Large screening mammography study restores faith in CAD

A new study based on nearly a quarter million mammograms suggests screening mammography with computer-aided detection is more sensitive than double reads. The findings contradict a key study published last year questioning its effectiveness.

CAD's potential for yielding too many false positives remains controversial. Radiologists argue its misuse drives up recall rates and, with them, the number of unwarranted biopsies and overall mammography costs. CAD came under close scrutiny after a study published in The New England Journal of Medicine in April 2007 concluded the technology could worsen the interpretation of mammograms instead of improving it.

A study published online this week by Dr. Matthew Gromet, section chief of breast imaging at Charlotte Radiology in North Carolina, could fan the flames again. Charlotte Radiology performed double reads of all screening mammograms until 2003, when it shifted to single reading with CAD. Gromet reviewed screening mammograms interpreted at the institution from 2001 through 2005 to cover evenly both time periods.

The study compared single reading, with and without CAD, and double reading in correlation to biopsy data. It found that CAD enhanced the sensitivity of single readers with only a small increase in the recall rate. Findings will be published in print in the April issue of the American Journal of Roentgenology.

"As a large single-institution program with relative consistency of methods, patients, radiologists, and data collection, we believe that results from our experience before and after CAD implementation could provide additional evidence about the usefulness of this technology," Gromet said.

Gromet retrospectively reviewed 231,221 screening mammograms interpreted at Charlotte Radiology by nine experienced mammographers between January 2001 through December 2005. Mammographers double-read 112,413 (48.6%) studies, while 118,808 (51.4%) were single-read with CAD.

The first interpretation of the double-reading process yielded a recall rate of 10.2%. Sensitivity and positive predictive values were 81.4% and 4.1%, respectively, while the biopsy-confirmed cancer detection rate was 4.12 per 1000. After the second read, the recall rate and sensitivity jumped - to 11.9% and 88%, respectively - as did the cancer detection rate (to 4.46 per 1000). The benefit came at the expense of 140 additional biopsies, however, resulting in a drop in positive predictive value to 3.7%.

Single-reading with CAD yielded a recall rate of 10.6%, sensitivity of 90.4%, positive predictive value of 3.9%, and cancer detection rate of 4.20 per 1000. CAD increased sensitivity and recall rate compared with single interpretation without CAD and reduced the recall rate compared with double reading. All findings were statistically significant (p<0.0001).

The study's potential weaknesses include its retrospective nature, as well as patient selection bias and radiologists' changing reading skills. The large number of malignancies and cases interpreted with double-reads and CAD, however, lend credence to Gromet's research, he said. The average number of CAD studies was 13,201 compared with 821 in the NEJM study.

A randomized study comparing double reading with single reading with CAD would provide the optimal assessment of these two competing strategies, Gromet said.

"With staffing and cost constraints limiting the use of double reading in the U.S., CAD appears to be an effective alternative that provides similar, and potentially greater, benefits," Gromet said. "Based on a historical methodology similar to that used by Fenton and colleagues, our study had fewer variables and yielded different results."

For more information from the Diagnostic Imaging archives:

Breast CAD marketing exceeds real results

Radiologists prove hasty in dismissing breast CAD

Breast MRI wins allies, while CAD survives attack

CAD scores well in digital mammography

Radiologists pan study critical of computer-aided detection