We prefer to treat suspected periprosthetic shoulder infections with a single stage revision: thorough debridement, antibiotic/Betadine irrigation, topical antibiotics, complete exchange of implants, and a course of postoperative antibiotics. Two stage revision is usually reserved for cases of draining sinus, infection with particularly virulent organisms, or failed prior single stage revision. Completion of a two stage procedure requires (a) removal of implants and insertion of a spacer at the first stage and (b) insertion of new implants at a second procedure. Not only does the two stage require two surgeries, it also subjects the patient to increased challenges of spacer fixation/removal and challenges of fixation of the final implants. A substantial percentage of patients planned for a two stage never end out having the second stage completed.
Evaluating whether a revision surgery has failed to eliminate an infection is a problem because the only sure evidence of a failed revision for infection is evidence that the organism cultured at the first stage persists in the shoulder. This evidence can come from cultures obtained at the second stage (or a joint aspirate, arthroscopic biopsy, or draining sinus) that are positive for the original causative organism.Â
Because the organisms that are most commonly isolated from failed shoulder arthroplasties tend to form biofilms on implants, it is preferable to remove all metal and plastic components and cement when performing a revision for suspected periprosthetic shoulder infection. However, complete removal of cement and retained fragments of broken hardware can be not only difficult, but also hazardous, risking the integrity of bone and the surrounding neurovascular structures.
Thirty-seven patients were included in the analysis. Stage one revision failed to resolve the periprosthetic infection in ten patients (27%). The authors found that the risk of recurrent infection was not associated with age, BMI, comorbidity index, patient, sex, or presence of diabetes.Â
Only two of the 10 (20%) of the cases of recurrent infection were culture positive for Cutibacterium at the index surgery, while 16 of the 27 (59%) of the cases without recurrent infection were culture positive for Cutibacterium at the index surgery. 70% of the recurrent infections were due to Staph Aureus or were polymicrobial.
Six patients had retained cement and one patient had two retained broken glenoid baseplate screws after first-stage revision.  All retained cement identified in this study was distal to the humeral stem
Of the ten cases of recurrent infection, one case involved retained cement/hardware.Â
The authors point out that surgeons must balance the potential benefit of complete cement/hardware removal against the risks.
Comment: It is interesting that while 1 of 7 patients with retained cement or hardware were documented has having recurrent infection, 9 of 30 patients without retained cement or hardware had recurrent infection. Thus we must wonder why the initial surgery failed to achieve its objective of successful infection resolution in these 9 cases of complete cement and hardware removal. The pie graphs shown above suggest that a determinant of the percentage of success in resolving the infection at the first stage may be whether the infection was due to Cutibacterium (lower risk of failure) versus Staph aureus or polymicrobial (higher risk of failure). Not assessed in this study are the possible effects on infection recurrence of adjunctive measures at the first stage procedure, such as antibiotic and Betadine irrigation, topical antibiotics, and postoperative antibiotic choice, route of administration and duration.