [58].

Ultrasonography is a viable option for the detection of fluid collections including abscesses and joint effusions as small as 1–2 mL and is particularly useful for guiding procedures in real time, such as needle aspiration of a fluid collection for fluid culture sampling [59]. Limitations include inability to assess bone stock as sound waves are not able to penetrate the hard outer cortical layer [60, 61].

Advanced imaging selections should be based on the type and location of the suspected infection and other related factors. For example, computed tomography (CT) provides some detail about bone quality and integrity, by detecting erosions, sclerosis, and other bony abnormalities [62]. To investigate suspected septic arthritis, however, MRI should be selected as it is superior in both sensitivity (100%) and specificity (75%) [63] and can better evaluate septic arthritis by analyzing soft tissue in addition to pertinent osseous structures. Imaging features which may suggest septic arthritis include evidence of joint effusion, chondral destruction, adjacent decreased bone density, soft tissue inflammation, or abscesses.

These advanced imaging techniques are very useful for evaluating inflammatory changes when there is no hardware; but when hardware is present, artifact effects significantly reduce their utility [64]. For suspected prosthetic joint infections, advanced imaging is not recommended due to lack of visualization, low specificity for infection, and high cost. An exception is titanium hardware, which permits better visualization of soft tissue in the same viewing field when evaluated by MRI, although some artifact effects will still be present [65] (Fig. 5.5).

Scintigraphy or bone scans are very common imaging modalities utilized in evaluating infections, but they are not reliable during the acute phase of the postoperative course. There are many different scintigraphy modalities available. The technetium (^99mTc MDP) study is the most basic and common bone scan but is the least specific for infection. In an attempt to increase the specificity for infection, white blood cell-labeled bone scans of various kinds have been developed, but still have limitations. They are all markers of metabolic activity so are very sensitive to any inflammatory process, including infection and normal postoperative healing. Although bone scans are sensitive, they are nonspecific for infection and should have limited use postoperatively since they can both produce false-positive results and misguide the physician’s treatment plan (Fig. 5.6). Palestro et al. compared the sensitivity and specificity of three different types of bone scans for their accuracy in diagnosing osteomyelitis. The ^99mTc MDP scan had a specificity of only 27%, while the indium-labeled white blood cell scan (¹¹¹In-WBC) had the same specificity as the monoclonal antigranulocyte antibody (Moab) in vivo white blood cell-labeled scan, but was still only 67% specific [66]. Although WBC-labeled bone scans are more specific than technetium scans, their use in the acute postoperative phase is still limited, and they are more useful in detecting chronic or late-presenting infections. Recent evidence has suggested that bone scans combining ¹¹¹In-WBC and ^99mTc-sulfur colloid single-photon emission computed tomography (SPECT) could diagnose prosthetic joint infections with a reported accuracy of 95–97%; however, further testing is needed to determine how to utilize this modality in a cost-effective way [67].

The other drawback of bone scans in evaluating infections is the inability to readily distinguish structures individually. Results are often generalized to a large relative area around the infection, due to superimposed focal uptake in all tissue layers such as bone, tissue, and joint areas. Anatomic structures are poorly visualized making evaluation difficult (Fig. 5.7).

MRI has been described as the imaging modality of choice for the evaluation of postoperative infection. It offers superior visualization of anatomic structures and can differentiate between soft tissue and osseous infection (Fig. 5.8). It can also be utilized for preoperative planning to determine the extent of infection and the level of debridement needed. Additionally, it can be utilized to look for other sources of infection, such as a deep abscess, if the patient is not responding adequately to the treatment protocol. Finally, it can also be utilized to monitor an infection and the response to treatment (Fig. 5.9). A meta-analysis was performed over a 40-year time frame comparing MRI with plain radiographs, bone scans, and white blood cell-labeled studies for diagnosing osteomyelitis. They concluded that MRI was a strong test to aid in both confirming and excluding osteomyelitis of the foot and that a positive MRI results in an 84% chance of having the diagnosis of osteomyelitis and other exam findings such as substantial wound depth and probing to bone cinch the diagnosis. They also stated that the use of technetium bone scanning in the diagnosis of osteomyelitis of the foot should be limited and the lack of adequate specificity creates many false-positive results [68] (Fig. 5.10). Another study added the probe to bone test and compared it with radiographs, bone scans, white blood cell-labeled scans, and MRI for the diagnosis of osteomyelitis. The probe to bone test was actually more specific than any imaging modality, and MRI was the second most specific test. Radiographs and white blood cell-labeled scans were equal in specificity, and bone scans were the least specific [69].

Microbiology studies such as gram stains, cultures, and sensitivities are an exceptionally important part of the diagnostic process and should be used to guide antibiotic selections whenever possible. Ideally, cultures should be taken prior to administration of antibiotics to reduce the risk of a false-negative result. However, in life-threatening situations such as suspected sepsis, or when cultures cannot be taken in a timely fashion, antibiotic treatment should not be delayed for the sake of acquiring more accurate cultures. In such cases, treatment should be administered, with cultures at the earliest opportunity.

When obtaining wound cultures , it is important to understand that superficial ulcerations and tissues are poor specimens which are contaminated with normal skin flora and are not recommended for attempting to identify pathogens responsible for deeper infections [70]. Whenever possible, deep wound cultures should be taken as they are more reliable for the accurate detection of offending pathogens [71]. If possible, augmentation of deep cultures with both purulent fluid cultures and deep tissue specimens can enhance the accuracy and precision of the resulting pathogenic profile [72]. Prior to deep culture acquisition, contamination should be prevented by reducing the superficial bacterial load overlying the infection site, via sterile water or normal saline lavage, debridement, and scrubbing with an iodine or chlorhexidine alcohol-based cleanser.

When bone infection is suspected, bone biopsy and cultures are predictive of responsiveness to targeted antibiotics [70] and should always be obtained when accessible or exposed through surgical site dehiscence. If a bone culture is not accessible, consider obtaining deep sinus tract cultures, as this has been shown to correlate well with potential osseous infections when properly executed and when the swab probes to bone [71].

When bacteremia or sepsis is suspected, due to abnormal vital signs or the presence of systemic symptoms such as nausea or vomiting and chills, blood cultures should be obtained as quickly as possible [64, 73]. March et al. have shown that a full antibacterial susceptibility panel can be produced within 120 min when taken from positive blood culture samples [74]. Blood cultures should be procured from two separate venous sites with 20 min in between, to avoid false-negative results, as patients with even severe sepsis can demonstrate rare or discontinuous bacteria in circulation at any one time.

Ideally, multiple culture samples should be taken from any tissues with suspected infection and with two or more cultures positive for the same species being highly indicative that infection is present and the identified pathogen is responsible [57]. Depending on the type of suspected infection, this may include sterilely acquired aspirate of any fluid collection, direct fluid sampling from a sinus tract to its terminal recess, tissue specimens from surgical debridement, and blood cultures. When hardware infection is suspected, it is recommended that five to six cultures be taken to adequately represent the infected area about the hardware.

The most common organisms responsible for postoperative infections are gram-positive agents including Staphylococci and Streptococci species, followed by the gram-negative organisms such as Escherichia coli. Postsurgical tissue infected with colonized Staphylococcus becomes intensely erythematous, edematous, and tender. Abscesses may then form consisting of either a thick, creamy, yellow purulence commonly seen with S. aureus or a white-colored purulence if S. epidermidis is present. The gas-forming gram-positive species, Clostridium perfringens, is a less common but potentially devastating organism which can cause gas gangrene or necrotizing fasciitis and should therefore be treated with extreme prudence. In rare cases, necrotizing fasciitis can also be caused by particularly virulent strains of Streptococcus pyogenes or Staphylococcus aureus (Fig. 5.11). Gram-positive cocci remain the most common bacteria associated with postoperative osteomyelitis [75]. S. aureus, S. epidermidis, and Streptococcal species account for up to 73% of the osteomyelitis following musculoskeletal surgery [76].

When prostheses or other hardware is involved, biofilms can present other unique challenges. Biofilms are formed when bacteria colonize implanted synthetic material, bind host fibrinogen and fibronectin, and produce barriers of protein complexes. These biofilm barriers impair both antibiotics and components of the host immune system and therefore complicate management of the infection [77, 78]. In prosthetic joint infections, the most common biofilm-producing agents include staphylococcal species such as S. aureus and S. epidermidis [79.

Arthrocentesis is a critically important diagnostic procedure which is used to analyze and culture the joint space fluid in cases of suspected septic arthritis [80, 81]. Many physicians are hesitant to proceed with arthrocentesis due to the possibility of iatrogenic contamination or “seeding the joint,” but the evidence shows that this is not a common occurrence. In a retrospective study by Geirsson et al., incidence of iatrogenic infection during joint fluid aspiration was only 0.037% [82]. Relative contraindications include anticoagulation or overlying infection, but this should be weighed against the indication of the procedure for diagnosing septic arthritis. The aspiration should be done in an aseptic manner and may be facilitated by ultrasound guidance. The collected fluid sample should be sent for gram stain, cytology, crystal analysis under polarized light, and culture with sensitivities. The Infectious Diseases Society of America recommends that antibiotics be stopped 2 weeks before the recovery of joint fluid whenever possible, in order to maximize the chances of a positive culture [40]. In cases where this is not feasible, Von Essen et al. showed that blood agar bottles can be utilized in place of traditional agar plates to grow aspirated samples with high dependability. This method was able to more accurately isolate organisms in patients receiving antibiotic therapy, compared to samples which had previously yielded negative results using conventional methods [83].

A typical septic joint will have a turbid appearance and a WBC count >50,000 mm⁻³. It does however need to be correlated with other elements of the patient’s presentation. Van der Bruggen et al. showed that a synovial fluid sample containing a WBC >9000/μl, in combination with an elevated ESR and CRP, had a 100% positive predictive value and 98% accuracy for detecting prosthetic joint infection [67]. Recently, molecular testing has become more commonplace in the diagnosis of acute infection. A wide array of PCR techniques have been developed for rapid diagnosis of pathological presence. However, simply having the DNA of pathologies in a sample can still result from contamination. At this time, PCR is recommended for those cases when infection is highly suspected but cultures have produced a negative result. The PCR should be calibrated to include staphylococcal and streptococcal species as well as any other common organisms found in a regional antibiogram.

A different profile of pathogens should be considered when septic arthritis is suspected. While MSSA is still the most common causative organism present in over one-third of reported cases, and up to 70% of pediatric cases [39, 84–86], there has been a recent increase in the proportion of cases caused by MRSA, such that it is now attributed to approximately 25% of all cases [39, 84, 86]. Other causative agents in order of frequency include Streptococcus species and gram-negative cocci such as Escherichia coli, Proteus mirabilis, Klebsiella, and Enterobacter, which are involved in about 20% of septic arthritis cases [64]. Additionally, anaerobic and biofilm-forming microorganisms should be considered when hardware or prostheses are present [87].

Surgical site infections are classified by the level of tissue that they affect and include superficial incisional, deep incisional, and organ/space infections [38]. Superficial incisional infections involve only the skin and subcutaneous structures and are diagnosed within 30 days of the initial surgical intervention by the presence of any one of the following criteria: (1) purulent drainage from the surgical site; (2) microorganisms aseptically obtained from the superficial incisional site; (3) clinical findings of erythema, pain, or edema at the incision site and a positive culture after intentional opening of the incision site by a surgeon; and (4) clinical diagnosis by a surgeon or physician based on experience.

Deep incisional infections involve deeper layers of the incision, such as muscle and facial layers, and present up to 1 year after surgical placement of implanted devices or hardware or within 30 days of most other types of surgery. The findings associated with this infection are the same as for superficial incisional infections, with addition of clinical signs of abscess formation, imaging results that suggest deep involvement, or spontaneous dehiscence of the incision.

Organ/space infections involve deep intra-compartmental tissues and structures which were surgically manipulated at the time of the intervention. In the realm of foot and ankle surgery, some examples include joint infections, osteomyelitis, and orthopedic hardware or prosthesis infections. Like deep incisional infections, organ/space infections also present up to 1 year after surgical placement of implanted devices or hardware, or within 30 days of most other types of surgery [38].