Development and Validation of Machine Learning Models to Predict Admission From Emergency Department to Inpatient and Intensive Care Units.

STUDY OBJECTIVE: This study aimed to develop and validate 2 machine learning models that use historical and current-visit patient data from electronic health records to predict the probability of patient admission to either an inpatient unit or ICU at each hour (up to 24 hours) of an emergency department (ED) encounter. The secondary goal was to provide a framework for the operational implementation of these machine learning models. METHODS: Data were curated from 468,167 adult patient encounters in 3 EDs (1 academic and 2 community-based EDs) of a large academic health system from August 1, 2015, to October 31, 2018. The models were validated using encounter data from January 1, 2019, to December 31, 2019. An operational user dashboard was developed, and the models were run on real-time encounter data. RESULTS: For the intermediate admission model, the area under the receiver operating characteristic curve was 0.873 and the area under the precision-recall curve was 0.636. For the ICU admission model, the area under the receiver operating characteristic curve was 0.951 and the area under the precision-recall curve was 0.461. The models had similar performance in both the academic- and community-based settings as well as across the 2019 and real-time encounter data. CONCLUSION: Machine learning models were developed to accurately make predictions regarding the probability of inpatient or ICU admission throughout the entire duration of a patient's encounter in ED and not just at the time of triage. These models remained accurate for a patient cohort beyond the time period of the initial training data and were integrated to run on live electronic health record data, with similar performance.

Investigations of Rhizobium biofilm formation.

The development of nitrogen-fixing nodules of the Rhizobium-legume symbiosis, especially the early stages of root hair deformation and curling, infection thread formation, and nodule initiation, has been well studied from a genetic standpoint. In contrast, the factors important for the colonization of surfaces by rhizobia, including roots-an important prerequisite for nodule formation-have not been as thoroughly investigated. We developed conditions for analyzing the ability of two fast-growing rhizobia, Sinorhizobium meliloti and Rhizobium leguminosarum bv. viciae, to produce biofilms on abiotic surfaces such as glass, plastic microtiter plates, sand and soil as a prelude to characterizing the genes important for aggregation and attachment. Factors involved in adherence to abiotic surfaces are likely to be used in rhizobial attachment to legume root cells. In this report, we show that S. meliloti exopolysaccharide-deficient mutants as well as exopolysaccharide overproducers exhibit reduced biofilm phenotypes that show parallels with their nodulation abilities. We also investigated two flagella-less S. meliloti mutants and found them to have reduced biofilming capabilities. To investigate whether there was a symbiotic phenotype, we tested one of the Fla- mutants on two different S. meliloti hosts, alfalfa and white sweetclover, and found that nodule formation was significantly delayed on the latter.

Positive outcome after intentional overdose of dabigatran.

INTRODUCTION: Dabigatran (Pradaxa™), an orally active direct thrombin inhibitor, was approved by the United States Food and Drug Administration for the prevention of stroke in patients with atrial fibrillation in October 2010. Life-threatening consequences from dabigatran therapy include hemorrhage and bleeding complications, but they typically occur after renal impairment. We describe the first case report of intentional, acute overdose with dabigatran. CASE REPORT: A 57-year-old woman with a medical history of depression and atrial fibrillation presented to the emergency department after ingesting 11.25 g of dabigatran in a suicide attempt. Despite an ecchymosis indicative of prior trauma, there was no evidence of acute bleeding. After receiving gastric lavage and activated charcoal therapy in the emergency department, she was admitted to the ICU. On presentation, dabigatran blood levels measured 970 ng/mL and thrombin clot times measured above the testable limits (>120 s) until 52 h post-arrival. The remainder of her clinical course was uncomplicated, and the patient was transferred to an inpatient psychiatric unit for depression follow-up. DISCUSSION: This case shows the clinical course of a patient with an acute, massive dabigatran overdose with no significant clinical consequences. Currently, there is no ideal method to monitor anticoagulation levels; there is no pharmacologic reversal method, and hemodialysis is an undesirable treatment option.

Intraoperative pulmonary artery stenting: an alternative technique for the management of pulmonary artery stenosis.

BACKGROUND: Pulmonary artery reconstruction remains a challenging clinical problem. We sought to evaluate the use of intraoperative stenting in conjunction with open heart surgical repairs. METHODS: We retrospectively analyzed data from 22 patients between 1998 and 2003 who underwent intraoperative pulmonary artery stenting. RESULTS: The mean age was 9.3 years (range, 9 months to 24 years). The left pulmonary artery was stented in 8 patients, the right in 13, the main in 1, and bilateral in 1. Available echocardiograms in 20 patients revealed a mean preoperative diameter of 7.6 mm (range, 3 to 16.5 mm) and a mean peak gradient of 45.4 mm Hg (range, 20 to 120 mm Hg). Augmentation patch angioplasty of the contralateral pulmonary artery was performed in 11 patients. Thirteen patients had pulmonary valve replacement and 11 had conduit replacement or augmentation. Three patients underwent tricuspid valve repair, 6 had Fontan, and 4 underwent a modified Maze procedure. Follow-up echocardiograms demonstrated a decrease in mean peak gradient to 4.3 mm Hg (range, 0 to 15 mm Hg), a change in mean peak gradient of 41.1 mm Hg (93% reduction). After stenting the mean pulmonary artery diameter increased to 10.9 mm (range, 6 to 17.9 mm). There was no perioperative mortality. At a mean follow-up of 22.8 months (range, 6 to 57 months), there were no surgical or percutaneous reinterventions. CONCLUSIONS: Intraoperative pulmonary artery stenting is a safe and effective technique that can be used as an alternative to patch angioplasty. Close follow-up is needed to examine the long-term outcome of this procedure, with special attention to the growth and development of the stented vessel.