Keywords
Endoscopic ultrasonography - Fine-needle aspiration/biopsy - Intervention EUS
Introduction
Endoscopic ultrasound-guided fine-needle aspiration (EUS-FNA) is a minimally invasive
method of obtaining tissue for diagnosis of pancreatobiliary, hepatic, and gastrointestinal
tract diseases [1]
[2]
[3]
[4]
[5]. This procedure involves insertion of an echoendoscope and a fine needle, which
can be guided to the targeted area under real-time EUS guidance. A meta-analysis of
33 studies involving 4984 patients showed that EUS-FNA had a pooled sensitivity of
85% (95% confidence interval [CI], 84%–86%) and a pooled specificity of 98% (95% CI,
97%–99%) for diagnosis of pancreatic cancer [6]. Despite the high diagnostic performance of EUS-FNA, several factors, such as endoscope
position and angle, use of an elevator, and tumor hardness, can affect tissue sampling.
Moreover, cytological aspiration by EUS-FNA has potential limitations, including an
inability to determine histologic architecture, and quantitative samples that are
too small for additional immunohistochemical assays. EUS-guided fine-needle biopsy
(EUS-FNB), with its larger core biopsy needle, was designed to overcome these potential
limitations. The need to obtain larger tissue specimens for precision medicine has
favored use of EUS-FNB. A meta-analysis of 18 studies including EUS-FNB (n=993) and
EUS-FNA (n=1017) showed that pooled diagnostic accuracy and tissue core rate were
significantly higher for EUS-FNB (87% and 80%) than for EUS-FNA (80% and 62%). EUS-FNB
also required significantly fewer passes for diagnosis [7]. For differentiation of mucinous pancreatic cystic lesions, the network ranking
of the superiority index for EUS-guided needle-based confocal laser endomicroscopy
and EUS-guided through-the-needle biopsy were significantly better than for other
techniques in a meta-analysis of 40 studies including 3,641 patients [8]. Concerns have arisen about peritoneal seeding after EUS-FNA of mucinous pancreatic
cystic lesions. In a meta-analysis of 10 studies, the pooled rate of peritoneal seeding
was 0.4% in patients with solid masses and 0.3% in patients with pancreatic cystic
lesions. Moreover, there was no difference between patients who underwent EUS-FNA/FNB
and non-sampled patients (odds ratio 1.02, 0.72–1.46; P=0.31) in terms of metachronous peritoneal dissemination [9]. It is important to assess whether fluid leakage occurs after EUS-FNA/B puncture
of cystic lesions.
The relative procedural performance of needles used for EUS-FNA/B puncture of cystic
lesions is unclear. The present study compared the performance of six types of 22-gauge
EUS-FNA/B needles using a bench simulator designed to provide standardized, reproducible,
and comparative performance data.
Methods
Needle and echoendoscope
The performances of six commercially available 22-gauge FNA/B needles were compared:
(1) EZShot3 Plus (Olympus Medical Systems, Tokyo, Japan), (2) Expect (Boston Scientific,
Marlborough, Massachusetts, United States), (3) ProCore (Cook Medical, Bloomington
Indiana, United States), (4) Acquire (Boston Scientific, Marlborough, Massachusetts,
United States), (5) SharkCore (Covidien, Japan Inc.), and (6) TopGain (Medi-Globe,
Achenmuhle, Germany). Each experiment tested five needles of each type. EUS was performed
with an EG-580UT echoendoscope (Fujifilm, Tokyo, Japan).
Bench simulation
Measurement of resistance forces during puncture and removal
Each needle was moved toward a 0.3-mm thick polyvinyl chloride board at a speed of
500 mm/min. Maximum resistance of each needle during puncture through the polyvinyl
chloride was measured using a rheometer (IMADA CO., LTD, Aichi, Japan) ([Fig. 1]). Maximum needle resistance during removal from the polyvinyl chloride board was
also measured.
Fig. 1 Measurement of resistance forces during needle puncture and removal. Each needle was
advanced toward a 0.3-mm thick polyvinyl chloride board at a speed of 500 mm/min.
The maximum resistance of each needle during puncture through the polyvinyl chloride
board was measured using a rheometer (IMADA CO., LTD, Aichi, Japan).
Needle tip abrasion damage before and after puncture
The geometry of each needle was evaluated by three investigators (Y.Y., H.Y., and
C.G.) before and after puncture. Needle shape was checked prior to puncture. The needle
was inserted into and retracted from a polyvinyl chloride board at a speed of 500
mm/min. This procedure was performed 100 times while altering the site of insertion
into the board. Abrasion damage to the needle tip was subsequently evaluated using
a microscope.
Measurement of leakage after puncture of a mucinous cyst model
A mucinous cyst model was composed of a pressure vessel filled with echo gels, to
which was attached a 2-mm thick silicone sponge rubber. The needles were used to puncture
the silicone sponge rubber, and the volume of fluid that leaked from the inside to
the outside of the cyst model was measured. The pressure in the vessel can be artificially
raised from 0 to 50 Kpa. The silicone sponge was punctured, and the volumes of fluid
leaking from the inside to the outside of the cyst model at pressures of 5 and 10
Kpa per minute were measured ([Fig. 2]). The pressure at which the liquid leaks out was also determined by gradually increasing
the pressure from 0 to 50 Kpa.
Fig. 2 Measurement of leakage after puncture of a mucinous cyst model. A mucinous cyst model
was composed of a pressure vessel filled with echo gels, to which was attached a 2-mm
thick silicone sponge rubber. The needles were used to puncture the silicone sponge
rubber, and the volume of fluid that leaked from the inside to the outside of the
cyst model was measured.
Evaluation of puncture surface during and after puncture
The form of the puncture tract (i.e., a hole or flap) was evaluated using a tissue
model composed of polyimide. Penetration during puncture was evaluated by examining
video recordings.
Amounts and histology of porcine liver tissue samples
Porcine livers in ex vivo were punctured 20 times with each needle and the sample
amounts were measured. Each tissue sample was removed from the needle and deposited
on a filter paper. The weights were then measured using an electronic weighing instrument
(AS ONE CO., LTD, Osaka, Japan). Samples were also examined histologically to determine
the presence of core samples.
Ranges of needle deflection angle using an elevator device
Each puncture needle was attached to the EUS and the ranges (from minimal to maximum)
of needle deflection angles using an elevator device from no elevation to maximum
elevation were measured ([Fig. 3]).
Fig. 3 Ranges of needle deflection angles using an elevator device. Each puncture needle
was attached to the EUS, and the ranges of needle deflection angles were evaluated
with an elevator device.
Measurement of durability with deformation angle of the needle after 20 punctures
at full endoscopic angle and full elevator.
Needle durability was measured with deformation angle of the needle after 20 punctures
at full endoscopic angle and full elevator. Needle deformation was evaluated by measuring
its bending angle after puncture ([Fig. 4]).
Fig. 4 Measurement of durability with deformation angle of the needle after 20 punctures
at full endoscopic angle and full elevator. Each puncture needle was attached to the
EUS and raised to full endoscopic angle and full elevator. The maximum angle of each
puncture needle was measured, and the deformation of the needle was evaluated after
20 punctures.
Statistical analysis
Resistance, leakage, and sample amounts were expressed as mean ± standard deviation.
Continuous variables in the two groups were compared by Mann-Whitney U tests. All
statistical analyses were performed using JMP Pro version 14 (SAS Institute Inc.,
Cary, North Carolina, United States), with P <0.05 considered statistically significant.
Results
Measurement of resistance forces during puncture and removal
Expect needles had the lowest and ProCore needles had the highest maximum resistance
forces during puncture. FNB needles had significantly higher maximum resistance forces
during puncture than FNA needles (2.16±0.4 N vs. 1.71±0.48 N, P=0.028). Of the FNB needles, TopGain and SharkCore needles had significantly lower
maximum resistance forces during puncture than ProCore and Acquire needles ([Table 1]).
Table 1 Summary of results of needle evaluation with ranking of superiority.
|
EZShot3 Plus
|
Expect
|
ProCore
|
Acquire
|
SharkCore
|
TopGain
|
|
Resistance during puncture (N)
(ranking in order of excellence)
|
2.16±0.16
(4)
|
1.26±0.03
(1)
|
2.53±0.10
(6)
|
2.40± 0.11
(5)
|
1.79±0.15
(3)
|
1.65±0.04
(2)
|
|
Resistance during removal
(ranking in order of excellence) (N)
|
1.03±0.12
(5)
|
0.70±0.04
(1)
|
1.33±0.04
(6)
|
0.70±0.08
(2)
|
0.97±0.02
(3)
|
0.82±0.03
(4)
|
|
Puncture tract (shape)
|
Flap
|
Flap
|
Whole
|
Whole
|
Whole
|
Whole
|
|
Amount of tissue sampling
(ranking in order of excellence) (mg)
|
0.018±0.009
(4)
|
0.018±0.010
(5)
|
0.017±0.008
(6)
|
0.030±0.006
(2)
|
0.025±0.050
(3)
|
0.032±0.008
(1)
|
|
Range of needle deflection (from minimum to maximum)
(ranking in order of excellence)
|
10–45°
(1)
|
10–40°
(2)
|
10–30°
(4)
|
10–30°
(4)
|
10–30°
(4)
|
10–35°
(3)
|
|
Needle deformation angle
(ranking in order of excellence)
|
1°
(1)
|
16°
(3)
|
32°
(6)
|
8°
(2)
|
28°
(4)
|
31°
(5)
|
|
Leakage volume at 5Kpa
(ranking in order of excellence) (mL)
|
0
(1)
|
0
(1)
|
0.15±0.25
(3)
|
1.14±0.65
(5)
|
1.78±1.16
(6)
|
0.35±0.29
(4)
|
|
Leakage volume at 10Kpa
(ranking in order of excellence) (mL)
|
0
(1)
|
0
(1)
|
0.3±0.19
(3)
|
1.49±1.05
(5)
|
2.04±1.41
(6)
|
0.73±0.49
(4)
|
|
Pressure at the beginning of leak out (KPa)
|
–
|
–
|
10.6±14.0
|
0
|
0.2±0.45
|
0
|
ProCore needles had the highest maximum resistance force during removal. There was
no significant difference between the maximum resistance of FNB (0.95±0.25 N) needled
and those of FNA needles (0.86±0.19 N) in maximum resistance forces during removal
([Table 1]).
Needle tip abrasion damage before and after puncture
Before puncture, the tips of Acquire needles were blunter than the tips of TopGain
needles, and the needle slits at the tips were longer in Acquire than TopGain needles.
Procore needles had reverse bevels. None of the FNA/B needles experienced abrasion
damage after puncturing. Visibility markers for ProCore/EZShot3 Plus and SharkCore
needles were the dimple points and sandblasting, respectively. Visibility marker shapes
of Acquire and Expect Plus needles were the circle slit. A coil sheath was used to
visualize EZShot3 Plus needle only ([Fig. 5]).
Fig. 5 Photographs showing the tip shapes of the six needles. a Acquire needle, made of cobalt-chromium with a circle slit visibility marker and
a crown-shaped tip with three symmetric prongs. b SharkCore needle, made of stainless steel with a sandblasting visibility marker,
and having a bevel tip incorporating two sharp prongs of different lengths. c TopGain needle, made of stainless steel with a unique visibility marker and a crown-shaped
tip with three symmetric prongs. d ProCore needle, made of stainless steel with dimple points as a visibility marker
and a reverse bevel. e EZShot3 Plus needle, made of nitinol with a dimple points visibility marker and coil
sheath. f Expect needle, made of cobalt-chromium with a circle slit visibility marker.
Measurement of leakage after puncture in a mucinous cyst model.
Puncture of the cyst model with any of the FNA needles tested showed no leakage occurred
at either 5 or 10 Kpa per minute pressure. In contrast, leakage was observed with
all FNB needles. Leakage following puncture with ProCore, Acquire, SharkCore, and
TopGain needles (FNB needles) started to occur at pressures of 10.6±14.0 Kpa, 0 Kpa,
0.2±0.45 Kpa, and 0 Kpa, respectively. By contrast, no leakage following puncture
by EZShot3 Plus and Expect needles (FNA needles) was observed when the pressure was
raised from 0 to 50 Kpa per minute ([Table 1]).
Evaluation of puncture surface during and after puncture
Puncture surface was most deformed during puncture by ProCore needles ([Fig. 6] and [Video 1]). Evaluation of puncture surfaces showed that puncture sites of the FNA needles,
EZShot3 Plus (100% [5/5]) and Expect (100% [5/5]) remained in the form of a flap,
whereas the puncture sites of the FNB needles – Acquire (100% [5/5]), TopGain (100%
[5/5]), SharkCore (100% [5/5]), and ProCore (60% [3/5]) – were broken off, with holes
remaining in the puncture sites ([Table 1]) ([Fig. 7]). These holes were supplemented within the puncture needle.
Fig. 6 Evaluation of needle puncture tracts. Examination of needle puncture tracts, showing
that ProCore needles resulted in the most deformed tracts.
Evaluation of deformation in needle puncture tracts. Examination of deformation in
needle puncture tracts, showing that ProCore needles resulted in more deformed tracts
than Expect needles.Video 1
Fig. 7 Evaluation and deformation of needle puncture surfaces. The puncture sites of EZShot3
Plus and Expect needles appeared as flaps, whereas the puncture sites of Acquire,
TopGain, SharkCore, and ProCore needles were broken off, with holes remaining at each
puncture site.
Amounts and histology of porcine liver tissue samples
The amounts of samples obtained with FNB needles (0.026±0.008 mg) were significantly
greater than those obtained with FNA needles (0.018±0.009 mg) (P=0.03). The amounts of samples obtained with TopGain needles were significantly larger
than those obtained with ProCore, EZShot3 Plus, and Expect needles. The amounts of
samples obtained with Acquire needles were significantly larger than those obtained
with ProCore and EZShot3 Plus needles ([Table 1]). However, there were no significant differences between Franseen and Fork-tip needles.
Histological examination showed that core samples were obtained with ProCore, Acquire,
TopGain, and SharkCore needles (FNB needles), but not with Expect and EZShot3 Plus
needles (FNA needles) ([Fig. 8]).
Fig. 8 Histologic assessment of porcine liver tissue samples. Histological examination showed
that ProCore, Acquire, TopGain, and SharkCore needles acquired core samples, whereas
Expect and EZShot3 Plus needles did not.
Range of needle deflection angle using an elevator device
The ranges of needle motion increased from minimum to maximum in the following order:
EZShot3 Plus, Expect, ProCore, and Acquire/SharkCore/TopGain needle ([Table 1]).
Measurement of durability with deformation angle of the needle after 20 punctures
at full endoscopic angle and full elevator.
Deformation angle of needles after puncture at full endoscopic angle and full elevator
decreased in the following order: EZShot3 Plus, Acquire, Expect, SharkCore, TopGain,
and ProCore ([Table 1]).
Discussion
EUS-FNA/B is widely used in diagnosis of digestive diseases, with many types of needles
available in clinical practice. Choice of needle depends on the preference of individual
operators. Few reports to date have utilized experimental methods to objectively evaluate
needle performance [10]
[11]
[12]. The present experimental study objectively evaluated performance of six types of
needles under the same objective conditions.
Evaluation of maximum resistance forces during puncture showed that Expect needles
had significantly lower resistance, whereas ProCore needles had the highest resistance.
Among FNB needles, SharkCore and TopGain had significantly lower maximum resistance
forces during puncture than ProCore and Acquire needles. Therefore, if another needle
cannot penetrate a tumor or be advanced into a tumor due to its deep position, Expect
needles, with the lowest maximum resistance force during puncture, or SharkCore or
TopGain needles, with the lowest maximum resistance forces during puncture among FNB
needles, may be selected. A comparison of Franseen needles showed that Acquire needles
had significantly higher maximum resistance forces during puncture than TopGain needles.
These differences in resistance forces during puncture may be due to differences in
the degree of polishing of the tip (sharpness) and/or the length of the needle slit.
The reverse bevel of ProCore needles provided more resistance.
Evaluation of maximum resistance forces during needle removal showed that ProCore
needles had the highest and Acquire/Expect needles had the lowest resistance. These
differences may be due to differences in visibility markers. The visibility markers
for the two needles with the highest resistance during removal were dimple points,
whereas those for two needles with the lowest resistance were circle slits. ProCore
needles had the highest resistance during removal because the combination of reverse
bevels and dimple points increased resistance. Expect and Acquire needles, therefore,
should be selected when mobility within a tumor is poor.
FNA needles showed no leakage of fluid, perhaps because a flap remained at the puncture
site after needle removal. By contrast, all FNB needles tested showed leakage because
their puncture sites were broken off and holes remained at these sites. FNA needles
should be used to puncture cystic lesions for diagnosis and treatment, because these
needles prevented leakage of cystic fluid. Because the holes remaining after puncture
with FNB needles were filled by the collected tissue, FNB needles are recommended
for collecting tissue samples from solid lesions.
Regarding puncture surface, puncture sites of the FNA needles remained in the form
of a flap after needle removal, whereas puncture sites of the FNB needles were broken
off and a hole remained in the puncture site. In fact, FNB needles allowed more and
larger core samples compared with FNA needles. In tissue sampling, FNB needles were
superior to FNA needles. Yousri M et al. reported that FNB needles were better at
obtaining adequate tissue cores than FNA needles [13]. Moreover, Kovacevic B et al. reported that mean total tissue and mean diagnostic
tissue areas for FNB needles were 6-fold larger than those for FNA needles [14]. In terms of FNB needles, the Franseen needle was superior to the reverse bevel
needle. However, there was no significant difference between Franseen and Fork-tip
needles. Therefore, Franseen and Fork-tip needles are recommended when a large amount
of tissue samples is required for gene panel testing.
Assessments of performance during puncture showed that EZShot3 Plus needles had the
largest range of needle deflection using an elevator device and the lowest durability
in terms of deformation angle of the needle after 20 punctures at full endoscopic
angle and full elevator. These differences may be ascribable to the material from
which these needles are made. EZShot3 Plus needles are composed of nitinol and Acquire
and Expect needles of cobalt-chromium, whereas ProCore, SharkCore, and TopGain needles
are composed of stainless steel. These findings suggest that needles made of nitinol
have the highest durability and lowest degree of deformation.
EZShot3 Plus needles, which contain coil sheaths, had the largest range of deflection
angle ranges using an elevator device and superior mobility. Therefore, interventional
EUS with EZShot3 Plus needles may be better for drainage and small lesions because
their puncture performance is superior, including lower deflection angles and degree
of deformation.
This study had several limitations. First, it was experimental with a small sample
size. In addition, the experimental setting may be different from ordinary clinical
practice. However, it is difficult to compare performance of needles under the same
objective conditions. Experimental comparisons as in the present study enable determination
of objective differences in standardized settings.
Concusions
In conclusion, the present study found that the performances of the six needles differed
in various aspects. Understanding the characteristics of individual needles may allow
for selection of a needle appropriate for each situation
