Re-operation rates of arthroscopic management versus arthrotomy in treatment of septic arthritis of native shoulder joint in adults. A systematic review and meta-analysis

Aims Previous studies have reported concern regarding high reoperation rates when septic arthritis of the native shoulder is treated arthroscopically, compared to open arthrotomy. We aimed to compare re-operation rate between the two strategies. Patients and methods The review was registered prospectively at PROSPERO, (CRD42021226518). We searched common databases and references lists (8 February 2021). The inclusion criteria included interventional or observational studies of adult patients with a confirmed diagnosis of native shoulder joint septic arthritis and had either arthroscopy or arthrotomy. The exclusion criteria included patients with periprosthetic or post-surgical infections, patients who had atypical infections, and studies that did not report re-operation rate. Cochrane Collaboration’s tool for assessing risk of bias (ROBINS-I) was used. Results Nine studies (retrospective cohort studies) were included that involved 5,643 patients (5,645 shoulders). Mean age ranged from 55.6 to 75.5 years, and follow-up time ranged from 1–41 months. Mean duration of symptoms prior to presentation ranged from 8.3–23.3 days. Metanalysis observed a higher re-operation rate for reinfection at any time point following initial arthroscopy in comparison to arthrotomy, odds ratio 2.61 (95% confidence interval 1.04, 6.56). There was marked heterogeneity ( I 2 = 78.8%) among studies including surgical techniques and missing data. Conclusion This metanalysis observed a higher reoperation rate in arthroscopy in comparison to arthrotomy for the treatment of native shoulder septic arthritis in adults. The quality of the included evidence is low and the heterogeneity among included studies is marked. Higher quality evidence is still needed that address limitations of previous studies.