Purpose: Infection following implant-based breast reconstruction (IBR) remains a major cause of increased health care costs, morbidity and reconstruction failure. Rates of infection from 5% to 35% have been reported. While the microbiology of breast implant infections has been studied, however, associations of specific clinical variables with common pathogens are not well understood. The aim of this study is to identify clinical characteristics associated with specific microorganism in patients with breast implant infections, in order to better guide treatment and empirical antimicrobial therapy. Methods: We retrospectively reviewed all patients who underwent IBR performed at our institution from 2007 to 2017 to identify cases of implant infection. Surgical site infection (SSI) was defined using the Centers for Disease Control and Prevention criteria. Demographic characteristics, comorbidities, surgical data, clinical presentation, laboratory, and microbiology data were collected. Comparative analysis of continuous variables and categorical variables were performed using the Mann-Whitney-Wilcoxon and Fisher’s exact tests, respectively. A value of p < 0.05 was considered significant. Results: A total of 2153 patients underwent IBR during the study period, of which 198 met case definition for SSI. Implant removal was required in 140 (70%) and positive culture data were identified from 105 (75%) clinical samples. There were 30 (29%) infections with Gram positive bacteria (GPB) and 75 (71%) with Gram negative bacteria (GNB). Of the GPB, Staphylococcus spp. and Streptococcus spp. were most common, whereas Pseudomonas spp. and Serratia spp. were most common among the GNB. In the GNB infections, obesity and autoimmune diseases were more prevalent (46.8% vs 17.8%; p=0.003, 75% vs 25%; p=0.032, respectively). Radiation therapy and adjuvant chemotherapy were more common in the GPB infections (86.7% vs 64.4%; p=0.027, 89.7% vs 63.3%; p=0.009, respectively). Compared to GPB, patients with GNB infections were more likely to present within 30 days after surgery (55% vs 15%; p=0.001). Erythema on exam was more commonly reported in those with GPB (76.1% vs 50%; p=0.047) whereas wound drainage (46.7% vs 18.2%; p=0.005) and necrosis (66.7% vs 23%; p=0.006) were more frequent in GNB infections. In GPB infections the median white blood cell count was 10.5 x 1000/mm3 vs. 7 x 1000/mm3 in the GNB group; p=0.004. Conclusion: Patients with GNB infections after IBR presented earlier and were more likely to present with wound drainage and necrosis of the skin compared to GPB infections. In contrast, infections caused by GPB were more likely to present with breast erythema and higher white cell counts. These clinical characteristics should be considered when selecting empiric antimicrobial therapy in IBR infections.
Read full abstract