Bone SPECT/CT has excellent sensitivity and specificity for diagnosis of loosening and patellofemoral problems after total knee arthroplasty.

Abstract

The primary purpose of this retrospective study with prospectively collected data was to evaluate the clinical value of SPECT/CT, in particular its impact on clinical diagnosis and patient management, in patients with pain, stiffness, or swelling after primary total knee arthroplasty (TKA). The secondary purpose was to determine the sensitivity and the specificity of SPECT/CT for different pathologies such as loosening and patellofemoral overloading. A retrospective analysis was carried out on 214 knees in 202 consecutive patients (mean age 69 ± 11years) with persistent or recurrent knee pain after TKA. All patients underwent clinical examination, standardised radiographs, and 99mTc-hydroxymethane diphosphonate (HDP) SPECT/CT as part of a routine diagnostic algorithm. The diagnoses before and after SPECT/CT and final treatment were recorded. TKA component position was routinely measured on 3D-reconstructed images. Intensity and anatomical distribution of bone tracer uptake were determined. SPECT/CT changed the clinical diagnosis and final treatment in 139/214 (65%) knees. In 117 (54.7%) out of 214 patients, revision surgery was proposed following SPECT/CT. Early onset of pain was significantly (p = 0.011) correlated with tibiofemoral stress pattern in SPECT/CT. A total of 86 knees (40.2%) were surgically revised using semi-constrained or fully constrained TKA. In 35 patients, a secondary patellar resurfacing was performed. SPECT/CT showed the clinical diagnosis to be incorrect for suspicion of tibial component loosening in 56 patients (26.2%), femoral component loosening in 53 patients (24.8%), and patellofemoral overloading/progressive patellofemoral osteoarthritis in 26 patients (12.1%) and provided different underlying causes of persistent knee pain after TKA. Likewise, SPECT/CT identified tibial component loosening in 8 patients (3.7%), femoral component loosening in 4 patients (1.9%), and patellofemoral OA in 71 patients (33.2%) without prior clinical suspected diagnosis. The sensitivity and specificity of SPECT/CT for detection of patellofemoral OA was 96.5% and 96.2%, respectively. The sensitivity and specificity for detection of tibial component loosening was 96.0% and 100%, respectively. The sensitivity and specificity for detection of femoral component loosening was 95.0% and 100%, respectively. The diagnostic benefits of SPECT/CT in the challenging and complex cohort of patients with pain after primary TKA have been proven. The excellent sensitivity and specificity for detection of tibial or femoral component loosening and diagnosis of patellofemoral OA have been surgically confirmed. Due to the benefits in establishing the correct diagnosis, SPECT/CT should be implemented as part of the routine diagnostic algorithm for patients with pain after primary TKA. II.