Indexed In : Science Citation Index Expanded(SCIE), MEDLINE,
Pubmed/Pubmed Central, Elsevier Bibliographic, Google Scholar,
Databases(Scopus & Embase), KCI, KoreaMed, DOAJ
Gut and Liver is an international journal of gastroenterology, focusing on the gastrointestinal tract, liver, biliary tree, pancreas, motility, and neurogastroenterology. Gut atnd Liver delivers up-to-date, authoritative papers on both clinical and research-based topics in gastroenterology. The Journal publishes original articles, case reports, brief communications, letters to the editor and invited review articles in the field of gastroenterology. The Journal is operated by internationally renowned editorial boards and designed to provide a global opportunity to promote academic developments in the field of gastroenterology and hepatology. +MORE
Yong Chan Lee |
Professor of Medicine Director, Gastrointestinal Research Laboratory Veterans Affairs Medical Center, Univ. California San Francisco San Francisco, USA |
Jong Pil Im | Seoul National University College of Medicine, Seoul, Korea |
Robert S. Bresalier | University of Texas M. D. Anderson Cancer Center, Houston, USA |
Steven H. Itzkowitz | Mount Sinai Medical Center, NY, USA |
All papers submitted to Gut and Liver are reviewed by the editorial team before being sent out for an external peer review to rule out papers that have low priority, insufficient originality, scientific flaws, or the absence of a message of importance to the readers of the Journal. A decision about these papers will usually be made within two or three weeks.
The remaining articles are usually sent to two reviewers. It would be very helpful if you could suggest a selection of reviewers and include their contact details. We may not always use the reviewers you recommend, but suggesting reviewers will make our reviewer database much richer; in the end, everyone will benefit. We reserve the right to return manuscripts in which no reviewers are suggested.
The final responsibility for the decision to accept or reject lies with the editors. In many cases, papers may be rejected despite favorable reviews because of editorial policy or a lack of space. The editor retains the right to determine publication priorities, the style of the paper, and to request, if necessary, that the material submitted be shortened for publication.
Yu Han Zhao1 , Yu Zheng2
, Jie Sha3
, Hong Jin Hua4
, Ke Dong Li4
, Yu Lu1
, Yi Ni Dang1
, Guo Xin Zhang1
Correspondence to: Guo Xin Zhang
ORCID https://orcid.org/0000-0002-7103-3630
E-mail guoxinz@njmu.edu.cn
Yi Ni Dang
ORCID https://orcid.org/0000-0001-6449-516X
E-mail yeani_hi@126.com
Yu Han Zhao and Yu Zheng contributed equally to this work as first authors.
This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.
Gut Liver 2023;17(1):78-91. https://doi.org/10.5009/gnl220060
Published online September 2, 2022, Published date January 15, 2023
Copyright © Gut and Liver.
Background/Aims: The discrepancies between the diagnosis of preoperative endoscopic forceps biopsy (EFB) and endoscopic submucosal dissection (ESD) in patients with early gastric neoplasm (EGN) exist objectively. Among them, pathological upgrading directly influences the accuracy and appropriateness of clinical decisions. The aims of this study were to investigate the risk factors for the discrepancies, with a particular focus on pathological upgrading and to establish a prediction model for estimating the risk of pathological upgrading after EFB.
Methods: We retrospectively collected the records of 978 patients who underwent ESD from December 1, 2017 to July 31, 2021 and who had a final histopathology determination of EGN. A nomogram to predict the risk of pathological upgrading was constructed after analyzing subgroup differences among the 901 lesions enrolled.
Results: The ratio of pathological upgrading was 510 of 953 (53.5%). Clinical, laboratorial and endoscopic characteristics were analyzed using univariable and binary multivariable logistic regression analyses. A nomogram was constructed by including age, history of chronic atrophic gastritis, symptoms of digestive system, blood high density lipoprotein concentration, macroscopic type, pathological diagnosis of EFB, uneven surface, remarkable redness, and lesion size. The C-statistics were 0.804 (95% confidence interval, 0.774 to 0.834) and 0.748 (95% confidence interval, 0.664 to 0.832) in the training and validation set, respectively. We also built an online webserver based on the proposed nomogram for convenient clinical use.
Conclusions: The clinical value of identifying the preoperative diagnosis of EGN lesions is limited when using EFB separately. We have developed a nomogram that can predict the probability of pathological upgrading with good calibration and discrimination value.
Keywords: Early gastric neoplasm, Pathological upgrading, Prediction model, Endoscopic forceps biopsy, Endoscopic submucosal dissection
Early gastric neoplasm (EGN) is a group including low grade intraepithelial neoplasia (LGIN), high grade intraepithelial neoplasia (HGIN), and early gastric carcinoma (EGC).1-4 LGIN and HGIN, for which the differences lie in the degree of cellular or structural heterogeneity, are ascribed as dysplasia and classified as precursor lesions. In current guidelines, endoscopic submucosal dissection (ESD) is preferred for EGN to prevent the deterioration of EGC to progressive gastric cancer, which has the following advantages: minimally invasive, faster recovery, higher curative ratio, and preservation of gastric function.5-8 Advanced endoscopic instruments, such as chromoendoscopy, high-definition white-light endoscopy, narrow-band imaging, are not yet widely applied and not easily mastered, and conventional white-light endoscopy in combination with endoscopic forceps biopsy (EFB) is under routine practice in the clinical context.
EFB is an important tool for the histopathological diagnosis of EGN. With multiple EFB, the diagnostic efficiency would be improved to some extent based on literature review and our clinical experience.9,10 However, further biopsies were not able to increase positive diagnosis cumulative rates, and may increase workload and medical cost, and may lead to trauma, bleeding, or perforation.11,12 On the other hand, the pathological outcomes of EFB and ESD are often inconsistent. These discrepancies could be divided into downgrading and upgrading. It has been reported that the overall rate of discrepancy between EFB and ESD ranges from 20.1% to 76.3%,10,11 with 16.3% to 44.9% pathological upgrading ultimately.7-9,12-16 In addition, the pathological upgrading group showed a significant tendency of submucosal invasion and lymphovascular/perineural invasion, which would lead to endoscopic noncurative resection and suboptimal prognostic results.17,18 Therefore, the question in front of us is how to classify patients with increased risk of pathological upgrading upon EFB. Indeed, it may be attributed to the experience of the endoscopist, but also highly associated with morphological features or biological characteristics of the lesion itself. Hitherto the risk factors associated with pathological upgrading remain unclear and the lack of a visual assessment model in clinical practice has led to insignificant improvements in diagnostic compliance.16
Considering the above concerns, this retrospective study was designed: (1) to investigate the risk factors of histological discrepancies between EFB and ESD, with a particular focus in pathological upgrading, and (2) to establish a prediction model in estimating the risk of pathological upgrading upon EFB.
This is a retrospective single-center cohort study to explore the risk factors for pathological upgrading between EFB and ESD in patients with EGN. The secondary outcome was to construct a nomogram to show the predictive efficacy of detected risk factors on pathological upgrading. This clinical study consisted of two stages. The first stage (training set) retrospectively identified patients diagnosed with EGN according to EFB or ESD in the First Affiliated Hospital of Nanjing Medical University from December 2017 to December 2020. The second stage (validation set) included patients from January 2021 to July 2021. The detailed procedure of the study is shown in Fig. 1. This study was conducted in accordance with the Declaration of Helsinki and was approved by the Institutional Review Board of the First Affiliated Hospital of Nanjing Medical University (IRB number: 2021-SR-227). It has been previously registered in Chinese Clinical Trial Registry (ChiCTR2100048088). Informed consent was waived due to the nature of the study design.
From December 2017 to July 2021, a total of 1,863 patients underwent gastric ESD. Indications followed the Japanese gastric cancer treatment guidelines 2018 (5th edition)
Collection of biopsies before ESD were conducted in our center and the pathological diagnosis was confirmed by at least two experienced pathologists. However, according to the routine clinical practice, part of EFB may be conducted prior to the patients’ admission to our center, therefore repeating endoscopic procedures would be performed for non-neoplastic pathology or questionable discrepancies between the pathologic results obtained at our institution and at the referring hospital.14 The gastroenterologist would evaluate lesions for the indication of ESD based on pathological and endoscopic records. The ESD consent form was signed on the patient’s own wishes. In fact, even though additional surgeries may be required due to exceeding ESD indications, the majority of target patients in our institution chose to undergo ESD first, based on the advantages of ESD, even though additional surgeries may be required due to exceeding ESD indications,17,18,21 which ensured us screening up sufficient study population.
In this study, we followed the World Health Organization pathological diagnostic criteria in digestive system.1 We enrolled patients with different pathological outcomes (e.g., acute inflammation, chronic gastric atrophy, gastric intestinal metaplasia, LGIN, HGIN, and intramucosal carcinoma) and prespecified principles to define pathological upgrading and non-upgrading according to EFB and ESD examinations. For instance, if one patient showed LGIN during EFB while the pathological grade was reported as HGIN or carcinoma following ESD, this was identified as pathological upgrading. On the other hand, if one patient showed lower or consistent pathological grade after ESD as compared to that from EFB, they were finally assigned as non-pathological upgrading.14,22
Demographic (e.g., age, sex, and body mass index) and clinical data (e.g.,
Individual medical history and digestive comorbidity was defined positive as written down in medical records or outpatient documents. Symptoms of digestive system were considered as existing if the following words appeared in the records: abdominal pain, nausea, vomiting, dyspepsia, eructation, heartburn, loss of appetite, weight loss.
Ulceration or ulcer scar was considered positive after excluding those caused by the biopsy.23 Remarkable redness was defined as discoloration on the mucosal surface of the lesion compared to the surrounding mucosa, and uneven surface was defined as the presence of irregularly raised or nodular mucosa.24,25 Fig. 2 shows the following endoscopic features in sequence: remarkable redness, ulceration or scar and uneven surface.
All of the following statistical analyses were performed with SPSS 23.0 (IBM Corp., Armonk, NY, USA). Continuous variables were presented as means and standard deviations or medians with interquartile ranges. Categorical variables were demonstrated as numbers and percentages. Differences between groups were assessed using the chi-square test, the Student t-test, or the Mann-Whitney test in the univariate analysis, as appropriate. Variables detected with p<0.05 in the univariate logistic regression were further included in the multivariate logistic regression analysis to determine the variables that were independent influencing factors of pathological upgrading.
A predictive nomogram was constructed, based on the variables selected with the multivariate logistic regression analysis and by using the package of rms in R software (rms in R version 4.0.4, R Foundation for Statistical Computing, Vienna, Austria; https://www.r-project.org/). The model was validated internally in the training set and externally in the validation set. A calibration plot with bootstrapping (1,000 replications) was used to illustrate the association between the actual probability and the predicted probability. In addition, five training models were constructed to compare the robustness of each model. The performance of the prediction model was evaluated using the area under the curve (AUC) in receiver operating characteristic analysis. Decision curve analysis was used to explain the utility of benefits in clinical use.
We retrospectively collected data for 1,053 lesions from 978 patients who underwent ESD for EGN. The comparison of diagnosis between the original EFB and the final ESD is summarized in Table 1. The predominant pathological upgrading involved upgrading from LGIN to HGIN or carcinoma (n=178) and upgrading from HGIN to carcinoma (n=270). After excluding lesions of which pathological diagnosis with EFB were carcinoma, 822 lesions were enrolled into the training and 131 were enrolled into validation sets. Clinical and endoscopic characteristics are summarized in Tables 2 and 3. Lesions (54.0%, 444/822) showed higher-grade disease after ESD in the training set (LGIN from non-neoplasia, HGIN or carcinoma from LGIN or non-neoplasia, or carcinoma from HGIN).
Table 1 Comparison of Pathological Diagnosis by Initial Biopsy and Final Endoscopic Resection
Pathologic diagnosis with EFB | Pathologic diagnosis with ESD | ||||
---|---|---|---|---|---|
Non-neoplasia | LGIN | HGIN | Carcinoma | Total No. (%) | |
Non-neoplasia | 15† | 42* | 11* | 10* | 78 (7.4) |
LGIN | 11† | 293† | 108* | 70* | 482 (45.8) |
HGIN | 1† | 29† | 93† | 270* | 393 (37.3) |
Carcinoma | 0† | 5† | 4† | 91† | 100 (9.5) |
Total No. (%) | 27 (2.6) | 369 (35.0) | 216 (20.5) | 441 (41.9) | 1,053 |
EFB, endoscopic forceps biopsy; ESD, endoscopic submucosal dissection; LGIN, low grade intraepithelial neoplasia; HGIN, high grade intraepithelial neoplasia.
*Pathological upgrading; †Non-pathological upgrading.
Table 2 Clinical Characteristics of the Training and Validation Sets
Characteristics | Training set (n=822) | Validation set (n=131) |
---|---|---|
Age, yr | 63 (55–69) | 63 (57–70) |
Male sex | 592 (72.0) | 103 (78.6) |
Body mass index, kg/m2 | 23.8±3.3 | 24.1±2.9 |
Negative | 289 (35.1) | 45 (36.4) |
Previous eradicated or infected | 533 (64.9) | 86 (65.6) |
Symptoms of digestive system | 599 (72.9) | 56 (42.7) |
Smoking | 198 (24.1) | 44 (33.6) |
Drinking | 78 (20.6) | 24 (18.3) |
Comorbidity | ||
History of CAG | 466 (56.7) | 61 (46.6) |
History of peptic ulcer | 137 (16.7) | 13 (9.9) |
Medical history | ||
Hypertension | 258 (31.4) | 58 (44.3) |
Diabetes | 74 (9.0) | 14 (10.7) |
Coronary heart disease | 33 (4.0) | 3 (2.3) |
History of cholecystectomy | 58 (7.1) | 19 (14.5) |
History of malignancy | 80 (9.7) | 4 (3.1) |
Family history of malignancy | 133 (16.2) | 12 (9.2) |
Laboratory examination | ||
CEA, ng/mL | 3.4±4.6 | 2.8±4.4 |
CA19-9, U/mL | 12.7±28.2 | 12.8±8.0 |
CA724, U/mL | 3.9±8.3 | 4.0±5.0 |
HDL, mmol/L | 1.2±0.3 | 1.1±0.3 |
LDL, mmol/L | 2.9±0.7 | 2.8±0.6 |
Albumin, g/L | 39.9±3.9 | 39.0±3.0 |
Calcium, mmol/L | 2.3±0.1 | 2.3±0.1 |
Data are presented as median (range), number (%), or mean±SD.
CAG, chronic atrophic gastritis; CEA, carcinoembryonic antigen; CA19-9, carbohydrate antigen 19-9; CA724, carbohydrate antigen 724; HDL, high density lipoprotein; LDL, low density lipoprotein.
Table 3 Endoscopic Characteristics of the Training and Validation Sets
Characteristics | Training set (n=822) | Validation set (n=131) |
---|---|---|
Locations of lesions | ||
Upper third | 241 (29.3) | 40 (30.5) |
Middle third | 93 (11.3) | 21 (16.0) |
Lower third | 488 (59.4) | 70 (53.4) |
Macroscopic type of lesions | ||
Elevated | 405 (49.3) | 55 (42.0) |
Flat | 197 (24.0) | 26 (19.8) |
Depressed | 220 (26.8) | 50 (38.2) |
Pathologic diagnosis of EFB | ||
Non-neoplasia | 74 (9.0) | 4 (3.1) |
LGIN | 413 (50.2) | 69 (52.7) |
HGIN | 335 (40.8) | 58 (44.3) |
Pathologic diagnosis of ESD | ||
Non-neoplasia | 27 (3.3) | |
LGIN | 314 (38.2) | 50 (38.2) |
HGIN | 179 (21.8) | 33 (252) |
Carcinoma | 302 (36.7) | 48 (36.8) |
Pathologic upgrading | ||
Yes | 444 (54.0) | 66 (50.4) |
No | 378 (46.0) | 65 (49.6) |
Multiple lesions | 52 (13.8) | 24 (18.3) |
Endoscopic features | ||
Ulceration or scar | 167 (20.3) | 39 (29.8) |
Uneven surface | 364 (44.3) | 64 (48.9) |
Remarkable redness | 531 (64.6) | 88 (67.2) |
Lesion size | ||
≤2 cm | 581 (70.7) | 78 (59.5) |
>2 cm | 241 (29.3) | 53 (40.5) |
Procedure time, min | 60 (45–90) | 60 (45–90) |
Depth of invasion | ||
M | 763 (92.8) | 126 (96.2) |
SM1 | 29 (3.5) | 1 (0.8) |
SM2 | 30 (3.6) | 4 (3.1) |
Data are presented as number (%) or median (range).
EFB, endoscopic forceps biopsy; LGIN, low grade intraepithelial neoplasm; HGIN, high grade intraepithelial neoplasm; ESD, endoscopic submucosal dissection; M, intraepithelial and/or the deepest infiltration depth is within the mucosal lamina propria; SM1, the infiltration depth >500 μm from the muscularis mucosae; SM2, the infiltration depth >500 μm from the muscularis mucosae.
As shown in Table 4, statistical differences were detected in those with history of smoking, drinking, chronic atrophic gastritis (CAG), and symptoms of digestive system. History of CAG was less commonly detected in the pathological upgrading group than in the non-pathological upgrading group (50.2% vs 60.9%, p<0.001). The rate of
Table 4 Clinical Characteristics of Lesions with or without Pathological Upgrading
Characteristics | Non-pathological upgrading group (n=378) | Pathological upgrading group (n=444) | p-value* |
---|---|---|---|
Age, yr | 63 (55–69) | 65 (58–70) | <0.001 |
Male sex | 250 (66.1) | 342 (77.0) | 0.001 |
Body mass index, kg/m2 | 23.6±3.1 | 23.9±3.4 | 0.082 |
0.680 | |||
Negative | 131 (34.7) | 160 (36.0) | |
Previous eradicated or infected | 247 (65.3) | 284 (64.0) | |
Symptoms of digestive system | 260 (68.8) | 339 (76.4) | 0.015 |
Smoking | 105 (27.8) | 159 (35.8) | 0.014 |
Drinking | 78 (20.6) | 120 (27.0) | 0.033 |
Comorbidity | |||
History of CAG | 243 (64.3) | 223 (50.2) | <0.001 |
History of peptic ulcer | 71 (15.4) | 82 (18.5) | 0.224 |
Medical history | |||
Hypertension | 104 (27.5) | 154 (34.7) | 0.073 |
Diabetes | 30 (7.9) | 44 (9.9) | 0.270 |
Coronary heart disease | 11 (2.9) | 22 (5.0) | 0.269 |
History of cholecystectomy | 22 (5.8) | 36 (8.1) | 0.236 |
History of malignancy | 35 (9.3) | 45 (10.1) | 0.531 |
Family history of malignancy | 61 (16.1) | 72 (16.2) | 0.700 |
Laboratory examination | |||
CEA, ng/mL | 3.3±4.5 | 3.5±4.4 | 0.398 |
CA19-9, U/mL | 12.4±30.5 | 13.1±26.2 | 0.734 |
CA724, U/mL | 4.0±9.7 | 3.8±6.9 | 0.682 |
HDL, mmol/L | 1.2±0.3 | 1.1±0.2 | 0.015 |
LDL, mmol/L | 2.9±0.7 | 2.9±0.7 | 0.099 |
Albumin, g/L | 40.2±3.6 | 39.7±4.2 | 0.056 |
Calcium, mmol/L | 2.3±0.1 | 2.2±0.1 | 0.004 |
Data are presented as median (range), number (%), or mean±SD.
CAG, chronic atrophic gastritis; CEA, carcinoembryonic antigen; CA19-9, carbohydrate antigen 19-9; CA724, carbohydrate antigen 724; HDL, high density lipoprotein; LDL, low density lipoprotein.
*p-value was derived from the chi-square test, Student t-test, or Mann-Whitney test.
The contradistinction of endoscopic characteristics is presented in Table 5. Pathological upgrading more frequently occurred in lower third (51.1%), followed by upper third (36.5%) and middle third (12.4%) in the upgrading group. The predominant macroscopic type in both groups was elevated pattern, with a gradual tendency of depressed pattern in the pathological upgrading group. The endoscopic ulceration, uneven surface, remarkable redness was found to account for 26.8%, 56.1%, and 75.4% in cases of pathological upgrading, accordingly.
Table 5 Endoscopic Characteristics and Posttreatment Evaluation of the Lesions with or without Pathological Upgrading
Characteristics | Non-pathological upgrading group (n=378) | Pathological upgrading group (n=444) | p-value* |
---|---|---|---|
Locations of lesions | <0.001 | ||
Upper third | 79 (20.9) | 162 (36.5) | |
Middle third | 38 (10.1) | 55 (12.4) | |
Lower third | 261 (69.0) | 227 (51.1) | |
Macroscopic type of lesions | <0.001 | ||
Elevated | 206 (54.5) | 199 (44.8) | |
Flat | 106 (28.0) | 91 (20.5) | |
Depressed | 66 (17.5) | 154 (34.7) | |
Pathologic diagnosis of EFB | <0.001 | ||
Non-neoplasia | 16 (3.5) | 58 (13.1) | |
LGIN | 261 (56.7) | 152 (34.2) | |
HGIN | 101 (22.0) | 234 (52.7) | |
Multiple lesions | 52 (13.8) | 68 (15.3) | 0.528 |
Endoscopic features | |||
Ulceration or scar | 48 (12.7) | 119 (26.8) | <0.001 |
Uneven surface | 115 (30.4) | 249 (56.1) | <0.001 |
Remarkable redness | 197 (52.1) | 334 (75.4) | <0.001 |
Lesion size | <0.001 | ||
≤2 cm | 306 (81.0) | 275 (61.9) | |
>2 cm | 72 (19.0) | 169 (38.1) | |
Procedure time, min | 60 (45–90) | 60 (45–100) | <0.001 |
Depth of invasion | <0.001 | ||
M | 378 (95.0) | 385 (86.7) | |
SM1 | - | 29 (6.5) | |
SM2 | - | 27 (6.8) |
Data are presented as number (%) or median (range).
EFB, endoscopic forceps biopsy; LGIN, low grade intraepithelial neoplasm; HGIN, high grade intraepithelial neoplasm; M, intraepithelial and/or the deepest infiltration depth is within the mucosal lamina propria; SM1, the infiltration depth >500 μm from the muscularis mucosae; SM2, the infiltration depth >500 μm from the muscularis mucosae.
*p-value was derived from chi-square test or Mann-Whitney test.
Multivariate analysis (logistic regression analysis) was further performed to analyze the variables that were statistically significant in univariate analysis. Ultimately, a total of nine variables which pertained to clinical, laboratorial, endoscopic characteristics separately were enrolled to build the clinical prediction model. As revealed in Table 6, history of CAG (odds ratio [OR], 0.59; 95% confidence interval [CI], 0.42 to 0.84; p=0.003), blood high density lipoprotein (HDL) concentration (OR, 0.46; 95% CI, 0.24 to 0.86; p<0.016) were protective predictors. Age, symptoms of digestive system, macroscopic type of lesions, pathologic diagnosis of EFB, remarkable redness were risk factors, while uneven surface (OR, 2.35; 95% CI, 1.66 to 3.32; p<0.001), lesion size >2 cm (OR, 2.20; 95% CI, 1.50 to 3.22; p<0.001) were independent risk predictors.
Table 6 Univariate and Multivariate Regression Analyses of the Predictors Associated with Pathological Upgrading According to the Analyzed Variables
Factor | Univariate logistic analysis | Multivariate logistic analysis | |||
---|---|---|---|---|---|
OR (95% CI) | p-value | OR (95% CI) | p-value | ||
Age | 1.05 (1.03–1.06) | <0.001 | 1.04 (1.03–1.06) | <0.001 | |
Sex (ref. male) | 0.60 (0.44–0.82) | 0.001 | |||
Smoking (ref. no) | 1.38 (1.02–1.87) | 0.036 | |||
Drinking (ref. no) | 1.39 (1.03–1.89) | 0.034 | |||
Comorbidity: CAG (ref. no) | 0.47 (0.35–0.63) | <0.001 | 0.47 (0.33–0.68) | <0.001 | |
Symptoms of digestive system (ref. no) | 1.44 (1.05–1.97) | 0.025 | 1.60 (1.08–2.36) | 0.019 | |
Laboratory examination: HDL | 0.34 (0.20–0.58) | <0.001 | 0.44 (0.24–0.81) | 0.008 | |
Laboratory examination: calcium | 0.09 (0.02–0.37) | 0.001 | |||
Locations of lesions (ref. upper third) | 1.00 | ||||
Middle third | 0.71 (0.43–1.18) | 0.187 | |||
Lower third | 0.40 (0.29–0.56) | 0.001 | |||
Macroscopic type of lesions (ref. elevated) | 1.00 | 1.00 | |||
Flat | 0.84 (0.60–1.19) | 0.331 | 0.74 (0.48–1.13) | 0.164 | |
Depressed | 2.25 (1.68–3.20) | <0.001 | 1.63 (1.13–2.32) | 0.045 | |
Pathological diagnosis of EFB (ref. non-neoplasia) | 1.00 | 1.00 | |||
LGIN | 0.01 (0.01–0.08) | <0.001 | 0.01 (0.01–0.06) | <0.001 | |
HGIN | 0.04 (0.01–0.30) | 0.002 | 0.02 (0.01–0.13) | <0.001 | |
Endoscopic features: ulceration or scar (ref. no) | 2.63 (1.79–3.84) | <0.001 | |||
Endoscopic features: uneven surface (ref. no) | 2.69 (2.01–3.60) | <0.001 | 2.24 (1.58–3.19) | <0.001 | |
Endoscopic features: remarkable redness (ref. no) | 2.95 (2.19–3.99) | <0.001 | 2.26 (1.57–3.27) | <0.001 | |
Lesion size (ref. ≤2 cm) | 2.47 (1.78–3.41) | <0.001 | 2.20 (1.49–3.24) | <0.001 |
OR, odds ratio; CI, confidence interval; ref., reference group; CAG, chronic atrophic gastritis; HDL, high density lipoprotein; EFB, endoscopic forceps biopsy; LGIN, low grade intraepithelial neoplasm; HGIN, high grade intraepithelial neoplasm.
We developed a nomogram to predict the risk of pathological upgrading in patients with EGN (Fig. 3). For instance, a 65-year-old patient, with a previous history of CAG, without symptoms of digestive system, and blood HDL concentration of 1.0 mmol/L, was found to have an elevated lesion with remarkable redness and uneven surface, and less than 2 cm in diameter at endoscopy. If the EFB pathology was HGIN, the incidence of pathological upgrading would be 65.1% (95% CI, 0.522 to 0.761). The online version of our nomogram software is accessible at https://zyh-njmu.shinyapps.io/Dynamic_nomogram/.
We performed validation by using the bootstrap method with 1,000 repetitions to display the calibration of the nomogram (Fig. 4A and B). Accuracy of the nomogram was examined with AUC. The C-statistics for our nomogram model was 0.804 (95% CI, 0.774 to 0.834) and 0.748 (95% CI, 0.664 to 0.832) in the training and validation set, respectively (Fig. 4C and D).
To assess the predictive ability of our nomogram, several training schemes (namely training models 1 to 5) were constructed (Fig. 5). Afterwards, we performed internal validation and AUC computation for training models 1 to 5 and EFB, and the calibration plot and C-statistics for each model were shown in Figs 6 and 7, respectively.
EFB and ESD are used for diagnosis of EGN. However, EFB is not able to show histology of the entire lesion, which may lead to discrepancy between histological diagnosis of EFB and resected specimen. Therefore, it is urgently needed to establish a clinical predictive tool to identify the risk of pathological upgrading. Nomogram is a visualization prediction tool that can incorporate different variables affecting outcomes.26 We developed a nomogram for pathological upgrading in patients received EFB before ESD which showed a steady accuracy in evaluating the risk of pathological upgrading.
In the current study, rate of pathological upgrading was 44.9%. It was similar to the results of previous studies.15,27-31 Sixty-nine point nine percent of HGIN in EFB were diagnosed as EGC in our study, which was comparable to reported results.30,32,33 It has been found that pathological upgrading was positively correlated with longer operation time and deeper submucosal invasion. In addition, the depth of invasion was verified in predicting the risk of nodal metastasis in EGCs.17,34,35
In agreement with reported data, our study demonstrated that the depressed pattern, nodular surface, surface redness, lesion size >2 cm, and location in the upper third of the stomach were risk factors in pathological upgrading.23,36,37 EGCs located in the lower third of the stomach, especially in the antrum, might be detected easily, but for those in the upper third, detection may be difficult due to denser mucosal folds, thinner stomach, and limited endoscopic visual fields.36
Ideally, one would incline to have a validated noninvasive biomarker to predict cancer risk. However, no positive values of traditional tumor markers in predicting pathological upgrading have been observed in our study. We found that associations between pathological upgrading and metabolic features such as body mass index, hypertension, and dyslipidemia were not significant except for HDL. A 2015 meta-analysis, which summarized studies associating apolipoprotein E gene and cancer, demonstrated that reduced circulating HDL might be a potentially causal risk factor for the development of overall cancer in Asians.38 As studies from several other centers have shown, the decreased HDL level was related to increased risk of gastric cancer and advanced disease stage.39-42 One possible explanation is that HDL may promote cancer development by generating reactive oxygen species, increasing hormone production and availability (e.g., insulin-like growth factor, insulin, and adipokines), and forming an energy rich environment. This imbalance of hormones, the redox system, and energy availability plays a role in epithelium transition.43,44
It was interesting to note that our study showed significant statistical differences in both univariate and multivariate analyses for history of CAG and symptoms of digestive system between groups. The intestinal type of gastric cancer typically follows the Correa precancerous cascade of changes initiated by a non-self-limiting inflammation, of which CAG represents the relatively early stage.45 With the upgrading of gastric carcinogenesis, non-specific conditions such as gastritis and ulceration may appear.46,47 Poor prognosis from gastric cancer is mainly due to late presentation. Alarm symptoms of EGN include abdominal or epigastric pain or discomfort, nausea and vomiting, etc.48,49 EGN is usually asymptomatic. Therefore, referral guidelines have been developed to encourage early detection.50,51 Normally, endoscopists would be more careful with symptomatic patients during endoscopy.52,53 We assumed that patients who had CAG or suffered from digestive symptoms may be concerned about their digestive system health and would go through endoscopic checks.52,54 Therefore, the history of CAG and symptoms of digestive system seem to be protective factors.
Here, we reported predictors for pathological upgrading, also we established a nomogram in a visualized pattern. Another strength of our study was that our predictors included clinical, laboratorial and endoscopic features, and its performance was verified in the comparison with other training models (Figs 7 and 8). The nomogram yielded the highest AUC demonstrating the best discrimination in estimating the risk of pathological upgrading in patients with EGN. The final decision curve analysis showed that as long as setting the threshold probability of patients or clinicians beyond 10% (i.e., if no intervention was provided for patient with pathological upgrading was considered probably inappropriate), screening strategies based on our nomogram’s pathological upgrading risk estimates resulted in superior net benefit than screen-none or screen-all strategies (Fig. 7). When we set the threshold up to 40% or higher, the predictive effect of nomogram was still better than that of any training model. For practical application, we recommended explaining the upgrading risk predicted by nomogram to patients based on EFB findings. For EFB diagnosis of HGIN, intervention of ESD is necessary regardless of upgrading.6,55 For LGIN or non-neoplasia, a predicting value >50% indicates a greater risk of misdiagnosis, and secondary examination or immediate endoscopic intervention would be recommended.
This study has some limitations. First of all, this was a retrospective cohort study, in which data were mainly obtained by reviewing medical records or endoscopic results from a single center. Thus, it may restrict application to a wider population. Second, the operational differences between primary examiner at the local clinic and the procedural endoscopic resection operator, as well as the impact of random/guided biopsies, number of biopsies, could not be evaluated.4,56 Several studies have reported that white-light endoscopy combined with narrow-band imaging is superior to white-light endoscopy alone in identifying patients with gastric intestinal metaplasia and dysplasia.57,58 In fact, there were still many medical institutions in which virtual chromoendoscopy and high-definition processors and screens were not routinely available. Here, we enrolled no endoscopic features other than from white-light endoscopy.
In conclusion, pathological upgrading in EGN is common with specific clinical, laboratorial and endoscopic characteristics. Based on the above identified risk factors, a nomogram was developed to objectively and accurately predict individualized pathological upgrading risk of patients with EGN before ESD.
This study was supported by the Jiangsu Municipal Science and Technology Bureau (No. BE2020784), the National Natural Science Foundation of China (No. 82000529), and the Natural Science Foundation of Jiangsu Province (No. SBK 2020042595). The funding body had no role in the study design, data collection, analysis, and interpretation of data.
The authors appreciate the study participants and the inestimable support from the research center. The authors are grateful for support of the medical and research staff of this study.
No potential conflict of interest relevant to this article was reported.
Study concept and design: Y.H.Z., Y.N.D., G.X.Z. Data acquisition: Y.H.Z., Y.L., J.S. Pathological support: H.J.H., K.D.L. Data analysis and interpretation: Y.H.Z. Drafting of the manuscript: Y.H.Z., J.S. Critical revision of the manuscript for important intellectual content: Y.Z. Statistical analysis: Y.H.Z., Y.Z. Obtained funding: Y.N.D., G.X.Z. Study supervision: Y.N.D., G.X.Z. Approval of final manuscript: all authors.
Gut and Liver 2023; 17(1): 78-91
Published online January 15, 2023 https://doi.org/10.5009/gnl220060
Copyright © Gut and Liver.
Yu Han Zhao1 , Yu Zheng2
, Jie Sha3
, Hong Jin Hua4
, Ke Dong Li4
, Yu Lu1
, Yi Ni Dang1
, Guo Xin Zhang1
Departments of 1Gastroenterology and 2Rehabilitation Medicine, The First Affiliated Hospital of Nanjing Medical University, Nanjing, 3Department of Gastroenterology, Jingjiang People's Hospital, Jingjiang, and 4Department of Pathology, The First Affiliated Hospital of Nanjing Medical University, Nanjing, China
Correspondence to:Guo Xin Zhang
ORCID https://orcid.org/0000-0002-7103-3630
E-mail guoxinz@njmu.edu.cn
Yi Ni Dang
ORCID https://orcid.org/0000-0001-6449-516X
E-mail yeani_hi@126.com
Yu Han Zhao and Yu Zheng contributed equally to this work as first authors.
This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.
Background/Aims: The discrepancies between the diagnosis of preoperative endoscopic forceps biopsy (EFB) and endoscopic submucosal dissection (ESD) in patients with early gastric neoplasm (EGN) exist objectively. Among them, pathological upgrading directly influences the accuracy and appropriateness of clinical decisions. The aims of this study were to investigate the risk factors for the discrepancies, with a particular focus on pathological upgrading and to establish a prediction model for estimating the risk of pathological upgrading after EFB.
Methods: We retrospectively collected the records of 978 patients who underwent ESD from December 1, 2017 to July 31, 2021 and who had a final histopathology determination of EGN. A nomogram to predict the risk of pathological upgrading was constructed after analyzing subgroup differences among the 901 lesions enrolled.
Results: The ratio of pathological upgrading was 510 of 953 (53.5%). Clinical, laboratorial and endoscopic characteristics were analyzed using univariable and binary multivariable logistic regression analyses. A nomogram was constructed by including age, history of chronic atrophic gastritis, symptoms of digestive system, blood high density lipoprotein concentration, macroscopic type, pathological diagnosis of EFB, uneven surface, remarkable redness, and lesion size. The C-statistics were 0.804 (95% confidence interval, 0.774 to 0.834) and 0.748 (95% confidence interval, 0.664 to 0.832) in the training and validation set, respectively. We also built an online webserver based on the proposed nomogram for convenient clinical use.
Conclusions: The clinical value of identifying the preoperative diagnosis of EGN lesions is limited when using EFB separately. We have developed a nomogram that can predict the probability of pathological upgrading with good calibration and discrimination value.
Keywords: Early gastric neoplasm, Pathological upgrading, Prediction model, Endoscopic forceps biopsy, Endoscopic submucosal dissection
Early gastric neoplasm (EGN) is a group including low grade intraepithelial neoplasia (LGIN), high grade intraepithelial neoplasia (HGIN), and early gastric carcinoma (EGC).1-4 LGIN and HGIN, for which the differences lie in the degree of cellular or structural heterogeneity, are ascribed as dysplasia and classified as precursor lesions. In current guidelines, endoscopic submucosal dissection (ESD) is preferred for EGN to prevent the deterioration of EGC to progressive gastric cancer, which has the following advantages: minimally invasive, faster recovery, higher curative ratio, and preservation of gastric function.5-8 Advanced endoscopic instruments, such as chromoendoscopy, high-definition white-light endoscopy, narrow-band imaging, are not yet widely applied and not easily mastered, and conventional white-light endoscopy in combination with endoscopic forceps biopsy (EFB) is under routine practice in the clinical context.
EFB is an important tool for the histopathological diagnosis of EGN. With multiple EFB, the diagnostic efficiency would be improved to some extent based on literature review and our clinical experience.9,10 However, further biopsies were not able to increase positive diagnosis cumulative rates, and may increase workload and medical cost, and may lead to trauma, bleeding, or perforation.11,12 On the other hand, the pathological outcomes of EFB and ESD are often inconsistent. These discrepancies could be divided into downgrading and upgrading. It has been reported that the overall rate of discrepancy between EFB and ESD ranges from 20.1% to 76.3%,10,11 with 16.3% to 44.9% pathological upgrading ultimately.7-9,12-16 In addition, the pathological upgrading group showed a significant tendency of submucosal invasion and lymphovascular/perineural invasion, which would lead to endoscopic noncurative resection and suboptimal prognostic results.17,18 Therefore, the question in front of us is how to classify patients with increased risk of pathological upgrading upon EFB. Indeed, it may be attributed to the experience of the endoscopist, but also highly associated with morphological features or biological characteristics of the lesion itself. Hitherto the risk factors associated with pathological upgrading remain unclear and the lack of a visual assessment model in clinical practice has led to insignificant improvements in diagnostic compliance.16
Considering the above concerns, this retrospective study was designed: (1) to investigate the risk factors of histological discrepancies between EFB and ESD, with a particular focus in pathological upgrading, and (2) to establish a prediction model in estimating the risk of pathological upgrading upon EFB.
This is a retrospective single-center cohort study to explore the risk factors for pathological upgrading between EFB and ESD in patients with EGN. The secondary outcome was to construct a nomogram to show the predictive efficacy of detected risk factors on pathological upgrading. This clinical study consisted of two stages. The first stage (training set) retrospectively identified patients diagnosed with EGN according to EFB or ESD in the First Affiliated Hospital of Nanjing Medical University from December 2017 to December 2020. The second stage (validation set) included patients from January 2021 to July 2021. The detailed procedure of the study is shown in Fig. 1. This study was conducted in accordance with the Declaration of Helsinki and was approved by the Institutional Review Board of the First Affiliated Hospital of Nanjing Medical University (IRB number: 2021-SR-227). It has been previously registered in Chinese Clinical Trial Registry (ChiCTR2100048088). Informed consent was waived due to the nature of the study design.
From December 2017 to July 2021, a total of 1,863 patients underwent gastric ESD. Indications followed the Japanese gastric cancer treatment guidelines 2018 (5th edition)
Collection of biopsies before ESD were conducted in our center and the pathological diagnosis was confirmed by at least two experienced pathologists. However, according to the routine clinical practice, part of EFB may be conducted prior to the patients’ admission to our center, therefore repeating endoscopic procedures would be performed for non-neoplastic pathology or questionable discrepancies between the pathologic results obtained at our institution and at the referring hospital.14 The gastroenterologist would evaluate lesions for the indication of ESD based on pathological and endoscopic records. The ESD consent form was signed on the patient’s own wishes. In fact, even though additional surgeries may be required due to exceeding ESD indications, the majority of target patients in our institution chose to undergo ESD first, based on the advantages of ESD, even though additional surgeries may be required due to exceeding ESD indications,17,18,21 which ensured us screening up sufficient study population.
In this study, we followed the World Health Organization pathological diagnostic criteria in digestive system.1 We enrolled patients with different pathological outcomes (e.g., acute inflammation, chronic gastric atrophy, gastric intestinal metaplasia, LGIN, HGIN, and intramucosal carcinoma) and prespecified principles to define pathological upgrading and non-upgrading according to EFB and ESD examinations. For instance, if one patient showed LGIN during EFB while the pathological grade was reported as HGIN or carcinoma following ESD, this was identified as pathological upgrading. On the other hand, if one patient showed lower or consistent pathological grade after ESD as compared to that from EFB, they were finally assigned as non-pathological upgrading.14,22
Demographic (e.g., age, sex, and body mass index) and clinical data (e.g.,
Individual medical history and digestive comorbidity was defined positive as written down in medical records or outpatient documents. Symptoms of digestive system were considered as existing if the following words appeared in the records: abdominal pain, nausea, vomiting, dyspepsia, eructation, heartburn, loss of appetite, weight loss.
Ulceration or ulcer scar was considered positive after excluding those caused by the biopsy.23 Remarkable redness was defined as discoloration on the mucosal surface of the lesion compared to the surrounding mucosa, and uneven surface was defined as the presence of irregularly raised or nodular mucosa.24,25 Fig. 2 shows the following endoscopic features in sequence: remarkable redness, ulceration or scar and uneven surface.
All of the following statistical analyses were performed with SPSS 23.0 (IBM Corp., Armonk, NY, USA). Continuous variables were presented as means and standard deviations or medians with interquartile ranges. Categorical variables were demonstrated as numbers and percentages. Differences between groups were assessed using the chi-square test, the Student t-test, or the Mann-Whitney test in the univariate analysis, as appropriate. Variables detected with p<0.05 in the univariate logistic regression were further included in the multivariate logistic regression analysis to determine the variables that were independent influencing factors of pathological upgrading.
A predictive nomogram was constructed, based on the variables selected with the multivariate logistic regression analysis and by using the package of rms in R software (rms in R version 4.0.4, R Foundation for Statistical Computing, Vienna, Austria; https://www.r-project.org/). The model was validated internally in the training set and externally in the validation set. A calibration plot with bootstrapping (1,000 replications) was used to illustrate the association between the actual probability and the predicted probability. In addition, five training models were constructed to compare the robustness of each model. The performance of the prediction model was evaluated using the area under the curve (AUC) in receiver operating characteristic analysis. Decision curve analysis was used to explain the utility of benefits in clinical use.
We retrospectively collected data for 1,053 lesions from 978 patients who underwent ESD for EGN. The comparison of diagnosis between the original EFB and the final ESD is summarized in Table 1. The predominant pathological upgrading involved upgrading from LGIN to HGIN or carcinoma (n=178) and upgrading from HGIN to carcinoma (n=270). After excluding lesions of which pathological diagnosis with EFB were carcinoma, 822 lesions were enrolled into the training and 131 were enrolled into validation sets. Clinical and endoscopic characteristics are summarized in Tables 2 and 3. Lesions (54.0%, 444/822) showed higher-grade disease after ESD in the training set (LGIN from non-neoplasia, HGIN or carcinoma from LGIN or non-neoplasia, or carcinoma from HGIN).
Table 1 . Comparison of Pathological Diagnosis by Initial Biopsy and Final Endoscopic Resection.
Pathologic diagnosis with EFB | Pathologic diagnosis with ESD | ||||
---|---|---|---|---|---|
Non-neoplasia | LGIN | HGIN | Carcinoma | Total No. (%) | |
Non-neoplasia | 15† | 42* | 11* | 10* | 78 (7.4) |
LGIN | 11† | 293† | 108* | 70* | 482 (45.8) |
HGIN | 1† | 29† | 93† | 270* | 393 (37.3) |
Carcinoma | 0† | 5† | 4† | 91† | 100 (9.5) |
Total No. (%) | 27 (2.6) | 369 (35.0) | 216 (20.5) | 441 (41.9) | 1,053 |
EFB, endoscopic forceps biopsy; ESD, endoscopic submucosal dissection; LGIN, low grade intraepithelial neoplasia; HGIN, high grade intraepithelial neoplasia..
*Pathological upgrading; †Non-pathological upgrading..
Table 2 . Clinical Characteristics of the Training and Validation Sets.
Characteristics | Training set (n=822) | Validation set (n=131) |
---|---|---|
Age, yr | 63 (55–69) | 63 (57–70) |
Male sex | 592 (72.0) | 103 (78.6) |
Body mass index, kg/m2 | 23.8±3.3 | 24.1±2.9 |
Negative | 289 (35.1) | 45 (36.4) |
Previous eradicated or infected | 533 (64.9) | 86 (65.6) |
Symptoms of digestive system | 599 (72.9) | 56 (42.7) |
Smoking | 198 (24.1) | 44 (33.6) |
Drinking | 78 (20.6) | 24 (18.3) |
Comorbidity | ||
History of CAG | 466 (56.7) | 61 (46.6) |
History of peptic ulcer | 137 (16.7) | 13 (9.9) |
Medical history | ||
Hypertension | 258 (31.4) | 58 (44.3) |
Diabetes | 74 (9.0) | 14 (10.7) |
Coronary heart disease | 33 (4.0) | 3 (2.3) |
History of cholecystectomy | 58 (7.1) | 19 (14.5) |
History of malignancy | 80 (9.7) | 4 (3.1) |
Family history of malignancy | 133 (16.2) | 12 (9.2) |
Laboratory examination | ||
CEA, ng/mL | 3.4±4.6 | 2.8±4.4 |
CA19-9, U/mL | 12.7±28.2 | 12.8±8.0 |
CA724, U/mL | 3.9±8.3 | 4.0±5.0 |
HDL, mmol/L | 1.2±0.3 | 1.1±0.3 |
LDL, mmol/L | 2.9±0.7 | 2.8±0.6 |
Albumin, g/L | 39.9±3.9 | 39.0±3.0 |
Calcium, mmol/L | 2.3±0.1 | 2.3±0.1 |
Data are presented as median (range), number (%), or mean±SD..
CAG, chronic atrophic gastritis; CEA, carcinoembryonic antigen; CA19-9, carbohydrate antigen 19-9; CA724, carbohydrate antigen 724; HDL, high density lipoprotein; LDL, low density lipoprotein..
Table 3 . Endoscopic Characteristics of the Training and Validation Sets.
Characteristics | Training set (n=822) | Validation set (n=131) |
---|---|---|
Locations of lesions | ||
Upper third | 241 (29.3) | 40 (30.5) |
Middle third | 93 (11.3) | 21 (16.0) |
Lower third | 488 (59.4) | 70 (53.4) |
Macroscopic type of lesions | ||
Elevated | 405 (49.3) | 55 (42.0) |
Flat | 197 (24.0) | 26 (19.8) |
Depressed | 220 (26.8) | 50 (38.2) |
Pathologic diagnosis of EFB | ||
Non-neoplasia | 74 (9.0) | 4 (3.1) |
LGIN | 413 (50.2) | 69 (52.7) |
HGIN | 335 (40.8) | 58 (44.3) |
Pathologic diagnosis of ESD | ||
Non-neoplasia | 27 (3.3) | |
LGIN | 314 (38.2) | 50 (38.2) |
HGIN | 179 (21.8) | 33 (252) |
Carcinoma | 302 (36.7) | 48 (36.8) |
Pathologic upgrading | ||
Yes | 444 (54.0) | 66 (50.4) |
No | 378 (46.0) | 65 (49.6) |
Multiple lesions | 52 (13.8) | 24 (18.3) |
Endoscopic features | ||
Ulceration or scar | 167 (20.3) | 39 (29.8) |
Uneven surface | 364 (44.3) | 64 (48.9) |
Remarkable redness | 531 (64.6) | 88 (67.2) |
Lesion size | ||
≤2 cm | 581 (70.7) | 78 (59.5) |
>2 cm | 241 (29.3) | 53 (40.5) |
Procedure time, min | 60 (45–90) | 60 (45–90) |
Depth of invasion | ||
M | 763 (92.8) | 126 (96.2) |
SM1 | 29 (3.5) | 1 (0.8) |
SM2 | 30 (3.6) | 4 (3.1) |
Data are presented as number (%) or median (range)..
EFB, endoscopic forceps biopsy; LGIN, low grade intraepithelial neoplasm; HGIN, high grade intraepithelial neoplasm; ESD, endoscopic submucosal dissection; M, intraepithelial and/or the deepest infiltration depth is within the mucosal lamina propria; SM1, the infiltration depth >500 μm from the muscularis mucosae; SM2, the infiltration depth >500 μm from the muscularis mucosae..
As shown in Table 4, statistical differences were detected in those with history of smoking, drinking, chronic atrophic gastritis (CAG), and symptoms of digestive system. History of CAG was less commonly detected in the pathological upgrading group than in the non-pathological upgrading group (50.2% vs 60.9%, p<0.001). The rate of
Table 4 . Clinical Characteristics of Lesions with or without Pathological Upgrading.
Characteristics | Non-pathological upgrading group (n=378) | Pathological upgrading group (n=444) | p-value* |
---|---|---|---|
Age, yr | 63 (55–69) | 65 (58–70) | <0.001 |
Male sex | 250 (66.1) | 342 (77.0) | 0.001 |
Body mass index, kg/m2 | 23.6±3.1 | 23.9±3.4 | 0.082 |
0.680 | |||
Negative | 131 (34.7) | 160 (36.0) | |
Previous eradicated or infected | 247 (65.3) | 284 (64.0) | |
Symptoms of digestive system | 260 (68.8) | 339 (76.4) | 0.015 |
Smoking | 105 (27.8) | 159 (35.8) | 0.014 |
Drinking | 78 (20.6) | 120 (27.0) | 0.033 |
Comorbidity | |||
History of CAG | 243 (64.3) | 223 (50.2) | <0.001 |
History of peptic ulcer | 71 (15.4) | 82 (18.5) | 0.224 |
Medical history | |||
Hypertension | 104 (27.5) | 154 (34.7) | 0.073 |
Diabetes | 30 (7.9) | 44 (9.9) | 0.270 |
Coronary heart disease | 11 (2.9) | 22 (5.0) | 0.269 |
History of cholecystectomy | 22 (5.8) | 36 (8.1) | 0.236 |
History of malignancy | 35 (9.3) | 45 (10.1) | 0.531 |
Family history of malignancy | 61 (16.1) | 72 (16.2) | 0.700 |
Laboratory examination | |||
CEA, ng/mL | 3.3±4.5 | 3.5±4.4 | 0.398 |
CA19-9, U/mL | 12.4±30.5 | 13.1±26.2 | 0.734 |
CA724, U/mL | 4.0±9.7 | 3.8±6.9 | 0.682 |
HDL, mmol/L | 1.2±0.3 | 1.1±0.2 | 0.015 |
LDL, mmol/L | 2.9±0.7 | 2.9±0.7 | 0.099 |
Albumin, g/L | 40.2±3.6 | 39.7±4.2 | 0.056 |
Calcium, mmol/L | 2.3±0.1 | 2.2±0.1 | 0.004 |
Data are presented as median (range), number (%), or mean±SD..
CAG, chronic atrophic gastritis; CEA, carcinoembryonic antigen; CA19-9, carbohydrate antigen 19-9; CA724, carbohydrate antigen 724; HDL, high density lipoprotein; LDL, low density lipoprotein..
*p-value was derived from the chi-square test, Student t-test, or Mann-Whitney test..
The contradistinction of endoscopic characteristics is presented in Table 5. Pathological upgrading more frequently occurred in lower third (51.1%), followed by upper third (36.5%) and middle third (12.4%) in the upgrading group. The predominant macroscopic type in both groups was elevated pattern, with a gradual tendency of depressed pattern in the pathological upgrading group. The endoscopic ulceration, uneven surface, remarkable redness was found to account for 26.8%, 56.1%, and 75.4% in cases of pathological upgrading, accordingly.
Table 5 . Endoscopic Characteristics and Posttreatment Evaluation of the Lesions with or without Pathological Upgrading.
Characteristics | Non-pathological upgrading group (n=378) | Pathological upgrading group (n=444) | p-value* |
---|---|---|---|
Locations of lesions | <0.001 | ||
Upper third | 79 (20.9) | 162 (36.5) | |
Middle third | 38 (10.1) | 55 (12.4) | |
Lower third | 261 (69.0) | 227 (51.1) | |
Macroscopic type of lesions | <0.001 | ||
Elevated | 206 (54.5) | 199 (44.8) | |
Flat | 106 (28.0) | 91 (20.5) | |
Depressed | 66 (17.5) | 154 (34.7) | |
Pathologic diagnosis of EFB | <0.001 | ||
Non-neoplasia | 16 (3.5) | 58 (13.1) | |
LGIN | 261 (56.7) | 152 (34.2) | |
HGIN | 101 (22.0) | 234 (52.7) | |
Multiple lesions | 52 (13.8) | 68 (15.3) | 0.528 |
Endoscopic features | |||
Ulceration or scar | 48 (12.7) | 119 (26.8) | <0.001 |
Uneven surface | 115 (30.4) | 249 (56.1) | <0.001 |
Remarkable redness | 197 (52.1) | 334 (75.4) | <0.001 |
Lesion size | <0.001 | ||
≤2 cm | 306 (81.0) | 275 (61.9) | |
>2 cm | 72 (19.0) | 169 (38.1) | |
Procedure time, min | 60 (45–90) | 60 (45–100) | <0.001 |
Depth of invasion | <0.001 | ||
M | 378 (95.0) | 385 (86.7) | |
SM1 | - | 29 (6.5) | |
SM2 | - | 27 (6.8) |
Data are presented as number (%) or median (range)..
EFB, endoscopic forceps biopsy; LGIN, low grade intraepithelial neoplasm; HGIN, high grade intraepithelial neoplasm; M, intraepithelial and/or the deepest infiltration depth is within the mucosal lamina propria; SM1, the infiltration depth >500 μm from the muscularis mucosae; SM2, the infiltration depth >500 μm from the muscularis mucosae..
*p-value was derived from chi-square test or Mann-Whitney test..
Multivariate analysis (logistic regression analysis) was further performed to analyze the variables that were statistically significant in univariate analysis. Ultimately, a total of nine variables which pertained to clinical, laboratorial, endoscopic characteristics separately were enrolled to build the clinical prediction model. As revealed in Table 6, history of CAG (odds ratio [OR], 0.59; 95% confidence interval [CI], 0.42 to 0.84; p=0.003), blood high density lipoprotein (HDL) concentration (OR, 0.46; 95% CI, 0.24 to 0.86; p<0.016) were protective predictors. Age, symptoms of digestive system, macroscopic type of lesions, pathologic diagnosis of EFB, remarkable redness were risk factors, while uneven surface (OR, 2.35; 95% CI, 1.66 to 3.32; p<0.001), lesion size >2 cm (OR, 2.20; 95% CI, 1.50 to 3.22; p<0.001) were independent risk predictors.
Table 6 . Univariate and Multivariate Regression Analyses of the Predictors Associated with Pathological Upgrading According to the Analyzed Variables.
Factor | Univariate logistic analysis | Multivariate logistic analysis | |||
---|---|---|---|---|---|
OR (95% CI) | p-value | OR (95% CI) | p-value | ||
Age | 1.05 (1.03–1.06) | <0.001 | 1.04 (1.03–1.06) | <0.001 | |
Sex (ref. male) | 0.60 (0.44–0.82) | 0.001 | |||
Smoking (ref. no) | 1.38 (1.02–1.87) | 0.036 | |||
Drinking (ref. no) | 1.39 (1.03–1.89) | 0.034 | |||
Comorbidity: CAG (ref. no) | 0.47 (0.35–0.63) | <0.001 | 0.47 (0.33–0.68) | <0.001 | |
Symptoms of digestive system (ref. no) | 1.44 (1.05–1.97) | 0.025 | 1.60 (1.08–2.36) | 0.019 | |
Laboratory examination: HDL | 0.34 (0.20–0.58) | <0.001 | 0.44 (0.24–0.81) | 0.008 | |
Laboratory examination: calcium | 0.09 (0.02–0.37) | 0.001 | |||
Locations of lesions (ref. upper third) | 1.00 | ||||
Middle third | 0.71 (0.43–1.18) | 0.187 | |||
Lower third | 0.40 (0.29–0.56) | 0.001 | |||
Macroscopic type of lesions (ref. elevated) | 1.00 | 1.00 | |||
Flat | 0.84 (0.60–1.19) | 0.331 | 0.74 (0.48–1.13) | 0.164 | |
Depressed | 2.25 (1.68–3.20) | <0.001 | 1.63 (1.13–2.32) | 0.045 | |
Pathological diagnosis of EFB (ref. non-neoplasia) | 1.00 | 1.00 | |||
LGIN | 0.01 (0.01–0.08) | <0.001 | 0.01 (0.01–0.06) | <0.001 | |
HGIN | 0.04 (0.01–0.30) | 0.002 | 0.02 (0.01–0.13) | <0.001 | |
Endoscopic features: ulceration or scar (ref. no) | 2.63 (1.79–3.84) | <0.001 | |||
Endoscopic features: uneven surface (ref. no) | 2.69 (2.01–3.60) | <0.001 | 2.24 (1.58–3.19) | <0.001 | |
Endoscopic features: remarkable redness (ref. no) | 2.95 (2.19–3.99) | <0.001 | 2.26 (1.57–3.27) | <0.001 | |
Lesion size (ref. ≤2 cm) | 2.47 (1.78–3.41) | <0.001 | 2.20 (1.49–3.24) | <0.001 |
OR, odds ratio; CI, confidence interval; ref., reference group; CAG, chronic atrophic gastritis; HDL, high density lipoprotein; EFB, endoscopic forceps biopsy; LGIN, low grade intraepithelial neoplasm; HGIN, high grade intraepithelial neoplasm..
We developed a nomogram to predict the risk of pathological upgrading in patients with EGN (Fig. 3). For instance, a 65-year-old patient, with a previous history of CAG, without symptoms of digestive system, and blood HDL concentration of 1.0 mmol/L, was found to have an elevated lesion with remarkable redness and uneven surface, and less than 2 cm in diameter at endoscopy. If the EFB pathology was HGIN, the incidence of pathological upgrading would be 65.1% (95% CI, 0.522 to 0.761). The online version of our nomogram software is accessible at https://zyh-njmu.shinyapps.io/Dynamic_nomogram/.
We performed validation by using the bootstrap method with 1,000 repetitions to display the calibration of the nomogram (Fig. 4A and B). Accuracy of the nomogram was examined with AUC. The C-statistics for our nomogram model was 0.804 (95% CI, 0.774 to 0.834) and 0.748 (95% CI, 0.664 to 0.832) in the training and validation set, respectively (Fig. 4C and D).
To assess the predictive ability of our nomogram, several training schemes (namely training models 1 to 5) were constructed (Fig. 5). Afterwards, we performed internal validation and AUC computation for training models 1 to 5 and EFB, and the calibration plot and C-statistics for each model were shown in Figs 6 and 7, respectively.
EFB and ESD are used for diagnosis of EGN. However, EFB is not able to show histology of the entire lesion, which may lead to discrepancy between histological diagnosis of EFB and resected specimen. Therefore, it is urgently needed to establish a clinical predictive tool to identify the risk of pathological upgrading. Nomogram is a visualization prediction tool that can incorporate different variables affecting outcomes.26 We developed a nomogram for pathological upgrading in patients received EFB before ESD which showed a steady accuracy in evaluating the risk of pathological upgrading.
In the current study, rate of pathological upgrading was 44.9%. It was similar to the results of previous studies.15,27-31 Sixty-nine point nine percent of HGIN in EFB were diagnosed as EGC in our study, which was comparable to reported results.30,32,33 It has been found that pathological upgrading was positively correlated with longer operation time and deeper submucosal invasion. In addition, the depth of invasion was verified in predicting the risk of nodal metastasis in EGCs.17,34,35
In agreement with reported data, our study demonstrated that the depressed pattern, nodular surface, surface redness, lesion size >2 cm, and location in the upper third of the stomach were risk factors in pathological upgrading.23,36,37 EGCs located in the lower third of the stomach, especially in the antrum, might be detected easily, but for those in the upper third, detection may be difficult due to denser mucosal folds, thinner stomach, and limited endoscopic visual fields.36
Ideally, one would incline to have a validated noninvasive biomarker to predict cancer risk. However, no positive values of traditional tumor markers in predicting pathological upgrading have been observed in our study. We found that associations between pathological upgrading and metabolic features such as body mass index, hypertension, and dyslipidemia were not significant except for HDL. A 2015 meta-analysis, which summarized studies associating apolipoprotein E gene and cancer, demonstrated that reduced circulating HDL might be a potentially causal risk factor for the development of overall cancer in Asians.38 As studies from several other centers have shown, the decreased HDL level was related to increased risk of gastric cancer and advanced disease stage.39-42 One possible explanation is that HDL may promote cancer development by generating reactive oxygen species, increasing hormone production and availability (e.g., insulin-like growth factor, insulin, and adipokines), and forming an energy rich environment. This imbalance of hormones, the redox system, and energy availability plays a role in epithelium transition.43,44
It was interesting to note that our study showed significant statistical differences in both univariate and multivariate analyses for history of CAG and symptoms of digestive system between groups. The intestinal type of gastric cancer typically follows the Correa precancerous cascade of changes initiated by a non-self-limiting inflammation, of which CAG represents the relatively early stage.45 With the upgrading of gastric carcinogenesis, non-specific conditions such as gastritis and ulceration may appear.46,47 Poor prognosis from gastric cancer is mainly due to late presentation. Alarm symptoms of EGN include abdominal or epigastric pain or discomfort, nausea and vomiting, etc.48,49 EGN is usually asymptomatic. Therefore, referral guidelines have been developed to encourage early detection.50,51 Normally, endoscopists would be more careful with symptomatic patients during endoscopy.52,53 We assumed that patients who had CAG or suffered from digestive symptoms may be concerned about their digestive system health and would go through endoscopic checks.52,54 Therefore, the history of CAG and symptoms of digestive system seem to be protective factors.
Here, we reported predictors for pathological upgrading, also we established a nomogram in a visualized pattern. Another strength of our study was that our predictors included clinical, laboratorial and endoscopic features, and its performance was verified in the comparison with other training models (Figs 7 and 8). The nomogram yielded the highest AUC demonstrating the best discrimination in estimating the risk of pathological upgrading in patients with EGN. The final decision curve analysis showed that as long as setting the threshold probability of patients or clinicians beyond 10% (i.e., if no intervention was provided for patient with pathological upgrading was considered probably inappropriate), screening strategies based on our nomogram’s pathological upgrading risk estimates resulted in superior net benefit than screen-none or screen-all strategies (Fig. 7). When we set the threshold up to 40% or higher, the predictive effect of nomogram was still better than that of any training model. For practical application, we recommended explaining the upgrading risk predicted by nomogram to patients based on EFB findings. For EFB diagnosis of HGIN, intervention of ESD is necessary regardless of upgrading.6,55 For LGIN or non-neoplasia, a predicting value >50% indicates a greater risk of misdiagnosis, and secondary examination or immediate endoscopic intervention would be recommended.
This study has some limitations. First of all, this was a retrospective cohort study, in which data were mainly obtained by reviewing medical records or endoscopic results from a single center. Thus, it may restrict application to a wider population. Second, the operational differences between primary examiner at the local clinic and the procedural endoscopic resection operator, as well as the impact of random/guided biopsies, number of biopsies, could not be evaluated.4,56 Several studies have reported that white-light endoscopy combined with narrow-band imaging is superior to white-light endoscopy alone in identifying patients with gastric intestinal metaplasia and dysplasia.57,58 In fact, there were still many medical institutions in which virtual chromoendoscopy and high-definition processors and screens were not routinely available. Here, we enrolled no endoscopic features other than from white-light endoscopy.
In conclusion, pathological upgrading in EGN is common with specific clinical, laboratorial and endoscopic characteristics. Based on the above identified risk factors, a nomogram was developed to objectively and accurately predict individualized pathological upgrading risk of patients with EGN before ESD.
This study was supported by the Jiangsu Municipal Science and Technology Bureau (No. BE2020784), the National Natural Science Foundation of China (No. 82000529), and the Natural Science Foundation of Jiangsu Province (No. SBK 2020042595). The funding body had no role in the study design, data collection, analysis, and interpretation of data.
The authors appreciate the study participants and the inestimable support from the research center. The authors are grateful for support of the medical and research staff of this study.
No potential conflict of interest relevant to this article was reported.
Study concept and design: Y.H.Z., Y.N.D., G.X.Z. Data acquisition: Y.H.Z., Y.L., J.S. Pathological support: H.J.H., K.D.L. Data analysis and interpretation: Y.H.Z. Drafting of the manuscript: Y.H.Z., J.S. Critical revision of the manuscript for important intellectual content: Y.Z. Statistical analysis: Y.H.Z., Y.Z. Obtained funding: Y.N.D., G.X.Z. Study supervision: Y.N.D., G.X.Z. Approval of final manuscript: all authors.
Table 1 Comparison of Pathological Diagnosis by Initial Biopsy and Final Endoscopic Resection
Pathologic diagnosis with EFB | Pathologic diagnosis with ESD | ||||
---|---|---|---|---|---|
Non-neoplasia | LGIN | HGIN | Carcinoma | Total No. (%) | |
Non-neoplasia | 15† | 42* | 11* | 10* | 78 (7.4) |
LGIN | 11† | 293† | 108* | 70* | 482 (45.8) |
HGIN | 1† | 29† | 93† | 270* | 393 (37.3) |
Carcinoma | 0† | 5† | 4† | 91† | 100 (9.5) |
Total No. (%) | 27 (2.6) | 369 (35.0) | 216 (20.5) | 441 (41.9) | 1,053 |
EFB, endoscopic forceps biopsy; ESD, endoscopic submucosal dissection; LGIN, low grade intraepithelial neoplasia; HGIN, high grade intraepithelial neoplasia.
*Pathological upgrading; †Non-pathological upgrading.
Table 2 Clinical Characteristics of the Training and Validation Sets
Characteristics | Training set (n=822) | Validation set (n=131) |
---|---|---|
Age, yr | 63 (55–69) | 63 (57–70) |
Male sex | 592 (72.0) | 103 (78.6) |
Body mass index, kg/m2 | 23.8±3.3 | 24.1±2.9 |
Negative | 289 (35.1) | 45 (36.4) |
Previous eradicated or infected | 533 (64.9) | 86 (65.6) |
Symptoms of digestive system | 599 (72.9) | 56 (42.7) |
Smoking | 198 (24.1) | 44 (33.6) |
Drinking | 78 (20.6) | 24 (18.3) |
Comorbidity | ||
History of CAG | 466 (56.7) | 61 (46.6) |
History of peptic ulcer | 137 (16.7) | 13 (9.9) |
Medical history | ||
Hypertension | 258 (31.4) | 58 (44.3) |
Diabetes | 74 (9.0) | 14 (10.7) |
Coronary heart disease | 33 (4.0) | 3 (2.3) |
History of cholecystectomy | 58 (7.1) | 19 (14.5) |
History of malignancy | 80 (9.7) | 4 (3.1) |
Family history of malignancy | 133 (16.2) | 12 (9.2) |
Laboratory examination | ||
CEA, ng/mL | 3.4±4.6 | 2.8±4.4 |
CA19-9, U/mL | 12.7±28.2 | 12.8±8.0 |
CA724, U/mL | 3.9±8.3 | 4.0±5.0 |
HDL, mmol/L | 1.2±0.3 | 1.1±0.3 |
LDL, mmol/L | 2.9±0.7 | 2.8±0.6 |
Albumin, g/L | 39.9±3.9 | 39.0±3.0 |
Calcium, mmol/L | 2.3±0.1 | 2.3±0.1 |
Data are presented as median (range), number (%), or mean±SD.
CAG, chronic atrophic gastritis; CEA, carcinoembryonic antigen; CA19-9, carbohydrate antigen 19-9; CA724, carbohydrate antigen 724; HDL, high density lipoprotein; LDL, low density lipoprotein.
Table 3 Endoscopic Characteristics of the Training and Validation Sets
Characteristics | Training set (n=822) | Validation set (n=131) |
---|---|---|
Locations of lesions | ||
Upper third | 241 (29.3) | 40 (30.5) |
Middle third | 93 (11.3) | 21 (16.0) |
Lower third | 488 (59.4) | 70 (53.4) |
Macroscopic type of lesions | ||
Elevated | 405 (49.3) | 55 (42.0) |
Flat | 197 (24.0) | 26 (19.8) |
Depressed | 220 (26.8) | 50 (38.2) |
Pathologic diagnosis of EFB | ||
Non-neoplasia | 74 (9.0) | 4 (3.1) |
LGIN | 413 (50.2) | 69 (52.7) |
HGIN | 335 (40.8) | 58 (44.3) |
Pathologic diagnosis of ESD | ||
Non-neoplasia | 27 (3.3) | |
LGIN | 314 (38.2) | 50 (38.2) |
HGIN | 179 (21.8) | 33 (252) |
Carcinoma | 302 (36.7) | 48 (36.8) |
Pathologic upgrading | ||
Yes | 444 (54.0) | 66 (50.4) |
No | 378 (46.0) | 65 (49.6) |
Multiple lesions | 52 (13.8) | 24 (18.3) |
Endoscopic features | ||
Ulceration or scar | 167 (20.3) | 39 (29.8) |
Uneven surface | 364 (44.3) | 64 (48.9) |
Remarkable redness | 531 (64.6) | 88 (67.2) |
Lesion size | ||
≤2 cm | 581 (70.7) | 78 (59.5) |
>2 cm | 241 (29.3) | 53 (40.5) |
Procedure time, min | 60 (45–90) | 60 (45–90) |
Depth of invasion | ||
M | 763 (92.8) | 126 (96.2) |
SM1 | 29 (3.5) | 1 (0.8) |
SM2 | 30 (3.6) | 4 (3.1) |
Data are presented as number (%) or median (range).
EFB, endoscopic forceps biopsy; LGIN, low grade intraepithelial neoplasm; HGIN, high grade intraepithelial neoplasm; ESD, endoscopic submucosal dissection; M, intraepithelial and/or the deepest infiltration depth is within the mucosal lamina propria; SM1, the infiltration depth >500 μm from the muscularis mucosae; SM2, the infiltration depth >500 μm from the muscularis mucosae.
Table 4 Clinical Characteristics of Lesions with or without Pathological Upgrading
Characteristics | Non-pathological upgrading group (n=378) | Pathological upgrading group (n=444) | p-value* |
---|---|---|---|
Age, yr | 63 (55–69) | 65 (58–70) | <0.001 |
Male sex | 250 (66.1) | 342 (77.0) | 0.001 |
Body mass index, kg/m2 | 23.6±3.1 | 23.9±3.4 | 0.082 |
0.680 | |||
Negative | 131 (34.7) | 160 (36.0) | |
Previous eradicated or infected | 247 (65.3) | 284 (64.0) | |
Symptoms of digestive system | 260 (68.8) | 339 (76.4) | 0.015 |
Smoking | 105 (27.8) | 159 (35.8) | 0.014 |
Drinking | 78 (20.6) | 120 (27.0) | 0.033 |
Comorbidity | |||
History of CAG | 243 (64.3) | 223 (50.2) | <0.001 |
History of peptic ulcer | 71 (15.4) | 82 (18.5) | 0.224 |
Medical history | |||
Hypertension | 104 (27.5) | 154 (34.7) | 0.073 |
Diabetes | 30 (7.9) | 44 (9.9) | 0.270 |
Coronary heart disease | 11 (2.9) | 22 (5.0) | 0.269 |
History of cholecystectomy | 22 (5.8) | 36 (8.1) | 0.236 |
History of malignancy | 35 (9.3) | 45 (10.1) | 0.531 |
Family history of malignancy | 61 (16.1) | 72 (16.2) | 0.700 |
Laboratory examination | |||
CEA, ng/mL | 3.3±4.5 | 3.5±4.4 | 0.398 |
CA19-9, U/mL | 12.4±30.5 | 13.1±26.2 | 0.734 |
CA724, U/mL | 4.0±9.7 | 3.8±6.9 | 0.682 |
HDL, mmol/L | 1.2±0.3 | 1.1±0.2 | 0.015 |
LDL, mmol/L | 2.9±0.7 | 2.9±0.7 | 0.099 |
Albumin, g/L | 40.2±3.6 | 39.7±4.2 | 0.056 |
Calcium, mmol/L | 2.3±0.1 | 2.2±0.1 | 0.004 |
Data are presented as median (range), number (%), or mean±SD.
CAG, chronic atrophic gastritis; CEA, carcinoembryonic antigen; CA19-9, carbohydrate antigen 19-9; CA724, carbohydrate antigen 724; HDL, high density lipoprotein; LDL, low density lipoprotein.
*p-value was derived from the chi-square test, Student t-test, or Mann-Whitney test.
Table 5 Endoscopic Characteristics and Posttreatment Evaluation of the Lesions with or without Pathological Upgrading
Characteristics | Non-pathological upgrading group (n=378) | Pathological upgrading group (n=444) | p-value* |
---|---|---|---|
Locations of lesions | <0.001 | ||
Upper third | 79 (20.9) | 162 (36.5) | |
Middle third | 38 (10.1) | 55 (12.4) | |
Lower third | 261 (69.0) | 227 (51.1) | |
Macroscopic type of lesions | <0.001 | ||
Elevated | 206 (54.5) | 199 (44.8) | |
Flat | 106 (28.0) | 91 (20.5) | |
Depressed | 66 (17.5) | 154 (34.7) | |
Pathologic diagnosis of EFB | <0.001 | ||
Non-neoplasia | 16 (3.5) | 58 (13.1) | |
LGIN | 261 (56.7) | 152 (34.2) | |
HGIN | 101 (22.0) | 234 (52.7) | |
Multiple lesions | 52 (13.8) | 68 (15.3) | 0.528 |
Endoscopic features | |||
Ulceration or scar | 48 (12.7) | 119 (26.8) | <0.001 |
Uneven surface | 115 (30.4) | 249 (56.1) | <0.001 |
Remarkable redness | 197 (52.1) | 334 (75.4) | <0.001 |
Lesion size | <0.001 | ||
≤2 cm | 306 (81.0) | 275 (61.9) | |
>2 cm | 72 (19.0) | 169 (38.1) | |
Procedure time, min | 60 (45–90) | 60 (45–100) | <0.001 |
Depth of invasion | <0.001 | ||
M | 378 (95.0) | 385 (86.7) | |
SM1 | - | 29 (6.5) | |
SM2 | - | 27 (6.8) |
Data are presented as number (%) or median (range).
EFB, endoscopic forceps biopsy; LGIN, low grade intraepithelial neoplasm; HGIN, high grade intraepithelial neoplasm; M, intraepithelial and/or the deepest infiltration depth is within the mucosal lamina propria; SM1, the infiltration depth >500 μm from the muscularis mucosae; SM2, the infiltration depth >500 μm from the muscularis mucosae.
*p-value was derived from chi-square test or Mann-Whitney test.
Table 6 Univariate and Multivariate Regression Analyses of the Predictors Associated with Pathological Upgrading According to the Analyzed Variables
Factor | Univariate logistic analysis | Multivariate logistic analysis | |||
---|---|---|---|---|---|
OR (95% CI) | p-value | OR (95% CI) | p-value | ||
Age | 1.05 (1.03–1.06) | <0.001 | 1.04 (1.03–1.06) | <0.001 | |
Sex (ref. male) | 0.60 (0.44–0.82) | 0.001 | |||
Smoking (ref. no) | 1.38 (1.02–1.87) | 0.036 | |||
Drinking (ref. no) | 1.39 (1.03–1.89) | 0.034 | |||
Comorbidity: CAG (ref. no) | 0.47 (0.35–0.63) | <0.001 | 0.47 (0.33–0.68) | <0.001 | |
Symptoms of digestive system (ref. no) | 1.44 (1.05–1.97) | 0.025 | 1.60 (1.08–2.36) | 0.019 | |
Laboratory examination: HDL | 0.34 (0.20–0.58) | <0.001 | 0.44 (0.24–0.81) | 0.008 | |
Laboratory examination: calcium | 0.09 (0.02–0.37) | 0.001 | |||
Locations of lesions (ref. upper third) | 1.00 | ||||
Middle third | 0.71 (0.43–1.18) | 0.187 | |||
Lower third | 0.40 (0.29–0.56) | 0.001 | |||
Macroscopic type of lesions (ref. elevated) | 1.00 | 1.00 | |||
Flat | 0.84 (0.60–1.19) | 0.331 | 0.74 (0.48–1.13) | 0.164 | |
Depressed | 2.25 (1.68–3.20) | <0.001 | 1.63 (1.13–2.32) | 0.045 | |
Pathological diagnosis of EFB (ref. non-neoplasia) | 1.00 | 1.00 | |||
LGIN | 0.01 (0.01–0.08) | <0.001 | 0.01 (0.01–0.06) | <0.001 | |
HGIN | 0.04 (0.01–0.30) | 0.002 | 0.02 (0.01–0.13) | <0.001 | |
Endoscopic features: ulceration or scar (ref. no) | 2.63 (1.79–3.84) | <0.001 | |||
Endoscopic features: uneven surface (ref. no) | 2.69 (2.01–3.60) | <0.001 | 2.24 (1.58–3.19) | <0.001 | |
Endoscopic features: remarkable redness (ref. no) | 2.95 (2.19–3.99) | <0.001 | 2.26 (1.57–3.27) | <0.001 | |
Lesion size (ref. ≤2 cm) | 2.47 (1.78–3.41) | <0.001 | 2.20 (1.49–3.24) | <0.001 |
OR, odds ratio; CI, confidence interval; ref., reference group; CAG, chronic atrophic gastritis; HDL, high density lipoprotein; EFB, endoscopic forceps biopsy; LGIN, low grade intraepithelial neoplasm; HGIN, high grade intraepithelial neoplasm.