How to perform EUS-guided tattooing?

Recently, we introduced a series of papers describing on how to perform certain techniques and controversies in EUS. In the first paper, "What should be known before performing EUS examinations, Part I," the authors discussed clinical information and whether other imaging modalities should be needed before embarking in EUS examination. In Part II, some technical controversies on how EUS is performed are discussed from different points of view by providing the relevant available evidence. Herewith, we describe on how to perform EUS-guided fine needle tattooing (FNT) in daily practice. The aim of this paper is to discuss pros and cons for several issues including historical remarks, injecting material, technical approach, and how to perform EUS-FNT including argues in favor and against.

INTRODUCTION

Recently, we introduced a series of papers describing on how to perform certain techniques,[12345] controversies in EUS, and “what should be known before performing EUS examinations.”

In Part I, the authors discussed clinical information and whether other imaging modalities should be needed before embarking in EUS examination.[6] The content includes the “nihilistic” or “puristic” approach, “I need nothing before EUS” versus the clinical approach (“Performing EUS as a clinician, I prefer to review as much clinical data as possible before EUS”. Should transcutaneous ultrasound and EUS be performed by the same operator? The use of esophagogastroduodenoscopy [EGD] before EUS. Should consenting for EUS differ from consenting for EGD? The utility of coagulation tests before EUS and interventions).

In Part II, some technical controversies on how EUS is performed are discussed from different points of view by providing the relevant available evidence.[7] Does equipment design influence the complication rate? Should we have a standardized screen orientation? Radial versus longitudinal (linear) echoendoscopes. Should we search for incidental findings using EUS?

Herewith, we describe on how to perform EUS-guided fine needle tattooing (EUS-FNT) in daily practice. The aim of this paper is to discuss pros and cons for several issues including historical remarks, injecting material, technical approach, and how to perform EUS-FNT including argues in favor and against.

The authors declare that this paper is not intended as a guideline, but rather an opportunity to document the current practice, allowing readers to evaluate their own EUS procedures and to stimulate further discussion.

EUS-GUIDED TATTOOING

Historical remarks

Tattooing has been widely used in luminal endoscopy by injecting India ink or sterile carbon particles into the submucosa of the gastrointestinal tract adjacent to neoplastic lesions, to favor their recognition during subsequent surgery or to facilitate the identification of prior endoscopically resected areas during follow-up endoscopic examinations. In 2002, the utilization of EUS-FNT to help recognition of a pancreatic lesion during surgery has been successfully applied for theLtime by Gress et al.[8] The authors performed preoperative EUS-FNT of a 19 mm × 5 mm insulinoma localized at the pancreatic body–tail junction, in between the splenic vein and the splenic artery. Four milliliters of diluted, filtered, and presterilized India ink solution (Permark Inc., Edison, NJ, USA) was injected at the center of the lesion utilizing a standard 22-gauge fine needle aspiration (FNA) needle[8] This procedure was performed at the surgeon's request to obtain the best possible preoperative localization in an attempt to allow safe performance of lesion enucleation. EUS-FNT was well tolerated by the patient, the lesion could be visualized at laparotomy performed the same day, but unfortunately, enucleation could not be done and distal pancreatectomy with splenectomy was finally performed.

The increasing use of the laparoscopic approach to surgically resect pancreatic body and tail lesions, in particular, small solid and cystic lesions with high-risk features, has intensified the need for a very precise preoperative localization of the target lesion. Indeed, loss of capability to palpate the pancreas may impair tumor recognition or proper assessment of its position, thus impairing surgical resection with tumor-free margins or conservation of as much viable pancreatic tissue as possible. Moreover, the frequently confusing appearance of the surrounding retroperitoneal fat can render intraoperative ultrasound localization of pancreatic lesions difficult (possible in only 60%–90% of the cases).[91011] In one case series, intraoperative identification of the lesion failed in four out of 61 patients (6.5%), for whom re-operation was needed.[12] Thus, a role for EUS-FNT was foreseen particularly for small tumors, located deeply, or in cases in which the resection was deemed problematic because of the close relationship to local vessels or vicinity to the pancreatic duct.

Review of the literature

Table 1 summarizes all case reports and case series describing the use of EUS-FNT up to now.[81314151617181920] Different techniques and solutions were utilized. After few adverse events (AEs) associated with the use of undiluted unproperly sterilized India ink solution were described in the literature,[21] Ashida et al. utilized diluted indocyanine green that was injected using EUS-FNT into the pancreatic capsule located just above a 5-mm tumor in the pancreatic tail.[14] The area was easily recognized by the surgeons at open laparotomy performed the following day, whereas the tumor could not be found by palpation or intraoperative ultrasound.[14] In another report of a small insulinoma in the pancreatic tail, EUS-FNT by injecting methylene blue was successfully performed after the tumor could not be identified by intraoperative ultrasound at initial laparotomy.[13]

Table 1

Summary of the studies evaluating EUS-guided preoperative marking to guide surgical resection

Author, year	Diagnosis& (n)	Number of patients	Number of lesions	Lesion size (median)	Needle gauge	Injected solution	Quantity (median)	Intra-operative recognition	Adverse events
Gress, 2002CR	Insulinoma	1	1	19 mm×5 mm	22	India ink	4 mL	Yes	None
Zografos, 2005CR	Insulinoma	1	1	9.8 mm×8.2 mm	NR	Methylene blue	1 mL	Yes	None
Ashida, 2006CR	NR	1	2	5 mm	22	Diluted indo-cyanine green	0.5 mL	Yes	None
Farrell, 2009CR	IPMN with carcinoma in situ	1	1	5 mm	22	GI Spot®	2 mL	Yes	None
Lennon, 2010RS	PanNEN (6); MCN (2); PDAC (1); IPMN (1); serous cystadenoma (2); epidermoid cyst (1)	13$	13	8-50 mm (13 mm)	22	GI Spot®	1.25-5 mL (3 mL)	100%	None
Newman, 2010RS	PanNEN (5); MCN (2); PDAC (2); Serous cystadenoma (1)	10$	10	8-28 mm (13 mm)	22	GI Spot®	2-4 mL	100%	None
Rodriguez, 2011CR	Insulinoma	1	1	9 mm	NR	GI Spot®	NR	Yes	None
Leelasinjaroen, 2014CR	Insulinoma	1	1	15.5 mm	22	GI Spot®	2 mL	Yes	None
Okuzono, 2016RS	PanNEN (1); IPMN (3); SPN (1); PDAC (1)	6	6	7-35 mm (16 mm)	25	Sodium hyaluronate and India ink	0.06-0.1 mL (0.08 mL)	5/6 (83%)	None

&All lesions were located in the pancreas, $It is possible that some or most of the patients might have been reported in both studies. CR: Case report; RS: Retrospective study; NR: Not reported; IPMN: Intraductal papillary mucinous neoplasm; PanNEN: Pancreatic neuroendocrine neoplasm; MCN: Mucinous cystic neoplasm; PDAC: Pancreatic ductal adenocarcinoma; SPN: Solid pseudopapillary neoplasm.

Farrell and colleagues were the first to report the use of the sterile carbon-based Spot® solution (GI Supply, Camp Hill, PA, USA) for EUS-FNT of a very small pancreatic adenocarcinoma located at the level of the pancreatic body, close to the portal vein confluence, to guide laparoscopic resection.[15] The sterile carbon-based solution was injected into an area of normal pancreatic parenchyma located toward the pancreatic head close to the lesion. Tattooing was, in this case also, the only method to identify the tumor, which made possible a safe and uneventful laparoscopic distal pancreatic resection.

In the first large retrospective study on 30 patients with pancreatic body/tail lesions, 13 underwent preoperative EUS-FNT using a sterile carbon-based solution injected in 0.5-mL increments into the normal pancreatic parenchyma, a few millimeters to the right side of the lesion.[16] The indications for EUS-FNT were either the small size of the lesion to increase the chance for its intraoperative localization or the presence of indistinct borders in cases of larger lesions to increase the possibility of achieving negative resection margins. The pathologic diagnoses were similar between the groups with and without preoperative EUS-FNT, but the lesion size was significantly smaller in the former (median 1.3 cm vs. 4.0 cm, P = 0.03).[16] Importantly, all 13 tattoos were clearly identified intraoperatively despite the mean interval of 20 days (range 1–69) between EUS-FNT and surgical laparoscopic resection, and all these 13 patients had negative surgical resection margins. None of the patients in the EUS-FNT group needed conversion to an open surgery, with the advantage of a significantly shorter mean operative time for laparoscopic distal pancreatectomy in the EUS-FNT group as compared to no EUS-FNT (128.5 min vs. 180 min, P = 0.02), with no difference of postoperative AEs between the two groups.[16]

Leelasinjaroen et al. subsequently reported for the first time a case of a deep pancreatic head insulinoma that underwent EUS-FNT to guide its localization and surgical enucleation, which allowed the preservation of the pancreatic duct located posterior to the lesion and avoided pancreaticoduodenectomy.[19]

Finally, in a retrospective case series of six patients (two of them with lesions located in the pancreatic head), the authors reported the use of a newly developed tool allowing for injection of minuscule amounts of solution.[20] They used a maximum of 0.1 mL of a viscous solution comprising 4:1 sodium hyaluronate and India ink that was injected while slowly withdrawing the needle, to keep it in a strictly limited area, thus avoiding its leak into the peritoneal cavity or retroperitoneal space. In one of these patients (17%), however, the tattoo could not be identified at surgery, mostly related to the fact that the tattooing solution was injected too deep into the pancreatic parenchyma, which did not allow its recognition on the pancreatic surface.

Injecting material

EUS-FNT has been performed injecting several different types of solutions: India ink, indocyanine green, methylene blue, and GI Spot® (GI Supply, Camp Hill, PA, USA),[8131415] which have to be sterile to avoid infectious AEs. The initial solution that was used for EUS-FNT was India ink, which is composed of a variety of fine soot (impure carbon particles resulting from the incomplete combustion of hydrocarbons) in colloidal suspension in water to form a liquid.[22] However, since some reports of AEs have been described during its use in endoscopy,[21] it has been largely replaced by other solutions. The most frequently utilized is the GI Spot®, a purified and sterilized preparation of carbon black, which is internalized by macrophages without determining any inflammation in surrounding tissues.[20] The advantage of this solution is related to its carbon particle content, which is nondegradable and remains into the tissues indefinitely, thus creating a permanent tattoo. Differently, when indocyanine green or methylene blue dye is used, EUS-FNT has to be performed time-wise as close as possible to surgery, since these dyes are water soluble and cleared by the organism's cleaning mechanisms.

Technical approach

A standard EUS-FNA needle should be used for performing EUS-FNT, which should be of a sufficient size to allow easy injection of the tattooing solution but avoid its dispersion (the 22-gauge has been used in most cases). Before inserting it into the working channel of the echoendoscope, the needle should be preloaded with the solution to eliminate all the air from the needle lumen that, once injected, would obscure the ultrasonographic view. The needle is then inserted inside the target lesion where a small amount of the tattooing solution should be slowly injected inside the lesion or immediately near the tumor borders into the normal parenchyma. The process is done while slowly withdrawing the needle to allow for a small quantity of the ink to be also placed in the subcapsular pancreatic space. Care must be taken not to overinject the tattooing solution since the compound can migrate into the peritoneum or the retroperitoneal space and create major discoloration, which could have a negative impact on the intraoperative lesion localization and consequently surgical resection. For pancreatic body or tail lesions, injection should be performed in the nearby pancreatic parenchyma proximal to the lesion toward the pancreatic head [Figure 1].

Figure 1

Representative diagram on how to perform EUS-guided tattooing of a pancreatic lesion: (a) Marking of an area of normal pancreatic parenchyma located to the right of a pancreatic body/tail lesion as a landmark for limiting the pancreatic resection with tumor-free margins; (b) Tattooing of a pancreatic head lesion, when the marker is usually injected inside the lesion

How to perform EUS-guided fine needle tattooing

Step by step

Once preloaded with the solution to be injected [Figure 2], the needle is inserted into the channel of the echoendoscope. The needle tip should then be advanced into the target lesion or the adjacent parenchyma using a thrust movement as it is done for EUS-FNA. A small amount of solution should be injected until a hyperechoic area appears in the endosonographic image at the level of the needle tip. For lesions that are tattooed inside, injection can be repeated in other spots, if needed, in the same endoscopic session. Injection should be done by slowly retracting the needle while continuing injecting, to leave an inked tract and induce discoloration of the subcapsular pancreatic tissues to help surgeons identify the marked spot. Particular attention should be given to inject only anterior to the Wirsung duct and to avoid injecting outside the gastrointestinal tract into the peritoneal/retroperitoneal space.

Figure 2

EUS-guided tattooing of a small pancreatic insulinoma before surgical resection: (a) Priming of the 22-gauge EUS-FNA needle with the marking solution (Spot®; GI supply, Camp Hill, PA, USA); (b) EUS-guided puncture of the lesion; (c) hyperechoic spot appearing at the tip of the needle by injection of the sterile carbon-based ink initially inside the lesion, injection which is continued while withdrawing the needle; (d) resection specimen harboring the black discoloration determined by tattooing; its appearance at the surface of the lesion helped the surgeon find it intraoperatively and resect it laparoscopically

Routine pre-EUS coagulation testing and management of anticoagulants/antiplatelet drugs

Prothrombin time, INR, and activated partial thromboplastin time should be checked before the procedure equally to a standard EUS-FNA/FNB procedure. Withholding and restarting antiplatelet or anticoagulant treatment periprocedurally should follow the guidelines on the management on anticoagulants/antiplatelet agents for endoscopic procedures, similar to that described for EUS-FNA.[2324]

Antibiotic prophylaxis

Once a foreign solution is injected from a nonsterile organ (the upper gastrointestinal tract) into an extraparietal space, broad-spectrum antibiotics, such as ciprofloxacin, second-or third-generation cephalosporins, or clindamycin, should be used prophylactically just before and for 24–72 h after the procedure,[1516] even though no solid evidence is available supporting this suggestion.

Clear indications, the clinical approach

Pancreatic indications

The procedure is mostly indicated to aid in the intraoperative localization of small pancreatic tumors, especially when the laparoscopic approach is used. In this case, the surgeon loses the capability of palpating the organ and of finding small lesions; thus, EUS marking is used to create a recognizable spot that can be seen through the serosa covering the pancreatic parenchyma.

Possible other indications

Other possible indications could be marking of subepithelial lesions of the gastrointestinal tract (such as gastrointestinal stromal tumors), with exophytic development for surgical identification and resection, or marking of lymph nodes for intraoperative identification and harvest. However, none of these possible indications has been reported so far.

Pros

The technique is simple, easy to perform even during the same session of EUS-guided tissue acquisition, and cheap. Using carbon-based solutions, EUS-FNT can be performed at the time of initial EUS evaluation, without any concern about the time span between the tattooing procedure and the surgical intervention, thus avoiding the costs and risks associated with repeating the EUS-FNT procedure.[16]

Preoperative marking of small tumors is of great value in assisting in their localization during surgery, particularly during the laparoscopic approach.[25] Indeed, up to two-thirds of pancreatic insulinomas measure <2 cm, and up to a quarter of them have a diameter of <1 cm.[8] Because of these characteristics, up to one-fifth of insulinomas cannot be localized intraoperatively by palpation only,[18] in particular, if located in the pancreatic head.[12] Thus, EUS is not only superior to other methods in the identification of pancreatic insulinomas and of other pancreatic neuroendocrine neoplasms, when preoperative transabdominal ultrasound, computed tomography, and MRI scans failed,[26272829] but also provides a method for easier intraoperative recognition by performing EUS-FNT. In other cases, EUS-FNT should be used as a landmark for performing limited surgical resection (by laparoscopy) or tumor enucleation,[15] since preservation of pancreatic parenchyma is of utmost importance to decrease the risk for long-term endocrine and exocrine pancreatic insufficiency, especially after distal pancreatectomy, when new-onset diabetes mellitus is observed in from 8% to 23% of patients.[3031] Moreover, utilization of EUS-FNT for preoperative delineation of small pancreatic tumors results in a decreased operative time and in lower re-operation rates, due to the inappropriate primary resection of the lesion.[17]

Cons

There are limited studies that have evaluated the performance of EUS-FNT in terms of applicability and clinical effectiveness in a meaningful number of patients and with proper study design. The limited experience accumulated so far suggests that the use of the above-mentioned solutions for marking, when utilized in a proper manner, is not associated with significant AEs. However, although the clinically apparent AEs are so far nil in the reported cases, in two of the patients, mild local pancreatic inflammatory changes were noted at surgery performed very soon after EUS-FNT.[1617] Thus, since EUS-FNA of pancreatic lesions is associated with a small risk of acute pancreatitis (0.9%) and infection (0.9%),[32] and India ink colonic endoscopic tattooing has been reported to rarely cause infectious complications (0.22%),[3334] one should expect a small rate of AEs associated with this procedure. Moreover, the choice of the proper marking solution and the quantity needed for tattooing have not been clarified yet.

CONCLUSIONS

EUS-FNT is a safe and effective procedure to provide intraoperative identification of small pancreatic lesions, with the potential to increase their recognition, reduce operative time by helping surgeons to plan in advance pancreatic resections, achieve negative resection margins, and spare normal pancreatic parenchyma. Large prospective multicenter randomized studies are needed to clearly demonstrate the role of EUS-FNT in all patients undergoing surgical resection of small pancreatic lesions, as well as to assess its efficacy in reducing the incidence of exocrine/endocrine postoperative pancreatic insufficiency. Other applications for EUS-FNT are awaited.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.