Transradial Access Learning Curve Evaluated as Part of Neuroendovascular Fellowship Training

July 22, 2021—In a study published in Operative Neurosurgery, Al Saiegh et al found that the use of transradial access (TRA) can be incorporated into neuroendovascular training without increasing complications, procedure time, or contrast use.

The study aimed to compare the learning curves of transfemoral access (TFA) to TRA using diagnostic angiogram procedural data prospectively collected from two neuroendovascular fellows who trained at Thomas Jefferson University from July 2018 to June 2019. Angiograms were supervised by senior faculty with expertise in both TRA and TFA.

Procedural success was defined as successful catheterization of the target vessels and completion of angiography without resorting to TFA. Procedure time was extracted from the electronic medical record and defined as the time between sheath insertion and removal, and in cases of TFA, the time included sheath placement and femoral sheath removal.

Of 293 diagnostic angiograms analyzed, 57.7% (169) were TRA and 42.3% (124) were TFA. The overall complication rate was 1.4% (two groin hematomas, one wrist hematoma, one access site infection). The crossover rate to TFA was 2.1%, most commonly due to vasospasm or vessel tortuosity.

Overall mean procedure time was 59.55 +/- 26.88 minutes for TRA and 51.89 +/- 19.82 minutes for TFA (P = .005). Overall median procedure time was 52 minutes. In multivariate analysis, age > 75 years (odds ratio [OR], 3.59; 95% CI, 1.11-11.60; P = .032) and > four target vessels (OR, 3.10; 95% CI, 1.75-5.45; P < .0001) were predictors of increased procedure time. Access was not significantly associated with prolonged procedure time after controlling for fellow, quarter of training, targeted vessel, and patient age and sex (P = .075).

Mean fluoroscopy time was 9.95 +/- 7.10 minutes for TRA and 8.71 +/- 4.89 minutes for TFA (P = .077). Median fluoroscopy time was 8.3 minutes. TRA (OR, 1.7; 95% CI, 1.001-3.03; P = .05) and > four target vessels (OR, 11.53; 95% CI, 5.60-23.74; P < .0001) were predictors of prolonged fluoroscopy time in multivariate analysis, but the individual fellow was not a significant predictor (P = .21).

Mean radiation dose was 3.39 +/- 3.91 cGy in TFA and 2.65 +/- 3.68 in TRA (P = .104), and median radiation dose was 1.36 cGy. The median radiation dose was 1.36 cGy. Only advancing quarter of training was a predictor of decreased radiation after controlling for fellow, number of target vessels, and patient age and sex (OR, 0.61; 95% CI, 0.49-0.75; P < 0001).

Overall mean contrast dose was 73.7 +/- 38.1 mL in TRA and 72.1 +/- 49.0 in TFA (P = .76), and median contrast dose was 60 mL. After controlling for fellow, access, quarter of training, and patient sex, increasing number of injected vessels (OR, 14.60; 95% CI, 3.40-63.16; P < .0001) and age > 65 years (OR, 0.57; 95% CI, 0.33-0.99; P = .05) were significant predictors of increased contrast use.

The maximum Youden index was used to determine the number of cases required to predict trainee improvement beyond the median of each case proficiency measure. The analysis showed that technical proficiency was achieved after 60 TFA and 95 TRA cases based on fluoroscopy time, 52 TFA and 77 TRA cases based on procedure time, and 53 TFA and 64 TRA cases based on contrast volume. This demonstrates that the radial learning curve can be overcome during neurovascular fellowship training at a high-volume center, noted the investigators.