Jasmohan S. Bajaj, Bradley Reuter, Kara Walter, Scott W. Biggins, Julien Bissonnette, Christopher F. Rose, Jennifer C. Lai, Michael D. Leise, Patrick S. Kamath, Puneeta Tandon, James Wade.
Reliability of HE characterization and evaluation is essential for accurate management and conduct of multi-center trials but a systematic intervention of trainees and non-trainees is lacking. METHODS: US & Canadian GI trainee/non-trainee practitioners were shown standard- ized simulated patient videos during the same sitting. The videos demonstrated cirrhotic pts with HE below stage 2 (normal/covert), and overt stages (2, 3, & 4). Therespondents were asked to identify the HE stage and answer questions about management after each video. RESULTS: 108 respondents (62 trainees, 46 non-trainees) from 7 centers(77 from four 4 US & 31 from 3 Canadian centers) were included. Trainees included 18 1st yr, 16 2 nd yr, 12 3 rd yr, 14 4 th yr fellows & 2 residents. Non-trainees included 41 consultants & 5 mid-level practitioners. HE stages (Table): A higher % of respondents were better at correctly diagnosing stages ≥2 compared to stages <2 (91 vs 64%, p<0.001). HE sub-stages were identified similarly between trainees and non-trainees with a trend towards better “normal” identification in non-trainees(P=0.13).Management: Most respondents would send pts with stages < 2 rather than ≥2 HE for cognitive testing (70 vs 20%, p<0.001). The reverse was seen for ordering ammonia levels, higher in stages ≥2 than stage<2 HE (36 vs 11%, p<0.001). A higher percent of non-trainees tested ammonia levels in stage <2 HE compared to trainees (16 vs 7%, p=0.03). A higher percent of non-trainees than trainees would prescribe low protein diets in stage ≥2 HE(18 vs 8%, p=0.007). Most respondents would treat pts with stages ≥2 with lactulose (58 vs 93%, p<0.001) & rifaximin (14 vs 50%, p<0.001). More trainees would use rifaximin for stages ≥2 than non-trainees (68 vs 30%, p<0.001). CONCLUSION: In this multi-center study, North American GI practitioners demonstrated poor reliability in lower stages but not stages ≥2 HE. There were significant variations in ordering MHE testing, & lactulose/rifaximin and relatively few respondents order ammonia levels or a low-protein diet. Overt HE ≥stage 2 remains a reliable benchmark of diagnosis across training levels but there is a need for more standardized algorithms for HE management.