Goals/Purpose:
According to the American Society of Plastic Surgeons, approximately 63,000 breast reduction surgeries were performed in the United States in 20111. The high prevalence of reduction mammaplasty underscores the finding that non-surgical interventions have not been demonstrated to provide lasting relief of symptomatic macromastia in all cases2. Additionally, numerous studies have demonstrated increased patient satisfaction and better quality of life following reduction mammaplasty3-6.
The impact of BMI on reduction mammaplasty outcomes has been evaluated, but the results have not been conclusive. Several studies have found that increased BMI leads to increased complications7-9, while other studies have concluded that there is no difference in complication rates between ideal-weight and obese patients10-12. In order to investigate the impact of BMI on reduction mammaplasty for a generalized patient population, we utilized the American College of Surgeons (ACS) National Surgical Quality Improvement Database (NSQIP). Surgical patients across many specialties and procedures are selected at random from participating hospitals and 30-day outcomes are reported. Between 2006 and 2010, over 1.3 million surgical procedures were captured at over 240 hospitals. One hundred thirty-five independent variables describing patient demographics, comorbidities, and preoperative condition are recorded; predefined outcomes, including post-operative morbidity, reoperations, and mortality are tracked by specially trained nurses and subjected to audit13. With thorough data collection and reporting procedures and a broadly selected patient population, this database provides a powerful tool to examine the impact of BMI on reduction mammaplasty outcomes. We present the largest multi-institutional study of BMI and breast reduction to date, providing generalizable information for patient education and clinical decision-making.
Methods/Technique:
Retrospective cohort analysis was conducted with reduction mammaplasty procedures recorded in the NSQIP database. Outcomes of interest included 30-day post-operative morbidity, reoperation, and mortality.
Demographic and preoperative statistics were tabulated for obese (BMI at least 30) and non-obese (BMI less than 30) and compared between groups with chi-squared or ANOVA tests for dichotomous and continuous variables, respectively. Outcomes of interest were also tabulated and compared with chi-squared or Fisher's exact test as appropriate. Two-tailed p-values below .05 were considered significant. Multiple logistic regression was performed to compare complication rates between obese and non-obese patients with correction for confounders.
Results/Complications:
A total of 2,492 female breast reduction patients with BMI information were identified. Information about patient demographics and rates of diabetes and smoking are reported in Table 1. Obese and non-obese patients were not statistically different in their mean age; obese patients were more likely to be diabetics and smokers than non-obese patients.
Complication, reoperation, and mortality rates are shown in Table 2. Obese patients experienced nearly double the overall complication rate of non-obese patients (5.64% vs. 3.25%, p = .007). Surgical complications were also nearly twice as high in obese patients (4.99% vs. 2.71%, p = .004). Wound infections demonstrated a significant difference between cohorts (3.83% in obese patients vs. 2.25% in non-obese patients, p = .025). Graft failure was too rare an event to allow statistically significant conclusions, but wound disruption was almost three times more prevalent in the obese cohort (1.30% in obese patients, .45% in non-obese patients, p = .034).
While surgical complication rates were higher in the obese cohort, medical complication rates were not statistically different between groups, either in aggregate (.58% in obese patients, .54% in non-obese patients, p = .901) or for any individual complication. The total incidence of medical complications was very low (less than 1%). Reoperation rates were also similar between groups (2.17% in obese patients, 1.98% in non-obese patients, p = .748). There were no deaths in either cohort.
Regression analysis (Table 3) showed that, after controlling for confounders, obese patients had approximately 55% greater odds of developing a complication than non-obese patients (OR 1.547, 95% CI 1.021-2.344).
Conclusion:
Reduction mammaplasty has emerged as a therapeutic option for women suffering from chronic pain due to breast hypertrophy, with evidence suggesting that reduction mammaplasty is the only viable option for permanent relief of symptomatic macromastia in some cases. With a generalized patient population, large sample size, and robust data collection, NSQIP provides a powerful tool to examine the impact of BMI on reduction mammaplasty outcomes. We have found that obesity increases risk for post-operative morbidity in reduction mammaplasty patients. However, the majority of the complications were superficial surgical site infections. With a total complication rate of less than 5%, and zero mortalities in this cohort of 2,492 women, breast reduction mammaplasty continues to be a safe option in obese patients; an informed consent discussion should include information about increased risk of non-lethal surgical complications.
