Safety and indications for endoscopic retrograde cholangiopancreatography in liver transplant patients: an analysis of the United States’ National Inpatient Sample database

Zahid Ijaz Tarara, Umer Farooqb, Mustafa Gandhia, Muhammad Usman Zafarc, Saad Saleemd, Faisal Kamale

University of Missouri School of Medicine, Columbia, MO; Rochester General Hospital, NY; Lehigh Valley Health Network, Allentown, PA; Sunrise Hospital and Medical Center, Las Vegas, NY; Thomas Jefferson University Hospital, PA, USA

aDepartment of Medicine, University of Missouri School of Medicine, Columbia, MO (Zahid Ijaz Tarar, Mustafa Gandhi); bDepartment of Medicine, Rochester General Hospital, Rochester, NY (Umer Farooq); cLehigh Valley Health Network, Allentown, PA (Muhammad Usman Zafar); dDepartment of Medicine, Sunrise Hospital and Medical Center, Las Vegas, NV (Saad Saleem); eDepartment of Gastroenterology and Hepatology, Thomas Jefferson University Hospital, PA (Faisal Kamal), USA

Correspondence to: Zahid Ijaz Tarar, MD, Department of Medicine, University of Missouri, Columbia, MO, USA, e-mail: drzahidejaz@hotmail.com
Received 15 December 2022; accepted 28 February 2023; published online 25 May 2023
DOI: https://doi.org/10.20524/aog.2023.0801
© 2023 Hellenic Society of Gastroenterology

Abstract

Background Endoscopic retrograde cholangiopancreatography (ERCP) is performed to treat biliary complications after a liver transplantation; however, the previously available literature on the safety of ERCP in liver transplant patients is limited. We aimed to study the safety of ERCP in liver transplant patients.

Methods We used a National Inpatient Sample database from 2016-2019 to identify patients who underwent ERCP and had a history of a liver transplantation, using the international classification of diseases, 10th revision. Multivariate logistic regression analysis was conducted to determine the odds of post-ERCP complications in liver transplant recipients.

Results Liver transplant patients who underwent ERCP had a higher rate of post-ERCP pancreatitis and bleeding compared to the general adult population (11.39% vs. 9.19%, 0.83% vs. 0.53%, respectively). However, the adjusted odds of post-ERCP pancreatitis (adjusted odds ratio [aOR] 1.13, 95% confidence interval [CI] 0.86-1.49; P=0.36) and bleeding (aOR 1.41, 95%CI 0.58-3.46; P=0.45) were similar in both the liver transplant and no-transplant groups. There was no difference in the odds of post-ERCP cholangitis (aOR 1.26, 95%CI 0.80-2.01; P=0.32), and sepsis (aOR 0.94, 95%CI 0.66-1.34; P=0.76) between liver transplant and no transplant groups. Biliary stricture was the most common indication for ERCP in the liver transplant group, whereas choledocholithiasis was the main reason for ERCP in the general adult population.

Conclusions ERCP is a safe procedure for treating biliary complications in liver transplant patients. The odds of post-ERCP complications (pancreatitis, bleeding, sepsis, cholangitis) in liver transplant patients are comparable to those in patients with no transplantation.

Keywords Endoscopic retrograde cholangiopancreatography, liver transplant, pancreatitis, cholangitis, bleeding

Ann Gastroenterol 2023; 36 (4): 459-465

Introduction

Liver transplantation, first performed in 1963, is on a rising trend worldwide, resulting in significant improvements in the management of end-stage liver disease. Although surgical techniques and post-transplant care have been improving, the rate of complications is relatively high [1]. Biliary complications represent a substantial cause of post-liver transplant morbidity and mortality, affecting 6-34% of patients [2]. The most common biliary complications reported are biliary leaks and stricture formation. The bile leaks usually develop within 3 months of transplantation and are reported in 1-25% of cases, whereas stricture is a late complication with a reported frequency of 3-15% in deceased donor liver transplant recipients and about 28-32% in living liver donor cases [3-5]. The other notable biliary complications seen in liver transplant recipients are the formation of stones, casts, sludge, fistula, and sphincter of Oddi dysfunction [1,6].

Endoscopic retrograde cholangiopancreatography (ERCP) has emerged as the cornerstone of management for biliary complications in patients with liver transplantation, especially bile leaks and bile duct strictures, with a success rate of more than 90% [7,8]. However, ERCP entails the risk of adverse events, such as pancreatitis, cholangitis, perforation, bleeding and infection, in both general and liver transplant patient populations [6].

ERCP in post-liver transplant patients is considered safe. However, the existing data on ERCP complications in liver transplant patients lack consensus, while studies have shown a varying incidence of complications associated with ERCP [9,10]. Therefore, we analyzed the data from the National Inpatient Sample (NIS) and assessed the burden of complications in liver transplant recipients undergoing ERCP compared to the general adult population. We further investigated the most common indications for ERCP in both groups.

Materials and methods

Study design

This is a retrospective analysis of the data obtained from NIS, the largest publicly available all-payer database maintained by the healthcare cost and utilization agency (HCUP) [11]. The data include one primary and 39 secondary discharge diagnoses and 25 procedure codes, as well as other patient- and hospital-level information. The patient-level information includes age, sex, race, insurance status and median household income in the patient’s zip code, while the hospital-level information is arranged in 4 hierarchical categories: hospital location, size, teaching status and ownership. A 20% probability sample of patients from all hospitals is collected. Each discharge is then weighted (weight=total number of discharges from all acute care hospitals in the United States divided by the number of discharges included in the 20% sample), making it nationally representative. We recorded up to 40 discharge diagnoses and 25 procedures. The dataset from 2016-2019 consists of more than 7 million weighted discharges each year, which is a 20% stratified sample from over 4500 nonfederal acute care hospitals in more than 48 states of the United States. This is equivalent to about 35 million yearly discharges nationwide when weighted and representative of 95% of hospital discharges nationwide.

Study population

We analyzed the NIS from 2016-2019 and recruited all patients who underwent diagnostic or therapeutic ERCP. We used the international classification of diseases, 10th revision (ICD-10), to identify these patients. We further subdivided the study population into 2 groups based on the history of a liver transplantation. We excluded patients under the age of 18 and those admitted for a new liver transplantation.

Study outcomes and variable definition

The outcomes of interest were post-ERCP pancreatitis, bleeding, and acute cholangitis. We also investigated the most common causes of ERCP in both groups. We also analyzed the resource utilization, including mean length of stay and total hospital charges.

Statistical analysis

Analysis was performed using STATA version 17. Categorical variables were compared using the chi-square test, and a t-test was performed to find the difference between continuous variables. Multivariate analysis was conducted to obtain adjusted odds ratios. Variables included in the multivariate regression model were age, race/ethnicity, hospital teaching status, hospital size, hospital location, median household income, insurance status, Charlson’s comorbidity index, and history of chronic kidney disease, cerebrovascular accident, coronary artery disease, congestive heart failure, hyperlipidemia, obesity, alcohol use, cigarette smoking or malnutrition. These variables were selected as confounders in multivariable regression analysis based on the significant association seen in univariate analysis with a cutoff P-value of 0.2. Logistic regression analysis was performed for categorical variables (post-ERCP pancreatitis, bleeding, acute cholangitis, infection/sepsis). In contrast, linear regression was used for continuous variables (mean length of stay and total charges). Deyo’s modification of the Charlson comorbidity index was used to assess the comorbidity burden and was included in regression models [12].

Results

Patient characteristics

There were 28 million hospital discharges between the years 2016-2019, of which 580,040 received diagnostic or therapeutic ERCP. Five thousand three hundred ninety-five patients were under age 18, so they were excluded from the study. Of the adult patients who underwent ERCP, 0.53% (3030) had a history of a liver transplantation. Fig. 1 shows the inclusion and exclusion criteria of the study population. 66.34% patients in the transplant group were male compared to 43.52% in the no-transplant group. Most patients in the study and control groups were White (66.43% vs. 67.23%, P=0.51), followed by Hispanic and Blacks. Most patients with a history of liver transplantations had high Charlson comorbidity index scores compared to the control group (57.10% vs. 30.38%, P<0.001). Medicare insured most patients in both transplant and no-transplant groups (47.19% vs. 49.57%, P<0.001), followed by private/self-insurance. The liver transplant patients underwent their ERCP predominantly in large urban teaching hospitals; similar trends were seen in the patients without transplant history (Table 1).

thumblarge

Figure 1 Criteria for the study population. Patients were selected from National Inpatient Sample using ICD-10 codes

ERCP, endoscopic retrograde cholangiopancreatography

Table 1 Baseline characteristics of the study population

thumblarge

Liver transplant patients were more likely to have diabetes mellitus, cerebrovascular accidents, chronic kidney disease, and malnutrition. Dyslipidemia, congestive heart failure, coronary artery disease, and hypertension were present more often in patients without a liver transplantation (Fig. 2).

thumblarge

Figure 2 Disease burden in the study population

*P<0.05

Post-ERCP complications

The overall incidence of post-ERCP pancreatitis in liver transplant recipients was 11.39%, compared to 9.19% in the general population. On multivariate regression analysis, the odds of post-ERCP pancreatitis in transplant patients were 13% more, but these results did not achieve statistical significance (adjusted odds ratio [aOR] 1.13, 95% confidence interval [CI] 0.86-1.49; P=0.36). Post-ERCP bleeding was seen in 0.83% of transplant patients, while only 0.53% of non-transplant patients developed bleeding. On adjusted analysis, higher odds of bleeding were seen in transplant patients, but the results were non-significant (aOR 1.41, 95%CI 0.58-3.46; P=0.45). Of the general adult population, 2.40% developed acute cholangitis after the ERCP, while 3.45% of liver transplant patients had acute cholangitis. On multivariate regression analysis, no significant difference in cholangitis was seen in both groups (aOR 1.26, 95%CI 0.80-2.01; P=0.32). Liver transplant patients had more cases of post-ERCP sepsis (6.27% vs. 4.58%) in comparison to the general patient population, but the adjusted odds were similar in both study groups (aOR 0.94, 95%CI 0.66-1.34; P=0.76) (Table 2, Fig. 3).

Table 2 Adjusted odds of post-ERCP complication in liver transplant patients

thumblarge
thumblarge

Figure 3 Post-ERCP complication rate in liver transplant vs. no-transplant group

ERCP, endoscopic retrograde cholangiopancreatography

Indications for ERCP

Indications for ERCP in both groups were similar but with different frequencies. The main presentation for ERCP in the liver transplant group was biliary stricture. In that group 44.71% of patients underwent ERCP for biliary stricture, while only 18.99% in the no-transplant group underwent the ERCP for stricture. In contrast, 46.63% of patients in the no-transplant population underwent ERCP for choledocholithiasis, while only 10.56% of patients in the transplant group underwent ERCP for that reason. Acute cholangitis was the reason for ERCP in 25.74% of liver transplant patients, whereas 14.64% of patients without liver transplant history underwent ERCP for cholangitis. ERCP was performed for bile leak in 8.25% of cases in the transplant group, compared to only 3.94% in the no-transplant group. Biliary acute pancreatitis was a much more common reason for ERCP in the no-transplant group compared to transplant patients (13.74% vs. 1.00%) (Fig. 4A, B).

thumblarge

Figure 4 (A) Indications for ERCP in liver transplant patients. (B) Indications for ERCP in patients without a liver transplant

ERCP, endoscopic retrograde cholangiopancreatography

Complications in patients who underwent biliary sphincterotomy for biliary stricture

We performed a subgroup analysis in patients who underwent ERCP with biliary sphincterotomy for biliary stricture and found that liver transplant patients with biliary stricture had significantly higher rates of post-ERCP pancreatitis (16.9% vs. 10.6%, P=0.001), bleeding (9.3% vs. 5.6%, P=0.002), cholangitis (15.8% vs. 11.6%, P=0.02) and sepsis (2.6% vs. 0.47%, P=0.001) compared to the patient group without a liver transplant (Fig. 5).

thumblarge

Figure 5 Post-ERCP complications in biliary stricture patients in liver transplant vs. no-transplant group

ERCP, endoscopic retrograde cholangiopancreatography

Discussion

In the current study, we investigated ERCP complications in liver transplant recipients compared to the general adult population. The study demonstrates that post-ERCP complications, such as bleeding, pancreatitis, cholangitis and sepsis/infections were comparable to those in the general adult population. This study also determined that the most common indication for ERCP in the liver transplant group was biliary stricture, whereas the predominant indication for ERCP in the no-transplant group was choledocholithiasis. To the best of our knowledge, this is the largest study conducted on a nationally representative cohort to determine the safety of ERCP in liver transplant recipients.

The safety of ERCP in patients with liver transplant history is a matter of debate, given the limited published literature on the topic. Previous studies assessing the risk of ERCP complications in transplant recipients yielded conflicting evidence. Our study’s overall rate of post-ERCP pancreatitis in liver transplant recipients was 11.39%, compared to 9.19% in the general adult population. In our study, we found non-significantly higher odds of post-ERCP pancreatitis in liver transplant patients, similar to the analysis performed by Catron et al, who concluded that transplant patients had a higher risk of developing pancreatitis, but the results did not achieve significance. However, Li et al documented higher odds of post-ERCP pancreatitis in liver transplant patients, and their results were statistically significant [13,14]. Contrary to our results, Singh et al reported lower trends of post-ERCP pancreatitis in liver transplant patients [15]. Our results are also reinforced by those of Alomari et al, who conducted a meta-analysis and reported non-significantly higher pooled odds of post-ERCP pancreatitis in liver transplant patients [16].

The overall rate of bleeding in our study population was 0.83%, compared to 0.53% in the general adult population. On multivariate regression analysis, we did not find any difference in the odds of post-ERCP bleeding between liver transplant and no-transplant group which is in accordance with the result found in the study performed by Sanna et al [17].

ERCP carries a high risk of infection compared to other diagnostic procedures, and in the case of transplant recipients on long-term immunosuppressive agents, this risk increases even further [7,18]. In view of the high risk of infection, the American College of Gastroenterology recommends antibiotic prophylaxis in all liver transplant recipients undergoing ERCP [19]. Previously Kohli et al conducted a retrospective analysis of liver transplant patients and found that the odds of post-ERCP infection in liver transplant recipients are low [7]. We also found that 3.45% of liver transplant patients developed post-ERCP cholangitis, similar to the incidences of 4.5% and 3.3% reported in the previous studies by Ambrus et al and Law et al [20,21]. Furthermore, we found that the odds of post-ERCP cholangitis in liver transplant patients are comparable to the general adult population, reinforcing previously available studies [10,16].

We also noted that indications to perform ERCP were significantly different between the liver transplantation and the general adult population. Bile duct anastomotic stricture, bile leak and cholangitis are commonly reported indications in liver transplant patients, while choledocholithiasis, biliary pancreatitis and cholangitis are the most common indications in the general adult population for ERCP, according to previous studies [22-24]. Like these studies, bile duct stricture was the most common indication for ERCP in our study in transplant patients, while choledocholithiasis was the main reason for performing ERCP in the general adult population.

There are various shortcomings to our study. First, the retrospective nature of our study meant that the exposure was not entirely randomized. However, we employed multivariate regression, including diverse patient and hospital-level characteristics in the model to control for confounders. Second, using ICD-10 codes instead of clinical parameters can result in misclassification of the diagnosis. Third, data on laboratory parameters are not stored in the NIS; therefore, the severity of complications could not be accounted for. Regardless of these constraints, our study has numerous strengths. We used the NIS database, which includes data on patients and various hospital-level characteristics from more than 45 states, resulting in better external validity and generalizability; hence we believe that the results should reflect the patient population admitted to hospitals across the United States.

Moreover, NIS eliminates the commonly encountered limitation of single-center studies by allowing the use of a nationally representative large sample size. Another striking feature of our study is that we excluded patients admitted for new transplants, given the high comorbidity burden in the peri-transplant period. This resulted in a bias-free assessment of the outcomes.

Summary Box

What is already known:

  • Biliary complications represent a substantial cause of post-liver transplant morbidity and mortality, affecting about 6-34% of patients

  • Biliary stricture and bile leak are the 2 most common complications seen in liver transplant recipients

  • Endoscopic retrograde cholangiopancreatography (ERCP) is the main treatment modality for biliary complications in liver transplant recipients

What the new findings are:

  • In our analysis we found that ERCP is safe in live transplant recipients

  • No difference in the adjusted odds of post-ERCP pancreatitis, bleeding, cholangitis and sepsis was seen between the liver transplant and no-transplant groups

  • Biliary stricture was the most common indication for ERCP in liver transplant patients

  • Biliary stricture patients who underwent sphincterotomy have a higher rate of complications in the liver transplant group compared to no-transplant patients

References

1. Kochhar G, Parungao JM, Hanouneh IA, Parsi MA. Biliary complications following liver transplantation. World J Gastroenterol 2013;19:2841-2846.

2. Tung BY, Kimmey MB. Biliary complications of orthotopic liver transplantation. Dig Dis 1999;17:133-144.

3. Hampe T, Dogan A, Encke J, et al. Biliary complications after liver transplantation. Clin Transplant 2006;20 Suppl 17:93-96.

4. Londoño MC, Balderramo D, Cárdenas A. Management of biliary complications after orthotopic liver transplantation:the role of endoscopy. World J Gastroenterol 2008;14:493-497.

5. Sharma S, Gurakar A, Jabbour N. Biliary strictures following liver transplantation:past, present and preventive strategies. Liver Transpl 2008;14:759-769.

6. Hüsing A, Cicinnati VR, Maschmeier M, et al. Complications after endoscopic sphincterotomy in liver transplant recipients:A retrospective single-centre study. Arab J Gastroenterol 2015;16:46-49.

7. Kohli DR, Shah TU, BouHaidar DS, Vachhani R, Siddiqui MS. Significant infections in liver transplant recipients undergoing endoscopic retrograde cholangiography are few and unaffected by prophylactic antibiotics. Dig Liver Dis 2018;50:1220-1224.

8. Crismale JF, Ahmad J. Endoscopic management of biliary issues in the liver transplant patient. Gastrointest Endosc Clin N Am 2019;29:237-256.

9. Kobayashi N, Kubota K, Shimamura T, et al. Complications of the treatment of endoscopic biliary strictures developing after liver transplantation. J Hepatobiliary Pancreat Sci 2011;18:202-210.

10. Sanna C, Saracco GM, Reggio D, et al. Endoscopic retrograde cholangiopancreatography in patients with biliary complications after orthotopic liver transplantation:outcomes and complications. Transplant Proc 2009;41:1319-1321.

11. Healthcare Cost and Utilization Project (HCUP) Statistical Briefs [Internet]. Rockville (MD):Agency for Healthcare Research and Quality (US);2006 Feb-. Available from:https://www.ncbi.nlm.nih.gov/books/NBK52651/[Accessed 30 March 2023].

12. Deyo RA, Cherkin DC, Ciol MA. Adapting a clinical comorbidity index for use with ICD-9-CM administrative databases. J Clin Epidemiol 1992;45:613-619.

13. Catron TD, Shipley L, Johnson D, et al. Sa1330 Differences in post-ERCP adverse event rates between post-liver transplant patients and patients with native hepatobiliary anatomy. Gastroint Endosc 2019;89:AB211-AB212.

14. Li GZ, Wang F, Fang J, Zha HL, Zhao Q. Risk factors for post-endoscopic retrograde cholangiopancreatography pancreatitis:evidence from 1786 cases. Med Sci Monit 2018;24:8544-8552.

15. Singh S. Risk of post-ERCP pancreatitis in liver transplant patients:single center experience. Theses &Dissertations 2017. Available from:https://digitalcommons.unmc.edu/etd/240 [Accessed 30 March 2023].

16. Alomari M, Al Momani L, Alomari A, et al. The incidence of endoscopic retrograde cholangiopancreatography-related complications in patients with liver transplant:a meta-analysis and systematic review. Gastroenterology Res 2021;14:259-267.

17. Sanna C, Giordanino C, Giono I, et al. Safety and efficacy of endoscopic retrograde cholangiopancreatography in patients with post-liver transplant biliary complications:results of a cohort study with long-term follow-up. Gut Liver 2011;5:328-334.

18. Fishman JA. Infection in solid-organ transplant recipients. N Engl J Med 2007;357:2601-2614.

19. Chandrasekhara V, Khashab MA, Muthusamy VR, et al;ASGE Standards of Practice Committee. Adverse events associated with ERCP. Gastrointest Endosc 2017;85:32-47.

20. Ambrus RB, Svendsen LB, HillingsøJG, Hansen ML, Achiam MP. Post-endoscopic retrograde cholangiopancreaticography complications in liver transplanted patients, a single-center experience. Scand J Surg 2015;104:86-91.

21. Law R, Leal C, Dayyeh BA, et al. Role of immunosuppression in post-endoscopic retrograde cholangiopancreatography pancreatitis after liver transplantation:a retrospective analysis. Liver Transpl 2013;19:1354-1360.

22. Kröner PT, Bilal M, Samuel R, et al. Use of ERCP in the United States over the past decade. Endosc Int Open 2020;8:E761-E769.

23. Egea Valenzuela J, Jijón Crespín R, Serrano Jiménez A, Alberca de Las Parras F. Endoscopic retrograde cholangiopancreatography in the management of biliary complications after orthotopic liver transplantation. Rev Esp Enferm Dig 2019;111:909-913.

24. Fernández-Simon A, Díaz-Gonzalez A, Thuluvath PJ, Cárdenas A. Endoscopic retrograde cholangiography for biliary anastomotic strictures after liver transplantation. Clin Liver Dis 2014;18:913-926.

Notes

Conflict of Interest: None