S1196 Comparing Virtual Chromoendoscopy to Dye-Spraying Chromoendoscopy and White Light Endoscopy in Screening Patients With Inflammatory Bowel Disease: A Systematic Review and Meta-Analysis
Bibliographic record
Abstract
Introduction: Patients with inflammatory bowel disease have high risk for colon cancer and surveillance colonoscopy is crucial for early detection of dysplasia and neoplasia for this patient population. Virtual chromoendoscopy (VCE) techniques have shown promise in enhancing lesion detection rates. However, the comparative effectiveness of these methods to the traditional dye-spraying chromoendoscopy (CE) and white light endoscopy (WLE) remains unclear. Methods: A comprehensive search was performed in electronic databases, including PubMed, EMBASE, and Web of Science, from inception until November 2022. Studies reporting number of detected lesions or number of patients with colonic lesions were included. Studies that reported tandem colonoscopies were excluded. For studies with crossover design, we included data from the first procedure. Random-effect models were used to estimate pooled risk ratios (RR) and 95% confidence intervals (95% CI). Subgroup analyses were conducted based on the study type (randomized (RCT) vs observational) and based on the VCE technique. R version 4.0.5 (R foundation for statistical computing, Vienna, Austria) was used to conduct the statistical analysis. Results: A total of 4,888 studies were assessed and 16 of them were eligible for our analysis. Of the included studies, two were observational, three had a crossover design, and five were published as abstracts (Table 1). Per patient analysis, VCE improved detecting patients with colonic lesions compared to CE (RR 0.73; 95% CI, 0.59–0.9) and WLE (RR 0.69; 95% CI, 0.54–0.87). However, these findings became statistically insignificant if only RCT were used in the subgroup analysis based on the study type (Figure 1A,C). Per the number of lesion analyses, VCE was not statistically different compared to CE (RR 0.73; 95% CI, 0.50–1.06) or WLE (RR 1.04; 95% CI, 0.65–1.67) (Figure 1B,D). In the subgroup analysis based on the VCE technique, there were no statistical differences between autofluorescence imaging, Fuji intelligent color enhancement, I-scan, or narrow band imaging compared to CE or WLE. Conclusion: The efficacy of virtual chromoendoscopy in detecting colonic lesions for patients with inflammatory bowel disease seems to be higher when compared to dye-spraying chromoendoscopy and white light colonoscopy. However, it is important to note that these findings are primarily based on observational data and do not hold up when considering only randomized trials.Figure 1.: Forest plots comparing virtual chromoendoscopy to dye-spraying chromoendoscopy and white light endoscopy. Table 1. - Characteristics of the included studies Study Year Study type Study design Publication Study setting Sample size Virtual CE Comparator Van Den 2010 RCT Crossover Full Netherlands 48 NBI WLE Feitosa 2011 RCT Parallel Abstract Brazil 29 NBI CE Pellise 2011 RCT Crossover Full Spain 60 NBI CE Ignjatovic 2012 RCT Parallel Full UK 112 NBI WLE Cassinotti 2015 RCT Parallel Abstract Italy 91 FICE WLE Bisschops 2016 RCT Parallel Full Canada 131 NBI CE Gasia 2016 Observational Parallel Full Canada 454 I-scan CE & WLE Watanabe 2016 RCT Parallel Abstract Japan 263 NBI CE Iacucci 2017 RCT Parallel Full Canada 270 I-scan CE & WLE Lopez-Serrano 2017 RCT Parallel Abstract Spain 66 I-scan CE Gulati 2018 RCT Crossover Full UK 48 FICE CE Vleugels 2018 RCT Parallel Full Netherlands & UK 210 AFI CE Kandiah 2021 RCT Parallel Full UK 188 I-scan WLE González-Bernardo 2021 RCT Parallel Full Spain 129 I-scan CE Lopez-Serranoa 2021 Observational Parallel Full Spain 191 I-scan CE Sinonquel 2022 RCT Parallel Abstract Europe (4 countries) 136 I-scan CE
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.011 | 0.026 |
| Meta-epidemiology (narrow) | 0.002 | 0.001 |
| Meta-epidemiology (broad) | 0.017 | 0.027 |
| Bibliometrics | 0.007 | 0.008 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.003 | 0.002 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.002 | 0.001 |
| Insufficient payload (model declined to judge) | 0.006 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".