Abstract
BACKGROUND: Despite some studies on Gram-negative bacteria as difficult to treat pathogens in periprosthetic joint infections, there are no detailed analyses on Serratia periprosthetic joint infections. As such, we present two cases of Serratia periprosthetic joint infections and summarize all known cases to date in the course of a PRISMA criteria-based systematic review. CASE PRESENTATION: Case 1: a 72-year-old Caucasian female with Parkinson's disease and treated breast cancer developed periprosthetic joint infection caused by Serratia marcescens and Bacillus cereus, following multiple prior revisions for recurrent dislocations of her total hip arthroplasty. Two-stage exchange was performed, and the patient remained free of Serratia periprosthetic joint infection recurrence at 3 years. Case 2: an 82-year-old Caucasian female with diabetes and chronic obstructive pulmonary disease presented with a chronic parapatellar knee fistula after undergoing multiple failed infection treatments at external clinics. After performing two-stage exchange and gastrocnemius flap plastic for combined Serratia marcescens and Proteus mirabilis periprosthetic joint infection, the patient was released without any signs of infection, but was subsequently lost to follow-up. REVIEW: a total of 12 additional Serratia periprosthetic joint infections were identified. Merged with our two cases, the mean age of 14 patients was 66 years and 75% were males. Mean length of antibiotic therapy was 10 weeks with ciprofloxacin most commonly used (50%). Mean follow-up was 23 months. There was a total of four reinfections (29%), including one case of Serratia reinfection (7%). CONCLUSIONS: Serratia is a rare cause of periprosthetic joint infection affecting elderly with secondary diseases. While the overall reinfection rate was high, the risk of Serratia periprosthetic joint infection persistence was low. Treatment failure in patients may be attributable to the host, rather than the Serratia periprosthetic joint infection itself, thus challenging current concepts on Gram-negatives as a uniform class of difficult-to-treat pathogens. LEVEL OF EVIDENCE: Therapeutic level IV.