Authors: CE Marculescu, EF Berbari, JM Steckelberg, AD Hanssen and DR Osmon
Title: Medical Treatment and Outcome of Clinically Unsuspected Prosthetic Joint Infection (PJI) diagnosed by Multiple Positive Cultures Following Revision Arthroplasty
Address: Mayo Clinic and Foundation, 200 first St SW, Rochester, MN 55905
Purpose: The objective of this retrospective cohort study is to analyze the medical treatment and outcome of clinically unsuspected PJI diagnosed by multiple positive cultures following revision arthroplasty.
Methods: Records of all cases of THA and TKA infections, using a strict case definition, diagnosed at our institution during 1/95-1/98 were reviewed by the primary author. Treatment failure was defined as relapse, reinfection, presence of positive pathology, sinus tract or clinical, laboratory and /or radiological findings suggestive of PJI at any time after surgical therapy. Kaplan- Meier survival methods was used to determine the cumulative incidence of treatment failure at 5 years following PJI diagnosis.
Results: 408 episodes of PJI were identified during the study period. We identified thirteen episodes (3.1%) in 13 patients of clinically unsuspected PJI diagnosed by multiple positive cultures following revision arthroplasty. The median age of the cohort was 65 years (range: 54-83). Ten patients (76%) were males. Sixty one percent (8/13) and 39% (5/13) underwent revision THA or TKA respectively. Five of 8 THA revisions had only partial or complete revisions of the acetabular component. Diabetes mellitus, malignancy and rheumatoid arthritis were present in 7.6, 23, and 15% of cases. A preoperative sedimentation rate, synovial fluid cultures and intraoperative pathology were normal or negative in 5/7 (71%), 2/2 (100%), and 13/13 (100%) respectively. No patient had CRP performed or sinus tract present on examination. Revision arthroplasty was performed for pain and/or loosening in 92.4% of cases and instability (7.6%) All patients had 2 or more cultures positive from tissue culture. Pathogens identified were SCN 46% ( 6/13), Propionibacterium sp 30% (4/13), viridans group streptococci 8% (1/13), enterococci 8% (1/13), polymicrobial infection 8% (1/13). Antibiotic impregnated PMMA cement was used at the time of revision arthroplasty in 46% of cases. Intravenous antimicrobial therapy was administered in 10/13 (76%) patients for a median of 35 days (5-42). B-lactams were used in 8/13 (61%) patients. Chronic oral antimicrobial suppression was used in 8/13 (61%) patients. The median length of chronic suppression was 696 days (range 58-1645 days). Two patients received no antimicrobial therapy, and one received chronic oral suppression only. The 5 year cumulative incidence of treatment failure was 12.5% (95% CI: 1.8%-53.4%). One treatment failure occurred 2.11 years following PJI diagnosis and consisted of reinfection with same organism (SCN) but with a different antibiogram.
Discussion: Clinically unsuspected PJI diagnosed by multiple positive cultures following revision arthroplasty is an infrequent clinical entity at our institution. Pathogens causing this clinical entity are typically of low virulence. The outcome of patients clinically unsuspected PJI diagnosed by multiple positive cultures following revision arthroplasty is very good utilizing a variety of different medical treatment strategies including chronic oral antimicrobial suppression.
Significance: Optimal medical treatment strategies for patients with PJI diagnosed by multiple positive cultures following revision arthroplasty have yet to be defined and prospective clinical studies are warranted.