Outcomes of percutaneous transluminal biopsy of biliary lesions using a dedicated forceps system

Introduction

Despite the recent advances in diagnostic imaging modalities such as ultrasound (US), computed tomography (CT), and magnetic resonance imaging (MRI) that have notably increased the diagnostic accuracy in the detection and localization of biliary strictures (1), the distinction between benign and malignant biliary obstruction is not always feasible (2,3). Additionally, according to the current European Society for Medical Oncology (ESMO) guidelines, it is specified that histological confirmation is mandatory before any non-surgical treatment such as chemotherapy, radiation therapy, and biliary stenting (4). The importance of tissue acquisition and immunohistochemistry is further highlighted to distinguish different subtypes and promote personalized therapy decisions (4,5). In the same guidelines, the recommended tissue acquisition techniques include endoscopic retrograde cholangiopancreatography (ERCP)-guided core biopsies or brush cytology and endoscopic ultrasound (EUS)-guided fine needle aspiration (FNA). Percutaneous transhepatic biliary drainage (PTBD) is today considered a well-established, non-surgical method of relieving obstructive jaundice, with technical and clinical results comparable, and in certain cases superior, to those achieved by endoscopy (6). Although endoluminal biliary biopsy during PTBD has been first reported almost 40 years ago and several studies have described the safety and efficacy of the method using different forceps sets, the specific technique is neither cited nor recommended in the ESMO guidelines probably due to the fact that both the number of published cases as well as the level of evidence, remain low (4,7–11).

We sought to prospectively investigate the safety and efficacy of percutaneous transluminal biopsy of biliary lesions using a dedicated forceps system.

Material and Methods

This is a prospective, single-center, single arm study, investigating the safety and efficacy of percutaneous transluminal biopsy of biliary lesions during PTBD, using a dedicated, low-profile, transluminal biliary access and biopsy forceps set (Cook Medical, Bloomington, IN, USA). Between September 2014 and January 2017, 29 consecutive patients (17 male (58.6%), 12 female (41.4%); mean age: 60  ±  9 years) suffering from obstructive jaundice (increase of direct bilirubin plasma levels) due to biliary stenosis or occlusion from histologically unidentified lesions requiring PTBD, underwent 30 percutaneous transluminal biopsies during the drainage procedure using the specific biopsy set and were included in the study. All biliary obstructions were previously investigated and confirmed by magnetic resonance cholangiopancreatography (MRCP) (Fig. 1). The study’s primary efficacy endpoint was technical success defined as the successful acquisition of at least one tissue sample which was characterized by the pathologist as sufficient to render diagnosis. The primary safety endpoint was the procedure-related major complications rate. The study’s secondary endpoints were procedure-related minor complications rate, sensitivity, specificity, and diagnostic accuracy for the characterization of malignancy. Complications were classified as major or minor according to the SIR reporting standards for percutaneous transhepatic biliary procedures (12). Sensitivity, specificity, and accuracy were calculated for the biopsy technique. Univariate analysis was performed to identify an association between variables and false-negative results and included patients’ age and gender, location of the obstruction, number of specimens obtained, and type of tumor. Variables that were found to be statistically significant according to univariate analysis were subsequently assessed by multivariate regression analysis to identify possible independent predictors of false-negative results. Statistical analysis was performed with the SPSS statistical software package (version 21.0; IBM, USA). A P value < 0.05 was considered as the threshold of statistical significance.

OPEN IN VIEWER

Fig. 1.

Hilar biliary stricture. A 68-year-old man presented with obstructive jaundice. (a) MRCP showed the presence of hilar stenosis (white arrow) with bilateral intrahepatic bile duct dilatation. (b) Axial T2-weighted fat-saturated image confirmed bile duct dilatation, caused by slightly hyperintense tissue within the bile duct (white arrow).

Procedure

All procedures were performed in the angiography suite under local anesthesia (lidocaine 2%) at the puncture site and conscious sedation using intravenous fentanyl and midazolam just before and during the procedure. PTBD was performed according to the standard interventional radiology technique (12). Endoluminal biopsy technique was performed using the cross and push technique as previously described (11). In brief, following percutaneous transhepatic access using the 22-G needle/6-Fr sheath Nef® percutaneous access set, the biliary obstruction was negotiated using a 5-Fr Torcon NB®, BMC-shaped catheter, and a 0.035-inch Roadrunner® hydrophilic guide wire (Fig. 2a). Following lesion crossing, the Flexor® 7-Fr × 30 cm sheath was positioned within the obstruction, over a 0.035-inch Amplatz® superstiff, support guide wire, positioned within the duodenum. All the above-mentioned materials are included in the Cook biospy forceps set. The superstiff guide wire was left for safety and subsequently the biopsy forceps device was inserted by the wire, through the sheath, and was pushed and advanced open within the lesion under fluoroscopic guidance, using the sheath for support (Fig. 2b and 2c). Up to five samples were taken from each lesion, which were fixed with formalin and sent to the pathology department for analysis. Following biopsy an internal–external biliary draining catheter was positioned over the superstiff guide wire (Fig. 2d).

OPEN IN VIEWER

Fig. 2.

Biopsy procedure. Same patient as in Fig. 1. (a) The hilar biliary obstruction is crossed using a 5-Fr catheter, with a hydrophilic guide wire. A 7-Fr sheath is then positioned within the lesion, over a superstiff, wire. (b, c) The biopsy forceps device is inserted by the wire and is pushed and advanced within the tissue. (d) An internal–external 8.5-Fr biliary draining catheter is positioned.

Results

The median sample number obtained was three (range = 3–5). In total, 36% of the cases (11/30) were proximal and 30% (9/30) were mid CBD obstructions. Tissue samples allowed histological diagnosis in 27/30 cases, resulting in a technical success rate of 93.3%. In three cases (6.7%), the sample was characterized by the pathologist as non-diagnostic as necrotic or fibrinous material was provided: one case suspicious for pancreatic cancer and two cases were cholangiocarcinoma. In one case, biopsy was successfully repeated and had the diagnosis of pancreatic cancer. The histological diagnosis included cholangiocarcinoma in 16 cases (53.3%), colorectal metastasis in three cases (10.0%), pancreatic adenocarcinoma in three cases (10.0%), and inflammation in five cases (16.6%). There were two false-negative cases of inflammation proven to be cholangiocarcinoma. These two cases were one stricture of the hilum and one distal common bile duct lesion. Patients’ demographics and outcomes are analytically reported in Table 1. Percutaneous, endoluminal biopsy using then specific access and biopsy forceps set resulted in a sensitivity value of 91.67% and specificity value of 100%, while accuracy was 92.59%. In three cases, percutaneous endoluminal posed the diagnosis of pancreatic cancer (two cases) and inflammation (one case) of the distal common bile duct, following inadequate tissue sampling from ERCP‐guided transpapillary forceps biopsy. No major complications were noted. There were four cases of hemobilia (13%) manifested as right quadrant pain and blood drained from the catheter, without significant hemoglobin drop (<2 units). In all four cases, transfusion, further treatment, or prolonged hospitalization was not required as hemobilia auto-resolved within 48 h. According to univariate analysis, no factors were associated with false-negative results and therefore multivariate analysis was not performed.

OPEN IN VIEWER

Table 1.

Patients demographics and outcomes.

Patient	Gender	Age (years)	Final diagnosis	Insufficient sample	Location of stenosis	Complication	FN diagnosis
1	Male	56	PC	Yes	Distal CBD	–	–
1	Male	56	PC	–	Distal CBD	–	–
2	Female	60	PC	–	Distal CBD	–	–
3	Female	65	IFL	–	Proximal CBD	Hemobilia	Yes (CC)
4	Male	55	IFL	–	Distal CBD	–	–
5	Male	54	IFL	–	Distal CBD	–	–
6	Male	70	CC	–	Mid CBD	Hemobilia	–
7	Male	75	CC	–	Mid CBD	–	–
8	Male	68	CC	–	Mid CBD	–	–
9	Male	65	CC	–	Mid CBD	–	–
10	Female	57	CC	–	Mid CBD	–	–
11	Female	66	CC	–	Mid CBD	–	–
12	Male	50	CC	Yes	Proximal CBD	–	–
13	Male	50	CC	–	Proximal CBD	–	–
14	Male	67	CC	–	Proximal CBD	Hemobilia	–
15	Female	54	CC	–	Proximal CBD	–	–
16	Female	56	CC	–	Proximal CBD	–	–
17	Female	45	CC	–	Proximal CBD	–	–
18	Male	67	CC	Yes	Proximal CBD	–	–
19	Female	69	CRM	–	Proximal CBD	–	–
20	Female	82	CRM	–	Mid CBD	–	–
21	Female	59	CC	–	Distal CBD	–	–
22	Female	60	CC	–	Distal CBD	–	–
23	Female	62	CC	–	Mid CBD	–	–
24	Male	56	CC	–	Mid CBD	–	–
25	Male	50	IFL	–	Distal CBD	–	Yes (CC)
26	Male	62	PC	–	Distal CBD	–	–
27	Male	54	IFL	–	Proximal CBD	–	–
28	Male	54	CRM	–	Proximal CBD	–	–
29	Female	78	CC	–	Mid CBD	Hemobilia	–

FN, false negative; ERCP, endoscopic retrograde cholangiopancreatography; PC, pancreatic cancer; CC, cholangiocarcinoma; CRM, colorectal metastasis; IFL, inflammation; CBD, common bile duct.

Discussion

PTBD is currently the treatment of choice in patients with biliary steno-occlusion after the failure of ERCP or in cases of hilar strictures (4). Percutaneous forceps biopsy of the bile duct was first reported in 1980 and, since that time, several studies have been conducted demonstrating its higher sensitivity when compared to bile cytology and fine needle aspiration biopsy (3,5,8–10). This is the first prospective study investigating the specific percutaneous access and biopsy forceps set and its diagnostic value was confirmed by the excellent 91.67% sensitivity, 100% specificity, and 92.59% accuracy rates. These are among the highest of published outcomes. Specifically, according to recent studies, the reported sensitivity of endobiliary forceps biopsy was in the range of 78–93% (13). Park et al., in a large retrospective analysis of 271 percutaneous endobiliary biopsies, using a 5.4-Fr flexible biopsy forceps system (Cordis, Miami Lakes, FL, USA), reported lower sensitivity (77.2%) and diagnostic accuracy (78.9%) rates due to the superior rate of false-negative results (21.0%; (57/271 cases) compared to the 6.6% (2/30 cases) noted in the current study. On the other hand, specificity rates were equal to this study (100%), while complication rates were similarly low (13). In the same study, multivariate analysis demonstrated that periampullary lesions, intrahepatic bile duct obstruction, and metastatic disease were independent predictors of false-negative diagnosis. In the herein presented study, no factors were associated with false-negative results. However, the sample size (30 cases) should be considered insufficient to support a statistically meaningful multivariate analysis. The two false-negative cases of inflammation, which involved one hilar and one distal CBD stricture, have been finally proven to be cholangiocarcinomas based on the radiological and clinical evolution of the disease.

According to the authors’ opinion, considering that evidence demonstrating the safety and substantial diagnostic value of percutaneous forceps biopsy are building up, this procedure should be considered a first-line diagnostic tool in selected cases, such as hilar strictures lacking histological identification and requiring PTCB. This is also supported by data demonstrating the inferior sensitivity of endoscopic ultrasonography–fine needle aspiration (EUS-FNA) in hilar strictures (11). In the era of personalized drug therapy, in which immunohistochemistry is required for the optimization of therapy, safe and effective, percutaneous endobiliary tissue sampling should be included in updated guidelines and diagnostic protocols (5,12). The authors have applied the cross and push technique using the sheath to push and advance the open biopsy forceps into the stricture, which has been reported to provide enhanced outcomes and, indeed, sensitivity (93.3%) and specificity (100%) outcomes were similarly high (11).

In three cases, the sample was not sufficient for the pathologist. Notably, these cases have been performed in the early part of the learning curve and the authors speculate that insufficient sampling could be related to the lack of experience. Nevertheless, the procedure does not entail severe technical difficulties and a steep learning curve of few cases should be expected. Interestingly, in three cases in which ERCP failed to provide adequate sampling, percutaneous biopsy posed the diagnosis. This demonstrates that percutaneous transluminal biopsy offers an efficient investigative option and should be considered in the diagnostic algorithm of bile duct obstructions following a failed endoscopic attempt. Specifically, the sensitivity of ERCP‐based transpapillary forceps biopsy cannot be considered satisfactory and is in the range of 33–89%, while novel methods such as EUS-FNA or intraductal ultrasonography (IDUS)-FNA, demonstrate sensitivity values in the range of 43–86%, following prior non-diagnostic ERCP (14). Moreover, in a recent study investigating proximal biliary obstructions, percutaneous endobiliary forceps biopsy yield similar outcomes compared with EUS-FNA and the authors concluded that it should be considered as the first-line tissue-sampling strategy in cases in which PTBD is indicated (15).

Limitations of this study include the small number of patients investigated. A larger number of patients would have revealed more valid technical success and complication rates and would have provided a robust statistical analysis for the identification of factors associated with misdiagnosis. The lack of a control group did not allow a direct comparison with endoscopic modalities. Finally, the single-center design does not warrant the generalizability of the method.

In conclusion, percutaneous endobiliary forceps biopsy using the 5.2-Fr biopsy set (Cook Medical), yield very high sensitivity, specificity, and accuracy results, with minimum complications and should therefore be considered as a valid diagnostic option, especially during PTBD without previous histologic identification. Larger, multicenter trials are required to determine the role of percutaneous endobiliary forceps biopsy in the diagnostic algorithm of biliary strictures.