Diagnostic Arthroscopy for Detection of Periprosthetic Infection in Painful Shoulder Arthroplasty

Purpose To analyze the utility of arthroscopic biopsies for detection of periprosthetic infection in painful shoulder arthroplasty without objective signs of infection. Methods A retrospective analysis of all patients who underwent a diagnostic arthroscopy for painful shoulder arthroplasty from June 2012 through July 2018 was performed. Patients with a subsequent revision shoulder arthroplasty after diagnostic arthroscopy were included. Arthroscopic tissue culture results were compared with the culture results of intraoperative tissue samples obtained at the time of open revision surgery. A minimum of 3 tissue samples from synovia and bone-prosthesis interface with signs of synovitis or abnormal appearance was routinely collected. Cases with 2 or more positive cultures for the same microorganism obtained at open revision surgery were considered as true presence of infection. The study protocol was reviewed and approved by the institutional ethics committee. Results Twenty-three cases in 22 patients were included in this study. Five of these 23 cases were classified as true infection based on the samples obtained during open revision surgery, and 16 cases had a positive culture in diagnostic arthroscopy. Cutibacterium acnes was isolated in each case. Classifying any microbiologic growth in the arthroscopic biopsies as positive resulted in a sensitivity and negative predictive value of 100%, specificity of 39%, and positive predictive value of 31.3% for the detection of a periprosthetic shoulder infection (PPSI). If at least 2 positive samples with the same microbiologic growth in the arthroscopic biopsies were considered as positive, sensitivity and negative predictive value dropped to 80% and 94.4%, respectively, but the specificity and positive predictive value increased to 94.4% and 80%, respectively. Conclusions Diagnostic arthroscopy is a useful diagnostic tool in patients with suspicion but no clear evidence of PPSI. Arthroscopically obtained tissue biopsies for culture offer a high sensitivity and specificity in the diagnosis of PPSI if at least 2 cultures positive for the same microorganism are considered as infection. Level of Evidence Level III.