Abstract P1-09-08: Risk factors for lymphedema are dependent on level of axillary surgery

Abstract

Abstract Background: Differentiation of lymphoedema (LE) risk factors for those who have undergone a sentinel node biopsy (SNB) from those who have undergone axillary node dissection (AND) is not considered, even though the incidence rates for the two are vastly different. In addition, events women are typically cautioned against have not been well investigated. Methods: A prospective study was conducted in which women were recruited and assessed prior to surgery, and then seen within 4 weeks following surgery, and at 6, 12 and 18 months following surgery. Women were categorised as having LE if their bioimpedance interlimb ratio exceeded previously established thresholds. Following post-surgery assessment, women were asked to complete weekly diaries regarding events that occurred in the previous week. Risk factors were broadly grouped into demographic, lifestyle, breast cancer treatment-related, arm swelling-related, and post-surgical activities (eg, airplane travel). Crude association between each potential predictor and presence of arm swelling was then identified using unadjusted logistic regression. Those variables with P<0.2 at this initial screen were considered for inclusion in a logistic regression model. The final multivariable model retained all variables with P<0.1 or odds ratio> 2.0, taking into account biological plausibility. The final multivariable models were developed without and with consideration of the presence of swelling in the first year. Results: 450 women (SNB group: 241; AND group: 209) were recruited and attended the final assessment; a subgroup of 243 women, of whom 112 had AND completed >70% of the weekly diaries. The incidence of LE for the SNB group was 3.3% (n=8) and 18.2% (n=38) for the AND group. The unadjusted risk factors for LE at 18 months for SNB included high BMI and absolute body weight, living alone and presenting at diagnosis with three or more other medical conditions. The final model for the SNB group included a high BMI and not living with a partner, explaining 21.3% of the variance. Inclusion of post-operative swelling in the model explained 48.4% of the variance. The unadjusted risk factors for LE at 18 months for AND included being older, low education, Stage 3, high number of nodes removed and involved, and radiotherapy to the axilla, and receiving taxane chemotherapy. The final risk factors model for the AND group included clinical stage 3, being older, low education, and receiving taxane-based chemotherapy, explaining 20.4% of the variance. The addition of any swelling within the first 6 months following surgery explained 36.8% of the variance. Notably none of the factors related to air travel, arm trauma, medical procedures (eg, blood pressure, injections, blood drawn on the affected side), or exercise which women are typically cautioned against differentiated women who had and did not get LE at 18 months. Conclusion: Advice to women undergoing SNB should differ to that provided to those undergoing AND, and for both, we should not be burdening them with range of behaviors to avoid. Importantly, for women at high risk, periodic assessment in the first year should occur to identify and manage any arm swelling. Acknowledgement: Cancer Australia and NBCF. Citation Format: Sharon L Kilbreath, Kathryn M Refshauge, Jane M Beith, Leigh C Ward, Owen A Ung, James R French, Louise Koelmeyer, Katrina Kastania, Jasmine Yee. Risk factors for lymphedema are dependent on level of axillary surgery [abstract]. In: Proceedings of the Thirty-Seventh Annual CTRC-AACR San Antonio Breast Cancer Symposium: 2014 Dec 9-13; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2015;75(9 Suppl):Abstract nr P1-09-08.