One limitation to the use of the polymerase chain reaction (PCR) to identify orthopedic infections has been apparent falsepositive results, possibly due to the detection of dead bacteria. We recently showed that the use of DNA-binding agent propidium monoazide (PMA) could distinguish viable from heat-inactivated bacteria, and, in this study, we investigated whether the same technique can be applied to bacteria killed by two antibiotics with distinctly different mechanisms of action, a test of greater clinical relevance than thermal inactivation. Staphylococcus aureus and S. epidermidis were inactivated by vancomycin and gentamicin and treated with PMA or left untreated before DNA extraction. The threshold cycle difference of antibiotic-treated bacteria with and without PMA pretreatment was investigated with PCR primers for the 16S rDNA and tuf genes. Our results indicated that PMA effectively inhibited detection by PCR of bacteria, which had been inactivated by either vancomycin or gentamicin. The effect was statistically significant at 24 h after treatment (C t difference consistently >3; p < 0.05) and after 10 days of treatment (C t difference >4; p < 0.01), when compared to viable cells (C t difference 1-2). Vancomycin had a stronger effect on the C t value than gentamicin, reflecting the different mechanism of action of each antibiotic. Techniques of this type may help reduce clinically false-positive PCR results caused by the detection of dead bacteria, and may be especially useful in patients who have received antibiotics, such as patients undergoing the second stage of a two-stage revision for infected arthroplasty. ß