PurposeArthrofibrosis after total knee arthroplasty represents a considerable burden for the patient and a therapeutic challenge for the practitioner. One possible cause discussed in the literature is a low-grade infection. This hypothesis should be examined within the scope of this retrospective study.Patients and methodsNineteen patients with clinical symptoms of arthrofibrosis after primary total knee arthroplasty were examined between January, 1999 and January, 2012. Incorrect positioning was radiologically ruled out. All patients were examined clinically (score of Freeman as well as Blauth and Jäger), radiologically (component and leg alignment, patella height according to Insall and Salvati), microbiologically (culture-based procedures), molecular biologically (PCR) and histologically in the course of an open revision of the prosthesis.ResultsAccording to the score of Freeman et al. (1977), a highly significant improvement in pain (p = 0.007) and in the overall score (p = 0.003) was shown. The knee joint mobility did not change significantly (p = 0.795). PCR was negative in 17 patients. One patient showed a PCR-positive result of the synovial membrane for Corynebacterium spp., while Staphylococcus warneri was detected in the culture. Another patient had a positive result of synovia PCR for Enterococcus cecorum as well as Corynebacterium spp. However, this culture was sterile. In 16 patient samples, no bacterial growth was detectable. Two samples were not evaluable. The main histopathological findings were synovialitis and fibrosis.ConclusionThe hypothesis of low-grade-infection-induced arthrofibrosis after total knee arthroplasty could not be confirmed in this study. However, based on this small study population the conclusion needs to be confirmed by new and larger studies, ideally prospectively designed including a control group.