With the advent of new diagnostic technologies and the fear of low diagnostic accuracy, there has been a reduction in the use of fine needle aspiration (FNA) for diagnosis of breast lesions. The objectives of the present study were to establish the diagnostic accuracy and complications associated with FNA of breast lesions.
We retrospectively reviewed FNA of breast lesions done in a tertiary referral center in Mexico City from 1999 through 2001. We analyzed demographic, radiologic and pathological variables in order to establish diagnostic accuracy of FNA. The gold standard was considered the histopathologic study of the specimen or the clinical follow-up of benign lesions. Categorical variables were analyzed with the X2 method and statistical significance was considered at p < 0.05.
We identified 300 patients with a mean age of 50 years (range, 20-86). Fifty-three percent were postmenopausal. In ninety-three percent of patients, FNA was performed in the outpatient clinic. Nonpalpable lesions were aspirated under ultrasound guidance. Mean size of the lesion was 2.27 cm (range 0.7-10 cm). Thirty-one percent of patients had definite diagnosis of malignancy. The only variables associated with cancer diagnosis were the BI-RADS category and the presence of palpable adenopathy. Cancer was diagnosed in 6.5% of lesions categorized as BI-RADS 0-3 compared to 56.2% for lesions BI-RADS 4-5 (p < 0.0001). Positive predictive value and specificity of FNA were 100%. Sensitivity, negative predictive value and accuracy were 82.6%, 92.8% and 94.7%, respectively. The most common diagnosis of false negative lesions were lobular and papillary carcinomas. There were no significant complications associated with FNA.
Diagnostic accuracy of FNA for breast lesions is very high with minimal complications. Positive predictive value of 100% allows to establish therapy based on its results.