Frozen Section Histology: A New Tool to Identify Propionibacterium acnes Infections in Revision Shoulder Arthroplasty: Commentary on an article by Matthew J. Grosso, BS, et al., “Sensitivity of Frozen Section Histology for Identifying Propionibacterium ac

Propionibacterium acnes is one of the most frequent pathogens responsible for shoulder infection, especially with failed shoulder arthroplasty. This bacterium is a particularly slow-growing anaerobic gram-positive organism. The clinical symptoms of infection are subtle and diagnosis is often made by presumption. Aspiration of the joint can be an option to make the diagnosis. However, sensitivity and specificity of the aspirate analyses are not optimal; because of the slow-growing nature of P. acnes, the culture must be followed for fifteen to twenty-one days and, therefore, the presence of P. acnes can be missed if the cultures have been discarded earlier. Frozen section histology of tissues can be an option to make a more accurate diagnosis as with periprosthetic hip or knee infection. Appropriate diagnosis can be essential to determine surgical and medical management, including extent of removal of components and cement, one or two-stage revision, and length and type of postoperative antibiotic course. Few studies have evaluated the sensitivity of frozen section analysis in the shoulder, and no studies have previously examined the performance of frozen section in P. acnes infections.

This single-center study was retrospective and comparative with control groups and analyzed all patients who underwent revision shoulder arthroplasty between 2005 and 2012. Patients with proven periprosthetic joint infection were compared with a group who had undergone aseptic revisions. Because there is no gold standard to establish the diagnosis of infection in shoulder arthroplasty, the authors have arbitrarily created a spectrum of infection categories: no evidence for infection, possible infection, probable infection, or definite infection based on the existence of negative or positive cultures and preoperative or intraoperative findings of infection. For final analysis, infected patients were subsequently divided into groups of patients who had P. acnes infections and patients who had infections from other organisms, to create three groups: (1) the non-infection group, (2) the P. acnes infection group, and (3) the other infection group. Intraoperative frozen section histology specimens were collected for forty-five patients for reanalysis. All tissue samples that were evaluated were deep periprosthetic specimens, obtained from along the interfaces of the glenoid or humeral components or adjacent to the implants. Anaerobic tissue and fluid cultures were incubated systematically for fourteen days. For all specimens, the total number of polymorphonuclear leukocytes in the five high-power fields (400×) with the highest concentration of polymorphonuclear leukocytes was determined.

The results of this study have shown the low sensitivity of the erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP) to diagnose P. acnes infection; the P. acnes infection group had a sensitivity of 61% for ESR and 33% for CRP. The specificity of ESR and CRP was 85% for both the P. acnes infection group and the other infection group. The sensitivity of frozen section histology for P. acnes infection was 50% and the specificity was 100% on the basis of the current institutional guidelines. Using the American Academy of Orthopaedic Surgeons (AAOS) guidelines of ten polymorphonuclear leukocytes per high-power field for five or more fields, sensitivity was only 39%. However, there was no false positive using any of the thresholds, yielding a specificity of 100%. Using a cutoff from seven to ten polymorphonuclear leukocytes, the sensitivity of frozen section histology for P. acnes infection improved to 72%, while maintaining 100% specificity.

The results of this study showed that the guidelines recommended by the AAOS Practice Guidelines Committee for Diagnosing Periprosthetic Joint Infections of the Hip and Knee had high specificity (no false positives) but relatively low sensitivity when applied to revision shoulder arthroplasty to diagnose P. acnes infection. A new threshold of a total of ten or more polymorphonuclear leukocytes in five high-power fields may help increase the sensitivity of frozen sections to diagnose the commonly indolent periprosthetic infections of the shoulder with minimal impact on specificity.