Background: Currently, there is no consensus regarding the
principles of empiric antibiotic treatment of suspected periprosthetic
infection following total knee and hip arthroplasties. This study was
undertaken to attempt to establish such principles.
Methods: We performed a retrospective analysis of 146 patients who
had had a total of 194 positive cultures of specimens obtained at the time of
a reoperation following a total knee or total hip arthroplasty at one of two
institutions. Patient demographic data, comorbid conditions, bacterial
species, the antibiotic sensitivity profile, and the postoperative day on
which the culture tested positive were recorded.
Results: Specimens from 110 hips and eighty-four knees were positive
on culture. Seventy percent of the infections were classified as chronic; 17%,
as acute postoperative; and 13%, as acute hematogenous. The mean time between
the operation and the positive culture results was three days. Gram-positive
organisms caused the majority of the infections. In the series as a whole, 88%
of the bacteria were sensitive to gentamicin; 96%, to vancomycin; and 61%, to
cefazolin. The most antibiotic-resistant bacterial strains were from patients
for whom prior antibiotic treatment had failed. Acute postoperative infections
had a greater resistance profile than did chronic or hematogenous infections.
Bacteria isolated from patients with a hematogenous infection had a high
sensitivity to both cefazolin and gentamicin.
Conclusions: Empiric antibiotic treatment for suspected
periprosthetic infection should be guided by the class of the infection and
the findings of Gram staining. We believe that, until the final culture
results are available, acute hematogenous infections should initially be
treated by a combination of cefazolin and gentamicin therapy. All chronic and
acute postoperative infections with Gram-positive bacteria and all cases in
which a Gram stain fails to identify bacteria should be managed with
vancomycin. Infections with Gram-negative bacteria should be managed with a
third or fourth-generation cephalosporin. Infections with mixed Gram-positive
and Gram-negative bacteria should be managed with a combination of vancomycin
and a third or fourth-generation cephalosporin. Furthermore, we believe that
if culture results and other confirmatory tests are not positive by the fourth
postoperative day, termination of empiric antibiotic therapy should be
Level of Evidence: Diagnostic Level IV. See Instructions
to Authors for a complete description of levels of evidence.