Surgeons who treat osteomyelitis or infected implants think that microorganisms can live on and around implanted biomaterials and necrotic bone without clinical manifestations of infection. Gristina and Costerton, in their seminal work, suggested that such bacteria persist within biofilms and that they are often overlooked when diagnosis is based on standard microbiologic culture techniques. Subsequent studies using specialized techniques including sonication to remove adherent bacteria and direct detection using various forms of microscopy have confirmed that bacteria are present in many culture-negative cases. This led to the suggestion that at least some cases of failed orthopaedic implants that were considered aseptic loosening based on the absence of clinical signs of infection and the failure to isolate bacteria may actually have an infectious etiology. In addition to biofilms, potentially important concepts that also may contribute to false-negative culture results include the failure to recognize small colony variants induced during growth in vivo and the presence of bacteria inside host cells including osteoblasts. Importantly, bacteria persisting as small colony variants within biofilms and/or inside osteoblasts also may be an explanation for the recurrent nature of musculoskeletal infection.