Effect of estrogen and estrogen-progestin replacement regimens on mammographic breast parenchymal density.

PURPOSE: Hormone replacement therapy (HRT) may increase the mammographic density with a possible reduction in the sensitivity or specificity. If so, the benefit of mammographic screening in women using HRT could be compromised. We evaluated the hypothesis that HRT regimens have differential effects on the mammographic density depending on treatment regimens or on age. PATIENTS AND METHODS: Among 31,498 Swedish women who received mammographic screening, we selected 554 women who started HRT after the first examination and who were current users at the second, and 554 age-matched women who had never received HRT. Mammograms were examined in a blinded review. The changes in density between the two examinations, graded as moderate or weak reduction, no change, or weak, moderate, or substantial increase, were assessed. We studied four HRT regimens-estradiol compounds only, estradiol compounds cyclically or continuously combined with progestins, and weak estrogens-and used descriptive statistics and logistic regression to analyze the association between HRT and density change. RESULTS: Density increased in 10% and 28% of women who received estradiol compounds with cyclically or continuously combined progestins, respectively, but in only 3% of unexposed women. Logistic regression analyses showed an elevated risk of a density increase (relative risk RR = 3.6; 95% confidence interval CI, 1.6 to 7.7) in women who received cyclically combined regimens or continuously combined regimens (RR = 12.4; 95% CI, 6.3 to 24.4) compared with unexposed women. Women > or = 50 years of age had even stronger associations; RRs in women on estradiol only, the cyclically combined and the continuously combined regimens were 32.2 (95% CI, 3.9 to 267.5), 21.9 (95% CI, 1.9 to 251.5), and 176.9 (95% CI, 22.8 to 1,372.7), respectively. CONCLUSION: HRT with estradiol-progestin regimens, especially continuously combined, may increase the mammographic density in a substantial proportion of women.