Eur J Nucl Med Mol Imaging (2010) 37:1751–1759 DOI 10.1007/s00259-010-1431-3

ORIGINAL ARTICLE

Detection of low-grade prosthetic joint infections using 99m Tc-antigranulocyte SPECT/CT: initial clinical results Vera Graute & Markus Feist & Sebastian Lehner & Alexander Haug & Peter Ernst Müller & Peter Bartenstein & Marcus Hacker

Received: 12 January 2010 / Accepted: 24 February 2010 / Published online: 23 March 2010 # Springer-Verlag 2010

Abstract Purpose Low-grade joint infections are characterized by infiltration of granulocytes, which mediate aspects of inflammatory changes. We evaluated retrospectively the contribution of SPECT/CT as an addition to planar scintigraphy with 99mTc-labelled antigranulocyte antibodies for diagnosing and localizing low-grade joint infections. Methods Planar scintigraphy using 99mTc-labelled antigranulocyte BW 250/183 antibodies was performed in 31 patients with suspected joint infections at 5 min, 5 h and 24 h after injection, with additional SPECT/CT performed 6 h after injection. With reference to gold standard clinical data, we assessed the diagnostic sensitivity of scintigraphy alone and in conjunction with SPECT/CT. Results Joint infections were diagnosed clinically in 9 of the 31 patients (1 hip and 8 knee prostheses). Planar scintigraphy revealed 6 true-positives, 13 true-negatives, 9 false-positives and 3 false-negative results, indicating sensitivity, specificity, positive and negative predictive values of, respectively, 0.66, 0.60, 0.4 and 0.81. With the addition of SPECT images, corresponding sensitivity, specificity, positive and negative predictive values increased to 0.89, 0.45, 0.40 and 0.91. Implementation of fused SPECT/CT led to a further increase to 0.89, 0.73, 0.57 and 0.94. Conclusion Relative to planar scintigraphy, SPECT with and without CT substantially improved the utility of V. Graute : S. Lehner : A. Haug : P. Bartenstein : M. Hacker (*) Department of Nuclear Medicine, University of Munich, Marchioninistr. 15, 81377 Munich, Germany e-mail: [email protected] M. Feist : P. E. Müller Department of Orthopedic Medicine, University of Munich, Munich, Germany

imaging with 99mTc-labelled antigranulocyte antibodies for diagnosis and localization of suspected joint infections. Optimal accuracy was obtained through image fusion, which permitted anatomical allocation of foci of pathological tracer accumulation as well as providing information on the extent of the infection. This imaging method seems suited for selection of patients requiring surgical therapy. Keywords Antigranulocyte scintigraphy . SPECT/CT . Low-grade . Joint infection

Introduction Failure or complications of joint arthroplasty can arise due to joint loosening, induction of heterotopic ossification, fracture, luxation, osteolysis and joint infections. In particular, joint infections occur in 1–2% of the primary implants and 3–5% of revision implants [1]. Early diagnosis of joint infections is crucial for successful therapy and prevention of complications; whereas severe infections are easily detected, the diagnosis of low-grade joint infections is challenging when clinical and biological signs are subtle or absent [2]. Persistent postoperative pain within 2 years after surgery can be a sign of low-grade infection [3], but can also be due to early prosthetic loosening. Clinical signs of low-grade infection are absent or weak at best, while laboratory parameters, radiographic changes, preoperative aspiration and bacteriological culture, and frozen section histopathological examination of deep tissue specimens are all either unreliable or of controversial utility [4]. In general, diagnosis of infection is mainly based on laboratory tests such as C-reactive protein (CRP), erythrocyte sedimentation rate, peripheral leukocyte count, histological examination and cell cultures, as well as cell

1752

Eur J Nucl Med Mol Imaging (2010) 37:1751–1759

counting of the infected area [5]. Since these indicators are not joint specific, diagnosis is frequently supported by anatomical imaging procedures like conventional radiography, CT or MRI. However, these imaging methods frequently fail to differentiate aseptic mechanical loosening and periprosthetic infection [6, 7]. Nuclear medicine procedures can potentially provide more specific physiological information about joint infection. Available methods for the differential diagnosis of joint infection include bone scintigraphy, gallium scanning and white blood cell imaging, all of which have been used to evaluate suspected infection of prosthetic joints. Bone scintigraphy is sensitive for identifying the failed joint replacement, but is uninformative about the cause of failure [1]. Antigranulocyte scintigraphy with monoclonal antibodies or antibody fragments directly targets leukocyte antigens or receptors in vivo and thus reveals concentrations of granulocytes in the inflamed tissue surrounding the prosthesis. A recent meta-analysis of 13 publications on the scintigraphic diagnosis of prosthesis infection after total joint arthroplasty [2, 4, 8–19] reported a sensitivity of this method ranging between 57 and 100% and a specificity between 65 and 100% [20]. However, there are few reports of the diagnostic performance of antigranulocyte scintigraphy in low-grade infections following prosthetic joint replacement. A recent planar scintigraphy study in which granulocytes were detected with 99mTc-labelled monoclonal anti-NCA-90 Fab fragments reported a sensitivity of 100%, but specificity of only 67–69% in a series of patients with total knee or hip arthroplasties [4]. Single photon emission computed tomography (SPECT) techniques enable more exact localization of the radiotracer uptake than is afforded by planar imaging, resulting in greater specificity [7, 21]. Still better performance is obtained with the advent of integrated SPECT/CT, which enables the combination of physiological information from nuclear medicine studies and structural information from CT [7, 22]. The aim of the present study was to determine retrospectively the accuracy of 99mTc-antigranulocyte scintigraphy with additional SPECT/CT in the detection and localization of low-grade joint infections. We hypothesized that the accuracy of 99mTc-antigranulocyte scintigraphy alone should be enhanced through the addition of CT image fusion, due to the superior anatomical allocation of pathological tracer accumulation.

shoulder and 27 knee) and suspicion of low-grade joint infection, we performed 99mTc-antigranulocyte scintigraphy with additional SPECT/CT. We suspected low-grade joint infection when the patient complained of persistent or increasing pain in the area of the joint prosthesis after surgery, in the absence of clear clinical markers for acute local infection, such as joint swelling, hyperthermia or purulent wound. We verified scintigraphic findings by means of bacteriological cultures, histopathological findings, cell score in the joint aspirate, CRP, operative findings and follow-up. The clinical follow-up data were registered during regular visits to the outpatient Department of Orthopedic Medicine of our clinic, which included clinical assessment. The mean follow-up time was 6.4±4.7 months.

Materials and methods

Image analysis

Patient selection and follow-up

In the present retrospective study the images were interpreted by consensus of two experienced nuclear medicine physicians, who were blind to pathological, bacteriological and surgical data and the patient’s clinical follow-up. Visual

In 31 consecutive patients (18 women and 13 men, mean age: 68 years, range: 48–88 years) with prostheses (3 hip, 1

99m

Tc-antigranulocyte scintigraphy

Antigranulocyte scintigraphy was performed after intravenous injection of 727–877 MBq 99mTc-labelled monoclonal antibody BW 250/183 (Scintimun®, CIS Bio International, Gif-sur-Yvette Cedex, France). The labelling procedure was performed according to the instructions of the manufacturer. Whole-body planar scans were obtained at 5 min, 4–6 h and 23–25 h after injection using either the Siemens Symbia T or the e.cam (both Siemens Medical Systems, Hoffman Estates, IL, USA) dual-head, large field of view camera with high-resolution, low-energy collimators and acquisition speeds of 30 cm min−1, 12 cm min−1 and 5 cm min−1, respectively. Additional SPECT/CT imaging was performed on a dual-head multifunctional gamma camera equipped with a low-power X-ray system (Siemens Symbia T, Siemens Medical Solutions, Hoffman Estates, IL, USA) at 6 h after tracer injection. Emission data were recorded through 360° SPECT with high-resolution low-energy collimators (matrix size of 128 × 128, 6° angle steps, 20 s per frame), followed by a brief, low-energy transmission scan. Our patients were positioned on the camera bed with a hook-and-loop fastener about the pelvis and the feet to avoid changes in position during acquisition. Of the 31 patients, 14 additionally received mild analgesia to eliminate movements elicited by pain. In every single patient, quality control for movement was performed after image acquisition using the patient’s sinograms. Changes in position during acquisition of SPECT and CT could have been reoriented manually, but this eventuality did not arise.

Eur J Nucl Med Mol Imaging (2010) 37:1751–1759

analysis was performed at a dedicated workstation (Hermes hybrid viewer, Hermes Medical Solutions, Stockholm, Sweden) for inspection of the images for the venous blood pool (5 min) as well as antibody distribution and uptake (4– 6 h and 23–25 h). Uptake at 4–6 h and 23–25 h postinjection was graded according to a visual scale relative to uptake in the bone marrow of the contralateral iliac crest, with 0=no uptake, 1=uptake less than, 2=uptake equal to and 3=uptake greater than that in the reference region, as described previously [4]. For the final diagnosis of a present joint infection, tracer uptake was considered as positive when an increase occurred between the 6-h and the 24-h scans, e.g. progression from grade 1 to grade 2 or above. In the first phase of image analysis, planar scans of all patients were independently evaluated with consideration of changes in radiopharmaceutical uptake between 3–5 h and 23–25 h post-injection, as described above. In the second phase, SPECT data (without CT) were interpreted in addition to the planar scans side by side. In the third phase, SPECT/CT image fusion was added to planar scans and SPECT. We considered SPECT/CT to be contributory when it accurately localized the anatomical site of infection and, in particular, when it discriminated between bone and soft tissue involvement [23]. Clinical gold standard of joint infection Persistent postoperative pain after surgery was reported in every patient. The final diagnosis was verified for each suspected site by the clinical gold standard of joint infection. In particular, low-grade joint infection was assumed to be present when at least three of four of the criteria (laboratory, microbiology, cell score and followup) were positive. If culture results or haematological results were negative and surgery was not performed, verification was done by follow-up at 6.4±4.7 months, with repeated visits to the clinic. Patients without pathological findings, not requiring joint revision or antibiotic therapy, and with unsuspicious follow-up were considered to be uninfected.

Results A general overview of clinical and patient data is given in Table 1. In 9 of the 31 patients, joint infection (1 hip and 8 knee prostheses) was diagnosed according to the clinical gold standard. All patients with prosthetic joint infections presented abnormal laboratory data (elevated CRP). In five of the nine patients diagnosis was additionally verified by positive bacteriological cell aspiration, and in four patients a positive cell score was

1753

seen in the aspiration sample. Clinical follow-up confirmed the diagnosis of prosthetic joint infection in eight cases. Considering only the planar scans, we found 6 truepositives, 13 true-negatives, 9 false-positives and 3 falsenegative results, indicating sensitivity, specificity and positive and negative predictive values of 0.66, 0.60, 0.40 and 0.81, respectively (Table 2). Adding SPECT images, these scores increased to 0.89, 0.45, 0.40 and 0.91, and with further addition of SPECT/CT fusion images, there were further increases to 0.89, 0.73, 0.57 and 0.94. Thus, SPECT/CT showed an additive value as compared to SPECT alone, providing accurate anatomical allocation of positive lesions as well as precise information on the extent of the infection. Overall, SPECT/CT made a contribution to the final diagnosis in 7 of 31 patients (23%), whereas SPECT alone made a contribution in only 4 cases (13%). SPECT/CT correctly classified periprosthetic, intraprosthetic and corticomedullary uptake, allowing the correct diagnosis of joint infections. In these seven cases, the diagnosis based on planar scintigraphy was radically reevaluated by SPECT/CT, with planar scans giving falsepositive results in five of these cases. SPECT/CT allowed the exclusion of joint infection, leading to five true-negative cases, in whom the uptake was allocated to the proximal fibula (patient 2), the normal bone marrow (patients 9 and 16), an osseous tendinopathy (patient 10) and soft tissue (patient 31). In the two planar scans giving false-negative results (patients 15 and 18), SPECT/CT revealed slight uptake adjacent to the prostheses, leading to true-positive cases. In a single case (patient 6), planar scintigraphy and SPECT/CT gave a false-negative finding for suspicious knee prosthesis, despite positive culture for Staphylococcus aureus and elevated CRP and cell score. In another six cases, the positive SPECT/CT finding concurred with planar scans and SPECT alone.

Discussion To our knowledge, this is the first study evaluating the value of SPECT/CT using 99mTc-labelled monoclonal antibodies for the diagnosis of low-grade joint infections. While the addition of ordinary SPECT imaging markedly increased both sensitivity and negative predictive value relative to planar scintigraphy alone, image fusion with CT data gave still greater specificity and positive predictive value. Thus, the present study demonstrates the superiority of SPECT/CT with respect to accurate anatomical allocation of positive lesions as well as the precision of information on the extent of the infection. In contrast to SPECT alone, SPECT/CT was also able to distinguish between periprosthetic and corticomedullary tracer uptake,

1754

Eur J Nucl Med Mol Imaging (2010) 37:1751–1759

Table 1 Characteristics and clinical data of the 31 patients with suspected low-grade joint infections Pat. No.

Gender/ age (years)

Prostheses

Microbiology

Laboratory test

Cell score/ aspiration (/µl)

Follow-up

Final diagnosis

Verification

1

F/67

K

Negative

CRP ↑

Ø

Pos.

L, F

2

F/54

K

Negative

Ø

3

F/69

K

Negative

4

M/82

H

Negative

5 6

F/67 M/67

K K

7 8 9 10 11 12 13

F/70 M/72 F/48 M/60 M/71 F/52 F/61

K K H K K K K

Ø Coagulase-neg. Staphylococcus Negative Negative Negative Negative Negative Negative Negative

Normal values Normal values Normal values Ø CRP ↑

Surgery, low-grade infection No surgery, MRI

14

F/63

K

Coagulase-neg. Staphylococcus

15

M/68

K

16

F/70

K

Propionibacterium species Ø

17

M/80

H

18

M/63

K

Enterococcus faecalis Negative

19

M/64

K

20 21

M/65 M/67

22 23

CRP ↑ CRP ↑↑ CRP ↑ CRP ↑ Ø CRP ↑ Normal values CRP ↑, leukocytes ↑ CRP ↑

800

Neg.

Ø 25,810

Surgery, nickel allergy, aseptic loosening Surgery, aseptic loosening No surgery, MRI Ø

Neg. Pos.

Ø Ø 580 50 Ø Ø Ø

No surgery Surgery, arthrofibrosis Surgery, joint fracture No surgery Surgery, arthrofibrosis Surgery, arthrofibrosis No surgery

Neg. Neg. Neg. Neg. Neg. Neg. Neg.

Ø

Surgery, low-grade infection

Pos.

Mi, L, F

1,880

Surgery, infection

Pos.

Mi, L, C, F

4,770

Neg. Neg.

Mi, L, C

Normal values CRP ↑↑

Ø

No surgery

Neg.

Ø

Pos.

Mi, L, F

CRP ↑

1,800

Pos.

L, C, F

Negative

CRP ↑↑

Ø

Pos.

L, F

K K

Negative Negative

81 980

Neg. Neg.

F/54 F/71

K K

Ø 4,790

Surgery, infection Surgery, infection

Pos. Pos.

24

F/84

K

Negative Lactobacillus gasseri Negative

CRP ↑ Normal values CRP ↑ CRP ↑

Surgery, low-grade infection Surgery, low-grade infection Surgery, low-grade infection Surgery, loosening No surgery

25

M/82

K

Negative

26

F/69

K

Negative

27 28

F/55 F/69

K K

Negative Negative

29

F/88

S

Negative

30 31

M/84 F/67

K K

Negative Negative

Ø

Ø

Ø

Neg.

Normal values Normal values CRP ↑ CRP ↑

Ø

No surgery

Neg.

3,200

No surgery

Neg.

630 Ø

No surgery Surgery, arthrofibrosis

Neg. Neg.

Ø

Surgery, loosening

Neg.

1,210 610

Surgery, loosening Ø

Neg. Neg.

Normal values CRP ↑↑ Normal values

L, F Mi, L, C, F

M male, F female, K knee, H hip, S shoulder, Ø not done, Pos. positive septic loosening, Neg. negative septic loosening, L laboratory, Mi microbiology, F follow-up, C cell score Normal values: CRP S <0.5 mg/dl, leukocytes 4.0–11.0×109 /l; abnormal values: CRP S 0.5–2.0 mg/dl ↑, >2.0 mg/dl ↑↑, leukocytes 11–13×109 /l ↑, >13×109 /l ↑↑

Eur J Nucl Med Mol Imaging (2010) 37:1751–1759

1755

Table 2 Accuracy of 99mTc-antigranulocyte planar scintigraphy only and with additional SPECT and SPECT/CT for the whole study group (n= 31)

True-positive True-negative False-positive False-negative Sensitivity Specificity Positive predictive value Negative predictive value

Planar images only

With additional SPECT

With additional SPECT/CT

6 13 9 3 0.66 0.60 0.40 0.81

8 10 12 1 0.89 0.45 0.40 0.91

8 16 6 1 0.89 0.73 0.57 0.94

allowing comprehensive diagnosis of low-grade joint infections. While this is the first demonstration of the superiority of SPECT/CT over planar techniques or SPECT alone for the diagnosis of joint infections, analogous findings have been Fig. 1 Planar (a) and SPECT, SPECT/CT images (b) from a 61-year-old woman (patient 23) with total knee prosthesis on the left side, presented in ventral and dorsal orientation at 1 min, 6 h and 24 h for planar imaging and at 6 h after tracer injection for SPECT/CT. Planar imaging revealed tracer accumulation at the tibial component of the prosthesis (red arrows). SPECT/ CT images additionally demonstrated the extent of the infection in the cortical bone and lack of involvement of the bone marrow. The focal uptake of low intensity lying behind the prosthesis could be attributed to the arteria poplitea (green arrow). Haematology laboratory results were completely normal at the time of antigranulocyte scintigraphy (CRP 0.3 mg/dl, leukocytes 5.6×109/l). However, CRP had increased to 16.7 mg/dl 2 months later. The diagnosis of a low-grade infection was verified by needle aspiration of the left knee, from which Lactobacillus gasseri was isolated, such that the scintigraphic result was rated as true-positive. RVL right ventral left, LDR left dorsal right

reported in several other settings [7, 22, 24]. Particularly when accurate anatomical localization of the lesion is required for planning or guiding surgical treatment, supplementary anatomical data from CT has been desirable [23, 24].

1756

There are several previous reports on diagnosing prosthesis infection through radionuclide imaging with antigranulocyte antibodies [2, 4, 8–20]. These studies have likewise shown that imaging of inflammatory cells using antibodies or antibody fragments contributes to correct diagnosis and affords high diagnostic accuracy. However, these studies found that planar scintigraphy and SPECT alone for localizing infection had limitations arising from inadequate image resolution and the lack of anatomical landmarks. Horger et al. [24] studied the diagnostic advantages imparted by SPECT/CT with 99mTc-labelled antigranulocyte antibodies in 27 patients with 29 sites of suspected bone infections. Using BW 250/183, the same monoclonal antibody used for 99mTc-labelling as in the present study, they found that SPECT/CT enabled differentiation between soft tissue infection, septic arthritis and osteomyelitis, as well as between cortical, corticomedullary and subperiosteal foci, in accordance with the present findings in suspected low-grade joint infection. While Fig. 2 A 68-year-old man (patient 15) with a knee prosthesis on the left side. The planar images at 5 min, 6 h and 24 h after injection (a) showed a characteristic pattern indicating the presence of synovitis, such that the scan was rated negative. Although a slight increase of uptake between the early and late scans was registered, uptake could not be assigned to anything other than the synovia. With both SPECT and SPECT/ CT at 6 h after infection (b) this focal uptake could be allocated to the lateral part of the prosthesis, leading to revision as a positive result. Haematology laboratory results, needle aspiration of the left knee (Propionibacterium species) and cell scores were positive for infection. The patient was subsequently operated. RVL right ventral left, LDR left dorsal right

Eur J Nucl Med Mol Imaging (2010) 37:1751–1759

sensitivity was identical for SPECT alone and SPECT/CT, specificity was improved from 78 to 89%, in accordance with our findings, likewise with respect to the correct differentiation between soft tissue infection and bone involvement. In another study, Filippi and Schillaci [23] investigated the value of adding SPECT/CT to 99mTc-HMPAO-labelled leukocyte scintigraphy in 28 patients with suspected bone (n=15) and joint infections (n=13). They showed that SPECT/CT provided an accurate anatomical localization of all positive foci, such that prosthesis and soft tissue uptake could be distinguished in five of seven patients with a hip prosthesis. In two of six patients with knee implants SPECT/CT correctly localized leukocyte uptake in the synovium and excluded prosthesis involvement. However, the HMPAO labelling technique is relatively elaborate due to the requirement for blood handling, which increases the risk of infection compared to the present kit-based in vivo labelling techniques.

Eur J Nucl Med Mol Imaging (2010) 37:1751–1759

The diagnosis of acute and postoperative joint infections has been based on a range of diagnostic methods such as cell culture, blood examination, radiological examinations and clinical follow-up, which have been developed to confirm the final diagnosis. While acute and postoperative joint infections do not in most cases present major diagnostic difficulties, the clinical signs in low-grade infections are frequently weak or completely absent. Laboratory parameters, radiographic changes, preoperative aspiration and bacteriological culture, as well as histopathological examination of deep tissue specimens are not always reliable [25–27], of controversial importance [4] or are uninformative [2]. Consequently, the gold standard for the diagnosis of low-grade infections is subject to caveats related to the imperfect sensitivity and specificity of intraoperative tests [28–30]. As in previous studies, we have assumed a low-grade infection to be present in cases with elevated CRP levels, elevated cell score in the Fig. 3 Planar scans (a) of a 70year-old woman (patient 16) with right and left knee prostheses, suffering from persisting pain of the right knee, obtained 5 min, 6 h and 24 h after tracer injection and corresponding SPECT, SPECT/CT (b) images at 6 h after injection. All images were rated positive, but SPECT/ CT (b) additionally enabled the allocation of the uptake to the bone marrow, so that the final rating was negative (physiological uptake in the bone marrow, red arrows). Haematological parameters were normal, and she was consequently not operated. During a follow-up of 3.4 months, pathological findings did not emerge. RVL right ventral left, LDR left dorsal right

1757

aspiration, positive bacteriological findings from cell aspiration or surgery, or conversion of initially negative results during follow-up. Even if the combination of different reference standards is adequate to excluded serious misclassification, some errors may nevertheless be unavoidable, because of the low sensitivity of the different diagnostic methods [11, 31]. In this context, SPECT/CT may gain clinical relevance for the improved interpretation of histopathological findings; it is claimed that at least five negative intraoperative smear tests are required to avoid false-negative results and to distinguish between contamination and infection [31]. If the surgeon were aware of the SPECT/CT results, he or she could be guided to take the smear in the region of highest tracer uptake, plausibly resulting in increased sensitivity. Improved accuracy of histopathological findings would certainly lead to optimized therapy, e.g. an appropriate course of antibiotic treatment. In cases where surgical therapy is indicated, it is

1758

to be expected that the surgeon would gain additional information on the extent of the operation field, which may reduce operation time and postoperative complications. Our results indicate that SPECT/CT with 99mTclabelled antigranulocyte antibodies is superior to planar scintigraphy or SPECT alone for the diagnosis of patients with suspected low-grade joint infections. Image fusion provides accurate anatomical localization and information on the extent of the infection. At sites of physiological tracer uptake (e.g. vascular structures as in patient 23, see Fig. 1) SPECT/CT allowed the exclusion of disease. If prosthetic joint infection can indeed be excluded by SPECT/CT, further examination or even invasive therapies such as operation would be unnecessary. In cases of positive SPECT/CT findings, one can accurately discriminate bone from soft tissue involvement, which is clinically important because the appropriate therapies are different. Thus, SPECT/CT potentially impacts clinical management by helping to select and plan appropriate therapies (Figs. 2 and 3). SPECT/CT also provides information on the anatomical extent of the infection, which may likewise influence preoperative planning. In particular it has been reported in a previous study that SPECT/CT is especially useful in patients with recurrent osteomyelitis, and also those suffering from anatomical abnormality after trauma [23]. This imaging method may gain clinical acceptance in the selection of patients requiring surgical therapy and may also gain importance for guiding intraoperative histopathological examination. Acknowledgements The authors thank Dr. Paul Cumming for critical revisions of the manuscript.

References 1. Love C, Marwin SE, Palestro CJ. Nuclear medicine and the infected joint replacement. Semin Nucl Med 2009;39:66–78. doi:S00012998(08)00104-9 [pii] 10.1053/j.semnuclmed.2008.08.007. 2. Boubaker A, Delaloye AB, Blanc CH, Dutoit M, Leyvraz PF, Delaloye B. Immunoscintigraphy with antigranulocyte monoclonal antibodies for the diagnosis of septic loosening of hip prostheses. Eur J Nucl Med 1995;22:139–47. 3. Fitzgerald RH Jr, Nolan DR, Ilstrup DM, Van Scoy RE, Washington JA 2nd, Coventry MB. Deep wound sepsis following total hip arthroplasty. J Bone Joint Surg Am 1977;59:847–55. 4. Ivanćević V, Perka C, Hasart O, Sandrock D, Munz DL. Imaging of low-grade bone infection with a technetium-99m labelled monoclonal anti-NCA-90 Fab′ fragment in patients with previous joint surgery. Eur J Nucl Med Mol Imaging 2002;29:547–51. doi:10.1007/s00259-001-0744-7. 5. Spangehl MJ, Masri BA, O’Connell JX, Duncan CP. Prospective analysis of preoperative and intraoperative investigations for the diagnosis of infection at the sites of two hundred and two revision total hip arthroplasties. J Bone Joint Surg Am 1999;81:672–83.

Eur J Nucl Med Mol Imaging (2010) 37:1751–1759 6. Reinartz P. FDG-PET in patients with painful hip and knee arthroplasty: technical breakthrough or just more of the same. Q J Nucl Med Mol Imaging 2009;53:41–50. 7. Schillaci O, Danieli R, Manni C, Simonetti G. Is SPECT/CT with a hybrid camera useful to improve scintigraphic imaging interpretation? Nucl Med Commun 2004;25:705–10. doi:00006231200407000-00012 [pii]. 8. Iyengar KP, Vinjamuri S. Role of 99mTc Sulesomab in the diagnosis of prosthetic joint infections. Nucl Med Commun 2005;26:489–96. doi:00006231-200506000-00003 [pii]. 9. Kaisidis A, Megas P, Apostolopoulos D, Spiridonidis T, Koumoundourou D, Zouboulis P, et al. Diagnosis of septic loosening of hip prosthesis with LeukoScan. SPECT scan with 99mTc-labeled monoclonal antibodies. Orthopade 2005;34:462– 9. German. doi:10.1007/s00132-005-0777-x. 10. von Rothenburg T, Schoellhammer M, Schaffstein J, Koester O, Schmid G. Imaging of infected total arthroplasty with Tc-99mlabeled antigranulocyte antibody Fab′fragments. Clin Nucl Med 2004;29:548–51. doi:00003072-200409000-00006 [pii]. 11. Kordelle J, Klett R, Stahl U, Hossain H, Schleicher I, Haas H. Infection diagnosis after knee-TEP-implantation. Z Orthop Ihre Grenzgeb 2004;142:337–43. German. doi:10.1055/s-2004818772. 12. Rubello D, Casara D, Maran A, Avogaro A, Tiengo A, Muzzio PC. Role of anti-granulocyte Fab′ fragment antibody scintigraphy (LeukoScan) in evaluating bone infection: acquisition protocol, interpretation criteria and clinical results. Nucl Med Commun 2004;25:39–47. doi:00006231-20040100000006 [pii]. 13. Vicente AG, Almoguera M, Alonso JC, Heffernan AJ, Gomez A, Contreras PI, et al. Diagnosis of orthopedic infection in clinical practice using Tc-99m sulesomab (antigranulocyte monoclonal antibody fragment Fab′2). Clin Nucl Med 2004;29:781–5. doi:00003072-200412000-00001 [pii]. 14. Gratz S, Schipper ML, Dorner J, Höffken H, Becker W, Kaiser JW, et al. LeukoScan for imaging infection in different clinical settings: a retrospective evaluation and extended review of the literature. Clin Nucl Med 2003;28:267–76. doi:10.1097/01. RLU.0000057613.86093.73. 15. Ryan PJ. Leukoscan for orthopaedic imaging in clinical practice. Nucl Med Commun 2002;23:707–14. 16. Klett R, Khalisi A, Haas H, Puille M, Steiner D, Stürz H, et al. Staged diagnosis of septic hip endoprosthesis loosening with laboratory parameters and antigranulocyte scintigraphy. Z Orthop Ihre Grenzgeb 2001;139:415–9. German. 17. Devillers A, Garin E, Polard JL, Poirier JY, Arvieux C, Girault S, et al. Comparison of Tc-99m-labelled antileukocyte fragment Fab′ and Tc-99m-HMPAO leukocyte scintigraphy in the diagnosis of bone and joint infections: a prospective study. Nucl Med Commun 2000;21:747–53. 18. Sciuk J, Puskás C, Greitemann B, Schober O. White blood cell scintigraphy with monoclonal antibodies in the study of the infected endoprosthesis. Eur J Nucl Med 1992;19:497–502. 19. Wolf G, Aigner RM, Schwarz T, Lorbach MP. Localization and diagnosis of septic endoprosthesis infection by using 99mTcHMPAO labelled leucocytes. Nucl Med Commun 2003;24:23–8. doi:10.1097/01.mnm.0000051628.18733.cc. 20. Pakos EE, Trikalinos TA, Fotopoulos AD, Ioannidis JP. Prosthesis infection: diagnosis after total joint arthroplasty with antigranulocyte scintigraphy with 99mTc-labeled monoclonal antibodies—a meta-analysis. Radiology 2007;242:101–8. doi:2421052011 [pii] 10.1148/radiol.2421052011. 21. Weon YC, Yang SO, Choi YY, Shin JW, Ryu JS, Shin MJ, et al. Use of Tc-99m HMPAO leukocyte scans to evaluate bone infection: incremental value of additional SPECT images. Clin Nucl Med 2000;25:519–26.

Eur J Nucl Med Mol Imaging (2010) 37:1751–1759 22. Townsend DW, Cherry SR. Combining anatomy and function: the path to true image fusion. Eur Radiol 2001;11:1968–74. doi:10.1007/s003300101007. 23. Filippi L, Schillaci O. Usefulness of hybrid SPECT/CT in 99mTcHMPAO-labeled leukocyte scintigraphy for bone and joint infections. J Nucl Med 2006;47:1908–13. doi:47/12/1908 [pii]. 24. Horger M, Eschmann SM, Pfannenberg C, Storek D, Dammann F, Vonthein R, et al. The value of SPET/CT in chronic osteomyelitis. Eur J Nucl Med Mol Imaging 2003;30:1665–73. 25. Patel R, Osmon DR, Hanssen AD. The diagnosis of prosthetic joint infection: current techniques and emerging technologies. Clin Orthop Relat Res 2005;437:55–8. 26. Sadiq S, Wootton JR, Morris CA, Northmore-Ball MD. Application of core biopsy in revision arthroplasty for deep infection. J Arthroplasty 2005;20:196–201.

1759 27. Savarino L, Baldini N, Tarabusi C, Pellacani A, Giunti A. Diagnosis of infection after total hip replacement. J Biomed Mater Res B Appl Biomater 2004;70:139–45. doi:10.1002/jbm.b.30030. 28. Fitzgerald RH Jr. Total hip arthroplasty sepsis. Prevention and diagnosis. Orthop Clin North Am 1992;23:259–64. 29. Garvin KL, Hanssen AD. Infection after total hip arthroplasty. Past, present, and future. J Bone Joint Surg Am 1995;77:1576–88. 30. Mirra JM, Amstutz HC, Matos M, Gold R. The pathology of the joint tissues and its clinical relevance in prosthesis failure. Clin Orthop Relat Res 1976;117:221–40. 31. Atkins BL, Athanasou N, Deeks JJ, Crook DW, Simpson H, Peto TE, et al. Prospective evaluation of criteria for microbiological diagnosis of prosthetic-joint infection at revision arthroplasty. The OSIRIS Collaborative Study Group. J Clin Microbiol 1998;36:2932–9.