Int J Gastrointest Interv 2024; 13(3): 98-104
Published online July 31, 2024 https://doi.org/10.18528/ijgii240026
Copyright © International Journal of Gastrointestinal Intervention.
Hussein Okasha , Ahmed Ebrahim
, Ihab Samih
, and Mohammed Sayed*
Department of Internal Medicine, KasrAlAiny School of Medicine, Cairo University, Giza, Egypt
Correspondence to:*Department of Internal Medicine, KasrAlAiny School of Medicine, Cairo University, 995, 25st 5th district, 6 October City, Giza 12613, Egypt.
E-mail address: dr.msh81@gmail.com (M. Sayed).
This is an open-access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/bync/4.0) which permits unrestricted noncommercial use, distribution, and reproduction in any medium, provided the original work is properly cited.
Background: Endoscopic ultrasound (EUS) tissue acquisition requires rapid on-site evaluation to improve diagnostic yield with less needle passes; however, this is not readily available in all centers, also it increases the duration and the cost of the procedure. Macroscopic on-site evaluation (MOSE) is the direct assessment of the tissue sample provided by EUS-guided fine needle biopsy (EUS-FNB) macroscopically, to detect whether it is sufficient or not. Aim of this study was to define the role of MOSE in judging the adequacy of the tissue core specimens and to evaluate its diagnostic yield and accuracy.
Methods: This prospective study was conducted at our institution with 215 participants of both genders presenting for evaluation of solid or mixed solid and cystic lesions by MOSE technique by EUS-fine needle aspiration (FNA) or FNB.
Results: In obtaining conclusive sample MOSE score 1 sensitivity was 68.2%, specificity was 75%, positive predictive value (PPV) was 99.3, negative predictive value (NPV) was 4.3 and overall accuracy was 68.4, in MOSE score 2 sensitivity was 88%, specificity was 75%, PPV was 99.5, NPV was 10 and overall accuracy was 87, in MOSE 2 (FNB) sensitivity was 92%, Specificity was 67%, PPV was 99.4, NPV was 12, 5 and overall accuracy was 92. There was no statistically significant difference between the two groups with conclusive and inconclusive cytopathological results regarding size of the needle, number of needle passes, type of the needle and the specimen acquisition method.
Conclusion: Using MOSE for assessment of adequacy of the sample obtained by EUS-FNB showed higher diagnostic yield compared to conventional method using different needle sizes.
Keywords: Endoscopic ultrasound, Fine needle aspiration, Macroscopic on-site evaluation
Although fine needle aspiration (FNA) cytology has been postulated as the standard method for tissue sampling, its accuracy of diagnosis varies from 78% to 95% as it does not give information on tissue architecture.1 So endoscopic ultrasound-guided fine needle biopsy (EUS-FNB) has been suggested as it has the advantage of giving core tissue sample, thus allowing a more accurate diagnosis.2
EUS-tissue acquisition (EUS-TA) requires rapid on-site evaluation (ROSE) to improve accuracy with less needle passes, however this is not readily present in all centers, also it increases the duration and the cost of the procedure.3
Macroscopic on-site evaluation (MOSE) is the direct assessment of the tissue sample provided by EUS-FNB macroscopically to detect whether it is sufficient or not. Recent studies suggest that MOSE may be a good alternative to ROSE.3
Only limited number of studies have evaluated the adequacy of histology cores in MOSE during EUS-FNA/FNB. However, without robust data, MOSE has not been broadly adopted as a standard technique.4
Currently, MOSE, compared to traditional methods, its role still undefined. In theory, its advantage is that the endoscopist judges the adequacy of the material and based on that he determines the number of needle passes.5
The main aim of this study was to assess the efficacy of MOSE in judging the adequacy of core tissue and to evaluate its diagnostic yield and accuracy.
This prospective study included 215 male and female (inpatient and outpatient) patients aged 18 to 80 years at our institution who were admitted for evaluation of solid or mixed solid and cystic lesions by EUS-FNA or FNB by MOSE. Approval for tissue sampling during EUS was obtained. All procedures followed were in accordance with the ethical standards of the responsible committee on human experimentation (institutional and national) and with the Helsinki Declaration of 1975, as revised in 2008. Informed consent was obtained from all patients for being included in the study and our study was approved by the ethical committee of Cairo University (IRB approval number MD-319-2022).
Severe coagulopathy (INR > 1.5 or platelet count < 50,000) and patients unfit for deep sedation.
All patients (over the age of eighteen) who were referred for EUS-FNB/FNA of solid pancreatic lesions or extra-pancreatic lesions, including subepithelial lesions and lymphadenopathy, were recruited. The target lesion was assessed using power Doppler, color flow, and B-mode imaging.
FNB or FNA was performed using a 19 or 22 G Franseen FNB needle (Boston Scientific) or Echotip FNA needles (Cook Medical). Each target was sampled with one to 3 passes, median was 2 passes. Samples acquisition was done by performing “slow pull” or “suction” technique.
After EUS-TA, patients were observed for immediate complications for at least 3 hours. The sample was placed on a filter paper soaked in physiological saline solution and the endoscopist evaluates the tissue core according to the MOSE classification.
The following macroscopical evaluations were made of the core tissue: There are three types of visible tissue cores: (1) visible with few blood clots, (2) visible with considerable blood clots, or (3) scarce with mostly blood clots.
0: No material.
1: Only necrotic or bloody material (Fig. 1).
2: ≥ 1 core equal or less than 2 mm whitish yellow (Fig. 2).
3: ≥ 1 core more than 2 mm whitish yellow (Fig. 3).
After inspection of the core by the operator, all specimens were fixed in 10% formalin solution or 95% absolute alcohol for cytological analysis. Specimens were evaluated by an experienced gastrointestinal and pancreaticobiliary pathologist.
The number of passes, needle type and size, and procedural adverse events were assessed.
Sample histopathology was categorized as benign, malignant, or inconclusive. Inconclusive samples were defined as hypocellular or acellular smears, no sufficient to diagnose a malignant or benign disease.
The Statistical Package for Social Sciences version 27 (IBM Corp.) was used for data management and analysis. The means and standard deviations of the numerical data were used to summarize it. Numbers and percentages were used to summarize the categorical data. The percentages and figures were used to estimate the frequency. Mathematical data were checked for normality using Shapiro-Wilk and Kolmogrov-Smirnov tests. When comparing the independent groups with regard to categorical data, chi-square or Fisher’s tests were utilized. Using a 2 × 2 table, the MOSE technique’s sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and total accuracy were determined. Every test had two tails, and a probability (
In this study, according to sociodemographic characteristics, the average age was 60 ± 12 years, 44.9% were female, and 55.1% were male. In terms of location, 89.8% of lesions were found in the pancreas; 4.6% of lesions were found in the pancreas and liver, 3.7% of lesions were found in the pancreas and lymph nodes, and 1.9% were found in the pancreas + liver + lymph nodes (Table 1). According to EUS, the median size of the pancreatic lesion on EUS was 40 mm; The median size of the liver lesion was 20 mm on EUS and the median size of the lymph node lesion was 28 mm. Regarding needle type, 5.1% of needles were FNA-Echo Tip-Cook, 9.3% were FNA-Expect-Boston, and 85.6% were FNB-Acquire-Boston.
Table 1 . Size of Lesions, Type and Size of Needles, Scope Position, Specimen Acquisition Method and Number of Needle Passes.
Variables | No. of patients ( |
---|---|
Size of lesion on EUS (mm) | |
Pancreatic | 40 (6–120) |
Hepatic | 20 (10–35) |
Lymph node | 28 (15–50) |
Type of the needle | |
FNA-EchoTip-Cook | 10 (4.6) |
FNA-Expect-Boston | 20 (9.3) |
FNB-Acquire-Boston | 185 (86) |
Size of the needle (G) | |
19 | 2 (0.9) |
22 | 213 (99.1) |
Echo endoscope position during FNB | |
Transduodenal | 130 (60.5) |
Transgastric | 70 (32.5) |
Transgastric + transduodenal | 15 (7) |
Specimen acquisition method | |
Suction method | 170 (79) |
Both (capillary & suction) | 45 (21) |
Number of needle passes FNB 1 | |
1 | 67 (31.2) |
2 | 141 (65.6) |
3 | 7 (3.2) |
MOSE 1 | |
1 | 145 (67.4) |
2 | 67 (31.2) |
3 | 3 (1.4) |
MOSE 2 | |
1 | 2 (0.5) |
2 | 29 (13.5) |
3 | 184 (86.0) |
Values are presented as mean (range) or number (%)..
EUS, endoscopic ultrasound; FNB, fine needle biopsy; MOSE, macroscopic onsite evaluation..
The size of the needle was 19 G in 0.9% and 22 G in 99.1% depending on the location of the echoendoscope during FNB. Transduodenal were 60.5%, transgastric were 32.5%, and transgastric + transduodenal were 7%. The suction method was used in 79%, and two methods (capillary & suction) were used in 21%. The number of needle passes for FNA was 38.7% for 1 pass, 58.1% for 2 passes, and 0.03% for 3 passes. In FNB, 1st pass was 31.2%, 2nd pass was 65.6%, and 3rd pass was 3.2%. According to the MOSE score, 1, 67.4% received a score of 1, 31.2% received a score of 2, and 1.4% received a score of 3. According to MOSE score 2, 0.5% scored 1, 13.5% scored 2, and 86% scored 3 (Table 1). Lesions diagnosis was shown in Table 2.
Table 2 . Lesion Diagnosis and Post Procedural Adverse Events.
Lesion diagnosis and post procedural adverse events | No. of patients |
---|---|
Cytopathology-grade | |
Adenocarcinoma grade II | 8 (6.5) |
High grade adenocarcinoma | 115 (93.5) |
Post procedural adverse events | |
Small blood collection | 3 (1.4) |
No | 212 (98.6) |
Final diagnosis | |
Metastatic adenocarcinoma | 6 (2.8) |
Adenocarcinoma | 167 (77.7) |
Adrenocortical neoplasm | 1 (0.5) |
Bifocal pancreatic large cell neuroendocrine carcinoma | 1 (0.5) |
Chronic pancreatitis | 8 (3.7) |
Inconclusive | 4 (1.9) |
IPMN | 2 (0.9) |
IPMN + adenocarcinoma | 1 (0.5) |
IPMN with dysplasia | 1 (0.5) |
Metastatic of Merkel cell carcinoma | 1 (0.5) |
Neuroendocrine tumor | 9 (4.2) |
Pancreatic microcystic serous cystadenoma | 6 (2.8) |
Pancreatic mucinous cystadenoma | 1 (0.5) |
Pancreatic mucinous cystadenoma + serous cystadenoma | 1 (0.5) |
Solid pseudopapillary neoplasm | 6 (2.8) |
Values are presented as number (%)..
IPMN, intraductal papillary mucinous neoplasm..
There was no statistically significant difference between conclusive and non-conclusive regarding final diagnosis or MOSE 2 according to type and size of needle and specimen acquisition method (Tables 3, 4). MOSE score 1 (all) sensitivity was 68.2%, specificity was 75%, PPV was 99.3%, NPV was 4.3%, and overall accuracy was 68.4%. In MOSE score 2, (all) sensitivity was 68.4%, 88%, specificity 75%, PPV 99.5%, NPV 10%, and overall accuracy 87%. In MOSE 2 (FNB), sensitivity was 92%, specificity was 67%, PPV was 99.4%, NPV was 12.5%, and overall accuracy was 92% (Table 5).
Table 3 . Relation of MOSE Diagnosis to Final Diagnosis.
Final diagnosis | ||||
---|---|---|---|---|
Inconclusive | Conclusive | |||
Benign | Malignant | |||
MOSE 1 score | ||||
1 | 1 (0.7) | 8 (5.5) | 136 (93.8) | |
2 | 1 (1.5) | 10 (14.9) | 56 (83.6) | < 0.001 |
3 | 2 (66.7) | 0 (0.0) | 1 (33.3) | |
MOSE 2 score | ||||
1 | 1 (100) | 0 (0.0) | 0 (0.0) | |
2 | 2 (6.9) | 6 (20.7) | 21 (72.4) | < 0.001 |
3 | 1 (0.5) | 12 (6.5) | 172 (93.0) |
Values are presented as number (%)..
MOSE, macroscopic on-site evaluation..
Table 4 . Type and Size of the Needle in Relation to MOSE 2.
MOSE 2 | |||
---|---|---|---|
Inconclusive | Conclusive | ||
Type of the needle | |||
FNA-Echo Tip-Cook | 0 (0.0) | 11 (100) | 0.9 |
FNA-Expect-Boston | 0 (0.0) | 20 (100) | |
FNB-Acquire-Boston | 1 (0.5) | 183 (99.5) | |
Type of needle | |||
FNA | 0 (0.0) | 31 (100) | 0.6 |
FNB | 1 (0.5) | 183 (99.5) | |
Size of the needle (G) | |||
19 | 0 (0.0) | 2 (100) | |
22 | 1 (0.5) | 212 (99.5) | 0.9 |
Specimen acquisition method | |||
Suction method | 1 (0.6) | 168 (99.4) | |
Both (capillary & suction) | 0 (0.0) | 46 (100) | 0.6 |
Values are presented as number (%)..
MOSE, macroscopic on-site evaluation; FNA, fine needle aspiration; FNB, fine needle biopsy..
Table 5 . Accuracy of MOSE in Obtaining Conclusive Sample.
Sensitivity (%) | Specificity (%) | PPV (%) | NPV (%) | Overall accuracy (%) | |
---|---|---|---|---|---|
MOSE 1 (all) | 68.2 | 75.0 | 99.3 | 4.3 | 68.4 |
MOSE 2 (all) | 88.0 | 75.0 | 99.5 | 10.0 | 87.0 |
MOSE 2 (FNB) | 92.3 | 67.0 | 99.4 | 12.5 | 92.0 |
MOSE, macroscopic on-site evaluation; PPV, positive predictive value; NPV, negative predictive value; FNB, fine needle biopsy..
There was no statistically significant difference between scores 2 and 3 regarding the number of needles passed through the FNB. There is highly statistically significant difference between score 2 and score 3 regarding final diagnosis (Tables 6, 7).
Table 6 . Number of Needle Pass in Relation to MOSE.
No. of needle passes FNB 1 | ||||
---|---|---|---|---|
1 | 2 | 3 | ||
MOSE 1 score | ||||
1 | 42 (29.0) | 98 (67.6) | 5 (3.4) | |
2 | 23 (34.3) | 42 (62.7) | 2 (3.0) | 0.6 |
3 | 2 (66.7) | 1 (33.3) | 0 (0.0) | |
MOSE 2 score | ||||
1 | 1 (100) | 0 (0.0) | 0 (0.0) | |
2 | 14 (48.3) | 14 (48.3) | 1 (3.4) | 0.1 |
3 | 52 (28.1) | 127 (68.6) | 6 (3.2) | |
MOSE 2 | ||||
Inconclusive | 1 (100) | 0 (0.0) | 0 (0.0) | 0.3 |
Conclusive | 66 (30.8) | 141 (65.9) | 7 (3.3) |
Values are presented as number (%)..
MOSE, macroscopic on-site evaluation; FNB, fine needle biopsy..
Table 7 . Final Diagnosis and Number of Needle Passes in Relation to MOSE 2 after Excluding Score 1 (One Patient).
MOSE 2 | |||
---|---|---|---|
Score 2 ( | Score 3 ( | ||
Final diagnosis | |||
Inconclusive | 2 (6.9) | 1 (0.5) | 0.006 |
Conclusive (benign, malignant) | 27 (93.1) | 184 (99.5) | |
Final diagnosis | |||
Inconclusive | 2 (6.9) | 1 (0.5) | |
Benign | 6 (20.7) | 12 (6.5) | 0.0007 |
Malignant | 21 (72.4) | 172 (93.0) | |
No. of needle passes | |||
1 | 14 (48.3) | 52 (28.1) | 0.087 |
2 | 14 (48.3) | 127 (68.6) | |
3 | 1 (3.4) | 6 (3.2) |
Values are presented as number (%)..
MOSE, macroscopic on-site evaluation..
There was no statistically significant difference between FNA and FNB regarding final diagnosis, number of needles passes, and MOSE score 1. There is a significant difference with respect to MOSE score 2 (Table 8).
Table 8 . Final Diagnosis, Number of Needle Passes and MOSE Score in Relation to Type of Needle.
Type of needle | |||
---|---|---|---|
FNA ( | FNB ( | ||
Final diagnosis | |||
Inconclusive | 1 (3.2) | 2 (1.1) | 0.3 |
Conclusive (benign, malignant) | 30 (96.8) | 181 (98.9) | |
Final diagnosis | |||
Inconclusive | 1 (3.2) | 2 (1.1) | |
Benign | 5 (16.1) | 13 (7.1) | 0.1 |
Malignant | 25 (80.6) | 168 (91.8) | |
No. of needle passes | |||
1 | 12 (38.7) | 54 (29.5) | 0.587 |
2 | 18 (58.1) | 123 (67.2) | |
3 | 1 (3.2) | 6 (3.3) | |
MOSE score 1 | |||
1 | 16 (51.6) | 129 (70.5) | |
2 | 14 (45.2) | 53 (29.0) | 0.059 |
3 | 1 (3.2) | 1 (0.5) | |
MOSE score 2 | |||
Score 2 | 14 (45.2) | 15 (8.2) | < 0.001 |
Score 3 | 17 (54.8) | 168 (91.8) |
Values are presented as number (%)..
MOSE, macroscopic on-site evaluation; FNA, fine needle aspiration; FNB, fine needle biopsy..
Evaluating the suitability of EUS specimens obtained using MOSE is a simple method designed to improve the assessment of diagnostic accuracy and reduce the costs associated with cytopathologist involvement.6
A more accurate way to judge adequacy of histologic cores is by gross visual inspection; as a result, it may yield better diagnostic results with fewer needle passes.6
Our primary goal was to evaluate MOSE’s performance in determining whether histologic core specimens are adequate. The percentage of accurate diagnoses will be used to determine the total diagnostic accuracy. Evaluating the MOSE accuracy of solid lesions’ (EUS-FNB) versus (EUS-FNA) was an additional goal.
In current study there was highly statistical significant difference between the group with inconclusive results and the group with conclusive benign and malignant histopathological results regarding MOSE score (with
There was study done by Oh et al7 on 79 patients with pancreatic and extrapancreatic lesions underwent EUS-FNB and tissue core with MOSE was evaluated. Visible cores were observed in 94.9% (75/79). 5.1% (4/79) of the specimens had blood clots and sparse tissue cores. Ninety-two percent (73/79) of the histologic samples were deemed sufficient after microscopic evaluation. The results showed that the diagnostic accuracy, specificity, and sensitivity were, in order, 94.3%, 100%, and 94.5%. According to the study’s findings, MOSE provides a sufficient tissue core while lowering blood contamination. In the event that ROSE is unavailable, MOSE can also enhance the diagnostic yield.7
According to Ishikawa et al,8 a macroscopic visible tissue core equal to or greater than 4 mm on MOSE of sample acquired by EUS-FNA using a 19 G needle could be an indicator of tissue adequacy and could improve diagnostic accuracy.
However Mangiavillano et al9 found that there is no statistically significant differences between conventional EUS-FNB and EUS-FNB with MOSE as regard the diagnostic accuracy {87.8% (95% confidence interval [CI], 82.1%–92.2%) vs. 90.0% (95% CI, 84.8%–93.9%);
An randomized controlled trial (RCT) conducted recently on 244 patients undergoing EUS plus FNA with a 19 G needle revealed that the diagnostic yield of the MOSE technique (92.6 %) was comparable to that of the conventional technique (89.3 %;
Gaia et al11 assess the diagnostic yield of the MOSE approach in comparison to the standard technique during the FNB of pancreatic and extra-pancreatic lesions. Comparable diagnostic accuracy was demonstrated by the MOSE score compared to the traditional method. MOSE, on the other hand, enables the endosonographer to assess the histologic core, has a higher diagnostic yield than the traditional method, and may require fewer needle passes.11
Different sensitivities and specificities of MOSE were reported among various studies, for examples, Sundaram et al12 reported that in MOSE (FNB) sensitivity was 96.12%, specificity was 100%, PPV was 100%, NPV was 83.9% and overall accuracy was 96.78%. Mohan et al4 reported that the accuracy in yielding a pathologic diagnosis by MOSE was 91.3%, pooled sensitivity was 91.5%, pooled specificity was 98.9%, pooled PPV was 98.8%, and pooled NPV was 55.5%. Leung Ki et al6 reported that diagnostic accuracy was 94%, PPV, NPV, sensitivity and specificity for malignancy were 100%, 81%, 92%, and 100%, respectively.
In a study by Sundaram et al12 77% of the 155 individuals had pancreatic head disease, 26 of the final diagnoses were not malignant, while 129 had malignant lesions. For ROSE using cytology, the corresponding sensitivity and specificity were 96.9% and 100%. MOSE’s histopathological study demonstrated a sensitivity of 96.1% and a specificity of 100%. Between the histological examination using MOSE and the ROSE with cytology, there was no statistically significant difference (
A meta-analysis encompassing 14 trials and a sample of 1,508 lesions underwent EUS was conducted. The accuracy of MOSE in yielding a pathologic diagnosis was 91.3% (95% CI, 88.6%–93.3%;
Our study showed no statistical significant difference between the two groups with conclusive and inconclusive cytopathological results regarding the specimen acquisition method (suction vs. capillary and suction) and this finding was supported with several studies.7,13,14
While suctioning during EUS-FNB may cause bloody sample, the use of filter paper in MOSE absorbs blood, may minimize this risk.
In the current study there was no statistical significant difference between the two groups with conclusive and inconclusive cytopathological results regarding size of the needle, number of needle passes and type of the needle (whether FNA and FNB).
Our results were consistent with Bor et al14 who found that there was no statistically significant difference between the two group with conclusive and inconclusive cytopathological results regarding the size of the needle. Cheng et al2 compared the accuracy of diagnosis of FNB vs. FNA for patients with intra-abdominal masses. They found that there was no statistically significant difference between FNA and FNB regarding final diagnosis, andnumber of needles passes and the diagnostic accuracy.2
Our results were consistent with de Moura et al15 who aimed to compare diagnostic accuracy between EUS-FNA and EUS-FNB in sampling of pancreatic lesions. A total of 574 patients (
According to a recent meta-analysis by Renelus et al16 that comprised 12 RCTs evaluating both malignant and non-malignant lesions, EUS-FNB outperformed EUS-FNA (87% and 81%,
Hassan et al19 sought to evaluate the diagnostic yield of FNA against FNB in patients undergoing EUS tissue sample without (ROSE) and pancreatic lesions. A total of 151 patients (
Recent multicenter series was performed on a 378 patients Mangiavillano et al.21 Two needle passes was the median (IQR 2–4). MOSE’s total diagnostic yield was 90% (86%–92% CI). A bigger needle diameter and three or more needle passes were factors linked to improved MOSE diagnostic accuracy.21
For EUS-FNA, 45.2% of the samples had score 2 in MOSE classification 2 while 54.8% had score 3 in MOSE classification 2. For EUS-FNB, 8.2% of the samples had score 2 in MOSE classification 2 while it had a very high number of samples of score 3 in MOSE classification 2 reaching up to 91.8%.
The present investigation demonstrated a highly statistically significant difference between FNA and FNB with respect to MOSE score 2, with no impact on the ultimate diagnosis. These findings can be explained by the fact that FNB displays an unbroken core sample that reveals the true architecture of the tissue, whereas FNA displays tissue samples at the cellular level so the endoscopist can judge the adequacy of the material by gross visual inspection. A meta-analysis encompassing eight RCTs involving 921 cases corroborated these findings. While there was no discernible difference in the diagnostic accuracy between the FNA and FNB groups, the use of FNB resulted in more adequate tissue than FNA. When utilizing FNB, fewer needle passes were required to extract sufficient tissue. Despite requiring fewer needle passes than FNA, FNB yields a greater diagnostic value.22
In conclusion, High diagnostic yield and accuracy utilizing various needle sizes were demonstrated by MOSE during EUS-FNB sampling, adding new information to the body of existing literature.
With MOSE, an endosonographer can assess the specimen, improving the diagnostic yield and possibly reducing the number of needle passes.
However, in the current study, TA was performed at tertiary referral hospitals by a cytopathologist and an expert endosonographer. Thus, less experienced operators may be unable to implement the results.
The authors acknowledge subjects for their participation and cooperation in this study.
None.
The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.
No potential conflict of interest relevant to this article was reported.
© The Society of Gastrointestinal Intervention. Powered by INFOrang Co., Ltd.