Impact of pancreatic stent caliber on post-endoscopic retrograde cholangiopancreatogram pancreatitis rates in patients with confirmed sphincter of Oddi dysfunction

Abstract

Background

Despite proven clinical benefit, there are no studies that have examined the relationship between pancreatic stent caliber and its impact on PEP [post-endoscopic retrograde cholangiopancreatogram (ERCP) pancreatitis] in high-risk patients.

Aim

To study the relationship between stent caliber and PEP rates in patients with confirmed sphincter of Oddi dysfunction (SOD).

Methods

A retrospective review was conducted of ERCP’s in patients with SOD from 2002 to 2012 from a prospectively maintained, Institutional Review Board approved database.

Results

A total of 243/7659 (3.2%) patients underwent 3Fr or 5Fr pancreatic stent placement following sphincterotomy for manometry-proven SOD. Of these, 133 (54.7%) underwent 3Fr stent placement, while 110 (45.3%) underwent 5Fr stent placement. There was no significant difference between the two groups in terms of baseline characteristics, demographics, and previous cholecystectomy. Cannulation and stent placement success rates were 100% in both groups. There was no significant difference in rates of PEP and overall complications, 12% versus 12.7%; P = 0.89 and 13.5% versus 15.5%; P = 0.54, between the 3Fr and 5Fr cohorts, respectively. There were more mild PEP rates recorded in 5Fr group (93% vs 56% P = 0.0549) that was not statistically significant.

Conclusions

There appears to be no relationship between stent characteristics and the risk or severity of PEP in patients with manometrically proven SOD.

Background

Diagnosis of sphincter of Oddi dysfunction (SOD) is based on pancreatobiliary manometry and therapy involves endoscopic sphincter ablation performed at endoscopic retrograde cholangiopancreatogram (ERCP). Among all patients undergoing ERCP, this group is acknowledged to be at the highest risk for post-ERCP pancreatitis (PEP).^1,2 Several studies have shown the protective effect of prophylactic pancreatic duct (PD) stent placement^3,4 and meta-analyses have confirmed a consistent clear benefit in such high-risk individuals.⁵

Although not fully understood, the reason for PEP can be due to patient/procedure-related factors or a combination of both. Pathophysiological factors thought to contribute to the development of pancreatitis post-ERCP include thermal (cautery),⁶ chemical (contrast injection),^7–9 and mechanical (cannulation attempts, guidewire manipulation)¹⁰ injury causing ampullary edema.² This obstruction to the flow of pancreatic secretions causes ductal hypertension and subsequent acinar injury setting off a cascade of events culminating in clinical and biochemical pancreatitis. While it is generally accepted that placement of a pancreatic stent for a short period of time reduces the risk by ensuring uninterrupted flow of pancreatic secretions from the duct into the duodenum, there is still considerable debate regarding the optimal size and caliber of pancreatic stents required.

The benefit of pancreatic stent placement is weighed against the technical challenges of placing the stent, the risk of inducing chronic pancreatitis type ductal changes and the consequences of failure to stent placement.⁴ Whereas a prospective study concluded that placement of 3Fr stent can be time-consuming, requires more accessories, and is more technically challenging;¹¹ conflicting results are shown in studies with 3Fr and 5Fr stent placements for PEP outcome. A retrospective study showed reduced rates of PEP with 3Fr stents and two prospective randomized trials showed no statistical difference between the two sized stents. These reports, however, included a heterogeneous population with inclusion of subjects with differing degrees of attributable risk of PEP. This heterogeneity makes direct comparison of studies problematic and absolute recommendation for the appropriate caliber of prophylactic pancreatic stenting for SOD patients impossible.

This retrospective study evaluated the effect of stent caliber on the rates of PEP in patients undergoing pancreatic stent placements for manometrically confirmed SOD.

Methods

Patients

All patients undergoing ERCP were entered prospectively in an Institutional Review Board-approved database (#X030409001). Data were reviewed retrospectively from January 2002 to June 2013. Variables maintained in the database include demographics, indications, pre-ERCP investigations, procedure-related details, and outcomes. All patients who underwent ERCP and manometry that confirmed SOD (biliary or pancreatic) and who had placement of 3Fr or 5Fr stents were included (Fig. 1). Excluded were those patients in whom cannulation was unsuccessful, PD stent placement was unsuccessful or not performed and those who underwent placement of other caliber stents.

Figure 1.

Flowchart showing selection of patients and frequency of PEP [post-endoscopic retrograde cholangiopancreatogram (ERCP) pancreatitis] between the two groups.

Procedure

The procedure was performed by using standard side-viewing duodenoscopes (Olympus Medical Systems, Center Valley, PA, USA). Endoscopic SOM was performed under conscious sedation or general anesthesia using a previously described technique.¹² When SOD was diagnosed based on manometry readings, sphincter ablation and PD stent placement were undertaken.

Sphincterotomy and PD stent placement

If the sphincter pressures were elevated in both the biliary and pancreatic segments of the sphincter of Oddi, a dual sphincterotomy was undertaken. If the pressures were elevated in only segment, the corresponding segment was ablated. Sphincterotomy was performed by using an ERBE generator (ERBE USA, Marietta, GA, USA) in Endocut mode. While biliary sphincterotomy was performed using a standard papillotome, pancreatic sphincterotomy was performed using a standard papillotome or needle-knife catheter after placement of a prophylactic PD stent. PD stents (Zimmon pancreatic stents, Cook Medical, Winston-Salem, NC, USA) placed were either unflanged, 3Fr (8, 10, 12 cms in length) or a 5Fr 2 cm single pigtail stent with multiple internal side holes and a single flange. Stent caliber used was based on individual preference of the operator.

Follow up

Most patients were admitted overnight for observation and discharged the following day unless they developed PEP. A nurse coordinator contacted all patients by telephone call at 24 h and 30 days to assess for complications that was graded per consensus criteria.¹³ Patient and referring physicians were advised to obtain an abdomen flat plate X-ray in 4 weeks to assess for spontaneous stent passage; if the stents were still in situ, a standard gastroscopy was performed for stent retrieval.

Outcome measures

Primary outcome measure was to compare the rate and severity of PEP, as defined by consensus criteria, between patients undergoing 3Fr and 5Fr PD stent placement. Secondary outcome measures were to evaluate the overall complication rates between the two groups. Furthermore, we sought to evaluate for other confounding variables that may impact PEP rates in patients with SOD.

Statistical analysis

Chi-squared test and Fisher’s exact test were used in comparisons of categorical variables between groups. Because age for each group was normally distributed and had equal variance, Student’s t-test was conducted to test their mean difference. Mann–Whitney U-test was performed assessing whether one of two groups tends to have larger values than the other for liver function test results, which were not normally distributed. The Cochran–Armitage test was used to test for trends in the proportion of PEP over study period. Univariate logistic regression analyses were conducted to examine the association between the risk factors and the occurrence of PEP, but all P-values for univariate testing are uncorrected for multiple testing because those tests were taken as exploratory. All P-values less than 0.05 were considered to be statistically significant. Analyses were done using SAS software, version 9.3 (SAS Institute Inc., Cary, NC, USA).

Results

Demographics

Of a total of 7659 ERCP’s performed between 1/1/2002 and 6/30/2013, 536 procedures were undertaken to evaluate for SOD of which 330 (61.5%) were confirmed to have manometry-proven disease. In this group, we identified patients who underwent 3Fr (n = 133) or 5Fr (n = 110) stent placements who constitute the study cohort. A total of 93 patients were excluded because of lack of technical details, follow-up data on complications, or because of placement of other caliber stents. There was no statistical difference between the two groups for age, gender, race, prior cholecystectomy, jaundice, abnormality in liver enzymes, or cholangitis. There was significant difference in the proportion of patients with pancreatobiliary pain and history of acute pancreatitis (Table 1).

Table 1.

Shows the baseline characteristics and findings at ERCP for the two groups

	Abnormal (n = 243)‡		P-value
	3Fr (n = 133)	5Fr (n = 110)
Age‡	43.2 ± 16.1 (45)	45.8 ± 15.2 (44)	0.2099
Female	106 (79.7%)	85 (77.3%)	0.6462
White	114 (86.4%)	102 (92.7%)	0.3392
Prior cholecystectomy	110 (82.7%)	91 (82.7%)	0.9966
History of pancreatic/biliary pain	91 (68.4%)	46 (41.8%)	0.0001†
History of jaundice	1 (0.8%)	—	> 0.9999
History of cholangitis	—	—	—
History of pancreatitis	35 (26.3%)	7 (6.4%)	< 0.0001†
Bilirubin‡	0.7 ± 0.4 (0.7)	0.6 ± 0.3 (0.6)	0.1286
AST‡	52.9 ± 81.8 (26)	31.3 ± 29.7 (22)	0.1307
ALT‡	52.2 ± 100.2 (25)	36.1 ± 39 (25)	0.8824
Alk Phos‡	79 ± 35.5 (73)	97.1 ± 52.6 (89)	0.0879
GGT‡	91.9 ± 146.7 (46)	123.6 ± 140.9 (75)	0.4962
EBS	83 (62.4%)	58 (52.7%)	0.1281
EPS	118 (88.7%)	97 (88.2%)	0.8956

^†The proportions between two groups (3Fr vs 5Fr) are significantly different at the level of 0.05.

^‡Mean ± SD (median).

ALT, alanine aminotransferase; AST, aspartate amiotransferase; EBS, endoscopic biliary sphincterotomy; EPS, endoscopic pancreatic sphincterotomy; ERCP, endoscopic retrograde cholangiopancreatogram; GGT, gamma glutamyl transferase.

Complications

There was no difference in PEP rates between the 3Fr and 5Fr stent cohorts, 12% versus 12.7%; P = 0.8694, respectively. While 13 of 14 (93%) patients had mild PEP in the 5Fr group, 9 of 16 (56%) patients in the 3Fr group developed mild PEP (P = 0.0549). Also, while none of the patients in the 5Fr cohort developed severe pancreatitis, one patient (7%) in the 3Fr group experienced a severe episode (Table 2). There was also no significant trend or difference (P = 0.1675) in the rates of PEP in relation to annual procedural volume over the 12-year study period.

Table 2.

Enlists the complications post ERCP between the two groups

	Abnormal (n = 243)		P-value
	3Fr (n = 133)	5Fr (n = 110)
Pancreatitis	16 (12%)	14 (12.7%)	0.8694
Mild	9/16	13/14	0.0549
Moderate	6/16	1/14	—
Severe	1/16	—	—
Cholangitis	—	1 (0.9%)	> 0.9999
Bleeding	—	—	—
Perforation	1 (0.8%)	2 (1.8%)	0.5935
Other	1 (0.8%)	—	> 0.9999
Total	18 (13.5%)	17 (15.5%)	0.5172

ERCP, endoscopic retrograde cholangiopancreatogram.

Factors evaluated for risk of PEP

Univariate analysis conducted to examine for confounding risk factors for the occurrence of PEP for the whole group showed a significant risk for patients less than 40 years of age. Other parameters studied including gender, race, prior cholecystectomy, history of acute pancreatitis, normal bilirubin, normal liver enzymes, endoscopic biliary sphincterotomy, pancreatic sphincterotomy, dual sphincterotomy, stent size, and the time period before and after 2005 did not show a significant association (Table 3).

Table 3.

Shows the results of univariate analysis for 243 patients of factors associated with post-ERCP pancreatitis between the two groups

	PEP incidence (%)		P-value
	Present	Absent
Age ≤ 40 years versus > 40 years	19.8	7.5	0.0059†
Female	13.6	7.7	0.2566
White	12.5	11.1	0.7463
Prior cholecystectomy	13	9.5	0.5431
History of acute pancreatitis	11.5	14	0.5734
Normal LFT result (n = 81)	21.7	14.3	0.3953
EBS	12.8	11.8	0.8149
EPS	12.6	10.7	0.7806
Dual sphincterotomy	12	12.7	0.8661
Stent size (3Fr vs 5Fr)	12	12.7	0.8694
Prior to year 2005 versus after 2005	14.5	11.6	0.5479

^†The proportions of PEP between two groups are significantly different at the level of 0.05.

EBS, endoscopic biliary sphincterotomy; EPS, endoscopic pancreatic sphincterotomy; ERCP, endoscopic retrograde cholangiopancreatogram; LFT, liver function test; PEP, post-ERCP pancreatitis.

Discussion

The present study demonstrates that the caliber of pancreatic stents, whether 3Fr or 5Fr, does not impact the rates of PEP.

There are multiple studies that have examined the effect of different caliber pancreatic stents in high-risk patients undergoing ERCP. In a randomized trial¹⁴ examining the short 5Fr versus long 3Fr stents for patients with high risk of pancreatitis, while a trend toward lower risk of PEP was observed with 5Fr stents, a higher rate of spontaneous stent dislodgement and stent placement failures were observed with 3Fr stents. Another randomized trial noted that when compared with 3Fr stents, placement of 5Fr stents was technically easier, faster, and required fewer guide wires.¹⁵ In the present study, although we did not document the technical difficulty associated with stent placement, we observed no difference in rates or severity of PEP and overall complications between both cohorts of patients. The current study, in contrast to other series included exclusively patients with manometry-proven SOD. Hence, majority of our study subjects had normal, small caliber PDs with no dilation, strictures, or chronic pancreatitis and therefore constituted a designated cohort at the highest risk for PEP.

Although the exact mechanism of PEP is multifactorial, pancreatic ductal stenting reduces risk by ductal decompression and continuous, uninterrupted drainage of pancreatic secretions. While in clinical practice, long 3Fr stents are preferred by some because of its thin caliber (less likely to induce ductal injury) and the span to avoid the genu area (most vulnerable part for ductitis);¹⁶ others prefer 5Fr stents as its placement is relatively easy, avoids the genu area with no “foreign body” contact with the rest of the ductal mucosa. Our data, although, confirm that caliber of the pancreatic stents do not impact PEP rates in SOD cohort; definitive or long-term follow-up data to examine for effects on the ductal mucosa with either stent need further study.

Pharmacological intervention is becoming increasingly popular as a measure against PEP in high-risk patients. A randomized control trial¹⁷ showed the usefulness of rectal indomethacin for reducing PEP in high-risk patients undergoing ERCP. A network meta-analysis¹⁸ and a post-hoc efficacy and cost-effective analysis¹⁹ indicate that rectal indomethacin may be more effective than PD stenting alone and more cost-effective than PD stenting or combination therapy. In this select group of high-risk patients, it may be too premature to abandon the placement of pancreatic stents unless there is robust randomized control data to suggest the sole use of indomethacin as a prophylactic measure. Since the new randomized trial data,¹⁷ we have introduced a protocol to ensure all SOD patients receive rectal indomethacin 100 mg along with PD stent placement as standard treatment.

There are several limitations to our study. First, this is a retrospective single-center experience that may not be applicable to all centers and is subject to biases. Second, we did not classify the SOD subtypes, and the risks associated with the corresponding groups were not examined. Third, because the experience spans more than a decade, accessories and techniques have altered over this time period and the study was not designed to account for those factors. Fourth, the study does not report the migration rates of individual stents and its effect on PEP while a spontaneously migrating stent maybe an option.²⁰ Fifth, although there is 100% success rate with stent placement in this study, we did not document the technical difficulty associated with individual stent placement and it is unclear how many patients encountered failed 3Fr stent placement and were crossed over to 5Fr stent placement.

In conclusion, our study demonstrates that there is no relationship between PD stent caliber, whether 3Fr or 5Fr, and the risk or severity of pancreatitis in patients with manometrically confirmed SOD.