CT-guided biopsy is a minimally invasive diagnostic method of evaluating musculoskeletal lesions. Other options include incisional and excisional biopsy with the possibility of intraoperative frozen section. The clinician's decision to order a CT-guided biopsy requires an understanding of the likelihood that this biopsy will affect treatment This requires an understanding of both diagnostic yield and accuracy. Furthermore, the clinical utility of a biopsy is affected by factors other than the yield and accuracy as the clinical setting may render a technically diagnostic biopsy unhelpful.
A retrospective review of the electronic record at an orthopedic oncology referral center identified all patients who had undergone CT-guided percutaneous needle biopsy of mus-culoskeletal lesions after being evaluated by an orthopedic oncologist in clinic over a period of 5 years. 53 CT-guided biopsies of bone lesions and 16 CT-guided biopsies of soft tissue lesions were identified. The diagnostic yield (rate of obtaining tissue from which the pathologist could report a diagnosis) and clinical utility (rate at which biopsy results guided treatment decisions) were calculated and statistically compared.
The overall diagnostic yield of CT-guided bone biopsies was 94% (50 of 53 biopsies) and the clinical utility was 70% (37 of 53 biopsies). In the first 2 years of the study the diagnostic yield was 95% (21 of 22 biopsies) and the clinical utility was 86% (19 of 22 biopsies). In the remaining 3 years the diagnostic yield was 91% (28 of 31 biopsies) and the clinical utility was 58% (18 of 31 biopsies). This decrease in clinical utility over time was statistically significant (p = 0.01). Suspicion of metastasis resulted in a diagnostic yield of 100% (11/11) and a clinical utility of 91% (10/11). Suspicion of primary tumor resulted in a diagnostic yield and clinical utility of 93% (39/42) and 67% (28/42), respectively. This difference in clinical utility was statistically significant (p = 0.02). The diagnostic yield of CT-guided soft tissue biopsies was 75% (12 of 16 biopsies) and the clinical utility was 69% (11 of 16 biopsies). The diagnostic yield was significantly lower for soft tissue biopsy than bone biopsy (p = 0.01). There was no relationship between the rate of diagnostic biopsies and the evaluating pathologist or the location of the lesion within the body.
CT-guided biopsy is useful in the diagnosis of musculoskeletal lesions, however, its clinical utility is substantially lower than its diagnostic accuracy and yield due to a significant rate of diagnostic biopsies that fail to guide treatment, particularly when a primary lesion is suspected. The disparity in clinical utility based on preoperative suspicion of metastasis was even greater in our study than previously shown. CT-guided percutaneous needle biopsy is much more likely to guide treatment in the setting of suspected bone metastasis as opposed to biopsies of suspected primary bone lesions and soft tissue lesions.