Percutaneous Ultrasound-Guided Biopsy for Sampling Neck Lesions.

Importance: Studies suggest that percutaneous ultrasound-guided biopsy may be effective for diagnosing lymphoma, but the achievable effectiveness has not been reported. A clinical workflow eliminating percutaneous vs surgical biopsy referral biases and allowing measurement of up-front imaged-guided biopsy effectiveness is needed.

Objective: To assess diagnostic sufficiency of core-needle biopsy (CNB) and fine-needle aspiration (FNA) of neck lesions in a setting where all patients are referred for CNB or FNA prior to excisional biopsy.

Design, setting, and participants: This prospective cohort study was conduced at a quaternary care academic medical institution. Biopsy results of consecutive patients undergoing percutaneous ultrasound-guided neck biopsies and head/neck surgeon excisional neck biopsies performed from June 1 to August 31, 2024, were analyzed.

Exposures: CNB and FNA.

Main outcomes and measures: Primary outcomes included diagnostic yield overall and for lymphoma. Secondary outcomes included rate of subsequent excisional biopsy and yield stratified by biopsy type, location, and clinician type.

Results: Of 158 biopsies among patients (mean [SD] age, 60.0 [16.5] years; 82 [51.9%] female), which were performed by 7 neuroradiologists and 5 physician assistants, diagnostic yield was 94.3% (149 of 158 biopsies), with CNB yield higher than FNA (97.9% vs 88.5%; difference, 9.4 percentage points [pp]; 95% CI, -0.4 to 19.2 pp) and no missed lymphoma diagnoses. Diagnostic yield in cases with lymphoma as prebiopsy differential diagnosis was 97.4% (74 of 76 biopsies): 20 lymphoma, 27 inflammatory, 17 solid tumors, 10 benign lymphoid tissue, and 2 nondiagnostic. No immediate complications and 1 infection occurred. Only 2 patients underwent subsequent excisional biopsy, 1 due to pending pathology on the surgery date and the other to confirm a negative CNB result. Two additional excisional biopsies were performed during the study period, both by request due to insufficient tissue from outside hospital specimens. Neuroradiologist yield was slightly higher than physician assistant yield, with a wide confidence interval (95.3% vs 92.2%; difference, 3.1 pp; 95% CI, -6.7 to 13.0 pp). Flow cytometry analysis detected lymphoma signatures in 13 of 20 (65.0%) lymphoma specimens.

Conclusions and relevance: In this cohort study, up-front percutaneous ultrasound-guided biopsy was shown to be a highly effective and safe method for diagnosing neck lesions, with CNB providing a high diagnostic yield that may generally obviate the need for excisional biopsy, even when lymphoma is a prebiopsy differential diagnosis.