Conventional Factors Do Not Explain Higher Mastectomy Rate in US Compared to UK

EVANSTON, Indiana—Analysis of traditional factors that correlate with choice of mastectomy or breast-conserving surgery failed to explain the significantly higher mastectomy rates in the United States compared to the United Kingdom seen in the international Arimidex (anastrozole), Tamoxifen, Alone or in Combination (ATAC) trial. ATAC investigator Gershon Y. Locker, MD, noted the striking difference "in the mastectomy rates between the two largest accruing nations in ATAC-the United Kingdom with a 42% mastectomy rate and the United States with a 51% mastectomy rate. This is a 21% increased risk of having a mastectomy if enrolled in the United States compared to the United Kingdom."

Dr. Locker is associate professor of medicine at Northwestern University and head of the division of hematology/oncology at Evanston Northwestern Healthcare, in Evanston, Indiana. Noting that ATAC offered a unique opportunity to examine breast surgery practices around the world, he reported that the overall mastectomy rate in ATAC was 47%, a figure that was derived from 381 treatment centers in 21 countries. The percentage of women who underwent mastectomy varied widely among the countries represented in ATAC, Dr. Locker noted, from a low of 20% to a high of 97%. These differences may be explained in part by sampling errors due to small sample size.

Univariate Predictive Factors

Among ATAC patients overall, univariate analysis identified the predictive factors for mastectomy as:

• tumor size 2 cm or larger;

• nodal involvement (especially 4 or more nodes);

• poorly differentiated tumor;

• age 70 or older;

• receiving adjuvant chemotherapy; and

• being treated in a center that had enrolled 40 or fewer patients in the ATAC trial (considered a "rough correlate" of surgeon experience).

Factors predicting breast conservation were:

• unknown nodal status;

• estrogen-positive or progesterone-positive tumor;

• body weight greater than 70 kilograms, and

• being treated in a center that had enrolled more than 40 patients.

Compared to participants in the UK, more patients from the US were older, had received chemotherapy, and were being treated at a center enrolling 40 or fewer patients, all factors favoring mastectomy.

"But for every other factor, either there was no difference between the United States and the United Kingdom or the difference favored doing greater breast conservation in the United States rather than mastectomy," Dr. Locker reported.

Multivariate Analysis

After performing a multivariate analysis, the researchers found that the size of a center’s enrollment no longer correlated with type of surgery. The strongest predictor of mastectomy was tumor size, Dr. Locker noted, with an odds ratio of greater than 3.0. Having four or more positive nodes or being older than age 70 produced odds ratios greater than 2.0. Other statistically significant factors predicting mastectomy were 1 to 3 positive nodes, moderate to poorly differentiated tumor, age 60 to 69, and receipt of adjuvant chemotherapy.

"However, taking all of these factors into account in a multivariate analysis, being from the United States remained an independent predictor of having a mastectomy compared to the United Kingdom, with an odds ratio of 1.44 and a 95% confidence interval of 1.26 to 1.64," Dr. Locker said.

"Given the desirability of breast conservation, increased educational efforts should be supported to promote its greater use," he concluded. "But to end on a somewhat more optimistic note, it is possible to achieve higher breast conservation rates in the United States, as is seen in the 1996-1997 Surveillance, Epidemiology, and End Results (SEER) data for Connecticut, which achieved a breast conservation rate of 64%."