Abstract

Core biopsy (CB) is increasingly popular for assessing solid lesions in children. To date, pediatric literature is limited regarding factors contributing to diagnostically inadequate or inaccurate CB. Therefore, we retrospectively examined radiologic/pathologic factors associated with adequacy/accuracy of CB in pediatric patients. A search of the surgical pathology database for CB between January 2007 and December 2014 yielded 134 CB from 99 patients. Age, sex, anatomic site of lesion, CB diagnosis, and final diagnosis were acquired from the electronic medical record. Image guidance modality, lesion size, and CB sampling device were obtained from radiology records. CB hematoxylin and eosin slides were reviewed for fragmentation, percentage of fibrosis, and percentage of necrosis. Overall, CB length was measured using cellSens software and a DP71 camera. Groups were compared using 2-sided homoscedastic Student's t tests; 87.3% (117/134) CB were diagnostic; final diagnosis was available for 105 cases, with a concordance rate of 80.0% (84/105). Image guidance modality, lesion site (extremity vs nonextremity), and CB needle gauge did not significantly differ between diagnostic versus nondiagnostic CB or concordant versus discordant CB. Diagnostic CB had less necrosis and fibrosis than did nondiagnostic CBs (6.8% vs 29.7%, P = .0002 and 10.3% vs 29.1%, P = .0006). Nondiagnostic and discordant CB were more likely to be from bony lesions than soft tissue ( P = .01 and P = .0248). CB is valuable for diagnosing solid lesions in children, with good overall diagnostic rates regardless of lesion size, location, or imaging modality used for biopsy. Nondiagnostic and discordant CB were more often obtained from bony lesions; sampling via open biopsy may be more useful in that setting. Nondiagnostic and discordant CB have more necrosis and fibrosis, suggesting that on-site evaluation of CB tissue viability-for example, by touch imprint or fine needle aspiration-may be useful in further enhancing CB utility.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call