Core needle biopsy is the preferred technique for evaluating breast masses and abnormal mammographic findings. The frequency of detection of noninvasive lobular lesions by core needle biopsy is increasing. Historically, the diagnosis of lobular carcinoma in situ has been considered a risk factor for the development of invasive carcinoma, and treatment has consisted of careful clinical follow-up with or without chemopreventive therapeutic agents such as tamoxifen citrate. We retrospectively reviewed core needle biopsy material with the primary diagnoses of lobular carcinoma in situ, atypical lobular hyperplasia, and lobular neoplasia in conjunction with clinical and radiographic findings to make recommendations as to when excision may be merited. We searched our database for core needle biopsy cases with lobular carcinoma in situ, atypical lobular hyperplasia, and lobular neoplasia as the primary diagnosis. Microcalcifications had been sampled with a stereotactically guided, 11 G Mammotome biopsy device, and masses had been sampled with an ultrasound guided, 18 G core needle. Glass slides were reviewed and histological parameters assessed. Mammographic findings were reviewed, and clinical information was obtained from the medical record. When available, excisional biopsy material was reviewed. The 2337 breast core needle biopsies performed from January 1995 to December 2001 included 35 (1.5%) with classic lobular carcinoma in situ (14), lobular neoplasia (4), and atypical lobular hyperplasia (17) as the primary diagnosis. Twelve of these 35 cases (34%) had histological evidence of microcalcifications directly associated with the lobular carcinoma in situ, lobular neoplasia, atypical lobular hyperplasia. Radiologic review revealed 21 calcifications, 6 ultrasonographic masses, and 8 mammographic masses and/or architectural distortions. Excisional biopsy had been performed in 17 cases (49%). In six cases diagnosed as in situ on core needle biopsy, excisional biopsy revealed invasive carcinoma. All of these patients had radiographically detectable masses. Eleven cases had excisional biopsies that showed histology similar to that of the core needle biopsies. The most important predictor of invasive carcinoma on excision was a synchronous mass lesion. Lobular carcinoma in situ involving adenosis and lobular carcinoma in situ with pagetoid spread on core needle biopsies did not show a histologically more aggressive lesion on excision and, therefore, may not require additional surgery. Histologically identified calcifications were associated with lobular lesions 34% of the time; however, their presence inside an in situ lobular lesion did not portend worse pathology on re-excision and should not be a criterion for excision. Based on these findings, we recommend excisional biopsy of lobular carcinoma in situ, atypical lobular hyperplasia or lobular neoplasia only when it is associated with a synchronous mass lesion.