The clinical evaluation of needle-based confocal laser endomicroscopy in the assessment of pancreatic cystic lesion: A pilot study

Abstract

Background/aim:

Pancreatic cystic lesions are increasingly diagnosed from cross-sectional imaging done for other indications. The challenge lies in the ability to correctly identify the high-risk individuals for resection surgery, which carries high morbidity and mortality. Unfortunately, present diagnostic techniques are suboptimal. Needle-based confocal laser endomicroscopy (nCLE) has been designed to bridge this diagnostic gap. We aim to assess the feasibility and safety of nCLE in the assessment of pancreatic cystic lesions.

Methods:

We prospectively recruited patients referred for assessment of pancreatic cystic lesions from August 2014 until July 2015. All pancreatic cystic lesions were examined with nCLE miniprobe via endoscopic ultrasound followed by fine-needle aspiration. Information regarding the cysts (morphology, location, fluid analysis etc.) was documented. Adverse event was recorded.

Results:

Fourteen patients were recruited, six were male with a mean age of 66.5 (range 48–80) years. Only 12 completed nCLE examination of the pancreatic cystic lesions. nCLE imaging was successful in 83.3% (10/12). Average nCLE imaging duration was 5 min 18 s. There was one (10%) adverse event. nCLE impressions were correct in eight cases (five malignant and three benign) compared with final diagnosis. Three patients underwent surgery, histology showed ductal adenocarcinoma, pancreatic neuroendocrine tumor and gastric-subtype of intraductal papillary mucinous neoplasm. The sensitivity and specificity of the nCLE impression when compared with final diagnosis were 83.3% and 75% respectively. The accuracy of nCLE was 80%.

Conclusion:

Our results have demonstrated that nCLE assessment of pancreatic cystic lesions is safe and feasible. It may complement the existing diagnostic modalities to improve diagnostic yield.

Keywords

Pancreatic cystic lesion needle-based confocal laser endomicroscopy endoscopic ultrasound fine needle aspiration

Introduction

Pancreatic cystic lesions (PCLs) are being detected more often as sensitive abdominal imaging such as computed tomography scans or magnetic resonance imaging are being used for multiple indications. Most PCLs are identified incidentally when abdominal imaging is performed for unrelated indications. The prevalence of incidental PCLs detected on abdominal imaging is 2.6%, increasing to 8% in the elderly.^1–3 A significant percentage of patients with PCLs have malignant potential and can transform into pancreatic cancer.^4,5 Unfortunately, the natural history of these cysts is not well known and poorly understood. As pancreatic cancers often present at late stage and are usually inoperable at time of presentation, PCLs were thought to provide a window of opportunity for early resection as they are detected easily on cross-sectional imaging.

In the past, aggressive resection policy was adopted for all PCLs. According to Goh et al., it has been shown that of all patients with PCLs that underwent resection surgery, approximately 40% had benign lesions.⁶ This means that we are subjecting this group of patients to major operative procedure without added benefit, but with a 5% risk of mortality and up to 60% of morbidity (e.g. pancreatic leak, pancreatic fistula, infection). Whilst the majority of post-operative complications are not life threatening, they can, however, amount to increased lengths of hospital stay and medical costs, and decrease in quality of life for the patients.

In view of the unfavorable outcome of resection surgery, in 2012 a guideline was proposed for better management of PCLs.⁷ This consensus guideline recommends endoscopic ultrasound–fine needle aspiration (EUS-FNA) to be performed on patients who have pancreatic cysts with worrisome features. EUS can differentiate the cyst’s structure (unilocular, macrocystic, microcystic), location in the pancreas (head vs. body and tail) and relationship to the pancreatic duct, but these features are not diagnostic. FNA of the cyst is performed in order to determine fluid characteristics, including cytological information of the cyst. However, even after fluid analysis, it remains difficult to differentiate between cyst types, because the fluid is often acellular. Tumor markers, including carcinoembryonic antigen (CEA), improve the sensitivity of detection of mucinous cysts, but lack specificity.^2,8,9 Brugge et al. concluded that combination of all three modalities could improve the accuracy to only approximately 70%.¹⁰ In view of the limited capability of current diagnostic tools, this creates a constant dilemma for the gastroenterologist or surgeon as to how best manage indeterminate PCLs. Therefore, there is a pressing need to search for a better diagnostic modality to improve the diagnostic yield of the PCLs.

The advent of needle-based confocal laser endomicroscopy (nCLE) has shown initial promising results in the assessment of PCLs.^11–17 We aimed to study our clinical experience, in particularly the feasibility and safety of nCLE in the assessment of PCLs in a local tertiary center.

Method

Study population

This was a prospective cohort study involving the enrollment of consecutive patients who were referred for endoscopic assessment (EUS-FNA) of cystic lesion of pancreas between August 2014 and July 2015. The patients’ demographic characteristics and clinical data, including age, sex and comorbid disease, were reviewed. The clinical outcome of patients was analyzed. This study was performed in accordance with the guidelines of our Institutional Review Board. The inclusion criteria for the study were patients who were 18 years or older, PCLs greater than 15mm and patients able to provide written informed consent for the study. The exclusion criteria were patients with allergy to fluorescein, pregnant, breast-feeding, coagulopathy with international normalized ratio greater than 1.5, partial thromboplastin time greater than twice that of control and platelet count less than 50,000 × 10³/µl.

Endoscopic procedure

All the EUS-FNA procedures were done with Olympus GF-UCT180 or GF-UC140P-AL5 curvilinear echoendoscopes with a Hitachi-Aloka ProSound F75 EUS processor. The procedures were performed by a single operator (DMYT) who had been accredited for nCLE image interpretation. Before the beginning of the procedure, the locking device was attached to the proximal end of a 19G (EchoTip® Ultra Cook-Medical, Winston-Salem, NC, United States) EUS needle, once the stylet had been removed from the needle. The nCLE miniprobe (AQ-Flex, Cellvizio, Mauna Kea Technologies, Paris, France) was then inserted inside the needle through the locking device. Once the tip of the miniprobe protruded from the needle’s bevel, it was then slightly retracted into the needle in order to make the tip of needle sharp enough to be penetrate the target lesion. It stayed inside the needle for the insertion into the endoscope. Prophylactic antibiotic was administered before the start of the procedure. The pancreatic cyst was then identified on the EUS monitor and punctured. Once the needle was inside the pancreatic cyst, the miniprobe was advanced in the needle for it to be in contact with the cyst wall or solid component if present and nCLE imaging began. Then, 2.5ml of fluorescein 10% was injected intravenously, in order to allow fluorescent imaging. Microscopic real-time sequences were recorded during 5–10 min inside the pancreatic cysts, preferably against the cyst wall, to image cellular structures. An initial nCLE impression was made after nCLE imaging; the miniprobe was extracted from the needle and brought for disinfection. The cyst fluid was then aspirated and analyzed for chemistry, cytology and/or tumor markers. All adverse events were recorded. For patients who had surgical resection of pancreatic cyst, surgery histological specimen was considered the gold standard diagnosis, and was compared with the nCLE impression made during the endoscopic assessment. In cases where there is no gold standard diagnosis with histopathology, final diagnosis was made based on other clinical parameters such as characteristic radiological imaging, endosonographic morphology, cyst fluid analysis including biochemistry, cytology if available and clinical follow-up.¹⁰ The final diagnosis of PCLs for malignant lesions (intraductal papillary mucinous neoplasm (IPMN), neuroendocrine tumor (NET), adenocarcinoma) were established based on any positive results after surgery or further clear evidence of malignancy by other imaging or diagnostic methods. A PCL was considered benign (pseudocyst, serous cystadenoma (SCA)) if pathologic results were negative and no tumor masses could be shown within a follow-up period of at least six months. The method on how each final diagnosis derived was indicated.

Results

Patients

Fourteen patients who were scheduled for EUS-FNA of PCLs were recruited throughout the study period. Only 12 patients eventually had nCLE examination; one was unfit for the procedure due to anesthesiology contraindication and one withdrew consent. Six were female, with a mean age of 66.5 (range 48–80) years. Four lesions were in the head/uncinate of the pancreas, six in the body/neck and two in the tail. Six (50%) PCLs showed a solid component within the lesion. Table 1 shows the clinical characteristic of the patients.

Table 1.

Clinical characteristics of patients with pancreatic cystic lesion.

Age (years), mean	66.5 (range 48–80)
Sex, n
Male	6
Female	6
Location
Head and uncinate	4
Body/neck	6
Tail	2
Lesion size (mm)
Range	19–62
Mean	33.9
Cystic lesion component
Solid component present	6
Multilocular	8
Unilocular	4
Imaging time
Range	4 min 18 s–6 min 18 s
Mean	5 min 18 s

Table 2 illustrates the nCLE findings and impressions. Needle puncture of the PCLs was successful in all cases. nCLE imaging was successful in only 83.3% (n=10). Average nCLE imaging duration was 5 min 18 s. The majority (66.7%) of the cases were trans-gastric puncture. Initial clinical impression was made after nCLE imaging, features consistent with IPMN were seen in four cases (Figure 1), four were SCA (Figure 2) and two cases had features suspicious of malignancy (Figure 3). Three patients had an operation after EUS-FNA: one had well differentiated NET, one showed ductal adenocarcinoma with pancreatobiliary subtype and one had pancreatic intraepithelial neoplasia (PanIN); two with gastric subtype. The rest of the patients were placed on interval surveillance as recommended by the 2012 international consensus guideline. Only patients who successfully completed the nCLE assessment (n=10) were included for analysis, hence patients number 4 and number 9 were excluded from final analysis. The sensitivity and specificity for the nCLE impression when compared with final diagnosis were 83.3% and 75% respectively. The positive predictive value was 83.3% and negative predictive value was 75%. The accuracy of nCLE from our study cohort was 80% (Table 3).

Table 2.

nCLE, biochemistry, cytology and histopathology finding of pancreatic cystic lesions.

Patient no.	Sex	Age(years)	Cyst location	Cyst size(mm)	Location of puncture	nCLE impression	Structures identified	Complication	Cyst CEA(µg/l)	Cyst amylase (U/l)	Cytology	Final diagnosis	Diagnosis by:
1	F	78	Head	56	Duodenum	IPMN	Papillary like projection	Nil	ND	ND	Cyst contents with benign epithelial cells	IPMN	*
2	M	74	Head	28	Duodenum	IPMN	Papillary like projection	Nil	53.4	22	Dysplastic glandular epithelium, consistent with IPMN	IPMN	Cytology
3	M	68	Tail	27	Gastric	Serous cystadenoma	Superficial vascular network	Nil	45.2	1139	Poor cellularity	IPMN	*
4^a	M	66	Uncinate process	27	Duodenum	Unable to capture image	NA	Nil	ND	ND	Low yield benign epithelial cells present	NA	NA
5	F	79	Neck	40	Gastric	Suspicious of malignant lesion	Prominent blood vesselsAcinar cells with fibrous area	Nil	ND	ND	Well differentiated neuroendocrine tumor NET	Well differentiated NET	Cytology and histology
6	M	75	Body	19	Gastric	Serous cystadenoma	Superficial vascular network	Nil	0.2	150	Benign pancreatic acinar elements are seen	Serous cystadenoma	*
7	M	69	Neck	22	Gastric	Serous cystadenoma	Superficial vascular network	Nil	ND	ND	No neoplastic epithelium identified	Serous cystadenoma	*
8	F	72	Tail	62	Gastric	IPMN	Papillary like projectionDark cell clumps	Nil	5696	11	Strips of neoplastic mucinous epithelium presentFeatures in keeping with IPMN	Pancreatic ductal adenocarcinoma pancreatobiliary IPMN	Cytology and histology
9^a	M	63	Uncinate process	21	Duodenum	Unable to capture image	NA	Nil	325	>15,000	No epithelial cell present	NA	NA
10	F	48	Body	29	Gastric	Serous cystadenoma	Superficial vascular network	Nil	4.5	519	No malignant cells seen	Serous cystadenoma	*
11	F	55	Body	39	Gastric	Suspicious of malignant lesion	Dark cell clumps	Bleeding	ND	13	Inflammatory cell with debrisNo malignant cells present	Pseudocyst	*
12	F	52	Neck	45	Gastric	IPMN	Papillary like projection	Nil	640	12913	No epithelial cells present	Gastric IPMN	Histology

Presumed diagnosis based on endosonographic appearance, fluid analysis, cytology and clinical follow-up.

Excluded from final analysis.

nCLE: needle-based confocal laser endomicroscopy; CEA: carcinoembryonic antigen; F: female; M: male; IPMN: intraductal papillary mucinous neoplasm; ND: not detectable; NA: not available; NET: neuroendocrine tumor.

Figure 1.

Intraductal papillary mucinous neoplasm, needle-based confocal laser endomicroscopy: papillary projection characterized by a vascular core (in white) surrounded by an epithelial border (in gray).

Figure 2.

Serous cystadenoma, needle-based confocal laser endomicroscopy: superficial vascular network.

Figure 3.

Pancreatic adenocarcinoma, needle-based confocal laser endomicroscopy: dark cell clumps.

Table 3.

Result of nCLE examination.

		Pancreatic cystic lesions
		Malignant^a	Benign^b
nCLE	Positive	5	1
	Negative	1	3

Intraductal papillary mucinous neoplasm, adenocarcinoma, neuroendocrine tumor.

Pseudocyst, serous cystadenoma.

nCLE: needle-based confocal laser endomicroscopy.

Adverse events

There were no adverse events related to intravenous fluorescein injection. One (10%) post-procedure complication occurred after examination of the PCLs (patient number 11 in Table 2). The cystic lesion was at the pancreas body. The patient developed intracystic bleeding after the EUS-FNA procedure that required blood products transfusion. The patient subsequently developed pseudoaneurysm within the pancreatic cyst. On further analysis, it revealed that the patient already had Escherichia coli bacteremia, which had not manifested clinically prior to the procedure. Despite giving prophylactic antibiotic, the pseudoaneurysm still developed after the procedure as a result of bacteremic seeding of the arterial wall of the pancreatic cyst.¹⁸ The pseudoaneurysm was obliterated with histoacryl injection and no further bleeding event reported.

Discussion

PCLs are increasingly detected with the liberal use of advanced imaging techniques.³ The majority of the PCLs are benign; however, some still harbor malignant potential.⁵ Hence, the management PCLs poses a constant dilemma and huge challenge for the managing gastroenterologist and surgeon. Existing investigation modalities are far from ideal to clinch the diagnosis of the PCLs.⁹ Therefore, there is a constant need to search for the best diagnostic tool to provide the much needed answer.

Confocal laser endomicroscopy (CLE) is an imaging technology in which a low power laser illuminates and scans a single focal plane of the tissue. This allows for microscopic-detail examination of the surface epithelium. The technical principle is to have focused light passed through a confocal aperture, thereby reducing scattered light above and below the plane. This will function as an ‘optical biopsy’, which may provide microscopic detail of the tissue examined.^19,20 The advent of the nCLE probe has given hope that it may bridge the ‘gap’ in the management of PCLs. It provides real-time histopathology assessment and has the potential to improve sampling error and to reduce the number of passes needed for diagnosis.

We are reporting our clinical experience with nCLE in the assessment of PCLs in a tertiary center. Our findings confirm the report from the initial investigators with regard to safety and feasibility of nCLE in assessment of PCLs.^12,13 Adopting the pre-loading approach of nCLE into the FNA needle prior to PCL puncture and limiting the duration of nCLE imaging to less than 10 min increases the safety of the procedure. There was no pancreatitis reported in our study cohort. From our series, IPMN is by far the most common finding. The diagnosis was made based on the typical nCLE finding of papillary finger-like projection. Pancreatic cyst fluid CEA was detected in only four of the IPMN cases. Two of the IPMN patients (patients number 8 and number 12) underwent surgery with histology finding of pancreatobiliary and gastric subtype respectively. The former underwent the operation because the cytology from the FNA showed strips of neoplastic mucinous epithelium and the latter had an enhancing definite mural nodule seen within the lesion on EUS examination. The rest of the patients were placed on interval surveillance as recommended by the 2012 international consensus guideline. Recent work on pancreatic cystic CEA level suggested that a high CEA level is associated with the gastric subtype of IPMN and low risk imaging EUS features.²¹ However, this is not evident in our study (Table 2). There are three SCAs which showed typical superficial vascular network (SVN), which is highly specific for diagnosis of SCA.¹⁴ Napoleon et al. suggest that the presence of SVN during nCLE imaging can confidently diagnose SCA and rule out any unnecessary surgical resection or clinical follow-up. However, in our series, patient number 3 had a false negative result whereby the final diagnosis was changed to IPMN based on collateral findings of multilocularity EUS morphology and high amylase level, which is unusual for SCA. This highlights the intricacy and limitation of investigations in diagnosing PCLs. Perhaps the nCLE images acquired from this patient could be displaying a subtype of IPMN which is not currently defined. The suspicious malignant feature that was seen on nCLE was a dark cells clump within the PCLs of two patients;¹⁷ one of the patients had surgical resection with positive malignant histology. The false positive test (patient number 11) was probably due to debris within the pseudocyst being interpreted as dark clumps; however, subsequent fluid analysis and cytology were suggestive of a benign lesion. There is one NET in our study cohort: nCLE imaging showed fibrous areas with a few dark clumps (Figure 4). These features are consistent with findings from recent publications.^16,22 For patients without gold standard, the final diagnosis was made based on other clinical parameters such as characteristic radiological imaging, endosonographic morphology and cyst fluid analysis. These patients were followed up clinically for at least 12 months; none of the patients had disease progression.

Figure 4.

Neuroendocrine tumor, needle-based confocal laser endomicroscopy: neoplastic cell cluster (red arrows) with fibrous area (white arrows).

Both unsuccessful nCLE imaging cases in our study were due to the location of the cystic lesion at the uncinate of the pancreas. This is because the curvilinear echoendoscope assumed a flexed position when it was advanced into the distal duodenum. This poses technical difficulties in positioning a large bore 19G FNA needle in the distal duodenum to puncture the lesion. We had to abandon the nCLE imaging in both cases. The location of PCLs should be taken into consideration prior to nCLE imaging to ensure higher technical success. It is difficult to exactly pinpoint the adverse event being directly related to the nCLE procedure because the patient may still develop intracystic bleeding if only the FNA procedure is performed, without the nCLE imaging. Perhaps one might argue that a smaller gauge FNA needle may potentially prevent the adverse event.

There are emerging data suggesting a molecular target, cathepsin E, which is specifically expressed in pancreatic ductal adenocarcinoma and PanIN. It can be detected with CLE and its expression increases with disease progression.^23,24 This could serve as a potential avenue to improve the stratification of malignant risk of PCLs; however, these were data from animal experiments and further studies are required before it can be applied clinically.

There are limitations with our study. The promising performance of nCLE that is observed in our study is probably due the small sample size as this is only an initial pilot study. We do not have a gold standard comparison for every case, hence, we adopted other surrogate markers such as cytology, biochemistry analysis of the cyst fluid and interval imaging to reach the final diagnosis. This raises the question of the authenticity of the so-called ‘true negative or true positive’ PCLs in our study cohort, which could potentially be altered if gold standard comparison become available. This reflects the real world clinical practice whereby not all patients with PCLs will eventually undergo surgery, the diagnosis has to be derived from next best available clinical parameters. The follow-up interval for our patients in this study was a minimum period of 12 months, therefore, the actual long term risk of progression for the non-operated PCLs in our study cohort is unclear. Longer duration of follow-up is warranted to delineate the malignant lesion. However, patients who had positive histology on surgical resection demonstrated suspicious malignant features on nCLE assessment.

Conclusion

PCLs are increasingly being diagnosed with the use of sophisticated imaging; however, much still remains unknown about their nature and progression. This study has proven nCLE to be a safe and feasible option as part of the assessment for PCLs. Based on currently available data, nCLE should not be routinely used beyond the context of research due to its limited evidence and high operating cost. With the recent publication of nCLE criteria by Napoleon et al., nCLE may be added to existing investigational armamentariums to complement the assessment of PCLs. It may help to stratify the malignant risk in patients with PCLs and refer for surgical management or surveillance accordingly. However, more research is still required before nCLE can be recommended as the standard of care for the assessment for PCLs.

Footnotes

Acknowledgements

The authors would like to thank all the endoscopy staffs who have assisted in the procedure throughout the study period. Informed consent: all study participants, or their legal guardians, in this study provided informed written consent about personal and medical data collection prior to study enrollment. Ethical approval: the study was reviewed and approved by the Institutional Review Board of Singapore General Hospital. Author contribution: Yung Ka, CHIN: study concept and design, acquisition of funding, study development and coordination, acquisition of data, drafting of the manuscript and review of the manuscript for important intellectual content. Christopher Jen Lok, KHOR: acquisition of data and review of the manuscript for important intellectual content. Brian Kim Poh, GOH: acquisition of data and review of the manuscript for important intellectual content. Tony Kiat Hon, LIM: provide histopathology diagnosis and review of the manuscript for important intellectual content. Damien Meng Yew, TAN: study concept and design, acquisition of data, review of the manuscript for important intellectual content and overall study supervision

Declaration of conflicting interests

The authors declare that there is no conflict of interest.

Funding

This work was supported by Singapore General Hospital Research Grant Number SRG/C2/09/2014.

References

Laffan

Horton

Klein

et al . Prevalence of unsuspected pancreatic cysts on MDCT. Am J Roentgenol 2008; 191: 802-807.

Jong

Bruno

Fockens

. Epidemiology, diagnosis and management of cystic lesions of pancreas. Gastroenterol Res Pract 2012; 2012: 1-8.

Jong

Nio

Hermans

et al . High prevalence of pancreatic cysts detected by screening magnetic resonance imaging examinations. Clin Gastroenterol Hepatol 2010; 8: 806-811.

Fernandez-del Castillo

Targarona

Thayer

et al . Incidental pancreatic cysts: Clinicopathologic characteristics and comparison with symptomatic patients. Arch Surg 2003; 138: 427-423.

Spinelli

Fromwiller

Daniel

et al . Cystic pancreatic neoplasms. Observe or operate. Ann Surg 2004; 239: 651-659.

Goh

Tan

Cheow

et al . Cystic lesions of the pancreas: an appraisal of an aggressive resectional policy at a single institution during 15 years. Am J Surg 2006; 192: 146-154.

Tanaka

Fernández-del Castillo

Adsay

et al . International consensus guidelines 2012 for the management of IPMN and MCN of the pancreas. Pancreatology 2012; 12: 183–197.

Hutchins

Draganov

. Cystic neoplasms of the pancreas: A diagnostic challenge. World J Gastroenterol 2009; 15: 48-54.

Lee

. Diagnostic approach to pancreatic cysts. Curr Opin Gastroenterol 2014; 30: 511-517.

10.

Brugge

Lewandrowski

Lee-Lewandrowskiand

et al . Diagnosis of pancreatic cystic neoplasms: A report of the cooperative pancreatic cyst study. Gastrotenterology 2004; 126: 1330-1336.

11.

Konda

Harry

Wallace

et al . First assessment of needle-based confocal laser endomicroscopy during EUS-FNA procedures of the pancreas. Gastrointest Endosc 2011; 74: 1049-1060.

12.

Konda

Meining

Jamil

et al . A pilot study of in vivo identification of pancreatic cystic neoplasms with needle-based confocal laser endomicroscopy under endosonographic guidance. Endoscopy 2013; 45: 1006-1013.

13.

Jamil

Meining

Wallace

et al . Safety profile of needle-based confocal laser endomicrosocpy (nCLE) during EUS-FNA procedure of the pancreatic cystic lesions. Gastrointest Endosc 2012; 75: AB188-AB189.

14.

Napoléon

Lemaistre

Pujol

et al . A novel approach to the diagnosis of pancreatic serous cystadenoma: Needle-based confocal laser endomicroscopy. Endoscopy 2015; 47: 26-32.

15.

Nakai

Iwashita

Park

et al . Diagnosis of pancreatic cysts: Endoscopic Ultrasound, Through-the-needle Confocal Laser-induced Endomicroscopy and Cystoscopy Trial (DETECT Study). Gastrointest Endosc 2015; 81: 1204-1214.

16.

Napoleon

Lemaistre

Pujol

et al . In vivo characterization of pancreatic cystic lesions by needle-based confocal laser endomicroscopy (nCLE): Proposition of a comprehensive nCLE classification confirmed by an external retrospective evaluation. Surg Endosc 2016; 30: 2603-2612.

17.

Pittayanon

Kongkam

Sampatanukul

et al . Endoscopic ultrasound-guided needle-based confocal laser endomicroscopy for diagnosis of solid pancreatic lesions (ENES): A pilot study. Endosc Int Open 2016; 4: E17-E23.

18.

Spelman

. Overview of infected (mycotic) arterial aneurysm. In: Basow

(ed.) UpToDate. Waltham, MA: UpToDate Inc., https://www.uptodate.com/contents/overview-of-infected-mycotic-arterial-aneurysm (2014, accessed 3 September 2016).

19.

Polglase

McLaren

Delaney

. Pentax confocal endomicroscope: A novel imaging device for in vivo histology of the upper and lower gastrointestinal tract. Expert Rev Med Devices 2006; 3: 549-556.

20.

Becker

Wallace

Fockens

et al . Needle-based confocal endomicroscopy for in vivo histology of intra-abdominal organs: First results in a porcine model (with videos). Gastrointest Endosc 2010; 71: 1260-1266.

21.

Yoon

Daglilar

Mino-Kenudson

et al . Characterization of epithelial subtypes of intraductal papillary mucinous neoplasm of the pancreas with endoscopic ultrasound and cyst fluid analysis. Endoscopy 2014; 46: 1071-1077.

22.

Karia

Waxman

Konda

et al . Needle-based confocal endomicroscopy for pancreatic cysts: The current agreement in interpretation. Gastrointest Endosc 2015; 83: 924-927.

23.

Cui

et al . Monitoring pancreatic carcinogenesis by the molecular imaging of cathepsin E in vivo using confocal laser endomicroscopy. PLoS One 2014; 9: e106566.

24.

Cruz-Monserrate

Abd-Elgalliel

Grote

et al . Detection of pancreatic cancer tumours and precursor lesions by cathepsin E activity in mouse models. Gut 2012; 61: 1315-1322.