Dual-Energy CT for Detecting Painful Knee Prosthesis Loosening

Background Dual-energy CT (DECT) is an alternative to radiography and single-energy CT (SECT) for detecting prosthesis-related complications. Purpose To compare the diagnostic performance of DECT, SECT, and radiography for knee prosthesis loosening, with use of surgery or imaging follow-up reference standards. Materials and Methods In this prospective single-center study from December 2018 to June 2021, participants with unilateral painful knee prostheses underwent radiographic, SECT, and DECT imaging. Five blinded readers, four radiologists, and one orthopedic surgeon evaluated the images. Prosthesis loosening was diagnosed by a periprosthetic lucent zone greater than 2 mm. The sensitivity, specificity, and area under the receiver operating characteristic curve (AUC) of each method were determined and compared with use of a multireader multicase analysis. Results There were 92 study participants (mean age ± SD, 70 years ± 9.4; 67 women) evaluated. Tibial and femoral loosening were diagnosed in 47 and 24 participants, respectively. For the tibia, mean sensitivity and specificity for arthroplasty loosening were 88% and 91%, respectively, for DECT, 73% and 78% for SECT, and 68% and 81% for radiography. For the tibia, DECT demonstrated similar diagnostic performance (AUC, 0.90) to SECT (AUC: 0.90 vs AUC: 0.87, respectively; P = .13) but was superior to radiography (AUC: 0.90 vs AUC: 0.82; P = .002). Overall diagnostic performance of DECT (AUC, 0.87) for the femur was superior to both SECT and radiography (P < .001). Conclusion Dual-energy CT had generally better diagnostic performance in detecting loosening of tibial and femoral components after total knee arthroplasty compared with single-energy CT or radiography. Clinical trial registration no. 2942 © RSNA, 2022