Society: ASGE

[Introduction] Early detection of gastric cancer offers favorable treatment outcomes. Second-generation narrow-band imaging (NBI) is expected to improve the detection of early gastric cancer (EGC); however, the detection rate is not superior to that of white light imaging (WLI). The latest endoscopic system EVIS X1 (Olympus, Tokyo, Japan) includes a higher-definition WLI, third generation (3G)-NBI, and texture and color enhancement imaging (TXI). We conducted a randomized phase II trial to compare the ability of 3G-NBI, TXI, and WLI to detect gastric neoplasms (GN) (jRCT1032210213).
[Methods] The eligibility criteria included patients aged 20–85 years with either of the following: 1) scheduled surveillance endoscopy after endoscopic resection for GN, or endoscopic resection, chemotherapy, or radiotherapy for esophageal cancer, or 2) scheduled preoperative endoscopy for known GN or esophageal cancer. Written informed consent was obtained from all participants. The EVIS X1 system and a high-definition gastroscope with an optical zoom (GIF-XZ1200) were used. Patients were randomly assigned in a 1:1:1 ratio to the 3G-NBI (primary 3G-NBI and secondary WLI), TXI (primary TXI and secondary WLI), and WLI (primary and secondary WLI) arms, with the WLI arm set as the reference arm. Non-magnifying primary observation was performed to detect the GN lesions. After completing the primary observation of the entire stomach, a secondary WLI was immediately performed by the same endoscopist. All suspected GN lesions were biopsied at the end of the examination. Pathological diagnoses were made based on biopsied tissue or specimens obtained from endoscopic or surgical resection by expert pathologists at each institution. The primary endpoint was GN detection rate in the primary observation, including cancer and adenoma. The other endpoints were the miss rate for GN, EGC detection rate, and positive predictive value (PPV) for the diagnosis of GN in the primary observation. We assumed that the primary endpoint would be 3.0% for one image-enhanced endoscopy (IEE) and > 4.3% for the other. The sample size was set to 300 per arm to ensure that 80% or more of the participants correctly selected the most promising IEE.
[Results] 901 patients were enrolled from six institutions and assigned to the 3G-NBI, TXI, and WLI arms (300/300/301). Of these, 222 (24.6%) underwent preoperative examination. The GN detection rate in the 3G-NBI, TXI, and WLI arms were 7.3%, 5.0%, and 5.6%, respectively, with the 3G-NBI showing the highest detection rate. In addition, the miss rates for GN were 1.0%, 0.7%, and 1.0%, the EGC detection rates were 5.7%, 4.0%, and 5.6%, and the PPVs for the diagnosis of GN were 36.5%, 21.3%, and 36.8% in the 3G-NBI, TXI, and WLI arms, respectively.
[Conclusion] 3G-NBI is the most promising modality for the detection of GN when compared with TXI and WLI.

Background & Aims:
Gastrointestinal (GI) bleeding is a significant complication of left ventricular assist device (LVAD) placement and can be life-threatening. If LVAD patients at higher risk of bleeding could be identified, increased efforts to prevent and manage GI bleeding could be instituted. In this study we aimed to derive a GI-bleed risk prediction model in patients with LVAD, by means of supervised machine learning, using a large retrospective data from the University of Utah heart-transplantation database.

Methods:
We utilized prospectively maintained database comprising 491 LVAD patients between 2004-2022. Institutional review board approval was obtained prior to data collection. The penalized Least Absolute Shrinkage and Selection Operator (LASSO) regression was used to select the best predictors of GI bleeding in LVAD patients, using 10-fold cross validation to develop a practical estimation of the predictive performance, by means of the area under the curve (AUC). Bootstrapped Bias corrected 95% confidence intervals (CI) for the AUC were generated. Thirty potential clinical predictors were introduced in the model. The model disagreement between the predicted and the observed outcomes were evaluated using the Brier score and calibration belt.

Results:
The GI bleeding risk in patients with LVAD was 26.06% (n=128) in the current sample. Three out of the 30 predictors were retained by the prediction model. These were: duration of LVAD implantation (<12 months, 13-24 months, >24 months), anticoagulant use, and LVAD as destination therapy. The model's ability to distinguish LVAD patients at high risk of GI-bleed was outstanding based on the AUC=0.9 (95%CI 0.82-0.92) (Figure 1A). There was no evidence of miscalibration for this model (test statistic=4.61; p=0.10), demonstrating good performance (Brier score=0.05).
A score was assigned to each predictor with a maximum score of 6. Destination therapy as an indication for LVAD transplant =1, LVAD implantation duration =1 if the duration was <12 months, 2 if the duration was 13-24 months, and 3 if the duration was >24 months. A score of 2 was assigned to anticoagulant use. The purpose of score derivation from LASSO regression predictors was to facilitate an understanding, applicability, and practicality for the proposed risk prediction model. The AUC of the Score was 0.90 (95% CI 0.87-0.93) and showed no statistical difference from the AUC of the LASSO model (P=0.49) (Figure 1B). The Youden index determined that cut-off score of 3 had a sensitivity of 69.5% and specificity of 99.1% (Table 1).

Conclusions:
The proposed model presents prediction variables in identifying patients at risk of GI bleeding post LVAD placement. Duration of LVAD implantation, anticoagulant use, and LVAD as destination therapy were the top variables retained. Additional studies are warranted to validate our findings externally.

<br /> <b>Figure 1: (A) ROC curve with mean cross validated area under the curve (CvAUC) after 10 fold cross validation. AUC: 0.92 (Bootstrap Bias Corrected 95%CI: 0.82, 0.92) and (B) comparison with AUC 0.90 (Bootstrap Bias Corrected 95% CI 0.87-0.93) of proposed score generated from lasso-generated predictors</b>

Figure 1: (A) ROC curve with mean cross validated area under the curve (CvAUC) after 10 fold cross validation. AUC: 0.92 (Bootstrap Bias Corrected 95%CI: 0.82, 0.92) and (B) comparison with AUC 0.90 (Bootstrap Bias Corrected 95% CI 0.87-0.93) of proposed score generated from lasso-generated predictors

<br /> <b>Table 1: GI bleeding risk post LVAD placement and point specific sensitivity/specificity of proposed score</b>

Table 1: GI bleeding risk post LVAD placement and point specific sensitivity/specificity of proposed score

Background: Gastric per-oral endoscopic myotomy (GPOEM) is a recently developed minimally invasive procedure for the treatment of refractory gastroparesis. Gastroparesis post-lung transplant is a frequent debilitating complication affecting 20% to 90% of lung transplant recipients, as it can increase the risk for gastro-esophageal reflux disease and subsequent graft dysfunction. We performed a systematic review and meta-analysis to evaluate the efficacy and safety of GPOEM in lung transplant patients with refractory gastroparesis.
Methods: The present systematic review and meta-analysis was performed according to the Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) guidelines. PubMed, Cochrane and Embase databases were searched for relevant studies published from inception through November 2022. We selected studies reporting clinical efficacy rates and gastroparesis cardinal symptom index (GCSI) in the pre- and post-GPOEM period in lung transplant patients with refractory gastroparesis. Pooled effect estimates were calculated for clinical efficacy, defined as no residual gastroparesis symptoms requiring prokinetic agents or reintervention. Hedges’ g test was used to calculate the standard mean difference of GCSI in the pre- and post-GPOEM period. Random effects model was used and analysis was performed with STATA 17.
Results: Four observational studies (one conference abstract) with 104 patients were included in the meta-analysis. Prior treatments for gastroparesis included prokinetic agents and botulinum toxin in 78% (78/104) and 66.7% (66/99) respectively (Table 1). Pooled estimates for clinical efficacy of GPOEM was 83% (95% CI, 76%-90%) (Figure 1). The pooled mean reduction in GCSI following the procedure was -2.01 (-2.35, -1.65, p=0.014) (Figure 2). Three studies reported statistically significant improvement of gastro-esophageal retention or emptying in the post-GPOEM period; however analysis was not feasible due to a non-uniform reporting of results. 30-day postoperative complications included minor or major bleeding (11.6%), severe reflux (1.2%) and pyloric stenosis (1.2%) requiring re-intervention. 90-day mortality was 2.6% with one patient dying from severe allograft rejection.
Conclusion: Our study showed that GPOEM is an effective and safe strategy for lung transplant patients with refractory gastroparesis and should be considered as a therapeutic strategy in this population. Larger multicenter trials are needed in the future to further evaluate the effect of GPOEM on allograft function and rates of rejection.

Background:
Although gastric cancer (GC) incidence is higher in high human development index (HDI) countries, mortality is lower compared to those with low and medium HDI. It is the fourth leading cause of death worldwide, but the first in some countries in Latin America, for example. It is reported that up to 25.8% GC cases are missed during endoscopy exams. Recently, Artificial Intelligence (AI) diagnostic tools have been reported to have high performance for detecting GC and to help overcome this issue. However, AI is trained with images obtained from high end equipment, which makes the generalizability of the diagnostic performance on outdated equipment an important question to be addressed. Therefore, the aim of this study is to evaluate the performance of a convolutional neural network-based AI tool in images retrieved from older endoscopes and processors in Brazil.
Methods:
The test dataset consisted of 420 images (standard white light) from 111 patients that were retrospectively collected from two hospitals in Brazil. Images were captured using the following endoscopic video systems: CV-150, CV-160, CV-170 and CV-180 Olympus Medical Systems and 4400, 4450 and 600 Fujifilm. All lesions were histology proven. The dataset was processed by a previously described convolutional neural network-based model (1), trained with high quality Japanese endoscopic images to differentiate early gastric cancer (EGC) from benign lesions or normal mucosa. For patient analysis, an answer was defined as correct when CNN detected one EGC image in multiple images of the same lesion.
Results:
The CNN detected 119 of 150 images of EGC with an overall sensitivity of 79.3%, and 50 non-neoplastic lesions resulting in a positive predictive value of 70.4%. Specificity and accuracy were 81.4% and 80.7%, respectively. Neoplastic lesions varied from 5 to 35 millimeters. Per patient analysis sensitivity was 85.7%, resulting from a correct diagnosis of 14 of 16 cases of EGC (Figure 1).
Discussion and Conclusion:
We reported the first external validation of AI in the Endoscopy field focused on using images retrieved from outdated equipment. Although the retrospective design has intrinsic limitations such as selection bias, many of the selected images would have been excluded in other studies due to the lower quality resolution and lightning. Despite that, CNN-based AI tool was able to detect early gastric cancer with acceptable performance, which suggests that this diagnostic tool can potentially be used in practices with lower resources.
(1) Ikenoyama Y, et al. Dig Endosc. 2021;33(1):141-150.

Figure 1. Detail of the two false negative cases with corresponding image examples.

Figure 1. Detail of the two false negative cases with corresponding image examples.

Figure 2. Exemplary cases of lesions from the validation set.<br /> (A)True positive: 14 mm adenocarcinoma diffuse type, Paris Classification 0-IIc<br /> (B)True positive: 30 mm adenocarcinoma intestinal type, Paris Classification 0-IIa+IIb<br /> (C)False negative: scar<br /> (D)False negative: metaplasia

Figure 2. Exemplary cases of lesions from the validation set.
(A)True positive: 14 mm adenocarcinoma diffuse type, Paris Classification 0-IIc
(B)True positive: 30 mm adenocarcinoma intestinal type, Paris Classification 0-IIa+IIb
(C)False negative: scar
(D)False negative: metaplasia

Kenichiro Suzuki

Appearances