Effect of obesity on screening mammography: outcomes analysis of 88,346 consecutive examinations.

Abstract

We determined differences in the rates of recall, biopsy, and cancer detection for screening mammography as a function of adiposity. Eighty-eight thousand three hundred forty-six consecutive screening mammography examinations were performed from April 1985 to August 1997. Patient weights were normalized to ideal weight correcting for height and were subdivided into adiposity cohorts including underweight by greater than 10%; ideal weight +/- 10%; overweight by 11-24%; overweight by 25-39%; and overweight by greater than 40%. The rates of recall, biopsy, cancer detection, and cancer stage were calculated and were analyzed using the chi-square test for trend. Cancer size was analyzed using linear regression analysis. Reliable (p < 0.05) and meaningful differences were seen between cohorts of increasing adiposity for rates of recall, biopsy, and cancer detection. An increase in recall rate occurred with progressively increasing adiposity (3.88%, 4.89%, 5.11%, 5.47%, 5.55% [p < 0.0001]). The rate of biopsy also increased with increasing adiposity (0.98%, 1.31%, 1.35%, 1.59%, 1.65% [p < 0.0002]), as did the rate of screening-detected cancer (number of cases of cancer per 1000 women screened) (3.74, 4.29, 5.34, 4.70, 6.04 [p < 0.015]). Finally, increased adiposity also correlated with increased median cancer size (p < 0.02) and with more advanced stage at diagnosis (p = 0.046). Increasing adiposity correlates with progressive increases in the rates of recall, biopsy, and cancer detection for women undergoing screening mammography. Increasing adiposity also correlates with increased cancer size and stage, providing further support for obesity as a risk factor for breast cancer.