Comparison of methods to diagnose lymphoedema among breast cancer survivors: 6-month follow-up

Summary

One of the more problematic and dreaded complications of breast cancer is lymphoedema. Our objective was to determine the prevalence of lymphoedema 6-months following breast cancer treatment and to examine potential risk factors among a population-based sample of women residing in South-East Queensland (n=176). Women were defined as having lymphoedema if the difference between the sum of arm circumferences (SOAC) of the treated and untreated sides was >5 cm (prevalence=11.9%) or >10% (prevalence=0.6%), their multi- frequency bioelectrical impedance (MFBIA) score was ≥3 standard deviations above the reference impedance score (prevalence=11.4%), or they reported ‘yes’ when asked if arm swelling had been present in the previous 6 months (prevalence=27.8%). Of those with lymphoedema defined by MFBIA, only 35% were detected using the SOAC method (difference > 5 cm), while 65% were identified via the self-report method (i.e., respective sensitivities). Specificities for SOAC (difference > 5 cm) and self-report were 88.5% and 76.9%, respectively. When examining associations between presence of lymphoedema and a range of characteristics, findings also varied depending on the method used to assess lymphoedema. Nevertheless, one of the more novel and significant findings was that being treated on the non-dominant, compared to dominant, side was associated with an 80% increased risk of having lymphoedema (MFBIA). Our work raises questions about the use of circumferences as the choice of measurement for lymphoedema in both research and clinical settings, and assesses MFBIA as a potential alternative.