Identifying risk factors for lymphedema in patients treated for breast cancer has become increasingly important, given the current lack of standardization surrounding diagnosis and treatment. Reports on the association of body mass index (BMI) and weight change with lymphedema risk are conflicting. We sought to examine the impact of pre-operative BMI and post-treatment weight change on the incidence of lymphedema. From 2005 to 2011, 787 newly diagnosed breast cancer patients underwent prospective arm volume measurements with a Perometer pre- and post-operatively. BMI was calculated from same-day weight and height measurements. Lymphedema was defined as a relative volume change (RVC) of ≥ 10 %. Univariate and multivariate Cox proportional hazards models were used to evaluate the association between lymphedema risk and pre-operative BMI, weight change, and other demographic and treatment factors. By multivariate analysis, a pre-operative BMI ≥ 30 was significantly associated with an increased risk of lymphedema compared to a pre-operative BMI <25 and 25- <30 (p = 0.001 and p = 0.012, respectively). Patients with a pre-operative BMI 25- <30 were not at an increased risk of lymphedema compared to patients with a pre-operative BMI <25 (p = 0.409). Furthermore, a cumulative absolute weight fluctuation of 10 pounds gained/lost per month post-operatively significantly increased risk of lymphedema (HR: 1.97, p = < 0.0001). In conclusion, pre-operative BMI of ≥ 30 is an independent risk factor for lymphedema, whereas a BMI of 25- <30 is not. Large post-operative weight fluctuations also increase risk of lymphedema. Patients with a pre-operative BMI ≥ 30 and those who experience large weight fluctuations during and after treatment for breast cancer should be considered at higher-risk for lymphedema. Close monitoring or early intervention to ensure optimal treatment of the condition may be appropriate for these patients.