Summary Aim: Diagnosis of infection with help of antigranulocyte scintigraphy near body stem is difficult because of contemporary visualisation of bone marrow. Therefore, we investigated, whether it is possible to improve the accuracy in diagnosing septic loosening of hip endoprosthesis by changing the analyzing methods. Methods: In 28 patients, the results of a visual interpretation of late scan, a visual interpretation and a quantitative interpretation of time-activity-course were compared. These results were verified by histology respectively microbiology. Results: Histological and microbiological verification found 14 septic loosening and 14 aseptic loosening of the hip protheses. Therefore, sensitivity, specifity, negative and positive predictive value for the visual interpretation of late scan were 0.86,0.57,0.80 and 0.67. For visual and quantitative interpretation of time-activity-course we found 0.86,0.79,0.85 and 0.80 respectively 1,0.93, 1,0.93. For interobserver agreement we found kappa coefficients of 0.28 ± 0.2 for visual interpretation of late scan, 0.48 ±0.17 for visual interpretation and 1.0 ± 0 for quantitative interpretation of time-activity-course. Conclusion: In all investigated values quantitative interpretation of time-activity-course was superior to the other analyzing methods. Therefore, antigranulocyte scintigraphy for septic loosening of hip endoprosthesis should be interpreted quantitativly.