Optimizing Distal Biliary Stricture Diagnosis: Endoscopic Ultrasonography for Undetected Masses on Cross-Sectional Imaging

To prevent complications associated with ERCP, magnetic resonance cholangiopancreatography [MRCP] has become a diagnostic alternative to ERCP[5]. Because endoscopic ultrasound (EUS) is now the preferred method for detecting undetected small pancreatic tumors using other imaging techniques and lacks the risk of complications associated with ERCP, it is necessary to explore further its effectiveness in distinguishing between malignant and benign biliary strictures[6]. Endoscopic ultrasonography (EUS) allows for detecting an undetected mass lesion using other imaging techniques and provides detailed imaging of the stricture morphology. Furthermore, it aids in staging by evaluating the vascular involvement and regional lymphadenopathy. In addition, EUS-guided fine-needle aspiration (EUS-FNA) can be employed for tissue acquisition[7].

Results

Table 3. Laboratory findings of the studied patients [n = 36].

Endoscopic ultrasound examination

Table 4. EUS characteristics of malignant and benign strictures.

pancreatitis; PAD: periampullary diverticulum

Histopathological examination results for the 20 patients [74.1%] diagnosed as malignant strictures by EUS indicated 10 cases with ampullary adenocarcinoma, 3 with pancreatic adenocarcinoma, and 5 with cholangiocarcinoma (distal CBD mass), while 2 cases from the 12 ampullary masses [initially diagnosed as malignant by EUS] were proved to be negative for malignancy on biopsy and were diagnosed post-inflammatory fibrosis. However, in the seven patients (25.9%) diagnosed with benign strictures by EUS (5 as post-inflammatory fibrosis and 2 as autoimmune pancreatitis), histopathological findings revealed ampullary adenocarcinoma in two cases from the 5 cases initially diagnosed as benign post-inflammatory fibrosis by EUS. The remaining five patients were confirmed to have benign post-inflammatory fibrosis strictures [three patients] and autoimmune pancreatitis (AIP) (two patients). During the follow-up period, ten benign cases remained unchanged, while three cases of chronic pancreatitis (CP) developed a malignant pancreatic head mass after six months; one was detected by CT, and two were identified by EUS. The final diagnosis of the studied patients revealed 23 malignant strictures (12 ampullary adenocarcinomas, 6 pancreatic adenocarcinomas, 5 cholangiocarcinomas) and 13 benign strictures (5 post-inflammatory fibrosis, 5 CP, 2 AIP, and 1 large periampullary diverticulum) (Table 6)

Final diagnosis of the studied patients revealed that, there were 23 malignant strictures (12 ampullary adenocarcinoma, 6 pancreatic adenocarcinoma, and 5 cholangiocarcinoma, and 13 benign stricture (5 post inflammatory fibrosis, 5 chronic pancreatitis, 2 AIP, and 1 large periampullary diverticulum) (Table 8).

The CEA at a cutoff value of ≥ 55 demonstrates excellent discrimination between malignant and benign DBS without a mass, with a sensitivity, specificity, and accuracy of 100.0%, 76.9%, and 91.7%, respectively. At a cutoff value of ≥ 82.5, the CEA achieves sensitivity, specificity, and accuracy of 95.7%, 84.6%, and 91.7%, respectively. This indicates that a CEA value ≥ 82.5 is optimal for diagnosing DBS without a mass, highlighting the effectiveness of the CEA tumor marker in distinguishing between malignant and benign cases of DBS without a mass (Figure 4).

The investigation into DBS poses a substantial challenge in clinical practice, particularly when the absence of a detectable mass complicates the distinction between benign and malignant etiologies. Distinguishing between these conditions is pivotal for guiding appropriate therapeutic interventions and optimizing patient outcomes. While conventional diagnostic methods such as ERCP and MRCP have been employed, their limitations in cases without an identified mass underscore the need for alternative approaches. Endoscopic ultrasonography (EUS), as a cutting-edge imaging modality, offers a promising avenue for improved diagnostic accuracy in DBS cases where conventional methods fall short. In this discussion, we delve into the findings of our study, exploring the efficacy of EUS in delineating the nature of DBS without a discernible mass, considering both diagnostic accuracy and the correlation with clinical and laboratory parameters. The nuanced understanding gained from this investigation may contribute to refining the diagnostic algorithm for patients facing the diagnostic conundrum of distal biliary strictures without an evident mass.

The demographics of the 36 participants in our study revealed a male predominance of 55.6%, with an average age of 56.6 years, in line with findings reported by Oppong et al.[12]. Malignant strictures were more prevalent in males, constituting 60.9% of cases with a mean age of 57.5 years. Jaundice was the most common presenting symptom, observed in 83.3% of participants, consistent with Saifuku et al.'s observations [13]. Noteworthy differences emerged when comparing patients with malignant and benign biliary strictures. Those with malignant strictures exhibited significantly elevated liver enzymes (ALT, AST, and ALP) and bilirubin concentrations compared to their benign counterparts. This elevation can be attributed to the complete and prolonged obstruction seen in malignant cases, potentially due to hepatic metastases or infiltration. Furthermore, patients with extra-hepatic malignant obstruction displayed more direct hyperbilirubinemia compared to those with extra-hepatic benign obstruction[14].

In discrimination between malignant and benign strictures, we revealed that the sensitivity, specificity, and accuracy of both CEA at a cutoff value of  ≥  82.5 ng/mL and CA 19-9 at a cutoff value of ≥ 125 ng/mL were 95.7%, 84.6%, and 91.7%, respectively. Both CEA and CA 19-9 may show an increase in patients with malignancies such as cancer head of the pancreas, cholangiocarcinoma, and ampullary carcinoma[15]. Moreover, Tsukada et al.[16] set the cutoff level for CA 19-9 as 37 U/mL and demonstrated a nonsignificant difference between malignant and benign strictures. This is owing to the CA 19-9 level increases in the case of cholestasis, leading to false-positive cases, which caused a decrease in CA 19-9 specificity to 76.9% when using a cutoff value of 75 ng/mL. Our results manifested that 23 patients had malignant strictures (63.9%) and 13 had benign strictures (36.1%), which aligns with the findings of Abou Bakr et al.[17]. The occurrence of cancer in BS is not negligible in the literature, which justifies the use of surgical exploration as a proactive strategy. Specifically, when obstructive jaundice is present, most BSs are considered to be malignant, and surgical intervention is commonly performed[18]. Nonetheless, we revealed that 69% of benign biliary strictures were accompanied by obstructive jaundice, which was considered quite a high percentage. Therefore, recommending exploratory surgery in all patients may not be justifiable especially since such benign strictures can be managed endoscopically.

Regarding the etiology of strictures, malignant stricture cases were 12 (33.3%) ampullary adenocarcinomas being the most common cause, 6 pancreatic head cancers, and 5 cholangiocarcinomas. Meanwhile, benign stricture cases were five fibrotic inflammatory, five CP, two AIP, and one large PAD. This differed from the results of Abou Bakr et al.[17], who found that pancreatic cancer was the most common cause of malignant strictures (43.5%). This can be attributed to variations in the patient selection process, as their study included 60 patients with obstructive jaundice and CBD dilatation identified through abdominal ultrasonography. Additionally, the variability in the assessment tools they used should be considered. In this case, CT scans were not utilized to determine the cause of obstructive jaundice, which means that patients with a mass that was detected through cross-sectional imaging, which is highly effective in identifying pancreatic cancer cases, were not excluded.

The EUS imaging findings consistently revealed mass lesions as a hallmark feature of malignancy, characterized by hypoechogenicity, irregular shape, and larger size compared to benign masses. In certain cases, these malignant lesions had invaded adjacent tissues, disrupting the bile duct layers and LN affection. All these features were significantly more prevalent in malignant strictures compared to benign strictures, except LN affection, which was also observed in benign strictures without significant differences. These features agree with that of Tamada et al.[19], with a difference in the detection of mass size > 8 mm, which was smaller than our mean mass size (2.9 cm). This is because they used intraductal ultrasound (IDUS) to define malignancy criteria, which is more sensitive than traditional EUS. Furthermore, Menzel et al.[20] reported a mixed echo pattern of malignant masses, and they observed that malignant LNs were hypochogenic, round-shaped, and larger than benign LNs detected by EUS imaging. From these features, Lee and his colleagues[3] found that in diagnosing adenocarcinoma of biliary and pancreatic origin, the presence of pancreatic head mass, irregular bile duct wall, or both had a sensitivity, specificity, PPV, and NPV of 88%, 100%, 100%, and 84%, respectively. In this study, follow-up was performed for 13 benign cases [8 with strictures due to CP and 5 fibrotic inflammatory strictures]. After six months of follow-up, three cases of those who were diagnosed as CP were found to have malignant pancreatic focal lesions. They may have had small pancreatic focal lesions at the initial EUS examination but were missed and enlarged to be detected after six months. Notably, CP is considered a risk factor for pancreatic cancer[21]; accordingly, developing malignant focal lesions alongside the previously diagnosed CP is possible.

Diagnosis of ampullary tumors is usually challenging and needs multiple imaging protocols[22]. This study identified two cases initially diagnosed by EUS as having malignant ampullary lesions [false positives], which histopathologically proved to be benign fibrotic inflammatory strictures. Conversely, two cases diagnosed by EUS as having benign strictures (fibrotic inflammatory strictures; false negatives) were histopathologically confirmed to be ampullary adenocarcinoma. The accuracy of distinguishing between benign and malignant ampullary lesions through sonography, relying on factors like echogenicity, shape irregularity, and disruption of the bile duct wall, may be influenced by the operator's expertise. Moreover, the presence of a bile duct stent can contribute to an increase in wall thickness and irregularity, potentially leading to misdiagnoses of malignancy[3]. In addition, Chen et al.[23] conducted a comparative analysis of various diagnostic imaging techniques (EUS, ERCP, US, and CT) to identify and determine the stage of primary ampullary tumors. They found that EUS had a sensitivity of 95% for identifying the tumors, 75% for determining the T stage, and 50% for determining the N stage compared to surgical histological findings. They also stated that the presence of no biliary stents might decrease the accuracy of EUS in staging ampullary tumors. Furthermore, Maluf-Filho et al.[24] reported 90% sensitivity of EUS in diagnosing ampullary adenocarcinoma.

Our study highlights the high diagnostic efficacy of EUS in identifying DBS without a mass, surpassing biopsy results, with a sensitivity, specificity, and accuracy of 90%, 71.4%, and 91.7%, respectively. Furthermore, when comparing EUS findings to the outcome after six months of follow-up, the sensitivity, specificity, and accuracy were 87%, 100%, and 91.7%, nearly similar to the results of Saifuku et al.[13]. Compared to the study of Heinzow et al.[25] who compared different diagnostic tools for detecting bile duct malignancy, our results had more accuracy. This may be because their study included a large group of 234 patients, and their follow-up was at least one year and extended to three years in some cases. Moreover, Abou Bakr et al.[17] reported sensitivity and specificity of EUS for detecting malignant obstruction of 100% and 86.36%, respectively, with an overall accuracy of 95%. Their studied patients were cases of obstructive jaundice, and they excluded patients with biliary stones detected by transabdominal ultrasonography only. Consequently, they reported higher sensitivity. In addition, Maluf-Filho et al.[24] showed that the sensitivity and specificity values for pancreatic adenocarcinoma detected by EUS were 96.6% and 90.6%, respectively, with an accuracy of 93.4%. These were higher results because they did not use MRCP and ERCP to exclude diagnosed cases.

Ultimately, the sonographic presentation on EUS demonstrates significant diagnostic potential for extrahepatic BS, particularly when other cross-sectional imaging techniques fail to visualize the underlying lesion. This non-invasive approach allows for avoiding unnecessary invasive procedures like surgical exploration or prolonged follow-up assessments in cases of benign strictures. Moreover, the early detection of malignant strictures through EUS provides patients with very early-stage pancreaticobiliary cancer with a broader spectrum of treatment options. EUS proves to be a highly accurate, minimally invasive diagnostic technique with low risk and high precision for patients with indeterminate distal biliary strictures[26].

 Limitations and Future Directions

The study has several limitations that warrant consideration. Firstly, the study is constrained by a limited sample size, which may impact the generalizability of the findings. Larger and more diverse cohorts in future studies could strengthen the robustness of the results. Additionally, the operator-dependent nature of EUS introduces variability, emphasizing the need for standardized procedures and the inclusion of multiple operators to mitigate this limitation.

Notably, discrepancies between EUS diagnoses and histopathological results resulted in false positives and negatives. Further investigation into the factors contributing to these discrepancies is crucial for refining diagnostic accuracy. The specificity of CEA and CA 19-9, while effective, may be influenced by other factors, prompting the exploration of additional biomarkers for improved specificity.

In terms of future directions, larger prospective studies with extended follow-up periods could offer more comprehensive insights into the diagnostic accuracy and predictive value of EUS for biliary strictures. Implementing standardized training programs for EUS operators and assessing their impact on diagnostic accuracy could enhance consistency across practitioners. Exploring the integration of advanced imaging techniques, such as artificial intelligence or machine learning algorithms, with EUS represents a promising avenue for improving diagnostic precision and reducing operator dependency. Collaboration with multiple centers is encouraged to facilitate the recruitment of larger and more diverse patient populations, contributing to a broader understanding of biliary strictures and the utility of EUS.

Finally, the evaluation of emerging biomarkers, in addition to CEA and CA 19-9, holds the potential for enhancing the specificity of diagnostic markers in distinguishing between benign and malignant biliary strictures. Addressing these limitations and pursuing these future directions is essential for advancing the field, leading to more reliable diagnostic methods and improved patient outcomes.

Conclusion

This study highlights the value of EUS as a minimally invasive and effective diagnostic tool for evaluating biliary strictures, particularly when no mass is visible on imaging. EUS plays a key role in early detection and differentiation of malignant from benign strictures, supporting timely and appropriate management.

Institutional Review Board Statement: This research study has been ethically reviewed and approved by the Institutional Review Board (IRB) at the Faculty of Medicine, Zagazig University, Egypt [6249-5-7-2020]. The IRB has carefully examined the study's protocol, methodologies, and ethical considerations, ensuring alignment with the principles outlined in the Declaration of Helsinki.

Informed Consent Statement: Written informed consent was obtained from each participant, signifying their voluntary agreement to take part in the study.

Data Availability Statement: The datasets generated and analyzed during the current study are available from the corresponding author upon reasonable request.

Acknowledgments: We sincerely appreciate EdigenomiX Scientific Co., Ltd. for their expert editing and proofreading, which greatly improved the clarity and quality of our manuscript. Their meticulous attention to detail and support in refining the document for publication is gratefully acknowledged.

Conflicts of Interest: The authors declare no conflict of interest