National Trends in Immediate Breast Reconstruction: An Analysis of Implant-Based Versus Autologous Reconstruction After Mastectomy

2020 
Many factors affect access to immediate breast reconstruction (IR) after mastectomy. The present study was performed to assess trends, outcomes, and predictors of IR techniques using a nationally representative cohort. The 2009–2014 National Inpatient Sample (NIS) was used to identify adult women who underwent inpatient mastectomy with IR. Patients were compared by type of reconstruction: implant-based IR versus autologous reconstruction (AR). AR was classified as a microsurgical or pedicled flap procedure. Incidence, outcomes, and predictors were assessed using Chi squared univariate tests and multivariable logistic regression analyses. Of 194,073 women who underwent IR, 136,668 (70.4%) received implant-based IR and 57,405 (29.6%) received AR. Of those who underwent AR procedures, 31,336 (54.6%) received microsurgical flaps and 26,680 (46.5%) received pedicled flaps. Utilization of deep inferior epigastric perforator (DIEP) flaps increased significantly (28.6–42.5% of AR, P < 0.001). Predictors of AR were Black race [adjusted odds ratio (AOR) = 1.46, P < 0.001], lower Elixhauser Comorbidity Index (AOR = 1.25, P < 0.001), private insurance (AOR = 1.07, P = 0.030), body mass index (BMI) ≥ 30 kg/m2 (AOR = 1.38, P < 0.001), urban teaching hospital designation (AOR = 1.77, P < 0.001), and high hospital volume (AOR = 3.11, P < 0.001). Similar factors were associated with the use of microsurgical flaps. AR and microsurgical flaps were associated with higher rates of acute inpatient complications, resource utilization and length of stay (LOS) compared with implant-based IR and pedicled flaps, respectively. Implant-based IR remains the most common type of IR, although rates of microsurgical AR are on the rise. Follow-up of complications, costs, and quality-of-life measures may show that AR provides long-term high-value care despite upfront morbidity, cost, and use of hospital resources.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    38
    References
    4
    Citations
    NaN
    KQI
    []