Purpose: To examine the diagnostic effect of immediate on-site cytopathologic evaluation of tissue core touch preparations in computed tomographic (CT)-guided coaxial needle biopsy.
Materials and methods: The authors reviewed the records of 430 patients and included 413 patients with 432 biopsies (210 in the lungs, 222 in other locations). Each time the guiding needle was moved to a new location in the lesion for tissue core procurement with the cutting needle, it represented a new session. Core specimen touch preparations were obtained and immediately evaluated on-site for specimen adequacy and preliminary diagnosis. New sessions were considered and/or executed in the case of inconclusive cytopathologic readings. Each final diagnosis was reached according to the pathology report showing "positive for malignancy," "negative but with a specific diagnosis," or "unclear" for further surgical resection specimen or radiologic follow-up.
Results: The accuracy of on-site cytopathologic examination of touch preparations was 80.6% for the first session and increased to 85.9% and 86.3%, respectively, for the second and third sessions. The corresponding accuracies for biopsy were 88.2%, 93.8%, and 94.9%. The overall accuracy was 97.1% for lesions in the lungs and 92.8% for lesions at other sites. More biopsy sessions were deemed necessary in lesions measuring 2 cm or smaller (P = .0045). During CT-guided lung biopsy, 10 patients (4.8%) had major complications that necessitated chest tube insertion.
Conclusions: The diagnostic accuracy of CT-guided needle biopsy can be increased through repeated sessions with immediate on-site cytologic evaluation, especially for lesions of 2 cm or smaller and those from nonpulmonary sites.