gms | German Medical Science

Objectives: Septic arthritis is an orthopaedic emergency, which requires prompt treatment to avoid permanent joint damage and potentially life-threatening complications. Management of the septic arthritis includes antibiotic therapy as well as immediate joint-decompression. Open or arthroscopic lavage and debridement are recommended treatment options. The failure rates after a single surgery remain high. Between 26% and 50% of the affected patients require more than one surgical debridement. Information about risk factors for infection-recurrence are scarce, but necessary to develop individual treatment strategies.

The objective is to identify potential risk factors for a failure of a single surgical debridement.

Methods: We obtained data of patients who were treated at our clinic because of acute septic arthritis of the knee and the shoulder between 2003 and 2018. In total 268 patients were included for final analysis. All patients were either treated arthroscopic or by open arthrotomy. We collected demographic data, medical history, initial and postoperative laboratory parameters and co-morbidities from each patient. The Charlson Co-morbidity Index (CCI) was additionally applied to summarize comorbidities. Kellgren and Lawrence classification was used to determine osteoarthritis grades. We used non-parametric tests to compare patients with and without a reinfection. Further, we performed a logistic regression analysis to identify risk factors for recurrence of infection.

Results: From 268 patients 209 (78%) were treated because of septic arthritis of the knee and 59 because of a bacterial shoulder infection. 167 (62.3%) patients underwent an arthroscopy and 101 (36.7%) an open arthrotomy. In total 73 patients (27.2%) suffered a recurrence of infection. Patients with a reinfection had a significantly longer duration of symptoms (p=0.001), a higher CCI (p=0.002), higher C-reactive protein (CRP) levels on the third and fifth postoperative day (p=0.001 and p<0.001, respectively) and lower postoperative haemoglobin levels (p=0.024). In logistic regression the presence of insulin dependent diabetes mellitus (IDDM) (odds ratio, OR 2.838; 95% CI 1.019 - 7.903; p=0.046) and higher Kellgren and Lawrence grades (OR 1.609; 95% CI 1.223 - 2.117; p=0.001) increased the risk of reinfection. Further a CRP drop of less than 20% on the third postoperative day and a CRP drop of less than 30% on the fifth postoperative day was significantly associated with a higher risk of infection recurrence (OR 0.515; 95% CI 0.275 - 0.962; p=0.036 and OR 0.482; 95% CI 0.263 - 0.885; p=0.019, respectively).

Conclusions: We could identify IDDM, severe osteoarthritis, higher scores of the CCI, longer duration of symptoms, higher postOP CRP levels and postOP anemia as independent risk factor for the recurrence of infection. In the context of personalized medicine, the presented results should help physicians to develop individual risk profiles and consequently individualized treatment strategies for their patients.