**5.3 Cellulitis**

104 12 Chapters on Nuclear Medicine

imaging results for urgent management decisions. Gallium not only enhances the specificity of the diagnosis but provides information about surrounding soft tissue infection (Palestro & Torres, 1997). If gallium scan is negative, it effectively excludes the diagnosis of osteomyelitis (Horwich, 2011; as cited in Pineda, 2006). A gallium scan can be performed concurrently with a technetium labelled three-phase bone scan, and the information gathered may be more useful than that of either examination alone (Horwich, 2011; as cited in Tumed, 1986). Both radionuclides can be injected at the same time and the scintigraphic images can be obtained three to four hours after injection, while gallium images will be obtained up to 24 hours later (Horwich, 2011). This combination is probably the best nuclear medicine tool for the

Labelled leukocyte imaging is a good alternative in the evaluation of osteomyleitis, but is of little value in vertebral osteomyelitis because this entity often presents as a non specific photopenic defect (Gotthard et al., 2010; as cited in Van Der Bruggen et al., 2010). But, in the diabetic foot diagnosis, labelled leukocyte imaging alone is sufficient to determine the presence of osteomyelitis in the forefoot. In the midfoot and hindfoot it may be necessary to combine leukocyte scintigraphy with others radiotracers to precisely localize the infection (Palestro & Torres, 1997). The combined imaging approach of 99mTc-colloid bone marrow/labelled-leukocyte scanning enhances the sensitivity and specificity above 90%, avoiding the problem of physiologic uptake into bone marrow. Because in osteomyelitis bone marrow is replaced by the infectious process, bone marrow imaging will be negative whereas leukocyte scanning in the same location will be positive (Gotthardt et al., 2010, as

In chronic osteomyelitis the specificity of Tc-99m-MDP bone scans is very low even with active exacerbation because positive uptake also occurs with the healing process (El-Maghraby et al., 2006). Other false negative results are possible in areas of relative ischemia, since radiotracer may not be adequately delivered to the target site (Horwich, 2011). 67Ga combined with Tc-99m-MDP allows identification of active chronic osteomyelitis. Discordance between 67Ga and Tc-99m-MDP with more intense 67Ga or different distribution is highly specific at 80-100% (El-Maghraby et al., 2006). 18F-FDG-PET is a promising modality for imaging musculoskeletal infection and might play an important role in the evaluation of chronic osteomyelitis and spinal infection (Strobel & Stumpe, 2007). The specificity for spinal infection drops only if patients underwent surgery less than 6 months before PET and if osteosynthetic material is present (Gotthardt et al., 2010; as cited in De Winter et al., 2003). However, because MRI may not be an option in patients with metallic implants in situ, PET currently is the most sensitive imaging modality in the evaluation of such patients (Gotthardt et al., 2010; as cited in De Winter et al., 2003). Furthermore, in cases of severe defenerative disk disease with oedema-like changes in the endplates and the adjacent discs, MRI can give false-positive results (Gotthardt et al., 2010; as cited in Palestro et al., 2006). Other tracers for diagnosing spinal osteomyelitis are also under investigation,

including radiolabelled antibiotics and antifungical tracers (Gemmel et al., 2006).

Septic arthritis is the infection of the synovial tissues. It often occurs as a result of hematogenous seeding and less often by direct inoculation as a result of trauma or surgery (Díaz & De haro, 2005). Although it may occur at any age, it is most common in children under 3 years. Over 90% of cases are mono-articular (El-Maghraby et al., 2006). The most commonly involved joints are the hips and the knees (Díaz & De haro, 2005). Despite MRI

evaluation of vertebral osteomyelitis (Palestro & Torres, 1997).

cited in Palestro et al., 2006).

**5.2 Septic arthritis** 

Cellulitis is a soft tissue infection. The scan shows intense uptake in the first two phases of the study with diffuse extra-osseous activity, while the final image is normal after 2-4 hours. The presence of other processes that stimulate osteoblastic reaction, as in cases of suspected osteomyelitis in areas with trauma, surgery or arthritis, complicates the interpretation of the scintigraphic image which is completely non specific. Combined studies with different radiotracers such as 67Ga, 111In-leukocytes or 99mTc-HM-PAO improve the sensitivity and specificity, although MRI remains as gold standard technique (Díaz & De Haro, 2005).

#### **5.4 Painful joint replacement**

Prosthetic joint replacement is a common procedure and most patients have excellent results, but 20% of them develop pain during the follow-up. It may be secondary to infection, aseptic loosening or heterotopic bone formation (El-Maghraby et al., 2006). Differentiate betweeen loosening and infection is often a difficult problem, especially because clinical signs and symptoms, laboratory tests and radiographies are insensitive, nonspecific, or both. Crosssectional imaging modalities are hampered by artifacts produced by the prosthetic devices. Radionuclide imaging is not affected by the presence of metallic hardware and is therefore useful for evaluating the painful prosthesis (Love et al., 2001). Negative scintigraphic study rules out septic or aseptic loosening. Both conditions, however, may show increased tracer uptake in bone scans, but the pattern and site of uptake may help to differentiate each other. In aseptic loosening, focal localized uptake at the tips is seen, whereas in the infection diffuse intense uptake will be seen in the three phases of bone scan. The sensitivity for bone scan in infection is relatively high, ranging from 70% to 100% but the specificity is variable ranging

Nuclear Medicine in Musculoskeletal Disorders: Clinical Approach 107

osteonecrosis. Glucocorticoid use and excessive alcohol intake are reported to be associated with more than 90% of the cases. Osteonecrosis usually occurs in the anterolateral femoral head, although it may also affect the femoral condyles, humeral heads, proximal tibia, vertebrae, and small bones of the hand and foot. Many patients have bilateral involvement at the time of diagnosis, including disease of the hips, knees, and shoulders. The most common presenting symptom of osteonecrosis is pain and patients may have eventually limitation on range of motion. A limp may be present late in the course of lower extremity disease. A small proportion of patients are asymptomatic. In these cases the diagnosis is usually incidental.

There is no pathognomonic feature of osteonecrosis. A clinical diagnosis is appropriately made in a symptomatic patient when imaging findings are compatible with this disease and other causes of pain and bony abnormalities are either unlikely or have been excluded by appropriate testing. The evaluation for suspected osteonecrosis should begin with plain film radiography, althought it can remain normal for months after symptoms of osteonecrosis begin. Features of osteonecrosis on plain film radiographs, radionuclide scans (**Fig. 3**), and MRI are helpful diagnostically and provide the basis for classification and staging systems. Early diagnosis of osteonecrosis is crucial: the earlier the stage of the lesion at the time of diagnosis, the better the prognosis. Clinically, early diagnosis and treatment of osteonecrosis might prevent unnecessary surgery (Pape et al., 2004). Therefore, early diagnosis and location of osteonecrosis have prognostic value and determine the therapeutic alternatives.

Fig. 3. Radionuclide bone scan of the pelvis in a 68-year-old man with hip pain. Bilateral central area of diminished uptake surrounded by a zone of increased uptake in the femoral

Currently, MRI is the technique of choice for the diagnosis of avascular osteonecrosis in the early stages. This technique has been proven to be a highly accurate method both for early diagnosis (changes can be seen early in the course of disease when other studies are negative) and for staging of the disease (Malizos et al., 2007). MRI is far more sensitive than plain radiographs or bone scanning, with an overall reported sensitivity of 91% (Jones, 2011; as cited in Chang et al., 1993). Nevertheless, 99mTc bone scintigraphy also plays an important role in the early diagnosis of avascular necrosis and whole body bone scan is useful in patients with suspected polytopic osteonecrosis. The characteristic distribution of the radiopharmaceutical

head consistent with avascular necrosis.

Asymptomatic involvement contralateral to a symptomatic site is frequently noted.

from 20% to 90% and the reported accuracy is around 50-70% (El-Maghraby et al., 2006; as cited in Rosenthall et al., 1982, Turpin & Lambert, 2001, Wilson, 2004). The combination of 67Ga-citrate and Tc-99m-MDP bone scans has better results, with accuracy around 70-80% (El-Maghraby et al., 2006). But lower results for this combination has also been reported because the uptake of both tracers can be found not only in infection, but also in the postoperative patient, in heterotopic bone formation, and loosening or inflammatory reaction to cement fixators (El-Maghraby et al., 2006; as cited in Turpin & Lambert, 2001).

Labelled leukocyte imaging is, at least theoretically, the ideal technique for diagnosing the infected prosthesis because in general, white cells do not accumulate at sites of increased bone mineral turnover in the absence of infection (Palestro, 1998). However, labelled-leukocytes yield false positive results due to reactive or displaced bone marrow as a result of surgery are present up to more than 24 months after implantation (El-Maghraby et al., 2006; as cited in Oswald et al., 1990). Labelled leukocytes accumulate not only in infection but in the bone marrow as well. This problem has been overcome by the addition of complementary bone marrow imaging, which is usually performed with Tc-99m sulfur colloid. Both, labelled leukocytes and sulfur colloid accumulate in the bone marrow, but only labelled leukocytes accumulate in infection. In contrast to the results reported for labelled leukocyte imaging alone, the results of combined leukocyte-marrow imaging of prosthetic joints have been uniformly excellent, with an accuracy of 90% or greater (Love et al., 2001; as cited in Palestro 1990, 1991) and has become the method of choice to evaluate surgical prostheses (El-Maghraby et al., 2006; as cited in Love et al., 2001; Turpin & Lambert, 2001). Although extremely accurate, leukocyte-marrow scintigraphy is hampered by significant limitations. The in vitro labelling process is labor intensive, is not always available, and requires direct contact with blood products. The need for marrow imaging adds to the complexity and cost of the study and is an additional inconvenience to patients, many of whom are elderly and debilitated (Love et al., 2001). In an effort to maintain the accuracy of the study while reducing or eliminating the disadvantages, several methods of labelling leukocytes in vivo have been investigated, but their role in prosthetic joint infection has not been established.

FDG-PET has been extensively investigated, the high-resolution tomographic images, availability of the agent, and rapid completion of the procedure are all desirable traits. Published results to date, however, are inconclusive in this setting (Love et al., 2009; as cited in Chacko et al., 2002; Joseph et al., 2001; Love et al., 2004; Manthey et al., 2002; Pill et al., 2006; Reinartz et al., 2005; Stumpe et al., 2004; Zhuang et al, 2001).
