Introduction

Endoscopic ultrasound-guided fine needle biopsy (EUS-FNB) has been proposed as a reliable, safe, and effective method of obtaining tissue samples from various organs, including pancreas, adrenal mass, and liver, for histological analysis1,2. EUS-FNB offers enhanced accessibility at a greater depth than ultrasound-guided percutaneous biopsy3, is minimally invasive, and provides high diagnostic accuracy4. However, insufficient tissue samples can hinder diagnosis; this may be due to the design of the biopsy needle tip, its gauge, or the movement of the needle within the lesion5,6. Further development and improvement of FNB needles are therefore required to improve the efficacy of EUS-FNB.

FNB needles with differently-shaped tips, various designs, and diverse materials have been developed to enable the obtainment of large tissue samples and increase puncture ability7,8. These FNB needles include fork-tip needles (SharkCore™; Medtronic, Minneapolis, MN, USA), which have two opposite stainless steel cutting edges, and Franseen needles (Acquire™; Boston Scientific, Marlborough, MA, USA) with three crown-cut cutting edges at the cobalt-chromium needle tip1,9. Recently, a modified crown-cut cutting needle (TriCore™; Koswire, Busan, Korea) with two symmetrical short stainless steel cutting edges on opposite sides of a single long stainless steel cutting edge has been developed. The purpose of this study was to compare the effectiveness of the newly designed Tricore FNB needle with a commercially available Franseen FNB needle under EUS guidance in a porcine liver model.

Methods

Characterization of the two FNB needles

Two types of biopsy needles including Franseen (Acquire™; Boston Scientific) and Tricore (Koswire) with 22G size were used in this study. The puncture ability of the two needles was assessed in accordance with guideline DIN 13097-4 issued by the German Institute for Standardization (Fig. 1a). The resistance force during the puncture of 0.4-mm thick polyurethane foil was measured using an LS1 materials testing machine (Lloyd Instruments Ltd., Bognor Regis, UK) and the following settings: pre-test speed, 10 mm/min; test speed, 100 mm/min; test penetration length, 10 mm; and test direction, perpendicular. The maximum load was measured and recorded using NexygenPlus software (Lloyd Instruments Ltd.). Three samples of each FNB needle type were used, and each needle was tested ten times.

Fig. 1
figure 1

Characterization of fine-needle biopsy needles. (a) Puncture ability was measured using the test machine. (b) The angles at the end-tip of all three needle cutting edges were measured before and after endoscopic ultrasound-guided fine-needle biopsy (θ1, θ2, and θ3).

Full size image

The angle of the FNB needle tip before and after the EUS-FNB procedure was measured to confirm the sustainability of puncture performance. As shown in Fig. 1b, the angle between the tangent of the needle body and the tangent of the needle tip was defined as the bending angle10. The angle of the FNB needle was measured on all three split needle tips.

Animal study design

This study was approved by the Institutional Animal Care and Use Committee of the Asan Institute for Life Sciences (no. 2023-20-113) and conformed to the National Institutes of Health Guide for the Care and Use of Laboratory Animals. This study was carried out in compliance with ARRIVE guidelines. Four male Yorkshire pigs weighing 27.50–33.70 kg (mean, 30.85 kg) were obtained from the International Animal Experiment Center (Pocheon, Korea). Animals were supplied with food and water ad libitum and housed at a temperature of 24 ± 2°C under a 14/10-h light/dark cycle. Pigs were randomly divided into two groups: the Franseen group (n = 2) underwent EUS-FNB using a Franseen needle, and the Tricore group (n = 2) underwent EUS-FNB using a Tricore needle. Nine EUS-FNB procedures were performed on each pig using three FNB needles. The sample size was selected based on previous research in the field and practical considerations11. The study used four pigs, with nine samples collected from each animal, for a total of 36 samples. However, after accounting for within-subject correlations, the effective sample size per animal fell outside the typical range recommended by the ‘resource equation’ method12. The sample size was determined to achieve a balance between the need for ethical animal use and the ability to conduct meaningful statistical analysis.

Porcine liver EUS-FNB procedure

Pigs were anesthetized using a combination of 50 mg/kg zolazepam, 50 mg/kg tiletamine (Zoletil™ 50; Virbac, Carros, France), and 10 mg/kg xylazine (Rompun®; Bayer Healthcare Pharmaceuticals LLC, Berlin, Germany). Anesthesia was maintained by inhalation of 2% isoflurane (Ifran®; Hana Pharm Co., Ltd., Seoul, Republic of Korea) with a 1:1 oxygen ratio (510 mL/kg per min) through an endotracheal tube. The EUS-FNB procedure was performed by a gastroenterologist with more than 20 years’ experience using an EUS linear scope (GF-UCT260; Olympus Corporation, Tokyo, Japan). The puncture site under EUS guidance was selected based on accessibility and the absence of vascular structures to avoid risk of bleeding. The FNB needle was advanced through the working channel of the endoscope with a locking device set to fix the needle depth at 3 cm. Once the FNB needle reached the target lesion of the liver, the stylet was removed and a syringe with 10 mL of negative pressure was connected. The needle was moved back and forth 10 times in three different directions at approximately 45 degrees using the fanning technique. The suction applied to the FNB device was released, and the needle was removed from the working channel. Heparinized saline was used to expel the obtained specimen from the needle. Immediately after the final procedures, a laparotomy was performed to grossly evaluate bleeding at the puncture site and needle penetrating liver injuries. All pigs were euthanized immediately after the final procedure. Pigs were euthanized by the administration of 75 mg/kg potassium chloride (Dai Han Pharmaceutical Co., Ltd., Seoul, Republic of Korea) through the marginal ear vein.

Histological analysis

All obtained specimens were fixed in 10% neutral buffered formalin (Sigma-Aldrich, St. Louis, MO, USA), embedded in paraffin, and cut into 5-µm thick sections. Sections were stained with hematoxylin and eosin to determine if the tissue sample was sufficient for diagnosis, and Masson’s trichrome to identify complete portal tracts (CPTs), which are composed of a portal vein, hepatic artery, and bile duct. Histological images were visualized using CaseViewer software (3DHISTECH Kft., Budapest, Hungary) and analyzed using ImageJ software (National Institutes of Health, Bethesda, MD, USA).

Outcome measurement

Ultrasound visibility was evaluated and compared between the two FNB needles to ensure precise puncture during the EUS-FNB procedures. Ultrasound images were obtained immediately after needle penetration of the porcine liver. The entire area of the needle observed in the ultrasound image was selected as a region of interest, and the mean pixel intensity in this area was measured using ImageJ software. Pixel intensity was defined by gray-scale level between 0 (black) and 255 (white)13. Blood contamination of the needle was evaluated and compared between the two FNB needles by measuring the area of blood absorbed into the filter paper following specimen release using ImageJ software. Specimen acquisition time was also measured from the needle penetration of the liver to the retrieval of the specimen from the needle.

The sufficiency of the tissue sample for diagnosis was assessed by measuring: (1) the total specimen area, (2) number of fragments, (3) fragment areas, and (4) number of CPTs. Histological examinations were conducted by three independent observers, all of whom were blinded to the details of the study.

Statistical analysis

Data are presented as mean ± standard deviation. Differences between the two groups were analyzed using two-sample t-tests for normally distributed data and Mann–Whitney U tests for non-normally distributed data. Statistical analyses were performed using SPSS software version 27 (IBM Corporation, Armonk, NY, USA), and P < 0.05 was considered statistically significant.

Results

Characteristics of the two FNB needles

The characteristics of the two FNB needles are shown in Fig. 2. The mean resistance to the Franseen needle (5.97 ± 0.53 N) was significantly higher than that to the newly designed Tricore needle (3.83 ± 0.84 N, P < 0.001, two-sample t-test). The mean angle of the three symmetrical cutting edges of both the Franseen (19.02 ± 3.66° vs. 84.58 ± 39.69°, P < 0.001, two-sample t-test) and Tricore (20.43 ± 2.71° vs. 32.82 ± 14.17°, P = 0.003) needles differed significantly before and after the procedure, respectively, representing considerable bending of both needle types. The change in Franseen needle angle was significantly greater than the change in Tricore needle angle (P < 0.001).

Fig. 2
figure 2

Characterization of the two FNB needles. (a) The resistance to the Franseen needle was significantly higher than that to the Tricore needle. (b) Needle angle changed significantly after the procedure compared with before for both needle types; however, the Franseen needle became more significantly bent than the Tricore needle. (c) Representative images of needles before and after procedures. The white arrowheads indicate the bent needle tips. ** P < 0.01; *** P < 0.001. EUS-FNB, endoscopic ultrasound-guided fine needle biopsy; CI, confidence interval.

Full size image

Procedural outcomes

All EUS-FNB procedures were technically successful, and no procedural-related complications were observed. Specimens were collected with a 100% (36/36) success rate. Gross examination revealed no evidence of bleeding or hematoma formation at the site of needle penetration in the liver. The mean pixel intensity of the Tricore group (168.97 ± 30.02) was significantly higher than that of the Franseen group (125.04 ± 29.27, P = 0.004, two-sample t-test) (Fig. 3). The mean percentage of blood contamination did not differ significantly between the Franseen (49.95 ± 20.31%) and the Tricore (50.57 ± 19.52%, P = 0.926, two-sample t-test) groups (Fig. 4a,b). The mean specimen acquisition time in the Franseen group (59.90 ± 19.15 s) was significantly longer than that in the Tricore group (48.94 ± 9.26 s, P = 0.038, two-sample t-test) (Fig. 4c).

Fig. 3
figure 3

Ultrasound findings during EUS-FNB in the porcine liver. (a) Representative ultrasound images of the two FNB needles. The Tricore group needle (arrows) produced relatively fewer metal artifacts than the Franseen group needle (arrow heads). (b) The pixel intensity of the Tricore group was significantly higher than that of the Franseen group. ** P < 0.01. CI, confidence interval; EUS-FNB, endoscopic ultrasound-guided fine needle biopsy.

Full size image
Fig. 4
figure 4

Intraoperative findings. (a) Blood contamination was quantified as the percentage of the total area of the filter paper contaminated with blood following specimen release. (b) Blood contamination did not differ between the Franseen and Tricore groups. (c) The mean specimen acquisition time was significantly shorter in the Tricore group than in the Franseen group. * P < 0.05. CI, confidence interval.

Full size image

Histological findings

Histological findings are presented in Fig. 5. The mean total specimen area in the Franseen group (4.68 ± 3.04 mm2) was significantly lower than that in the Tricore group (6.67 ± 2.96 mm2, P = 0.049, two-sample t-test). The mean number of fragments in the Franseen group (31.94 ± 22.61) was considerable but not significantly higher than that in the Tricore group (23.94 ± 11.11, P = 0.190, two-sample t-test). The mean fragment area in the Franseen group (0.15 ± 0.24 mm2) was significantly lower than that in the Tricore group (0.28 ± 0.34 mm2, P < 0.001, Mann–Whitney U test). The mean number of CPTs in the Franseen group (9.33 ± 6.61) was significantly lower than that in the Tricore group (15.44 ± 7.88, P = 0.017, two-sample t-test).

Fig. 5
figure 5

Representative histological findings. (a) Hematoxylin and eosin staining was performed to evaluate the sufficiency of the tissue for diagnosis. The number and area of each fragment in the (b) Franseen group and (c) Tricore group were measured using ImageJ. (d) Masson’s trichrome staining was performed to identify complete portal tracts, composed of a PV, HA, and BD. BD, bile duct; HA, hepatic artery; PV, portal vein.

Full size image

Discussion

The results of our study demonstrate that the newly designed Tricore needle can successfully acquire liver tissue specimens under EUS guidance. The mechanical properties of the Tricore needle were similar or superior to those of the commercially available Franseen needle, and it demonstrated superior puncture ability. Furthermore, the Tricore needle exhibited greater ultrasound visibility and a shorter specimen acquisition time than the Franseen needle. Histological findings revealed that the specimens obtained by the Tricore needle were superior to those obtained by the Franseen needle in terms of total specimen area, number of fragments, fragment area, and number of CPTs. The selection of the FNB needle is an important aspect of the EUS-FNB procedure14; our findings suggest that the Tricore needle may be a promising additional option.

Large-diameter FNB needles have excellent puncture performance and produce tissue samples that enable diagnostic accuracy; however, their insertion is hampered due to limitations in elevator function and limited tip angle motion15,16,17. Therefore, we selected a diameter of 22G for the Tricore FNB needle for better access to the lesion. The newly designed needle features a trident-shaped design with one long and two short cutting edges designed to overcome the limitations of a blunt tip, which enhances puncture ability. Consistent with this, the Tricore needle encountered significantly lower resistance during puncture than the Franseen needle, indicative of superior puncture ability16. Furthermore, the angle of the Franseen needle bent to a significantly greater extent than that of the Tricore needle, perhaps due to the higher resistance encountered. The resistance of the Tricore needle during puncture was significantly higher for the Franseen needle than for the Tricore needle. These results back up the finding that the tissue acquisition time was significantly shorter with the Tricore needle than with the Franseen needle. Therefore, the Tricore needle may provide superior puncture ability and convenience compared to existing needles.

EUS-FNB procedures are known for their accuracy and efficiency, but adverse effects such as bleeding still occur18. Improvements in ultrasound visualization are essential to allow identification of puncture angle, needle depth, and location, and therefore improve puncture precision19,20. In the present study, the Tricore needle was more clearly visible under ultrasound than the Franseen needle. Enhanced ultrasound visibility has been shown to be crucial to enable a clear view of the needle’s trajectory and its proximity to adjacent structures during EUS-FNB21. Visibility is key to minimizing complications including vessel perforation and bleeding19. Enhancing ultrasound visibility can be achieved through needle surface modification methods such as sandblasting, dimpling, electric discharge machining, or polymer coating22. Metal instruments produce strong artifacts under ultrasound guidance, making visualization difficult23,24. Artifacts can be affected by factors such as device location, angle, and surface modification25, and artifacts-related evaluation was therefore not performed in the present study due to the inability to control all the variables. However, it was observed that the Tricore needle had relatively fewer artifacts than the Franseen needle. These findings suggest that the dual surface modification of the Tricore needle using the kerf pattern and heat treatment contributed to improved ultrasonic signal reflection; this may result in safer and more accurate procedures.

We introduced an FNB needle with a new geometric structure and evaluated its performance through histological examination of the specimens obtained with each needle. The Tricore needle samples were superior to the Franseen needle samples in terms of total specimen area, number of fragments, and fragment area. Fragmentation of the tissue specimen poses challenges to disease diagnosis as it obstructs the accurate identification of the core involved in the disease26. The issue of specimen quality is particularly important in the era of personalized medicine, as large and high-quality samples are crucial for immunohistochemistry. The number of CPTs, which is considered key in the evaluation of liver biopsy specimen adequacy7, was also significantly higher in the Tricore needle group than in the Franseen needle group. Numerous studies have proposed that more than six CPTs are required for specimen adequacy. However, some studies and the Academy for the Study of Liver Diseases recommend obtaining 11–15 CPTs to ensure the reliability of histological analysis27, a level achieved by the Tricore needle but not the Franseen needle in the present study. These findings indicate that the Tricore needle outperforms the Franseen needle in terms of specimen quantity and quality.

This study has several limitations. First, the study was performed in the livers of healthy pigs rather than in a clinically-relevant situation. While our results indicate the superiority of the new FNB needle in terms of mechanical properties and specimen quantity and quality, clinical confirmation of these results is necessary. Second, the lack of long-term follow-up data on adverse effects highlights the need to investigate long-term safety in future studies. Third, all EUS-FNB procedures were performed by one operator. While this minimizes variation, the results could be influenced by the specific skills and techniques of the individual, potentially limiting the generalizability of the findings. Forth, a crossover design could reduce between-subject variability by allowing direct comparison of needle performance within the same subject. In our study, we chose a parallel approach to allow for independent evaluation of adverse effects of each needle28. Future studies could benefit from comparing results from crossover designs with larger sample size to compare the needle performance within the same porcine liver. Fifth, the small sample size, with a total of 18 specimens per needle type, prevented us from performing robust statistical analyses. Finally, although the puncture depth and procedure method were consistent across all attempts, minor variations may occur due to anatomical differences in the enrolled pigs. Nevertheless, these findings provide preliminary evidence suggesting the superior performance of a new EUS-FNB needle.

In conclusion, the newly designed Tricore EUS-FNB needle demonstrated outstanding puncture ability and ultrasound visibility, and acquired specimens of greater quantity and quality than those obtained by the Franseen needle. Although further clinical studies are required to confirm these results, this newly designed needle may emerge as a promising choice for EUS-FNB procedures.