One-fourth of colorectal neoplasia are missed at screening colonoscopy, representing the main cause of interval colorectal cancer. Deep learning systems with real-time computer-aided polyp detection ...(CADe) showed high accuracy in artificial settings, and preliminary randomized controlled trials (RCTs) reported favorable outcomes in the clinical setting. The aim of this meta-analysis was to summarize available RCTs on the performance of CADe systems in colorectal neoplasia detection.
We searched MEDLINE, EMBASE, and Cochrane Central databases until March 2020 for RCTs reporting diagnostic accuracy of CADe systems in the detection of colorectal neoplasia. The primary outcome was pooled adenoma detection rate (ADR), and secondary outcomes were adenoma per colonoscopy (APC) according to size, morphology, and location; advanced APC; polyp detection rate; polyps per colonoscopy; and sessile serrated lesions per colonoscopy. We calculated risk ratios (RRs), performed subgroup and sensitivity analyses, and assessed heterogeneity and publication bias.
Overall, 5 randomized controlled trials (4354 patients) were included in the final analysis. Pooled ADR was significantly higher in the CADe group than in the control group (791/2163 36.6% vs 558/2191 25.2%; RR, 1.44; 95% confidence interval CI, 1.27-1.62; P < .01; I2 = 42%). APC was also higher in the CADe group compared with control (1249/2163 .58 vs 779/2191 .36; RR, 1.70; 95% CI, 1.53-1.89; P < .01; I2 = 33%). APC was higher for ≤5-mm (RR, 1.69; 95% CI, 1.48-1.84), 6- to 9-mm (RR, 1.44; 95% CI, 1.19-1.75), and ≥10-mm adenomas (RR, 1.46; 95% CI, 1.04-2.06) and for proximal (RR, 1.59; 95% CI, 1.34-1.88), distal (RR, 1.68; 95% CI, 1.50-1.88), flat (RR, 1.78; 95% CI, 1.47-2.15), and polypoid morphology (RR, 1.54; 95% CI, 1.40-1.68). Regarding histology, CADe resulted in a higher sessile serrated lesion per colonoscopy (RR, 1.52; 95% CI, 1.14-2.02), whereas a nonsignificant trend for advanced ADR was found (RR, 1.35; 95% CI, .74-2.47; P = .33; I2 = 69%). Level of evidence for RCTs was graded as moderate.
According to available evidence, the incorporation of artificial intelligence as aid for detection of colorectal neoplasia results in a significant increase in the detection of colorectal neoplasia, and such effect is independent from main adenoma characteristics.
Display omitted
One-fourth of colorectal neoplasias are missed during screening colonoscopies; these can develop into colorectal cancer (CRC). Deep learning systems allow for real-time computer-aided detection ...(CADe) of polyps with high accuracy. We performed a multicenter, randomized trial to assess the safety and efficacy of a CADe system in detection of colorectal neoplasias during real-time colonoscopy.
We analyzed data from 685 subjects (61.32 ± 10.2 years old; 337 men) undergoing screening colonoscopies for CRC, post-polypectomy surveillance, or workup due to positive results from a fecal immunochemical test or signs or symptoms of CRC, at 3 centers in Italy from September through November 2019. Patients were randomly assigned (1:1) to groups who underwent high-definition colonoscopies with the CADe system or without (controls). The CADe system included an artificial intelligence–based medical device (GI-Genius, Medtronic) trained to process colonoscopy images and superimpose them, in real time, on the endoscopy display a green box over suspected lesions. A minimum withdrawal time of 6 minutes was required. Lesions were collected and histopathology findings were used as the reference standard. The primary outcome was adenoma detection rate (ADR, the percentage of patients with at least 1 histologically proven adenoma or carcinoma). Secondary outcomes were adenomas detected per colonoscopy, non-neoplastic resection rate, and withdrawal time.
The ADR was significantly higher in the CADe group (54.8%) than in the control group (40.4%) (relative risk RR, 1.30; 95% confidence interval CI, 1.14–1.45). Adenomas detected per colonoscopy were significantly higher in the CADe group (mean, 1.07 ±1.54) than in the control group (mean 0.71 ± 1.20) (incidence rate ratio, 1.46; 95% CI, 1.15–1.86). Adenomas 5 mm or smaller were detected in a significantly higher proportion of subjects in the CADe group (33.7%) than in the control group (26.5%; RR, 1.26; 95% CI, 1.01–1.52), as were adenomas of 6 to 9 mm (detected in 10.6% of subjects in the CADe group vs 5.8% in the control group; RR, 1.78; 95% CI, 1.09–2.86), regardless of morphology or location. There was no significant difference between groups in withdrawal time (417 ± 101 seconds for the CADe group vs 435 ± 149 for controls; P = .1) or proportion of subjects with resection of non-neoplastic lesions (26.0% in the CADe group vs 28.7% of controls; RR, 1.00; 95% CI, 0.90–1.12).
In a multicenter, randomized trial, we found that including CADe in real-time colonoscopy significantly increases ADR and adenomas detected per colonoscopy without increasing withdrawal time. ClinicalTrials.gov no: 04079478
Display omitted
Main Recommendations
The following recommendations for post-polypectomy colonoscopic surveillance apply to all patients who had one or more polyps that were completely removed during a high quality ...baseline colonoscopy.
1
ESGE recommends that patients with complete removal of 1 – 4 < 10 mm adenomas with low grade dysplasia, irrespective of villous components, or any serrated polyp < 10 mm without dysplasia, do not require endoscopic surveillance and should be returned to screening.
Strong recommendation, moderate quality evidence.
If organized screening is not available, repetition of colonoscopy 10 years after the index procedure is recommended. Strong recommendation, moderate quality evidence.
2
ESGE recommends surveillance colonoscopy after 3 years for patients with complete removal of at least 1 adenoma ≥ 10 mm or with high grade dysplasia, or ≥ 5 adenomas, or any serrated polyp ≥ 10 mm or with dysplasia.
Strong recommendation, moderate quality evidence.
3
ESGE recommends a 3 – 6-month early repeat colonoscopy following piecemeal endoscopic resection of polyps ≥ 20 mm.
Strong recommendation, moderate quality evidence.
A first surveillance colonoscopy 12 months after the repeat colonoscopy is recommended to detect late recurrence.
Strong recommendation, high quality evidence.
4
If no polyps requiring surveillance are detected at the first surveillance colonoscopy, ESGE suggests to perform a second surveillance colonoscopy after 5 years.
Weak recommendation, low quality evidence.
After that, if no polyps requiring surveillance are detected, patients can be returned to screening.
5
ESGE suggests that, if polyps requiring surveillance are detected at first or subsequent surveillance examinations, surveillance colonoscopy may be performed at 3 years.
Weak recommendation, low quality evidence.
A flowchart showing the recommended surveillance intervals is provided (Fig. 1).
Use of artificial intelligence may increase detection of colorectal neoplasia at colonoscopy by improving lesion recognition (CADe) and reduce pathology costs by improving optical diagnosis (CADx).
A ...multicenter library of ≥ 200 000 images from 1572 polyps was used to train a combined CADe/CADx system. System testing was performed on two independent image sets (CADe: 446 with polyps, 234 without; CADx: 267) from 234 polyps, which were also evaluated by six endoscopists (three experts, three non-experts).
CADe showed sensitivity, specificity, and accuracy of 92.9 %, 90.6 %, and 91.7 %, respectively. Experts showed significantly higher accuracy and specificity, and similar sensitivity, while non-experts + CADe showed comparable sensitivity but lower specificity and accuracy than CADe and experts. CADx showed sensitivity, specificity, and accuracy of 85.0 %, 79.4 %, and 83.6 %, respectively. Experts showed comparable performance, whereas non-experts + CADx showed comparable accuracy but lower specificity than CADx and experts.
The high accuracy shown by CADe and CADx was similar to that of experts, supporting further evaluation in a clinical setting. When using CAD, non-experts achieved a similar performance to experts, with suboptimal specificity.
Abstract
We are currently living in the throes of the COVID-19 pandemic that imposes a significant stress on health care providers and facilities. Europe is severely affected with an exponential ...increase in incident infections and deaths. The clinical manifestations of COVID-19 can be subtle, encompassing a broad spectrum from asymptomatic mild disease to severe respiratory illness. Health care professionals in endoscopy units are at increased risk of infection from COVID-19. Infection prevention and control has been shown to be dramatically effective in assuring the safety of both health care professionals and patients. The European Society of Gastrointestinal Endoscopy (www.esge.com) and the European Society of Gastroenterology and Endoscopy Nurses and Associates (www.esgena.org) are joining forces to provide guidance during this pandemic to help assure the highest level of endoscopy care and protection against COVID-19 for both patients and endoscopy unit personnel. This guidance is based upon the best available evidence regarding assessment of risk during the current status of the pandemic and a consensus on which procedures to perform and the priorities on resumption. We appreciate the gaps in knowledge and evidence, especially on the proper strategy(ies) for the resumption of normal endoscopy practice during the upcoming phases and end of the pandemic and therefore a list of potential research questions is presented. New evidence may result in an updated statement.
Main Recommendations
1
ESGE suggests that high definition endoscopy, and dye or virtual chromoendoscopy, as well as add-on devices, can be used in average risk patients to increase the endoscopist’s ...adenoma detection rate. However, their routine use must be balanced against costs and practical considerations.
Weak recommendation, high quality evidence.
2
ESGE recommends the routine use of high definition systems in individuals with Lynch syndrome.
Strong recommendation, high quality evidence.
3
ESGE recommends the routine use, with targeted biopsies, of dye-based pancolonic chromoendoscopy or virtual chromoendoscopy for neoplasia surveillance in patients with long-standing colitis.
Strong recommendation, moderate quality evidence.
4
ESGE suggests that virtual chromoendoscopy and dye-based chromoendoscopy can be used, under strictly controlled conditions, for real-time optical diagnosis of diminutive (≤ 5 mm) colorectal polyps and can replace histopathological diagnosis. The optical diagnosis has to be reported using validated scales, must be adequately photodocumented, and can be performed only by experienced endoscopists who are adequately trained, as defined in the ESGE curriculum, and audited.
Weak recommendation, high quality evidence.
5
ESGE recommends the use of high definition white-light endoscopy in combination with (virtual) chromoendoscopy to predict the presence and depth of any submucosal invasion in nonpedunculated colorectal polyps prior to any treatment.
Strong recommendation, moderate quality evidence.
6
ESGE recommends the use of virtual or dye-based chromoendoscopy in addition to white-light endoscopy for the detection of residual neoplasia at a piecemeal polypectomy scar site.
Strong recommendation, moderate quality evidence.
7
ESGE suggests the possible incorporation of computer-aided diagnosis (detection and characterization of lesions) to colonoscopy, if acceptable and reproducible accuracy for colorectal neoplasia is demonstrated in high quality multicenter in vivo clinical studies. Possible significant risks with implementation, specifically endoscopist deskilling and over-reliance on artificial intelligence, unrepresentative training datasets, and hacking, need to be considered.
Weak recommendation, low quality evidence.
The miss rate of flat advanced colorectal neoplasia is still unacceptably high, especially in the Western setting, notwithstanding the widespread implementation of quality improvement programs and ...training. It is well known that flat morphology is associated with miss rate of colorectal neoplasia, and that this subset of lesions often shows a more aggressive biological behaviour. Artificial intelligence (AI) applied to the detection of colorectal neoplasia has been shown to increase adenoma detection rate, consistently across all lesion sizes and locations in the colon. However, there is still uncertainty whether AI can reduce the miss rate of flat advanced neoplasia, mainly because all published trials report a low number of flat colorectal lesions in their training sets, and this could reduce AI accuracy for this subset of lesions. In addition, flat lesions have different morphologies with variable prevalence and potentially different accuracy in their detection. For example, the subtle appearance and rarer frequency of a non‐granular laterally spreading tumor (LST) could be much harder to identify than a granular mixed LST. In this review, we present a summary of the evidence on the role of AI in the identification of colorectal flat neoplasia, with a focus on data regarding presence of LSTs in the training/validation sets of the AI systems currently available on the market.
Artificial intelligence (AI) may detect colorectal polyps that have been missed due to perceptual pitfalls. By reducing such miss rate, AI may increase the detection of colorectal neoplasia leading ...to a higher degree of colorectal cancer (CRC) prevention.
Patients undergoing CRC screening or surveillance were enrolled in 8 centers (Italy, UK, US), and randomized (1:1) to undergo 2 same-day, back-to-back colonoscopies with or without AI (deep learning computer aided diagnosis endoscopy) in 2 different arms, namely AI followed by colonoscopy without AI or vice-versa. Adenoma miss rate (AMR) was calculated as the number of histologically verified lesions detected at second colonoscopy divided by the total number of lesions detected at first and second colonoscopy. Mean number of lesions detected in the second colonoscopy and proportion of false negative subjects (no lesion at first colonoscopy and at least 1 at second) were calculated. Odds ratios (ORs) and 95% confidence intervals (CIs) were adjusted by endoscopist, age, sex, and indication for colonoscopy. Adverse events were also measured.
A total of 230 subjects (116 AI first, 114 standard colonoscopy first) were included in the study analysis. AMR was 15.5% (38 of 246) and 32.4% (80 of 247) in the arm with AI and non-AI colonoscopy first, respectively (adjusted OR, 0.38; 95% CI, 0.23–0.62). In detail, AMR was lower for AI first for the ≤5 mm (15.9% vs 35.8%; OR, 0.34; 95% CI, 0.21–0.55) and nonpolypoid lesions (16.8% vs 45.8%; OR, 0.24; 95% CI, 0.13–0.43), and it was lower both in the proximal (18.3% vs 32.5%; OR, 0.46; 95% CI, 0.26–0.78) and distal colon (10.8% vs 32.1%; OR, 0.25; 95% CI, 0.11–0.57). Mean number of adenomas at second colonoscopy was lower in the AI-first group as compared with non-AI colonoscopy first (0.33 ± 0.63 vs 0.70 ± 0.97, P < .001). False negative rates were 6.8% (3 of 44 patients) and 29.6% (13 of 44) in the AI and non-AI first arms, respectively (OR, 0.17; 95% CI, 0.05–0.67). No difference in the rate of adverse events was found between the 2 groups.
AI resulted in an approximately 2-fold reduction in miss rate of colorectal neoplasia, supporting AI-benefit in reducing perceptual errors for small and subtle lesions at standard colonoscopy. ClinicalTrials.gov, Number: NCT03954548.
When the endoscopist is assisted by artificial intelligence, the risk of missing colorectal neoplasia decreased substantially, reassuring a better stratification of the neoplastic risk of the individual patient, and a higher degree of colorectal cancer prevention.
Clinical management after complete endoscopic resection of pT1 colorectal cancers (CRCs) is still under debate. Follow-up data are heterogeneous and poorly reported, resulting in variable clinical ...management. Our aim was to meta-analyze recurrence and cancer-specific mortality (CSM) occurring after endoscopic resection of low- and high-risk pT1 CRCs undergoing conservative (nonsurgical) management.
Literature was systematically searched until February 2019 for studies describing patients with pT1 CRCs, histologically classifiable as low or high risk, endoscopically resected without complementary surgery and with ≥12 months of follow-up. Pooled cumulative incidence (and incidence rate when specific follow-up intervals were available) of recurrence and CSM were calculated separately for low- and high-risk pT1 CRCs. Quality, publication bias, and heterogeneity were explored.
Pooled cumulative incidences of recurrence and CSM among high-risk lesions (5 studies, 571 patients) were, respectively, 9.5% (95% confidence interval CI, 6.7%-13.3%; I2 = 38.4%) and 3.8% (95% CI, 2.4%-5.8%; I2 = 0%), whereas among low-risk lesions (7 studies, 650 patients) they were, respectively, 1.2% (95% CI, .6%-2.5%; I2 = 0%) and .6% (95% CI, .2%-1.7%; I2 = 0%). Pooled incidence rates of recurrence and CSM among high-risk lesions (3 cohorts, 237 patients) were, respectively, 11 (95% CI, 2-20; I2 = 43.3%) and 4 (95% CI, 1-7; I2 = 0%) per 1000 patient-years, whereas among low-risk lesions (3 cohorts, 229 patients) they were 3 (95% CI, 0-6; I2 = 0%) and 2 (95% CI, 0-4; I2 = 0%) per 1000 patient-years, respectively. No publication bias or significant heterogeneity was found.
Pooled estimates of adverse events after endoscopic resection of pT1 CRCs suggest a conservative approach for low-risk lesions. In high-risk lesions, increased surgical risk might justify a conservative management, whereas fitness for surgery makes surgical completion appropriate.
Display omitted