Laparoscopic cholecystectomy is the gold standard for treating acute cholecystitis (AC).¹ However, percutaneous cholecystostomy (PC) is an alternative treatment for patients who are not candidates for early surgery due to underlying severe comorbidities, severe AC, or shock.² Interval cholecystectomy is the next step for definitive treatment in patients with initial PC. However, some patients remain at high risk for surgery even after their AC has improved using PC and antibiotics. To date, there are no consensus guidelines for the treatment of patients with AC undergoing PC who are unfit for interval cholecystectomy. In clinical practice, most such cases are observed after PC tube removal.

Endoscopic gallbladder (GB) drainage, consisting of endoscopic transpapillary GB stenting (ET-GBS) and endoscopic ultrasound-guided GB drainage (EUS-GBD), has emerged as an alternative to surgery or PC in patients with AC who are at high surgical risk.³ Many studies have reported that primary ET-GBS and EUS-GBD have high technical and clinical success rates with infrequent procedure-related complications,^4-11 and several recent studies have assessed the conversion of PC to ET-GBS or EUS-GBD in patients at high surgical risk.^12-16 However, no studies have compared endoscopic internalization of percutaneous cholecystostomy (EIPC) and conservative treatment following PC removal. The current study aimed to compare the clinical outcomes of EIPC followed by PC removal and those of conservative treatment following PC removal.

1. Patients

Data from consecutive patients treated with PC as a first-line treatment for AC were retrospectively collected between January 2017 and December 2020 using the Wonkwang University Hospital database. Patients who met any of the following criteria were excluded: (1) interval cholecystectomy, (2) underlying malignancy, (3) death without PC removal, and (4) loss to follow-up. The diagnosis and severity of AC were defined according to the 2018 Tokyo guidelines.¹⁷ The patients’ surgical risk was determined by a multidisciplinary team that included an anesthesiologist, a surgeon, and a physician. The following were considered as high-risk factors for surgery: age ≥80 years, an age-adjusted Charlson Comorbidity Index score of ≥4, and ≥grade 3 of the American Society of Anesthesiologists physical status classification. We proposed EIPC before PC removal in patients who underwent PC and were identified to be at high surgical risk. Whether to perform EIPC or conservative treatment with PC removal was determined with patient consent after explaining the risks and benefits associated with each option. EUS-GBD conversion was attempted when there was no cystic duct patency, prior endoscopic retrograde cholangiopancreatography (ERCP) history, or bile duct stones. In contrast, ET-GBS conversion was performed in patients with cystic duct patency, a history of ERCP, or concomitant bile duct stones.

The enrolled patients were divided into two groups (EIPC followed by PC removal or conservative treatment following PC removal). Background demographics, severity of AC, and disease progression in both groups were collected. Comorbidities were assessed using the age-adjusted Charlson Comorbidity Index and American Society of Anesthesiologists score. In the EIPC group, procedural details and procedure-related complications were recorded based on a retrospective review of electronic medical records.

The study protocol was reviewed and approved by the Institutional Review Board of Wonkwang University Hospital (IRB number: 2022-03-045). All procedures were performed after obtaining informed consent from patients or their family members.

2. Endoscopic procedures

1) Conversion of ET-GBS from PC

All ERCP procedures for ET-GBS were performed under conscious sedation by two skilled experts (T.H.K. and H.K.C.). After selective biliary cannulation, endoscopic sphincterotomy was performed in patients with naïve papillae, and if necessary, a respective bile duct procedure such as bile duct stone removal or biopsy of biliary stricture was performed. A cholangiogram was then obtained. If cystic duct patency was observed, a hydrophilic-coated guidewire (0.025-in VisiGlide; Olympus Medical Systems, Tokyo, Japan) with simultaneous manipulation of the catheter was inserted into the GB via the cystic duct. After successful placement of the catheter along the guidewire into the GB, a cholecystogram was obtained to ensure proper positioning of the catheter in the GB and to check for procedure-related cystic duct injuries. The guidewire was coiled 2 to 3 times in the GB to prevent its loss. Based on the endoscopist’s decision, a double-pigtail plastic stent (7-F, 12 or 15 cm in length, Zimmon^Ⓡ; Cook Medical, Bloomington, IN, USA) was placed between the GB and the lumen of the second portion of the duodenum (Supplementary Video 1). Digital cholangioscopy (SpyGlass DS Direct Visualization System; Boston Scientific, Natick, MA, USA) was performed in cases where guidewire insertion into the cystic duct was difficult. The stents were not regularly changed.

2) Conversion of EUG-GBD from PC

All the EUS-GBD procedures were performed by a highly experienced endosonographer (T.H.K.). Before starting the procedure, a significant amount of saline was infused through the PC to inflate the GB and free up the space for stent placement. A linear echoendoscope (GF-UCT 260; Olympus Medical Systems) was placed in the proper position for the procedure, and the GB neck was punctured using a 19-gauge fine-needle aspiration needle (EZ shot3 plus; Olympus Medical Systems). Contrast media were injected through the 19-gauge fine-needle aspiration needle, and the GB was visualized. The 19-gauge fine-needle aspiration needle was then removed, leaving 2 to 3 coils of the guidewire (0.025-in VisiGlide; Olympus Medical Systems) in the GB. Tract dilatation was performed using a 6-F cystotome (Cysto Gastro Set; Endoflex GmbH, Voerde, Germany), and an anti-migrating tubular metal stent (BONA-AL stent; Standard Sci Tech Inc., Seoul, Korea) was placed between the GB and gastrointestinal lumen. Based on the endoscopist’s decision preference, a lumen-apposing metal stent (LAMS) pre-loaded with an electrocautery delivery system (Hot Niti-S Spaxus; Taewoong Medical Co., Ltd., Goyang, Korea) was used (Supplementary Video 2).

3. Timing of the PC removal

In the conservative treatment group, the PC tube was clamped after confirming cystic duct patency by contrast medium injection through the PC tube at least 2 weeks after PC tube insertion. The PC tube was removed if the patient had no symptoms of AC recurrence approximately 1 week after clamping of the PC tube. However, if cholecystography showed no patency of the cystic duct, cholecystography was repeated 1 week later. In the EIPC group, the PC tube was clamped after EUS-GBD or ET-GBS conversion and was removed at least 3 weeks after the initial PC tube insertion in clinically stable patients. However, if the inadvertent PC tube was dislodged before the scheduled PC tube removal or confirmation of cystic duct patency, the PC tube was removed, and the patient was observed.

4. Outcome measurements

The primary outcome was AC recurrence, and the secondary outcome was unplanned readmission owing to gallstone-related disease. Technical and clinical success rates and procedure-related complications were measured in the EIPC group.

AC recurrence was defined as the recurrence of typical symptoms, laboratory results, and radiologic findings compatible with AC according to the 2018 Tokyo guidelines, after complete resolution of the clinical signs and symptoms of a previous episode of AC.¹⁷ Unplanned readmission due to gallstones was recorded during hospitalization during the study period.

In ET-GBS, technical success was defined as the placement of a double-pigtail plastic stent between the GB and lumen of the duodenum. In EUS-GBD, technical success was defined as the maintenance of proper GB drainage via the placement of a metal stent. Clinical success was defined as no AC recurrence within 14 days of PC removal after EIPC conversion. Procedure-related complications were defined as events that occurred within seven days after the procedure. The gallstone-related disease was defined as the occurrence of acute calculus cholecystitis or acute cholangitis secondary to gallstones.

5. Statistical analyses

Statistical analyses were performed using SPSS Statistics version 22.0 (IBM Corp., Armonk, NY, USA). Comparisons were conducted using the chi-square test or Fisher exact test for categorical data, and the Wilcoxon signed-rank test or Student t-test for continuous data, where appropriate. The results are presented as the median and interquartile range (IQR) or mean and standard deviation. Statistical significance was set at p<0.05. Recurrence curves were plotted using the Kaplan-Meier method, and differences between groups were assessed using the log-rank test. Finally, multivariate analysis was performed using a logistic model using data that were significantly different between groups.

1. Baseline characteristics

Between January 2017 and December 2020, 571 patients underwent PC. Among them, 333 patients were excluded for the following reasons: interval cholecystectomy (213 patients), loss to follow-up (36 patients), malignancy (24 patients), and death without PC removal (60 patients). Therefore, 238 patients (63 EIPC cases, including 55 ET-GBS and 8 EUS-GBD cases, and 175 cases of conservative treatment following PC removal) who satisfied the inclusion criteria were included (Fig. 1). Demographic characteristics of the enrolled patients are shown in Table 1. There were no statistically significant differences between the two groups with respect to sex, age, age-adjusted Charlson Comorbidity Index, severity of AC, underlying dementia, use of antiplatelets/anticoagulants, or median follow-up period. However, calculus cholecystitis and concomitant bile duct stones were significantly more frequently observed in the EIPC group than in the conservative treatment group (calculus cholecystitis, 60 [95.24%] vs 117 [66.86%], p<0.001; concomitant bile duct stones, 29 [46.03%] vs 27 [15.43%], p<0.001). The median periods from PC to EIPC in the EIPC group and from PC placement to PC removal in the conservative treatment group were 17 days (IQR, 10 to 42 days) and 32 days (IQR, 25 to 45 days), respectively.

Figure 1. Flowchart of enrolled patients.
PC, percutaneous cholecystostomy; E-GBD, endoscopic gallbladder drainage; ET-GBS, endoscopic transpapillary gallbladder stenting; EUS-GBD, endoscopic ultrasound-guided gallbladder drainage.

Table 1 . Demographics and Baseline Characteristics of the Study Cohort.

Characteristic	EIPC group (n=63)	Conservative treatment group (n=175)	p-value
Male sex	35 (55.56)	99 (56.57)	0.889
Age, yr	82.48±7.37	80.63±8.99	0.146
Age ≥80 yr	47 (74.60)	114 (65.14)	0.169
Age-adjusted CCI, yr	6.48±1.38	6.49±1.71	0.944
ASA-PS (III/IV)	60/3	162/13	0.570*
Dementia	30 (47.62)	64 (36.57)	0.124
Antiplatelet/anticoagulant agent	34 (53.97)	87 (49.71)	0.563
Severity of acute cholecystitis			0.783
II	38 (60.32)	109 (62.29)
III	25 (39.68)	66 (37.71)
Calculus cholecystitis	60 (95.24)	117 (66.86)	<0.001
Acalculous cholecystitis	3 (4.76)	58 (33.14)	<0.001
Concomitant bile duct stone	29 (46.03)	27 (15.43)	<0.001
Period of PC removal from PC insertion, day	41 (32–56)	32 (25–45)	0.192
Length of hospital stay, day	14.68±13.20	15.24±14.16	0.783
Follow-up period, day	656 (398–1,198)	822 (565–1,282)	0.057†
Patient status on follow-up			0.116
Dead	11 (17.46)	48 (27.43)
Alive	52 (82.54)	127 (72.57)

Data are presented as number (%), mean±SD, or median (interquartile range)..

EIPC, endoscopic internalization of percutaneous cholecystostomy; CCI, Charlson Comorbidity Index; ASA-PS, American Society of Anesthesiologists physical status; PC, percutaneous cholecystostomy..

*Fisher exact test; ^†Wilcoxon two-sample test..

2. Clinical outcomes

The EIPC group had a significantly lower cumulative incidence rate of AC recurrence than the conservative treatment group (p=0.017, log-rank test) (Fig. 2). The clinical outcomes of the two groups are shown in Table 2. The EIPC group had significantly less frequent AC recurrence (3 [4.76%] vs 31 [17.71%], p=0.012) and fewer unplanned readmissions due to gallstone-related diseases (6 [9.52%] vs 40 [22.86%], p=0.022). However, in terms of 30-day events, including AC recurrence and unplanned readmission due to gallstone-related disease, there were no significant differences between the two groups. In the EIPC group, AC recurrence occurred 2, 110, and 371 days after PC removal following EIPC and was managed by repeat endoscopic GB drainage (2 ET-GBS cases and 1 EUS-GBD case). The median periods of AC recurrences after PC removal was 147 days (IQR, 47 to 209 days) in the conservative treatment group. In the conservative treatment population with AC recurrence, the treatments performed were as follows: 12 patients underwent repeat PC, nine underwent ET-GBS conversion following PC, seven underwent ET-GBS, and three underwent EUS-GBD. In the univariate and multivariate analyses, calculus cholecystitis (odds ratio, 13.75; 95% confidence interval, 1.83 to 102.83; p=0.011) and EIPC (odds ratio, 0.23; 95% confidence interval, 0.06 to 0.78; p=0.019) were significant predictive factors for recurrent AC (Table 3). No PC tube removal-related adverse events, such as bile leaks or peritonitis, were observed.

Figure 2. The Kaplan-Meier curve of the cumulate incidence of acute cholecystitis recurrence with comparison between the endoscopic gallbladder drainage conversion group and the conservative treatment group.
PC, percutaneous cholecystostomy; E-GBD, endoscopic gallbladder drainage.

Table 2 . Comparison of Clinical Outcomes between the Endoscopic Gallbladder Drainage Conversion and Conservative Treatment Groups.

	EIPC conversion group (n=63)	Conservative treatment group (n=175)	p-value
30-day recurred acute cholecystitis	1 (1.59)	2 (1.14)	0.604
30-day acute cholangitis requiring ERCP	0	0	>0.999
30-day unplanned readmission due to gallstone-related diseases	1 (1.59)	2 (1.14)	0.604
Overall recurred acute cholecystitis	3 (4.76)	31 (17.71)	0.012
Overall acute cholangitis requiring ERCP	5 (7.94)	18 (10.29)	0.588
Overall unplanned readmission due to gallstone-related diseases	6 (9.52)	40 (22.86)	0.022
Re-intervention for acute cholecystitis	3 (4.76)	31 (17.71)	0.012

EIPC, endoscopic internalization of percutaneous cholecystostomy; ERCP, endoscopic retrograde cholangiopancreatography..

Table 3 . Univariate and Multivariate Analyses for Predicting Factors of Recurred Acute Cholecystitis.

Factor		Univariate analysis			Multivariate analysis
		OR (95% CI)	p-value		OR (95% CI)	p-value
Age		0.99 (0.95–1.03)	0.633
Sex	Male	1.50 (0.70–3.20)	0.288
	Female	1
Age-adjusted CCI		1.23 (0.98–1.53)	0.064
ASA-PS (III/IV)	IV	0.38 (0.04–2.98)	0.359
	III	1
Dementia	Yes	1.43 (0.69–2.98)	0.331
	No	1
Severity of acute cholecystitis	III	0.86 (0.40–1.84)	0.703
	II	1
Calculus cholecystitis	Yes	13.75 (1.83–102.83)	0.011		20.83 (2.74–158.02)	0.003
	No	1			1
Concomitant bile duct stone	Yes	1.00 (0.42–2.35)	>0.999
	No	1
Cystic duct stone	Yes	1.06 (0.29–3.84)	0.924
	No	1
Gallbladder perforation	Yes	0.87 (0.28–2.67)	0.814
	No	1
Period from PC to tube removal		1.00 (0.99–1.01)	0.300
Cystic duct patency via tube cholangiogram	Yes	0.50 (0.18–1.37)	0.181		1.92 (0.65–5.61)	0.232
	No	1
PC removal type	Scheduled	0.62 (0.26–1.50)	0.296
	Accident	1
Group	EIPC	0.23 (0.06–0.78)	0.019		0.13 (0.03–0.47)	0.000
	Conservative treatment	1

OR, odds ratio; CI, confidence interval; CCI, Charlson Comorbidity Index; ASA-PS, American Society of Anesthesiologists physical status; PC, percutaneous cholecystostomy; EIPC, endoscopic internalization of percutaneous cholecystostomy..

3. Procedure details, clinical outcomes, and procedure-related complications in ET-GBS conversion from PC

The technical and clinical success rates of ET-GBS conversion from PC were 88.7% (55/62) and 98.1% (54/55), respectively. Five patients underwent digital cholangioscopy-guided cystic duct cannulation. Among the seven cases of technical failure, five involved failure of catheter insertion into the cystic duct due to severe cystic duct stricture and two involved failure of guidewire insertion into the cystic duct. Most procedures used 7-F, 12-cm double-pigtail plastic stents, and the median procedure time was 15.0 minutes (IQR, 9.8 to 22.5 minutes). Procedure-related complications occurred in three patients (one patient each had post-ERCP pancreatitis, endoscopic sphincterotomy bleeding, and a hematoma around the cystic duct) and they were managed with conservative treatment or endoscopic therapy. Stent migration occurred in three patients, one of whom experienced AC recurrence and was treated with EUS-GBD. The median stent patency was 656 days (IQR, 397 to 1,167 days) (Table 4).

Table 4 . Procedural Details, Clinical Outcomes, and Adverse Events in Patients with ET-GBS Conversion from PC.

Variable	ET-GBS conversion from PC (n=55)
Technical success	55/62 (88.7)
Clinical success	54/55 (98.1)
Length of stent (12/15 cm)	52/3
Caliber of stent (7-F)	55 (100)
Native papilla	36 (58.1)
Prior endoscopic biliary sphincterotomy	26 (41.9)
Procedure time, min	15.0 (9.8–22.5)
Spyglass-guided intervention	5 (9.1)
Procedure-related adverse events
Post-ERCP pancreatitis	1 (1.8)
EST bleeding	1 (1.8)
Hematoma	1 (1.8)
Recurred acute cholecystitis	3 (5.4)
Stent migration	3 (5.4)
Biliary events	5 (9.1)
Procedure-related mortality	0
Sent patency, day	656 (397–1,167)
Patient status on follow-up
Dead	9 (16.4)
Alive	46 (83.6)

Data are presented as number (%) or median (interquartile range)..

ET-GBS, endoscopic transpapillary gallbladder stenting; PC, percutaneous cholecystostomy; ERCP, endoscopic retrograde cholangiopancreatography; EST, endoscopic sphincterotomy..

4. Procedure details, clinical outcomes, and procedure-related complications in the EUS-GBD conversion from PC

Technical and clinical success were achieved in 100% (8/8) of cases without any procedure-related complications. An anti-migrating tubular metal stent and LAMS with cautery enhancement were used in two and six patients, respectively. Most puncture sites involved the duodenal bulb (n=6, 75%), and the median duration of the procedure was 6 minutes (IQR, 5 to 8 minutes). There was no recurrence of AC, and spontaneous stent migration occurred in one patient. The median stent patency was 377 days (IQR, 304 to 385 days). Cholecystoscopy was not performed with EUS-GBD (Table 5).

Table 5 . Procedural Details, Clinical Outcomes, and Adverse Events in Patients with EUS-GBD Conversion from PC.

Variable	EUS-GBD conversion from PC (n=8)
Technical success	8
Clinical success	8
Metal stent type
LAMS with cautery enhanced (hot) (10 mm×2 cm)	6
Anti-migrating tubular metal stent (10 mm×6 cm)	2
Puncture root (duodenum/antrum)	6/2
Dilation device
Cystotome	2
Dilation balloon	0
Needle knife	0
Procedure time, median (IQR), min	6 (5–8)
Procedure-related adverse events	0
Recurred acute cholecystitis	0
Procedure-related mortality	0
Median stent patency, median (IQR), day	377 (304–385)
Patient status on follow-up
Dead	2
Alive	6

EUS-GBD, endoscopic ultrasound-guided gallbladder drainage; PC, percutaneous cholecystostomy; LAMS, lumen-apposing metal stent; IQR, interquartile range..

In the current study, we demonstrated that compared with conservative treatment following PC removal, EIPC significantly reduced AC recurrence and unplanned readmissions due to gallstone-related diseases. In addition, EIPC has high technical and clinical success rates, and infrequent procedure-related complications. To the best of our knowledge, this is the first study to compare the clinical outcomes of EIPC followed by PC removal and conservative treatment following PC removal.

Laparoscopic cholecystectomy is the cornerstone of AC, but perioperative morbidity and mortality rates are considerable in patients at high risk due to severe AC, underlying serious comorbidities, or old age.¹⁸ In clinical practice, PC can still be considered an alternative to urgent cholecystectomy in patients who are unfit for surgery. However, the optimal strategy for patients unable to undergo interval cholecystectomy after reassessing the anesthesia risk remains undefined. PC maintenance is often cumbersome, and the recurrence rate of AC after PC removal without subsequent surgery has been reported to range from 22% to 45%.^19-21 In the current study, the recurrence rate of AC was 17.7%, which is comparable to the findings of other studies, despite the exclusion of patients with malignant disease. However, EIPC lowered the recurrence rate by 4.7%.

Currently, there are no definitive guidelines on the timing of PC tube removal. Tract maturation is important to avoid complications, such as bile leakage or peritonitis, after PC tube removal. There is currently no method for determining tract maturity. However, various studies have shown that tract maturation occurs within 20 days and tube removal is safe and effective after a minimum of 3 weeks.^22-24 In the current study, PC tube removal was scheduled with clamping tests at least 3 weeks after PC tube insertion based on patients’ comorbidities and findings from previous studies, except for patients who were spontaneously dislodged. Therefore, the PC tube was removed a while after the fistula track matured, resulting in no adverse events associated with PC tube removal.

EUS-GBD and ET-GBS have been established as alternatives to PC. A systematic review reported that EUS-GBD led to superior outcomes, including technical success, clinical success, and reduced AC recurrence, compared to ET-GBS.²⁵ Moreover, with the advent of LAMS with and without an electrocautery tip, EUS-GBD is technically easier, safer, and has good long-term outcomes.^26,27 However, most existing studies included endoscopic GB drainage as the primary purpose. A few studies have reported the conversion of PC to EUS-GBD as a secondary internal GB drainage approach.^12-15 The technical success rate of EUS-GBD conversion ranges from 90.5% to 100%, and the clinical success rate ranges from 89.5% to 100%. Additionally, procedure-related complications occurred in 0% to 21.4% of the cases, and the AC recurrence rate was 0% to 28.6%. In the current study, an anti-migrating tubular metallic stent was inserted in the first two procedures, and a LAMS with cautery enhancement was used for the remainder of the procedure. If the scope is stable and can be tolerated by the patient, an anti-migrating tubular metallic stent can be safely used. However, we assumed that the use of a LAMS with cautery enhancement could shorten the procedure time, which would be helpful for patients with severe underlying diseases. Furthermore, unlike primary EUS-GBD, fibrotic changes in the thickened GB wall following AC improvement in patients with PC can cause technical difficulties during GB puncture or tract dilatation. Furthermore, a collapsed GB can be problematic because of inadequate space for stent placement. However, these problems can be solved by infusing normal saline before and during the procedure and by using a LAMS with a pre-loaded electrocautery delivery system that does not require tract dilatation. The LAMS used in this study has the advantage of allowing the placement of a stent, even in a small GB, by moving the delivery device up and down. In the current study, clinical and technical success were achieved in all cases of attempted EUS-GBD conversion. In addition, no AC recurrences or procedure-related complications were observed.

The technical and clinical success rates of the conversion of PC to ET-GBS in the present study were 88.7% and 98.1%, respectively. Secondary ET-GBS from the PC can be beneficial for recognizing the patency and course of the cystic duct by contrast medium injection via the PC. The guidewire was inserted into the GB, but sometimes the catheter along the guidewire could not be advanced because of a severe cystic duct stricture. This was also the cause of most failures in this study. When ameliorating severe cystic duct strictures, caution is required when using a mechanical or balloon dilator, which can cause cystic duct injury.

The advantages of ET-GBS conversion include the use of the standard technique and devices of ERCP, maintenance of normal biliary anatomy, simultaneous treatment of bile duct stones, and lower costs than those of EUS-GBD conversion. The disadvantages of ET-GBS conversion are the relatively low technical success rate, post-ERCP complications, and the lack of guidelines for stent change. The advantages of EUS-GBD conversion include possible intervention with cholecystoscopy, the evolution of dedicated devices, and a high technical success rate. The disadvantages of EUS-GBD conversion are its relatively high cost, technical difficulty, lack of standardization, and stent-related complications. Therefore, the procedure should be selected considering the advantages and disadvantages of each procedure, patient’s condition, and available procedural experts.

The present study has several limitations. First, this was a single-center, retrospective investigation. Thus, a selection bias and confounding variables may have been present. Second, the sample size of the current study was small, and a type 2 error may have occurred. In particular, the cohort that underwent EUS-GBD conversion from PC was small; however, several studies on EUS-GBD conversion from PC with metallic stents also had very small sample sizes, ranging from six to eight cases. Third, EIPC procedures were conducted by experienced endoscopists, and the results may not be generalizable to other centers. In addition, EUS-GBD and ET-GBS are completely different methods, so there was a problem in analyzing the two methods as EIPC procedures. However, ET-GBS and EUS-GBD can be considered for the same purpose in terms of endoscopic internalization. The purpose of the current study was not to compare the two modalities, but rather to determine the clinical usefulness of endoscopic internalization in indwelling PC with benign disease. Fourth, the cost-effectiveness of the procedures was not compared. EUS-GBD requires dedicated devices and stents; thus, the procedure-related costs are relatively high. However, unplanned readmissions for gallstone-related diseases occurred significantly more frequently in the conservative treatment group than in the EIPC group. In addition, the rate of concomitant bile duct stones in the ET-GBS conversion group was 50.9% (28/55); thus, ERCP for bile duct stone removal was required in approximately half of this group, regardless of whether ET-GBS conversion was performed. Considering these aspects, total hospital costs may be higher in the conservative treatment group. Based on the current study, our hospital is currently attempting EIPC in all high-risk surgical patients with PC as the first-line treatment for AC if the patient consents to the procedure. We propose a treatment algorithm for patients with AC and an indeterminate surgical risk at initial presentation (Fig. 3).

Figure 3. Proposed algorithm for the treatment of patients with AC and indeterminate surgical risk at initial presentation.
AC, acute cholecystitis; ERCP, endoscopic retrograde cholangiopancreatography; ET-GBS, endoscopic transpapillary gallbladder stenting; EUS-GBD, endoscopic ultrasound-guided gallbladder drainage; PC, percutaneous cholecystostomy.

In conclusion, EIPC led to more favorable outcomes with lower procedure-related complication rates than conservative treatment after PC removal. Therefore, EIPC can be considered a promising treatment option for patients at high surgical risk for PC. In particular, EUS-GBD conversion by an expert led to high technical and clinical success rates with few adverse events. Future well-designed prospective studies with a larger number of patients are required to validate our results.

Supplementary materials can be accessed at https://doi.org/10.5009/gnl230019.

This paper was sponsored by Soongsan Fellowship at Wonkwang University in 2022.

The authors thank So Hyun Park, In Hee So, Seo Hee Choi, and Jun Ah Park of Wonkwang University Hospital for their assistance with the procedures. The authors are grateful to Hee Heun Kim and Ju Hee Ahn of Wonkwang University Hospital who served as physician assistants.

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

Study concept and design: H.K.C. Data acquisition: H.K.C. Data analysis and interpretation: H.K.C. Drafting of the manuscript: H.K.C. Critical revision of the manuscript for important intellectual content: T.H.K., S.H.K. Statistical analysis: H.K.C. Obtained funding: H.K.C. Administrative, technical, or material support; study supervision: T.H.K., S.H.K. Approval of final manuscript: all authors.