Back to 2016 Annual Meeting

Introduction:
Infections are a recognized complication of breast reconstructions and may require unplanned reoperations, including washouts, implant exchanges, and implant explantation. The authors sought to characterize factors associated with successful treatment of breast reconstruction infections as well as the prevalence of and outcomes of tertiary reconstruction.
Methods:
Patients undergoing procedures consistent with breast reconstruction since 2003 with follow-up in our institution’s medical record system were included. Reconstruction failure was defined as 1) the need to start reconstruction over 2) the need to regress a stage of reconstruction (i.e. permanent implant back to tissue expander), or 3) the need to change modalities of reconstruction due to infection. Chi-squared and independent t-tests to assess differences between groups.
Results:
Four hundred thirty three patients were included, of whom 206 were Caucasian (47.6%) and 256 had private insurance (59.1%). The average age was 50.6±14 years and average BMI 27.2±5.7. 110 had two or more co-morbidities (25.4%). The majority of patients (282) underwent implant-based reconstruction (65.1%). 88 patients developed infections - 23 had only superficial infections, 54 had deep infections, and 11 developed both. 71 of the patients with infections had undergone implant-based reconstruction (25.2% of implant reconstructions) versus 17 autologous reconstruction patients (11.3% of autologous patients, p=0.001).
A regression controlling for factors significant on univariate analysis (Caucasian ethnicity, number of comorbidities, smoking, radiation, chemotherapy, implant-based reconstruction, dehisce, seroma, and implant exposure) was constructed. Implant-based reconstruction increased odds of infection by 90.6% (p=0.004), dehiscence increased odds by 65.1% (p=0.034), and implant exposure by 165% (p=0.003).
46 patients had reconstructive failure (53% of those with infections). There were no differences between patients whose reconstructions failed and those whose didn’t based on BMI, co-morbidities, or cancer and adjuvant therapy variables. In a regression controlling for implant exposure and implant-based reconstruction, implant- based reconstruction was found to increase chances of reconstruction failure by 316% (p=0.003). There were no differences in number of tissue expander (p=0.395) or implant (p=0.667) exchanges between patients whose reconstructions were salvaged and those whose were not. Patients with reconstruction failures were not any less likely to complete their reconstructions (p=0.128). Thirty patients underwent tertiary reconstruction (65% of patients with reconstructive failure). On multivariate regression, each year of age decreased likelihood of tertiary reconstruction by 10.3% (p=0.016) while a history of dehiscence increased tertiary reconstruction by 193% (p=0.042).
Conclusions:
Fewer than half of infected reconstructions were salvaged in this sample. Implant reconstruction increased odds of infections and reconstruction failures on multivariate analysis, reinforcing that this risk is something that must be stressed to potential implant-based reconstruction candidates. Fortunately, patients with
reconstructive failure were not any less likely to complete their reconstructions, likely due to the high incidence and success of tertiary reconstruction. The optimal approach to reconstruction salvage, particularly of implant-based reconstructions, must be explored further in a prospective manner, as better preventative and treatment strategies would affect a large number of women.