The Use of Erythromycin and Colistin Cement in Total Knee Arthroplasty Does Not Reduce the Incidence of Infection: A Randomized Study in 2,893 Knees With a 9-year Average Follow-Up.

BACKGROUND

One of the most severe complications of primary total knee arthroplasty (TKA) is prosthetic joint infection. Currently, the use of antibiotic-loaded cement for the prevention of infection is still controversial. The aim of the present study was to evaluate if the use of antibiotic-loaded cement reduces the infection rate in primary TKA in long-term follow-up (more than 5 years average follow-up).

METHODS

This study is the follow-up extension of a prospective randomized study, with 2,893 cemented TKA performed between 2005 and 2010 at our institution. There were 2 different cohorts depending on which bone cement was used: without antibiotics (control group) or those loaded with erythromycin and colistin (study group). All patients received the same systemic prophylactic antibiotics. The patients were followed for a minimum of twelve months. The diagnosis of prosthetic joint infection was done according to Zimmerli criteria.

RESULTS

In 1,452 patients, the prosthetic components were fixed using bone cement without antibiotics, whereas in 1,441 patients, bone cement was loaded with erythromycin and colistin. Both groups were comparable in terms of all the possible risk factors studied. We found a total of 53 deep infections, with a mean rate of 1.8%. There were no differences between the groups as to whether bone cement with or without antibiotics had been used (P = .58). The average duration of follow-up was 8.7 years. In terms of prosthetic revision due to aseptic loosening, there were no differences between groups (P = .32), with 33 revision arthroplasties in the control group and 37 in the study group. Moreover, we analyzed the erythromycin resistance rate, with no differences between both groups (P = .6).

CONCLUSIONS

The use of erythromycin and colistin-loaded bone cement in TKA did not lead to a decrease in the rate of infection in long-term follow-up, a finding that suggests that its use would not be indicated in the general population.