PET/CT-Guided Tissue Sampling in Patients With a Failed or Inconclusive CT-Guided Procedure: Outcomes and Contributing Factors.

Abstract

CT-guided tissue sampling is a very effective tool. However, false-negative results are obtained when regions such as necrotic core or surrounding reactive fibrosis and inflammation are sampled. PET/CT-guided sampling can circumvent these limitations. The aim of this study was to analyze the effectiveness of PET/CT-guided sampling in patients with at least 1 instance of failed or inconclusive CT-guided procedure and factors determining the accurate sampling and complications. One hundred eleven patients were prospectively included. After feasibility analysis in a diagnostic F-FDG PET/CT, sampling was performed in 106 patients (45 women, 61 men; mean age, 48.09 ± 15.42 years; biopsy in 80 and fine-needle aspiration cytology [FNAC] in 26 patients), using robotic arm and a lower IV injection dose of 74 to 111 MBq (2-3 mCi) F-FDG. In all patients, final check scans revealed needle at the target site. Using planned needle path as reference, deviations in first check scan were measured. Patient (n = 30) and respiratory motion (n = 57) were also recorded. Accurate lesion targeting was achieved in 81 cases (63 positive lesions, 12 confounding lesions, and 7 inadequate samples). Lesion was missed in 5 instances, and blood/necrotic tissue sampled in 19. Overall F-FDG-avid lesions were accurately targeted in 77.36% of patients (86.25% [biopsy] + 50% [FNAC]). Significant variables affecting targeting were needle gauge, deviation from intended entry point, procedure duration, procedure type, and patient movement. Using binomial regression, the significant parameters were procedure type (biopsy vs FNAC; odds ratio [OR], 5.916; P = 0.002), patient movement (OR, 0.275; P = 0.023), and procedure duration (OR, 1.195; P = 0.011). Overall complication rate was 21.70%, with 4.71% major complications. It was dependent on target depth (mean depth, 69.74 ± 20.29 mm [complications] vs 47.18 ± 22.60 mm; P < 0.001). Positive correlation was seen between the target depth and distance of needle from the intended target (Spearman ρ = 0.307; P = 0.001). In 28 procedures, the physician was asked to wear a pocket dosimeter, who received a mean dose of 2.52 (SD, 3.10) μSv. PET/CT-guided sampling should be considered where CT-guided biopsy has failed or is inconclusive. The outcome is impacted by needle gauge and patient movement, and complication rate is dependent on target depth.