Pseudoinfection of a total knee arthroplasty.

We were very interested in a recent report by Quale and Reese regarding pseudoinfection of prosthetic hip implants. 1 The references cited were inclusive and documented the varied kinds of situations in which pseudoepi-demics occur as well as the potentially significant consequences of such events. A similar situation recently occurred in our institution, suggesting that contamination of transport media/ swabs may be much more common than currently is appreciated. A 7&year-old woman with degen-erative arthritis had left total knee arthro-plasty in July 1992, without complication. On October 14, 1992, she fell backward while at home and incurred a lateral subluxed patella. Conservative measures failed and she was admitted on December 14, 1992, with dislocating left patella, and underwent lateral reti-nacular repair and revision of the patel-lar component of the total knee. Her leukocyte count was 6,100 with 59% neutrophils, the erythrocyte rate was 40, and urinalysis showed 5 to 10 white cells per high-power field; urine culture was negative. Pathologic examination demonstrated grossly unremarkable orthopedic prosthetic material and fragments of soft tissue. Gram stain of swab from the deep tissues placed in transport media showed many neutrophils, many red cells, and few (2 to 3/hpf) slender gram-negative rods. Aerobic and anaerobic cultures showed no growth. Following telephone report of the gram stain, with culture still pending , the patient was treated with oral ciprofloxacin. Immediate epidemiologic investigation was requested by the ortho-pedic surgeons, who found no evidence of infection at surgery and therefore were inclined to disbelieve the gram stain report. The original slide was reviewed, and the presence of gram-negative organisms was documented. The original swab still was available, and the lot number was identified. Unopened unused transport media from the same lot number were obtained from surgery and gram stained after plunging the swab into the media. Gram stains were positive for gram-negative rods from that lot number and two additional randomly selected lot numbers. We initiated the following procedure: a) notified the department of surgery ; b) removed all contaminated lot numbers and used only lots with no evidence of contamination; c) notified the supplier and changed supplier; and d) instituted routine screening gram stain on each new lot of transport media purchased for use in the operating room from the new supplier. Because prosthetic joint infections frequently are indolent, time is available to rule out pseudoinfection and to be sure that any aggressive surgery required is …