Stereotactic core biopsy of the breast: results of one-year follow-up of 101 patients.

Stereotactic core biopsy (SCB) is being used as a cost-effective alternative to needle localized biopsy (NLB). However, an area of concern is the potential for sampling error, with sparse surgical data available concerning follow-up and failure rates. We therefore reviewed our results in patients undergoing SCB for mammographically detected breast abnormalities. Between January 1994 and February 1995, 128 patients underwent SCB. Average age was 56.4 years. Nine patients (7.0%) had histologic evidence of malignancy, with 111 (86.7%) benign diagnoses requiring no further initial intervention. Eight patients (6.3%) proceeded directly to NLB, five because of technical failure of SCB and three because of suspicious initial histology. One of the latter patients had ductal carcinoma in situ. The remaining 111 SCB patients were evaluated at 6 months and 1 year by mammographic and physical examination. Ten patients were lost to follow-up. Of the remaining 101 patients, 98 (97%) had stable mammograms and normal physical examinations. Three patients (3.9%) required subsequent NLB due to progression of the mammographic lesion. Two cases were histologically benign, and 1 patient had ductal carcinoma in situ adjacent to the previous SCB biopsy site. An additional patient underwent NLB for a new radiographic abnormality at a separate location in the ipsilateral breast, which was invasive ductal carcinoma. SCB appears to be an effective alternative to NLB for the majority of patients deemed eligible. Careful mammographic follow-up is warranted for these patients given the small, but real, possibility of sampling error.

Differentiation of benign from malignant pancreatic masses by endoscopic ultrasound.

BACKGROUND: It is often difficult to determine whether a mass in the pancreas is benign or malignant. The goal was to evaluate whether endoscopic ultrasound (EUS) can reliably establish whether a mass is benign or malignant. METHODS: One hundred five patients with possible pancreatic tumors were referred for EUS. Those who were found to have a lesion suspicious for carcinoma and did not have a known malignancy also underwent EUS-guided FNA. RESULTS: A mass suspicious for cancer was identified in 73 patients, whereas inflammatory changes or a normal pancreas was noted in 32 patients. Four of the latter 32 patients were subsequently found to have cancer. EUS-guided FNA was performed on 47 of the 73 patients with a suspicious mass and was read as cancer in 27 patients, atypia in 10 patients, and benign in 10 patients. All 10 patients with atypia were subsequently confirmed to have cancer, and 6 of the 10 patients with a benign FNA were proved to have a tumor at surgery. EUS could differentiate the lesion as malignant with a sensitivity of 95%, specificity 88%, positive predictive value 95%, and negative predictive value 88%. CONCLUSIONS: Radial array EUS is helpful in supporting or refuting a diagnosis of cancer in a patient with a pancreatic mass. Although EUS-guided FNA can confirm the diagnosis, a negative FNA should not preclude exploration when clinically indicated.