Streamlining Clinical Breast Examination

There has been a sharp decline in the practice of screening clinical breast examination (CBE) during periodic health examinations. In 1987, 95% of women who had a screening mammogram had a CBE within 6 months ( 1 ) ; by 2004, that rate was approximately 50% ( 2 , 3 ) . This decline in the practice of CBE decreases opportunities to detect breast cancer by screening and probably decreases the proficiency with which CBE is performed, which may explain why misinterpretation of CBE results is the leading cause of delayed diagnosis of breast cancer ( 4 ) .

Abandonment of screening CBE in favor of mammography is driven by confidence in mammography ( 1 , 2 ) and also, we believe, because the multiple steps taught for traditional CBE (i.e., palpation of nodes and visual inspection and manual palpation of the breast) ( 5 ) must compete with other tasks for the scarce time allotted for most periodic health examinations ( 6 ) .

On the basis of finding only three (1%) of 286 cancers by inspection for nipple retraction, Mahoney and Csima ( 7 ) reasoned that it would be appropriate to streamline screening CBE to the single “simple, brief, efficient” step of palpation while the patient is supine (hereafter called supine palpation) if so few cancers would be missed. (That is, skin and nipple retraction are difficult to observe in a supine patient, but these are so rarely the only sign of cancer that one can forego looking for them.)

To reevaluate this proposal, we studied (with institutional review board approval) a HIPAA-compliant database of breast cancer patients that was derived from prospective independent databases. These databases were maintained by three surgeons (WHG, NAG, and FMD) in separate practices to facilitate analysis of their quality of care. After excluding three cancers found by screening MRI and 13 cancers found during prophylactic or cosmetic surgery, 1189 cases of invasive cancer and 212 cases of pure ductal carcinoma in situ remained for analysis. The median patient age at diagnosis was 53 years (range = 18–98 years) and tumors were 0.1–18.5 cm in diameter (median = 1.5 cm).

A total of 238 patients (17%) had negative mammograms ( Table 1 ). Of these, 227 had palpable masses and six had visible nipple changes during supine palpation. Two patients had nipple retraction as the only sign of cancer; in both instances, the patient found the retraction herself. Two of the three patients who had adenopathy only detected the swollen node themselves. Thus, only one (0.07%) of 1401 cancers was found during a step of CBE other than supine palpation.

Overall, skin and nipple retraction were rare (being detected in only 48 [3%] of the 1401 patients); inspection for skin and nipple retraction by a clinician never identified cancer when both supine palpation and mammograms were negative, and four of five instances of isolated retraction or adenopathy were discovered by the patient herself.

Although supine palpation alone would have missed fewer than 0.1% of breast cancers, endorsing even a remote possibility of missing cancer is awkward. However, a quicker, more efficient CBE might reverse the trend to drop this examination from periodic health examinations. Given that more than 17% of cancers are detected by CBE alone, it is important to maintain, or even increase, the frequency of CBE as part of the overall periodic health examination.

We therefore recommend that screening CBE be streamlined to the single step of supine palpation so that clinicians can concentrate on the examination skill most likely to detect breast cancer.

References