Endoscopic biliary drainage techniques for bilirubin normalization in distal malignant biliary obstruction: A comprehensive review

Abstract

Endoscopic retrograde cholangiopancreatography (ERCP) is currently recommended as the gold standard treatment modality for distal malignant biliary obstruction (DMBO), with percutaneous transhepatic biliary drainage being historically regarded as the second-line strategy in case of ERCP failure. Following the recent advent of therapeutic endoscopic ultrasound, various endoscopic ultrasound-guided biliary drainage techniques have been advocated as the preferred second-line options, as they are associated with fewer adverse events and morbidity compared to percutaneous transhepatic biliary drainage. Moreover, they have also been proposed as promising first-line alternatives to ERCP. However, the capability of both the historical and emerging biliary drainage modalities to normalize total bilirubin in patients with DMBO is underreported and evidence is still lacking, especially regarding the emerging ones. Notably, bilirubin normalization is regarded as mandatory before starting chemotherapy, thus representing a crucial outcome of any biliary drainage method for DMBO. Our study summarizes and discusses the current evidence regarding the effectiveness of currently available endoscopic biliary drainage techniques for bilirubin normalization in DMBO.

Key Words: Endoscopic retrograde cholangiopancreatography; Endoscopic ultrasound-guided biliary drainage; Endoscopic ultrasound-guided choledochoduodenostomy; Endoscopic ultrasound-guided gallbladder drainage; Biliary drainage; Distal malignant biliary obstruction; Malignant biliary obstruction; Bilirubin normalization

Core Tip: Endoscopic retrograde cholangiopancreatography remains the gold standard for distal malignant biliary obstruction (DMBO), with endoscopic ultrasound-guided biliary drainage emerging as a promising alternative. However, evidence regarding bilirubin normalization rates across these modalities is lacking. Our study reviews the available evidence to evaluate the efficacy of endoscopic biliary drainage techniques in normalizing bilirubin levels in patients with DMBO.

INTRODUCTION

Distal malignant biliary obstruction (DMBO) is a severe complication associated with advanced bilio-pancreatic and gastrointestinal malignancies. The leading cause of DMBO is pancreatic cancer, followed by cholangiocarcinoma[1,2]. Less common etiologies include ampullary cancer, lymphoma, metastatic disease, and gallbladder cancer[1,2]. DMBO is associated with significant morbidity and mortality, along with impaired quality of life and prognosis[3,4]. Its prompt diagnosis and treatment are crucial not only to alleviate symptoms and improve the quality of life and survival but also to allow the administration of systemic oncological therapies[3-6].

The therapeutic armamentarium for DMBO currently includes endoscopic retrograde cholangiopancreatography (ERCP), percutaneous transhepatic biliary drainage (PTBD), and various endoscopic ultrasound-guided biliary drainage (EUS-BD) techniques. ERCP is currently recommended as the first-line treatment modality for DMBO[7,8]. In case of ERCP failure, PTBD has historically represented the preferred second-line modality. However, following the advent of therapeutic endoscopic ultrasound, EUS-BD has recently been recommended as the second-line modality of choice, due to the lower associated morbidity[7-10]. Moreover, various EUS-BD techniques have been investigated and proposed as promising first-line alternatives to ERCP[11-14]. Nevertheless, evidence regarding the effectiveness of available biliary drainage (BD) modalities for bilirubin normalization in DMBO is limited, particularly for the emerging modalities. Indeed, bilirubin normalization is underreported in favor of clinical success defined by a bilirubin reduction of > 50% within 2-4 weeks and/or not specifically reported within composite endpoints[11-13,15].

Bilirubin normalization is a crucial oncological outcome for any BD technique in DMBO. Indeed, most of the chemotherapeutic agents (i.e. gemcitabine, doxorubicin, irinotecan, and the taxanes) require intact mechanisms of bilirubin excretion and BD to prevent toxicity[16,17]. Given the higher risk of toxicity, a total bilirubin (TB) level < 1.5 times the upper normal level (UNL) is mandatory before starting standard neoadjuvant or palliative chemotherapy and for its continuation[18-20]. In the case of TB < 5 UNL, a dose reduction and/or a less toxic regimen may be considered. Conversely, a TB level > 5 UNL represents a contraindication to systemic anticancer treatment[21,22]. Thus, bilirubin normalization is essential to establish the proper allocation of any BD modality within the management algorithm of DMBO.

Our narrative review summarizes and discusses the current available evidence concerning the effectiveness of endoscopic BD modalities for bilirubin normalization in DMBO.

LITERATURE SEARCH

A comprehensive search of PubMed/MEDLINE, EMBASE, and Google Scholar databases was performed through March 2026 to identify relevant studies evaluating the effectiveness of available endoscopic BD modalities for bilirubin normalization in DMBO, either as a “rescue” or a first-line approach. The evaluated endoscopic BD modalities included ERCP, endoscopic ultrasound-guided choledochoduodenostomy (EUS-CDS), endoscopic ultrasound-guided gallbladder drainage (EUS-GBD), and endoscopic ultrasound-guided hepaticogastrostomy (EUS-HGS).

Studies omitting data on the bilirubin normalization rate were excluded. Studies addressing bilirubin normalization within a composite outcome without specifically providing data on its rate were also excluded. Finally, studies reporting only mean/median post-drainage TB level were excluded. Given a standardized definition of bilirubin normalization is currently lacking, a priori exclusion was not contemplated based on its definition.

The search terms included “distal malignant biliary obstruction”, “DMBO”, “malignant distal biliary obstruction”, “MDBO”, “bilirubin normalization”, “bilirubin”, “endoscopic retrograde cholangiopancreatography”, “ERCP”, “endoscopic ultrasound-guided choledochoduodenostomy”, “EUS-CDS”, “endoscopic ultrasound-guided gallbladder drainage”, “EUS-GBD”, “endoscopic ultrasound-guided hepaticogastrostomy”, and “EUS-HGS” in various combinations, using the Boolean operators AND, OR, and NOT. The search was restricted to human studies and English-language articles. Abstracts, case reports/series (< 10 cases), reviews, position papers, editorials, and book chapters were excluded. References from the included studies and pertinent reviews were carefully hand-searched for potential inclusion.

EFFICACY OF ENDOSCOPIC BD MODALITIES FOR BILIRUBIN NORMALIZATION IN DMBO

Evidence

A total of nine studies specifically reported data on the bilirubin normalization rate following endoscopic BD for DMBO and were included in our final analysis[23-31]. All but one prospective study[29] were retrospective in nature[23-28,30,31]. Seven of the included studies evaluated ERCP[23-29], while the remaining two evaluated EUS-GBD[30,31]. No studies assessing EUS-CDS or EUS-HGS were eligible for inclusion. The main features of the included studies are summarized in Tables 1 and 2.

Table 1 Characteristics of the included studies.

Ref.	Study design	Country	Study type	Enrollment period
Kahaleh et al[23]	Retrospective	United States	Monocentric	2001-2003
Weston et al[24]	Retrospective	United States	Monocentric	2002-2005
Weston et al[25]	Retrospective	United States	Monocentric	2006-2007
Cabral et al[26]	Retrospective	Brazil	Monocentric	2015
Pausawasdi et al[27]	Retrospective	Thailand	Monocentric	2007-2017
Mikalsen et al[28]	Retrospective	Norway	Monocentric	2015-2021
Naidu et al[29]	Prospective	India	Monocentric	2023-2024
Martínez-Moreno et al[30]	Retrospective	Spain	Multicentric	2016-2024
Chieng et al[31]	Retrospective	New Zealand	Multicentric	2017-2023

Table 2 Summary of studies reporting on the effectiveness of endoscopic biliary drainage for bilirubin normalization in distal malignant biliary obstruction.

Ref.	Number	Cancer type (n)	BD modality	Stent type (n)	Intention	Bilirubin normalization (%)	Bilirubin normalization definition	Time to bilirubin normalization
Kahaleh et al[23]	80	Pancreas (62). Cholangiocarcinoma (6). Ampullary (3). Metastasis (9)	ERCP	Covered SEMS 10 mm (80)	First-line	73	TB ≤ 1.2 mg/dL at 30 days
Weston et al[24]	156	Pancreas (63). Cholangiocarcinoma (16). Gallbladder (7). Metastasis (45). Other (25)	ERCP	PS 7/8.5/10 Fr (75). SEMS 6/8/10 mm (81)	First-line	73.2	TB ≤ 2 mg/dL within 6 weeks	Pre-stent TB < 5 mg/dL: 2.6 weeks. Pre-stent TB 5-10 mg/dL: 2.7 weeks. Pre-stent TB > 10 mg/dL: 5.6 weeks
Weston et al[25]	177	Pancreas (125). Cholangiocarcinoma (16). Ampullary (8). Metastasis (18). Other (10)	ERCP	Uncovered SEMS 8/10 mm (177)	First-line	Total: 59.9. N-SEMS: 65.4. SS-SEMS: 58.3	TB ≤ 2 mg/dL
Cabral et al[26]	58	Pancreas (44). Neuroendocrine tumor (3). Ampullary (2). Metastasis (9)	ERCP	PS 8.5/10 Fr (23). SEMS 10 mm (35)	First-line	Total: 62.1. SPS: 41.2. MPS: 83.3. SEMS: 68.6	TB < 2 mg/dL	SPS: 16.6 days. MPS: 38.4 days. SEMS: 19.5 days
Pausawasdi et al[27]	511: DC: 383; VC: 128 (295 DMBO)	Intrahepatic cholangiocarcinoma (43). Hilar cholangiocarcinoma (138). Extrahepatic cholangiocarcinoma (60). Pancreas (160). Gallbladder (45). Ampullary (19)	ERCP	PS (173). Uncovered SEMS (293). Covered SEMS (9). Partially-covered SEMS (4). Two SEMS (19). Two PS (10). One SEMS and one PS (3)	First-line	DC: 31.2. VC: 26.9	TB ≤ 1.2 mg/dL within 6 weeks
Mikalsen et al[28]	596 (436 DMBO)	Pancreas (308). Cholangiocarcinoma (95). Ampullary (21). Gallbladder (12). Metastasis (117). Other (43)	ERCP	PS (198). Covered SEMS (272). SEMS + PS (4). None (122)	First-line	60.2 (284/472)	TB < 4.68 mg/dL within 30 days
Naidu et al[29]	101 (33 DMBO)		ERCP	PS (94). SEMS (7)	First-line	Within 7 days: 21.3. Within 15 days: 60.6. Within 30 days: 78.7	TB < 3 mg/dL within 7 days, 15 days, and 30 days
Martínez-Moreno et al[30]	96		EUS-GBD	EC-LAMS 6/8/10/15 mm ± coaxial PS	After ERCP and/or EUS-BD failure	65.6	TB < 3 mg/dL	15 (7-27)
Chieng et al[31]	26	Pancreas (20). Cholangiocarcinoma (2). Ampullary (1). Other (3)	EUS-GBD	EC-LAMS 8/10/15 mm ± coaxial PS	After ERCP and/or EUS-BD failure	19.2	Full bilirubin normalization over the follow-up

BD: Biliary drainage; DC: Derivation cohort; DMBO: Distal malignant biliary obstruction; EC-LAMS: Electrocautery-enhanced lumen apposing metal stent; ERCP: Endoscopic retrograde cholangiopancreatography; EUS-BD: Endoscopic ultrasound-guided biliary drainage; EUS-GBD: Endoscopic ultrasound-guided gallbladder drainage; MPS: Multiple plastic stents; N-SEMS: Nitinol self-expandable metal stent; PS: Plastic stent; SEMS: Self-expandable metal stent; SPS: Single plastic stent; SS-SEMS: Stainless steel self-expandable metal stent; TB: Total bilirubin; VC: Validation cohort.

Bilirubin normalization

ERCP: In 2005, Kahaleh et al[23] showed retrospectively a normalization of bilirubin (≤ 1.2 mg/dL) within 30 days in 73 of 80 included patients with unresectable DMBO following the insertion of a biliary fully-covered self-expandable metal stent (SEMS).

Subsequently, Weston et al[24] specifically evaluated the bilirubin normalization rate following endoscopic transpapillary BD. Ninety-three of 156 included patients achieved a post-drainage TB level ≤ 2 mg/dL, whereas 29 patients failed because of stent malfunction, and 34 patients failed because of inadequate follow-up. Focusing on DMBO, the bilirubin normalization was achieved in 73.2% of cases. The time to bilirubin normalization was also precisely assessed. Notably, the time required for 80% of patients to achieve normalization was more than doubled in those with pre-drainage bilirubin levels > 10 mg/dL (5.6 weeks) compared with those with pre-drainage bilirubin levels < 10 mg/dL (2.7 weeks). Moreover, high preoperative bilirubin level, stricture outside the common bile duct, diffuse liver metastasis, and international normalized ratio ≥ 1.5 (suggestive of either prolonged cholestasis and/or underlying hepatic dysfunction) showed a significant negative association with bilirubin normalization. Conversely, the cancer type, recent chemotherapy, stent type/diameter, and sphincterotomy showed no significant association[24]. In 2010, the same group published a retrospective study comparing stainless steel with biliary uncovered nitinol SEMS. The reported bilirubin normalization rate (≤ 2 mg/dL) was 65.4% for nitinol SEMS and 58.3% for stainless steel ones[25].

In a retrospective study from Brazil, 36/58 of the included patients with unresectable DMBO were finally able to achieve a post-drainage TB level < 2 mg/dL. In detail, the success rates in achieving TB < 2 mg/dL for SEMS, multiple plastic stents (MPS), and single plastic stents (SPS) were 68.5% (24/35), 83.3% (5/6), and 41.1% (7/17), respectively. Although there was mainly divergence between SPS and MPS and SEMS, this difference was not found to be significant. Furthermore, the mean time in days to bilirubin normalization in the SPS, MPS, and SEMS groups was 6.6 days, 38.4 days, and 19.5 days, respectively[26].

In 2022, a large retrospective study by Pausawasdi et al[27] reported a bilirubin normalization rate, defined by TB regression below 1.2 mg/dL within 6 weeks after stenting, of 31.2% in the derivation cohort and 26.9% in the validation one. To be noted, this study included both distal and proximal malignant biliary obstructions, with cholangiocarcinoma being the most common etiology and accounting for about half of the included cases. Thus, the reported suboptimal bilirubin normalization rate may have been affected by the etiology and the level of biliary obstruction, and by the very low cut-off adopted for the definition of bilirubin normalization, which was lower than the acceptable TB level needed for chemotherapy administration. Nonetheless, 66.2% of cases with an early improvement in bilirubin levels failed to achieve the normalization of bilirubin at 6 weeks. Thus, a TB regression exceeding 50% within 2 weeks could not guarantee bilirubin normalization. Notably, univariate and multivariate analyses of baseline variables, extrahepatic biliary obstruction, pre-endoscopic TB levels, showed that the type of biliary stent were independent variables for predicting bilirubin normalization. Intriguingly, the authors developed a pre-endoscopic scoring system comprising biliary obstruction level, liver biochemistry, and type of stent for the prediction of bilirubin normalization[27]. Subsequently, Mikalsen et al[28] reported retrospectively that 60.2% of the included patients undergoing ERCP for malignant biliary obstruction achieved TB normalization. Worth mentioning, it was defined by bilirubin levels < 4.68 mg/dL at post-drainage day 30[28].

Finally, a recent single-center prospective study showed a bilirubin normalization rate, defined as bilirubin < 3 mg/dL by day 15 and 30, of 70.3% and 85.1%, respectively, despite the prevalent adoption of biliary plastic stenting (94%). This study included 101 patients affected by benign or malignant extrahepatic biliary obstruction (EHBO). In further details, among 61 enrolled patients with malignant EHBO, including 33 cases of DMBO, the observed bilirubin normalization rates by day 15 and 30 were 60.6% and 78.7%, respectively. Based on the baseline bilirubin level, a significantly higher proportion of patients with baseline bilirubin < 10 mg/dL achieved bilirubin normalization by day 15 (91.7% vs 56.9%) and 30 (97.2% vs 80%). Worth mentioning, baseline bilirubin and alkaline phosphatase levels were found to be independent predictors of obtaining bilirubin < 3 mg/dL by day 15, while failure to achieve this outcome was shown to be an independent predictor of mortality[29].

EUS-GBD: Martínez-Moreno et al[30], in their retrospective multicenter study, reported a bilirubin normalization rate (TB reduction below 3 mg/dL) of 66.3% within a median time of 15 days following EUS-GBD. Finally, Chieng et al[31] showed retrospectively that, despite 100% technical and clinical success, “rescue” EUS-GBD was able to fully normalize TB in only 5 of 26 included cases (19.2%). Notably, the time settled by the authors to achieve bilirubin normalization was very long, lasting from the time of intervention until the patient’s death, with a reported median survival of 103 (38-192) days. Conversely, bilirubin normalization at 2-4 weeks post-drainage was omitted[31].

DISCUSSION

From a strict oncological point of view, a key outcome of BD in DMBO is a rapid and sustained bilirubin normalization, which is mandatory before starting the full-dose standard chemotherapy and for its continuation[13-17]. Moreover, bilirubin normalization not only provides patients with DMBO the opportunity for further treatment in the form of systemic chemotherapy, but it also improves chemotherapy tolerance, symptoms of jaundice, quality of life, performance status, and survival[32-36].

However, despite its crucial role in the multidisciplinary management of patients with DMBO, our review ruled out that evidence on bilirubin normalization following endoscopic BD is currently limited, especially with regard to emerging EUS-BD modalities. Indeed, while seven studies about ERCP were found eligible for inclusion in our study[23-29], only two studies concerning EUS-GBD were finally retrieved[30,31], with no studies evaluating EUS-CDS and/or EUS-HGS specifically addressing the bilirubin normalization rate. Of note, a multicenter randomized clinical trial comparing EUS-CDS or EUS-HGS and ERCP for the primary palliation of unresectable DMBO reported the mean TB level at week 4 post-drainage (1.5 ± 2.9 for ERCP vs 1.5 ± 2.4 for EUS-BD)[37]. However, the bilirubin normalization rate was not specifically reported, and the post-drainage mean TB level may not be considered synonymous with the bilirubin normalization. Similarly, Tarantino et al[38] reported the mean TB level at 2 weeks following “rescue” EUS-CDS, while omitting the bilirubin normalization rate.

Excluding the study by Pausawasdi et al[27], which included a significant percentage of hilar malignant obstructions (37.3% in the derivation cohort and 39.1% in the validation cohort), with cholangiocarcinoma as the most common etiology, and did not specifically report the bilirubin normalization rate in DMBO only, we found a bilirubin normalization rate ranging from 59.9% to 78.7% for ERCP. To be noted, Mikalsen et al[28] included distal (73.2%), perihilar (18.6%), and intrahepatic (1.7%) biliary obstructions, reporting only the overall bilirubin normalization rate. Similarly, Naidu et al[29] reported the bilirubin normalization rate among patients affected by malignant EHBO, including both distal (54.1%) and perihilar (45.9%) biliary obstruction.

Conversely, the reported bilirubin normalization rate ranged widely from 65.6% to 19.2% for EUS-GBD. It is noteworthy that PTBD, historically the preferred second-line modality, has been associated with higher bilirubin normalization rates compared to EUS-GBD. Indeed, despite being addressed in studies not limited to DMBO, the bilirubin normalization rate (TB level ≤ 2-3 mg/dL) reported for PTBD in malignant biliary obstruction ranged from 57% up to 70%[39-41]. Conversely, a large retrospective study by Thornton et al[42] reported a bilirubin normalization rate of 31% for PTBD in malignant biliary obstruction. However, a very strict definition of bilirubin normalization (TB level ≤ 1 mg/dL by 100 days) was adopted by the authors[42].

The definition of bilirubin normalization differed consistently among the included studies[23-31]. However, a standardized definition of bilirubin normalization is currently lacking. In our opinion, although a post-drainage TB level ≤ 2 mg/dL is finally desirable, bilirubin normalization defined by TB ≤ 3 mg/dL within 2-6 weeks seems to be reasonable and clinically appropriate in the DMBO setting. Indeed, the kinetics of bilirubin normalization vary widely depending on the underlying DMBO etiology, baseline bilirubin burden, duration of biliary obstruction, presence of cholangitis or sepsis, hepatic parenchymal compromise, preexisting known or occult chronic liver disease, and the completeness of drainage achieved[24,27,29,43].

We believe that the bilirubin normalization defined by TB ≤ 3 mg/dL within 2-6 weeks should be systematically addressed in future trials, in order to better clarify the efficacy of the currently available BD modalities for DMBO patients. The current lack of evidence regarding the effectiveness of endoscopic BD, especially EUS-BD, for bilirubin normalization in DMBO is a critical issue. This is particularly relevant as EUS-BD techniques have been recommended as the preferred “rescue” strategies over PTBD and also proposed as promising first-line alternatives to ERCP. Nevertheless, according to currently available evidence, concerns still exist about the effectiveness of EUS-GBD for bilirubin normalization in DMBO[44-46]. Thus, this research gap needs to be properly addressed.

According to the European Society of Gastrointestinal Endoscopy guidelines, clinical success is defined by a TB regression of 50%-75% within 2-4 weeks[10]. Conversely, a ≥ 50% reduction or normalization of TB within 2 weeks is recommended by the latest Tokyo criteria[47]. Anyway, bilirubin normalization is frequently omitted or not specifically reported when included in the clinical success composite endpoint, as per Tokyo criteria[47]. Nonetheless, clinical success, defined as TB reduction to below 50% of baseline at 2-4 weeks, is not synonymous with bilirubin normalization. Indeed, in many cases, this TB reduction may not be sufficient to initiate full-dose standard chemotherapy. The main limitations of our study included its narrative nature, the prevalent retrospective design of the analyzed studies[23-28,30,31], the inhomogeneous definition of bilirubin normalization, and the inclusion of three studies that reported the overall bilirubin normalization rate in malignant biliary obstruction without providing data restricted to DMBO only[27-29].

FUTURE PERSPECTIVES

Large-scale randomized controlled trials prioritizing the bilirubin normalization rate as the primary endpoint are warranted to rigorously compare ERCP, the current standard of care, with emerging BD techniques. Furthermore, long-term follow-up studies are advised to address the impact of different BD techniques and their associated bilirubin normalization rate on patients’ survival and quality of life. Finally, a multidisciplinary consensus primarily involving the oncologist on a standardized definition of bilirubin normalization should be encouraged. This is critical for ensuring the reproducibility and comparability of clinical outcomes.

CONCLUSION

The bilirubin normalization rate is a crucial oncological outcome for any BD modality in DMBO. Its careful, systematic, and standardized assessment should be deeply encouraged in future trials, in order to better define the “oncological” efficacy of both the standard of care and the emerging BD modalities, and their proper allocation within the management algorithm for DMBO.

ACKNOWLEDGEMENTS

We are grateful to Velia De Magistris for English editing.