The aim of this study was to evaluate pitfalls and technical limitations of MR imaging in diagnosing relapse of chronic posttraumatic osteomyelitis of the lower extremities. Retrospective analysis of MR examinations in 15 patients (17 body areas) with suspected relapse of chronic posttraumatic osteomyelitis (at least 1.5 years duration/mean number of surgical procedures per patient: 5.8). The MRI findings were compared with postoperative bacteriology (n = 11) and clinical follow-up (n = 4). Five patients had additional CT examination. Magnetic resonance imaging identified all infected areas correctly, but five uninfected regions were diagnosed false positive due to postoperative scarring/oedema in bone defects (n = 4) and soft tissue (n = 1). Specificity of MRI in diagnosing active bone infection was 63% and sensitivity 100%. Additional CT was preoperatively necessary in 5 patients (33%) to further examine osteomyelitic and reparative bone remodeling. Metal artefacts were present in 11 patients, rendering complete evaluation impossible (n = 2) or considerably more difficult (n = 4). Scarring/oedema in postoperative bone defects occurs up to 13 months postoperatively and represents a major pitfall leading to low specificity. Definitive evaluation of suspected fistula, bony fragments and mineralization by MRI may be limited in this special patient group and requires additional CT in one third of patients. Metal artefacts occur in most patients and may impair or even prevent correct film evaluation in 23 and 11%, respectively.