Abstract
Introduction
The surgical management of late PJI was usually done in two stages with the placement of a temporary cement spacer. The development of one-stage surgical care raises questions about the two-stage strategy. The objectives of this study are to identify the complications related to the presence of the cement spacer within a two-stage strategy. The septic recurrence rate is also evaluated after a minimum follow-up of two years.
Material and methods
Medical files of 208 patients (101 knees and 107 hips) who underwent a two-stage revision for late PJI prosthesis infection were retrospectively reviewed. Antibiotic loaded articulated homemade cement spacers were used. Second stage was usually planned on average 4 to 6 weeks after the first stage. Patients were allowed to walk without loading. The success rate was defined as the absence of septic recurrence after a minimum follow-up of two years. Descriptive statistics and uni- and multivariate analysis were conducted.
Results
The spacers were left in place for an average of 42 days for the knees and 30 days for the hips. Six patients (3%) died before performing the second stage. Hip spacers were associated with 8 fractures for only one observed in the knee (4%). Spacer dislocation (11%) was observed in 23 cases (13 for the hip and 10 for the knee respectively). Treatment failure with recurrence of the infection within 2 years was observed in 26 patients (12%). Resistance to the antibiotic present in the cement was found in one third of infectious failures. The presence of a prior cemented prosthesis was significantly associated with the presence of a germ resistant to gentamicin and the persistence of the germ at the second stage. However, it was not associated with failure at two years.
Discussion
the two-stage management of PJI is associated with a non-negligible mortality rate before the second stage, rarely reported in studies. The presence of an initially cemented prosthesis is associated with the presence of germs resistant to the antibiotic contained in the cement and exposes to the persistence of the resistant germ at the second stage. Spacer fractures are observed more at the hip, but less frequently than in previous reports, while dislocations are observed at both the hip and the knee, particularly due to loss of tibial fixation in this area.
These observations are all arguments for further consideration of revision surgery in 1 stage.