Clinical significance of additional lateral imaging in diagnosis of prosthesis infection by combined Tc-99m labeled leukocyte/bone marrow scintigraphy

Abstract

Introduction: Recently the most accurate diagnostic method in the diagnosis of prosthesis infection is the combination of Tc99m labeled leukocyte/bone marrow imaging. However the diagnostic efficiency of the test is influenced because of its low spatial resolution. Recently SPECT/CT imaging is implicated in this test with efficient additional diagnostic improvement. In the departments without capability of SPECT/CT lateral imaging may provide important clinical information. The aim of this study was to analyze the additional diagnostic information provided by additional lateral spot images to the diagnostic accuracy of combined Tc-99m labeled leukocyte/bone marrow imaging in diagnosis of prosthesis infection. Methods: Retrospective evaluation of 11 patients (9 F, 2 M; 61 +/- 8,6 years old) with pre-diagnosis of prosthesis infection (4 patients with right knee, 3 left, 2 bilateral, 2 right hip) by two experienced Nuclear Medicine physicians were included into the study. The age of the prosthesis were 7,8 +/- 15,3 months by the time of imaging. The imaging was performed from the region of interest at the first, second and fourth hour of injection of approximately 5 mCi (185 MBq) labeled leukocyte imaging and one hour after injection of 5 mCi (185 MBq) Tc-99m nanocolloid in anteroposterior and in additional lateral position. The diagnosis of prosthesis infection was decided in case of increased activity accumulation in leukocyte imaging higher than in bone marrow imaging. The final diagnosis was decided by additional aspiration biopsy (n=1), culture results (n=3), revision arthroplasty (n=1) or according to follow up results (n=6)(sedimentation, CRP) in 3,4 +/- 3,4 months follow up. Results: The labeling efficiency of the patients was in an acceptable range (33-67%) in leukocyte labeling. The diagnosis of prosthesis infection was decided in two patients according to scintigraphy results and these two patients were documented to have prosthesis infection (revision arthroplasty and biopsy results). One patient was decided to have prosthesis infection despite negativity of scintigraphy (false negative) due to persistently high sedimentation, CRP levels in follow up despite proper antibiotic treatment. One patient was diagnosed as negative due to demonstration of anterior cellulites rather than prosthesis infection in the lateral images. The diagnostic sensitivity, specificity, accuracy, negative and positive predictive value of the test were 60%, 87.5%, 81%, 60%, 87.5%, respectively for anteroposterior imaging only with additional lateral imaging the rates were increased to 60%, 100%, 91%, 89% and 89%, respectively. Conclusion: Lateral imaging increases the diagnostic accuracy of the combined Tc-99m labeled leukocyte/bone marrow scintigraphy compared to anteroposterior imaging only and might be implemented in combined imaging protocol in case of non-availability of SPECT/CT imaging.