Introduction
Endoscopic ultrasound-guided fine-needle aspiration (EUS-FNA) is a useful technique
to obtain specimens for histopathologic examination. However, there is a small risk
of tumor cell seeding along the needle track or within the peritoneum caused by EUS-FNA
[1]
[2]. We report needle track seeding following EUS-FNA in a patient with a mediastinal
tumor.
Case Report
The patient was a 53-year-old man who presented with hoarseness. Chest computed tomography
(CT) revealed a 6-cm mass in the mediastinum ([Fig. 1]). EUS showed a heterogeneous, septated tumor. Transesophageal EUS-FNA was performed
( [Fig. 2a, b]). Three passes were made with a 22-gauge needle (Olympus Medical Systems, Tokyo,
Japan). During the second pass, necrotic tissue was obtained. No immediate adverse
events developed after the EUS-FNA. Pathology showed carcinoma with embryonal features
([Fig. 2c, d, e]).
Fig. 1 Computed tomography (CT) scan findings. a, b CT scan revealed a 6-cm mediastinal mass. c, d Four weeks after endoscopic ultrasound-guided fine-needle aspiration, repeat CT scan
revealed a mass in the esophageal wall.
Fig. 2 Endoscopic ultrasound findings and pathologic findings of biopsy specimens. a Endoscopic ultrasound showed a tumor with a heterogeneous appearance and a septum.
b Endoscopic ultrasound-guided fine-needle aspiration was performed with a 22-gauge
needle. c The biopsy specimen from the tumor showed tumor cells and necrotic tissue. (hematoxylin-eosin, × 4).
d Tumor cells show a papillary pattern. Nuclei are overlaid and cell polarity is lost.
e Cells stain positive for AFP ( × 4).
Two weeks after the EUS-FNA, the patient presented with mediastinitis. Symptoms improved
with medical therapy. Two weeks later, a repeat CT scan showed a mass in the esophageal
wall. Upper gastrointestinal endoscopy showed a protruding lesion in the middle esophagus
([Fig. 3]), which was biopsied using forceps and confirmed to be histologically similar to
the mediastinal mass. This was believed to be consistent with needle tract seeding
from the EUS-FNA. Tumor resection was not performed because the mass was felt to be
technically unresectable, in part due to the tumor seeding. Chemotherapy was administered
and the patient died less than 2 years later.
Fig. 3 Endoscopic findings. a Upper gastrointestinal endoscopy showed a mass effect in the middle esophagus. b Four weeks after endoscopic ultrasound-guided fine-needle aspiration, a protruding
lesion was seen in the middle esophagus
Discussion
This case underscores the potential risk of needle tract seeding after EUS-FNA. In
this patient, the location of tumor seeding corresponded to the entry point of the
EUS-FNA. The endoscopic unusual appearance of mushroom-shaped tumor protruding from
the FNA puncture site was suggestive of seeding as spontaneous direct tumor invasion
would be expected to appear as a flat elevation. In addition, the tumor became mucosally
based. We believe this is the reason for the rapid development of seeding. The tumor
seeding also might have been associated with the post-FNA mediastinitis.
The literature from 2003 to 2016 contains reports on only 14 previous patients with
needle tract
seeding to gastrointestinal tract wall by a malignancy following EUS-FNA ( [Table 1]). Needle size, number of passes, needle movement during puncture, suction, and characteristics
of the tumor, might be factors in tumor seeding [3]. According to previous reports, the number of needle passes and tumor characteristics
(poorly differentiated or cystic tumor) are considered to be risk factors [2]. Interestingly, needle size was not associated with seeding. In this patient, we
speculate that multiple needle passes and tumor characteristics (a poorly differentiated
and cystic tumor) may have contributed to development of seeding. Although EUS showed
the lesion to be solid, the aspirate showed an abundance of necrotic tissue. We speculated
the tumor had a fluid component similar to a cystic tumor.
Table 1
Summary of previous reports of endoscopic ultrasound-guided fine-needle aspiration
seeding.
No.
|
Author year
|
Aspiration site
|
Pathological diagnosis
|
Seeding site
|
Needle size
|
Passes
|
Interval from EUS-FNA until identification of seeding
|
1
|
Hirooka, 2003
|
Pancreatic body mass
|
Adenocarcinoma (IPMC)
|
Posterior gastric wall (and peritoneal seeding)
|
22G
|
3
|
10 days
|
2
|
Shah, 2004
|
Perigastric lymph node
|
Melanoma
|
Posterior gastric wall
|
22G
|
1
|
6 months
|
3
|
Paquin, 2005
|
Pancreatic tail cancer
|
Adenocarcinoma
|
Posterior gastric wall
|
22G
|
5
|
16 months
|
4
|
Doi, 2008
|
Lymph node metastasis of gastric cancer
|
Adenocarcinoma
|
Esophageal wall
|
19G
|
1
|
18 months
|
5
|
Ashmed, 2011
|
Cystic mass in the pancreatic body
|
Adenocarcinoma
|
Gastric wall of antrum and body
|
Unknown
|
Multiple
|
4 years
|
6
|
Chong, 2011
|
Pancreatic tail cancer
|
Adenocarcinoma
|
Posterior gastric wall of body
|
22G
|
2
|
2 years
|
7
|
Katanuma, 2012
|
Pancreatic body cancer
|
Adenocarcinoma
|
Posterior gastric wall of upper body
|
22G
|
4
|
22 months
|
8
|
Anderson, 2013
|
Celiac lymph node (and pancreatic head mass)
|
Adenocarcinoma
|
Gastroesophageal junction
|
Unknown
|
Unknown
|
Unknown
|
9
|
Minaga, 2015
|
Pancreatic body cancer
|
Adenocarcinoma
|
Posterior gastric wall
|
22G
|
3
|
8 months
|
10
|
Sakurada, 2015
|
Pancreatic body cancer
|
Adenosquamous carcinoma
|
Posterior gastric wall
|
22G
|
Unknown
|
19 months
|
11
|
Tomonari, 2015
|
Pancreatic body cancer
|
Adenocarcinoma
|
Posterior gastric wall
|
22G
|
2
|
8 months
|
12
|
Iida, 2016
|
Pancreatic body cancer
|
Adenocarcinoma
|
Posterior gastric wall
|
22G
|
3
|
6 months
|
13
|
Kita, 2016
|
Pancreatic body-tail cancer
|
Adenocarcinoma
|
Posterior gastric wall
|
22G
|
2
|
7 months
|
14
|
Yamabe, 2016
|
Pancreatic mass beside cystic lesions
|
Adenocarcinoma (IPMC)
|
Posterior gastric wall
|
25G
|
Unknown
|
3 months
|
15
|
Our Case
|
Mediastinal tumor
|
Embryonal adenocarcinoma
|
Esophageal wall
|
22G
|
3
|
4 weeks
|
IPMC, intraductal papillary mucinous adenocarcinoma
Peritoneal dissemination has been reported more frequently than needle tract seeding.
The exact etiology is unknown but once a patient suffers needle tract seeding, the
prognosis is worse.
Conclusion
In conclusion, in patients who have lesions that are surgically resectable for curative
intent, we must carefully consider appropriate indications for performing EUS-FNA
and inform these individuals about the potential for esophageal seeding, which is
rare.