Return To Abstract Listing

Efficacy, Education, Administration, Informatics

E2633. Technique and Tissue: Diagnostic Rate of Image-Guided Biopsies and Associated Factors

Lee J,  Chen F,  Cen S,  Romero M. University of Southern California Keck School of Medicine, Los Angeles, CA

Address correspondence to M. Romero (

Objective: Resources are limited at our large county hospital. Percutaneous image-guided biopsies are less invasive and less expensive than open operative biopsies. However, nondiagnostic biopsies can delay treatment and increase costs. While most radiologists within the department maintain individual procedure logs, our overall diagnostic rate was unknown. Our goal was to determine our diagnostic rate and identify technical factors associated with nondiagnostic biopsies.

Materials and Methods: All percutaneously image-guided biopsies performed at a large urban county hospital over an 8-month period, excluding bone and breast biopsies, were reviewed. Technique (fine needle aspiration vs core), imaging modality used (ultrasound vs CT), needle gauge, and number of passes for each biopsy were recorded. Pathologic results were obtained and final surgical pathology was recorded where available. Images from the nondiagnostic biopsies were retrospectively reviewed by two experienced radiologists to evaluate needle placement and lesion characteristics.

Results: A total of 456 biopsies were performed. Of these, 399 were diagnostic (87.5%) and 57 were nondiagnostic (12.5%). Results from four biopsies were unavailable. Diagnostic biopsy did not correlate with technique (core versus fine needle aspiration) or needle size. The number of passes was statistically significant with a median number of three passes performed in diagnostic biopsies and four passes performed in nondiagnostic biopsies. Of the 57 nondiagnostic biopsies, one lymph node actually appeared benign in retrospect, one lymph node was poorly visualized due to technique as repeat biopsy with better visualization was successful, four large masses were necrotic on preprocedure CT (two masses were biopsied twice), one lesion was heavily calcified, and needle positioning was poor for three lesions. Limited imaging on one biopsy precluded assessment of needle positioning. The lack of diagnosis in the remaining 44 biopsies could not be explained based on imaging. Diagnostic rates were highest for solid organs and superficial structures (95.5% overall). Lower rates were seen for abdominopelvic masses, chest, muscle, head and neck, omental, and retroperitoneal lesions (80.1% overall).

Conclusion: Our overall diagnostic rate for percutaneous image-guided biopsies was 87.5%. Of the nondiagnostic biopsies, only five were definitely due to suboptimal technique. The choice of core versus fine needle aspiration and needle size did not affect diagnostic rates. Nondiagnostic biopsies required one more pass than diagnostic biopsies which may reflect more difficult cases. Biopsies of large, necrotic masses may be less successful, despite relative ease of accessing the lesion. Diagnostic rates were highest for solid organs and superficial structures. Lower diagnostic rates for other biopsies may be secondary to a combination of factors, including more technically challenging biopsies and target characteristics. Further study may help identify means of improving diagnostic rates in these biopsies.