Endoscopic retrograde cholangiopancreatography (ERCP) is generally considered to be a safe procedure. Still, there is a substantial risk of complications. Overall, the incidence of post-ERCP pancreatitis (PEP) has been estimated to be 3.4%, with a 3% rate of related mortality.¹ However, the incidence may reach 15% in high-risk patients.² Female sex, young age, sphincter of Oddi dysfunction, and a past history of PEP are all definite or major risk factors.³ Due to the risk of PEP, both endoscopic approaches and pharmacologic therapies have been investigated in an attempt to reduce the occurrence. In high-risk patients, prophylactic pancreatic stent placement has been proven to decrease the chance of developing PEP.⁴ However, in certain instances, the ductal morphology may complicate deep cannulation and stent implantation. The implications of many unsuccessful pancreatic stent implantation procedures are undesirable, since the likelihood of PEP increases significantly.⁵ As such, pharmacologic prophylaxis with nonsteroidal anti-inflammatory medications (NSAIDs) via the rectum is a non-invasive and presumably non-toxic method of avoiding PEP. Rectal diclofenac is a cost-effective, widely accessible medication with an uncomplicated administration technique and a favorable side effect profile, making it an appealing alternative. However, there has been minimal research on the usefulness of NSAIDs in preventing PEP. Rectal diclofenac has been investigated in a few studies, but the majority of these trials included low-risk patients and had a limited sample size. There is currently a dearth of data from Asian populations evaluating rectal diclofenac for the prevention of PEP. The aim of this trial was to determine whether the combination of rectal diclofenac and vigorous hydration with Intravenous Ringer's lactate is superior to each of those individual treatments in preventing PEP in high-risk patients.

Study design

This study was approved by the Institutional Ethical Committee (IEC-NO NIMSUNI/IEC219/22) and informed consent was obtained from all participants.

This randomized, open-label, controlled trial was conducted at the Department of Medical Gastroenterology, National Institute of Medical Science and Research, Jaipur, India from August 2020 to January 2022. In this study, we included patients aged more than 15 years, who were at increased risk of developing PEP. A high risk for PEP was defined as having at least one major or two minor risk factors for PEP (Table 1).⁶ We excluded patients who had a low risk of PEP (i.e., a history of biliary sphincterotomy, routine biliary stent exchange; or chronic pancreatitis), patients who had contraindications for NSAID use (e.g., recent gastrointestinal bleeding), chronic heart failure (New York Heart Association class > 2), hypoxemia (saturated oxygen < 90%), renal failure (glomerular filtration rate < 40 mL/min), liver failure/dysfunction (prolonged international normalized ratio and low albumin level), or evidence of fluid overload (e.g., pulmonary edema, hypo/hypernatremia), as well as pregnant patients.

Table 1 . Risk Factors for Post-Endoscopic Retrograde Cholangiopancreatography Pancreatitis (PEP)⁶.

Major	Minor
Sphincter of Oddi dysfunction	Female patients aged less than 60 years
History of previous PEP	Nondilated CBD (≤ 6 mm)
Pancreatic duct injection	Normal serum bilirubin (< 2 mg/dL)
Freehand needle knife sphincterotomy	Failure to remove all bile duct stones
Balloon sphincter dilation without sphincterotomy	Failed cannulation
More than 1 deep pancreatic guidewire passage	Difficult cannulation (time to CBD cannulation more than 10 minutes or more than 5 attempts at cannulation)

Sample size

We calculated the sample size using the formula N = (Z_α/2 + Z_β) × PQ × 2 /d², where n = sample size, Z_α/2 – Z value at 5% error (1.96), Z_β – Z value at 20% (0.84), P – (p1 + p2)/2, Q – 1 - P. The FLUYT trial by Sperna Weiland et al⁷ reported PEP in 8% of patients in the aggressive hydration with NSAID group and in 9% of patients in the NSAID group. Considering a 10% effect size, we calculated a minimum sample size of 43 patients in each group. To allow for unexpected dropouts and missing data, we increased the sample size by 10%, resulting in 48 patients in each intervention group.

Intervention and randomization

We allocated patients to one of three prophylaxis groups in this study:

Group A: Diclofenac sodium suppository (100 mg) immediately before or after the procedure.

Group B: Aggressive intravenous hydration with Ringer’s lactate at a rate of 3 mL/kg per hour during ERCP, a bolus of 20 mL/kg immediately post-ERCP, followed by a maintenance rate of 3 mL/kg per hour for 8 hours thereafter.

Group C: Diclofenac sodium suppository (100 mg) immediately before or after the procedure, as well as aggressive intravenous hydration with Ringer’s lactate at a rate of 3 mL/kg per hour during the ERCP, a bolus of 20 mL/kg immediately post-ERCP, followed by a maintenance rate of 3 mL/kg per hour for 8 hours thereafter.

Patients were randomized (1 : 1 : 1) to group A, B, or C. The research coordinator used a web-based computer application with hidden, permuted blocks of varied sizes to randomly assign patients (2, 4, or 6). Masking of the treatment personnel and patients was impractical due to the large differences in visible volume intake and urine volume output across groups. However, the data collectors and analysts were blinded to the prophylactic administered.

Data collection and data analysis

We obtained demographic information on patients from their medical records. The medical records were also consulted to obtain patients’ personal history, past medical history, and laboratory investigations. Serum amylase and lipase levels were determined 2 hours after ERCP and repeated after 24 hours. Intra-procedural observations of all patients were noted from the procedure notes. PEP was defined as a serum amylase level more than three times the upper limit of normal with associated epigastric pain and tenderness within 24 hours after ERCP.

Baseline patient data were described as means and frequency distributions. Means were compared between groups A, B, and C using one-way analysis of variance, while frequencies between the groups were compared using the chi-square test. A univariate analysis was performed to assess patient variables associated with the occurrence of PEP. Variables found to have a P-value < 0.25 were included in the multiple regression model to calculate adjusted odds ratios with 95% confidence intervals. A P-value less than 0.05 was considered statistically significant. All data were analyzed using IBM SPSS Statistics ver. 22.0 (IBM Corp., Armonk, NY, USA).

During the study period, we randomized 48 cases to each of the intervention groups. The mean age of the patients was 51.7 years, and 75.7% of the patients were women. A past medical history of hypertension was present in 10.4%. The mean age, sex, smoking, alcohol, and past medical history were not significantly different among the patients in the three intervention groups (Table 2). Similarly, none of the laboratory investigations, including pancreatic enzyme levels at 2 hours and 24 hours post-ERCP, were found to be significantly different among patients in the three intervention groups. We observed that more than one pancreatic duct guidewire cannulation occurred in 29.2% of patients, and biliary sphincterotomy was performed in 93.8%. In addition, cannulation was difficult in 27.8% of patients, needle knife sphincterotomy was performed in 34.7%, all stones could not be removed in 17.4%, a plastic biliary stent was required in 91.7%, and self-expandable metal stents were inserted in 3.5%. Furthermore, we observed common bile duct stones, strictures, and tumors in 86.1%, 1.4% and 9.7% of patients, respectively. No statistically significant difference was found among the three interventional groups with respect to the ERCP findings and observations (Table 3). In our sample of 144 high-risk patients, the overall incidence of PEP was 9%. There was no significant difference among groups A, B, and C (8.3%, 10.4%, and 8.3% respectively, P = 0.92) with respect to the incidence of PEP (Fig. 1). Table 4 describes the factors that were found to be significantly associated with PEP in our sample of patients. Multiple logistic regression analysis showed that a personal history of alcohol consumption (adjusted odds ratio, 2.31; 95% confidence interval, 1.12–3.02, P < 0.05) and more than one pancreatic duct guidewire cannulation (adjusted odds ratio, 2.81; 95% confidence interval, 1.79–4.52, P < 0.05) were found to be significantly associated with the development of PEP (Table 4). Not using hydration was not found to be a significant factor in causing or preventing PEP. None of the patients developed post-surgical complications, hemorrhage, perforation, or fever.

Table 2 . Baseline Characteristics of the Patients Included in the Study.

Patient variable	Intervention group			Total	P-value*
	Group A	Group B	Group C
Mean age (yr)	53.7 ± 16.74	53.38 ± 16.12	48.1 ± 13.6	51.7 ± 15.6	0.14
Sex
Male	15 (31.3)	9 (18.8)	11 (22.9)	35 (24.3)	0.39
Female	33 (68.7)	39 (81.2)	37 (77.1)	109 (75.7)
Personal history
Smoking	14 (29.2)	9 (18.8)	11 (22.9)	34 (23.6)	0.51
Alcohol	2 (4.2)	3 (6.3)	0 (0)	5 (3.5)	0.37
Past medical history
Hypertension	6 (12.5)	7 (14.6)	2 (4.2)	15 (10.4)	0.16
Diabetes mellitus	2 (4.2)	2 (4.2)	1 (2.1)	5 (3.5)	0.79
Coronary artery disease	1 (2.1)	1 (2.1)	0 (0)	2 (1.4)	0.44
Laboratory investigations
Hemoglobin (g%)	10.979 ± 1.76	11.267 ± 1.49	11.61 ± 1.78	11.285 ± 1.69	0.18
Total leucocyte count (/mm3)	9,711.45 ± 4,929.14	8,971.94 ± 5,186.31	9,144.27 ± 5,315.51	9,275.89 ± 5,119.87	0.76
Platelet (/mL)	198,542.04 ± 118,423.17	196,470.83 ± 101,034.32	202,691.92 ± 105,587.56	199,234.93 ± 107,867.34	0.96
Total bilirubin (mg/dL)	3.21 ± 4.27	3.82 ± 6.65	2.82 ± 5.2	3.28 ± 5.4	0.66
Direct bilirubin (mg/dL)	2.36± 3.7	2.81 ± 5.4	1.79 ± 3.6	2.32 ± 4.3	0.52
Indirect bilirubin (mg/dL)	0.85 ± 0.72	1.22 ± 2.7	0.97 ± 1.7	1.01 ± 1.9	0.63
SGOT (U/L)	101.38 ± 97.2	107.71 ± 143.7	102.42 ± 112.9	103.83 ± 118.7	0.96
SGPT (U/L)	143.67 ± 155.4	137.54 ± 150.5	135.44 ± 141.5	138.88 ± 148.2	0.96
Alkaline phosphatase (U/L)	414.81 ± 351.5	330.48 ± 267.02	379.1 ± 336.3	374.8 ± 320.07	0.43
Serum creatinine (mg/dL)	0.75 ± 0.24	0.73 ± 0.25	0.75 ± 0.17	0.74 ± 0.22	0.86
Urea (mg/dL)	24.50 ± 7.7	26.85 ± 9.35	24.38 ± 7.36	25.24 ± 8.2	0.25
Prothrombin time (sec)	13.76 ± 3.59	14.04 ± 2.03	14.24 ± 3.04	14.01 ± 2.95	0.73
International normalized ratio	1.06 ± 0.21	1.05 ± 0.18	1.36 ± 2.08	1.16 ± 1.21	0.36
Pancreatic enzymes
Amylase at 2 hr (U/L)	170.42 ± 184.76	367.1 ± 1,098.86	191.88 ± 351.37	243.13 ± 675.64	0.29
Amylase at 24 hr (U/L)	201.38 ± 211.83	400.48 ± 1,149.18	226.83 ± 370.28	276.23 ± 708.335	0.32
Lipase at 2 hr (U/L)	424.58 ± 595.609	988.27 ± 2,083.164	530.81 ± 1,359.601	647.89 ± 1,486.833	0.14
Lipase at 24 hr (U/L)	510.46 ± 711.822	959.69 ± 2,010.446	549.27 ± 1,317.85	673.14 ± 1,451.688	0.24

Table 3 . ERCP Findings Among Different Intervention Groups.

ERCP findings and observations	Intervention group			Total	P-value*
	Group A	Group B	Group C
History of post-ERCP pancreatitis	0 (0)	0 (0)	0 (0)	0 (0)	NA
>1 pancreatic duct guidewire cannulation	17 (35.4)	10 (20.8)	15 (31.3)	42 (29.2)	0.31
Biliary sphincterotomy	44 (91.7)	45 (93.8)	46 (95.8)	135 (93.8)	0.91
Balloon dilation without sphincterotomy	0 (0)	0 (0)	0 (0)	0 (0)	NA
Failure to remove all stones	11 (22.9)	6 (12.5)	8 (16.7)	25 (17.4)	0.43
Plastic biliary stent	42 (87.5)	46 (95.8)	44 (91.7)	132 (91.7)	0.32
Difficult cannulation	12 (25.0)	16 (33.3)	12 (25)	40 (27.8)	0.58
Needle knife sphincterotomy	16 (33.3)	19 (39.6)	15 (31.3)	50 (34.7)	0.67
Pancreatic sphincterotomy	0 (0)	0 (0)	0 (0)	0 (0)	NA
Non-dilated common bile duct	26 (54.2)	25 (52.1)	26 (54.2)	77 (53.5)	0.97
Self-expandable metal stents	2 (4.2)	2 (4.2)	1 (2.1)	5 (3.5)	0.79
Pancreatic stent	0 (0)	0 (0)	0 (0)	0 (0)	NA
Common bile duct stone	38 (79.2)	41 (85.4)	45 (93.8)	124 (86.1)	0.1
Common bile duct stricture	1 (2.1)	1 (2.1)	0 (0)	2 (1.4)	0.44
Common bile duct tumor	6 (12.5)	5 (10.4)	3 (6.3)	14 (9.7)	0.68
Sphincter of Oddi dysfunction	0 (0)	0 (0)	0 (0)	0 (0)	NA

Table 4 . Factors Associated with the Occurrence of Post-ERCP Pancreatitis.

Variable	Univariate analysis					Multivariate model
	OR	95% CI for OR		P-value*		Adjusted OR	95% CI for OR		P-value
		Lower	Upper				Lower	Upper
Age (18–40 yr)	0.64	0.14	2.91	0.57		NA
Age (41–60 yr)	0.74	0.21	2.72	0.65		NA
Female	4.2	0.52	9.56	0.17		1.1	0.88	2.67	0.54
Smoking	0.96	0.25	3.73	0.96		NA
Alcohol	2.64	0.27	25.61	0.21		2.13	1.12	3.02	< 0.05
Normal bilirubin (≤ 1 mg/dL)	0.84	0.26	2.64	0.17		0.44	0.21	1.84	0.72
> 1 pancreatic duct guidewire cannulation	3.2	1.01	10.17	< 0.05		2.81	1.79	4.52	< 0.05
Biliary sphincterotomy	0.78	0.09	6.77	0.82		NA
Failure to remove all stones	2.32	0.65	8.26	0.19		1.45	0.87	2.05	0.44
Plastic biliary stent	1.1	0.13	9.26	0.93		NA
Difficult cannulation	3.45	1.08	11.04	< 0.05		1.42	0.69	2.36	0.62
Needle knife sphincterotomy	3.39	1.04	10.99	< 0.05		0.58	0.31	2.54	0.34
Non-dilated common bile duct	0.51	0.15	1.64	0.26		NA
Self-expandable metal stents	2.64	0.27	14.6	0.41		NA
Common bile duct stone	0.49	0.12	1.99	0.32		NA
Common bile duct tumor	1.81	0.35	9.1	0.47		NA
No hydration	0.87	0.25	3.01	0.83		NA

Figure 1. Incidence of post-ERCP pancreatitis among different intervention groups. Group A: diclofenac sodium suppository (100 mg), group B: aggressive hydration with Ringer’s lactate, and group C: a combination of diclofenac and aggressive hydration. ERCP, endoscopic retrograde cholangiopancreatography.

The development of PEP is multifactorial and incompletely understood. However, the most widely accepted view is that damage to the papilla results in transient blockage of pancreatic output. Another widely recognized view is that hydrostatic pressure in the pancreatic duct is caused by contrast or saline injection. This lesion triggers an inflammatory cascade that results in the intra-ductal activation of proteolytic enzymes, resulting in pancreatic autodigestion and decreased secretions. This activates the inflammatory cascade, resulting in both local and systemic inflammation.⁸ Numerous prophylactic actions are recommended to prevent PEP.

In the early stages of acute pancreatitis, a considerable volume of fluids (aggressive hydration) is indicated.⁹ Hydration is justified according to the theory that early derangements of pancreatic microcirculatory perfusion correlate with the severity of acute pancreatitis. In comparison to other crystalloid preparations, aggressive intravenous fluid resuscitation (IVFR) using lactated Ringer’s solution may create a more favorable acid-base balance and induce an anti-inflammatory response.¹⁰ The positive impact of IVFR is believed to be greatest during the first 24 hours of acute pancreatitis, when proinflammatory cytokines are believed to produce a variety of physiological changes that result in pancreatic hypoperfusion. Additionally, the majority of patients undergoing ERCP fast for a minimum of 8 to 12 hours before the procedure. This relative volume loss may have an effect on their susceptibility to postoperative pancreatitis. In a study conducted by Park et al¹¹ on 395 patients, the researchers discovered that vigorous hydration resulted in a lower rate of PEP than routine hydration (1.6% vs. 11.6%). Buxbaum et al¹² observed similar findings in a smaller study. Choi et al¹³ discovered that 36 of 510 individuals had PEP (7.1%). The main outcome of PEP was considerably lower (4.3%) in the vigorous hydration group than in the regular hydration group (9.8%) (P = 0.016).

Diclofenac inhibits phospholipase A2, a key mediator of several processes in the inflammatory cascade, hence interrupting this cycle.¹⁴ Rectal diclofenac reaches its peak concentration between 30 and 90 minutes after insertion, with 100% absorption. The half-life of elimination is 2 hours. In a controlled experiment, Hajalikhani et al¹⁵ administered a 100-mg diclofenac sodium suppository along with standard intravenous hydration with Ringer's lactate (1.5 mL/kg/h) during ERCP and continued for 8 hours thereafter. Another group received intensive hydration with Ringer's lactate (3 mL/kg/h) throughout ERCP, followed by a bolus dose of 20 mL/kg/h at the procedure's conclusion and then 3 mL/kg/h for 8 hours after the procedure ended. The authors found that combining preventive NSAID medication with vigorous hydration did not seem to provide further clinically meaningful advantages in avoiding PEP.

In our study, the type of prophylactic intervention had no effect on the incidence of PEP. The aggressive hydration group had the highest incidence (10.4%), whereas the other two groups had a similar incidence of PEP (8.3%); however, the difference was not statistically significant. Mok et al¹⁶ performed a randomized, double-blind, placebo-controlled experiment and found that Ringer’s lactate with rectal indomethacin had a lower incidence of PEP than normal saline with placebo. The patients in that trial received 1 L of Ringer's lactate before ERCP, but in our study, patients received the fluid depending on their weight during ERCP and as a bolus immediately after the procedure. Variations in the amount and timing of fluid administration may account for some of the observed differences in the findings of the trials.¹⁷ Park et al¹¹ also investigated the effects of aggressive fluid treatment in 3 groups of patients: with aggressive Ringer’s lactate, aggressive normal saline, and normal hydration with Ringer’s lactate. Their research demonstrated a significant difference in the PEP rate between patients who received aggressive Ringer’s lactate (3.0%), aggressive normal saline (6.7%), and conventional Ringer’s lactate (11.6%) (P < 0.05).

Wu et al¹⁸ conducted a systematic review and meta-analysis on the efficacy of aggressive hydration with Ringer's lactate. They discovered a lower incidence of PEP (odds ratio = 0.29) and a lower risk of pain (odds ratio = 0.17). Additionally, Zhang et al¹⁹ conducted a meta-analysis of randomized trials on aggressive Ringer’s lactate and discovered that Ringer’s lactate is an effective and safe treatment for PEP prophylaxis. Pasha and colleagues found that vigorous hydration combined with NSAIDs might lower the risk of PEP in the high-risk group, but not to the same degree as in the low-risk group.²⁰

Numerous research studies have evaluated the effects of NSAIDs, either oral or rectal, on reducing the incidence of PEP.²¹ However, the majority of these trials examined the impact of NSAIDs on high-risk patients, with little attention paid to the effectiveness of these therapies in reducing PEP in low-risk patients. Additionally, these studies have infrequently examined the synergistic impact of NSAIDs and vigorous hydration on the incidence of PEP in high-risk patients. PEP was predicted to occur at a rate of roughly 3% to 4% in individuals receiving rectal indomethacin in previous investigations. Aghajanpoor Pasha et al²⁰ administered 100 mg of rectal diclofenac prior to ERCP to both high-risk categories and one of the low-risk arms. Only one patient had severe PEP. Additionally, NSAIDs seem to be useful for reducing the length and severity of PEP, based on prior research. Hosseini et al²¹ observed that concurrent pre-ERCP indomethacin prescription and hydration not only decreased the incidence of PEP (4%) compared to the control group (17%–19%), but also had the potential to lessen the severity of this complication.

We found that alcohol intake and more than one pancreatic duct cannulation were substantially linked with the development of PEP. Choi et al¹³ found that difficult cannulation (odds ratio = 3.9; P < 0.001), pneumatic dilatation of an intact biliary sphincter (odds ratio = 3.9; P < 0.003), and non-use of forceful hydration (odds ratio = 2.4; P < 0.016) were all associated with an increased risk of death. Additionally, intensive hydration proved protective in high-risk individuals for PEP. According to Freeman et al,²² the variables associated with PEP risk included female sex, young age of the patient, suspected sphincter of Oddi dysfunction, difficult cannulation, pancreatic duct injection, and precut sphincterotomy.

This study has a few limitations. First, the treatment personnel and patients were not properly blinded due to practical constraints. Second, the utilization of pancreatic duct stenting is concurrently disputed. As a result, we left the insertion of pancreatic duct stents to the discretion of the treating endoscopist. Finally, we employed strict exclusion criteria to exclude individuals in whom we considered that aggressive hydration could not be safely administered. The strict eligibility restrictions may have impaired this clinical trial's external validity. The study's strengths include the fact that our treatment protocol adhered to worldwide guidelines for preventing PEP, boosting the generalizability of the findings. Additionally, including individuals at a high risk of PEP resulted in a higher incidence of pancreatitis and a lower risk of type II statistical error.

In conclusion, no difference in the incidence of PEP was observed with or without the use of aggressive hydration. At this moment, combining aggressive hydration in patients receiving rectal NSAIDs for preventing PEP cannot be recommended, although the procedure appears to be safe for the patients. A multicentric trial with a larger sample size is recommended to assess the usefulness of combining rectal NSAIDs with aggressive hydration.

None.

No potential conflict of interest relevant to this article was reported.