Sepsis following total knee replacement (TKR) is a disastrous complication. The knee arthrodesis comes as the final solution to solve the chronic infection after multiple failed surgeries. Our objective was to review these procedures performed in our institution for the past 20 years.

We present a retrospective study with 5 cases of severe unsolved knee infection after TKR, who have been finally submitted to knee arthrodesis performed between 1993 and 2008. The patients are 4 males and 1 female, with a mean age of 62 years (ranging from 55 to 74 years) at the time of surgery. They presented MRSA infection (3 cases), P aeruginosa infection (1 case) and Mycobacterium tuberculosis (1 case). The average follow up was 25 months (between 12–48 months). The Visual Analogic Scale (VAS) value was registered.

The mean number of surgeries before the arthrodesis was 3,6. In 4 of the 5 cases the surgeon used an external fixator to achieve the fusion. In the other patient, an intramedullary nail was used. The arthrodesis was performed in a single-time surgery in every patient. All cases achieved knee fusion and the mean time of consolidation was 5,1 months, with the longest being 8 months until fusion. No bone graft was used in any of these cases, and the mean size of leg length discrepancy was 2,8 cm. The average VAS was 3,6. In one patient with an external fixator there was a relapse of the infection (MRSA infection), but with adequate antibiotic therapy (vancomycin) ended up to resolve. There are no records of any other complication.

The knee arthrodesis appears as last but useful resort in extreme cases of relapse infection after a TKR and multiple unsuccessful surgeries. Although it comprises an elevated level of morbidity, it also seems to allow to heal difficult and multi resistant infections and with few complications.

Septic arthritis is a rare form of infection in the healthy adult population, especially adults who have never been submitted to any articular procedure. Our objective is to describe a rare case of an infectious arthritis of a hip caused by bone abscess in a previously healthy patient with few risk factors.

This is the case of a 40 year-old female, who had a history of heavy smoking and low socioeconomic status. She was examined complaining with pain in the left hip and the x-ray showed a lytic image in the intracapsular femoral neck, but it was overlooked and the diagnosis of bone abscess was missed. She was to the hospital 3 weeks later with intense pain and a suppurative and swollen right thigh. Blood analyses were compatible with infection (elevated white-blood cells and C-reactive protein). Pelvic CT scan showed a lytic image in the intracapsular femoral neck with pus inside the hip. It also showed a huge abscess and infiltration of the tissues surrounding. She was then submitted to multiple surgeries in order to drain the abscess and articular cleaning. A S. aureus MSSA was isolated and an adequate antibiotic was instituted. Although this measures, there was only partial clinical and analytical success.

We eventually ended-up by resecting the femoral head, which had no viability, as shown in an MRI, and was behaving as a sequestrum, justifying the impossibility in controlling the infection. The femoral head was replaced by a temporary cement-spacer impregnated with vancomycin, and the patient maintained adequate antibiotic therapy until normalization of the infection parameters. She now has no signs of infection, and is waiting to be submitted to a total hip replacement.

This is a rare case of a septic arthritis with an uncommon onset in a healthy woman who had never been submitted to a hip procedure and had no local risk factors for the development of such pathology. In the research we did, we found no other case similar to this, in which the arthritis started as an intracapsular bone abscess with hematogenous origin. Reporting this case, we would also like to emphasize that, although its rarity, it is extremely important to always exclude septic arthritis in cases of joint pain, in order to prevent catastrophic consequences.

Intramedullary nailing (IMN) has been frequently indicated to treat long bone open and closed fractures, but infection following internal fixation may have devastating consequences, with higher costs. Treatment of intramedullary nail-associated infections (IMNI) is challenging and based upon surgery and adequate antibiotic administration, which requires the correct identification of causative microorganisms. However, there have been difficulties for the microbial diagnosis of IMNI, as the peri-prosthetic tissue cultures may show no microbial growth, particularly in patients with previous use of antibiotics. Sonication have shown higher sensitivity and specificity for microbial identification on a variety of orthopedic implant-associated infections. Aim: To compare clinical and microbiological results and sensitivity for the pathogen identification obtained by conventional peri-implant tissue culture samples with culture of samples obtained by sonication of explanted IMN implants, among patients presenting IMNI of long bones.

Methods: Longitudinal prospective cohort study performed at a tertiary public hospital, ongoing since August 2011. We analyzed all patients with indication for IMN implant removal, and orthopedic-implant associated infections was defined according to previous publications addressing osteosynthesis-associated infections (Yano 2014). Minimal of 2 samples from the peri-implant tissue were taken and sent under sterile conditions to the laboratory for culture. Statistical analysis was performed McNemar's test for related proportions.

Results: We included 26 patients presenting clinical signs of IMNI, of which tissue and sonication cultures were performed for 26 (100%) and 20 (77%) patients, respectively. Among them, 88% were male, with mean age was 35.9 years (range, 19–59 yo). Causes of trauma were mainly motorcycle crashes accounting 54% of accidents; tibia and fibula were affected in 65% and 27%, respectively. Gustilo open fracture classification was grade II (35%) and IIIA (35%). First stage management with external fixation for fracture stabilization was performed in 75% of trauma patients. Sensitivity of peri-prosthetic tissue culture and sonication was 80.7% (21/26), and 95% (19/20) (p< 0.05), respectively. Only one infected patient presented negative tissue and fluid cultures. Gram-positive cocci were isolated in 75% and 79% in tissue and sonication fluid cultures, respectively. Staphylococcus aureus, coagulase-negative staphylococci, Enterococcus sp., were isolated from tissue and sonication culture in 43.5% and 36.3%, 8.7% and 22.7%, 13% and 13.7%, respectively. Polymicrobial infection was diagnosed in 3.8% (1/26) and 15.8% (3/19), patients by tissue and sonication fluid cultures (p< 0,01), respectively.

Conclusion: Sonication of retrieved infected intramedullary nails has the potential for improving the microbiological diagnosis of IMNI.