CT-fluoroscopy-guided lumbar epidural steroid injections: trainee participation, radiation dose, and procedural time

Compare patient radiation, procedure time, and immediate post-procedural pain relief for CT-fluoroscopy (CTF)-guided lumbar epidural steroid injections (ESIs) performed by attendings alone versus with trainee participation.

This retrospective single-institution study analyzed 371 consecutive CTF-guided lumbar ESIs (2018-2025) performed by attendings (n = 93) or by trainees under direct attending supervision (n = 278). Patient and procedural characteristics and outcomes were compared using Welch t-tests and chi-squared tests. Total dose length product (DLP) was separated into planning-scan and procedural components. Linear mixed-effects models were adjusted for age, sex, injection approach, planning CT omission, and academic year, with a random intercept for supervising attending.

Total DLP was higher for trainee injections (53.7 ± 52.0 vs 31.7 ± 40.6 mGy∙cm, p < 0.001), driven by the planning CT (37.7 ± 47.3 vs 16.4 ± 32.2 p < 0.001); procedural DLP was similar (15.6 ± 10.3 vs 15.3 ± 11.9 mGy∙cm, p = 0.83). After adjustment, trainee participation was not associated with total or procedural DLP (geometric mean ratio 0.99 for both; 95% CIs 0.81-1.21 and 0.84-1.17). Proceduralist group explained 7% of total DLP variance; planning CT omission and supervising attending added 38% and 10%, respectively. Planning CT omission, practiced by two of seven attendings, declined from 53% of procedures in 2019-2020 to none after 2022. Trainee procedures took longer (20.0 ± 6.3 vs 18.3 ± 4.2, p = 0.005). Immediate pain reduction was similar (3.5 ± 2.9 vs 3.4 ± 2.6 points, p = 0.81).

Trainee participation in CTF-guided lumbar ESIs was associated with modestly longer procedures but not increased procedural radiation. Higher total radiation with trainees reflected increasing use of planning CT and attending practice variation rather than trainee performance.