Similar Efficacies of Endoscopic Ultrasound Gallbladder Drainage With a Lumen-Apposing Metal Stent Versus Percutaneous Transhepatic Gallbladder Drainage for Acute Cholecystitis

Accepted Manuscript Similar Efficacies of Endoscopic Ultrasound Gallbladder Drainage with a LumenApposing Metal Stent vs Percutaneous Transhepatic Gallbladder Drainage for Acute Cholecystitis Shayan Irani, Saowanee Ngamruengphong, Anthony Teoh, Uwe Will, Jose Nieto, Barham K. Abu Dayyeh, S. Ian Gan, Michael Larsen, Hon Chi Yip, Mark D. Topazian, Michael J. Levy, Christopher C. Thompson, Andrew C. Storm, Gulara Hajiyeva, Amr Ismail, Yen-I. Chen, Majidah Bukhari, Yamile Haito Chavez, Vivek Kumbhari, Mouen A. Khashab PII: DOI: Reference:

S1542-3565(16)31246-0 10.1016/j.cgh.2016.12.021 YJCGH 55046

To appear in: Clinical Gastroenterology and Hepatology Accepted Date: 12 December 2016 Please cite this article as: Irani S, Ngamruengphong S, Teoh A, Will U, Nieto J, Abu Dayyeh BK, Gan SI, Larsen M, Yip HC, Topazian MD, Levy MJ, Thompson CC, Storm AC, Hajiyeva G, Ismail A, Chen Y-I, Bukhari M, Chavez YH, Kumbhari V, Khashab MA, Similar Efficacies of Endoscopic Ultrasound Gallbladder Drainage with a Lumen-Apposing Metal Stent vs Percutaneous Transhepatic Gallbladder Drainage for Acute Cholecystitis, Clinical Gastroenterology and Hepatology (2017), doi: 10.1016/ j.cgh.2016.12.021. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

Title: Similar Efficacies of Endoscopic Ultrasound Gallbladder Drainage with a LumenApposing Metal Stent vs Percutaneous Transhepatic Gallbladder Drainage for Acute Cholecystitis Shayan Irani1, Saowanee Ngamruengphong2, Anthony TEOH3, Uwe Will4, Jose Nieto5,

RI PT

Barham K. Abu Dayyeh6, S. Ian Gan1, Michael Larsen1, Hon Chi Yip3, Mark D. Topazian6, Michael J. Levy6, Christopher C. Thompson7, Andrew C. Storm7, Gulara Hajiyeva2, Amr

Ismail2, Yen-I Chen2, Majidah Bukhari2, Yamile Haito Chavez2, Vivek Kumbhari2, Mouen

M AN U

SC

A. Khashab2

1) Division of Gastroenterology and Hepatology, Virginia Mason Medical Center, Seattle, USA

2) Division of Gastroenterology and Hepatology, Johns Hopkins University, Baltimore,

TE D

USA

3) Department of Surgery, Prince of Wales Hospital, The Chinese University of Hong Kong

EP

4) Department of Gastroenterology, Municipal Hospital, Gera, Germany. 5) Advanced Therapeutic Endoscopy Center. Borland Groover Clinic. Jacksonville, Florida

AC C

6) Division of Gastroenterology and Hepatology, Mayo Clinic, Rochester, Minnesota, USA.

7) Division of Gastroenterology and Hepatology, Brigham and Women’s Hospital Abbreviations EUS-GBD: EUS-guided gallbladder drainage; LAMS: Lumen-apposing metal stent; PT-GBD percutaneous transhepatic gallbladder drainage

ACCEPTED MANUSCRIPT

Correspondence Shayan Irani, Division of Gastroenterology and Hepatology, Virginia Mason Medical

RI PT

Center, 1100 Ninth Avenue, C3-GAS, Seattle, WA 98101, [email protected], (P) 206-223-2319, (F) 206-625-7195

SC

Conflicts of interest

Shayan Irani is a consultant for Boston Scientific remittance to clinic.

M AN U

Mouen Khashab is a consultant for Boston Scientific. Barham K Abu Dayyeh consultant for Boston Scientific.

Jose Nieto is a consultant for Boston Scientific and Olympus

Christopher Thompson is a consultant for Boston Scientific, Olympus and Apollo

TE D

Endosurgery

Manuscript inception: Shayan Irani1, Mouen A. Khashab2

EP

Drafting / Critical Revision: Shayan Irani1, Mouen A. Khashab2, Saowanee Ngamruengphong2.

AC C

Data acquisition: Shayan Irani1, Saowanee Ngamruengphong2, Anthony TEOH3, Uwe Will4, Jose Nieto5, Barham K Abu Dayyeh6, Seng-Ian Gan1, Michael Larsen1, Hon Chi Yip3 Mark D. Topazian6, Michael J. Levy6, Christopher C Thompson7, Andrew C Storm7, Gulara Hajiyeva2, Amr Ismail2, Yen-I Chen2, Majidah Bukhari2, Yamile Haito Chavez2, Vivek Kumbhari2 Critical Revision / final approval: all authors

ACCEPTED MANUSCRIPT

RI PT

BACKGROUND & AIMS: Acute cholecystitis in patients who are not candidates for surgery is often managed with percutaneous transhepatic gallbladder drainage (PT-GBD). Endoscopic ultrasound-guided gallbladder drainage (EUS-GBD) with a lumen-apposing metal stent (LAMS) is an effective alternative to PT-GBD. We compared the technical success of EUS-GBD vs PT-GBD, as well as patient outcomes, numbers of adverse events (AEs), length of hospital stay, pain scores, and repeat interventions.

M AN U

SC

METHODS: We performed a retrospective study to compare EUS-GBD vs PT-GBD at 7 centers (5 in the United States, 1 in Europe, and 1 in Asia), from 2013 through 2015, in management of acute cholecystitis in patients who are not candidates for surgery. A total of 90 patients (56 men) with acute cholecystitis (61 calculous, 29 acalculous) underwent EUS-GBD (n=45) or PT-GBD (n=45). Data were collected on technical success, clinical success (resolution of symptoms or laboratory and/or radiologic abnormalities within 3 days of intervention), and need for repeat intervention. Characteristics were compared using t tests for continuous variables and the χ2 test, or the Fisher exact test, when appropriate, for categorical variables. Adverse events were graded according to American Society for Gastrointestinal Endoscopy definitions and compared using the Fisher exact test. Post-procedure pain scores were compared using the Mann-Whitney U test.

AC C

EP

TE D

RESULTS: Baseline characteristics, type, and clinical severity of cholecystitis were comparable between groups. In the EUS-GBD group, noncautery LAMS were used in 30 patients and cautery-enhanced LAMS were used in 15. Technical success was achieved for 98% of patients in the EUS-GBD and 100% of the patients in the PT-GBD group (P=.88). Clinical success was achieved by 96% of patients in the EUS-GBD group and 91% in the PT-GBD group (P=.20). There was a nonsignificant trend toward fewer AEs in the EUS-GBD group (5 patients, 11%) than in the PT-GBD group (14 patients, 32%)(P=.065). There were no significant differences in the severity of the AEs: mild, 2 in the EUS-GBD group vs 5 in the PT-GBD group (P=.27); moderate, 4 vs 3 (P=.98); severe, 1 vs 3 (P =.62); or deaths, 1 vs 3 (P=.61). The mean post-procedure pain score was lower in the EUS-GBD group than in the PT-GBD group (2.5 vs 6.5; P<.05). The EUS-GBD group had a shorter average length of stay in the hospital (3 days) than the PTGBD group (9 days) (P<.05) and fewer repeat interventions (11 vs 112)(P<.05). The average number of repeat interventions per patients was 0.2 ± 0.4 EUS-GBD group vs 2.5±2.8, in the PT-GBD group (P<.05). Median follow-up after drainage was comparable in EUS-GBD group (215 days; range 1–621 days) vs the PT-GBD group (265 days; range, 1–1638 days). Conclusion: EUS-GBD has similar technical and clinical success compared to PT-GBD and should be considered an alternative for patients who are not candidates for surgery. Patients who undergo EUS-GBD seem to have shorter hospital stays, lower pain scores, and fewer repeated interventions, with a trend toward fewer AEs. A prospective, comparative study is needed to confirm these results. KEY WORDS: nonsurgical candidate, transmural gallbladder drainage, liver

ACCEPTED MANUSCRIPT

Although cholecystectomy is the treatment of choice for acute cholecystitis, conservative therapy as a temporizing measure is preferred in severely ill patients and

RI PT

as a definitive measure in poor surgical candidates.1,2 Non-surgical gallbladder drainage options include percutaneous and endoscopic techniques. Percutaneous transhepatic gallbladder drainage (PT-GBD), performed since

SC

the 1980s, is the most established technique.3,4 Endoscopic methods include

transpapillary gallbladder stenting, first reported by Kozarek in 1984,1 or nasogallbladder

Bookmark not defined.

M AN U

catheter placement via Endoscopic Retrograde Cholangiopancreatogram (ERCP).Error! A systematic review of the available case series (7 studies, 127 patients)

found a success rate of over 95% and adverse event rate of 6%.6,7 The clinical response to PT-GBD varies from 56% to 100%.8,9,10,11 with adverse events reported in at least 10% of

TE D

patients, including bile leaks and peritonitis, bleeding, and pneumothorax.12,13 In addition, PT-GBD may be inappropriate for patients with ascites or coagulopathy.14 Finally,

EP

catheter dislodgement could cause peritonitis and require additional procedures.2,14 EUS-guided transmural gallbladder drainage (EUS-GBD) is an alternative

AC C

endoscopic drainage technique. A recent review of 155 patients with acute cholecystitis (8 series and 12 case reports) treated with EUS-GBD noted a technical and clinical success of 97% and 99%, respectively.15 Adverse events were reported in less than 8% of patients, and were managed conservatively. EUS-GBD has been performed using plastic stents, nasobiliary drainage catheters, standard self-expandable metal stents (SEMS), defined. Error! Bookmark not defined. and most recently, lumen-apposing metal stents (LAMS).Error! Bookmark not defined.,Error!

ACCEPTED MANUSCRIPT

Comparison studies between the different stent types are lacking. The theoretical advantage of LAMS over other stents is the ability to approximate the gallbladder wall

RI PT

to the intestinal lumen, thus “sealing off” potential bile leaks once placed. The first multicenter US case series using LAMS to treat cholecystitis noted a clinic success in all 15 patients and 1 mild adverse event.16 More recently, a comparative study in non-

SC

surgical candidates with acute cholecystitis, between 43 patients undergoing PT-GBD and 30 patients undergoing endoscopic gallbladder drainage (24 transpapillary, 6

M AN U

transmural), was published.17 Clinical success was comparable in the 2 groups, however, adverse events, post procedure pain scores, length of stay, and need for repeat interventions were fewer in the endoscopic group. Our multicenter study retrospectively compared outcomes of EUS-GBD vs PT-GBD in 90 non-surgical patients

Methods

TE D

with acute cholecystitis.

EP

Patients and Outcome Variables

From July 2013 to December 2015, a retrospective chart review was performed

AC C

of patients diagnosed with acute calculous or acalculous cholecystitis from 7 tertiary care referral centers. Any patient above the age of 18 years who was deemed a nonsurgical candidate and thus requiring a permanent non-surgical treatment (PT-GBD or EUS-GBD) was included in the analysis. Patient demographics including age, sex, and underlying comorbidity precluding surgery were recorded. Type of cholecystitis (calculous vs acalculous) was determined radiologically (CT or US). Procedural details

ACCEPTED MANUSCRIPT

including type and diameter of stent or drainage catheter, technique of placement, number of interventions, and technical success were recorded. Post-procedure details

RI PT

including length of stay after the procedure, pain scores, adverse events, and need for repeat interventions were recorded. Patients were followed post-procedure with

telephone calls or clinic visits and/or imaging studies. This study was approved by the

M AN U

SC

Institutional Review Board at our institutions.

Definitions

TE D

Acute cholecystitis was diagnosed and graded in severity by the Tokyo guidelines as a combination of clinical symptoms (fever, right upper quadrant pain, positive Murphy’s sign, and elevated leukocyte count or C-reactive protein level) and radiological

EP

findings consistent with acute cholecystitis.Error! Bookmark not defined. Grade 1 cholecystitis was therefore defined as mild inflammatory changes with no organ dysfunction. Grade 2

AC C

cholecystitis was moderate inflammatory changes with increased operative difficulty. Grade 3 cholecystitis was associated with one or more organ dysfunction making the patient critically ill.1

Technical success was defined as the ability to place a percutaneous drainage catheter (PT-GBD) or a transmural LAMS (EUS-GBD) into the gallbladder as determined by flow of bile or pus and confirmed radiographically. Clinical success was defined as

ACCEPTED MANUSCRIPT

resolution of symptoms, laboratory, and or radiological abnormalities within 3 days of intervention. Any procedure required to replace, check, or remove the previously placed

RI PT

drain or stent was considered a reintervention. Adverse events were defined as any procedure, drain or stent-related event occurring during or within 30 days after the procedure as previously defined.18 Coagulopathy was defined by an INR ≥ 2, and

SC

thrombocytopenia by a platelet count ≤ 50,000/dL. The postprocedure pain score was

Technique and Follow-up:

M AN U

the highest pain score noted the day after the intervention.

TE D

PT-GBD was performed by interventional radiologists from the various centers mostly using mild intravenous sedation (89%). Using sterile techniques, an 8Fr (predominantly) or 10Fr self-locking pigtail catheter (Boston Scientific, Natick, MA) was

EP

placed ideally via a transhepatic route. A tube check was performed in majority of patients (82%) at 4–8 weeks to assess removability of the catheter. Most catheters were

AC C

left open to drainage unless patency of the cystic duct was established at a subsequent tube check.

EUS-GBD was predominantly performed under general anesthesia (89%).

Procedures were performed by experienced therapeutic endoscopists with more than 5 years of experience in the endoscopy suite or who are rarely in the operating room. A LAMS with an inner diameter of 10mm or 15mm, and outer flange diameters of 21mm

ACCEPTED MANUSCRIPT

or 24mm and a saddle length of 10mm was used in all patients (Axios stent, Boston Scientific, Natick, MA). EUS-GBD was performed using an oblique-viewing (GF-UCT-180;

RI PT

Olympus Optical, Tokyo, Japan) or forward viewing therapeutic linear array echoendoscope (TGF-UC 180J; Olympus Optical, Tokyo, Japan). The puncture site was either the duodenal bulb or the prepyloric antrum into the body or neck of the

SC

gallbladder using a 19-gauge needle (EUSN-19-T; Cook Endoscopy, Winston-Salem, NC, and 19G Expect needle Boston Scientific, Natick MA). The distance between the

M AN U

gallbladder wall and the intestinal wall was measured and confirmed to be ≤10mm, ie, the saddle length of the LAMS. Bile was aspirated, after which a 0.035-inch Jagwire (Boston Scientific, Natick, Ma) or 0.025-inch Visiglide wire (Olympus Corporation, Center Valley, PA) was coiled into the gallbladder. For the noncautery enhanced (cold) LAMS,

TE D

dilation of the tract was performed using a 4mm balloon dilation (Fusion Titan Biliary Dilation Balloon; Cook Endoscopy, Winston-Salem, NC, or Hurricane Balloon, Boston Scientific, Natick, MA), or an over the wire needle knife or 10Fr cystotome (Cook

EP

Endoscopy, Winston-Salem, NC). Cautery-enhanced (hot) LAMS were placed with or without initial guidewire access. (Video 1) In over half the patients, a plastic pigtail stent

AC C

through the LAMS. Our intent was to leave the stents in place indefinitely. No planned repeat endoscopy was scheduled.

Statistical Analysis

Characteristics of the study group were compared using t tests for continuous variables and the chi-square test (or the Fisher exact test, when appropriate) for categoric variables. Adverse events were compared using the Fisher exact test, and

ACCEPTED MANUSCRIPT

post-procedure pain scores were compared using the Mann–Whitney U test. P values of less than .05 were considered statistically significant. Analyses were performed using

RI PT

SPSS 18.0 (SPSS, Inc, Chicago, IL), except for 95% CIs, which were obtained by asymptotic normal distribution.

SC

Results

Between 2013 and 2015, 90 patients (56 M, 34F) were included in the study

M AN U

from 7 centers (5 US, 1 Europe, 1 Asia). There was a nonstatistical trend toward younger patients (65 years vs 75 years, P = 0.06) and more patients with acalculous cholecystitis in the EUS-GBD group (18 vs 11, P = 0.06). Underlying comorbidities precluding cholecystectomy, median duration of cholecystitis prior to intervention, and severity of

TE D

cholecystitis was comparable in both groups. All patients who underwent EUS-GBD, either refused PT-GBD or preferred internal drainage. In addition, 5 patients had significant ascites, 4 patients were hypercoagulable or thrombocytopenic and in 4

EP

patients, there was concern for drain dislodgement (3 advanced dementia, 1 patient legally blind) (Table 1). ERCP transpapillary gallbladder stent placement was attempted

AC C

but failed in 8 patients requiring an ERCP for ductal obstruction from stones, biliary stents, or cancer, who subsequently underwent EUS-GBD. Median gallbladder wall thickness was 4mm in both groups. The sites of

obstruction were comparable in the 2 groups. General anesthesia was used in 40/45 (89 %) patients undergoing EUS-GBD versus only 5/45 (11%) of patients undergoing PT-GBD (P < 0.001). Median procedure times were longer with EUS-GBD compared to PT-GBD

ACCEPTED MANUSCRIPT

(28 min vs 22 min, P < 0.05) (Table 2). In EUS-GBD group, non-cautery LAMS were used in 30 patients and cautery-enhanced LAMS were used in 15. Thirty-seven 10mm and

RI PT

eight 15mm diameter LAMS were used. No difference was noted in the adverse events between the 2 stent sizes or the non-cautery versus cautery-enhanced LAMS [5/30

(17%) vs. 3/15 (20%) p=0.78 ]. The 15mm LAMS was selected in some cases when there

SC

was a concern for larger stones potentially occluding the stent. Pigtail stents were

placed through the LAMS in 24/45 (53%) given possibility of injury of the stent to the

M AN U

contralateral gallbladder wall or risk of being buried (recurrent cholecystitis), as was seen in 1 patient. One technical failure was due to misdeployment of the gallbladder flange and had to be salvaged with a longer 10mm x 60mm fully covered biliary metal stent (Table 3).

TE D

Technical success of EUS-GBD and PT-GBD was high in both groups (98% vs 100%, respectively; P = 0.98). Clinical success was similar in the EUS-GBD and PT-GBD groups (96% vs 91%; P = 0.12). Median post-procedure pain score were lower in the

EP

EUS-GBD group vs PT-GBD group (2.5 vs 6.5; P < 0.01). Median post-intervention hospital length of stay was shorter (3 days vs 9 days; P < 0.05). Reinterventions were

AC C

significantly fewer [(total 11 vs 112; P < 0.01, reinterventions per patient 0.2 ± 0.4 vs 2.5 ± 2.8 P < 0.005)] in the EUS-GBD group vs PT-GBD group (Table 4). There was a nonsignificant trend toward fewer AEs in the EUS-GBD group vs PT-

GBD group, [8 (18%) vs 14(31%); P = 0.07]. There were no differences in the severity of the AEs [mild (2 vs 5; P = 0.27), moderate (4 vs 3; P = 0.98), severe (1 vs 3; P = 0.62), and deaths (1 vs 3; P = 0.61)]. Two episodes of bleeding were encountered 3 days and 6

ACCEPTED MANUSCRIPT

months after placement of 10 mm LAMS. Both patients were anticoagulated (1 with heparin and the other supratherapeutic on warfarin). The first patient was treated by

RI PT

evacuating the clot with a snare and Roth Net® and placing a pigtail stent through the LAMS (bleeding site never seen), while the second stopped spontaneously with reversal of the INR without need for endoscopy (Figure 1). Three patients in the EUS-GBD group

SC

experienced recurrent cholecystitis at 6, 8, and 12 months. One patient was treated

with antibiotics alone, while 2 patients were treated endoscopically, including 1 patient

M AN U

with granulation overgrowth on the gastric side causing stent occlusion. He was treated with an additional pigtail stent through the LAMS. A bile leak noted 3 days after EUSGBD resulting in peritonitis was treated successfully with a percutaneous drain. Abdominal pain without fever (food occluding a transgastric LAMS n = 1, granulation

TE D

overgrowth on gastric side n = 1) was treated with evacuation of the food and a pigtail stent through the LAMS and balloon dilation of the granulation overgrowth and pigtail stent through the LAMS. One death was due to ongoing sepsis in a patient (type 3

EP

cholecystitis) who also refused a PT-GBD despite findings of a ruptured gallbladder noted at EUS-GBD. Contrast was injected into the gallbladder to look for this possibility

AC C

given his clinical deterioration an hour prior to the procedure, with development of septic shock on 3 pressure support medications. Recurrent cholecystitis was seen in 4 patients with PT-GBD at 8 days (due to

drain dislodgement), 2, 4, and 6 months (drain occlusion n = 1, removal n = 2), treated with antibiotics and tube change or placement. Pain without cholecystitis necessitating admission from tube occlusion was seen in 3 patients. One patient developed cellulitis

ACCEPTED MANUSCRIPT

around the PT-GBD treated with 7 days of oral antibiotics. Bile leak with additional perihepatic collections needing drainage was seen in 3 patients, of which 1 patient died

RI PT

from ongoing sepsis. Two additional patients (type 3 cholecystitis) died of ongoing sepsis. Injury to an intervening loop of jejunum at PT-GBD was successfully treated by allowing the fistula to mature over 2 weeks and then internalizing the gallbladder

SC

drainage with EUS-GBD (Figure 2). Twenty-three patients in the EUS-GBD group and 20 patients in the PT-GBD group died from underlying diseases unrelated to the stent or

M AN U

procedure. Median follow-up after EUS-GBD was 215 days (1–621 days) and 265 days (1–1638 days) in the in the PT-GBD group (P = 0.25).

Discussion

Historically, there were 2 non-surgical options to treat acute cholecystitis in

TE D

patients unfit for surgery. The more frequently performed, almost universally available, defined. Error! Bookmark not defined. percutaneous transhepatic gallbladder drainage (PT-GBD)Error! Bookmark not defined.,Error!

versus the less frequently performed, less well known endoscopic option of ERCP

EP

Error! Bookmark not defined. transpapillary gallbladder stenting.1–Error! As EUS-guided therapies became

AC C

more common, transmural drainage of the gallbladder with pigtail or fully covered biliary metal stents emerged.Error! Bookmark not defined. Most recently, with the availability of LAMS, these stents have been used to provide drainage with the theoretical benefit of better lumen apposition and lower risk of bile leak. However, only a few case reports and defined. Error! Error! series have described EUS-GBD for the treatment of acute cholecystitis.Error! Bookmark not defined.,Err Bookmark not defined.,19 defined.

Additionally, to date there are no comparison studies between EUS-GBD

ACCEPTED MANUSCRIPT

using a LAMS and PT-GBD. In a recent, retrospective comparison study, Kedia et al17, demonstrated endoscopic drainage of the gallbladder (24 transpapillary, 6 transmural)

RI PT

had similar technical and clinical success compared to PT-GBD, with need for fewer hospital resources, reinterventions, adverse events, and lower pain scores. They

suggested that endoscopic gallbladder drainage may provide a less-invasive, safer, cost-

SC

effective option than PT-GBD with improved clinical outcomes. In another, recent

retrospective study comparing ERCP transpapillary gallbladder stenting (n=35) to PT-

M AN U

GBD (n=29), the authors found statistically significant a lower rate of recurrent cholecystitis 0% vs. 17.2%.20 Neither of these studies however, compared the transmural approach (EUS-GBD) with PT-GBD, which is what we set out to examine. Although our study was retrospective, the 2 groups were similar in their baseline

TE D

clinical characteristics, etiologies, and severity of cholecystitis. There was a slight trend toward more acalculous cases in the EUS-GBD group. Technical and clinical success rates were high and comparable in both groups as expected, and previously demonstrated in

Bookmark not defined.

EP

Error! defined. Error! Bookmark not defined.,Error! defined. Error! Bookmark not defined. prior case seriesError! Bookmark not defined.,Error! and comparative studies.2,Error!

AC C

PT-GBD has the inconvenience and almost universal discomfort of an external

drainage catheter, at least initially. Two prior studies including a randomized controlled trial by Jang et al2 and the retrospective comparison study by Kedia et al Error! Bookmark not defined. demonstrated lower pain scores post-intervention with endoscopic drainage versus PTGBD. We confirmed this finding, demonstrating significantly lower median pain scores after EUS-GBD versus PT-GBD (2.5 vs 6.5, P < 0.005). Post-procedure discomfort,

ACCEPTED MANUSCRIPT

adjustment and learning to care for an external drain may also explain the significantly longer hospital lengths of stay after PT-GBD versus EUS-GBD. Although the adverse event profile is similar in both groups, pneumothorax and

RI PT

injury to intervening small bowel is specific to PT-GBD.21,22 PT-GBD may be inappropriate for patients with ascites or coagulopathy. EUS-GBD has a theoretical lower risk of

Error! Bookmark not defined.,3 defined.

SC

bleeding, given the tract from the gallbladder to the bowel is less vascular than the liver. Also ascites may not be as significant a risk with EUS-GBD using LAMS as

M AN U

it might be with PT-GBD. Four patients who were not candidates for PT-GBD due ascites underwent successful LAMS placement without bile leaks. It should be noted that pneumoperitoneum can be expected with EUS-GBD, as with any transmural endoscopic procedure, and less so with PT-GBD, but is inconsequential unless associated with a fluid

TE D

collection or bile leak.

In the absence of an ongoing obstruction of the cystic duct or gallbladder neck, an attempt to remove the PT-GBD is usually made at 4–8 weeks after placement. Thus

EP

resulting in at least 1 more intervention/tube check to remove the drain. However, catheter dislodgement or obstruction in the interim, and especially for drains that need

AC C

to be left in indefinitely, result in the need for multiple procedures. Our study clearly demonstrated a significantly higher number of reinterventions in the PT-GBD group than the EUS-GBD group (112 vs 11, P < 0.005). Our practice is to leave LAMS in indefinitely, given the frailty of most of these patients, to reduce the number of interventions and their associated risks, unless future evidence suggests the benefit of removing these stents.

ACCEPTED MANUSCRIPT

Finally, no differences were seen between the non-cautery (n=30) and cauteryenhanced LAMS (n=15) with regards to technical success, clinical success, and adverse

RI PT

events. However, the numbers of patients in the 2 groups were not similar and not large enough to show any differences given the overall very high technical and clinical success of EUS-GBD. The cautery-enhanced LAMS does eliminate the need for balloon dilation of

SC

the tract and can also be placed directly without the need for needle puncture or

guidewire access, thus shortening the procedure time. However, the cautery-enhanced

M AN U

LAMS costs $1,300 more than the non-cautery enhanced LAMS at our institution. Limitations of this study include the retrospective design, small sample size, and relatively short follow-up. Extrapolation of these results to hospitals with varying levels of endoscopic experience cannot be made at this time. In addition, the other endoscopic

TE D

option, ERCP transpapillary gallbladder stenting was not compared. EUS-GBD creates a chronic cholecysto-enteric fistula, and this will most likely increase the complexity of future surgery. Therefore, it should be limited at this time to patients who will not

EP

undergo future cholecystectomy as was the case in all our patients. Even though a single randomized controlled trial did not show any increased rates of conversion to open

AC C

cholecystectomy when EUS-GBD was performed with a nasobiliary drain versus PT-GBD, the nasobilairy drain was 5Fr, and the median time to cholecystectomy was 6 days.2 ERCP transpapillary gallbladder stenting maintains anatomic integrity, doesn’t preclude future cholecystectomy, and has good long-term efficacy data. A retrospective series in 46 patients showed no recurrent cholecystitis in 94% at 5 years23 and another study in 29 patients showed stent patency of 80% at 2 years.24 The limitations of transpapillary

ACCEPTED MANUSCRIPT

drainage via ERCP are the risks of pancreatitis and guidewire perforations of the cystic duct which are not seen with EUS-GBD and PT-GBD, and the obvious situation of an

RI PT

occluded cystic duct precluding placement of a stent through it. In conclusion, EUS-GBD has similar rates of technical and clinical success

compared to PT-GBD but is associated with lower post-intervention pain scores, shorter

SC

lengths of hospital stay, decreased need for reinterventions, and a trend toward fewer adverse events, which may make it a more attractive alternative to PT-GBD. Long-term

M AN U

data on EUS-GBD and a comparison with transpapillary stenting via ERCP are lacking. So the question remains as to the ideal management of acute cholecystitis in the nonoperative candidate. A 3-way prospective comparison of PT-GBD versus EUS-GBD and

Figure Legends

TE D

ERCP transpapillary gallbladder stenting could answer this question.

Figure 1. (A) CT scan demonstrates hyperdense material in the gallbladder 3 days after

EP

lumen apposing metal stent (LAMS) placement. (B) (C) (D) Endoscopy demonstrates large clot occluding the LAMS and filling the gallbladder, evacuated with a snare and

AC C

Roth Net®.

Figure 2. With severe ongoing pain 1 week after placement of a percutaneaous transhepatic gallbladder drain (PT-GBD), (A) (B) a CT scan demonstrates intervening loop of jejunum between abdominal wall and liver, (C) (D) with the drain traversing the

ACCEPTED MANUSCRIPT

jejunum. This was subsequently treated with internalization with a lumen apposing

RI PT

metal stent (LAMS) and drain removal with resolution of pain.

Video 1. Endoscopic ultrasound-guided gallbladder drainage (EUS-GBD) in a patient with

SC

acute cholecystitis.

References

2

M AN U

1 Hirota M, Takada T, Kawarada Y, et al. Diagnostic criteria and severity assessment of acute cholecystitis: Tokyo Guidelines. J Hepatobiliary Pancreat Surg 2007;14:78–82. Nahrwold D. Acute cholecystitis. Philadelphia (PA): WB Saunders; 1997.

3

Elyaderani M, Gabriele OF. Percutaneous cholecystostomy and cholangiography in patients with obstructive jaundice. Radiology 1979;130:601–2. 4

5

TE D

Glenn F. Cholecystostomy in the high risk patient with biliary tract disease. Ann Surg 1977;185:185–91. Kozarek. Selective cannulation of the cystic duct at time of ERCP. J Clin Gastroenterol 1984;6:37–40. 6

7

EP

Hasan MK, Itoi T, Varadarajulu S. Endoscopic management of acute cholecystitis. Gastrointest Endosc Clin North Am 2013;23:453–9.

8

AC C

Itoi T, Sofuni A, Itokawa F, et al. Endoscopic transpapillary gallbladder drainage in patients with acute cholecystitis in whom percutaneous transhepatic approach is contraindicated or anatomically impossible (with video). Gastrointest Endosc 2008;68:45–60. Kiviniemi H, Makela JT, Autio R, et al. Percutaneous cholecystostomy in acute cholecystitis in high-risk patients: an analysis of 69 patients. Int Surg 1999;83:299–302.

9

Lo LD, Vogelzang RL, Braun MA, et al. Percutaneous cholecystostomy for the diagnosis and treatment of acute calculous and acalculous cholecystitis. J Vasc Interv Radiol 1995;6:629–634.

ACCEPTED MANUSCRIPT

10

Lee MJ, Saini S, Brink JA, et al. Treatment of critically ill patients with sepsis of unknown cause: value of percutaneous cholecystostomy. AJR Am J Roentgenol 1991;156:1163–1166. 11

RI PT

Davis CA, Landercasper J, Gundersen LH, et al. Effective use of percutaneous cholecystostomy in high-risk surgical patients: techniques, tube management, and results. Arch Surg 1999;134: 727–731; discussion, 731–732. 12

SC

England RE, McDermott VG, Smith TP, et al. Percutaneous cholecystostomy: who responds? AJR Am J Roentgenol 1997;168:1247–1251. 13

McGahan JP, Lindfors KK. Percutaneous cholecystostomy: an alternative to surgical cholecystostomy for acute cholecystitis? Radiology 1989;173:481–485. Jang JW1, Lee SS, Song TJ, et al. Endoscopic ultrasound-guided transmural and percutaneous transhepatic gallbladder drainage are comparable for acute cholecystitis. Gastroenterology. 2012;142(4):805–811.

M AN U

14

Peñas-Herrero I1, de la Serna-Higuera C, Perez-Miranda M.1 Endoscopic ultrasoundguided gallbladder drainage for the management of acute cholecystitis (with video).J Hepatobiliary Pancreat Sci. 2014 Nov 13. doi: 10.1002/jhbp.182. 15

TE D

Irani S1, Baron TH2, Grimm IS2, et al. EUS-guided gallbladder drainage with a lumenapposing metal stent (with video). Gastrointest Endosc. 2015;82(6):1110–1115. doi: 10.1016/j.gie.2015.05.045. Epub 2015 Jul 2.

16

18

EP

Kedia P1, Sharaiha RZ1, Kumta NA1, et al. Endoscopic gallbladder drainage compared with percutaneous drainage. Gastrointest Endosc. 2015;82(6):1031–1036. doi: 10.1016/j.gie.2015.03.1912.

17

19

AC C

Cotton PB, Eisen GM, Aabakken L, et al. A lexicon for endoscopic adverse events: report of an ASGE workshop. Gastrointest Endosc. 2010;71:446–454. Itoi T, Binmoeller KF, Shah J, et al. Clinical evaluation of a novel lumen-apposing metal stent for endosonography-guided pancreatic pseudocyst and gallbladder drainage (with videos). Gastrointest Endosc. 2012;75:870–876. 20

Inoue T, Okumura F, Kachi K, Fukusada S, Iwasaki H, Ozeki T, Suzuki Y, Anbe K, Nishie H, Mizushima T, Sano H. Long-term outcomes of endoscopic gallbladder stenting in highrisk surgical patients with calculous cholecystitis (with videos). Gastrointest Endosc. 2016 May;83(5):905-13. doi: 10.1016/j.gie.2015.08.072. Epub 2015 Sep 10.

ACCEPTED MANUSCRIPT

21

England RE, McDermott VG, Smith TP, et al. Percutaneous cholecystostomy: who responds? AJR Am J Roentgenol 1997;168:1247–1251. 22

RI PT

McGahan JP, Lindfors KK. Percutaneous cholecystostomy: an alternative to surgical cholecystostomy for acute cholecystitis? Radiology 1989;173:481–485. 23

Maekawa S, Nomura R, Murase T, et al. Endoscopic gallbladder stenting for acute cholecystitis: a retrospective study of 46 elderly patients aged 65 years or older. BMC Gastroenterol 2013;13:65.

SC

24

AC C

EP

TE D

M AN U

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.

ACCEPTED MANUSCRIPT

Table 1. Indications and Baseline Characteristics of Patients Undergoing EUS-Guided Transmural Gallbladder Drainage (EUS-GBD), n = 45 vs Percutaneous Gallbladder Drainage (PT-GBD), n = 45 PT-GBD (n = 45) 75 (34– 94) 27M 18F 74 (44–125)

P Value 0.06

RI PT

EUS-GBD (n = 45) 65 (25– 87) 29M 16F 75 (48118)

Median age (yr) (range) Sex (M/F) Weight (Kg)

0.53 0.91

AC C

EP

TE D

M AN U

SC

Indication 0.06 Calculous cholecystitis 27 34 Acalculous cholecystitis 18 11 Severity of cholecystitis 0.13 Type 1 6 8 Type 2 29 23 Type 3 10 14 Underlying comorbidity precluding cholecystectomy 0.23 ASA 4 26 31 ASA 5 3 5 Widespread malignancy 16 9 Underlying condition precluding percutaneous transhepatic gallbladder drainage (PT-GBD) Perihepatic ascites 5 Coagulopathy/Need to resume anticoagulation 4 Concern for drain dislodgement (dementia, legally blind) 4 Patient refused 31 Median duration of cholecystitis prior to intervention 4 3 0.09 (days) (range)

ACCEPTED MANUSCRIPT

Table 2. Procedural Details of Patients Undergoing EUS-Guided Transmural Gallbladder Drainage (EUS-GBD), n = 45 vs Percutaneous Gallbladder Drainage (PT-GBD), n = 45.

TE D EP AC C

P Value 0.98 0.72

RI PT

PT-GBD (n = 45) 4 (2–7)

42 34 8 3 5 (11%) <0.0001 22 (12–30) 0.02 8 (2–12) 0.93

SC

43 27 16 2 40 (89%) 28 (18–52) 7 (1–16)

M AN U

Median gallbladder wall thickness (mm) (range) Location of obstruction Cystic duct Stone Malignant Gallbladder neck General anesthesia, n (%) Median procedure time (minutes) (range) Median duration for antibiotics for patients with cholecystitis (days) (range)

EUS-GBD (n = 45) 4 (2–6)

ACCEPTED MANUSCRIPT

Table 3. Procedural Details Relevant Only to Patients Undergoing EUS-Guided Transmural Gallbladder Drainage (EUS-GBD), n = 45

Dilation Device in case of cold LAMS

RI PT

32 13

30 15

M AN U

Dilation balloon

1 (1–2)

SC

Median number of EUS punctures (range) Site of puncture and drainage Transduodenal Transgastric LAMS type: Non cautery-enhanced (cold) Cautery-enhanced (hot) LAMS diameter (mm)/ length (mm) 10/10 15/10

cystotome Needle knife Blank

37 8

30 19 0 8 3

Additional Stent placed through LAMS

7

10mm x 6 cm fully covered biliary metal stent

1

AC C

EP

TE D

7Fr x 3 cm double pigtail plastic biliary stent

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT