Purpose: This study was performed to evaluate the usefulness of $^{99m}Tc$-HMPAO-labelled leucocyte scintigraphy for diagnosing prosthetic infection after total knee replacement arthroplasty without the aid of following bone marrow scintigraphy Materials and Methods: The study subjects were 25 prostheses of 17 patients (one man and 16 women, mean age. 65 years) who had total knee replacement arthroplasty. After injection of $^{99m}Tc$-HMPAO-labelled leucocyte, the whole body planar and knee SPECT images were obtained in all patients. The subjects were classified into three groups according to clinical suspicion of prosthetic infection. Group A (n=11) with high suspicion of infection; Group B (n=6) with equivocal suspicion of infection, and Group C (n=8) with asymptomatic contralateral prostheses. Final diagnosis of infection was based on surgical, histological and bacteriological data and clinical follow-up. Results: Infection was confirmed in 13 prostheses (11 in Group A and 2 in Group B). All prostheses in Group A were true positive. There were two true positives, one false positive and three true negatives in Group B, and six true negatives and two false positives in Group C. Overall sensitivity, specificity, and accuracy for diagnosis of the infected knee prosthesis were 100%, 75% and 88%, respectively Conclusion: $^{99m}Tc$-HMPAO-labelled leucocyte scintigraphy is a sensitive method for the diagnosis of infected knee prosthesis. However, false positive uptakes even in asymptomatic prosthesis suggest that bone marrow scintigraphy may be needed to achieve improved specificity.