Application of the Mayo Periprosthetic Joint Infection Risk Score for Total Ankle Arthroplasty

Background: Assessing patient's risk of infection is fundamental for prevention of periprosthetic joint infection (PJI) after total ankle arthroplasty (TAA). The Mayo Prosthetic Joint Infection Risk Score (Mayo Score) is based on data from total hip and knee arthroplasty and has not been validated for application for TAA. Methods: A total of 405 consecutive TAA cases were followed for 6 months for PJI. Individual patients' Mayo Scores were calculated and analyzed with logistic regression and receiver operating characteristic (ROC) for predictability for PJI. A critical cut-off Mayo Score for patients at high risk of PJI was determined by best Youden index. Among the Mayo Score-defined high-risk patients, the contribution of different risk factors were compared between the PJI and non-PJI patients. Results: There were 10 cases of PJI (2.5%) among the 405 cases within 6 months after TAA. Of the 405 patients, the Mayo Scores ranged between -4 and 13 (median 2; interquartile range 0-5). The average Mayo Score was 2.5 +/- 3.4 in the non-PJI patients and 7.7 +/- 3.1 in the PJI patients (P < .001). Logistic regression showed that the probability of PJI increased with higher Mayo Scores (odds ratio 1.48, 95% CI 1.23-1.78). All but 1 PJI patients had a Mayo Score >5. The sensitivity and specificity were 90.0% and 84.3%, respectively, when a Mayo Score >5 was used as a criterion for high risk of PJI. Conclusion: This study demonstrated that the Mayo Score could similarly predict PJI risk after TAA as in total hip and knee arthroplasty. Data analysis suggests that a Mayo Score >5 could be a criterion for identifying high-risk patients for PJI, although further validation with a large number of PJI cases is necessary.

Automated summary

The Mayo Score can predict the risk of periprosthetic joint infection after total ankle arthroplasty. A Mayo Score >5 can identify high-risk patients with a sensitivity of 90.0% and specificity of 84.3%. However, further validation with a larger number of infection cases is necessary.