The efficacy of bedside percutaneous cholecystostomy in critically ill patients: comparison with cholecystostomy performed in angiography suite

Change history

6/9/2026

An Editorial Expression of Concern to this paper has been published: 10.1007/s00423-026-04093-4

Abstract

Background

Critically ill patients frequently develop acute acalculous cholecystitis (AAC), particularly those requiring prolonged mechanical ventilation, ECMO, or CRRT. Percutaneous cholecystostomy is a minimally invasive alternative to cholecystectomy, yet transportation to the angiography suite may not be feasible in unstable ICU patients. This study aimed to compare the efficacy and outcome of bedside percutaneous cholecystostomy with the conventional technique performed in the angiography suite.

Methods

We retrospectively reviewed 70 ICU patients who underwent percutaneous cholecystostomy between January 2019 and December 2023 at a tertiary referral center. Patients were divided into two groups based on procedure setting: bedside trocar group (n = 26) and angiography suite group (n = 44). Baseline characteristics, SOFA scores, procedural outcomes, and complications were compared between the groups.

Results

The bedside group had higher rates of ECMO use (50% vs. 6.8%, p < 0.001), inotrope support (76.9% vs. 52.3%, p = 0.041), and NO ventilation (30.8% vs. 6.8%, p = 0.008). Despite significantly higher pre-procedural SOFA scores (12.85 ± 2.59 vs. 9.77 ± 3.38, p < 0.001), there were no significant differences in complication rates (42.3% vs. 27.3%, p = 0.196), catheter dislodgement, or in-hospital mortality (53.8% vs. 50.0%, p = 0.756) between groups. Catheter dislodgement was the most common complication and was not significantly associated with the technique used.

Conclusion

Bedside percutaneous cholecystostomy is a viable alternative for critically ill patients who are unable to be moved. This method provides effective biliary decompression even in patients with high disease severity who require multi-organ support.

Key points

Question

How do bedside and angiography suite percutaneous cholecystostomy compare in terms of safety and efficacy for ICU patients with acute cholecystitis?

Findings

Bedside cholecystostomy (n = 26) used the Trocar technique with a transperitoneal approach, while angiography suite cholecystostomy (n = 44) used either the Trocar (n = 12) or Seldinger technique with a transhepatic approach (n = 32). There were no significant differences in procedure duration, length of hospital or ICU stay, complications, or mortality between the two methods. The SOFA score was higher in the bedside group, reflecting more severe conditions.

Meaning

Bedside percutaneous cholecystostomy is as effective and safe as the angiography suite approach, especially for immobile ICU patients.

Introduction

Acute cholecystitis is a common surgical emergency typically managed by cholecystectomy, which remains the gold standard of care [1, 2]. However, in critically ill or high-risk surgical patients, percutaneous cholecystostomy has emerged as a valuable alternative, offering effective biliary decompression without the risks of general anesthesia or major surgery [3].

Percutaneous cholecystostomy can be performed using either the Seldinger technique, commonly done under fluoroscopic guidance in the angiography suite [4], or the trocar technique, which can be safely executed at the patient’s bedside under ultrasound guidance alone [5].

With advancements in critical care, the survival of ICU patients has improved significantly. On the contrary, prolonged ICU stays and use of variable interventions have contributed to an increased incidence of acute acalculous cholecystitis (AAC) in ICU patients. Unfortunately, moreover, critically ill patients are connected with a lot of life support system such as ventilator, extracorporeal Membrane Oxygenation (ECMO) or continuous renal replacement therapy (CRRT), and often exhibit unstable vital signs, making transport to the angiography suite for image-guided interventions hazardous or impractical. In these cases, bedside cholecystostomy using the trocar method could be a potentially life-saving solution.

While bedside cholecystostomy using the trocar technique offers logistical and potential clinical advantages, it is not yet widely adopted as the standard approach for critically ill patients [6, 7]. This is partly due to concerns about the safety and technical feasibility of performing invasive procedures in a non-sterile ICU environment, as well as the potential limitations of ultrasound guidance alone compared to fluoroscopic imaging. The bedside approach may also pose challenges in terms of catheter placement accuracy and complication control, especially in patients with poor anatomical windows or severe coagulopathy [6, 8].

The aim of this study was to determine whether bedside percutaneous cholecystostomy represents a feasible and effective therapeutic alternative for critically ill patients who are unable to undergo transfer or conventional procedures. To this end, we compared the efficacy, complication and outcomes of percutaneous cholecystostomy performed at the bedside in the intensive care unit (ICU) with those performed in the angiography suite.

Materials & methods

Study design

This is a retrospective study was designed to include consecutive patients undergoing percutaneous cholecystostomy at single large tertiary university hospital. Institutional board approval was obtained, and informed consent was waived.

During the 5-year period from January 2019 to December 2023, a total of 452 patients underwent percutaneous cholecystostomy, as documented in the electronic medical records. Technical failure of the procedure occurred in 2 patients. Additionally, 380 procedures were conducted on patients in the emergency room or general ward, resulting in their exclusion. Ultimately, 70 patients from the ICU were included in the study.

In the majority of ICU patients, classical symptoms or signs such as fever, right upper quadrant pain, and Murphy’s sign are typically absent. Therefore, the diagnosis of acute cholecystitis relies heavily on a combination of laboratory findings (elevated levels of inflammatory markers such as C-reactive protein [CRP], white blood cell count, and bilirubin) and imaging results from ultrasound (US) or computed tomography (CT). When feasible, CT is preferred due to its ability to comprehensively assess the entire abdomen.

The indications for percutaneous cholecystostomy in our study population included: (1) Suspected or confirmed acute cholecystitis in critically ill patients who were deemed poor surgical candidates, (2) Sepsis or systemic inflammatory response syndrome (SIRS) suspected to be of biliary origin, and (3) Gallbladder distention with signs of inflammation or infection on imaging, such as gallbladder wall thickening, pericholecystic fluid, or a positive Murphy’s sign on ultrasound.

When possible, the procedure was performed in the angiography suite with patient transfer. For patients unable to be moved due to various reasons, the procedure was conducted bedside. Among these, percutaneous cholecystostomy was performed at the bedside for 26 patients, classified as the bedside group. The remaining 44 patients underwent the procedure in the angiography suite, forming the angiography suite group (Fig. 1).

Fig. 1.

Flow chart in study patient

We investigated patients’ age, gender, comorbidities, ECMO status, presence of NO ventilation, CRRT status, use of inotropics, feeding status, vital signs (i.e., SBP, DBP, HR, RR, BT, SpO2), laboratory findings (i.e., WBC, Hb, PLT, PaO2, creatinine, PT, INR, aPTT, bilirubin, CRP), SOFA score, and diagnostic tools (i.e., ultrasound, CT) across two groups (Table 1). We defined clinical improvement as discontinuation of inotropics within 3 days post-procedure, a decrease in SOFA score by 2 points or more, a significant reduction in WBC compared to baseline, and a decrease in CRP of more than 50% or a clear continuous decline. A pre-procedural transfusion was administered in cases where the platelet count was less than 50,000 or the PT INR was 1.5 or above.

Table 1.

Patients base characteristics

	Bedside (N = 26)	Angiography suite (N = 44)	P
Age(y), mean ± SD	65.31 ± 8.44	71.09 ± 13.061	0.047
Sex			0.181
Male	20 (76.9)	27 (61.4)
Female	6 (23.1)	17 (38.6)
Comorbidity
Hypertension	14 (53.8)	18 (40.9)	0.294
Diabetes mellitus	8 (30.8)	14 (31.8)	0.927
Cardiac disease	9 (34.6)	10 (22.7)	0.280
Pulmonary disease	5 (19.2)	12 (27.3)	0.448
Gastrointestinal disease	2 (7.7)	0 (0)	0.062
Malignant disease	7 (26.9)	7 (15.9)	0.036
Cerebrovascular disease	1 (3.8)	10 (22.7)	0.279
Liver disease	2 (7.7)	1 (2.3)	0.279
End-stage renal disease	5 (19.2)	8 (18.2)	0.913
ECMO	13 (50)	3 (6.8)	< 0.001
NO vent	8 (30.8)	3 (6.8)	0.008
CRRT	14 (53.8)	16 (36.4)	0.153
Inotropics	20(76.9)	23 (52.3)	0.041
Feeding	11(42.3)	40 (90.9)	< 0.001
Vital sign
SBP (mmHg) mean ± SD	106.269 ± 22.324	145.41 ± 24.276	< 0.001
DBP (mmHg) mean ± SD	59 ± 9.338	82.52 ± 17.263	< 0.001
HR mean ± SD	96.42 ± 20.967	89.59 ± 17.533	0.148
RR mean ± SD	18 ± 5.012	19.68 ± 3.183	0.133
BT (℃) mean ± SD	36.6 ± 0.384	36.45 ± 0.285	0.094
SpO2 (%) mean ± SD	96.5 ± 4.111	98.73 ± 2.061	0.015
Laboratory findings
WBC (X103㎕)	9.41 ± 7.945	11.08 ± 6.044	0.324
Hb	8.72 ± 0.867	8.945 ± 1.549	0.444
PLT	83.35 ± 59.797	117.82 ± 86.796	0.078
PaO2	100.5 ± 35.15	106.18 ± 29.949	0.475
Creatinine	0.98 ± 0.383	1.02 ± 0.715	0.774
PT INR	1.45 ± 0.537	1.25 ± 0.233	0.077
aPTT	59.84 ± 33.944	50.04 ± 26.483	0.195
bilirubin	5.59 ± 4.721	3.93 ± 3.786	0.111
CRP	14.84 ± 8.909	13.09 ± 10.143	0.47
SOFA	12.85 ± 2.588	9.77 ± 3.375	< 0.001
Diagnostic method			0.001
Ultrasound	25 (96.2)	27 (61.4)
CT	1 (3.8)	17 (38.6)
Approach (technique)			< 0.001
Transhepatic (Seldinger)	0 (0)	32 (72.7)
Transperitoneal (Trocar)	26 (100)	12 (27.3)

Procedures

In the bedside group, a total of 26 patients underwent cholecystostomy using the trocartechnique. All patients were positioned supine, and a transperitoneal approach was employed. Cholecystostomy was executed with an 8.5-French trocar pigtail catheter (Dawson-Mueller Drainage catheter, Cook, Bloomington, IN). In the trocar technique, the catheter’s loop is straightened over the stylet and stiffening cannula. Local anesthesia was administered using 2% lidocaine, followed by puncturing of the skin, subcutaneous tissues, and gallbladder wall under ultrasound guidance. Once the catheter’s position in the gallbladder was confirmed, it was advanced over the cannula so that the distal loop reformed into a pigtail shape, secured by pulling the string. Typically, 25 cc of bile was aspirated post-procedure for culture. In cases where the bile was too concentrated to drain naturally, manual aspiration was performed as much as feasible. This technique involves a single puncture, making this a very simple procedure that can be performed at the patient’s bedside.

In 12 of 44 patients who underwent cholecystostomy in the angiography suite, trocar technique was performed in a same way. For the remaining 32 patients, cholecystostomy was performed using the Seldinger technique. All patients undergoing the Seldinger technique were approached via the transhepatic route. Initial access was achieved under ultrasound guidance using an 18-gauge Chiba needle under local anesthesia. Subsequently, a 0.035-inch, 150 cm-long hydrophilic guidewire (Terumo, Tokyo, Japan) was advanced through the needle, followed by placement of an 8.5-French pigtail catheter (Multipurpose Drainage Catheter; Sungwon medical, Cheongju, Korea) over the wire under continuous fluoroscopic guidance.

All steps of both procedures, including preparation, implementation, and completion, were conducted in compliance with the relevant standard clinical guidelines. Both procedures were conducted by an interventional radiologist with over three years of experience in both trocar and Seldinger techniques.

Results

Patients base characteristics

Patient demographic and clinical characteristics, the imaging method used for diagnosis and anatomical approach route are summarized in Table 1.

Significant differences in age were observed between the two groups. The bedside group exhibited a younger age profile compared to the angiography suite group (5.31 ± 8.44 vs. 71.09 ± 13.061, P = 0.047). Among comorbidities, only cardiovascular disease demonstrated a significant difference (1 vs. 10, P = 0.036). Bedside group patients more frequently received ECMO (13 vs. 3, P < 0.001), NO vent (8 vs. 3, P = 0.008), and inotropic support (20 vs. 23, P = 0.041) compared to the angiography suite group. Conversely, more patients in the angiography suite group were fed (11 vs. 40, P < 0.001).

Systolic blood pressure (106.269 ± 22.324 vs. 145.41 ± 24.276, P < 0.001), diastolic blood pressure (59 ± 9.338 vs. 82.52 ± 17.263, P < 0.001), and SpO2 (96.5 ± 4.111 vs. 98.73 ± 2.061, P = 0.015) were significantly lower in the bedside group. Laboratory findings did not differ significantly between the two groups. However, the Sequential Organ Failure Assessment (SOFA) Score was significantly higher in the bedside group (12.85 ± 2.588 vs. 9.77 ± 3.375, P < 0.001).

Regarding diagnostic methods, the bedside group showed a greater incidence of diagnosing acute cholecystitis using ultrasound (25 vs. 27, P = 0.001). In the context of percutaneous cholecystostomy, all patients in the bedside group underwent the procedure via the transperitoneal route with trocar technique, which was significantly more common compared to the angiography suite group (26 vs. 12, P < 0.001).

Compare clinical outcomes between bedside group and angiography suite group

Clinical outcomes are summarized in Table 2. Statistically significant differences were observed only in the SOFA score after the procedure (12.5 ± 3.289 vs. 9 ± 3.41, P < 0.001).There were no significant difference between two groups in procedure duration, change in SOFA score, length of hospital stay, length of ICU stay, complications and mortality. According to the preceding definition, clinical improvement was observed in 12 of 26 patients (46.2%) in the bedside group and in 25 of 44 patients (56.8%) in the angiography room group.

Table 2.

Outcomes compared 2 groups

	Bedside (N = 26)	Angiography suite (N = 44)	P
Procedure duration (min), mean ± SD	16.04 ± 8.884	13.73 ± 6.428	0.253
Post SOFA	12.5 ± 3.289	9 ± 3.41	< 0.001
ΔSOFA	−0.35 ± 2.297	−0.77 ± 1.764	0.386
Length of Hospital (d)	87.58 ± 77.119	64.05 ± 55.676	0.181
Length of ICU stay (d)	48.42 ± 55.546	34.39 ± 34.408	0.195
Complications (%)	11(42.3%)	12(27.3%)	0.196
Mortality (%)	14 (53.8)	22 (50)	0.756

Δ Changes in SOFA scores before and after treatment

In the bedside group, excluding patients who were lost to follow-up (n = 1), whose catheter was removed shortly after insertion due to patient death (n = 13), or whose catheter was dislodged and not reinserted (n = 7), five patients were available for follow-up. All of these patients showed clinical improvement, and the catheter was subsequently removed in each case. In the angiography suite group, excluding those who were lost to follow-up (n = 8), whose catheter was removed shortly after insertion due to patient death (n = 18), or whose catheter was dislodged and not reinserted (n = 4), fourteen patients were available for follow-up. Among them, five patients had their catheter removed, and nine patients underwent cholecystectomy.

Complication event according two groups

As previously mentioned, there was no significant difference observed between the two groups in terms of complications (P = 0.196).

In the bedside group (Fig. 2), there were 11 episodes of complications affecting 11 patients (42.3%). Five mild complications were noted, primarily due to catheter dislodgement. Reinsertion was attempted in three patients; however, the procedure failed in two cases. No further procedures were performed in the remaining two patients. A drop in hemoglobin by ≥ 1 g/dL was considered a moderate complication and occurred in three patients, with two requiring blood transfusions. Three patients experienced severe complications, specifically catheter dislodgement leading to bile leakage. All three patients required surgical interventions such as open cholecystectomy with explorative laparotomy or laparoscopic cholecystectomy; however, there were no procedure-related deaths.

Fig. 2.

Complications in bedside group

In the angiography suite group (Fig. 3), 14 episodes of complications were noted in 12 patients (27.3%). Minor complications, also primarily due to catheter dislodgement, occurred in eight patients. Four tubes were replaced via tube change or reinsertion. Six moderate complications were observed, with one patient requiring a blood transfusion. There were no severe complications in this group.

Fig. 3.

Complications in angiography suite group

Catheter dislodgement was the most frequent procedure-related complication, affecting a total of 16 out of 70 patients. These patients are often unable to reposition themselves, which increases the risk of pressure sores. As a result, medical staff frequently reposition them, and during this process, accidental dislodgement of the catheter may occur. In addition, some patients with altered mental status have been observed to pull out the catheter themselves. There was no statistically significant difference in incidence between the two groups (8 vs. 8, P = 0.226). Among the 38 patients who underwent the transperitoneal route with trocar technique, 10 tubes dislodged (26.3%), compared to 6 out of 32 tubes placed via the transhepatic route with Seldinger technique (18.8%).

However, there was no significant difference between the two routes (techniques) (P = 0.453). Therefore, catheter dislodgement appeared unrelated to the route or technique of cholecystostomy placement.

Discussion

Acute cholecystitis is a common inflammatory condition of the gallbladder associated with significant morbidity and mortality, and remains one of the leading surgical causes of emergency department visits [9, 10]. Although cholecystectomy—especially the laparoscopic approach—remains the standard of care in operable patients [1, 2, 11, 12], emergency surgery may carry a high mortality risk of up to 30% in elderly or critically ill populations, particularly those in the ICU or with multiple comorbidities [13, 14].

To mitigate surgical risks, conservative management with systemic antibiotics, with or without gallbladder decompression, is recommended as an initial step [15]. Percutaneous cholecystostomy provides a minimally invasive option for biliary decompression, serving either as a bridge to delayed surgery or as definitive management in high-risk patients or those with acute acalculous cholecystitis [16–19].

However, in clinical practice, percutaneous cholecystostomy may not always be feasible. Hemodynamic instability, respiratory failure, or the need for multiple vasopressors often make it unsafe to transport ICU patients to the interventional suite. In addition, diagnosing acute cholecystitis in sedated or ventilated patients is inherently challenging due to suppression of clinical manifestations.

Acalculous cholecystitis, which constitutes 2–14% of all acute cholecystitis cases [20–22], tends to follow an insidious course. Its pathogenesis is multifactorial, associated with prolonged fasting, parenteral nutrition, sepsis, sedation, shock, and vasopressor use [23]. Clinical manifestations are often masked in ICU settings, and sonographic findings—such as gallbladder wall thickening or biliary sludge—can be nonspecific in fasting or sedated patients [22, 24].

These diagnostic and logistical challenges often lead to delays in intervention, contributing to poorer outcomes. In such scenarios, performing percutaneous cholecystostomy at the bedside offers a crucial alternative. The trocar technique, which requires only ultrasound guidance, is particularly suitable for bedside procedures, eliminating the need for fluoroscopy or patient transport.

In our study, we found no significant differences in clinical outcomes or complication rates between patients who underwent cholecystostomy at the bedside versus those treated in the angiography suite. This suggests that, when performed under appropriate conditions, bedside cholecystostomy using the trocar method is not only feasible but also effective. These findings underscore the importance of developing flexible, patient-centered procedural strategies for managing acute cholecystitis in critically ill populations.

Another important consideration in percutaneous cholecystostomy is the access route—transhepatic versus transperitoneal—which may influence the risk of catheter-related complications, particularly bile leakage upon catheter dislodgement or removal. The transhepatic route is generally considered to provide more secure catheter placement due to tract formation through liver parenchyma, which may reduce the risk of free intraperitoneal bile leakage. In contrast, the transperitoneal route traverses the peritoneal cavity and may carry a higher theoretical risk of bile leak.

In our study, catheter dislodgement occurred in 16 patients, making it the most common procedure-related complication. When comparing the bedside and angiography suite groups, there was no statistically significant difference in the incidence of catheter dislodgement. Similarly, although not statistically significant, dislodgement was numerically more frequent in the transperitoneal route (10/38, 26.3%) compared to the transhepatic route (6/32, 18.8%). These findings align with prior literature suggesting a possible trend toward greater catheter stability with the transhepatic approach, although large-scale studies and meta-analyses have not shown consistent statistical significance in terms of bile leak or overall complication rates between the two routes [25, 26].

Further prospective studies are warranted to more definitively evaluate whether the choice of access route influences clinical outcomes, particularly in high-risk ICU populations where catheter stability and complication avoidance are critical.

This study has several limitations, most notably its retrospective design, which may be subject to selection bias and confounding factors that cannot be fully controlled. A prospective randomized controlled trial (RCT) would provide a higher level of evidence to directly compare bedside and angiography suite cholecystostomy. However, conducting an RCT in this clinical context poses significant ethical and logistical challenges, as many critically ill patients are hemodynamically unstable and may not tolerate transport or randomization. Despite these limitations, our findings provide valuable insights into real-world practice and may help inform clinical decision-making for this high-risk patient population.

Conclusion

In critically ill ICU patients with acute cholecystitis who are unsuitable for transport or surgery, bedside percutaneous cholecystostomy using the trocar technique is an effective alternative.

Our study demonstrates comparable outcomes between bedside and angiography suite procedures, supporting the integration of bedside cholecystostomy into clinical practice for ICU patients with severe comorbidities or instability.

Author contributions

Conceptualization, J, K and K; Formal analysis, K, R; Investigation, J and K; Validation, K and K; Writing – original draft, J and K; Writing – review & editing, K & J. All authors read and approved the final version of the manuscript.

Funding

none.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Competing interests

The authors declare no competing interests.

Ethics approval

This study was approved by the Institutional Review Board of Pusan National University Yangsan Hospital (IRB No. 55-2024-138). The requirement for informed consent was waived due to the retrospective nature of the study and use of anonymized data.