Risk factors for lymphedema in a prospective breast cancer survivorship study: the Pathways Study

Abstract

Objective: To determine the incidence of breast cancer-related lymphedema (BCRL) during the early survivorship period as well as demographic, lifestyle, and clinical factors associated with BCRL development.

Design: The Pathways Study, a prospective cohort study of breast cancer survivors with a mean follow-up time of 20.9 months.

Setting: Kaiser Permanente Northern California medical care program.

Participants: We studied 997 women diagnosed from January 9, 2006, through October 15, 2007, with primary invasive breast cancer and who were at least 21 years of age at diagnosis, had no history of any cancer, and spoke English, Spanish, Cantonese, or Mandarin.

Main outcome measure: Clinical indication for BCRL as determined from outpatient or hospitalization diagnostic codes, outpatient procedural codes, and durable medical equipment orders.

Results: A clinical indication for BCRL was found in 133 women (13.3%), with a mean time to diagnosis of 8.3 months (range, 0.7-27.3 months). Being African American (hazard ratio, 1.93; 95% confidence interval, 1.00-3.72) or more educated (P for trend = .03) was associated with an increased risk of BCRL. Removal of at least 1 lymph node (hazard ratio, 1.04; 95% confidence interval, 1.02-1.07) was associated with an increased risk, yet no significant association was observed for type of lymph node surgery. Being obese at breast cancer diagnosis was suggestive of an elevated risk (hazard ratio, 1.43; 95% confidence interval, 0.88-2.31).

Conclusions: In a large cohort study, BCRL occurs among a substantial proportion of early breast cancer survivors. Our findings agree with those of previous studies on the increased risk of BCRL with removal of lymph nodes and being obese, but they point to a differential risk according to race or ethnicity.

Figure 1

Cumulative incidence function estimates ^a for occurrence of BCRL in the Pathways Study cohort, as of July 31, 2008 ^a Derived from the Kaplan-Meier method ^b Transient and persistent BCRL diagnosis from KP electronic data sources, as defined by outpatient clinic or hospitalization diagnosis, BCRL procedure, or DME order anytime after breast cancer diagnosis; n=133 total BCRL events, mean time (8.3 months), range (0.7–27.3 months), and interquartile range (3.8–11.0 months)