A Nationwide Survey of Gastroenterology Program Leadership Regarding Implementation of the GI Hospitalist Model.

BACKGROUND: Academic gastroenterology (GI) hospitalists are increasing, however the impacts on fellowship training and clinical care are unclear. Motivations for implementation of the GI hospitalist model are uninvestigated. AIMS: We aimed to determine the prevalence of GI hospitalists, explore motivations for and against adoption of a GIH model, and investigate the model's effects on fellowship training. METHODS: Leadership at current general GI fellowships were surveyed about current staffing models, as well as effects and perceptions of the hospitalist model. RESULTS: There was a total of 52 (26%) respondents and 12 (23%) reported having a GI hospitalist at their institution. A majority of respondents stated burnout and reduced time on service for other faculty was a primary reason for hiring a GI hospitalist. DISCUSSION: The largest perceived benefit of a hospitalist is reduced burnout and time on service for outpatient GI faculty. Many respondents also believed a GIH would improve fellowship education and quality of inpatient care.

In adults using oral anticoagulants, PPI use may be associated with lower risk for upper GI bleeding.

SOURCE CITATION: Kurlander JE, Barnes GD, Fisher A, et al. Association of antisecretory drugs with upper gastrointestinal bleeding in patients using oral anticoagulants: a systematic review and meta-analysis. Am J Med. 2022;135:1231-43.e8. 35679879.

Comprehensive National Inpatient Sample data reveals low but rising Pneumocystis jiroveci pneumonia risk in inflammatory bowel disease patients.

Background: There are only limited data to guide the management of infectious risk for Pneumocystis jiroveci pneumonia (PCP) in patients with inflammatory bowel disease (IBD). We evaluated the frequency of admissions for PCP among patients with IBD, as well as the temporal trend in PCP admission rates and the contribution of non-IBD risk factors to the development of infection. Methods: The National Inpatient Sample from 2016-2017 was queried for all admissions involving both PCP and either Crohn's disease or ulcerative colitis. Inpatient outcomes associated with PCP and additive risk factors for development of PCP within the IBD patient population were assessed using multivariate regression. Linear regression was performed on data from 2002-2017 to measure infectious trends over time. Results: There were an estimated 225 admissions involving PCP among patients with IBD from 2016-2017 nationwide, representing 0.035% of total admissions. IBD patients with PCP faced a 4.67-fold higher adjusted odds of inpatient mortality (95% confidence interval 1.72-12.66), while 49% of patients with IBD who developed PCP had an unrelated risk factor. The most common factors were HIV and congenital immunodeficiency, both of which were associated with PCP in adjusted regression. The infectious incidence of PCP increased by 141% from 2002 to 2017 (P=0.003). Conclusions: National admissions data indicate that significant PCP is rare in IBD patients. Routine PCP prophylaxis is probably not necessary, although further study of high-risk subgroups of patients is required. The rising incidence of PCP indicates a need for continued surveillance.

Outcomes in patients with inflammatory bowel disease and acute gastrointestinal symptoms who test indeterminate for Clostridioides difficile.

Background: Inflammatory bowel disease (IBD) and Clostridioides difficile infection (CDI) can present with similar symptoms. The current preferred method for diagnosing CDI is the nucleic acid amplification test (NAAT) for C. difficile in stool, followed by reflex toxin enzyme immunoassay (EIA) when NAAT is positive. The clinical significance of NAAT(+)/EIA(-) in the IBD population is uncertain. Methods: This retrospective cohort included IBD patients who presented with acute onset of gastrointestinal symptoms and a C. difficile NAAT(+) test. The primary outcome was C. difficile recurrence within 12 months. Other outcomes examined included hospital admissions within 30 days of CDI, change of IBD maintenance therapy within 90 days of CDI, and complications such as bowel resection or death. Results: A total of 71 patients were included. Eighty-four percent of the tests were EIA(-) and among the EIA(-) 88% were treated with antibiotics. Outcomes between EIA(+) and EIA(-) were not significantly different in terms of recurrences, admissions, changes to IBD medications or complications. Outcomes were also similar when comparing those who received antibiotic therapy to those who did not. Conclusions: Our cohort did not demonstrate a significant difference in outcomes between EIA(+) and EIA(-) C. difficile patients. Treatment for EIA(-) patients did not improve outcomes. Even though there may be a role for antibiotic therapy in IBD patients who test NAAT(+)/EIA(-) for C. difficile, further studies will be needed to identify that subpopulation.

Comparative Efficacy of Endoscopic Assist Devices on Colonic Adenoma Detection: A Systematic Review With Network Meta-analysis.

BACKGROUND AND AIMS: Numerous endoscopic assist devices exist, yet data surrounding their comparative efficacy is lacking. We conducted a systematic review with network meta-analysis to determine the comparative efficacy of endoscopic assist devices on colonic adenoma detection. METHODS: A systematic search was performed using multiple electronic databases through July 2020, to identify all randomized controlled trials and dual-arm observational studies compared with either other endoscopic assist devices and/or standard colonoscopy. The primary outcome was adenoma detection rate (ADR). Secondary outcomes included polyp detection rate (PDR), serrated adenoma detection rate (SADR), right-sided adenoma detection rate (RADR), and proximal adenoma detection rate (PADR). RESULTS: Fifty-seven studies (31,051 patients) met inclusion criteria and were analyzed. Network meta-analysis identified an enhanced ADR among (clear) cap [odds ratio (OR): 2.69, 95% confidence interval (CI): 1.45-4.99], endocuff, (OR: 4.95, 95% CI: 3.15-7.78), and endoring (OR: 3.68, 95% CI: 1.47-9.20)-with no significant difference amongst any particular device. Similar findings for PDR were also seen. Enhanced SADR was identified for endocuff (OR: 9.43) and endoring (OR: 4.06) compared with standard colonoscopy. Enhanced RADR (OR: 5.36) and PADR (OR: 3.78) were only identified for endocuff. Endocuff comparatively demonstrated the greatest ADR, PDR, and SADR, but this was not significant when compared with the other assist devices. Subgroup analysis of randomized controlled trials identified enhanced PDR and ADR for both cap and endocuff. CONCLUSIONS: Endoscopic assist devices displayed increased ADR and PDR as compared with standard colonoscopy and thus should be widely adopted. A nonsignificant trend was seen toward higher efficacy for the endocuff device.

Impact of Cirrhosis on Outcomes in Inflammatory Bowel Disease Hospitalizations.

BACKGROUND: Evidence regarding outcomes in inflammatory bowel disease (IBD) hospitalizations with coexisting cirrhosis is scant. We queried the National Inpatient Sample (NIS) database to evaluate the impact of cirrhosis on hospitalization characteristics and outcomes in patients with Crohn's disease and ulcerative colitis. METHODS: All admissions that listed IBD as a primary diagnosis by ICD-10-CM code (K50.X for Crohn's disease and K51.X for ulcerative colitis) in the NIS for 2016 and 2017 were included. Attributes of admissions with cirrhosis (K74.XX, 70.3, 78.81, and 71.7) were compared with noncirrhosis IBD admissions. The primary outcome was inpatient mortality. Length of stay and total hospital charges comprised secondary outcomes. RESULTS: A total weighted sample of 276,430 IBD admissions were identified, including 4615 with a concomitant diagnosis of cirrhosis. In a multivariate model, after adjusting for comorbidities, age, alimentary surgery during the admission and hospital type (teaching, urban nonteaching or rural), the presence of cirrhosis was associated with a higher inpatient mortality [odds ratio: 1.57; 95% confidence interval (CI): 1.16-2.15] and increased cost of admission (mean difference $11,651; 95% CI: 3830-19,472). No difference was noted in length of stay (difference: 0.44 d; 95% CI: -0.12-1.02) among these groups. Among admission diagnoses, infectious complications were the primary cause of death in 93.3% (95% CI: 87.1%-99.5%) of all inpatient mortality in the IBD with cirrhosis cohort as compared with 80.1% (95% CI: 77.6%-82.7%) of the mortality among IBD patients without cirrhosis ( P =0.01). CONCLUSIONS: This study demonstrates that the presence of cirrhosis has an independent negative impact on outcomes for hospitalized patients with IBD as reflected by increased in-hospital mortality and higher cost of admission. A majority of the mortality was attributable to infections.

High Rates of Mortality in Geriatric Patients Admitted for Inflammatory Bowel Disease Management.

GOAL: The goal of this study was to evaluate the inpatient mortality risk among geriatric patients with inflammatory bowel disease (IBD). BACKGROUND: The challenges of caring for elderly patients with IBD will increase with the aging of the US population. Given the complications of hospitalization, we set to examine if elderly patients age older than 65 were at higher risk of mortality. MATERIALS AND METHODS: All patients with ulcerative colitis (UC) or Crohn's disease (CD) in the National Inpatient Sample (NIS) from 2016 and 2017 as the primary diagnosis or secondary diagnosis with an IBD-related cause of admission were included. Outcomes for patients aged above 65 were compared with below 65 using multivariable survey-adjusted regression. CD and UC were analyzed separately. RESULTS: In 2016-2017, there were an estimated 162,800 admissions for CD and related complications compared with 96,450 for UC. In total, 30% of UC and 20% of CD admissions were geriatric. Geriatric status was associated with higher odds of mortality for CD [odds ratio (OR)=3.47, 95% confidence interval (CI): 2.72-4.44] and UC (OR=2.75, 95% CI: 2.16-3.49) after adjustment for comorbidities, admission type, hospital type, inpatient surgery, and IBD subtype. The cause of death was ∼80% infectious in both CD and UC in all groups. An average of 0.19 days (95% CI: 0.05-0.34) and $2467 (95% CI: 545-4388) increase was seen for geriatric CD patients. No significant change was seen for UC. CONCLUSIONS: Age over 65 was independently associated with higher odds of death in both UC and CD patients, even after appropriate adjustment. Further research is needed to optimize care for this growing patient population.

Pros and Cons of Performing Early Endoscopy in Geriatric Patients Admitted with Non-variceal Upper Gastrointestinal Bleeding: Analysis of the US National Inpatient Database.

BACKGROUND: Age greater than 65 years is a well-defined risk factor for increased mortality in patients with non-variceal upper gastrointestinal bleeding (NVGIB). Endoscopy is indicated in most patients at any age but presents unique risks in the elderly cohort, and ideal timing is unclear. This study examined the association between outcomes and early (within 24 h) esophagogastroduodenoscopy (EGD) among elderly patients with NVGIB. METHODS: All patients over age 65 admitted primarily for NVGIB who underwent EGD were included from the National Inpatient Sample 2016-2017. Clinical outcomes stratified by early EGD versus late EGD were compared after adjustment for comorbidities and bleeding severity using inverse probability of treatment weighting with survey-adjusted linear and logistic regression. RESULTS: Out of estimated 625,530 admissions with a primary diagnosis of NVGIB, 120,835 met eligibility criteria; 24,830 underwent early EGD. Mean length of stay and total charges decreased by 1.17 days (95%CI 1.04-1.30, P < 0.001) and $5717.24 (95%CI 4034.57-7399.91, P < 0.001), respectively, in the early EGD group. Early EGD increased the odds ratio of death 1.32 (95%CI 1.06-1.64, P 0.01) and transfer to other hospitals 1.48 (95%CI 1.22-1.81, P < 0.001). No change was seen in the requirement for surgery or angiography. Rates of discharge to a nursing facility or home health were similar. CONCLUSION: In a comprehensive cohort of geriatric patients with NVGIB, early EGD is associated with decreased hospital stay and charges, but also with increased mortality and inter-hospital transfer. Further research is needed to determine the optimal management of this vulnerable population.

Serum trough levels of infliximab are not associated with peripheral arthralgia activity in patients with inflammatory bowel disease.

BACKGROUND: Infliximab is an efficacious therapy for inflammatory bowel disease and may play a role in management of some extraintestinal manifestations. While higher trough levels of infliximab are associated with higher rates of disease remission, the association between trough levels of infliximab and arthralgia activity characterised as an extraintestinal manifestation has yet to be defined. OBJECTIVE: We aimed to assess the association between serum trough levels of infliximab and peripheral arthralgia activity in patients with inflammatory bowel disease. DESIGN: In this cross-sectional study, we identified patients with inflammatory bowel disease on infliximab therapy with known history of arthralgias attributed to an extraintestinal manifestation. Collected variables included disease phenotype, medications (such as thiopurines or methotrexate), Harvey Bradshaw Index, partial Mayo score, C reactive protein, trough levels of infliximab and anti-infliximab antibodies. The primary outcome was active patient-reported arthralgia. RESULTS: Out of 267 patients included, 65 (24.4%) had active arthralgias at the time the trough level of infliximab was measured. No significant differences in trough levels were seen between those patients with and without arthralgias. Patients on combination therapy with methotrexate or thiopurines or those with detectable anti-infliximab antibodies were not more likely to have inactive arthralgias (OR 0.99, 95% CI 0.57 to 1.74, p=0.99 and OR 1.94, 95% CI 0.9 to 4.1, p=0.09, respectively). CONCLUSIONS: This study suggests that although therapeutic drug monitoring of infliximab can have a role in the management of Crohn's disease and ulcerative colitis, it does not seem to be useful in managing arthralgias associated with inflammatory bowel disease.

The impact of obesity on hospitalized patients with ulcerative colitis.

BACKGROUND: Obesity is the fifth leading risk factor for mortality in the world and it has increased among patients with ulcerative colitis in recent years. We examined the impact of obesity on the hospitalized patients admitted primarily with a diagnosis of ulcerative colitis. METHODS: We used the National Inpatient Sample data for the year 2016 to identify patients with ulcerative colitis and compared obese and non-obese patients in terms of length of hospital stay, total charges, and mortality. We used multiple imputations to estimate missing values and survey analysis to estimate the outcomes, and we adjusted for confounders by implementing the inverse probability of treatment weighting using propensity score. RESULTS: A total of 61,075 admissions with ulcerative colitis were identified. Among these, 6020 were diagnosed with obesity. Baseline hospital and patient characteristics between the 2 groups were notable for differences in age and sex. Patients with obesity were found to have a mean hospital stay longer by 0.57 days (95% confidence interval [CI] 0.22-0.93; P=0.002) and charges $6341.71 higher (95%CI 2499.72-10,183.71; P=0.001) compared to non-obese patients. There was no difference in hospital mortality, with an odds ratio of 0.28 (95%CI 0.04-2.05; P=0.212). CONCLUSION: In a comprehensive review of inpatient admissions in 2016, primarily for ulcerative colitis, obesity was associated with a longer hospital stay and higher total charges per admission after balancing of confounders.

Drugmonizome and Drugmonizome-ML: integration and abstraction of small molecule attributes for drug enrichment analysis and machine learning.

Understanding the underlying molecular and structural similarities between seemingly heterogeneous sets of drugs can aid in identifying drug repurposing opportunities and assist in the discovery of novel properties of preclinical small molecules. A wealth of information about drug and small molecule structure, targets, indications and side effects; induced gene expression signatures; and other attributes are publicly available through web-based tools, databases and repositories. By processing, abstracting and aggregating information from these resources into drug set libraries, knowledge about novel properties of drugs and small molecules can be systematically imputed with machine learning. In addition, drug set libraries can be used as the underlying database for drug set enrichment analysis. Here, we present Drugmonizome, a database with a search engine for querying annotated sets of drugs and small molecules for performing drug set enrichment analysis. Utilizing the data within Drugmonizome, we also developed Drugmonizome-ML. Drugmonizome-ML enables users to construct customized machine learning pipelines using the drug set libraries from Drugmonizome. To demonstrate the utility of Drugmonizome, drug sets from 12 independent SARS-CoV-2 in vitro screens were subjected to consensus enrichment analysis. Despite the low overlap among these 12 independent in vitro screens, we identified common biological processes critical for blocking viral replication. To demonstrate Drugmonizome-ML, we constructed a machine learning pipeline to predict whether approved and preclinical drugs may induce peripheral neuropathy as a potential side effect. Overall, the Drugmonizome and Drugmonizome-ML resources provide rich and diverse knowledge about drugs and small molecules for direct systems pharmacology applications. Database URL: https://maayanlab.cloud/drugmonizome/.

Appyters: Turning Jupyter Notebooks into data-driven web apps.

Jupyter Notebooks have transformed the communication of data analysis pipelines by facilitating a modular structure that brings together code, markdown text, and interactive visualizations. Here, we extended Jupyter Notebooks to broaden their accessibility with Appyters. Appyters turn Jupyter Notebooks into fully functional standalone web-based bioinformatics applications. Appyters present to users an entry form enabling them to upload their data and set various parameters for a multitude of data analysis workflows. Once the form is filled, the Appyter executes the corresponding notebook in the cloud, producing the output without requiring the user to interact directly with the code. Appyters were used to create many bioinformatics web-based reusable workflows, including applications to build customized machine learning pipelines, analyze omics data, and produce publishable figures. These Appyters are served in the Appyters Catalog at https://appyters.maayanlab.cloud. In summary, Appyters enable the rapid development of interactive web-based bioinformatics applications.

Prioritizing Pain-Associated Targets with Machine Learning.

While hundreds of genes have been associated with pain, much of the molecular mechanisms of pain remain unknown. As a result, current analgesics are limited to few clinically validated targets. Here, we trained a machine learning (ML) ensemble model to predict new targets for 17 categories of pain. The model utilizes features from transcriptomics, proteomics, and gene ontology to prioritize targets for modulating pain. We focused on identifying novel G-protein-coupled receptors (GPCRs), ion channels, and protein kinases because these proteins represent the most successful drug target families. The performance of the model to predict novel pain targets is 0.839 on average based on AUROC, while the predictions for arthritis had the highest accuracy (AUROC = 0.929). The model predicts hundreds of novel targets for pain; for example, GPR132 and GPR109B are highly ranked GPCRs for rheumatoid arthritis. Overall, gene-pain association predictions cluster into three groups that are enriched for cytokine, calcium, and GABA-related cell signaling pathways. These predictions can serve as a foundation for future experimental exploration to advance the development of safer and more effective analgesics.

Artificial intelligence-based prediction of transfusion in the intensive care unit in patients with gastrointestinal bleeding.

OBJECTIVE: Gastrointestinal (GI) bleeding commonly requires intensive care unit (ICU) in cases of potentialhaemodynamiccompromise or likely urgent intervention. However, manypatientsadmitted to the ICU stop bleeding and do not require further intervention, including blood transfusion. The present work proposes an artificial intelligence (AI) solution for the prediction of rebleeding in patients with GI bleeding admitted to ICU. METHODS: A machine learning algorithm was trained and tested using two publicly available ICU databases, the Medical Information Mart for Intensive Care V.1.4 database and eICU Collaborative Research Database using freedom from transfusion as a proxy for patients who potentially did not require ICU-level care. Multiple initial observation time frames were explored using readily available data including labs, demographics and clinical parameters for a total of 20 covariates. RESULTS: The optimal model used a 5-hour observation period to achieve an area under the curve of the receiving operating curve (ROC-AUC) of greater than 0.80. The model was robust when tested against both ICU databases with a similar ROC-AUC for all. CONCLUSIONS: The potential disruptive impact of AI in healthcare innovation is acknowledge, but awareness of AI-related risk on healthcare applications and current limitations should be considered before implementation and deployment. The proposed algorithm is not meant to replace but to inform clinical decision making. Prospective clinical trial validation as a triage tool is warranted.

Improving All-Cause Inpatient Mortality After Percutaneous Endoscopic Gastrostomy.

BACKGROUND AND AIMS: Percutaneous gastrostomy (PEG) is a common inpatient procedure. Prior data from National Inpatient Sample (NIS) in 2006 reported a mortality rate of 10.8% and recommended more careful selection of PEG candidates. This study assessed for improvement in the last 10 years in mortality rate and complications for hospitalized patients. METHODS: A retrospective cohort analysis of all adult inpatients in the NIS from 2006 to 2016 undergoing PEG placement compared demographics and indication for PEG placement per ICD coding. Survey-based means and proportions were compared to 2006, and rates of change in mortality and complication rates were trended from 2006 through 2016 and compared with linear regression. Multivariable survey-adjusted logistic regression was used to determine predictors of mortality and complications in the 2016 sample. RESULTS: A total of 155,550 patients underwent PEG placement in 2016, compared with 174,228 in 2006. Mortality decreased from 10.8 to 6.6% without decreased comorbidities (p < 0.001). This trend was gradual and persistent over 10 years in contrast to a stable overall inpatient mortality rate (p = 0.113). Stroke remained the most common indication (29.7%). The majority of patients (64.6%) had Medicare. Indications for placement were stable. Complication rates were stable from 2006 (4.4%) to 2016 (5.1%) (p = 0.201). CONCLUSIONS: Inpatient PEG placement remains common. Despite similar patient characteristics, mortality has decreased by approximately 40% over the last 10 years without a decrease in complications likely reflecting improved patient selection.