Comparison of Adverse Events of Different Endoscopic Ultrasound-Guided Tissue Acquisition Methods: A Single-Center Retrospective Analysis

The efficacy of new generation endoscopic ultrasound-guided biopsy needles has been promising in recent years. Yet, comparing these needles’ diagnostic yield and safety to conventional needles is not well-known. Our study aims to compare the adverse events of endoscopic ultrasound-guided tissue acquisition (EUS-TA) with different types of needles, including FNA needles, FNB needles with a Franseen tip and FNB needles with a reverse bevel. Furthermore, we will analyze the risk factors, including tumor vascularity, different needle types, and the underlying disease, which may impact the safety of the procedures. From May 2014 to December 2021, 192 consecutive EUS-TAs were performed on pancreatic and peripancreatic lesions in our hospital using different types of FNA and FNB needles. We retrospectively reviewed the data and identified the risk factors for EUS-TA-related complications. As a result, the hypervascular tumor is a significant risk factor for adverse events in our multivariate analysis, with an odds ratio of 4.96 (95% CI 1.33–18.47), while liver cirrhosis is one of the risk factors for adverse events during EUS-TA, with an odds ratio of 5.3 (95% CI 1.1–25.6). However, the risk of adverse events did not increase using Franseen-tip needles, compared to conventional FNA or FNB needles with a reverse bevel. In conclusion, we must be more cautious in patients with liver cirrhosis and hypervascular tumors, such as pancreatic neuroendocrine tumors, when performing EUS-guided tissue acquisition.


Introduction
Endoscopic ultrasound-guided tissue acquisition, including endoscopic ultrasoundguided fine-needle aspiration (EUS-FNA) and endoscopic ultrasound-guided fine-needle biopsy (EUS-FNB), has emerged as an accurate and safe modality in the diagnosis of pancreatic masses as well as one of the valuable diagnostic tools for intra-abdominal masses [1]. Although the specimen obtained from EUS-FNA is fragile and often insufficient for pathological evaluation, EUS-FNA has acceptable sensitivity and specificity for diagnosis with multiple needle passes [2]. In recent years, needles dedicated to fine-needle biopsy have been modified. As a result, more intact tissue sampling has become possible using this new generation of fine-needle biopsy devices. The accuracy of ultrasound-guided fineneedle biopsy without rapid on-site evaluation (ROSE) is not inferior to ultrasound-guided fine-needle aspiration with the aid of ROSE, and EUS-FNB alone is associated with fewer needle passes, shorter procedure time, and excellent histological yield [3,4].
The EUS-FNB needles' designs are divided into side fenestrations, Franseen tip, and Fork tip. Theoretically, these designs could carry a higher risk for tissue trauma than conventional aspiration needles, leading to a higher risk of bleeding, pancreatitis, or other adverse events. Some studies compared the efficacy and safety of EUS-FNA with EUS-FNB. Still, limited data exist regarding the impact of different FNB needles and the patient's comorbidity, which may be associated with procedure-related adverse events. Therefore, our study aims to compare the adverse events of Endoscopic ultrasound-guided tissue acquisition (EUS-TA) with different types of needles, namely FNA needles, FNB needles, Franseen-tip needles, and FNB needles with reverse bevel [5][6][7]. Furthermore, we will analyze the risk factors, including tumor vascularity, different needle types, and the underlying disease, that may impact the safety of the procedures.

Patients and Data Preparation
From May 2014 to December 2021, our hospital performed consecutive EUS-TAs on pancreatic and peripancreatic masses. After excluding simple fluid analysis (N = 21) and missing data (N = 6), a total of 192 procedures of EUS-FNA/FNB on pancreatic and peripancreatic lesions was performed on 185 patients in our hospital (Table 1). We retrieved the data of every patient undergoing EUS-TA registered in the database from 2014 and retrospectively reviewed the medical file, including the complications and final diagnosis. We checked the platelet count and prothrombin time before EUS-TA, and all enrolled patients had platelet count > 50,000/µL, without prolongation of prothrombin time. Those on anticoagulant therapy were instructed to discontinue the medication before the procedure, according to ESGE guidelines [8]. In addition, before undergoing EUS-TA, all patients provided written informed consent. The local ethics committee approved the study in our hospital (IRB no. 220423).

EUS-TA Procedures
The physician who performed EUS-TA procedures in this study had performed more than 200 EUS procedures (including 120 EUS-TA procedures) in a high-volume center before the study period. All the patients were placed in the left lateral position, and conscious sedation was performed using either intravenous pethidine hydrochloride (35 mg) or intravenous fentanyl (50 µg), along with intravenous midazolam (5 mg).
The After the needle was inserted into the target lesion, the stylet was removed. For FNA needles, to and fro motions were performed 20 times using negative pressure in the syringe. For the FNB needle, to and fro motions were performed 10 times, with the so-called "slow pull method" or without negative pressure, depending on the needle size and the vascularity of the target. Since rapid on-site evaluation (ROSE) is not available in our institution, we applied the method, called "macroscopic on-site quality evaluation" (MOSE), right after the method was first reported in 2015 by a Japanese medical team [9]. As a result, the final diagnosis of pancreatic masses was confirmed by a combination of smear cytology, histopathology, surgery, imaging, and clinical follow-up for more than six months.

Assessment of Adverse Events and Variables
After the procedure, the medical team confirmed the symptoms and physical findings on the day following the procedure and followed up on the case at the outpatient department one week after discharge. We arranged diagnostic imaging such as abdominal sonography or computed tomography (CT), if the patients had any complaints suggestive of acute pancreatitis or peritonitis. According to the Atlanta criteria, pancreatitis was diagnosed based on abdominal pain, abdominal sonography, or CT findings and elevated pancreatic enzyme greater than three times normal limits. Besides, gastrointestinal bleeding was defined as a drop in the hemoglobin level by 2 g/dL, as compared with the baseline levels and clinical signs of bleeding, along with any events undergoing endoscopic hemostasis after tissue acquisition. If a large coagulum on the puncture site was noted right after EUS-TA, we also defined it as minor (mild) bleeding. The severity of the adverse events was defined according to a workshop held by the American Society for Gastrointestinal Endoscopy (ASGE), published in 2010 [10].

Statistics
Data are expressed as n (%) and median (interquartile range, IQR) (range). The onesample Kolmogorov-Smirnov test examined the distribution of continuous variables. In comparing the three groups (conventional FNA needles, FNB with Franseen-tip needles, and FNB needles with a reverse bevel), categorical variables were compared using the chi-squared test or Fisher's exact test with Bonferroni correction, and continuous variables were compared using Kruskal-Wallis test. In the comparisons between the two groups (with/without complication), categorical variables were compared using the chi-squared test or Fisher's exact test, and continuous variables were compared using the Mann-Whitney U test (continuous data), as appropriate. To enter the multivariate adjustment model 1 and model 2, a multivariate logistic regression analysis was used to analyze factors associated with the complication and selected variables with a p-value < 0.2 from the crude model. Moreover, model 2 used the stepwise backward elimination procedure. The results were statistically significant if the p-value was <0.05 for all tests. All statistical data were analyzed using IBM SPSS version 22.0 (IBM Corp., Armonk, NY, USA).

Results
During the study period, 192 procedures of EUS-TA were performed. The overall complication rate was 6.7% (13), and the overall gastrointestinal bleeding rate was 4.2% (8). Among them, six cases with a gastrointestinal bleeding event were mild or moderate bleeding, while two cases needed blood transfusions or prolonged admission. Besides those hemorrhage episodes, there were one case of mild pancreatitis and two cases of fever after tissue acquisition. In two patients, duodenal perforation occurred after EUS-TA, so the patients underwent emergent surgery (case 8 and case 11).
The baseline characteristics are listed in Table 1. Since perforation is related to scope manipulation rather than EUS-FNA/FNB-related complications, the two cases of perforation were excluded from our statistical analysis. We divided the needle type into three groups, which were the FNA needle (N = 76), FNB needle with a Franseen tip (N = 95), and FNB needle with a reverse bevel (N = 19). In these three groups, there was no significant difference between patient factors (age, gender, comorbidity), tumor factors (tumor size, location, vascularity, cystic component), or technique factors such as needle passes and sampling route. However, we tended to select a 22 G needle when we performed FNB with a Franseen-tip needle (69.5% vs. 43.4%, p = 0.002) and to select a 25 G needle when we performed FNA with a conventional FNA needle (55.3% vs. 30.5%, p = 0.003). In our institution, a 19 G FNA needle was selected when we performed EUS-guided pancreatic pseudocyst drainage or other EUS-guided local treatment such as fluid analysis or ethanol injection [11]. In terms of adverse events, there was no adverse event in the group using an FNB needle with a reverse bevel. At the same time, there were six cases (7.9%) of adverse events in the FNA group and five cases (5.3%) of adverse events in the group using an FNB needle with a Franseen tip (p = 0.482).
The variables were examined in 190 patients who underwent EUS-TA for the pancreatic and peripancreatic masses to identify the risk factors of adverse events during the procedures. They are patient factors such as age, gender, comorbidity (compensated liver cirrhosis and end-stage renal disease), tumor factors such as tumor size, tumor vascularity (hypervascular or hypovascular), cystic component in the mass, and technique factors such as needle types (FNA needles, FNB with a Franseen tip, or FNB with a reverse bevel), route of sampling, size of the needle, and the number of needle passes. The diagnosis of liver cirrhosis in our hospital was made by gastroenterologists based on clinical examination, imaging findings (CT, MRI, abdominal sonography), and laboratory findings suggestive of liver cirrhosis and portal hypertension. Besides, the vascularity of the tumors was evaluated by contrast CT/MRI and Doppler EUS. A hypervascular lesion was defined as a hyperdense/hyperechoic mass compared to adjacent pancreas parenchyma on either contrast CT, contrast MRI, or doppler EUS. End-stage renal disease was defined as having a glomerular filtration rate of less than 15 mL/min. Univariate analysis showed that the incidence of procedural complications was significantly high in patients with old age (p = 0.039), liver cirrhosis (p = 0.024), and hyper-vascular tumors (p = 0.010) ( Table 2). Multivariate analysis identified hypervascular tumors (OR 4.96, 95% CI 1.33-18.47) and liver cirrhosis (OR 5.3, 95% CI 1.1-25.6) as independent risk factors of adverse events (Table 3). All the details of adverse events are listed in Table 4.

Discussion
Our study's overall complication rate associated with EUS-TA was 5.8%, slightly higher than that reported in previous studies [7,[12][13][14][15][16]. For example, Hamada et al. [13] reported a severe bleeding rate of 0.23% in 212 patients undergoing EUS-FNA. In addition, Akio Katanuma et al. reported a complication rate of 3.4% in a retrospective study in 2013 [17]. Unlike prior studies, our study enrolled patients with compensated liver cirrhosis and end-stage renal disease, which may explain the slightly higher complication rate. Among our adverse events, the majority were bleeding events, with an overall bleeding rate of 4.2% (N = 8). However, most of them were mild to moderate in severity (6/8), and severe bleeding events requiring prolonged admission or ICU care were observed in two cases (2/8).
Peter Vilmann first performed EUS-TA for the cytologic diagnosis of a pancreatic lesion in 1991 [18].This technique has now become the standard method for tissue acquisition from pancreatic and peripancreatic lesions [14,19]. A recent network meta-analysis demonstrated better accuracy with 22-gauge FNB needles than conventional 22-gauge FNA needles [20], but it did not compare the performance between different types of FNB needles, namely, side-fenestrated needles, Franseen-tip needles, and Fork-tip needles. About this issue, another network meta-analysis in 2022 demonstrated the higher accuracy of Franseen-tip needles and Fork-tip needles than reverse-bevel needles and conventional FNA needles [21,22], especially in the absence of ROSE [3].
However, the safety of novel FNB needles, compared to conventional FNA needles and reverse-bevel needles, has rarely been discussed. A retrospective study in 2021 revealed that the use of FNB needles (most of them Franseen-tip needles and Fork-tip needles) and lower platelet count were two parameters associated with minor bleeding after EUS-TA [23]. In our multivariate analysis, there is no statistically significant difference between Franseen-tip needles, conventional FNA needles, and reverse-bevel needles regarding adverse events. The difference between our study and that prior study may be due to the high proportion of antithrombotic agents used in that study (29.8%, in the solid tumor group). Similarly, a recent meta-analysis comparing the efficacy and safety of EUS-FNA and EUS-FNB also revealed a rare occurrence of AE, with no difference between FNA and FNB. However, due to the rarity of AE and variations in reporting, that meta-analysis was unable to provide meaningful analysis regarding specific complications [24,25]. Given the conflicting result compared to prior studies, we considered that a prospective study enrolling a more significant number of patients is still needed to determine the risk factors of adverse events during EUS-TA.
Hypervascular tumors are important risk factors for adverse events in our multivariate analysis, with an odds ratio of 4.96 (95% CI 1. 33-18.47). When we perform EUS-TA, pancreatic neuroendocrine and gastrointestinal stromal tumors are the most common hypervascular tumors we may encounter. Small pNETs are almost hypervascular tumors, and the risk of bleeding from these tumors is increased compared with hypovascular tumors. When bleeding occurs around the pancreatic parenchyma after EUS-FNA/FNB, it may present as both upper gastrointestinal or intra-abdominal bleeding (case 13 in Table 4; Figure 1). A retrospective study by Akio Katanuma et al. demonstrated that pNET is the most critical risk factor for an adverse event, with an odds ratio of 36.5, compared to benign lesions using FNA needles (14). Our study showed that hypervascular tumors remained the most crucial risk factor for an adverse event in the era of EUS-FNB. Recently, EUS-guided through-the-needle biopsy (EUS-TTNB) has been developed and proven to be a safe and effective tool in EUS-guided tissue acquisition, especially when we encountered cystic lesions in a pancreas with suspicion of a cystic neoplasm or a malignant change in a pancreatic cyst. In the cystic neoplasm, such as our case 13, we should keep in mind that EUS-guided through-the-needle biopsy is one of the choices because this method can avoid a hemorrhage into the retroperitoneal cavity after tissue acquisition [26]. However, a recent study has identified the risk factors and risk classes of an adverse event after EUS-TTNB, and it becomes crucial to select patients before EUS-TTNB to optimize the benefit and risk of this procedure [27]. Besides, a recent meta-analysis demonstrated that routine use of hemostatic powder is helpful for patients with upper gastrointestinal bleeding, suggestive of the possibility of applying this method to EUS-TA for a subepithelial lesion [28].
Diagnostics 2021, 11, x FOR PEER REVIEW 8 o classes of an adverse event after EUS-TTNB, and it becomes crucial to select patien before EUS-TTNB to optimize the benefit and risk of this procedure [27]. Besides, a rece meta-analysis demonstrated that routine use of hemostatic powder is helpful for patien with upper gastrointestinal bleeding, suggestive of the possibility of applying this meth to EUS-TA for a subepithelial lesion [28]. Our study demonstrated that liver cirrhosis is one of the risk factors for adver events during EUS-TA, with an odds ratio of 5.3 (95% CI 1. 1-25.6). Reports about t adverse events of cirrhotic patients in the English literature are rare. A retrospective stu of 692 patients who underwent ERCP revealed that overall complications, includi pancreatitis and bleeding, were higher in those with cirrhosis than those without cirrho (p = 0.015) [29], suggesting that liver cirrhosis is a significant risk factor during pancreaticobiliary endoscopic procedure. On the other hand, a study of EUS-guided liv biopsy that included 15.8% cirrhotic patients showed that adverse events were uncomm (1.8%), most of which were self-limited [30]. However, as shown in our data, it reasonable to say that liver cirrhosis is a risk factor for an adverse event during EUS-T since a low platelet count is common in cirrhotic patients, and thrombocytopenia itself a well-known risk factor of bleeding during EUS-TA [23].
The study has some limitations. First of all, due to the retrospective design a relatively small case number, selection bias may occur [31][32][33][34]. However, although o study was retrospective, all the patients undergoing EUS-TA were registered in a databa every time we performed these procedures. Therefore, we can ensure the accuracy of o database. Second, this study only enrolled pancreatic and peripancreatic tumors, wh pseudocysts and subepithelial tumors (SET) in the gastrointestinal tract were exclude Hence, the result of our study cannot be valid for SET. Third, although the use antithrombotic agents before EUS-TA was based on the ESGE guideline, we had detailed information about all the drug history before the patients were admitted. Final Our study demonstrated that liver cirrhosis is one of the risk factors for adverse events during EUS-TA, with an odds ratio of 5.3 (95% CI 1. 1-25.6). Reports about the adverse events of cirrhotic patients in the English literature are rare. A retrospective study of 692 patients who underwent ERCP revealed that overall complications, including pancreatitis and bleeding, were higher in those with cirrhosis than those without cirrhosis (p = 0.015) [29], suggesting that liver cirrhosis is a significant risk factor during a pancreaticobiliary endoscopic procedure. On the other hand, a study of EUS-guided liver biopsy that included 15.8% cirrhotic patients showed that adverse events were uncommon (1.8%), most of which were self-limited [30]. However, as shown in our data, it is reasonable to say that liver cirrhosis is a risk factor for an adverse event during EUS-TA, since a low platelet count is common in cirrhotic patients, and thrombocytopenia itself is a well-known risk factor of bleeding during EUS-TA [23].
The study has some limitations. First of all, due to the retrospective design and relatively small case number, selection bias may occur [31][32][33][34]. However, although our study was retrospective, all the patients undergoing EUS-TA were registered in a database every time we performed these procedures. Therefore, we can ensure the accuracy of our database. Second, this study only enrolled pancreatic and peripancreatic tumors, while pseudocysts and subepithelial tumors (SET) in the gastrointestinal tract were excluded. Hence, the result of our study cannot be valid for SET. Third, although the use of antithrombotic agents before EUS-TA was based on the ESGE guideline, we had no detailed information about all the drug history before the patients were admitted. Finally, since the period of data analysis is long (2014-2021), there was a learning curve for both the examiner and the cytopathologist, which may influence our data analysis [35,36].

Conclusions
In conclusion, different needle types (conventional FNA, Franseen-tip, or reverse-bevel needles) did not impact the safety of EUS-TA in our study. On the contrary, hypervascular tumors and cirrhotic patients were risk factors for adverse events during EUS-TA, so we must be more cautious in patients with such risk factors. Informed Consent Statement: Patient consent was waived by the institutional review board due to the retrospective nature of the present study.

Data Availability Statement:
The datasets generated and/or analyzed during the current study are not publicly available, but these may be requested from the corresponding author upon reasonable request.

Conflicts of Interest:
The authors declare no conflict of interest.