Search

Objective: To investigate the impact of contrast dispersion pattern/location during lumbar CT-guided transforaminal epidural steroid injection (TFESI) and experience of the performing radiologist on therapeutic outcome. Materials and methods: In this single-center retrospective cohort study, two observers analyzed contrast dispersion during CT-guided TFESI of 204 patients (age 61.1 ± 14 years) with discogenic unilateral single-level L4 or L5 radiculopathy. The contrast dispersion pattern was classified as "focal," "linear," or "tram-track"; the location was divided into "extraforaminal," "foraminal," or "recessal." Pain was assessed before and 4 weeks after treatment using a numerical rating scale (0, no pain; 10, intolerable pain). Additionally, the patient global impression of change (PGIC) was assessed. The TFESI was performed by musculoskeletal radiologists (experience range: first year of musculoskeletal fellowship training to 19 years). Contrast pattern/location and radiologist's experience were compared between "good responder" (≥ 50% pain reduction) and "poor responder" (< 50%). A p-value < 0.05 was considered to be statistically significant. Results: Overall, CT-guided TFESI resulted in a substantial pain reduction in 46.6% of patients with discogenic radiculopathy. The contrast dispersion pattern and location had no effect on pain relief (p = 0.75 and p = 0.09) and PGIC (p = 0.70 and p = 0.21) 4 weeks after TFESI. Additionally, the experience of the radiologist had no influence on pain reduction (p = 0.92) or PGIC (p = 0.75). Regarding pre-interventional imaging findings, both the location and grading of nerve compression had no effect on pain relief (p = 0.91 and p = 0.85) and PGIC (p = 0.18 and p = 0.31). Conclusion: Our results indicate that neither contrast agent dispersion/location nor the experience of the radiologist allows predicting the therapeutic outcome 4 weeks after the procedure. [ABSTRACT FROM AUTHOR]

Deep learning-based MRI diagnosis of internal joint derangement is an emerging field of artificial intelligence, which offers many exciting possibilities for musculoskeletal radiology. A variety of investigational deep learning algorithms have been developed to detect anterior cruciate ligament tears, meniscus tears, and rotator cuff disorders. Additional deep learning-based MRI algorithms have been investigated to detect Achilles tendon tears, recurrence prediction of musculoskeletal neoplasms, and complex segmentation of nerves, bones, and muscles. Proof-of-concept studies suggest that deep learning algorithms may achieve similar diagnostic performances when compared to human readers in meta-analyses; however, musculoskeletal radiologists outperformed most deep learning algorithms in studies including a direct comparison. Earlier investigations and developments of deep learning algorithms focused on the binary classification of the presence or absence of an abnormality, whereas more advanced deep learning algorithms start to include features for characterization and severity grading. While many studies have focused on comparing deep learning algorithms against human readers, there is a paucity of data on the performance differences of radiologists interpreting musculoskeletal MRI studies without and with artificial intelligence support. Similarly, studies demonstrating the generalizability and clinical applicability of deep learning algorithms using realistic clinical settings with workflow-integrated deep learning algorithms are sparse. Contingent upon future studies showing the clinical utility of deep learning algorithms, artificial intelligence may eventually translate into clinical practice to assist detection and characterization of various conditions on musculoskeletal MRI exams. [ABSTRACT FROM AUTHOR]