Intraoperative frozen section analysis in revision total joint arthroplasty

A prospective study of the sensitivity, specificity, and predictive values for frozen sections against cultures obtained at the time of revision total joint replacement was done. One hundred twenty-one revision total joint replacements were done in 92 men and 29 women. A positive frozen section with more than 10 polymorphonuclear leukocytes per high power field was compared with the intraoperative cultures. Twenty-one patients who had revision surgery had greater than 10 polymorphonuclear leukocytes per high power field. Of these, 14 patients had positive cultures. The remaining 100 patients had less than 10 polymorphonuclear leukocytes per high power field, but seven had positive cultures. Statistical analysis of frozen sections for all total joint arthroplasties revealed a 67% sensitivity, 93% specificity, 67% positive predictive value, and 93% negative predictive value. Analysis of frozen sections for total hip arthroplasties revealed a 45% sensitivity, 92% specificity, 55% positive predictive value, and 88% negative predictive value. Analysis for total knee arthroplasties revealed 100% sensitivity, 96% specificity, 82% positive predictive value, and 100% negative predictive value. Comparisons of sensitivity, positive predictive value, and negative predictive value between total knee arthroplasty and total hip arthroplasty were significant. The results indicate that the use of intraoperative frozen section analysis with greater than 10 polymorphonuclear leukocytes per high power field as an indication of infection lacks the positive predictive value and sensitivity for accurate determination of prosthetic infection at the time of revision total hip arthroplasty. Frozen sections have an acceptable sensitivity and positive predictive value in total knee arthroplasty. The results of the current study show the limitation of using frozen sections as a diagnostic test for infection in revision total hip arthroplasty.