Early Colonoscopy Does Not Improve Outcomes of Patients With Lower Gastrointestinal Bleeding: Systematic Review of Randomized Trials
Clinical gastroenterology and hepatology : the official clinical practice journal of the American Gastroenterological Association. 2019
BACKGROUND & AIMS Guidelines recommend colonoscopy evaluation within 24 hours of presentation or admission in patients with high-risk or severe acute lower gastrointestinal bleeding (LGIB). Meta-analyses of the timing of colonoscopy have relied primarily on observational studies that had major potential for bias. We performed a systematic review of randomized trials to determine optimal timing of colonoscopy for patients hospitalized with acute LGIB. METHODS We searched publication databases through July 2019 and abstracts from gastroenterology meetings through November 2019 for randomized trials of patients with acute LGIB or hematochezia. We searched for studies that compared early colonoscopy (within 24 hours) with elective colonoscopy beyond 24 hours and/or other diagnostic tests. Our primary outcome was further bleeding, defined as persistent or recurrent bleeding after index examination. Secondary outcomes included mortality, diagnostic yield (identifying source of bleeding), endoscopic intervention, and any primary hemostatic intervention (endoscopic, surgical, or interventional radiologic). We performed dual independent review, data extraction, and risk of bias assessments. We performed the meta-analysis using a random-effects model. RESULTS Our final analysis included data from 4 randomized trials. Further bleeding was not decreased among patients who received early vs later, elective colonoscopy (relative risk [RR] for further bleeding with early colonoscopy, 1.57; 95% CI. 0.74-3.31). We did not find significant differences in the secondary outcomes of mortality (RR, 0.93; 95% CI, 0.05-17.21), diagnostic yield (RR, 1.09; 95% CI, 0.99-1.21), endoscopic intervention (RR, 1.53; 95% CI, 0.67-3.48), or any primary hemostatic intervention (RR, 1.33; 95% CI, 0.92-1.92). CONCLUSIONS In a meta-analysis of randomized trials, we found that colonoscopy within 24 hours does not reduce further bleeding or mortality in patients hospitalized with acute LGIB. Based on these findings, patients hospitalized with acute LGIB do not generally require early colonoscopy.
Machine Learning to Predict Outcomes in Patients with Acute Gastrointestinal Bleeding: A Systematic Review
Digestive diseases and sciences. 2019
Risk stratification of patients with gastrointestinal bleeding (GIB) is recommended, but current risk assessment tools have variable performance. Machine learning (ML) has promise to improve risk assessment. We performed a systematic review to evaluate studies utilizing ML techniques for GIB. Bibliographic databases and conference abstracts were searched for studies with a population of overt GIB that used an ML algorithm with outcomes of mortality, rebleeding, hemostatic intervention, and/or hospital stay. Two independent reviewers screened titles and abstracts, reviewed full-text studies, and extracted data from included studies. Risk of bias was assessed with an adapted Quality in Prognosis Studies tool. Area under receiver operating characteristic curves (AUCs) were the primary assessment of performance with AUC ≥ 0.80 predefined as an acceptable threshold of good performance. Fourteen studies with 30 assessments of ML models met inclusion criteria. No study had low risk of bias. Median AUC reported in validation datasets for predefined outcomes of mortality, intervention, or rebleeding was 0.84 (range 0.40-0.98). AUCs were higher with artificial neural networks (median 0.93, range 0.78-0.98) than other ML models (0.81, range 0.40-0.92). ML performed better than clinical risk scores (Glasgow-Blatchford, Rockall, Child-Pugh, MELD) for mortality in upper GIB. Limitations include heterogeneity of ML models, inconsistent comparisons of ML models with clinical risk scores, and high risk of bias. ML generally provided good-excellent prognostic performance in patients with GIB, and artificial neural networks tended to outperform other ML models. ML was better than clinical risk scores for mortality in upper GIB.