ABSTRACT
PURPOSE: The objective of the present study was to determine whether sonication yields greater sensitivity when compared with the traditional tissue culture in detecting peri-implant infections in shoulder surgery. METHODS: It is a retrospective study that includes 99 shoulder surgeries with implants explanted. The inclusion criteria required at least four tissue cultures, sonication of the material explanted, and a minimum follow-up of two years. Patients were classified according to the definition of periprosthetic shoulder infection of the 2018 International Consensus Meeting on Orthopedic Infections. The classifications are definitive infection, probable infection, possible infection, and unlikely infection. RESULTS: Among the 99 surgical procedures, 31 were considered definitive infections, 11 possible/probable infections, and 57 unlikely infections. Considering the cases with a definitive infection, the sensitivity of the tissue culture was 87.09% and the sensitivity of sonication stood at 80.64% (p = 0.406). Analyzing the cases with a definitive infection and those having a possible/probable infection together and comparing them with those with unlikely infection, the sensitivity of sonication was 80.4% and the sensitivity of the tissue culture came to 91.4%. The specificity of the sonication was 98.1% and the specificity of the tissue culture was 99.6%. CONCLUSION: The sensitivity of sonication in shoulder surgery (80.64%) is not superior to the sensitivity of the tissue culture (87.09%). Specificity remains high with both methods, being 98.1% in the sonication group and 99.6% in the tissue culture. Sonication brings no benefit to the detection of shoulder per-implant infections.
