A Pilot Machine Learning Study Using Trauma Admission Data to Identify Risk for High Length of Stay.

INTRODUCTION: Trauma patients have diverse resource needs due to variable mechanisms and injury patterns. The aim of this study was to build a tool that uses only data available at time of admission to predict prolonged hospital length of stay (LOS). METHODS: Data was collected from the trauma registry at an urban level one adult trauma center and included patients from 1/1/2014 to 3/31/2019. Trauma patients with one or fewer days LOS were excluded. Single layer and deep artificial neural networks were trained to identify patients in the top quartile of LOS and optimized on area under the receiver operator characteristic curve (AUROC). The predictive performance of the model was assessed on a separate test set using binary classification measures of accuracy, precision, and error. RESULTS: 2953 admitted trauma patients with more than one-day LOS were included in this study. They were 70% male, 60% white, and averaged 47 years-old (SD: 21). 28% were penetrating trauma. Median length of stay was 5 days (IQR 3-9). For prediction of prolonged LOS, the deep neural network achieved an AUROC of 0.80 (95% CI: 0.786-0.814) specificity was 0.95, sensitivity was 0.32, with an overall accuracy of 0.79. CONCLUSION: Machine learning can predict, with excellent specificity, trauma patients who will have prolonged length of stay with only physiologic and demographic data available at the time of admission. These patients may benefit from additional resources with respect to disposition planning at the time of admission.

Prolonged wait time is associated with increased mortality for Chilean waiting list patients with non-prioritized conditions.

BACKGROUND: Most data on mortality and prognostic factors of universal healthcare waiting lists come from North America, Australasia, and Europe, with little information from South America. We aimed to determine the relationship between medical center-specific waiting time and waiting list mortality in Chile. METHOD: Using data from all new patients listed in medical specialist waitlists for non-prioritized health problems from 2008 to 2015 in three geographically distant regions of Chile, we constructed hierarchical multivariate survival models to predict mortality risk at two years after registration for each medical center. Kendall rank correlation analysis was used to measure the association between medical center-specific mortality hazard ratio and waiting times. RESULT: There were 987,497 patients waiting for care at 77 medical centers, including 33,546 (3.40%) who died within two years after registration. Male gender (hazard ratio [HR] = 1.17, 95% confidence interval [CI] 1.1-1.24), older age (HR = 2.88, 95% CI 2.72-3.05), urban residence (HR = 1.19, 95% CI 1.09-1.31), tertiary care (HR = 2.2, 95% CI 2.14-2.26), oncology (HR = 3.57, 95% CI 3.4-3.76), and hematology (HR = 1.6, 95% CI 1.49-1.73) were associated with higher risk of mortality at each medical center with large region-to-region variations. There was a statistically significant association between waiting time variability and death (Z = 2.16, P = 0.0308). CONCLUSION: Patient wait time for non-prioritized health conditions was associated with increased mortality in Chilean hospitals.

Using the Electronic Medical Record to Reduce Unnecessary Ordering of Coagulation Studies for Patients with Chest Pain.

INTRODUCTION: Our goal was to reduce ordering of coagulation studies in the emergency department (ED) that have no added value for patients presenting with chest pain. We hypothesized this could be achieved via implementation of a stopgap measure in the electronic medical record (EMR). METHODS: We used a pre and post quasi-experimental study design to evaluate the impact of an EMR-based intervention on coagulation study ordering for patients with chest pain. A simple interactive prompt was incorporated into the EMR of our ED that required clinicians to indicate whether patients were on anticoagulation therapy prior to completion of orders for coagulation studies. Coagulation order frequency was measured via detailed review of randomly sampled encounters during two-month periods before and after intervention. We classified existing orders as clinically indicated or non-value added. Order frequencies were calculated as percentages, and we assessed differences between groups by chi-square analysis. RESULTS: Pre-intervention, 73.8% (76/103) of patients with chest pain had coagulation studies ordered, of which 67.1% (51/76) were non-value added. Post-intervention, 38.5% (40/104) of patients with chest pain had coagulation studies ordered, of which 60% (24/40) were non-value added. There was an absolute reduction of 35.3% (95% confidence interval [CI]: 22.7%, 48.0%) in the total ordering of coagulation studies and 26.4% (95% CI: 13.8%, 39.0%) in non-value added order placement. CONCLUSION: Simple EMR-based interactive prompts can serve as effective deterrents to indiscriminate ordering of diagnostic studies.

Bedside sonography for the diagnosis of esophageal food impaction.

BACKGROUND: Esophageal foreign body (EFB) and impaction are common gastrointestinal emergencies. Detection with standard imaging can be challenging. Computed tomography is a commonly used non-invasive imaging modality, but is not 100% sensitive and not always feasible. Sensitivity of plain film x-ray varies widely and the addition of a barium swallow can obscure evaluation by subsequent esophagogastroduodenoscopy (EGD). Use of emergency ultrasound (EUS) for detection of EFB in adults has not been previously studied. OBJECTIVE: To evaluate the role of EUS in detection of EFB and to characterize sonographic findings. METHODS: A case control series of five patients with clinical suspicion of EFB underwent EUS, and findings were compared to five healthy controls. Patients were evaluated for persistent air-fluid levels after swallowing, esophageal dilatation, and visualization of EFB. RESULTS: All patients with suspected EFB had esophageal dilatation (17.5mm vs 9.3mm in healthy controls; p=0.0011) and persistent air-fluid levels after swallowing. EFB was visualized on EUS in 60% of patients. All patients had EFB confirmed on EGD except one, who vomited a significant food bolus during EUS and prior to EGD. CONCLUSION: In patients with suspected EFB, point-of-care ultrasound may identify those with impaction. Suggestive findings include cervical esophageal dilatation and persistent intraluminal air-fluid levels after swallowing. EUS is a rapid, convenient test with the potential to expedite definitive management while decreasing cost and radiation exposure in this patient population.

Sphingosine 1-phosphate stimulates smooth muscle cell differentiation and proliferation by activating separate serum response factor co-factors.

Sphingosine 1-phosphate (S1P) is a lipid agonist that regulates smooth muscle cell (SMC) and endothelial cell functions by activating several members of the S1P subfamily of G-protein-coupled Edg receptors. We have shown previously that SMC differentiation is regulated by RhoA-dependent activation of serum response factor (SRF). Because S1P is a strong activator of RhoA, we hypothesized that S1P would stimulate SMC differentiation. Treatment of primary rat aortic SMC cells with S1P activated RhoA as measured by precipitation with a glutathione S-transferase-rhotekin fusion protein. In SMC and 10T1/2 cells, S1P treatment up-regulated the activities of several transiently transfected SMC-specific promoters, and these effects were inhibited by the Rho-kinase inhibitor, Y-27632. S1P also increased smooth muscle alpha-actin protein levels in SMC but had no effect on SRF binding to the smooth muscle alpha-actin CArG B element. Quantitative reverse transcriptase-PCR showed that S1P treatment of SMC or 10T1/2 cells did not increase the mRNA level of either of the recently identified SRF co-factors, myocardin or myocardin-related transcription factor-A (MRTF-A). MRTF-A protein was expressed highly in SMC and 10T1/2 cultures, and importantly the effects of S1P were inhibited by a dominant negative form of MRTF-A indicating that S1P may regulate the transcriptional activity of MRTF-A. Indeed, S1P treatment increased the nuclear localization of FLAG-MRTF-A, and the effect of MRTF-A overexpression on smooth muscle alpha-actin promoter activity was inhibited by dominant negative RhoA. S1P also stimulated SMC growth by activating the early growth response gene, c-fos. This effect was not attenuated by Y-27632 but could be inhibited by the MEK inhibitor, UO126. S1P enhanced SMC growth through ERK-mediated phosphorylation of the SRF co-factor, Elk-1, as measured by gel shift and Elk-1 activation assays. Taken together these results demonstrate that S1P activates multiple signaling pathways in SMC and regulates proliferation by ERK-dependent activation of Elk-1 and differentiation by RhoA-dependent activation of MRTF-A.