Introduction
Introduction
The advantage of endoscopic ultrasonography (EUS) is its high resolution and capability
of local observation [1]
[2]. It is one of the most accurate diagnostic methods, particularly for the pancreatobiliary
region, where various advanced imaging diagnostic methods available today, such as
CT and MRI [3]
[4]. Since EUS is most useful for diagnosing a small lesion, its usage for early diagnosis
of pancreatobiliary cancer has being anticipated.
Endoscopic ultrasound-guided fine-needle aspiration biopsy (EUS-FNA) was developed
to enhance the diagnostic capability of EUS by providing additional pathological findings
[5]
[6]
[7].
However, it is not clearly known yet whether EUS is actually useful for the diagnosis
of small pancreatic tumor, whether EUS-FNA is indispensable when a small tumor is
detected. It is also necessary to understand for what size of tumor EUS-FNA can be
used, and if there is no risk of complication or not. This time, I introduce the present
state of diagnosing small pancreatic tumors, the application of EUS-FNA, and the discussion
of the points to be solved.
Detection of small pancreatic tumors
Detection of small pancreatic tumors
It is well known that pancreatic cancer is hard to cure and is often fatal. Since
it presents no characteristic symptoms, and there is no effective medical check-up
for the disease, it is difficult to diagnose it at an early stage. Thus how to detect
a small pancreatic cancer has been a long standing question. With the rapid advancement
in less invasive imaging diagnostic methods such as CT and MRI, it has become fairly
easy in recent years to indirectly find abnormalities such as dilatation of the pancreatic
duct, presence of pancreatic cyst, or dilatation of the bile duct. Nevertheless, there
is limitation in diagnosing small pancreatic tumor by those methods.
Table 1 shows a comparison of US, CT and EUS in the ability to detect pancreatic carcinomas
to be resected in our institute. Detection of tumors, particularly those smaller than
2 cm (TS1), was possible by EUS in most cases, except in the case of carcinoma in
situ. However, detection by US was successful in 52 % of the cases and by CT in 43
% (Fig. [1]) [4]. EUS was also helpful for detecting two different tumors located in one area (Fig.
[2]). From these results, it would be a key to detecting a small pancreatic cancer to
assemble indirect findings obtained by less invasive imaging examinations, and then
resort to EUS [8].
Fig. 1 A case of small pancreatic carcinoma (T1). (a) CT revealed the dilated main pancreatic duct in the pancreatic tail; however, it
did not show the pancreatic tumor. (b) EUS revealed a low echoic mass.
Fig. 2 A case of pancreatic cancer concomitant with endocrine tumor. (a) CT revealed a strongly enhanced area (arrow) in the pancreatic head. (b) By EUS, two tumorous images were shown. The one close to the papilla was a homogeneous
echoic mass (arrow), which was suspected to be an endocrine tumor. The other on the
pancreatic body side presented an irregular periphery (arrow head), and was a hypoechoic
and hetrogenious mass; it was diagnosed as pancreatic cancer.
Table 1 Detection rate of pancreatic carcinoma by each imaging modality
|
US |
CT |
EUS |
| TS1 |
11/21 (52.4 %) |
9/21 (42.8 %) |
20/21 (95.2 %) |
| TS2 |
42/54 (77.8 %) |
45/54 (83.3 %) |
50/51 (98.0 %) |
| TS3 |
23/24 |
23/24 |
20/20 |
| TS4 |
4/4 |
4/4 |
4/4 |
Differential diagnosis of small pancreatic lesions
Differential diagnosis of small pancreatic lesions
A number of reports have already shown the utility of EUS in the differential diagnosis
of pancreatic tumors [9]
[10]
[11]. It is suitable for the diagnosis of cystic tumors, particularly the qualitative
diagnosis of solid tumors presenting cystic changes (Fig. [3]). Although EUS is useful in the differential diagnosis of tumors, we cannot differentiate
cancerous fibrosis from inflammatory fibrosis only by the B-mode images of ultrasound.
Therefore, in some cases, additional information by other modalities is required.
For instance, to evaluate the vascularity of tumors is very important to differentiate
pancreatic cancer from inflammatory changes. The common type of pancreatic carcinoma
is usually hypovascular, in contrast to an inflammatory pancreatic mass, which is
isovascular. Enhanced CT/MRI and contrast-enhanced US are useful for the differential
diagnosis [3]
[4]
[12]. Despite the advancement in diagnostic imaging techniques, however, differentiation
between pancreatic cancer and focal pancreatitis remains difficult, especially in
a case of a small lesion (Fig. [4]). Therefore it is sometimes required to have a tissue diagnosis. There are two methods
for obtaining tissue samples from a pancreatic mass, transpapillary biopsy and/or
cytology, and EUS-FNA [13]
[14]
[15].
Fig. 3 A case of endocrine tumor with cystic changes. (a) CT showed a tumor-like cystic mass in the body of the pancreas. (b) EUS visualized the internal components more precisely. This finding by EUS helped
us to diagnose it as endocrine tumor with cystic changes.
Fig. 4 A case of focal pancreatits which was difficult to differentiate from small pancreatic
carcinoma. (a) ERCP demonstrated a stricture of the main pancreatic duct in the pancreatic body.
(b) EUS revealed a low echoic small mass measuring 6 mm.
Transpapillary biopsy and/or cytology vs. EUS-FNA
Transpapillary biopsy and/or cytology vs. EUS-FNA
Both methods have some advantages and disadvantages (Table 2). The advantages of transpapillary biopsy and/or cytology are being free from the
risk of cancer cells dissemination and suitable for obtaining tissue samples from
carcinoma in situ [16]. The disadvantages arise in the difficulty to obtain tissue samples from the branch
duct of the pancreas, risk of ERCP-related pancreatitis, and that the techniques are
sometimes difficult and the sensitivity is less than what we expected. In contrast,
the advantages of EUS-FNA are being free from the ERCP-related pancreatitis and suitable
for obtaining tissue samples even if cancer has not invaded the main pancreatic duct.
The disadvantages are the risk of dissemination of cancer cells after EUS-FNA and
the difficulty to obtain tissue samples from the carcinoma in situ; and it is not
clear yet on what size of tumor EUS-FNA can be applied to.
Table 2 Transpapillary biopsy/cytology versus EUS-FNA
| Transpapillary approaches |
EUS-FNA |
| Advantage |
|
- No risk of dissemination
- Suitable for obtaining samples from MPD |
- Free of ERCP-related pancreatitis
- Possible to obtain samples even if the MPD is not invated |
| Disadvantage |
|
- Limitation in obtaining samples from the branch duct
- Risk of ERCP-related pancreatitis
- Technical difficulty
- Low accuracy |
- Risk of dissemination
- Limitation in obtaining samples from the carcinoma in situ or small pancreatic mass? |
Preoperative histological diagnosis of small pancreatic
carcinoma
Preoperative histological diagnosis of small pancreatic
carcinoma
There are very few reports of histologically diagnosed small pancreatic carcinoma
preoperatively. Yamao et al. [17] reported the results of biopsies by ERCP and EUS-FNA in operated patients with small
pancreatic cancer less than 2 cm in diameter (TS1); the cancer-positive rate was 50
%(4/8) in the use of ERCP and 100 %(2/2) in the use of EUS-FNA. We have investigated
the results of transpapillary biopsy/cytology in our operated patients with TS1; the
cancer-positive rate was 56 %(10/18). We have not performed EUS-FNA on patients with
TS1 because we are very concerned with dissemination of malignant cells after EUS-FNA.
Indication of EUS-FNA for small pancreatic mass lesions
Indication of EUS-FNA for small pancreatic mass lesions
A fundamental principle in establishing indication for EUS-FNA is the determination
as to whether or not the information obtained has the potential to affect patient
management [18]. According to this principle, EUS-FNA is currently indicated for pancreatic mass
lesions for (1) histological evidence required for chemotherapy and/or radiation therapy,
(2) differential diagnosis between localized pancreatitis and pancreatic cancer which
shows an atypical imaging pattern and/or is negative for biopsy/cytology by transpapillary
approach, and (3) tumor staging in a patient with a small amount of ascitis or lymph
node swelling [17]. However, the role of EUS-FNA in patients suspected of having pancreatic cancer
that appears to be respectable on imaging studies especially small lesion is controversial.
One view is that a tissue diagnosis will not alter patient management, and is therefore
unnecessary. This is because as with EUS-FNA for pancreatic masses have a low negative
predictive value for malignancy. Thus, FNA negative for cancer will not exclude the
suspected diagnosis of cancer and the patient will be explored anyway. In addition,
the risk of tumor seeding by EUS-FNA is strongly stressed in the opinions against
the indication of this approach, especially in Japan [17]. It has not fully been determined whether tumor seeding occurs with EUS-FNA. There
have been three reports of seeding possibly caused by EUS-FNA [18]
[19]
[20]
[21]. Paquin et al. [21] reported of tumor seeding after EUS-FNA in patients with pancreatic cancer even
though the tumor size was 8 mm.
Theoretically tumor seeding may occur more frequently in cystic lesions than in solid
lesions [17]
[18]
[19], and the complication from EUS-FNA appears to have a greater risk of infections
in cystic lesions; the incidence rate is 14 % compared to 0.5 % in solid masses. For
this reason, indication or contraindication of EUS-FNA for pancreatic cystic lesions
suspected of being neoplastic cystic tumors is a matter of controversy in Japan, which
explains why EUS-FNA has actively been performed for such lesions in the United States
and Europe [22]
[23], but not in Japan. In view of the above, it will be our future task to further the
study of tumor seeding and to devise a reliable method of tissue sampling in the use
of EUS-FNA.
