Outcomes of ERCP in Prone versus Supine Position: A Large Multi-center Study

Introduction: Endoscopic retrograde cholangiopancreatography (ERCP) may be performed with the patient in either prone or supine position. Meta-analyses comparing ERCP in prone versus supine position show heterogeneous results, possibly skewed due to higher prone case volumes. Our study aimed to compare a more equal distribution between ERCP performed in prone versus supine position from multi-center high volume tertiary care centers to assess technical success and procedural outcomes. Methods: We performed a multi-center retrospective study on all adult consecutive patients undergoing ERCP in either prone or supine position over a 1-year period. IRB was approval was obtained. Demographics and procedural data were collected via manual chart review. Patient position during ERCP and other variables were adjusted for fellow presence, Schutz score for procedural difficulty, ASA classification and type of anesthesia used. Odds ratios (OR) and 95% confidence intervals (95% CI) pertaining to patient position during ERCP were reported from binary logistic regression models. All analyses were performed using SAS 9.4. Results: A total of 960 adult patients met our inclusion criteria, of which there were 465 prone and 495 supine patients, with median age of 64 years, and 51.5% were male. A majority of ERCPs in prone position 83.9% were performed with endotracheal intubation, compared to 42.6% in supine position. ERCP in prone position had a statistically significant higher rate of successful cannulation of the desired duct (98.7% vs 95.2%), shorter procedure time (22 mins vs 32 mins), but longer fluoroscopy time (5 mins vs 1.9 mins) compared to ERCP in supine position. (Table 1) Prone patients were less likely to experience a procedural complication (1.3% vs 3.4%) but were more likely to have intraprocedural cardiovascular instability (64.5% vs 41.4%) compared to supine patients. There was no statistical difference between 30-day mortality rates between the 2 groups. (Table 1). Conclusion: The debate between prone versus supine patient positioning for ERCP is ongoing. Our multi-center study suggests that prone positioning during ERCP has higher desired duct cannulation rate, shorter procedure time and lower complication rate; however, with higher patient risk for intraprocedural cardiovascular instability and longer fluoroscopy time, as compared to supine positioning. The final determination of patient positioning should be individualized to each patient’s comorbidities and the endoscopists’ expertise. Table 1. - Frequency and Percent of Patient, ERCP, and Outcome Characteristics Across ERCP Start Position, with Unadjusted and Adjusted Comparisons of Outcomes for Prone ERCP Position vs Supine Patient & ERCP Characteristics Prone (n=465) Supine (n=495) Median Age (IQR) (year) 64 (51, 74) 65 (53, 78) Male Sex,n (%) 259 (55.7) 235 (47.5) Fellow Present During ERCP Procedure,n (%) 75 (16.1) 358 (72.3) Anesthesia Type,n (%) Natural Airway 75 (16.1) 284 (57.4) Intubation 390 (83.9) 211 (42.6) Schutz Score for Procedural Difficulty,n (%) 1-Simple Diagnostic 65 (14.0) 29 (5.9) 2-Simple Therapeutic 62 (13.3) 126 (25.2) 3-Complex Diagnostic 3 (0.6) 26 (5.3) 4-Complex Therapeutic 242 (52.0) 166 (33.5) 5-Very Advanced 93 (20.0) 148 (29.9) ASA Patient Classification Score a , n (%) II-Mild systemic disease 159 (34.2) 104 (21.0) III-Severe systemic disease 261 (56.1) 332 (67.1) IV-Severe systemic life threatening disease 41 (8.8) 4 (10.9) ERCP Procedure Outcomes of Interest: Unadj OR (95% CI) P-value Adjd OR (95% CI) P-value Desired Duct Successfully Cannulated,n (%) 459 (98.7) 471 (95.2) 0.26 (0.10, 0.63) 0.0032* 0.26 (0.09, 0.77) 0.0148* Median Procedure Duration (IQR) (min) 22 (15, 33) 32 (21, 49) 1.41 (1.39, 1.43) < 0.0001* 1.21 (1.19, 4.09) < 0.0001* Median Fluoro Time (IQR) (min) 5 (3, 8.5) 1.9 (1.2, 3.2) 0.40 (0.33, 0.46) < 0.0001* 0.38 (0.30, 0.46) < 0.0001* Occurrence of Any Intra-operative Instabilityb,n (%) 301 (64.7) 241 (48.7) 0.44 (0.34, 0.57) < 0.0001* 0.70 (0.52, 0.96)e 0.0280* Occurrence of Cardiovascular Instability, n (%) 300 (64.5) 205 (41.4) 0.39 (0.30, 0.51) < 0.0001* 0.63 (0.46, 0.86)e 0.0034* Occurrence of Procedural Complicationc, n (%) 6 (1.3) 17 (3.4) 2.72 (1.06, 6.96) 0.0368* 5.56 (1.91, 16.21) 0.0017* Mortality at 30 Days, n (%) 12 (2.6) 14 (2.8) 1.10 (0.50, 2.40) 0.8133 1.27 (0.45, 3.63) 0.6499 a: ASA Scores “I-Normal, healthy” (n=5, 80% Prone vs 20% Supine)” and “V - Moribund, not expected to survive” (n=4, 0% Prone vs 100% Supine) omitted. b:Represents occurrence of any cardiovascular or respiratory instabilities, where results pertaining to the latter where omitted due to unreliable estimation (respiratory instability: n=37, 3% Prone vs 97% Supine; OR = 36.39, 95% CI: 4.97, 266.54). c: Represents occurrence of PEP (n=14, 0% Prone vs 100% Supine), Bleeding (n=5, 80% Prone vs 20% Supine), or Cholangitis (n=4, 50% Prone vs 50% Supine). Relationship between Outcome and Supine vs Prone position was adjusted for by ASA Patient Classification Score and either d: presence of fellow during ERCP procedure and Schutz Score for Procedural Difficulty, or e:anesthesia type. *Statistically significant at α=0.05.