Publications by Year: 2023

BACKGROUND AND AIMS: The adenoma detection rate is a key quality metric for colonoscopy and is inversely related to the post-colonoscopy colorectal cancer rate. Natural language processing can be used to automate the generation of such quality metrics from colonoscopy reports. We performed a systematic review and meta-analysis on the performance of natural language processing (NLP) in identifying adenoma detection in colonoscopy and paired pathology reports.

METHODS: We performed a systematic review and meta-analysis according to Preferred Reporting Items for Systematic Reviews and Meta-Analyses recommendations. A literature query was conducted on MEDLINE, Embase, and Cochrane Database of Systematic Reviews through July 2022. Studies were included if they reported on the operator characteristics of an NLP algorithm in interpreting adenoma detection in colonoscopy and pathology reports. Two authors independently screened studies and abstracted data using an a priori designed data collection form. Performance characteristics were pooled by first using a univariate analysis, followed by a bivariate analysis of sensitivity and specificity.

RESULTS: The pooled specificity and sensitivity for identifying adenoma detection were .997 (95% confidence interval [CI], .984-.999) and .978 (95% CI, .938-.992). The pooled positive predictive value, negative predictive value, and F1 score were .997 (95% CI, .979-1.00), .977 (95% CI, .938-.992), and .982 (95% CI, .957-.993), respectively. In the bivariate analysis, the pooled specificity and sensitivity were .992 (95% CI, .978-.997) and .973 (95% CI, .929-.990). The NLP systems performed similarly well in identifying the detection of sessile serrated lesions and advanced adenomas.

CONCLUSIONS: NLP systems can identify adenoma detection from colonoscopy and pathology reports with strong operator characteristics.

BACKGROUND AND AIMS: Despite the widespread prevalence of post-traumatic stress disorder (PTSD), there are few data on its effects on utilization of colonoscopy, an invasive procedure involving heightened vulnerability. The aim of this study was to describe factors, including PTSD, associated with screening and surveillance colonoscopy utilization.

METHODS: We designed a retrospective cohort study examining colonoscopy patterns from 2009 to 2019 among patients aged 50 to 74 years. Primary outcomes were factors associated with screening and surveillance colonoscopy, assessed with multivariable logistic regression. A stratified analysis of screening sedation modality was also performed.

RESULTS: A total of 65,062 patients were included, with PTSD reported among 267 (.41%). Most patients with a history of PTSD underwent screening colonoscopy (58.4%) compared with those without a history of PTSD (10.9%). The adjusted odds ratio (OR) for undergoing screening colonoscopy if patients had a history of PTSD was 3.15 (95% confidence interval [CI], 2.37-4.21). Similarly, the adjusted OR for undergoing surveillance colonoscopy if patients had a history of PTSD was 2.05 (95% CI, 1.46-2.85). When stratifying according to procedural sedation versus monitored anesthesia care (MAC), patients without PTSD were more likely to undergo surveillance if screening was with MAC (adjusted OR, 3.90; 95% CI, 2.77-5.45). However, this did not have a similarly positive effect among patients with PTSD (adjusted OR, 1.60; 95% CI, .40-5.41).

CONCLUSIONS: This study validates known characteristics associated with colonoscopy utilization and identifies PTSD as an unexpectedly important protective factor in both screening and surveillance, which has not previously been reported. The lack of positive impact of MAC on surveillance colonoscopy in this population underscores the need for an individualized decision on anesthesia to shape best practices in endoscopic trauma-informed care.

BACKGROUND AND AIMS: Magnetically controlled capsule endoscopy (MCCE) has the potential to allow an operator to move a video capsule endoscope inside the foregut. The primary objective of this pilot study was to demonstrate that MCCE could visualize the major anatomic regions of the stomach in symptomatic patients before an EGD. Secondary objectives were measuring patient satisfaction, patient safety, and comparing MCCE with a follow-up EGD in the detection of any significant gastric lesions.

METHODS: In this prospective, single-arm, single-center, comparative study, adult patients aged ≥18 years who were referred for an EGD as part of a standard evaluation for symptoms were approached for participation in the study. Participants received MCCE before the EGD. MCCE videos were reviewed by 2 independent physicians and compared with subsequent EGD. Patients were followed for 30 days for safety outcomes and satisfaction.

RESULTS: In this study of 40 patients, MCCE detected each of the 6 preidentified major gastric anatomic landmarks with a greater than 95% rate of visualization. Thirty-five patients received a follow-up EGD, and no high-risk lesions were missed with MCCE. Patients preferred MCCE to EGD (80%-13%), and there were no adverse events.

CONCLUSIONS: In the first pilot study of MCCE in the United States, a high rate of visualization of all regions of the gastric mucosa was achieved. In addition, high satisfaction and no adverse events were recorded. Future studies will focus on higher-risk cohorts to confirm the accuracy of detection of benign, premalignant, and malignant gastric lesions.

BACKGROUND AND AIMS: Hybrid endoscopic submucosal dissection (ESD) may overcome the complexity of conventional ESD associated with gastric lesions. The aim of this study was to perform a systematic review and meta-analysis to compare the efficacy and safety of hybrid versus conventional ESD for the treatment of gastric lesions.

METHODS: Individualized search strategies were developed in accordance with Preferred Reporting Items for Systematic Reviews and Meta-Analyses and Meta-Analysis of Observational Studies in Epidemiology guidelines. Pooled proportions were calculated with rates estimated using random-effects models. Measured outcomes included en-bloc resection, procedure-associated adverse events, and procedure duration. Heterogeneity was assessed with the I 2 statistic and publication bias using funnel plots and Egger regression testing.

RESULTS: Of 5 included comparator studies (hybrid ESD, 184 patients; conventional ESD, 289 patients), 1 was a randomized trial and 4 were retrospective observational studies. Mean patient age was 68.89 ± 4.75 years, and the average lesion size was 17.81 ± 5.58 mm. Hybrid ESD patients were older (P < .001) with smaller lesions (15.75 ± 4.72 mm vs 19.12 ± 5.69 mm; P < .001). Overall, the en-bloc resection rate was significantly decreased for hybrid ESD (odds ratio [OR], .11; 95% confidence interval [CI], .02-.62; P = .010). Total adverse events were not different between groups (OR, 1.56; 95% CI, .44-5.53; P = .490). Rates of delayed bleeding (OR, 1.47; 95% CI, .34-6.40; P = .610) and perforation (OR, 2.41; 95% CI, .65-9.12; P = .194) were also not significantly different. Procedure time was significantly shorter for hybrid ESD (mean difference, 15.13 minutes; 95% CI, 4.05-26.21; P = .007).

CONCLUSIONS: Although hybrid ESD for gastric lesions was associated with significantly shorter procedure times compared with conventional ESD, hybrid ESD was associated with lower rates of en-bloc resection and similar adverse events.

Coronavirus disease-2019 (COVID-19) is caused by severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) infection. SARS-CoV-2 virus is internalized by surface receptors, e.g., angiotensin-converting enzyme-2 (ACE2). Clinical reports suggest that non-insulin dependent diabetes mellitus (DM-II) may enhance COVID-19. This bioinformatics study investigated how DM-II augments COVID-19 complications through molecular interactions with cytokines/chemokines, using QIAGEN Ingenuity Pathway Analysis (IPA) and CLC Genomics Workbench 22 (CLCG-22). "(Iβ-CG) RNA-sequencing of (Iβ-CG) through CLCG-22 (SRA SRP287500) were analyzed to identify differential expression of (Iβ-CG). IPA's QIAGEN Knowledge Base (QKB) was also used to retrieve 88 total molecules shared between DM-II and SARS-CoV-2 infection to characterize and identify Iβ-CG, due to close association with DM-II. Molecules directly associated with ACE2 and cytokines/chemokines were also identified for their association with SARS-CoV-2 infection. Using IPA, it was found that 3 Iβ-CG (SCL2A2, PPARγ, and CPLX8) are common in both diseases that were downregulated by DM-II. Their downregulation occurred due to increased activity of cytokines/chemokines and ACE2. Collectively, this network meta-analysis demonstrated that interaction of SARS-CoV-2 with ACE2 could primarily induce endothelial cell dysfunction. Identification of common molecules and signaling pathways between DM-II and SARS-CoV-2 infection in this study may lead to further discovery of therapeutic measures to simultaneously combat both diseases.

BACKGROUND: Post-stroke cognitive impairment and dementia (PSCID) is a sequel of ischemic stroke (IS), highly prevalent and linked to poor long-term outcomes. Thus, early recognition of the clinical determinants of PSCID is urgent for identifying high-risk individuals who are susceptible to PSCID. And investigating objective measures of PSCID in relation to patient-reported outcome measures (PROMs) is essential for understanding the impact of IS. Here we identify the clinical determinants associated with PSCID and the relationship of PSCID to patient-reported outcomes in a population with IS.

METHODS: This was a cohort study. We enrolled 138 patients who were admitted to our hospital between February 2017 and February 2020, with IS and no pre-stroke diagnosis of dementia. Clinical variables were acquired on admission. At 3 months, patients underwent a follow-up evaluation including the Telephone Interview for Cognitive Status (TICS), modified Rankin scale (mRS), Barthel Index (BI), and PROMs, using the Patient-Reported Outcomes Measurement information System Global Health (PROMIS GH). MCI/Dementia was defined as a TICS score of <36. Regression analyses were used to identify clinical, functional, and patient-reported outcome determinants of the 3-month TICS score. Analyses were adjusted for age, stroke severity, and prior IS.

RESULTS: At follow-up, 113 participants (82%) were found to have MCI/Dementia. Patients with PSCID were more likely to be older, and at 3-months post-stroke they had lower rates of PROMIS GH T Mental (mean 47.69 vs. 52.13) and T Physical (mean 46.75 vs. 50.64). In multivariable linear regression analyses, increasing age (β = -0.07, p = 0.03) and Peripheral Artery Disease (PAD; β = -3.60, p = 0.03) were independently associated with a lower TICS score. Functional and patient-reported outcomes were also associated with worse TICS, including mRS ≥ 2, BI, T Mental, Global Mental, T Physical, and Global Physical in adjusted analyses. Individual components of PROMs were also associated with TICS, including quality of life, mental health, social satisfaction, and physical activities.

CONCLUSIONS: In patients with IS, increased age and a pre-admission diagnosis of PAD are independently associated with worse objective measures of PSCID. Worse functional and patient-reported outcomes are also strongly linked to PSCID.

Some argue that the term transient ischemic attack (TIA) has become obsolete in the current age of advanced modern technology. Let us look back and analyze the term, its support and detraction, and its potential continued usefulness. The best way to begin is a review of the history of the term.