ORIGINAL ARTICLE
Transpapillary Gallbladder Stents Can Stabilize or Improve Decompensated Cirrhosis in Patients Awaiting Liver Transplantation Shannan R. Tujios, MD,* Sahand Rahnama-Moghadam, MD,w Joseph B. Elmunzer, MD,z Richard Kwon, MD,z Amit G. Singal, MD,* Michelle A. Anderson, MD,z Erik-Jan Wamsteker, MD,z Jason R. Taylor, MD,y James Scheiman, MD,z Grace Elta, MD,z Robert J. Fontana, MD,z and Cyrus R. Piraka, MD8
Goals: To describe the short-term and long-term outcomes in 34 consecutive decompensated cirrhotic patients with symptomatic gallbladder disease undergoing transpapillary gallbladder stent (TGS) placement. Background: Endoscopic TGS placement is a minimally invasive means of treating symptomatic gallbladder disease in poor surgical candidates. Study: Between June 2005 and June 2011, 34 patients with cirrhosis and symptomatic gallbladder disease underwent attempted TGS placement. Results: Median patient age was 52 years, 56% were hospitalized, and 48% were listed for liver transplantation. The median model for end-stage liver disease (MELD) score was 15 (range, 6 to 40) and 88% were Child-Turcotte-Pugh class B/C. A double pigtailed stent was successfully placed in 94% of the patients. At 1-month follow-up, clinical improvement was noted in 82% of the treated subjects and the MELD scores in 14 of 22 (64%) evaluable subjects improved or stabilized. Actuarial transplant-free survival was 53% in the liver transplant candidates with a mean follow-up of 352 days, whereas survival was 44% in the 18 nontransplant candidates with a mean follow-up of 1.5 years. Periprocedural complications included pancreatitis in 5 patients, cholangitis in 3, and 1 patient with cystic duct perforation. In addition, 2 subjects had symptomatic bleeding from traumatic duodenal ulcerations 2 years after TGS placement that necessitated stent removal.
Received for publication July 21, 2014; accepted October 19, 2014. From the *Department of Internal Medicine, Division of Digestive and Liver Diseases, University of Texas Southwestern Medical Center, Dallas; wDepartment of Internal Medicine, University of Texas at San Antonio, San Antonio, TX; zDepartment of Internal Medicine, Division of Gastroenterology, University of Michigan, Ann Arbor; 8Department of Internal Medicine, Division of Gastroenterology, Henry Ford Hospital, Detroit, MI; and yDepartment of Gastroenterology & Hepatology, St Louis University, St Louis, MO. All authors had full access to the data and approved the final version of the manuscript. S.R.T. and R.J.F.: concept, analysis, and writing; C.R.P.: concept, analysis, and editing; A.G.S. statistics analysis and editing; S.R.-M., J.B.E., R.K., M.A.A., E.-J.W., J.R.T., J.S., and G.E.: data capture, analysis, and editing. R.J.F. has received grant funding from Vertex and Gilead and consulting for GlaxoSmithKline and Tibotec. G.E. is a consultant for Olympus Medical. A.G.S. is a consultant for Onyx and Bayer. The remaining authors declare that they have nothing to disclose. Reprints: Robert J. Fontana, MD, Department of Internal Medicine, Division of Gastroenterology, University of Michigan, 3912 Taubman Center, Ann Arbor, MI 48109-0362 (e-mail: [email protected]). Copyright r 2014 Wolters Kluwer Health, Inc. All rights reserved.
J Clin Gastroenterol
Volume 49, Number 9, October 2015
Conclusions: Endoscopic TGS placement was technically feasible in 94% of decompensated cirrhotics and was associated with a relatively low rate of periprocedural (26%) and long-term complications (6%). Stabilization or improvement in clinical status and MELD scores was seen in the majority of treated patients. Key Words: cholelithiasis, cholecystitis, cirrhosis, endoscopic retrograde cholangiopancreatography, cystic duct
(J Clin Gastroenterol 2015;49:771–777)
C
holelithiasis frequently develops in patients with cirrhosis. Mechanisms that enhance stone formation include impaired hepatic transport of bile salts, gallbladder dysmotility secondary to elevated estrogens that enhance bile lithogenicity, and the formation of black pigment stones due to hyperbilirubinemia.1 The prevalence of cholelithiasis also increases as liver disease worsens with 24% of Child-Turcotte-Pugh (CTP) class A cirrhotics harboring stones compared with 37% of CTP class C patients.2,3 The standard of care for patients with symptomatic gallbladder disease is cholecystectomy, as studies have shown that the development of complications such as acute cholecystitis, biliary obstruction, cholangitis, and pancreatitis are more frequent in symptomatic patients.1 However, patients with advanced liver disease and symptomatic gallstones have an increased risk of bleeding and infectious complications with cholecystectomy due to their underlying coagulopathy and portal hypertension. In addition, these patients are at increased risk of clinical decompensation and death due to the adverse effects of general anesthesia and laparotomy on their hepatic function.4 Although laparoscopic cholecystectomy can be applied to highly selected patients with early and wellcompensated cirrhosis, overall morbidity (21%) and mortality (< 1%) remain higher than in patients without liver disease.5 Percutaneous transhepatic gallbladder drainage through placement of a cholecystostomy tube is frequently used to treat acute cholecystitis in nonsurgical patients and is associated with a reduction in overall mortality from 30% to 15%.6 However, bleeding and infectious complications are more likely to be encountered in patients with cirrhosis given the presence of ascites and coagulopathy. Furthermore, this approach has primarily been used as a temporizing measure to allow recovery from sepsis or optimization of underlying medical conditions before undergoing cholecystectomy.7 www.jcge.com |
Copyright r 2015 Wolters Kluwer Health, Inc. All rights reserved.
771
J Clin Gastroenterol
Tujios et al
Endoscopic biliary sphincterotomy in patients with choledocholithiasis can help prevent gallstone pancreatitis but carries a higher complication rate in patients with cirrhosis and more importantly has limited efficacy in those with biliary colic and cholecystitis.8 Recently, endoscopically placed transpapillary gallbladder stents (TGS) have been used to manage cirrhotic patients and other poor surgical candidates with symptomatic gallbladder disease.9–14 In a recent systematic review of TGS in 127 patients with various contraindications to surgery, 88% of the patients experienced short-term improvement and only 6% reported adverse events.15 However, these high-risk and technically challenging endoscopic therapies are not widely available; reports are limited to small retrospective case series with variable data on the indications, complications, and long-term follow-up. The aims of the current study were to determine the technical success and complication rates of TGS in consecutive nonoperative patients with endstage liver disease managed at a single liver transplant (LT) center using an intention-to-treat approach. In this manuscript, we describe the presenting characteristics and indications for TGS, the short-term and long-term clinical and laboratory outcomes in 34 decompensated patients with cirrhosis, and the potential role that TGS may have in LT candidates with symptomatic gallbladder disease.
PATIENTS AND METHODS Patients This study was conducted after receiving approval from the University of Michigan Institutional Review Board. A retrospective review of patients undergoing endoscopic retrograde cholangiopancreatography (ERCP) with gallbladder stent placement between June 1, 2005 and June 1, 2011 was performed. Patients were identified using the ProVation endoscopy database using the search terms “ERCP, gallbladder, gallbladder stent, GB stent, cystic duct, cholecystitis, biliary colic, biliary pancreatitis, and gallstone pancreatitis,” which initially yielded 158 unique procedures. After chart review, 52 patients were identified that underwent a total of 77 ERCPs for TGS. In this analysis, the indications and outcomes of the 34 patients with end-stage liver disease who underwent TGS are reported. Data abstracted during chart review included: patient demographics (age, sex, race), clinical characteristics (hospitalization status, comorbidities, etiology of cirrhosis, CTP score), and procedural characteristics (indication, need for blood products, antibiotics given, sedation type, need for sphincterotomy, cholecystostomy wire or cholangioscopy, stent diameter and length). In addition, available laboratory data at the time of the procedure, 4 weeks, 3 months, 6 months, and 1 year later were reviewed [white blood cell count, platelets, albumin, aspartate aminotransferase, alanine aminotransferase, alkaline phosphatase, total bilirubin, creatinine, international normalized ratio (INR), calculated model for end-stage liver disease (MELD)]. Finally, radiographic data [ultrasound, CT scan, hepatobiliary imino-diacetic acid (HIDA) scan, cholangiogram], symptom relief, complications, and long-term outcomes (stent removal, cholecystostomy, cholecystectomy, orthotopic LT, or death) were recorded.
Procedures Eight advanced endoscopists at a single tertiary referral center performed all of the ERCPs to place TGS
Volume 49, Number 9, October 2015
using a previously described technique.13 Three patients underwent a rendezvous procedure with interventional radiology inserting a percutaneous cholecystostomy tube and advancing a hydrophilic guidewire through the gallbladder and cystic duct into the duodenum to facilitate cystic duct access.16 In 6 cases where the cystic duct could not be accessed initially, SpyGlass cholangioscopy (Boston Scientific, Boston, MA) helped identify the cystic duct orifice and allow for guidewire advancement under direct visualization followed by TGS placement in 5 patients. Technical success was defined by placement of a double pigtail 5 or 7 Fr catheter that was 7 to 18 cm in length into the gallbladder from the duodenum. Stent exchanges were not routinely performed.
Outcomes Patients were followed after TGS with cholecystectomy, orthotopic LT, death, or the last available follow-up through June 1, 2011. Clinical success was defined as resolution of their symptoms of epigastric pain, nausea and vomiting, or biliary infection. Stent migration and procedural complications were reported per patient. The MELD score was calculated at the time of the procedure, 1 month, 3 months, 6 months, 1 year, and then annually thereafter.
Statistics Data are presented as median with ranges for nonnormally distributed variables and mean with SD for normally distributed variables. Statistical analysis included w2 test for categorical measures and paired Student t test to assess change over time. Kaplan-Meier survival curves were used to calculate transplant-free survival in the LT candidates and nontransplant patients. Patients were censored at death, transplant, cholecystectomy, or the end of the study period. All P < 0.05 were considered statistically significant.
RESULTS Baseline Patient Characteristics Thirty-four patients with symptomatic gallbladder disease underwent endoscopic TGS placement (Table 1). The median age of the patients was 52 years, 53% were male, and 44% were listed for liver transplantation. The most common cause of liver disease was chronic hepatitis C (35%) followed by alcoholic liver disease (26%). Both the listed and nonlisted patients were very ill with over half of them hospitalized at the time of procedure and requiring intravenous antibiotics and pain medications at the time of ERCP. The median MELD score was 15 (range, 6 to 40) with 53% CTP class C, 35% CTP class B, and 12% CTP class A. Psychosocial issues (32%) and other comorbidities (42%) were the most common contraindications for transplant listing. Nearly a third of the patients had previously well-compensated or stable liver disease before their presentation of symptomatic gallbladder disease. The most common indication for TGS placement was recurrent biliary colic (41%) followed by biliary pancreatitis (21%) and acute cholecystitis (21%). The endoscopic TGS helped facilitate the removal of cholecystostomy tubes previously inserted for acute cholecystitis in 5 patients (15%). Liver imaging aided in the diagnosis of gallbladder pathology in all of the patients with the majority of patients undergoing ultrasound (85%), abdominal CT scans (53%), HIDA
772 | www.jcge.com Copyright r 2014 Wolters Kluwer Health, Inc. All rights reserved. Copyright r 2015 Wolters Kluwer Health, Inc. All rights reserved.
J Clin Gastroenterol
Volume 49, Number 9, October 2015
TABLE 1. Baseline Characteristics of Study Population
Characteristic
Listed for Not Listed for LT (n = 15) LT (n = 19)
Age (y) 52 (21-60) Sex (male) 9 (60) Race White 10 (67) Black 2 (13) Asian 2 (13) Hispanic 1 (7) 28.8 ± 5.6 Body mass index (mean ± SD) 2 (kg/m ) Indication for ERCP Biliary colic 4 (27) Acute cholecystitis 1 (7) Cholecystostomy 2 (13) tube removal Acalculous 3 (20) cholecystitis Biliary pancreatitis 5 (33) Choledocholithiasis 0 Status Outpatient 6 (40) Inpatient 8 (53) ICU 1 (7) Etiology of cirrhosis Hepatitis C 7 (47) Alcohol 4 (27) Nonalcoholic 1 (7) steatohepatitis Primary sclerosing 1 (7) cholangitis Other liver diseases 2 (14) CTP score 9 (6-13) CTP class A 1 (7) CTP class B 8 (53) CTP class C 6 (40) Median MELD 16 (12-29) Baseline laboratories 5.6 (2.5-12.9) WBC (103/mL) 63 (25-175) Platelets ( 103/mL) INR 1.4 (1.1-2.4) Serum AST (U/L) 59 (15-141) Serum ALT (U/L) 35 (19-72) Alkaline 114 (63-467) phosphatase (U/ L) Bilirubin (mg/dL) 4.7 (1.6-33.9)
TABLE 2. Transpapillary Gallbladder Stent Procedures
Total (n = 34)
51 (28-69) 9 (47)
52 (21-69) 18 (53)
14 (74) 4 (21) 1 (5) 0 31.3 ± 12.1
24 (71) 6 (18) 3 (9) 1 (3) 30.2 ± 9.8
10 (53) 3 (16) 3 (16) 0
14 (41) 4 (12) 5 (15) 3 (9)
2 (11) 1 (5)
7 (21) 1 (3)
8 (42) 11 (58) 0
14 (41) 19 (56) 1 (3)
5 (26) 5 (26) 4 (21)
12 (35) 9 (26) 5 (15)
2 (11)
3 (9)
3 10 3 4 12 15
(16) (5-14) (16) (21) (63) (6-40)
5 10 4 12 18 15
(15) (5-14) (12) (35) (53) (6-40)
5.2 (1.6-16.7) 5.4 (1.6-16.7) 104 (11-182) 68 (11-182) 1.5 61 46 140
(0.9-2.5) 1.45 (0.9-2.5) (17-226) 60 (15-226) (7-117) 40 (7-117) (57-750) 123 (57-750)
3.1 (0.3-19.4) 3.1 (0.3-33.9)
Data presented as median (range) or n (%). ALT indicates alanine aminotransferase; AST, aspartate aminotransferase; CTP, Child-Turcotte-Pugh; ERCP, endoscopic retrograde cholangiopancreatography; ICU, intensive care unit; LT, liver transplant; MELD, model for end-stage liver disease; WBC, white blood cell count.
scans (29%), or magnetic resonance cholangiopancreatography (3%) immediately before TGS placement.
Procedural Characteristics A TGS was successfully placed in 94% of the patients with most (76%) on the first attempt (Table 2). Only 10 patients (29%) underwent a sphincterotomy and there was no significant postsphincterotomy bleeding in any of these patients. Cholangioscopy was used in 6 cases and a Copyright
r
Gallbladder Stents in Decompensated Cirrhosis
n (%) Procedure
Listed for LT (n = 15)
Successful TGS 14 (93) placement 1 attempt 11 (73) > 1 attempt 3 (20) Sphincterotomy 5 (33) Cholangioscopy 3 (20) Cholecystostomy 1 (7) wire Preoperative 9 (60) antibiotics Preoperative 2 (13) platelets Preoperative fresh8 (53) frozen plasma Sedation Conscious 8 (53) MAC 1 (7) General 6 (40) anesthesia Endobiliary stent diameter (Fr) 5 3 (20) 7 11 (73) Stent length (cm) 7 0 10 0 12 2 (13) 15 9 (60) 18 3 (20)
Not Listed for LT (n = 19)
Total (n = 34)
18 (95)
32 (94)
15 3 5 3 2
26 6 10 6 3
(79) (16) (26) (16) (11)
(76) (18) (29) (18) (9)
14 (74)
23 (68)
3 (16)
5 (15)
4 (21)
12 (35)
10 (53) 3 (16) 6 (32)
18 (53) 4 (12) 12 (35)
7 (37) 11 (58)
10 (31) 22 (69)
1 1 4 10 2
1 1 6 19 5
(5) (5) (21) (53) (11)
(3) (3) (19) (59) (16)
LT indicates liver transplant; TGS, transpapillary gallbladder stent.
cholecystostomy wire aided cystic duct access in 3 patients. Of the 6 patients that required >1 ERCP attempt, 3 were successful using cholangioscopy, 1 required a cholecystostomy wire placed by interventional radiology, 1 was successful after biliary sphincterotomy, and 1 was successful with repeat ERCP. The most common ERCP finding was cholelithiasis in 22 patients. An additional 4 patients had gallstones on ultrasound, 1 on MRI, and 1 patient with normal imaging and ERCP whose symptoms did not improve after TGS placement. The 6 remaining patients without gallstones seen on ERCP underwent TGS placement for acalculous cholecystitis or removal of cholecystostomy tube previously placed for acalculous cholecystitis. Both of the patients with unsuccessful TGS placement were hospitalized with acute cholecystitis. A tortuous, inflamed cystic duct with perforation recognized during the procedure was noted in 1 patient and later required cholecystostomy tube placement. This nontransplant candidate with morbid obesity died 90 days later from sepsis, variceal bleeding, and multiorgan failure. The other patient had an obstructing cystic duct stone that was managed with a cholecystostomy tube. He eventually underwent open cholecystectomy 54 days later and despite initial decompensation after surgery did well and is now awaiting LT at 4 months of follow-up. Prophylactic fresh-frozen plasma was administered before the procedure in 33% of the patients who had a median INR of 1.7 (range, 1.3 to 2.5) and 15% also required platelets for a median platelet count of 28,000/mL
2014 Wolters Kluwer Health, Inc. All rights reserved.
www.jcge.com |
Copyright r 2015 Wolters Kluwer Health, Inc. All rights reserved.
773
J Clin Gastroenterol
Tujios et al
TABLE 3. Outcomes and Complications After Endobiliary Stent Placement*
35
11 2 1 1 4
(73) (13) (7) (7) (27)
5 7 3 3
(33) (47)z (20) (100)
15 (79) 12 3 1 1 10
30
28 (82)
(63) (16) (5) (5) (53)
23 5 2 2 14
(68) (15) (6) (6) (44)
11 (58) 8 (42) 5 (63)
5 7 11 8
(15) (53) (32) (73)
5 11 5 2 3 1 1
(15) (32) (15) (6) (9) (3) (3)
MELD
13 (87)
Listed Patients 40
Listed for LT Not Listed for LT Total (n = 15) (n = 19) (n = 34)
Clinical symptoms resolved Complete Partial Cholecystectomy Cholecystostomyw Stent in place >1 y Transplanted Alive at last visit Died Liver-related death Stent complications Stent removal Complications Pancreatitis Ulcer Cholangitis Biliary stricture Cystic duct perforation Variceal bleed
Volume 49, Number 9, October 2015
A
n (%) Variable
25
20
15
10
5 Time of TGS
1 month
B 1 (7) 3 (20) 2 (13) 0 0 0 0 1 (7)
4 8 3 2 3 1 1
(21) (42) (16) (11) (16) (5) (5) 0
1 (3)
*Follow-up calculated until liver transplant, death, or last available follow-up through June 2011. wDoes not include patients with cholecystostomy before gallbladder stent. zDoes not include the 5 patients who are alive posttransplant. LT indicates liver transplant.
(range, 11,000/mL to 44,000/mL). The majority (68%) of patients were on antibiotics at the time of TGS placement. Half of the procedures were completed under conscious sedation with the remainder performed under propofol (12%) or general endotracheal anesthesia (35%), although sedation practice changed over time with all cases being performed with anesthesia monitoring after 2009. The length of the endobiliary stent placed was at the discretion of the endoscopist but intended to result in 1 pigtail in the gallbladder lumen (rather than the cystic duct) and 1 within the duodenum. A 15 cm 7 Fr double pigtailed stent was most commonly placed (35%) and a prophylactic pancreatic stent was not placed in any of the subjects.
Clinical Efficacy Long-term Outcomes In the 32 patients where TGS was technically successful, 88% demonstrated clinical improvement with 72% experiencing no further symptoms of gallbladder disease during a median follow-up of 9 months (mean 13 ± 14.2 mo) (Table 3). In addition, 44% have remained asymptomatic with the TGS in place for over 1 year. Follow-up laboratory MELD scores at 1 month after TGS placement in 22 evaluable patients demonstrated a reduction or stabilization compared with baseline scores in 14 of the patients [MELD 16 vs. 18, median improvement 2.5 points (0 to 7 points)] (Fig. 1).
FIGURE 1. MELD scores at the time of TGS placement and 1 month later. Eleven patients listed for liver transplantation (A) and 11 nonlisted patients with advanced liver disease (B). MELD indicates model for end-stage liver disease; TGS, transpapillary gallbladder stent.
Of the 15 patients listed for transplant, 5 (33%) underwent uncomplicated LT and 7 (47%) are still alive on the waitlist, whereas 3 (20%) died of progressive liver failure at 24, 34, and 173 days after TGS placement. The median MELD score of the patients still awaiting transplant 1 month to 1 year after TGS placement is 12. One listed patient is alive with stable liver function 5.4 years after initial TGS placement. The cirrhotic patients not listed for transplant have had their TGS in place longer than those listed (median 14.7 vs. 2.9 mo), with 10 patients having stents in place for over 1 year. In the 8 patients with MELD scores available at 1 year, 6 remained stable or improved with median MELD score of 13. One patient with continued abdominal pain underwent an open cholecystectomy 2 years after TGS placement with a complicated postoperative course but did well now 21 months after surgery. During follow-up, 8 of the nonlisted patients died with most of these deaths due to liver disease progression. The median time to death after TGS was 8.7 months and varied from 7 days to 2 years. The overall actuarial survival of these patients was 76% at 1 year and 58% at 2 years. Using competing risk analysis,
774 | www.jcge.com Copyright r 2014 Wolters Kluwer Health, Inc. All rights reserved. Copyright r 2015 Wolters Kluwer Health, Inc. All rights reserved.
J Clin Gastroenterol
Volume 49, Number 9, October 2015
FIGURE 2. Transplant-free survival in the 19 patients listed for liver transplantation and 15 nontransplant candidates. During follow-up, the transplant-free survival rate was similar (P = 0.29).
there was no difference in survival between listed and nonlisted patients (P = 0.29). The 1- and 2-year transplantfree survival rates in the listed patients were both 72%, compared with 78% and 54% in the nonlisted patients (Fig. 2).
Complications A total of 14 complications occurred in 11 patients (32%), with pancreatitis most commonly reported. The majority of these complications (11 of 14) developed within 48 hours of the TGS placement, including a single episode of variceal bleeding. One patient who had not received antibiotics had both post-ERCP pancreatitis and cholangitis. Four patients required stent exchange; 3 patients had the initial TGS coil in the cystic duct and 1 patient had the stent migrate into the duodenum. The incidence of late complications was low. However, severe bleeding from a duodenal ulcer that arose from traumatic mucosal injury from the pigtail catheter tip was noted in 1 patent 2.2 years after TGS placement. In addition, another nontransplant candidate had a mild common bile duct stricture that responded to balloon cholangioplasty and a nonbleeding duodenal ulcer that was detected at 2.1 years after initial TGS placement. Finally, a third patient developed cholangitis 2 days after replacement of a suspected occluded stent that had been in place for 74 days despite the use of prophylactic antibiotics. The increased frequency of complications in the nonlisted patients (42% vs. 20%, P = 0.18 using ordinal logistic regression) may in part relate to the longer duration of follow-up in these patients (median 509 vs. 96 d, respectively, P = 0.11, Wilcoxon rank-sum test). Patients with complications had a higher body mass index (34 ± 14 vs. 28 ± 6 kg/m2, P = 0.14) and underwent more repeat ERCPs (1.8 + 1.0 vs. 1.4 + 0.7, P = 0.31) compared with those without complications, but this was not statistically significant. However, there was no difference in stent size or length, use of antibiotics, CTP, or MELD scores among those with and without complications.
DISCUSSION Effective treatment of symptomatic gallbladder disease in patients with advanced liver disease has been problematic Copyright
r
Gallbladder Stents in Decompensated Cirrhosis
with open cholecystectomy historically carrying a mortality rate as high as 80%.17 Even with the advent of laparoscopic cholecystectomy, patients with cirrhosis experience more bleeding complications, higher conversion to open cholecystectomy, and more morbidity and mortality than patients without liver disease due to their underlying coagulopathy and portal hypertension.5 In particular, cirrhotic patients with MELD scores >13 seem to be at the greatest risk for perioperative complications and morbidity.18 Furthermore, surgery is rarely, if ever, attempted in Child’s C cirrhotics even in experienced referral centers as supported by the 1992 NIH consensus statement on laparoscopic cholecystectomy.19,20 Nonsurgical therapies for symptomatic gallstones are also limited as methyl tert-butyl ether contact dissolution therapy and mechanical extracorporeal shock wave lithotripsy have been essentially abandoned due to complications and high recurrence rates.21 Oral bile acid therapy also offers little utility in cirrhotic patients with gallbladder dyskinesia and pigment stones as ursodiol is effective only with small cholesterol stones in a functioning gallbladder.21 Biliary sphincterotomy is an effective intervention for recurrent gallstone pancreatitis in poor surgical candidates but will not prevent future episodes of cystic duct obstruction and is unsafe in coagulopathic patients with liver disease.8 Percutaneous cholecystostomy can also be used to treat acute cholecystitis in poor surgical candidates but short-term mortality remains high.22 Kalloo and colleagues first described endoscopic TGS placement in 4 cirrhotic patients in 1994 as a potential alternative palliative treatment. Since then, there have been multiple small case series describing the feasibility and clinical efficacy of endoscopic TGS for poor operative candidates.9–15 However, the technique is not commonly performed even in LT centers and there remains little data on the long-term safety and outcomes with TGS in decompensated cirrhotic patients or LT candidates. Our current study of 34 patients is the largest singlecenter study of endoscopic TGS in cirrhosis that we are aware of. Using an intention-to-treat design, the technical success rate was 94% with a clinical success rate of 82%, confirming that this approach is an effective procedure for carefully selected patients. Our high rate of technical success is similar to that reported by others.15 Over time, we have found that use of guidewires, cholangioscopy, and rendezvous with percutaneous cholecystostomy wires can facilitate endobiliary stenting in these patients. In the 2 cases where TGS was not possible, the cystic duct was either obstructed by a stone or significantly inflamed, highlighting the technical difficulties in attempting to safely cannulate the cystic duct in patients with acute cholecystitis. The majority of patients undergoing endoscopic TGS placement not only had improvement in their clinical status and symptoms but also experienced a stabilization or improvement in their MELD score at 1-month follow-up (Fig. 1). In addition, many of our patients have experienced further improvement in their liver disease status with more prolonged follow-up, including 6 of the waitlisted patients. Most of these patients have not required a stent exchange. Endoscopically placed gallbladder stents act to prevent stone impaction of the cystic duct as well as provide a wick for biliary drainage. As stent patency is not required, smalldiameter stents are preferentially used and routine stent changes do not seem to be necessary. In support of this, a recent prospective study of 16 patients undergoing TGS
2014 Wolters Kluwer Health, Inc. All rights reserved.
www.jcge.com |
Copyright r 2015 Wolters Kluwer Health, Inc. All rights reserved.
775
Tujios et al
noted that stents may be safely left in place for up to 2 years in mostly noncirrhotic patients.23 Although late complications including stent migration, duodenal ulceration, and biliary stricture were observed in our study, selecting appropriate stent length and smaller diameter might avert these. On the basis of our experience and that of other groups, we would suggest careful and frequent laboratory and clinical monitoring of decompensated cirrhotics undergoing TGS, but not recommend routine endobiliary stent change at this time. Periprocedural complications occurred in a third of patients but were generally mild and resolved within a few days with conservative management. Five patients met consensus definition of post-ERCP pancreatitis with increased abdominal pain, elevated amylase, and hospitalization but only 1 case was severe. Identifiable risk factors for post-ERCP pancreatitis included age in one young woman and normal bilirubin in another.24 The relatively high rate of post-ERCP pancreatitis of 15% is likely due to small sample size, with 95% confidence intervals estimated 16.7% (95% CI, 6.0%-30.6%) using the adjusted Wald method. Longer procedure times with multiple cannulations due to the technical challenge of negotiating wire into cystic duct and gallbladder may also contribute to increased incidence of post-ERCP pancreatitis, but we would expect this to improve with endoscopist experience. Rectal indomethacin has been shown to reduce the rate of post-ERCP pancreatitis in high-risk patients; however, indomethacin use should be considered carefully in this patient population given their increased risk of bleeding and their increased susceptibility to renal failure.25 Although most of our patients were receiving antibiotics before and after the procedure, the 1 patient with severe pancreatitis, cholangitis, and sepsis did not receive prophylactic antibiotics. Although there are no clear data, based upon our experience and other series of ERCP in immunosuppressed patients with biliary obstruction and incomplete drainage, we would recommend considering a prophylactic antibiotic regimen such as broad-spectrum cephalosporin or fluoroquinolone to be given to cirrhotic patients undergoing TGS in light of their known increase for severe bacterial infections.26,27 Although 10 of our patients underwent sphincterotomy, we did not encounter significant postspincterotomy bleeding. The low rate of periprocedural bleeding may, in part, be due to the limited size sphincterotomy that we made as well as the selective use of periprocedural plasma and platelet transfusions in 3 patients. We would encourage others to consider using prophylactic plasma to achieve an INR of 50,000/mL if a sphincterotomy is planned and to observe these high-risk patients in house after blood product administration. The role of endoscopic TGS placement in LT candidates requires further study. In some centers, physicians have advocated to obtain MELD exception points for their decompensated cirrhotic patients with cholecystitis similar to patients with recurrent bacterial cholangitis.28,29 However, a review of MELD exception practices demonstrates substantial heterogeneity in the number of points requested and success rates with nonstandard exemptions across centers.29 In our series, 5 patients have already been successfully bridged to liver transplantation and 7 others remain stable on the LT waiting list with a median followup of 63 days. The 3 LT candidates who died had been placed on hold for cardiac and active infectious issues
J Clin Gastroenterol
Volume 49, Number 9, October 2015
unrelated to their gallbladder disease, with eventual progression to hepatorenal syndrome and liver failure. In light of the high rate of short-term clinical improvement we observed in both our listed and nonlisted decompensated cirrhotics, we would encourage others to consider this treatment option for cirrhotic patients who are poor surgical candidates with symptomatic gallbladder disease. However, several potential limitations with our current retrospective series should be noted. First, the number of patients studied was limited and complete follow-up laboratory data were not available in all patients. In addition, functional studies of gallbladder function, such as quantitative HIDA scans, were not routinely obtained before or after TGS to document improved gallbladder function. However, HIDA scans are known to have limited utility in cirrhotic patients and those with elevated bilirubin. Furthermore, over 88% reported improvement in their clinical symptoms during follow-up. In addition, the optimal endobiliary stent to use remains unknown as well as the utility of prophylactic pancreatic duct stenting, rectal indomethacin, and/or bile acid therapy to reduce bile lithogenicity. Finally, the safety and efficacy of endoscopic TGS placement has not been directly compared with percutaneous cholecystostomy tube placement.22 Endoscopic ultrasound-guided gallbladder drainage is another therapeutic option but in the absence of acute cholecystitis, the gallbladder wall may not be tethered to the antrum or duodenal bulb resulting in a reasonably high risk of bile leak or pneumoperitoneum. For these reasons, we believe that TGS has a more favorable risk-benefit ratio than transmural drainage, especially in cirrhotics who are harder to rescue from major complications. Additional research may clarify whether novel “tissue approximating stents” will change our approach to endoscopic therapy in this patient population.30 In light of the relatively low rate of periprocedural complications in our series and others, we would advocate consideration of an endoscopic TGS placement first in decompensated cirrhotics, as they are at increased risk for bleeding, infectious, and fluid leak complications with a percutaneous cholecystostomy tube. In conclusion, our data demonstrates that endoscopic TGS placement is a safe and effective treatment for decompensated cirrhotics with symptomatic gallbladder disease. Although nearly 90% of our patients were CTP class B or C, our rate of periprocedural complications was low and did not lead to worsening liver disease status in most of the treated patients. Our study and other series also suggest that TGS may prove to be an effective long-term management approach to symptomatic gallbladder disease in decompensated cirrhosis patients or perhaps stabilize them for future cholecystectomy. Finally, TGS placement seems to be a safe and effective bridge to liver transplantation.
REFERENCES 1. Thistle JL, Cleary PA, Lachin AM. The natural history of cholelithiasis: The National Cooperative Gallstone Study. Ann Intern Med. 1984;107:171–175. 2. Elzouki AN, Nilsson S, Nilsson P, et al. The prevalence of gallstones in chronic liver disease is related to degree of liver dysfunction. Hepatogastroenterology. 1999;46:2946–2950. 3. Conte D, Fraquelli M, Fornari F, et al. Close relation between cirrhosis and gallstones: cross-sectional and longitudinal survey. Arch Intern Med. 1999;159:49–52.
776 | www.jcge.com Copyright r 2014 Wolters Kluwer Health, Inc. All rights reserved. Copyright r 2015 Wolters Kluwer Health, Inc. All rights reserved.
J Clin Gastroenterol
Volume 49, Number 9, October 2015
4. Bloch R, Allaben R, Walt A. Cholecystectomy in patients with cirrhosis. Arch Surg. 1985;120:669–971. 5. Puggioni A, Wong LL. A metaanalysis of laparoscopic cholecystectomy in patients with cirrhosis. J Am Coll Surg. 2003;197:921–926. 6. Kortram K, de Vries Reilingh TS, Wiezer MJ, et al. Percutaneous drainage for acute calculous cholecystitis. Surg Endosc. 2011;25:3642–3646. 7. Winbladh A, Gullstrand P, Svanvik J, et al. Systematic review of cholecystostomy as a treatment option in acute cholecystitis. HPB. 2009;11:183–193. 8. Freeman ML, Nelson DB, Sherman S, et al. Complications of endoscopic biliary sphincterotomy. N Engl J Med. 1996; 335:909–918. 9. Kalloo AN, Thuluvath PJ, Pasricha PJ. Treatment of high-risk patients with symptomatic cholelithiasis by endoscopic gallbladder stenting. Gastrointest Endosc. 1994;40:608–610. 10. Gaglio PJ, Buniak B, Leevy CB. Primary endoscopic retrograde cholecystoendoprosthesis: a nonsurgical modality for symptomatic cholelithiasis in cirrhotic patients. Gastrointest Endosc. 1996;44:339–342. 11. Shrestha R, Trouillot TE, Everson GT. Endoscopic stenting of the gallbladder for symptomatic gallbladder disease in patients with end-stage liver disease awaiting orthotopic liver transplantation. Liver Transpl Surg. 1999;5:275–281. 12. Conway JD, Russo MW, Shrestha R. Endoscopic stent insertion into the gallbladder for symptomatic gallbladder disease in patients with end-stage liver disease. Gastrointest Endosc. 2005;61:32–36. 13. Schlenker C, Trotter JF, Shah RJ, et al. Endoscopic gallbladder stent placement for treatment of symptomatic cholelithiasis in patients with end-stage liver disease. Am J Gastroenterol. 2006;101:278–283. 14. Pannala R, Peterson BT, Gostout CJ, et al. Endoscopic transpapillary gallbladder drainage: 10-year single center experience. Minerva Gastroenterol Dietol. 2008;54:107–113. 15. Itoi T, Coelho-Prabhu N, Baron TH. Endoscopic gallbladder drainage for management of acute cholecystitis. Gastrointest Endosc. 2010;71:1038–1045. 16. Elmunzer BJ, Novelli PM, Taylor JR, et al. Percutaneous cholecystostomy as a bridge to definitive endoscopic gallbladder stent placement. Clin Gastroenterol Hepatol. 2011;9:18–20.
Copyright
r
Gallbladder Stents in Decompensated Cirrhosis
17. Aranha GV, Sontag SJ, Greenlee HB. Cholecystectomy in cirrhotic patients: a formidable operation. Am J Surg. 1982; 143:55–60. 18. Delis S, Bakoyiannis A, Madariaga J, et al. Laparoscopic cholecystectomy in cirrhotic patients: the value of MELD score and Child-Pugh classification in predicting outcome. Surg Endosc. 2010;24:407–412. 19. Nguyen KT, Kitisin K, Steel J, et al. Cirrhosis is not a contraindication to laparoscopic cholecystectomy: results and practical recommendations. HPB. 2011;13:192–197. 20. NIH consensus conference statement on gallstones and laparoscopic cholecystectomy Am J Surg. 1993;165:390–398. 21. Portincasa P, Moschetta A, Palasciano G. Cholesterol gallstone disease. Lancet. 2006;368:230–239. 22. Hatzidakis AA, Prassopoulos P, Pertinarakis I, et al. Acute cholecystitis in high-risk patients: percutaneous cholecystomy vs. conservative treatment. Eur Radiol. 2002;12: 1778–1784. 23. Lee TH, Park DH, Lee SS, et al. Outcomes of endoscopic transpapillary gallbladder stenting for symptomatic gallbladder diseases: a multicenter prospective follow-up study. Endoscopy. 2011;43:702–708. 24. Anderson MA, Fisher L, Jain R, et al. Complications of ERCP. Gastrointest Endosc. 2012;75:467–473. 25. Elmunzer BJ, Scheiman JM, Lehman GA, et al. A randomized trial of rectal indomethacin to prevent post-ERCP pancreatitis. N Engl J Med. 2012;366:1414–1422. 26. Banerjee S, Shen B, Baron TH, et al. Antibiotic prophylaxis for GI endoscopy. Gastrointest Endosc. 2008;67:791–798. 27. Jalan R, Fernandez J, Wiest R, et al. Bacterial infections in cirrhosis. A position statement based on the EASL special conference 2013. J Hepatol. 2014;60:1310–1324. 28. Goldberg D, Bittermann T, Makar G. Lack of standardization in exception points for patients with primary sclerosing cholangitis and bacterial cholangitis. Am J Transplant. 2012;12:163–1609. 29. Goldberg DS, Makar G, Bittermann T, et al. Center variation in the use of non-standardized model for end-stage liver disease exception points. Liver Transpl. 2013;19:1330–1342. 30. Choi JH, Park DH, Lee SS, et al. Can endoscopic ultrasound help to drain the gallbladder? Gastrointest Interv. 2013;2: 30–35.
2014 Wolters Kluwer Health, Inc. All rights reserved.
www.jcge.com |
Copyright r 2015 Wolters Kluwer Health, Inc. All rights reserved.
777
Transpapillary Gallbladder Stents Can Stabilize or Improve Decompensated Cirrhosis in Patients Awaiting Liver Transplantation Shannan R. Tujios, MD,* Sahand Rahnama-Moghadam, MD,w Joseph B. Elmunzer, MD,z Richard Kwon, MD,z Amit G. Singal, MD,* Michelle A. Anderson, MD,z Erik-Jan Wamsteker, MD,z Jason R. Taylor, MD,y James Scheiman, MD,z Grace Elta, MD,z Robert J. Fontana, MD,z and Cyrus R. Piraka, MD8
Goals: To describe the short-term and long-term outcomes in 34 consecutive decompensated cirrhotic patients with symptomatic gallbladder disease undergoing transpapillary gallbladder stent (TGS) placement. Background: Endoscopic TGS placement is a minimally invasive means of treating symptomatic gallbladder disease in poor surgical candidates. Study: Between June 2005 and June 2011, 34 patients with cirrhosis and symptomatic gallbladder disease underwent attempted TGS placement. Results: Median patient age was 52 years, 56% were hospitalized, and 48% were listed for liver transplantation. The median model for end-stage liver disease (MELD) score was 15 (range, 6 to 40) and 88% were Child-Turcotte-Pugh class B/C. A double pigtailed stent was successfully placed in 94% of the patients. At 1-month follow-up, clinical improvement was noted in 82% of the treated subjects and the MELD scores in 14 of 22 (64%) evaluable subjects improved or stabilized. Actuarial transplant-free survival was 53% in the liver transplant candidates with a mean follow-up of 352 days, whereas survival was 44% in the 18 nontransplant candidates with a mean follow-up of 1.5 years. Periprocedural complications included pancreatitis in 5 patients, cholangitis in 3, and 1 patient with cystic duct perforation. In addition, 2 subjects had symptomatic bleeding from traumatic duodenal ulcerations 2 years after TGS placement that necessitated stent removal.
Received for publication July 21, 2014; accepted October 19, 2014. From the *Department of Internal Medicine, Division of Digestive and Liver Diseases, University of Texas Southwestern Medical Center, Dallas; wDepartment of Internal Medicine, University of Texas at San Antonio, San Antonio, TX; zDepartment of Internal Medicine, Division of Gastroenterology, University of Michigan, Ann Arbor; 8Department of Internal Medicine, Division of Gastroenterology, Henry Ford Hospital, Detroit, MI; and yDepartment of Gastroenterology & Hepatology, St Louis University, St Louis, MO. All authors had full access to the data and approved the final version of the manuscript. S.R.T. and R.J.F.: concept, analysis, and writing; C.R.P.: concept, analysis, and editing; A.G.S. statistics analysis and editing; S.R.-M., J.B.E., R.K., M.A.A., E.-J.W., J.R.T., J.S., and G.E.: data capture, analysis, and editing. R.J.F. has received grant funding from Vertex and Gilead and consulting for GlaxoSmithKline and Tibotec. G.E. is a consultant for Olympus Medical. A.G.S. is a consultant for Onyx and Bayer. The remaining authors declare that they have nothing to disclose. Reprints: Robert J. Fontana, MD, Department of Internal Medicine, Division of Gastroenterology, University of Michigan, 3912 Taubman Center, Ann Arbor, MI 48109-0362 (e-mail: [email protected]). Copyright r 2014 Wolters Kluwer Health, Inc. All rights reserved.
J Clin Gastroenterol
Volume 49, Number 9, October 2015
Conclusions: Endoscopic TGS placement was technically feasible in 94% of decompensated cirrhotics and was associated with a relatively low rate of periprocedural (26%) and long-term complications (6%). Stabilization or improvement in clinical status and MELD scores was seen in the majority of treated patients. Key Words: cholelithiasis, cholecystitis, cirrhosis, endoscopic retrograde cholangiopancreatography, cystic duct
(J Clin Gastroenterol 2015;49:771–777)
C
holelithiasis frequently develops in patients with cirrhosis. Mechanisms that enhance stone formation include impaired hepatic transport of bile salts, gallbladder dysmotility secondary to elevated estrogens that enhance bile lithogenicity, and the formation of black pigment stones due to hyperbilirubinemia.1 The prevalence of cholelithiasis also increases as liver disease worsens with 24% of Child-Turcotte-Pugh (CTP) class A cirrhotics harboring stones compared with 37% of CTP class C patients.2,3 The standard of care for patients with symptomatic gallbladder disease is cholecystectomy, as studies have shown that the development of complications such as acute cholecystitis, biliary obstruction, cholangitis, and pancreatitis are more frequent in symptomatic patients.1 However, patients with advanced liver disease and symptomatic gallstones have an increased risk of bleeding and infectious complications with cholecystectomy due to their underlying coagulopathy and portal hypertension. In addition, these patients are at increased risk of clinical decompensation and death due to the adverse effects of general anesthesia and laparotomy on their hepatic function.4 Although laparoscopic cholecystectomy can be applied to highly selected patients with early and wellcompensated cirrhosis, overall morbidity (21%) and mortality (< 1%) remain higher than in patients without liver disease.5 Percutaneous transhepatic gallbladder drainage through placement of a cholecystostomy tube is frequently used to treat acute cholecystitis in nonsurgical patients and is associated with a reduction in overall mortality from 30% to 15%.6 However, bleeding and infectious complications are more likely to be encountered in patients with cirrhosis given the presence of ascites and coagulopathy. Furthermore, this approach has primarily been used as a temporizing measure to allow recovery from sepsis or optimization of underlying medical conditions before undergoing cholecystectomy.7 www.jcge.com |
Copyright r 2015 Wolters Kluwer Health, Inc. All rights reserved.
771
J Clin Gastroenterol
Tujios et al
Endoscopic biliary sphincterotomy in patients with choledocholithiasis can help prevent gallstone pancreatitis but carries a higher complication rate in patients with cirrhosis and more importantly has limited efficacy in those with biliary colic and cholecystitis.8 Recently, endoscopically placed transpapillary gallbladder stents (TGS) have been used to manage cirrhotic patients and other poor surgical candidates with symptomatic gallbladder disease.9–14 In a recent systematic review of TGS in 127 patients with various contraindications to surgery, 88% of the patients experienced short-term improvement and only 6% reported adverse events.15 However, these high-risk and technically challenging endoscopic therapies are not widely available; reports are limited to small retrospective case series with variable data on the indications, complications, and long-term follow-up. The aims of the current study were to determine the technical success and complication rates of TGS in consecutive nonoperative patients with endstage liver disease managed at a single liver transplant (LT) center using an intention-to-treat approach. In this manuscript, we describe the presenting characteristics and indications for TGS, the short-term and long-term clinical and laboratory outcomes in 34 decompensated patients with cirrhosis, and the potential role that TGS may have in LT candidates with symptomatic gallbladder disease.
PATIENTS AND METHODS Patients This study was conducted after receiving approval from the University of Michigan Institutional Review Board. A retrospective review of patients undergoing endoscopic retrograde cholangiopancreatography (ERCP) with gallbladder stent placement between June 1, 2005 and June 1, 2011 was performed. Patients were identified using the ProVation endoscopy database using the search terms “ERCP, gallbladder, gallbladder stent, GB stent, cystic duct, cholecystitis, biliary colic, biliary pancreatitis, and gallstone pancreatitis,” which initially yielded 158 unique procedures. After chart review, 52 patients were identified that underwent a total of 77 ERCPs for TGS. In this analysis, the indications and outcomes of the 34 patients with end-stage liver disease who underwent TGS are reported. Data abstracted during chart review included: patient demographics (age, sex, race), clinical characteristics (hospitalization status, comorbidities, etiology of cirrhosis, CTP score), and procedural characteristics (indication, need for blood products, antibiotics given, sedation type, need for sphincterotomy, cholecystostomy wire or cholangioscopy, stent diameter and length). In addition, available laboratory data at the time of the procedure, 4 weeks, 3 months, 6 months, and 1 year later were reviewed [white blood cell count, platelets, albumin, aspartate aminotransferase, alanine aminotransferase, alkaline phosphatase, total bilirubin, creatinine, international normalized ratio (INR), calculated model for end-stage liver disease (MELD)]. Finally, radiographic data [ultrasound, CT scan, hepatobiliary imino-diacetic acid (HIDA) scan, cholangiogram], symptom relief, complications, and long-term outcomes (stent removal, cholecystostomy, cholecystectomy, orthotopic LT, or death) were recorded.
Procedures Eight advanced endoscopists at a single tertiary referral center performed all of the ERCPs to place TGS
Volume 49, Number 9, October 2015
using a previously described technique.13 Three patients underwent a rendezvous procedure with interventional radiology inserting a percutaneous cholecystostomy tube and advancing a hydrophilic guidewire through the gallbladder and cystic duct into the duodenum to facilitate cystic duct access.16 In 6 cases where the cystic duct could not be accessed initially, SpyGlass cholangioscopy (Boston Scientific, Boston, MA) helped identify the cystic duct orifice and allow for guidewire advancement under direct visualization followed by TGS placement in 5 patients. Technical success was defined by placement of a double pigtail 5 or 7 Fr catheter that was 7 to 18 cm in length into the gallbladder from the duodenum. Stent exchanges were not routinely performed.
Outcomes Patients were followed after TGS with cholecystectomy, orthotopic LT, death, or the last available follow-up through June 1, 2011. Clinical success was defined as resolution of their symptoms of epigastric pain, nausea and vomiting, or biliary infection. Stent migration and procedural complications were reported per patient. The MELD score was calculated at the time of the procedure, 1 month, 3 months, 6 months, 1 year, and then annually thereafter.
Statistics Data are presented as median with ranges for nonnormally distributed variables and mean with SD for normally distributed variables. Statistical analysis included w2 test for categorical measures and paired Student t test to assess change over time. Kaplan-Meier survival curves were used to calculate transplant-free survival in the LT candidates and nontransplant patients. Patients were censored at death, transplant, cholecystectomy, or the end of the study period. All P < 0.05 were considered statistically significant.
RESULTS Baseline Patient Characteristics Thirty-four patients with symptomatic gallbladder disease underwent endoscopic TGS placement (Table 1). The median age of the patients was 52 years, 53% were male, and 44% were listed for liver transplantation. The most common cause of liver disease was chronic hepatitis C (35%) followed by alcoholic liver disease (26%). Both the listed and nonlisted patients were very ill with over half of them hospitalized at the time of procedure and requiring intravenous antibiotics and pain medications at the time of ERCP. The median MELD score was 15 (range, 6 to 40) with 53% CTP class C, 35% CTP class B, and 12% CTP class A. Psychosocial issues (32%) and other comorbidities (42%) were the most common contraindications for transplant listing. Nearly a third of the patients had previously well-compensated or stable liver disease before their presentation of symptomatic gallbladder disease. The most common indication for TGS placement was recurrent biliary colic (41%) followed by biliary pancreatitis (21%) and acute cholecystitis (21%). The endoscopic TGS helped facilitate the removal of cholecystostomy tubes previously inserted for acute cholecystitis in 5 patients (15%). Liver imaging aided in the diagnosis of gallbladder pathology in all of the patients with the majority of patients undergoing ultrasound (85%), abdominal CT scans (53%), HIDA
772 | www.jcge.com Copyright r 2014 Wolters Kluwer Health, Inc. All rights reserved. Copyright r 2015 Wolters Kluwer Health, Inc. All rights reserved.
J Clin Gastroenterol
Volume 49, Number 9, October 2015
TABLE 1. Baseline Characteristics of Study Population
Characteristic
Listed for Not Listed for LT (n = 15) LT (n = 19)
Age (y) 52 (21-60) Sex (male) 9 (60) Race White 10 (67) Black 2 (13) Asian 2 (13) Hispanic 1 (7) 28.8 ± 5.6 Body mass index (mean ± SD) 2 (kg/m ) Indication for ERCP Biliary colic 4 (27) Acute cholecystitis 1 (7) Cholecystostomy 2 (13) tube removal Acalculous 3 (20) cholecystitis Biliary pancreatitis 5 (33) Choledocholithiasis 0 Status Outpatient 6 (40) Inpatient 8 (53) ICU 1 (7) Etiology of cirrhosis Hepatitis C 7 (47) Alcohol 4 (27) Nonalcoholic 1 (7) steatohepatitis Primary sclerosing 1 (7) cholangitis Other liver diseases 2 (14) CTP score 9 (6-13) CTP class A 1 (7) CTP class B 8 (53) CTP class C 6 (40) Median MELD 16 (12-29) Baseline laboratories 5.6 (2.5-12.9) WBC (103/mL) 63 (25-175) Platelets ( 103/mL) INR 1.4 (1.1-2.4) Serum AST (U/L) 59 (15-141) Serum ALT (U/L) 35 (19-72) Alkaline 114 (63-467) phosphatase (U/ L) Bilirubin (mg/dL) 4.7 (1.6-33.9)
TABLE 2. Transpapillary Gallbladder Stent Procedures
Total (n = 34)
51 (28-69) 9 (47)
52 (21-69) 18 (53)
14 (74) 4 (21) 1 (5) 0 31.3 ± 12.1
24 (71) 6 (18) 3 (9) 1 (3) 30.2 ± 9.8
10 (53) 3 (16) 3 (16) 0
14 (41) 4 (12) 5 (15) 3 (9)
2 (11) 1 (5)
7 (21) 1 (3)
8 (42) 11 (58) 0
14 (41) 19 (56) 1 (3)
5 (26) 5 (26) 4 (21)
12 (35) 9 (26) 5 (15)
2 (11)
3 (9)
3 10 3 4 12 15
(16) (5-14) (16) (21) (63) (6-40)
5 10 4 12 18 15
(15) (5-14) (12) (35) (53) (6-40)
5.2 (1.6-16.7) 5.4 (1.6-16.7) 104 (11-182) 68 (11-182) 1.5 61 46 140
(0.9-2.5) 1.45 (0.9-2.5) (17-226) 60 (15-226) (7-117) 40 (7-117) (57-750) 123 (57-750)
3.1 (0.3-19.4) 3.1 (0.3-33.9)
Data presented as median (range) or n (%). ALT indicates alanine aminotransferase; AST, aspartate aminotransferase; CTP, Child-Turcotte-Pugh; ERCP, endoscopic retrograde cholangiopancreatography; ICU, intensive care unit; LT, liver transplant; MELD, model for end-stage liver disease; WBC, white blood cell count.
scans (29%), or magnetic resonance cholangiopancreatography (3%) immediately before TGS placement.
Procedural Characteristics A TGS was successfully placed in 94% of the patients with most (76%) on the first attempt (Table 2). Only 10 patients (29%) underwent a sphincterotomy and there was no significant postsphincterotomy bleeding in any of these patients. Cholangioscopy was used in 6 cases and a Copyright
r
Gallbladder Stents in Decompensated Cirrhosis
n (%) Procedure
Listed for LT (n = 15)
Successful TGS 14 (93) placement 1 attempt 11 (73) > 1 attempt 3 (20) Sphincterotomy 5 (33) Cholangioscopy 3 (20) Cholecystostomy 1 (7) wire Preoperative 9 (60) antibiotics Preoperative 2 (13) platelets Preoperative fresh8 (53) frozen plasma Sedation Conscious 8 (53) MAC 1 (7) General 6 (40) anesthesia Endobiliary stent diameter (Fr) 5 3 (20) 7 11 (73) Stent length (cm) 7 0 10 0 12 2 (13) 15 9 (60) 18 3 (20)
Not Listed for LT (n = 19)
Total (n = 34)
18 (95)
32 (94)
15 3 5 3 2
26 6 10 6 3
(79) (16) (26) (16) (11)
(76) (18) (29) (18) (9)
14 (74)
23 (68)
3 (16)
5 (15)
4 (21)
12 (35)
10 (53) 3 (16) 6 (32)
18 (53) 4 (12) 12 (35)
7 (37) 11 (58)
10 (31) 22 (69)
1 1 4 10 2
1 1 6 19 5
(5) (5) (21) (53) (11)
(3) (3) (19) (59) (16)
LT indicates liver transplant; TGS, transpapillary gallbladder stent.
cholecystostomy wire aided cystic duct access in 3 patients. Of the 6 patients that required >1 ERCP attempt, 3 were successful using cholangioscopy, 1 required a cholecystostomy wire placed by interventional radiology, 1 was successful after biliary sphincterotomy, and 1 was successful with repeat ERCP. The most common ERCP finding was cholelithiasis in 22 patients. An additional 4 patients had gallstones on ultrasound, 1 on MRI, and 1 patient with normal imaging and ERCP whose symptoms did not improve after TGS placement. The 6 remaining patients without gallstones seen on ERCP underwent TGS placement for acalculous cholecystitis or removal of cholecystostomy tube previously placed for acalculous cholecystitis. Both of the patients with unsuccessful TGS placement were hospitalized with acute cholecystitis. A tortuous, inflamed cystic duct with perforation recognized during the procedure was noted in 1 patient and later required cholecystostomy tube placement. This nontransplant candidate with morbid obesity died 90 days later from sepsis, variceal bleeding, and multiorgan failure. The other patient had an obstructing cystic duct stone that was managed with a cholecystostomy tube. He eventually underwent open cholecystectomy 54 days later and despite initial decompensation after surgery did well and is now awaiting LT at 4 months of follow-up. Prophylactic fresh-frozen plasma was administered before the procedure in 33% of the patients who had a median INR of 1.7 (range, 1.3 to 2.5) and 15% also required platelets for a median platelet count of 28,000/mL
2014 Wolters Kluwer Health, Inc. All rights reserved.
www.jcge.com |
Copyright r 2015 Wolters Kluwer Health, Inc. All rights reserved.
773
J Clin Gastroenterol
Tujios et al
TABLE 3. Outcomes and Complications After Endobiliary Stent Placement*
35
11 2 1 1 4
(73) (13) (7) (7) (27)
5 7 3 3
(33) (47)z (20) (100)
15 (79) 12 3 1 1 10
30
28 (82)
(63) (16) (5) (5) (53)
23 5 2 2 14
(68) (15) (6) (6) (44)
11 (58) 8 (42) 5 (63)
5 7 11 8
(15) (53) (32) (73)
5 11 5 2 3 1 1
(15) (32) (15) (6) (9) (3) (3)
MELD
13 (87)
Listed Patients 40
Listed for LT Not Listed for LT Total (n = 15) (n = 19) (n = 34)
Clinical symptoms resolved Complete Partial Cholecystectomy Cholecystostomyw Stent in place >1 y Transplanted Alive at last visit Died Liver-related death Stent complications Stent removal Complications Pancreatitis Ulcer Cholangitis Biliary stricture Cystic duct perforation Variceal bleed
Volume 49, Number 9, October 2015
A
n (%) Variable
25
20
15
10
5 Time of TGS
1 month
B 1 (7) 3 (20) 2 (13) 0 0 0 0 1 (7)
4 8 3 2 3 1 1
(21) (42) (16) (11) (16) (5) (5) 0
1 (3)
*Follow-up calculated until liver transplant, death, or last available follow-up through June 2011. wDoes not include patients with cholecystostomy before gallbladder stent. zDoes not include the 5 patients who are alive posttransplant. LT indicates liver transplant.
(range, 11,000/mL to 44,000/mL). The majority (68%) of patients were on antibiotics at the time of TGS placement. Half of the procedures were completed under conscious sedation with the remainder performed under propofol (12%) or general endotracheal anesthesia (35%), although sedation practice changed over time with all cases being performed with anesthesia monitoring after 2009. The length of the endobiliary stent placed was at the discretion of the endoscopist but intended to result in 1 pigtail in the gallbladder lumen (rather than the cystic duct) and 1 within the duodenum. A 15 cm 7 Fr double pigtailed stent was most commonly placed (35%) and a prophylactic pancreatic stent was not placed in any of the subjects.
Clinical Efficacy Long-term Outcomes In the 32 patients where TGS was technically successful, 88% demonstrated clinical improvement with 72% experiencing no further symptoms of gallbladder disease during a median follow-up of 9 months (mean 13 ± 14.2 mo) (Table 3). In addition, 44% have remained asymptomatic with the TGS in place for over 1 year. Follow-up laboratory MELD scores at 1 month after TGS placement in 22 evaluable patients demonstrated a reduction or stabilization compared with baseline scores in 14 of the patients [MELD 16 vs. 18, median improvement 2.5 points (0 to 7 points)] (Fig. 1).
FIGURE 1. MELD scores at the time of TGS placement and 1 month later. Eleven patients listed for liver transplantation (A) and 11 nonlisted patients with advanced liver disease (B). MELD indicates model for end-stage liver disease; TGS, transpapillary gallbladder stent.
Of the 15 patients listed for transplant, 5 (33%) underwent uncomplicated LT and 7 (47%) are still alive on the waitlist, whereas 3 (20%) died of progressive liver failure at 24, 34, and 173 days after TGS placement. The median MELD score of the patients still awaiting transplant 1 month to 1 year after TGS placement is 12. One listed patient is alive with stable liver function 5.4 years after initial TGS placement. The cirrhotic patients not listed for transplant have had their TGS in place longer than those listed (median 14.7 vs. 2.9 mo), with 10 patients having stents in place for over 1 year. In the 8 patients with MELD scores available at 1 year, 6 remained stable or improved with median MELD score of 13. One patient with continued abdominal pain underwent an open cholecystectomy 2 years after TGS placement with a complicated postoperative course but did well now 21 months after surgery. During follow-up, 8 of the nonlisted patients died with most of these deaths due to liver disease progression. The median time to death after TGS was 8.7 months and varied from 7 days to 2 years. The overall actuarial survival of these patients was 76% at 1 year and 58% at 2 years. Using competing risk analysis,
774 | www.jcge.com Copyright r 2014 Wolters Kluwer Health, Inc. All rights reserved. Copyright r 2015 Wolters Kluwer Health, Inc. All rights reserved.
J Clin Gastroenterol
Volume 49, Number 9, October 2015
FIGURE 2. Transplant-free survival in the 19 patients listed for liver transplantation and 15 nontransplant candidates. During follow-up, the transplant-free survival rate was similar (P = 0.29).
there was no difference in survival between listed and nonlisted patients (P = 0.29). The 1- and 2-year transplantfree survival rates in the listed patients were both 72%, compared with 78% and 54% in the nonlisted patients (Fig. 2).
Complications A total of 14 complications occurred in 11 patients (32%), with pancreatitis most commonly reported. The majority of these complications (11 of 14) developed within 48 hours of the TGS placement, including a single episode of variceal bleeding. One patient who had not received antibiotics had both post-ERCP pancreatitis and cholangitis. Four patients required stent exchange; 3 patients had the initial TGS coil in the cystic duct and 1 patient had the stent migrate into the duodenum. The incidence of late complications was low. However, severe bleeding from a duodenal ulcer that arose from traumatic mucosal injury from the pigtail catheter tip was noted in 1 patent 2.2 years after TGS placement. In addition, another nontransplant candidate had a mild common bile duct stricture that responded to balloon cholangioplasty and a nonbleeding duodenal ulcer that was detected at 2.1 years after initial TGS placement. Finally, a third patient developed cholangitis 2 days after replacement of a suspected occluded stent that had been in place for 74 days despite the use of prophylactic antibiotics. The increased frequency of complications in the nonlisted patients (42% vs. 20%, P = 0.18 using ordinal logistic regression) may in part relate to the longer duration of follow-up in these patients (median 509 vs. 96 d, respectively, P = 0.11, Wilcoxon rank-sum test). Patients with complications had a higher body mass index (34 ± 14 vs. 28 ± 6 kg/m2, P = 0.14) and underwent more repeat ERCPs (1.8 + 1.0 vs. 1.4 + 0.7, P = 0.31) compared with those without complications, but this was not statistically significant. However, there was no difference in stent size or length, use of antibiotics, CTP, or MELD scores among those with and without complications.
DISCUSSION Effective treatment of symptomatic gallbladder disease in patients with advanced liver disease has been problematic Copyright
r
Gallbladder Stents in Decompensated Cirrhosis
with open cholecystectomy historically carrying a mortality rate as high as 80%.17 Even with the advent of laparoscopic cholecystectomy, patients with cirrhosis experience more bleeding complications, higher conversion to open cholecystectomy, and more morbidity and mortality than patients without liver disease due to their underlying coagulopathy and portal hypertension.5 In particular, cirrhotic patients with MELD scores >13 seem to be at the greatest risk for perioperative complications and morbidity.18 Furthermore, surgery is rarely, if ever, attempted in Child’s C cirrhotics even in experienced referral centers as supported by the 1992 NIH consensus statement on laparoscopic cholecystectomy.19,20 Nonsurgical therapies for symptomatic gallstones are also limited as methyl tert-butyl ether contact dissolution therapy and mechanical extracorporeal shock wave lithotripsy have been essentially abandoned due to complications and high recurrence rates.21 Oral bile acid therapy also offers little utility in cirrhotic patients with gallbladder dyskinesia and pigment stones as ursodiol is effective only with small cholesterol stones in a functioning gallbladder.21 Biliary sphincterotomy is an effective intervention for recurrent gallstone pancreatitis in poor surgical candidates but will not prevent future episodes of cystic duct obstruction and is unsafe in coagulopathic patients with liver disease.8 Percutaneous cholecystostomy can also be used to treat acute cholecystitis in poor surgical candidates but short-term mortality remains high.22 Kalloo and colleagues first described endoscopic TGS placement in 4 cirrhotic patients in 1994 as a potential alternative palliative treatment. Since then, there have been multiple small case series describing the feasibility and clinical efficacy of endoscopic TGS for poor operative candidates.9–15 However, the technique is not commonly performed even in LT centers and there remains little data on the long-term safety and outcomes with TGS in decompensated cirrhotic patients or LT candidates. Our current study of 34 patients is the largest singlecenter study of endoscopic TGS in cirrhosis that we are aware of. Using an intention-to-treat design, the technical success rate was 94% with a clinical success rate of 82%, confirming that this approach is an effective procedure for carefully selected patients. Our high rate of technical success is similar to that reported by others.15 Over time, we have found that use of guidewires, cholangioscopy, and rendezvous with percutaneous cholecystostomy wires can facilitate endobiliary stenting in these patients. In the 2 cases where TGS was not possible, the cystic duct was either obstructed by a stone or significantly inflamed, highlighting the technical difficulties in attempting to safely cannulate the cystic duct in patients with acute cholecystitis. The majority of patients undergoing endoscopic TGS placement not only had improvement in their clinical status and symptoms but also experienced a stabilization or improvement in their MELD score at 1-month follow-up (Fig. 1). In addition, many of our patients have experienced further improvement in their liver disease status with more prolonged follow-up, including 6 of the waitlisted patients. Most of these patients have not required a stent exchange. Endoscopically placed gallbladder stents act to prevent stone impaction of the cystic duct as well as provide a wick for biliary drainage. As stent patency is not required, smalldiameter stents are preferentially used and routine stent changes do not seem to be necessary. In support of this, a recent prospective study of 16 patients undergoing TGS
2014 Wolters Kluwer Health, Inc. All rights reserved.
www.jcge.com |
Copyright r 2015 Wolters Kluwer Health, Inc. All rights reserved.
775
Tujios et al
noted that stents may be safely left in place for up to 2 years in mostly noncirrhotic patients.23 Although late complications including stent migration, duodenal ulceration, and biliary stricture were observed in our study, selecting appropriate stent length and smaller diameter might avert these. On the basis of our experience and that of other groups, we would suggest careful and frequent laboratory and clinical monitoring of decompensated cirrhotics undergoing TGS, but not recommend routine endobiliary stent change at this time. Periprocedural complications occurred in a third of patients but were generally mild and resolved within a few days with conservative management. Five patients met consensus definition of post-ERCP pancreatitis with increased abdominal pain, elevated amylase, and hospitalization but only 1 case was severe. Identifiable risk factors for post-ERCP pancreatitis included age in one young woman and normal bilirubin in another.24 The relatively high rate of post-ERCP pancreatitis of 15% is likely due to small sample size, with 95% confidence intervals estimated 16.7% (95% CI, 6.0%-30.6%) using the adjusted Wald method. Longer procedure times with multiple cannulations due to the technical challenge of negotiating wire into cystic duct and gallbladder may also contribute to increased incidence of post-ERCP pancreatitis, but we would expect this to improve with endoscopist experience. Rectal indomethacin has been shown to reduce the rate of post-ERCP pancreatitis in high-risk patients; however, indomethacin use should be considered carefully in this patient population given their increased risk of bleeding and their increased susceptibility to renal failure.25 Although most of our patients were receiving antibiotics before and after the procedure, the 1 patient with severe pancreatitis, cholangitis, and sepsis did not receive prophylactic antibiotics. Although there are no clear data, based upon our experience and other series of ERCP in immunosuppressed patients with biliary obstruction and incomplete drainage, we would recommend considering a prophylactic antibiotic regimen such as broad-spectrum cephalosporin or fluoroquinolone to be given to cirrhotic patients undergoing TGS in light of their known increase for severe bacterial infections.26,27 Although 10 of our patients underwent sphincterotomy, we did not encounter significant postspincterotomy bleeding. The low rate of periprocedural bleeding may, in part, be due to the limited size sphincterotomy that we made as well as the selective use of periprocedural plasma and platelet transfusions in 3 patients. We would encourage others to consider using prophylactic plasma to achieve an INR of 50,000/mL if a sphincterotomy is planned and to observe these high-risk patients in house after blood product administration. The role of endoscopic TGS placement in LT candidates requires further study. In some centers, physicians have advocated to obtain MELD exception points for their decompensated cirrhotic patients with cholecystitis similar to patients with recurrent bacterial cholangitis.28,29 However, a review of MELD exception practices demonstrates substantial heterogeneity in the number of points requested and success rates with nonstandard exemptions across centers.29 In our series, 5 patients have already been successfully bridged to liver transplantation and 7 others remain stable on the LT waiting list with a median followup of 63 days. The 3 LT candidates who died had been placed on hold for cardiac and active infectious issues
J Clin Gastroenterol
Volume 49, Number 9, October 2015
unrelated to their gallbladder disease, with eventual progression to hepatorenal syndrome and liver failure. In light of the high rate of short-term clinical improvement we observed in both our listed and nonlisted decompensated cirrhotics, we would encourage others to consider this treatment option for cirrhotic patients who are poor surgical candidates with symptomatic gallbladder disease. However, several potential limitations with our current retrospective series should be noted. First, the number of patients studied was limited and complete follow-up laboratory data were not available in all patients. In addition, functional studies of gallbladder function, such as quantitative HIDA scans, were not routinely obtained before or after TGS to document improved gallbladder function. However, HIDA scans are known to have limited utility in cirrhotic patients and those with elevated bilirubin. Furthermore, over 88% reported improvement in their clinical symptoms during follow-up. In addition, the optimal endobiliary stent to use remains unknown as well as the utility of prophylactic pancreatic duct stenting, rectal indomethacin, and/or bile acid therapy to reduce bile lithogenicity. Finally, the safety and efficacy of endoscopic TGS placement has not been directly compared with percutaneous cholecystostomy tube placement.22 Endoscopic ultrasound-guided gallbladder drainage is another therapeutic option but in the absence of acute cholecystitis, the gallbladder wall may not be tethered to the antrum or duodenal bulb resulting in a reasonably high risk of bile leak or pneumoperitoneum. For these reasons, we believe that TGS has a more favorable risk-benefit ratio than transmural drainage, especially in cirrhotics who are harder to rescue from major complications. Additional research may clarify whether novel “tissue approximating stents” will change our approach to endoscopic therapy in this patient population.30 In light of the relatively low rate of periprocedural complications in our series and others, we would advocate consideration of an endoscopic TGS placement first in decompensated cirrhotics, as they are at increased risk for bleeding, infectious, and fluid leak complications with a percutaneous cholecystostomy tube. In conclusion, our data demonstrates that endoscopic TGS placement is a safe and effective treatment for decompensated cirrhotics with symptomatic gallbladder disease. Although nearly 90% of our patients were CTP class B or C, our rate of periprocedural complications was low and did not lead to worsening liver disease status in most of the treated patients. Our study and other series also suggest that TGS may prove to be an effective long-term management approach to symptomatic gallbladder disease in decompensated cirrhosis patients or perhaps stabilize them for future cholecystectomy. Finally, TGS placement seems to be a safe and effective bridge to liver transplantation.
REFERENCES 1. Thistle JL, Cleary PA, Lachin AM. The natural history of cholelithiasis: The National Cooperative Gallstone Study. Ann Intern Med. 1984;107:171–175. 2. Elzouki AN, Nilsson S, Nilsson P, et al. The prevalence of gallstones in chronic liver disease is related to degree of liver dysfunction. Hepatogastroenterology. 1999;46:2946–2950. 3. Conte D, Fraquelli M, Fornari F, et al. Close relation between cirrhosis and gallstones: cross-sectional and longitudinal survey. Arch Intern Med. 1999;159:49–52.
776 | www.jcge.com Copyright r 2014 Wolters Kluwer Health, Inc. All rights reserved. Copyright r 2015 Wolters Kluwer Health, Inc. All rights reserved.
J Clin Gastroenterol
Volume 49, Number 9, October 2015
4. Bloch R, Allaben R, Walt A. Cholecystectomy in patients with cirrhosis. Arch Surg. 1985;120:669–971. 5. Puggioni A, Wong LL. A metaanalysis of laparoscopic cholecystectomy in patients with cirrhosis. J Am Coll Surg. 2003;197:921–926. 6. Kortram K, de Vries Reilingh TS, Wiezer MJ, et al. Percutaneous drainage for acute calculous cholecystitis. Surg Endosc. 2011;25:3642–3646. 7. Winbladh A, Gullstrand P, Svanvik J, et al. Systematic review of cholecystostomy as a treatment option in acute cholecystitis. HPB. 2009;11:183–193. 8. Freeman ML, Nelson DB, Sherman S, et al. Complications of endoscopic biliary sphincterotomy. N Engl J Med. 1996; 335:909–918. 9. Kalloo AN, Thuluvath PJ, Pasricha PJ. Treatment of high-risk patients with symptomatic cholelithiasis by endoscopic gallbladder stenting. Gastrointest Endosc. 1994;40:608–610. 10. Gaglio PJ, Buniak B, Leevy CB. Primary endoscopic retrograde cholecystoendoprosthesis: a nonsurgical modality for symptomatic cholelithiasis in cirrhotic patients. Gastrointest Endosc. 1996;44:339–342. 11. Shrestha R, Trouillot TE, Everson GT. Endoscopic stenting of the gallbladder for symptomatic gallbladder disease in patients with end-stage liver disease awaiting orthotopic liver transplantation. Liver Transpl Surg. 1999;5:275–281. 12. Conway JD, Russo MW, Shrestha R. Endoscopic stent insertion into the gallbladder for symptomatic gallbladder disease in patients with end-stage liver disease. Gastrointest Endosc. 2005;61:32–36. 13. Schlenker C, Trotter JF, Shah RJ, et al. Endoscopic gallbladder stent placement for treatment of symptomatic cholelithiasis in patients with end-stage liver disease. Am J Gastroenterol. 2006;101:278–283. 14. Pannala R, Peterson BT, Gostout CJ, et al. Endoscopic transpapillary gallbladder drainage: 10-year single center experience. Minerva Gastroenterol Dietol. 2008;54:107–113. 15. Itoi T, Coelho-Prabhu N, Baron TH. Endoscopic gallbladder drainage for management of acute cholecystitis. Gastrointest Endosc. 2010;71:1038–1045. 16. Elmunzer BJ, Novelli PM, Taylor JR, et al. Percutaneous cholecystostomy as a bridge to definitive endoscopic gallbladder stent placement. Clin Gastroenterol Hepatol. 2011;9:18–20.
Copyright
r
Gallbladder Stents in Decompensated Cirrhosis
17. Aranha GV, Sontag SJ, Greenlee HB. Cholecystectomy in cirrhotic patients: a formidable operation. Am J Surg. 1982; 143:55–60. 18. Delis S, Bakoyiannis A, Madariaga J, et al. Laparoscopic cholecystectomy in cirrhotic patients: the value of MELD score and Child-Pugh classification in predicting outcome. Surg Endosc. 2010;24:407–412. 19. Nguyen KT, Kitisin K, Steel J, et al. Cirrhosis is not a contraindication to laparoscopic cholecystectomy: results and practical recommendations. HPB. 2011;13:192–197. 20. NIH consensus conference statement on gallstones and laparoscopic cholecystectomy Am J Surg. 1993;165:390–398. 21. Portincasa P, Moschetta A, Palasciano G. Cholesterol gallstone disease. Lancet. 2006;368:230–239. 22. Hatzidakis AA, Prassopoulos P, Pertinarakis I, et al. Acute cholecystitis in high-risk patients: percutaneous cholecystomy vs. conservative treatment. Eur Radiol. 2002;12: 1778–1784. 23. Lee TH, Park DH, Lee SS, et al. Outcomes of endoscopic transpapillary gallbladder stenting for symptomatic gallbladder diseases: a multicenter prospective follow-up study. Endoscopy. 2011;43:702–708. 24. Anderson MA, Fisher L, Jain R, et al. Complications of ERCP. Gastrointest Endosc. 2012;75:467–473. 25. Elmunzer BJ, Scheiman JM, Lehman GA, et al. A randomized trial of rectal indomethacin to prevent post-ERCP pancreatitis. N Engl J Med. 2012;366:1414–1422. 26. Banerjee S, Shen B, Baron TH, et al. Antibiotic prophylaxis for GI endoscopy. Gastrointest Endosc. 2008;67:791–798. 27. Jalan R, Fernandez J, Wiest R, et al. Bacterial infections in cirrhosis. A position statement based on the EASL special conference 2013. J Hepatol. 2014;60:1310–1324. 28. Goldberg D, Bittermann T, Makar G. Lack of standardization in exception points for patients with primary sclerosing cholangitis and bacterial cholangitis. Am J Transplant. 2012;12:163–1609. 29. Goldberg DS, Makar G, Bittermann T, et al. Center variation in the use of non-standardized model for end-stage liver disease exception points. Liver Transpl. 2013;19:1330–1342. 30. Choi JH, Park DH, Lee SS, et al. Can endoscopic ultrasound help to drain the gallbladder? Gastrointest Interv. 2013;2: 30–35.
2014 Wolters Kluwer Health, Inc. All rights reserved.
www.jcge.com |
Copyright r 2015 Wolters Kluwer Health, Inc. All rights reserved.
777
