Risk factors for post-endoscopic retrograde cholangiopancreatography pancreatitis in the Chinese population: a systematic review and meta-analysis

Study design and registration

The systematic review and meta-analysis was conducted in accordance with the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) reporting standards. A predefined protocol was registered with the International Prospective Register of Systematic Reviews (PROSPERO, CRD42024514093) prior to the study’s initiation to ensure transparency and minimize bias (accessible at https://www.crd.york.ac.uk/PROSPERO/). The protocol outlined the objectives, search strategies, inclusion/exclusion criteria, and analytical methods, with any deviations from the original plan documented and justified.

Eligibility criteria

Study eligibility was determined according to the PICOS framework - Population, Intervention, Comparator, Outcomes, and Study design. The target population comprised Chinese patients undergoing diagnostic or therapeutic ERCP. Interventions or exposures included procedural techniques [e.g., endoscopic sphincterotomy (EST), pancreatic duct cannulation], biochemical markers (e.g., SOD levels), or clinical parameters (e.g., history of pancreatitis). Patients who underwent ERCP yet remained free of PEP served as the comparison group. The principal endpoint was PEP occurrence, operationalized according to ASGE criteria - new-onset abdominal pain after the procedure together with amylase or lipase values ≥ three times the upper reference limit and compatible imaging findings. Eligible evidence comprised RCTs, prospective and retrospective cohort studies, and case-control investigations. Non-research articles, animal studies, conference abstracts, and studies lacking extractable risk estimates [e.g., incomplete odds ratios (ORs) or confidence intervals (CIs)] were excluded. Studies finally included were those published in either Chinese or English and evaluated as moderate to high quality based on NOS and Cochrane risk of bias tools (section 2.8).

Information sources

Systematic searches were conducted in the following databases: PubMed (https://pubmed.ncbi.nlm.nih.gov/), China National Knowledge Infrastructure (CNKI) (https://www.cnki.net/), Wanfang Data (https://www.wanfangdata.com.cn/), VIP Database (https://www.cqvip.com/), Chinese Biomedical Literature Database (SinoMed) (https://www.sinomed.ac.cn), Cochrane Library (https://www.cochranelibrary.com/), EMBASE (https://www.embase.com/), and Web of Science (https://www.webofscience.com/). Grey literature and unpublished trials were identified through the WHO International Clinical Trials Registry Platform (https://clinicaltrials.gov/). The final database search was conducted on October 31, 2023.

Search

The search strategies for PubMed, SinoMed, Cochrane Library, and Web of Science utilized both MeSH terms and keywords. Grey literature and unpublished trials were identified through the WHO International Clinical Trials Registry Platform: [“Endoscopic Retrograde Cholangiopancreatography”(Mesh) OR “ERCP”] AND [“Pancreatitis”(Mesh) OR “Post-ERCP Pancreatitis”] AND [“Risk Factors”(Mesh) OR “Epidemiologic Factors”]. The Chinese search terms were (“zhong zheng yi xian yan” OR “yi xian yan” OR “ji xing yi xian yan” OR “zhong du ji xing yi xian yan” OR “ji xing chu xue huai si xing yi xian yan” OR “ji xing chu xue xing huai si xing yi xian yan” OR “zhong xing ji xing yi xian yan” OR “ji xing zhong zheng yi xian yan”) AND (“yi dan guan zao ying shu, nei kui jing ni xing” OR “ERCP” OR “jing nei jing ni xing yi dan guan zao ying” OR “nei kui jing ni xing yi dan guan zao ying” OR “nei jing xia ni xing yi dan guan zao ying shu” OR “ni xing yi dan guan zao ying shu”) AND (“wei xian yin su” OR “ying xiang yin su” OR “bing yin” OR “xiang guan yin su”) AND (“bing li dui zhao yan jiu” OR “bing li dui zhao” OR “dui lie yan jiu” OR “dui lie”) Filters were applied to restrict results to human studies and publications from October 1, 2003, to October 31, 2023. Equivalent Chinese-language search terms were used in domestic databases.

Study selection

Two independent reviewers conducted title and abstract screening using Rayyan QCRI software. Full-text articles were then assessed against the eligibility criteria. Discrepancies were resolved through consensus discussions with a third investigator. Studies that met the inclusion criteria proceeded to data extraction and quality assessment.

Data collection process

Data extraction was performed using standardized forms that were pilot-tested on 5% of the included studies. Dual independent extraction was conducted for study characteristics (author, year, sample size), risk factor definitions, and outcome measures. If studies reported risk ratios, these were converted to ORs. ORs were synthesized under a random-effects model, and any data ambiguities were resolved by contacting the respective study authors for clarification.

Data items

The extracted variables included demographic parameters (age, gender), procedural factors (cannulation attempts, duct visualization), biochemical markers (SOD levels), and clinical history (prior pancreatitis, biliary pathology). Operational definitions were consistent with the American Society for Gastrointestinal Endoscopy criteria for diagnosing PEP.

Risk of bias in individual studies

The methodological quality was assessed using the Newcastle-Ottawa Scale (NOS) for observational studies, which evaluates studies based on three domains: selection (0-4 stars), comparability (0-2 stars), and outcome/exposure assessment (0-3 stars). RCTs were appraised with the Cochrane Risk of Bias 2.0 instrument, which evaluates bias related to randomization, protocol deviations, attrition, outcome assessment, and selective reporting. Scores of 5-7 denoted moderate methodological quality, whereas 8-9 signified high quality. For the sensitivity analyses, we excluded studies that attained ≤ 5 points on the Newcastle-Ottawa Scale or were judged high-risk by the RoB 2.0 tool to safeguard the robustness of our pooled estimates.

Summary measures

Effect sizes were ORs. Both ORs and their 95%CIs were extracted from individual studies and transformed to natural logarithms. Continuous variables were expressed as mean differences, where applicable.

Synthesis of results

All analyses were performed using RevMan 5.3 (Cochrane Collaboration). A two-sided P value ≤ 0.05 was considered statistically significant unless otherwise specified.

Heterogeneity was quantified using the I² statistic - classified as low (≤ 50 %), moderate (51%-75%), or high (> 75 %). Where I² exceeded 50%, pooled ORs with 95%CI were estimated under a random-effects framework to accommodate clinical variability; otherwise, a fixed-effects model was applied. Sensitivity analyses were conducted by excluding one study at a time and replacing a random-effects model with a fixed-effects one or vice versa to examine the robustness of the findings.

Risk of bias across studies

Potential publication bias was interrogated for predictors represented by ten or more studies - including female sex, prior pancreatitis, EST, pancreatic-duct opacification, difficult cannulation, repeated guidewire passage into the pancreatic duct, and antecedent biliary calculi - by examining funnel plot symmetry and applying Egger’s regression analysis. When funnel plot asymmetry was detected, a trim-and-fill analysis was performed to estimate adjusted effect sizes and assess the potential impact of publication bias on the overall results.

Additional analyses

Pre-specified subgroup analyses were conducted to examine potential differences across characteristics [age (≥ 60 years vs. < 60 years, sex, ***] and procedural variations (with or without pancreatic stenting). Meta-regression was conducted to explore sources of heterogeneity, including sample size and study design.

Study selection

A comprehensive search of seven electronic databases retrieved 1,155 records. After de-duplication (n = 427), 728 titles and abstracts were screened, of which 421 were excluded as irrelevant or ineligible. Full-text evaluation of 119 articles led to the additional exclusion of 81 studies. Thirty-eight studies met the eligibility criteria and were included in the meta-analysis. The selection process is detailed in Figure 1.

Figure 1. Flowchart of Literature Screening. This PRISMA flowchart outlines the systematic selection process of relevant literature. A total of 1,155 studies were identified through initial database searches, with 427 duplicates removed. After screening titles and abstracts, 728 records were reviewed, and 119 full-text articles were assessed for eligibility. Thirty-eight studies met the inclusion criteria for the meta-analysis, focusing on PEP risk factors in the Chinese population. Excluded studies were those with non-relevant outcomes, non-Chinese cohorts, or insufficient data.

Study characteristics

Thirty-eight observational studies (2003-2023) were analyzed [Table 1]. These studies investigated 22 predefined risk factors based on the PICOS criteria, including patient demographics (e.g., female gender, age), procedural factors (e.g., cannulation attempts, procedure duration), and biochemical markers (e.g., SOD levels). Quality assessment scores ranged from 5 to 9 on the NOS, indicating moderate to high methodological rigor.

Risk of bias within studies

Funnel plots revealed publication bias in pancreatic duct visualization, difficult cannulation, and multiple guidewire passages into the pancreatic duct. For 18 factors with high heterogeneity (I² > 50%), random-effects models were applied, while fixed-effects models were used for four factors with low heterogeneity (e.g., bile duct stenosis). Sensitivity analysis confirmed the robustness of the results, although heterogeneity was likely due to methodological or population differences. Retrospective study designs predominated, introducing the risk of selection and recall bias, highlighting the need for standardized prospective protocols.

Results of individual studies

Forest plot analysis revealed significant heterogeneity (I² = 51%-95%) for 18 risk factors, necessitating the use of random-effects models. Fixed-effects models were applied to four factors with low heterogeneity (I² ≤ 50%). Key findings included a significantly elevated ORs for PEP associated with prior pancreatitis (OR = 3.26, 95%CI: 2.44-4.35) and pancreatic duct visualization (OR = 2.24, 95%CI: 1.72-2.91). Non-significant associations were observed for hypertension (OR = 1.93, 95%CI: 0.78-4.78) and biliary stenting (OR = 1.65, 95%CI: 0.41-6.58).

Synthesis of results

The meta-analysis identified 22 statistically significant risk factors [Table 2 and Figure 2]. The strongest predictors included prior PEP (OR = 6.56, 95%CI: 2.40-17.92), papillary abnormalities (OR = 3.38, 95%CI: 1.41-8.08), and difficult cannulation (OR = 3.39, 95%CI: 2.07-5.54). Moderate heterogeneity (I² = 51%-76%) characterized factors such as elevated SOD levels (OR = 2.81, 95%CI: 1.24-6.34) and Endoscopic papillary balloon dilation (EPBD) (OR = 2.75, 95%CI: 1.61-4.72). Fixed-effects models confirmed bile duct stenosis (OR = 2.19, 95%CI: 1.64-2.93) and periampullary diverticulum (PAD) (OR = 2.09, 95%CI: 1.54-2.83) as stable predictors.

Figure 2. Meta-analysis of the associations between influence factors and risk of post-ERCP pancreatitis. This figure presents forest plots showing the pooled ORs with 95%CIs for 22 identified risk factors for post-ERCP pancreatitis. Significant predictors include prior PEP, papillary abnormalities, and difficult cannulation. Random-effects models were applied to account for substantial heterogeneity (I² = 51%-95%), while fixed-effects models confirmed the stability of the associations. ERCP: Endoscopic retrograde cholangiopancreatography; ORs: odds ratios; PEP: post-ERCP pancreatitis.

Risk of bias across studies

Funnel plot asymmetry [Figure 3] indicated potential publication bias for risk factors such as pancreatic duct visualization, difficult cannulation, and repeated guidewire insertion.

Figure 3. Bias of risk factors. This figure presents the visual evaluation of funnel plot asymmetry for key outcomes. Trim-and-fill analysis was applied to adjust the effect sizes for selected factors, addressing potential publication bias.

Additional analyses

Sensitivity analyses using the leave-one-out approach confirmed the robustness of all significant predictors. The results remained within the 95%CI boundaries throughout, and the exclusion of any single study did not alter the overall conclusions [Figure 4].

Figure 4. Sensitivity analysis of different factors associated with risk of post-ERCP pancreatitis. This figure demonstrates the robustness of significant predictors. Sequential exclusion of individual studies showed that the OR fluctuations remained within the 95%CI bounds, confirming the stability of the findings. ERCP: Endoscopic retrograde cholangiopancreatography; OR: odds ratio.

Gender (Female)

In this study, the incidence of PEP in women was found to be 2.77 times higher than in men. This suggests that women are at a higher risk of developing PEP compared to men, a finding consistent with previous studies. Among female patients, SOD levels were found to be higher, which may explain the increased incidence of PEP in women. The more frequent occurrence of SOD in elderly women, compared to men, is thought to elevate the risk of post-procedure pancreatitis^[14].

History of pancreatitis

A history of previous pancreatitis is a known risk factor for the development of PEP^[15]. Patients with a history of pancreatitis have often sustained significant damage to the pancreatic parenchyma, resulting in impaired pancreatic secretory function^[16]. The data from this meta-analysis indicate that patients with a history of pancreatitis are 3.26 times more likely to develop PEP compared to those without such a history, which aligns with findings from other studies. Consequently, the risk of developing pancreatitis following the procedure is significantly elevated^[15].

EST

Upon pooling the available data, EST exhibited no statistically significant association with the incidence of PEP. EST, a routine procedure in ERCP, remains controversial regarding its role as a risk factor for PEP. Some studies suggest that the thermal effects of EST may cause edema in the tissues surrounding the pancreatic duct, obstructing pancreatic fluid drainage and potentially triggering PEP^[17]. However, EST may also reduce pancreatic duct pressure, alleviate the severity of pancreatitis, and provide a protective effect by lowering the tension at the duct opening^[18]. Furthermore, although some studies indicate that EST may reduce the incidence of PEP, its potential complications, such as bleeding and perforation, could counterbalance these benefits^[19]. Therefore, the relationship between EST and PEP occurrence warrants further investigation through larger sample studies.

Pancreatic duct injection

In this study, pancreatic duct injection was identified as a risk factor for PEP. The injection of contrast agents during ERCP can directly damage pancreatic tissue, and if performed too rapidly or under excessive pressure, it can cause further mechanical injury to the pancreatic duct and acini, thus triggering PEP^[20]. During ERCP procedures, repeated placement of the guidewire within the pancreatic duct increases the risk of duct injury and contrast agent entry, leading to more frequent pancreatic duct visualization, which significantly elevates the risk of PEP^[21]. The placement of a metal biliary stent can disrupt biliary and pancreatic fluid drainage, causing pressure imbalance and pancreatic juice reflux, which further increases the likelihood of abnormal pancreatic duct imaging and the risk of PEP^[21].

At the molecular scale, experimental evidence suggests that administration of contrast media or isotonic saline activates NF-κB p65 and STAT3 signalling, alters the expression of apoptosis-related proteins, and downregulates tight-junction components, collectively fostering the initiation and progression of PEP^[20].

Moreover, increased pancreatic duct pressure has been confirmed as a key trigger for PEP. Animal studies have demonstrated that elevated intraductal pressure can activate the Ca²⁺/calmodulin-dependent phosphatase signaling pathway, leading to calcium homeostasis disruption in pancreatic acinar cells, mitochondrial dysfunction, and tight junction damage^[22]. This cascade of events ultimately results in pancreatic edema, inflammation, and necrosis, closely mirroring the pathological features of clinical PEP.

Cannulation factors (Difficult cannulation, multiple cannulations)

In this study, difficult cannulation and multiple cannulations were identified as risk factors for PEP. Difficult cannulation is widely recognized as an independent risk factor for PEP^[23]. Concepción-Martín et al. suggested that repeated cannulation causes spasm and damage to the papillary muscle, leading to increased pressure within the pancreatic duct. This obstructs the flow of bile and pancreatic juice, causing pancreatic juice reflux and ultimately activating pancreatic enzymes^[24]. Similarly, Zhu noted that multiple cannulations increase the risk of direct pancreatic injury, which may facilitate bacterial infection and thereby induce secondary PEP^[25]. Therefore, when multiple or difficult cannulations occur during an ERCP procedure, early and close monitoring of serum amylase levels is crucial^[26]. If abnormalities are detected, prompt intervention should be undertaken to prevent the onset of PEP.

Multiple guidewire insertions into the pancreatic duct

In this study, multiple guidewire insertions into the pancreatic duct were identified as a risk factor for PEP. Research has shown that repeated unintended guidewire insertions into the pancreatic duct, often due to difficult cannulation, are relatively common. The risk of PEP increases with the frequency and extent of these insertions, which contribute to a higher incidence of PEP and induce acute postoperative pancreatic inflammation^[27].

History of biliary stone disease

A history of biliary stone disease is a recognized risk factor for PEP. When the number of stones is large, multiple interventions may be required, which increases the risk of mucosal injury and infection^[28]. Asymptomatic stones, due to the lack of inflammatory adhesions, have fragile bile duct walls, making them more prone to perforation or bleeding during surgery^[29]. Additionally, factors such as stone fragments or inflammatory mediators being refluxed into the pancreatic duct during the procedure may trigger pancreatitis^[30]. Moreover, biliary obstruction caused by stones can lead to bacterial overgrowth, further increasing the risk of infection and PEP^[31].

Age

Accumulating evidence indicates that patients younger than 60 years are at heightened risk of developing PEP^[32]. Younger age appears to predispose patients to more complicated manifestations of PEP^[32,33]. Pancreatic exocrine output increases until roughly 43 years of age and then gradually wanes, diminishing the gland’s responsiveness to mechanical stimuli such as ERCP^[34]. Experts suggest that with advancing age, pancreatic exocrine function diminishes, potentially due to pancreatic atrophy and decreased enzyme activity. This may explain the reduced incidence of PEP in older individuals^[34]. However, in this study, no significant association was found between age and the occurrence of PEP, indicating that larger sample size studies are needed for further validation.

Cannulation time

Consistent with this study, cannulation time is a recognized risk factor for PEP^[25]. Prolonged cannulation time increases the risk of adverse events, including pancreatitis, following ERCP^[35]. Some studies suggest that extended cannulation time may lead to ductal contrast injection, SOD spasm, and papillary edema, which in turn cause pancreatic duct obstruction and trigger pancreatitis^[36]. Evidence from a retrospective analysis suggests that PEP stems from the interplay of multiple contributory factors, culminating in acinar cell injury and the downstream inflammatory cascade^[37]. Other literature indicates that cannulation time decreases with increasing physician experience, suggesting that improper cannulation or lack of skill may contribute to prolonged cannulation time^[38]. As such, cannulation time can, to some extent, serve as a representative factor for related procedural issues.

Duodenal papillary abnormalities

Duodenal papillary abnormalities are a recognized risk factor for pancreatitis following ERCP. These abnormalities include papillary microstone impaction, papillary inflammatory stricture, and anomalous pancreaticobiliary duct junctions. When such abnormalities occur, they can lead to biliary and pancreatic duct obstruction, increasing biliary pressure and elevating the risk of pancreatitis after ERCP^[39]. Additionally, the size and anatomical variations of the papilla can affect the success rate of cannulation^[40]. Small papillae are associated with a higher risk of PEP^[41]. Abnormal papillary shapes can lead to increased incidences of pancreatitis, more attempts at cannulation, longer cannulation times, and a higher rate of inadvertent pancreatic duct cannulation, further raising the risk of PEP^[42]. These known risk factors make the occurrence of PEP more likely after the procedure, thereby increasing the risk of pancreatitis following ERCP.

Total bilirubin abnormalities

In this study, total bilirubin abnormalities were not found to be statistically significant when considered alongside other variables, suggesting that they are not an independent risk factor. However, elevated total bilirubin serves as a highly specific marker of residual common bile duct stones in cholelithiasis-related pancreatitis and can therefore guide the selective use of ERCP, averting unnecessary procedures^[43].

Endoscopic nasobiliary drainage

In this study, the analysis of data related to endoscopic nasobiliary drainage (ENBD) did not yield statistically significant results. This suggests that, based on the data from this study, there is no significant correlation between ENBD and the occurrence of PEP. However, some existing studies present differing viewpoints. One study indicated that the insertion of an ENBD catheter had a preventive effect on PEP, with the mechanism likely being that ENBD can effectively prevent pancreatic duct obstruction, thereby reducing the risk of PEP^[42,44]. Nonetheless, it should be noted that in clinical practice, there are certain drawbacks to placing an ENBD catheter. These include potential discomfort, the risk of pressure ulcers, and the possibility of the catheter being removed by non-compliant patients^[45]. Given these considerations, further clinical research is needed to determine whether ENBD serves as a protective or risk factor in the occurrence of PEP. More in-depth analysis from various perspectives is necessary to clarify its exact role and provide more reliable guidance for clinical practice.

Biliary or pancreatic stent placement

The combined results for this factor were not statistically significant, suggesting that biliary or pancreatic stent placement is not significantly associated with the occurrence of PEP. Part of the reason for this may be the current understanding of the challenges in pancreatic stent placement, along with skepticism regarding its effectiveness in preventing PEP. Some studies suggest that stent placement may fail if the duct is small or curved, potentially increasing the risk^[46]. However, a related study by Martin Freeman indicates that pancreatic stent placement remains a critical intervention for high-risk patients, young patients with normal pancreatic function, or those who encounter cannulation difficulties during surgery^[47]. Although this study initially concluded that biliary or pancreatic stent placement is not significantly associated with the occurrence of PEP, the complexity of research in this area and the limitations of the current sample size suggest that this conclusion requires further validation through large-scale studies.

SOD

SOD abnormalities are a well-established risk factor for PEP^[48]. During ERCP, abnormal increases in biliary and pancreatic duct pressure can lead to ductal obstruction. This obstructive state can cause local circulatory disturbances, resulting in insufficient oxygen supply to the pancreatic tissue. In this hypoxic microenvironment, the redox balance within cells is disrupted, leading to the excessive production of oxygen free radicals^[49]. Normally, SOD, a key antioxidant enzyme, clears oxygen free radicals in a timely manner, maintaining the body’s oxidative-antioxidant balance. However, when SOD levels become excessively high, its overactivity in clearing free radicals can disrupt cellular redox signaling pathways, overstimulating the oxidative stress response^[50]. This enhanced oxidative stress further promotes the release of inflammatory mediators, exacerbating the inflammatory cascade and causing direct cytotoxic effects on pancreatic tissue^[51]. As a result, pancreatic acinar cells are damaged, and pancreatic parenchymal inflammation occurs, significantly increasing the risk of pancreatitis.

Common bile duct stricture

Common bile duct stricture is a recognized risk factor for PEP. Bile duct stricture can lead to biliary obstruction, increasing the risk of elevated biliary pressure and bile stasis during the ERCP procedure. This can result in bile reflux into the pancreatic duct, triggering pancreatitis^[52].

Previous PEP

In this study, a history of previous PEP was identified as a risk factor for PEP. This finding is consistent with the 2020 European Society of Gastrointestinal Endoscopy guidelines, which identified a history of previous PEP as an independent risk factor for PEP. The underlying reason may be that patients with a history of previous PEP often have a higher number of comorbidities and susceptibility factors, making them more sensitive to the procedure and increasing the likelihood of developing PEP^[53].

PAD

PAD is considered a risk factor for PEP. PAD is a cystic structure closely associated with the biliary tract, where bile may accumulate. It has long been recognized as a risk factor for PEP^[54]. Historically, PAD was considered one of the major causes of ERCP failure, likely due to the location of the diverticulum near the papillary orifice^[55]. This proximity could lead to prolonged procedure times and a higher failure rate of cannulation. If the diverticulum is too close to the orifice, it may exert pressure on the distal bile duct, disrupting pancreatic juice drainage and thereby leading to pancreatitis^[56]. However, some studies suggest that the presence of PAD does not affect the technical success of ERCP^[56]. Research has shown that there is no significant difference between PAD and non-PAD groups among patients undergoing ERCP for various indications^[57]. Therefore, the impact of PAD on the success of ERCP and the occurrence of complications remains controversial.

Sphincter of oddi dysfunction

Sphincter of Oddi dysfunction is a recognized risk factor for PEP. Oddi dysfunction may contribute to the pathogenesis of acute pancreatitis. When the function of the Sphincter of Oddi is impaired, it can lead to impaired drainage of pancreatic juice and bile, potentially triggering pancreatitis^[58,59]. A study by Chen et al. demonstrated in animal experiments that Sphincter of Oddi dysfunction is an independent risk factor for inducing pancreatitis^[60].

Hypertension

Based on the hypertension data from this study, no statistically significant association was observed between hypertension and the development of PEP. This finding aligns with the study by Liu^[61], while the research by Dong et al. suggests that hypertension may be a risk factor for PEP^[62]. Consequently, the role of hypertension as a risk factor for PEP remains controversial and warrants further investigation. However, renin-angiotensin system inhibitors, including captopril, ramipril, enalapril, lisinopril, quinapril, benazepril, losartan, and valsartan, have been implicated in cases of drug-induced acute pancreatitis^[63]. Moreover, certain combinations of antihypertensive drugs, such as ramipril-hydrochlorothiazide, along with medications like formoterol and budesonide, have been associated with an increased risk of PEP following ERCP^[64]. The evidence implies that certain classes of antihypertensive agents may heighten the risk of PEP. Given the limitations of this preliminary study, including potential sample selection bias and small sample size, further studies with larger cohorts are needed to determine whether hypertension is a risk factor for PEP.

Endoscopic papillary balloon dilation

In this study, EPBD was identified as a risk factor for PEP. The sensitivity of EPBD-related PEP may vary across different populations. Some studies suggest that balloon dilation itself does not directly cause PEP, but insufficient dilation may damage the pancreatic duct, and post-dilation papillary edema or spasm may contribute to the development of PEP^[65].

Therapeutic potential and research imperatives of TCM for PEP prevention

Beyond conventional pharmacotherapy, TCM has emerged as a potential strategy for PEP prophylaxis. Meta-analyses indicate that TCM formulations (e.g., Gegen Qinlian decoction, Shakuyakukanzoto) may reduce the incidence of PEP and hyperamylasemia by modulating inflammatory pathways such as NLRP3 inflammasome-mediated pyroptosis and suppressing oxidative stress^[66]. Experimental studies demonstrate that topical application of TCM extracts (e.g., Shakuyakukanzoto spray) to the duodenal papilla inhibits sphincter of Oddi spasm, thereby reducing pancreatic duct pressure and serum amylase elevation^[67]. Mechanistically, TCM compounds exhibit antioxidative properties that counteract free radical-mediated acinar cell injury, a key pathogenic mechanism in PEP^[66].

However, clinical adoption remains constrained by heterogeneity in TCM formulations, limited multicenter randomized trials, and insufficient standardization of dosing protocols. While preliminary evidence suggests efficacy comparable to rectal NSAIDs in specific cohorts, large-scale validation is needed to establish safety profiles, optimize delivery methods (e.g., enteral vs. topical), and identify patient subgroups most likely to benefit^[68]. Future research should prioritize high-quality RCTs evaluating TCM as adjunctive therapy alongside established PEP prophylaxis protocols.

Comparison of PEP risk factors in chinese vs. global populations

Our findings reveal distinctive patterns in PEP risk factors within the Chinese population compared to global cohorts. While female gender and difficult cannulation universally elevate PEP risk globally (OR 1.47-1.70 and 2.60-3.39, respectively), their prevalence in Chinese patients (OR 2.77 and 3.39) exceeds pooled estimates from multinational meta-analyses^[69]. Notably, the attenuated association between EST and PEP in our cohort contrasts with Western reports where EST remains a significant procedural risk (OR 1.08 vs. global OR 1.98-2.64)^[69,70]. This divergence may arise from ethnic variations in pancreatic ductal anatomy or differential operator techniques in sphincterotomy execution. Furthermore, the prominence of elevated SOD as a biochemical predictor (OR 2.81) is underrepresented in non-Asian studies, suggesting population-specific pathophysiological pathways^[71]. Conversely, prior pancreatitis consistently confers high PEP risk across ethnicities, though its effect size in Chinese patients (OR 6.56) surpasses global estimates (OR 3.26-4.40)^[69,72]. These distinctions underscore the necessity of ethnicity-tailored risk stratification, particularly as emerging machine learning algorithms (e.g., RF AUC 0.947 in Chinese cohorts vs. 0.70-0.74 in multinational models) exhibit population-dependent performance variances^[73,74]. Future guidelines should integrate region-specific procedural factors - such as the reduced PEP incidence in acute cholangitis patients observed in East Asia (Saito et al., 2022) - to optimize prophylaxis strategies^[75].

Methodological limitations of the meta-analysis

This study has several methodological limitations requiring careful consideration. Substantial heterogeneity (I² > 50%) was observed for 18 of the 22 risk factors analyzed, with I² values ranging from 51% to 95%, attributable to clinical and methodological variations across studies. These include differences in ERCP procedural techniques (e.g., cannulation protocols, stent placement criteria), definitions of risk exposures (e.g., “difficult cannulation”, “papillary abnormalities”), and PEP diagnostic thresholds across institutions. Although random-effects models were employed to incorporate this heterogeneity, residual variability may influence the precision of pooled estimates. Furthermore, variations in study quality were evident, with NOS scores spanning 5-9 points, reflecting disparities in design robustness, particularly among retrospective studies that may be susceptible to selection and recall biases. The predominance of single-center observational studies also limits generalizability. Publication bias, indicated by funnel plot asymmetry for factors including pancreatic duct visualization, difficult cannulation, and multiple guidewire passages [Figure 3], suggests possible underrepresentation of smaller studies reporting null associations. While trim-and-fill analysis was performed to adjust effect sizes, the exclusion of non-English literature and reliance on Chinese databases (CNKI, Wanfang) may have omitted relevant data, potentially skewing risk estimates. These limitations underscore the need for cautious interpretation of results and highlight the imperative for future prospective, multicenter studies with standardized PEP diagnostic criteria and risk exposure definitions.

In conclusion, this synthesis outlines a distinct set of procedural, anatomical, and biochemical risk factors for PEP in Chinese patients, highlighting cannulation difficulty, papillary abnormalities, and elevated SOD levels as particularly notable. Interestingly, the lack of a clear association between EST and PEP risk in this population stands in contrast to global data, suggesting the presence of ethnic-specific protective mechanisms that warrant further investigation. Future research should focus on prospectively validating these risk factors, developing ethnicity-tailored risk prediction models, and exploring the mechanisms behind the SOD-PEP paradox. Additionally, the integration of AI for real-time assessment of cannulation difficulty and genetic studies on PEP susceptibility loci offer promising avenues to enhance ERCP safety, particularly for high-risk populations^[76].