PAD is a factor that has been inconsistently linked to the success rates of SBC in patients undergoing ERCP3–5. As precut sphincterotomy in difficult cases increases the overall success rate of SBC by 25% points, this study adds valuable insights by examining the influence of PAD classification and MDP morphology on the outcomes of NKPS9.
There is little data regarding precut sphincterotomy in patients with PAD17, 18. Park et al. reported that among 33 patients with PAD who underwent NKF, type II PAD was the most common, and type I PAD was the least common, which was similar to our findings18. However, due to the small number of cases, they did not analyze the success rate of each type18. Our results show that type III PAD had the highest success rate (75%) and type I PAD had the lowest success rate (50%). This may be because that the location of the MDP in type III PAD is outside the diverticulum and is therefore minimally affected by the PAD, allowing the endoscopist to perform NKPS commonly with the highest success rate. In comparison, types I and II PAD have more papillae that are difficult to detect or access, or bile duct directions that are more unpredictable19. Fernandes et al compared NKF for flat papillae, intradiverticular papillae (corresponding to type I PAD, n = 14), and diverticular marginal papillae (corresponding to type II PAD, n = 14) and reported success rates of 93.9%, 64.3%, and 71.4% (p = 0.005), respectively17. They concluded that NKF is feasible in diverticular papillae, but the success rate of initial ERCP is relatively low, which is consistent with our results.
The clinical influences of PAD size have rarely been studied. Kim et al. reported that CBD diameter was significantly related to PAD size and that patients with type I PAD had significantly larger CBD diameters than patients with type III PAD20. That is, type I PAD presented as larger while type III PAD tended to be smaller20. These findings were consistent with our results that type I PAD were significantly larger in size, followed by type II PAD, and type III PAD were significantly smaller. Furthermore, we found that the NKPS success group had a significantly smaller PAD size. However, the precise determination of PAD size is not straightforward. In our study, PAD diameter was measured through endoscopic imaging, whereas other studies used CT or MRCP, resulting in the median PAD diameter in our study being smaller compared with other reports20.
Several studies have explored the influence of MDP morphology on the technical success of ERCP14, 15, 21. Since its introduction in 2017, the Scandinavian classification of MDP morphology has gradually gained acceptance due to its remarkable intra- and inter-observer agreement 9, 14, 21, 22. Based on this classification, Haraldsson et al. found that cannulating type II and type III MDPs is more difficult21. However, the Scandinavian classification lacks some important papillary subtypes, such as those associated with PAD. Nevertheless, the inclusion of type D to account for PAD, as reported by Mohamed et al., yielded similar findings, indicating increased difficulty in cannulating Type II and III MDPs compared to Type I15. Additionally, they observed a higher frequency of NKPS in types II, III, and IV MDP compared to Type I (p < 0.001). Our study further explored the correlation between MDP morphology and NKPS outcomes in patients with PAD: type II MDP had the lowest NKPS success rate (38.2%), whereas type III MDP had the highest NKPS success rate (92.9%). Because most endoscopists consistently perform the same type of precut sphincterotomy over time, different precut techniques based on MDP morphology may be considered to improve success rates in patients with PAD, especially type II MDP9, 23, 24.
Several studies have suggested a potential link between MDP type and adverse events during ERCP15. However, in a study by Lyu et al. comparing the safety of NKP among various MDP types in patients with difficult biliary cannulation (8.3% of patients had PAD), no significant difference was found in the incidence of adverse events among the four MDP types25. Our findings were in line with this, albeit limited to patients with PAD. Additionally, our study revealed no significant differences in adverse event rates among the three types of PAD. Notably, perforation did not occur in our study, potentially due to its small sample size. Therefore, larger-scale studies may be necessary in the future.
This study is subject to various limitations. It was conducted retrospectively and involved a sample size in which procedures were carried out by five endoscopists. Consequently, there may have been an introduction of selection bias and operator bias. Furthermore, although the incidence of adverse events in this study is consistent with previous studies, the limited sample size could potentially impact the results. In addition, because pancreatic stents can reduce the risk of pancreatitis after ERCP and may improve the outcome of NKPS26, the utilization rate of pancreatic stents was only 26.2%, which may have affected the results of this study. Therefore, future large-scale prospective studies are necessary to validate our findings.
In conclusion, type II MDP morphology and significant bleeding during NKPS were significantly associated with NKPS failure in PAD patients with difficult bile duct cannulation, whereas type III MDP morphology had a trend toward increased NKPS success. These findings may have important implications for practitioners of ERCP, particularly in patients with type II MDP morphology, in whom techniques other than NKPS may be considered.