The value of an open biopsy in the diagnosis of periprosthetic joint infection

Background

Determining the presence of bacteria in the shoulder prior to shoulder arthroplasty can be challenging especially in the case of revision arthroplasty. An open biopsy provides an opportunity to obtain tissue samples with minimal patient morbidity. The purpose of this study was to characterize the diagnostic utility of an open shoulder biopsy.

Methods

A retrospective cohort study was performed at an academic medical center. All patients that underwent an open shoulder biopsy using a small proximal deltopectoral incision between 2008 and 2021 were included. Demographics, surgical history, culture results, and development of subsequent infection were recorded. Subsequent infection was defined as the development of a sinus tract, purulent drainage, or revision surgery with greater than or equal to two tissues specimens with growth of the same bacterial species. Sensitivity and negative predictive value (NPV) of an open biopsy were calculated based on the development of subsequent infection. As culture positive patients were treated for their infection, positive predictive value and specificity could not be determined.

Results

We identified 55 patients that underwent 75 open biopsies. Most patients had a shoulder arthroplasty in place at the time of biopsy (69.1%), while 23.6% had an antibiotic spacer, and 7.3% had a native shoulder. Patients with a history of infection were more likely to have a spacer in place at the time of biopsy (65% vs. 0%; P < .001). The sensitivity of an open biopsy was 60% and the NPV was 83% among patients with a shoulder arthroplasty with no history of infection to predict infection after revision arthroplasty. The sensitivity was 63% and NPV was 75% among patients with any history of shoulder infection. The sensitivity was 67% and the NPV was 83% among patients with an arthroplasty in place at the time of biopsy independent of prior infection. The sensitivity was 62% and the NPV was 75% among patients with an antibiotic spacer at the time of biopsy. Mean operative time was 32.2 ± 10.5 minutes.

Discussion

This diagnostic utility of an open shoulder biopsy is not influenced by whether there is a history of infection or whether there is currently a spacer or an arthroplasty in place, with a sensitivity between 60% and 67% and a NPV between 75% and 83%.