Radiation Doses to ERCP Patients Are Significantly Lower with Experienced Endoscopists

Abstract

Purpose: Patients undergoing ERCP receive a significant dose of radiation which may increase their risk of developing future cancers, especially in young patients. Mean effective doses are comparable to the radiation received from an abdominal/pelvic CT. Radiation doses to patients during ERCP correlate most closely with fluoroscopy time (FT). We hypothesized that FT is lower with experienced ERCPists. Methods: We analyzed the prospectively collected ERCP Quality Database to determine if the number of years of ERCP experience and number of ERCP's performed in the last year correlated with FT, adjusting for clustering of complexity of cases and fellow involvement. Multivariate mixed linear regression models of log-transformed FT were used for the analysis to determine adjusted percentage differences in FT. Results: 8,399 ERCPs by 68 providers from 6 countries had complete data. ERCP experience ranged from 1 to 36 years and 10 to 940 ERCPs performed in the last year. Median FT was 3.0 minutes (min), (range: 0.1-60.0 minutes). Cases associated with greater FT included hilar stent placement, stricture dilation, biliary brushing, extraction of biliary stone >10 mm, pancreatic stone extraction, and need for pre-cut sphincterotomy (difficult cases median=4.9 min, less difficult cases=2.5 min). There was significant interaction between annual volume of ERCPs and case difficulty for FT (p<0.0001). Adjusting for clustering within endoscopists and fellow involvement in the case, ERCPist experience was inversely associated with FT in the less difficult cases. ERCPists who performed less than 100 ERCPs per year had 104% increase in FT (p<0.0001), and those who performed 100-200 ERCPs per year had 27% increase (p<0.0001) compared to those performing greater than 200 ERCPs. Additionally, for every 10 years of additional ERCP experience, FT was associated with a 21% decrease in FT (p<0.0001). For difficult cases, ERCP experience was also inversely associated with FT (<100 ERCPs: 59% increase, p<0.0001; 100-200 ERCPs: 11%, p=0.005; every 10 years of ERCP: 20% decrease, p<0.0001). For both difficult and less difficult cases, fellow involvement was associated with greater FT (involvement ≤50% of difficult case: 83% increase, p<0.0001; ≤50% of less difficult case: 112% increase, p<0.0001). If the case was performed entirely by the fellow, the differences were marginal (9 or 14% increase). Conclusion: Fluoroscopy time is lower when ERCP is performed by experienced endoscopists, adjusting for clustering by case difficulty and fellow involvement. Both years performing ERCP and annual volume of ERCP contribute independent effects. These findings may have important ramifications for radiation-induced cancer risk.