Diagnosis of orthopaedic prosthesis infections with radionuclide techniques; clinical application of various imaging methods

Martti Larikka

Laboratory, Länsi-Pohja Central Hospital
Department of Clinical Chemistry, University of Oulu
Department of Surgery, University of Oulu
Department of Infection Control, Oulu University Hospital

Abstract

A variety of radiopharmaceuticals and imaging techniques are currently available for scintigraphic imaging of infections. However, comparisons on the clinical value of such techniques have been limited, especially in prosthesis infections.

This series included 138 cases with suspected prosthesis infections – 94 in hip and 44 in knee prostheses of patients whose final diagnoses were based on clinical, operative and microbiological findings, and who underwent three-phase bone, ^99mTc -leukocyte and ^99mTc-ciprofloxacin imaging in the Department of Clinical Chemistry, Oulu University Hospital and in the Laboratory, Länsi-Pohja Central Hospital, during the years from 1993 to 2001.

The normal arterial and soft-tissue phase images of three-phase bone imaging practically excluded infection in hip prostheses, whereas these techniques frequently yielded false positive findings in patients with knee prostheses, resulting in specificity of 23% or less. In combined ^99mTc-leukocyte/bone imaging, diagnostic accuracy was 80–86% at two- to four-hour images and 87–98% at 24-hour images. The ^99mTc-ciprofloxacin images showed unspecific accumulation of tracer in the one-hour and four-hour images, which disappeared in the 24-hour images in most hip and knee prostheses. ^99mTc-ciprofloxacin imaging yielded almost as good diagnostic accuracy as combined ^99mTc-leukocyte/bone imaging.

In conclusion, in suspected hip prosthesis infections, normal findings in three-phase bone imaging exclude infection, whereas abnormal results in the arterial and soft-tissue phases should be confirmed with ^99mTc-leukocyte imaging using 24-hour images. Contrary-wise, in suspected knee prosthesis infections, ^99mTc-leukocyte imaging with 24-hour images is the first-line examination, and abnormal results in 24-hour images should be confirmed by using ^99mTc-bone-metabolic imaging. ^99mTc-ciprofloxacin yielded almost equally good results as ^99mTc-leukocyte/bone-metabolic imaging, but unfortunately, the tracer is not commercially available, although it has been patented.

Table of Contents

Acknowledgements

Abbreviations

List of original publications

1. Introduction

2. Review of the literature

2.1. History of prosthesis surgery

2.2. Indications and clinical statistics of hip and knee prosthesis surgery, the increasing need

2.3. Complications of prosthesis surgery

2.3.1. Pathophysiology involved in the loosening of prostheses
2.3.2. Incidence of prosthesis infection and contamination with bacteria
2.3.3. Risk factors of joint prosthesis infections
2.3.4. Countermeasures against prosthesis infections related to operations
2.3.5. Epidemiology of prosthesis contamination and infection
2.3.6. Pathogenesis of prosthesis infection, biofilm formation

2.4. Classification of prosthesis infections

2.5. Microbiological diagnosis of prosthesis infections

2.6. Non-invasive diagnosis of prosthesis infections

2.7. Nuclear medicine procedures for imaging infection and inflammation

2.7.1. Non-specific infection and inflammation imaging
2.7.2. Specific infection and inflammation imaging

2.8. Examples of studies utilising nuclear medicine procedures in the diagnosis of orthopaedic infections

3. Aims of the study

4. Patients and methods

4.1. Patients

4.1.1. Patient series
4.1.2. Diagnostic criteria of infections

4.2. Methods

4.2.1. Bone-imaging radiopharmaceutical
4.2.2. Leukocyte labelling
4.2.3. Preparation of ^99mTc-ciprofloxacin
4.2.4. Imaging procedures
4.2.5. Interpretation of images
4.2.6. Statistics

5.1. Verified infections
5.2. Hip prostheses
5.3. Knee prostheses
5.4. Total series
5.5. Figures

6. Discussion

6.1. Patient series, incidence of infections and verification of final diagnoses
6.2. Three-phase bone imaging
6.3. ^99mTc-leukocyte imaging
6.4. ^99mTc-ciprofloxacin imaging
6.5. Comparisons and recommendations

7. Summary and conclusions

List of Tables
1. Main uptake mechanisms of radiopharmaceuticals used to image infection and inflammation.
2. Examples of studies utilising nuclear medicine procedures in the diagnosis of orthopaedic infections.
3. Main patient characteristics in the studies 1–4.
4. Main patient characteristics in the studies 1–4
5. Efficacies of the different methods for identifying hip prosthesis infections. Results derived from the original studies I (n=64) and IV (n=30).
6. Efficacies of the different methods for identifying knee prosthesis infections. Results derived from the original studies II (n=30) and III (n=16).
7. Total efficacies of ^99mTc-ciprofloxacin imaging (studies III and IV, n=46) and ^99mTc-leukocyte/bone-metabolic imaging (studies I and II, n=94) in hip and knee prosthesis infections.

List of Figures
1. A 76-year-old male patient with an infected (9 months old) cemented prosthesis of the left hip. First row: arterial and soft-tissue phase images from a positive bone scan. The first three images demonstrate the arterial phase, while the fourth image demonstrates the soft-tissue phase. Second row: left, static image of the metabolic phase of the bone scan; centre, 4-hour leukocyte image; right 24-hour leukocyte image. The images demonstrate diffusely increased metabolic activity around the femoral shaft of the hip prosthesis and focal uptake of leukocytes at the prosthesis shaft, which becomes clearly visible over time. Bacterial cultures taken at revision operation 4 months after the imaging indicated the growth of Staphylococcus aureus and Enterococcus faecalis.
2. Upper row: a 71-year-old woman with post-traumatic arthrosis and a left knee prosthesis operation 1 year and 1 month before the imaging. She had had pain in her left knee after the operation, which continued until the imaging. ^99mTc-ciprofloxacin images (from left to right) at 1, 4 and 24 hours after injection demonstrate fading accumulation of the tracer. No infection could be verified, and the knee became symptomless during follow-up. Lower row: a 68-year-old man with arthrosis, a left knee revision prosthesis operation because of a worn prosthesis 10 months and patelloplasty operation 3.5 months before the imaging. He had suffered from pain and swelling of the left knee after the last operation. ^99mTc-ciprofloxacin images (from left to right) at 1, 4 and 24 hours after injection demonstrate increased uptake, which remains visible. The patient had verified Staphylococcus epidermidis prosthesis infection.

		Next
		Acknowledgements

Homepage of this publication

Library Units | Collections | Databases | Library News | Library Services | Electronic Collection | Links elsewhere | Alphabetical Index

© 2003 Oulu University Library