Abstract
Purpose To define operator learning curve inflection points for prostatic artery embolization (PAE) and their impact on technical efficiency, clinical outcomes, and adverse events. Materials and Methods Between May 2013 and May 2021, 296 consecutive patients with moderate-to-severe lower urinary tract symptoms, urinary retention, or gross hematuria from benign prostatic hyperplasia underwent PAE by an interventional radiologist without prior PAE-specific experience. Operator learning curves plotted procedure time, fluoroscopy time, contrast volume, and embolic endpoint data against sequential procedure number. Multiple regression analysis evaluated for improvements in these parameters, with segmented linear regression to detect learning curve inflection points. Linear and logistic regression evaluated for learning curve impacts on 6-month clinical outcomes and 90-day adverse events. Results No baseline patient characteristic varied over the series apart from decreasing pre-procedural gland volume (P < 0.01). Multiple regression analysis demonstrated experience-dependent improvements in procedure time, fluoroscopy time, and contrast volume (P < 0.01), with corresponding learning curve inflection points at 76 (P < 0.01), 78 (P < 0.01), and 73 (P = 0.10) procedures. Embolic endpoints did not vary with experience (P > 0.05). Post-procedure reductions in International Prostate Symptom Score (21.5 +/- 6.2 to 6.7 +/- 4.7), Quality of Life score (4.5 +/- 1.2 to 1.3 +/- 1.2), post-void residual (190 +/- 203 to 97 +/- 148 mL), and gland volume (142 +/- 97 to 76 +/- 47 mL) were substantial (P < 0.01) but did not vary with experience (P > 0.05), nor did adverse event frequency/severity (P > 0.05).Conclusion Operator technical efficiency plateaued after 73-78 PAE procedures. Clinical improvements were sub-stantial and adverse event frequency/severity low, and neither varied with experience. Operators without prior PAE-specific experience may perform PAE safely and effectively from the outset.
