Can Periprosthetic Joint Infection of Tumor Prostheses Be Controlled With Debridement, Antibiotics, and Implant Retention?

Abstract

Background: Two-stage revision for periprosthetic joint infection (PJI) in patients who have undergone segmental replacement of the distal femur or proximal tibia after tumor resection can be associated with considerable morbidity, pain, and risk of complications because the procedure often results in removal of long, well-fixed stems from the diaphysis. A less-aggressive surgical approach, such as debridement, antibiotics, and implant retention (DAIR), may be attractive to patients and surgeons because of less morbidity, but the likelihood of eradicating infection in comparison to the traditional two-stage revision is not well established for oncology patients. Furthermore, the relative risk of subsequent amputation for DAIR versus two-stage revision has not been defined for this population.

Questions/purposes: (1) How does DAIR compare with two-stage revision in terms of infection control for patients with distal femoral or proximal tibial segmental modular endoprostheses? (2) Is DAIR as an initial procedure associated with an increased risk of amputation compared with two-stage revision for infection?

Methods: From the longitudinally maintained orthopaedic oncology surgical database at our institution, we identified 69 patients who had been treated for a clinical diagnosis of PJI at the knee between 1993 and 2015. We excluded 32% (22) of patients who did not meet at least one of the major criteria of the Musculoskeletal Infection Society (MSIS) for PJI, 3% (2) of patients who underwent immediate amputation, 3% (2) of patients who had a follow-up time of < 24 months, and 7% (5) of patients who did not have a primary tumor of the distal femur or proximal tibia. The study consisted of 38 patients, of whom eight underwent two-stage revision, 26 underwent DAIR, and four underwent extended DAIR (removal of all segmental components but with retention of stems and components fixed in bone) for their initial surgical procedure. To be considered free of infection, patients had to meet MSIS standards, including no positive cultures, drainage, or surgical debridement for a minimum of 2 years from the last operation. Factors associated with time-dependent risk of infection relapse, clearance, amputation, and patient survival were analyzed using Kaplan-Meier survivorship curves and the log-rank test to compare factors. Association of demographic and treatment factors was assessed using chi-square and Fisher exact tests.

Results: Continuous infection-free survival at 5 years was 16% (95% CI 2% to 29%) for patients undergoing DAIR compared with 75% (95% CI 45% to 100%) for patients undergoing two-stage revision (p = 0.006). The median (range) number of total surgical procedures was 3 per patient (1 to 10) for DAIR and 2 (2 to 5) for two-stage revision. Twenty-nine percent (11 of 38) of patients eventually underwent amputation. Survival without amputation was 69% (95% CI 51% to 86%) for DAIR compared with 88% (95% CI 65% to 100%) for two-stage revision at 5 years (p = 0.34). The cumulative proportion of patients achieving infection-free status (> 2 years continuously after last treatment) and limb preservation was 58% (95% CI 36% to 80%) for patients initially treated with DAIR versus 87% (95% CI 65% to 100%) for patients first treated with two-stage revision (p = 0.001).

Conclusion: Infection control was better with two-stage revision than DAIR. The chance of eventual clearance of infection with limb preservation was better when two-stage revision was chosen as the initial treatment. However, the loss to follow-up in the two-stage revision group would likely make the true proportion of infection control lower than our estimate. Our experience would suggest that the process of infection eradication is a complex and difficult one. Most patients undergo multiple operations. Nearly one-third of patients eventually underwent amputation, and this was a serious risk for both groups. While we cannot strongly recommend one approach over the other based on our data, we would still consider the use of DAIR in patients who present with acute short duration of symptoms (< 3 weeks), no radiographic signs of erosion around fixed implants, and organisms other than Staphylococcus aureus . We would advocate the extended DAIR procedure with removal of all segmental or modular components, and we would caution patients that there is a high likelihood of needing further surgery. A prospective trial with strict adherence to indications may be needed to evaluate the relative merits of an extended DAIR procedure versus a two-stage revision.

Level of evidence: Level III, therapeutic study.