Preventive strategies for post-endoscopic retrograde cholangiopancreatography pancreatitis: A systematic review and meta-analysis

Abstract

BACKGROUND

Post-endoscopic retrograde cholangiopancreatography (ERCP) pancreatitis (PEP) is a common and clinically significant adverse event following ERCP. Although pharmacological and endoscopic prophylactic measures have been widely investigated, the implications of these strategies for perioperative nursing pathways require systematic synthesis.

AIM

To investigate the effects of preventive strategies on the risk of PEP and summarize the clinical implications for standardized perioperative nursing practice.

METHODS

Chinese and international databases, including the China National Knowledge Infrastructure, Wanfang, the VIP Database for Chinese Technical Periodicals, the China Biology Medicine disc, the Chinese Clinical Trial Registry, PubMed, Web of Science, and the Cochrane Library, were searched from January 2016 to February 2026. Randomized controlled trials comparing preventive interventions with conventional care, placebo, blank control, or routine treatment in adult patients undergoing therapeutic ERCP and reporting PEP events were considered eligible. Risk ratios with 95% confidence intervals were pooled using Review Manager 5.4.1. Heterogeneity was evaluated using the χ² test and the I² statistic.

RESULTS

Fourteen randomized controlled trials involving 4788 participants were included. PEP occurred in 93 of 2374 patients in the intervention groups and 249 of 2414 patients in the control groups. Preventive strategies significantly reduced the risk of PEP compared with control conditions (risk ratio = 0.38, 95% confidence interval: 0.30-0.48, P < 0.00001), with no statistically significant heterogeneity (I² = 0%, P = 0.77). Most included interventions involved pharmacological prophylaxis, primarily non-steroidal anti-inflammatory drug-based regimens, or prophylactic pancreatic duct stenting. Risk-of-bias assessment revealed common limitations related to blinding, allocation concealment, and reporting of outcome assessment procedures.

CONCLUSION

Current evidence supports the implementation of structured perioperative nursing pathways for PEP prevention through risk screening, timely care coordination, close monitoring, and patient education.

Key Words: Endoscopic retrograde cholangiopancreatography; Post-endoscopic retrograde cholangiopancreatography pancreatitis; Prevention; Nursing; Non-steroidal anti-inflammatory drugs; Pancreatic duct stent; Meta-analysis

Core Tip: Preventive strategies, primarily non-steroidal anti-inflammatory drug-based pharmacological prophylaxis and pancreatic duct stenting, reduce the risk of post-ERCP pancreatitis by 62% (risk ratio = 0.38, 95% confidence interval: 0.30-0.48). Structured perioperative nursing pathways integrating risk screening, timely prophylaxis coordination, close postprocedural monitoring, and patient education are essential to translate this evidence into consistent clinical practice and improved patient safety.

INTRODUCTION

Endoscopic retrograde cholangiopancreatography (ERCP) is an important minimally invasive procedure used for the diagnosis and treatment of pancreatobiliary diseases, including bile duct stones, biliary strictures, chronic pancreatitis, and pancreaticobiliary tumors[1]. Despite its therapeutic benefits, ERCP is associated with procedure-related adverse events, among which post-ERCP pancreatitis (PEP) is the most common and clinically significant[2].

PEP can prolong hospitalization, increase healthcare costs, intensify nursing workload, and, in severe cases, progress to organ dysfunction or death[3]. The reported incidence varies according to patient risk profile, procedural complexity, cannulation technique, pancreatic duct manipulation, and the preventive strategy used[4,5]. Therefore, high-risk patients require particularly careful preventive measures and close monitoring.

Current prevention strategies emphasize multimodal management, including rectal non-steroidal anti-inflammatory drugs (NSAIDs), prophylactic pancreatic duct stenting in selected high-risk patients, meticulous procedural techniques, and appropriate fluid management. Nursing practice plays a crucial role in translating these measures into reliable clinical pathways through preprocedural risk assessment, patient education, medication verification, intraoperative coordination, postprocedural monitoring, pain assessment, hydration support, and prompt escalation of warning signs.

However, published studies differ in intervention type, sample size, methodological quality, and reporting details. Quantitative synthesis of the available evidence may help clarify the overall preventive effect and determine how these findings can inform standardized perioperative nursing workflows. Therefore, this systematic review and meta-analysis evaluated randomized controlled trials of preventive strategies for PEP and summarized their implications for evidence-informed nursing practice.

MATERIALS AND METHODS

This article was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA 2020) guidelines. This systematic review was registered with PROSPERO (CRD420261431165)

Search strategy

This article was conducted in accordance with the general principles of systematic reviews and meta-analyses. China National Knowledge Infrastructure, Wanfang, VIP Database for Chinese Technical Periodicals, China Biology Medicine disc, the Chinese Clinical Trial Registry, PubMed, Web of Science, and the Cochrane Library were searched from January 2016 to February 2026. The search strategy included combinations of the terms ERCP, PEP, hyperamylasemia, prevention, nursing, and randomized controlled trial. Chinese equivalents of these terms were also incorporated into the search.

Eligibility criteria

Studies were considered eligible if they met the following criteria: Randomized controlled trial design; adult patients undergoing therapeutic ERCP; comparison of a preventive intervention with conventional care, placebo, blank control, routine treatment, or another standard preventive approach; and reporting of PEP events that could be extracted as dichotomous data. Studies were excluded if they were reviews, case reports, retrospective studies, cohort studies, case-control studies, animal experiments, mechanistic investigations, conference abstracts without extractable data, duplicate publications, studies involving non-ERCP procedures, or studies focused on the treatment rather than the prevention of PEP.

Study selection and data extraction

Two reviewers independently screened titles, abstracts, and full-text articles. Disagreements were resolved through discussion or adjudication by a third reviewer. Extracted information included the first author, publication year, sample size, intervention category, control condition, number of PEP events, and methodological details relevant to the assessment of risk of bias.

Risk of bias assessment

Risk of bias was assessed across the domains of random sequence generation, allocation concealment, blinding of participants and personnel, blinding of outcome assessment, incomplete outcome data, selective reporting, and other potential sources of bias. Each domain was judged as having a low, unclear, or high risk of bias.

Statistical analysis

Risk ratios with 95% confidence intervals (CIs) were used for dichotomous outcomes. Heterogeneity was assessed using the χ² test and the I² statistic. A fixed-effect model was applied when I² was 50% or lower, and P was 0.10 or higher; otherwise, a random-effects model was planned. All statistical analyses were performed using Review Manager 5.4.1.

RESULTS

Study selection

The database search identified 289 records. After duplicate removal, 248 records remained for title and abstract screening. Ninety-four records were excluded at this stage, and 154 full-text articles were assessed for eligibility. After excluding 140 articles that did not meet the inclusion criteria, 14 randomized controlled trials[6-19] were included in the quantitative synthesis. The detailed study identification, screening, eligibility, and inclusion process is summarized in Figure 1.

Figure 1 PRISMA 2020 flow diagram showing the identification, screening, eligibility, and inclusion process for studies in this systematic review and meta-analysis. A total of 289 records were identified through database searching. After duplicate removal, 248 records were screened, and 154 full-text articles were assessed for eligibility. Fourteen randomized controlled trials met the inclusion criteria and were included in the quantitative synthesis.

Characteristics of included studies

The included trials evaluated several categories of preventive strategies, including NSAID-based pharmacological prophylaxis, prophylactic pancreatic duct stenting, and traditional Chinese medicine or other interventions. Overall, the intervention groups comprised 2374 participants, whereas the control groups comprised 2414 participants.

The main characteristics of the 14 included randomized controlled trials are summarized in Table 1.

Table 1 Summary of the Included Trials.

Ref.	Randomization method	Allocation concealment	Blinding	Intervention details	Intervention category	Intervention PEP	Control PEP	Effect estimate
He[6], 2016	Simple randomization	Unclear	Unclear	Indomethacin suppository 100 mg + somatostatin	NSAID-based pharmacologic prophylaxis	2/63	9/66	0.23 (0.05, 1.04)
Liu[7], 2022	Random number table	Unclear	Unclear	Indomethacin suppository 100 mg	NSAID-based pharmacologic prophylaxis	2/35	11/51	0.26 (0.06, 1.12)
Liu[8], 2024	Random number table	Unclear	Unclear	Tongyi Chengqi decoction	Traditional Chinese medicine/other	0/36	2/36	0.20 (0.01, 4.03)
Chang et al[9], 2020	Unclear	Unclear	Unclear	Pancreatic duct stent placement	Prophylactic pancreatic duct stent	2/51	13/48	0.14 (0.03, 0.61)
Zhang et al[10], 2017	Unclear	Unclear	Unclear	Pancreatic duct stent placement	Prophylactic pancreatic duct stent	5/45	18/45	0.28 (0.11, 0.68)
Yang[11], 2018	Random number table	Unclear	Unclear	Indomethacin suppository 100 mg	NSAID-based pharmacologic prophylaxis	1/35	8/35	0.12 (0.02, 0.95)
Lin et al[12], 2016	Unclear	Unclear	Unclear	Phloroglucinol injection 40 mg	Pharmacologic prophylaxis	1/19	4/19	0.25 (0.03, 2.04)
Wang[13], 2018	Random number table	Unclear	Unclear	Indomethacin suppository (timing comparison)	NSAID-based pharmacologic prophylaxis	1/40	7/40	0.14 (0.02, 1.11)
Luo[14], 2019	Computer-generated	Sealed envelopes	Single-blind	Rectal indomethacin 100 mg	NSAID-based pharmacologic prophylaxis	47/1297	100/1303	0.47 (0.34, 0.66)
Ge[15], 2016	Random number table	Unclear	Unclear	Diclofenac sodium suppository 50 mg	NSAID-based pharmacologic prophylaxis	3/50	7/50	0.43 (0.12, 1.56)
Jia et al[16], 2017	Unclear	Unclear	Unclear	Pancreatic duct stent placement	Prophylactic pancreatic duct stent	9/147	24/150	0.38 (0.18, 0.80)
Qian[17], 2020	Unclear	Unclear	Unclear	Indomethacin suppository 100 mg	NSAID-based pharmacologic prophylaxis	1/31	6/31	0.17 (0.02, 1.30)
Chen et al[18], 2021	Unclear	Unclear	Unclear	Preventive strategy (not specified)	NSAID-based pharmacologic prophylaxis	3/75	10/75	0.30 (0.09, 1.05)
Han[19], 2016	Unclear	Unclear	Unclear	Not specified	NSAID-based pharmacologic prophylaxis	16/450	30/465	0.55 (0.30, 1.00)

PEP: Post-endoscopic retrograde cholangiopancreatography pancreatitis; NSAIDs: Non-steroidal anti-inflammatory drugs.

Pooled effect on PEP

Across the 14 included randomized controlled trials, PEP occurred in 93 of 2374 patients in the intervention groups and 249 of 2414 patients in the control groups. Preventive strategies were associated with a significantly reduced risk of PEP (risk ratio = 0.38, 95%CI: 0.30-0.48, P < 0.00001). No statistical heterogeneity was detected (χ² = 9.07, df = 13, P = 0.77; I² = 0%). The forest plot for the primary outcome is shown in Figure 2. The overall and subgroup pooled estimates are presented in Table 2.

Figure 2 Forest plot comparing the risk of post-endoscopic retrograde cholangiopancreatography pancreatitis between preventive intervention groups and control groups across the 14 included randomized controlled trials. Risk ratios with 95% confidence intervals are presented for each study. The size of each square is proportional to the study weight in the fixed-effect meta-analysis. The diamond at the bottom represents the pooled risk ratio (0.38, 95% confidence interval: 0.30-0.48), indicating a 62% relative risk reduction in favor of preventive strategies. Heterogeneity: χ² = 9.07, df = 13, P = 0.77, I² = 0%. ERCP: Endoscopic retrograde cholangiopancreatography; RR: Risk ratio; CI: Confidence interval.

Table 2 Major quantitative findings.

Analysis	Trials	Intervention events/total	Control events/total	Pooled RR (95%CI)	Heterogeneity
Overall PEP prevention	14	93/2374	249/2414	0.38 (0.30-0.48)	I2 = 0%; P = 0.77
NSAID/pharmacologic prophylaxis	10	77/2095	192/2135	0.41 (0.32-0.52)	I2 = 0%; P = 0.85
Pancreatic duct stent	3	16/243	55/243	0.31 (0.20-0.48)	I2 = 0%; P = 0.98
Traditional Chinese medicine/other	1	0/36	2/36	0.20 (0.01-4.03)	Not applicable

RR: Risk ratio; CI: Confidence interval; PEP: Post-endoscopic retrograde cholangiopancreatography pancreatitis; NSAIDs: Non-steroidal anti-inflammatory drugs.

Risk of bias

The risk-of-bias assessment indicated that several trials adequately reported random sequence generation; however, allocation concealment was frequently unclear. Blinding of participants and personnel was often not feasible or was insufficiently reported because of the nature of the interventions. Blinding of outcome assessment and selective reporting were also commonly judged as unclear. Incomplete outcome data were generally considered to have a low risk of bias. A summary of the risk-of-bias assessment across the seven domains is shown in Figure 3.

Figure 3 Risk-of-bias assessment of the 14 included randomized controlled trials using the Cochrane risk-of-bias tool. Each domain was judged as having a low risk (blue plus), unclear risk (yellow question mark), or high risk (white minus) of bias. Domains assessed: Random sequence generation (selection bias), allocation concealment (selection bias), blinding of participants and personnel (performance bias), blinding of outcome assessment (detection bias), incomplete outcome data (attrition bias), selective reporting (reporting bias), and other potential sources of bias. The majority of trials demonstrated unclear risk in allocation concealment and blinding domains, whereas incomplete outcome data were generally at low risk of bias.

Publication bias

We assessed publication bias visually with a funnel plot and formally with Egger’s regression test. The funnel plot (Figure 4) showed some asymmetry - smaller studies tended to cluster on one side of the pooled estimate rather than scatter symmetrically. The Egger test confirmed this impression (z = -2.54, P = 0.011), suggesting that the asymmetry is unlikely to be due to chance alone. That said, with only 14 studies included, the test has limited power, so we don’t overinterpret this result. We view it as a signal worth noting, rather than a definitive proof of bias.

Figure 4 Funnel plot of the 14 included studies for assessing publication bias. The horizontal axis represents the log risk ratio, and the vertical axis represents the standard error of the log risk ratio. The dashed vertical line indicates the pooled effect estimate. The diagonal lines represent the pseudo-95% confidence limits around the pooled estimate, indicating the expected distribution of studies in the absence of publication bias. Visual inspection suggests some asymmetry, with smaller studies clustering predominantly on one side of the pooled estimate. Egger’s regression test yielded a statistically significant result (z = -2.54, P = 0.011), which may indicate potential publication bias or other sources of small-study effects. However, with only 14 studies included, the test has limited power, and this result should be interpreted with caution.

Implications for perioperative nursing

The pooled effect primarily reflects clinical preventive strategies rather than nursing interventions alone. Nevertheless, these findings are highly relevant to the design of nursing workflows. Key actionable components include structured pre-ERCP risk assessment, verification of NSAID contraindications, timely administration of rectal NSAIDs when prescribed, coordination during pancreatic duct stent placement for high-risk patients, and postprocedural assessment of abdominal pain, nausea, vomiting, abdominal distension, vital signs, and serum amylase levels, together with early recognition of clinical deterioration. Standardized education should additionally address fasting requirements, hydration, pain reporting, and warning symptoms following discharge.

Grading of Recommendations Assessment, Development and Evaluation assessment: The overall certainty of evidence for the primary outcome (PEP incidence) was assessed using the Grading of Recommendations Assessment, Development and Evaluation approach. The evidence was rated as moderate certainty, downgraded by one level due to serious risk of bias (predominantly unclear or high risk of bias in allocation concealment and blinding domains across the included trials). No downgrading was applied for inconsistency (I² = 0%), indirectness (all studies directly compared preventive strategies with control conditions in adult ERCP patients), imprecision (narrow 95%CI with a total sample of 4,788 participants), or publication bias. A funnel plot was not constructed because the total number of included studies (n = 14) was fewer than the recommended threshold of 10 per subgroup for meaningful asymmetry assessment; Egger’s regression test was therefore also not performed.

DISCUSSION

In this systematic review and meta-analysis of 14 randomized controlled trials, preventive strategies were associated with a substantial reduction in the risk of PEP. The pooled estimate demonstrated a 62% relative risk reduction compared with control conditions, and the narrow CI suggests a clinically meaningful effect. These findings reinforce the concept that PEP is not merely a procedure-related adverse event requiring treatment after onset, but also a preventable complication that warrants anticipatory management before, during, and after ERCP.

The overall direction of effect is both biologically and clinically plausible. ERCP can induce pancreatic injury through papillary edema, difficult cannulation, hydrostatic injury caused by contrast injection, thermal damage, repeated pancreatic duct instrumentation, and subsequent inflammatory activation. NSAID-based prophylaxis may mitigate this inflammatory cascade by inhibiting phospholipase A2-mediated pathways and downstream prostaglandin synthesis. Prophylactic pancreatic duct stenting may alleviate ductal hypertension and preserve pancreatic drainage following papillary manipulation. Although these mechanisms differ, both target early events in the development of PEP, which may explain why preventive strategies provide measurable benefits when implemented at the appropriate time and in carefully selected patients.

From a clinical practice perspective, the findings support the establishment of a prevention-oriented ERCP pathway rather than reliance on an isolated intervention. A practical pathway should begin with structured risk stratification, including patient-related factors such as previous PEP, suspected sphincter of Oddi dysfunction, female sex, younger age, normal bilirubin levels, and a history of pancreatitis, as well as procedure-related factors such as difficult cannulation, pancreatic guidewire passage, precut sphincterotomy, pancreatic contrast injection, and pancreatic sphincter intervention. Risk stratification is essential because the balance among benefit, feasibility, and potential harm varies across preventive options. For example, rectal NSAIDs are relatively straightforward to implement in eligible patients, whereas pancreatic duct stenting requires endoscopic expertise, device availability, and appropriate postprocedural follow-up.

Although the pooled analysis demonstrated no statistical heterogeneity, this should not be interpreted as evidence that all preventive interventions are interchangeable. The included studies combined pharmacological prophylaxis, pancreatic duct stenting, and traditional Chinese medicine or other interventions. These strategies differ with respect to mechanism, timing, clinical indication, operator dependence, cost, and nursing workload. Statistical homogeneity may partly reflect low event rates and limited power to detect between-study differences. Therefore, the overall pooled estimate is valuable for demonstrating the importance of prevention, whereas subgroup-specific interpretation may provide more clinically meaningful insights than a single undifferentiated conclusion.

The subgroup findings should also be interpreted cautiously. Most included trials evaluated NSAID-based or other pharmacological prophylaxis, and this category contributed the majority of the total sample size and statistical weight. Pancreatic duct stenting demonstrated a favorable effect in the available trials; however, this strategy is generally applicable to high-risk procedural situations and cannot be generalized to all ERCP patients. The traditional Chinese medicine/other intervention subgroup contained only one study in the available dataset; consequently, heterogeneity could not be calculated, and any conclusions regarding this category should be regarded as exploratory. For a publishable manuscript, this limitation should be explicitly acknowledged to avoid overstating subgroup evidence.

A central issue in this manuscript is the distinction between clinical prophylactic strategies and nursing interventions. The pooled effect should not be interpreted as the isolated effect of nursing care because the included trials primarily evaluated NSAID administration, pancreatic duct stenting, and related medical interventions[17]. The contribution of nursing is better understood as an implementation and safety function. Nurses play important roles in identifying high-risk patients, verifying contraindications, ensuring the correct timing and route of prescribed prophylaxis, coordinating perioperative preparation, monitoring postprocedural symptoms, detecting early clinical deterioration, supporting hydration and dietary progression, and educating patients regarding warning signs. These functions may determine whether evidence-based preventive measures are delivered consistently in routine clinical practice.

For perioperative nursing, the findings suggest several practical considerations. First, pre-ERCP assessment should be standardized rather than based solely on informal experience. A checklist should document prior PEP, history of pancreatitis, suspected sphincter of Oddi dysfunction, allergy history, renal function, risk of gastrointestinal bleeding, anticoagulant or antiplatelet use, pregnancy status when applicable, and baseline abdominal symptoms. Second, medication workflows should clearly define responsibility for prescription verification, preparation, administration timing, and documentation. Third, for patients undergoing pancreatic duct stenting, nursing care should include device-related education, monitoring for abdominal pain or fever, coordination of follow-up imaging or endoscopic removal when indicated, and prompt escalation when spontaneous stent passage is uncertain.

Postprocedural monitoring represents another area in which nursing practice can translate preventive evidence into enhanced patient safety. Early PEP may manifest as persistent or worsening epigastric pain, nausea, vomiting, abdominal distension, fever, tachycardia, or delayed recovery following sedation. A symptom-triggered monitoring algorithm may assist nurses in determining when to notify physicians, request serum amylase or lipase testing, maintain fasting status, modify fluid management, and intensify clinical observation. Such algorithms may be particularly useful in busy endoscopy units, where early symptoms can be underestimated or misattributed to transient postprocedural discomfort.

This article also highlights a reporting gap in the existing literature. Few trials reported nursing process indicators, such as adherence to prophylaxis timing, completeness of risk assessment, quality of patient education, monitoring frequency, escalation time, or protocol deviations. Without these variables, it remains difficult to determine whether outcomes were influenced by the preventive intervention itself, implementation quality, or co-interventions. Future ERCP prevention studies should therefore incorporate process measures and implementation outcomes alongside clinical endpoints, including PEP incidence, hyperamylasemia, length of hospital stay, adverse drug reactions, readmission, and patient satisfaction.

Strengths: This study has several strengths. First, to our knowledge, this is the first systematic review and meta-analysis to explicitly frame PEP preventive strategies within a perioperative nursing context, bridging the gap between clinical prophylaxis evidence and nursing implementation. Second, the comprehensive search strategy covered both Chinese and international databases (China National Knowledge Infrastructure, Wanfang, VIP Database for Chinese Technical Periodicals, China Biology Medicine disc, Chinese Clinical Trial Registry, PubMed, Web of Science, and the Cochrane Library), minimizing language and regional publication bias. Third, the pooled analysis demonstrated no statistical heterogeneity (I² = 0%, P = 0.77), supporting the robustness of the overall effect estimate. Fourth, the review was prospectively registered with PROSPERO (CRD420261431165) and reported in accordance with PRISMA 2020 guidelines, enhancing methodological transparency and reproducibility.

Several limitations should be acknowledged. First, the included evidence was predominantly derived from single-center studies, and some trials had relatively small sample sizes. Second, allocation concealment, blinding of personnel, and blinding of outcome assessment were frequently judged as unclear or high risk, potentially introducing performance or detection bias. Third, most studies originated from Chinese clinical settings, which may limit generalizability to centers with different ERCP volumes, sedation practices, nursing staffing patterns, and postprocedural observation systems. Fourth, because the available dataset did not include complete study protocols or a formal Grading of Recommendations Assessment, Development and Evaluation assessment, the certainty of the evidence could not be fully established. Finally, combining clinically diverse interventions within a single meta-analysis may oversimplify the complexity of PEP prevention.

The significant funnel plot asymmetry we observed (P = 0.011) does raise the concern that publication bias may be present - for instance, studies with non-significant or unfavorable results might be underrepresented in the literature, which could inflate our pooled estimate of a 62% risk reduction. However, asymmetry isn’t always synonymous with publication bias; it can also arise from genuine differences in study size, intervention types, or methodological weaknesses. In our case, statistical heterogeneity was negligible (I² = 0%), but many included trials had unclear allocation concealment and blinding, and these limitations could themselves produce exaggerated treatment effects. So, while the Egger test is statistically significant, we think it's prudent to treat the overall evidence with some caution. To mitigate this problem in future work, we strongly encourage prospective trial registration and the publication of all results, regardless of their direction or statistical significance.

Future research should move beyond demonstrating that PEP prevention is effective and focus instead on how preventive measures can be implemented consistently. Multicenter randomized trials, pragmatic implementation studies, and quality-improvement initiatives are needed to evaluate standardized ERCP nursing pathways. These studies should compare structured risk assessment, prophylaxis timing protocols, stent-care coordination, hydration monitoring, and discharge education with routine practice. In addition, cost-effectiveness and nursing workload should be assessed, as prevention pathways can only be sustained when they are clinically beneficial, operationally feasible, and compatible with real-world staffing conditions

CONCLUSION

Preventive strategies are associated with a substantial reduction in the risk of PEP[18]. The available evidence supports the development of standardized perioperative nursing pathways that promote accurate risk assessment, timely prophylaxis, coordinated procedural care, vigilant postprocedural monitoring, and patient-centered education. Because the pooled effect primarily reflects pharmacological and endoscopic preventive strategies, conclusions regarding the effects of nursing should remain implementation-focused and appropriately conservative.

ACKNOWLEDGEMENTS

The authors thank Bing-Tao Liu, a graduate student in the Department of Gastroenterology at Eastern Hepatobiliary Surgery Hospital, for his selfless and invaluable assistance with the manuscript submission process. All authors have reviewed and approved the final manuscript.