Evidence Based Scarce Resource Allocation During the COVID-19 Pandemic: A Case Study of Bamlanivimab Administration in the Emergency Department.

Background: Patients presenting for emergency department (ED) evaluation may be appropriate for treatment with monoclonal antibodies for mild to moderate COVID-19. While many sites have implemented infusion centers for these agents, EDs will continue to evaluate these patients where appropriate identification and efficient infusion of eligible patients is critical. Objectives: Patients receiving bamlanivimab in the EDs of an academic medical center are described. The primary objective was to describe operational metrics and secondary objectives reported clinical outcomes. Methods: Patients receiving bamlanivimab and discharged from the ED were included from November 16, 2020 to January 16, 2021 in the retrospective, observational cohort. Primary outcome was adherence to institutional criteria. Secondary outcomes included ED visit metrics, clinical characteristics, and return visits within 30 days. Risk factors for return visits were assessed with regression. Results: One hundred nineteen patients were included. Most (71%) were diagnosed with COVID-19 during the ED visit and median symptom duration was 3(IQR 2-5) days. Median number of risk factors for progression to severe disease was 2 (IQR 1-2). Thirty percent had a documented abnormal chest x-ray. Institutional criteria adherence was 99.2%. Median time from ED room to bamlanivimab was 4 (IQR 3.1-5.2) hours. Thirty patients had return visit within 30 days; 19 were COVID-19 related. Two multivariable regression models were analyzed for COVID-19 related return visit. Characteristics on ED presentation were considered in Model I: male gender (OR 3.01[0.97-9.31]), age (per 10 years) (OR 1.49[1.05-2.12]), African-American race (OR 3.46[1.09-11.06]), and symptom duration (per day) (OR 1.34[1.05-1.73]). Model II included labs and imaging acquired in ED. In Model II, age (per 10 years) (OR 1.52[1.07-2.16]) and abnormal CXR (OR 5.74[1.95-16.9]) were associated with COVID-19 related return visits. Conclusions: Administration of bamlanivimab to ED patients can be done efficiently, with the potential to reduce COVID-19 related return visits. Age and abnormal imaging were independent predictors of COVID-19 return visits.

Arranging Hospice Care from the Emergency Department: A Single Center Retrospective Study.

BACKGROUND: Arranging hospice services from the Emergency Department (ED) can be difficult due to physician discomfort, time constraints, and the intensity of care coordination needed. We report patient and visit characteristics associated with successful transition from the ED directly to hospice. METHODS: Setting: Academic ED with 82,000 annual visits. POPULATION: ED patients with a referral to hospice order placed during the ED visit from January 2014-December 2018. Charts were abstracted by trained, non-blinded personnel. Primary goal was to evaluate patient and visit factors associated with requiring admission for hospice transition. RESULTS: Electronic Health Record inquiry yielded 113 patients, 93 of which met inclusion criteria. Patients were aged 65.8 years (range 32-92), 54% were female, and 78% were white, non-hispanic. The majority had cancer (78%, n = d72) and were on public insurance (60%, n = 56). Half (55%, n = 51) were full code upon arrival. Average ED length of stay was 4.6 ± 2.6 hours. Discharge from the ED to hospice was successful for 38% (n = 35), a few (n = 5) were dispositioned to an ED observation unit, and 57% (n = 53) were admitted. Only 10 (11%) required an inpatient length of stay longer than an observation visit (2 days). Case management and social work team arranged for transportation (54.8%, n = 51), hospital beds (16.1%, n = 16), respiratory equipment (18.3%, n = 17), facility placement (33.3%, n = 31), and home health aides (29.0%, n = 27). CONCLUSION: Transitioning patients to hospice care from the ED is possible within a typical ED length of stay with assistance from a case manager/social work team.

Predicting in vivo MRI Gradient-Field Induced Voltage Levels on Implanted Deep Brain Stimulation Systems Using Neural Networks.

INTRODUCTION: MRI gradient-fields may induce extrinsic voltage between electrodes and conductive neurostimulator enclosure of implanted deep brain stimulation (DBS) systems, and may cause unintended stimulation and/or malfunction. Electromagnetic (EM) simulations using detailed anatomical human models, therapy implant trajectories, and gradient coil models can be used to calculate clinically relevant induced voltage levels. Incorporating additional anatomical human models into the EM simulation library can help to achieve more clinically relevant and accurate induced voltage levels, however, adding new anatomical human models and developing implant trajectories is time-consuming, expensive and not always feasible. METHODS: MRI gradient-field induced voltage levels are simulated in six adult human anatomical models, along clinically relevant DBS implant trajectories to generate the dataset. Predictive artificial neural network (ANN) regression models are trained on the simulated dataset. Leave-one-out cross validation is performed to assess the performance of ANN regressors and quantify model prediction errors. RESULTS: More than 180,000 unique gradient-induced voltage levels are simulated. ANN algorithm with two fully connected layers is selected due to its superior generalizability compared to support vector machine and tree-based algorithms in this particular application. The ANN regression model is capable of producing thousands of gradient-induced voltage predictions in less than a second with mean-squared-error less than 200 mV. CONCLUSION: We have integrated machine learning (ML) with computational modeling and simulations and developed an accurate predictive model to determine MRI gradient-field induced voltage levels on implanted DBS systems.

Beyond Bones: Assessing Whether Ultrasound-Aided Instruction and Practice Improve Unassisted Soft Tissue Palpation Skills of First-Year Medical Students.

OBJECTIVES: Our purpose was to determine whether ultrasound (US)-aided instruction and practice on musculoskeletal anatomy would improve first-year medical students' ability to locate and identify specific soft tissue structures by unaided palpation in the upper and lower extremities of healthy human models. METHODS: This study was a randomized crossover design with 49 first-year medical students randomly assigned to 1 of 2 groups. Each group was provided expert instruction and hands-on practice using US to scan and study soft tissue structures. During session 1, group A learned the anatomy of the upper extremities, whereas group B learned the lower. Students were then tested on their proficiency in locating 4 soft tissue structures (2 upper and 2 lower extremities) through palpation of a human model. During session 2, group A learned lower extremities, and group B learned upper. At the end of session 2, students repeated the assessment. RESULTS: After the first instructional session, neither group performed significantly better on identifying and locating the soft tissue landmarks they learned aided by US. After the second instructional session, however, scores for both groups increased approximately 20 percentage points, indicating that both groups performed significantly better on palpating and identifying both the upper and lower extremity soft tissue landmarks (Cohen d = 0.89 and 0.82, respectively). CONCLUSIONS: Time and practice viewing soft tissue structures with US assistance seems to have a "palpation-with-eyes" effect that improves students' abilities to correctly locate, palpate, and identify limb-specific soft tissue structures once the US assistance is removed.

Does Ultrasound-Enhanced Instruction of Musculoskeletal Anatomy Improve Physical Examination Skills of First-Year Medical Students?

OBJECTIVES: Ultrasound imaging is commonly used to teach basic anatomy to medical students. The purpose of this study was to determine whether learning musculoskeletal anatomy with ultrasound improved performance on medical students' musculoskeletal physical examination skills. METHODS: Twenty-seven first-year medical students were randomly assigned to 1 of 2 instructional groups: either shoulder or knee. Both groups received a lecture followed by hands-on ultrasound scanning on live human models of the assigned joint. After instruction, students were assessed on their ability to accurately palpate 4 anatomic landmarks: the acromioclavicular joint, the proximal long-head biceps tendon, and the medial and lateral joint lines of the knee. Performance scores were based on both accuracy and time. A total physical examination performance score was derived for each joint. Scores for instructional groups were compared by a 2-way analysis of variance with 1 repeated measure. Significant findings were further analyzed with post hoc tests. RESULTS: All students performed significantly better on the knee examination, irrespective of instructional group (F = 14.9; df = 1.25; P = .001). Moreover, the shoulder instruction group performed significantly better than the knee group on the overall assessment (t = -3.0; df = 25; P < .01). Post hoc analyses revealed that differences in group performance were due to the shoulder group's higher scores on palpation of the biceps tendon (t = -2.8; df = 25; P = .01), a soft tissue landmark. Both groups performed similarly on palpation of all other anatomic structures. CONCLUSIONS: The use of ultrasound appears to provide an educational advantage when learning musculoskeletal physical examination of soft tissue landmarks.

Impact of checklists on peri-intubation care in ED trauma patients.

BACKGROUND: Checklists have successfully been used in intensive care units (ICUs) to improve metrics of critical care. Proper peri-intubation care including use of appropriate induction agents and postintubation sedation is crucial when performing endotracheal intubation (ETI) on critically ill patients, especially in the emergency department (ED). We sought to evaluate the impact of checklists on peri-intubation care in ED trauma patients. METHODS: We performed a retrospective review of all trauma patients intubated in the ED of an urban, level 1 academic center from November 2010 to October 2012. As part of a quality improvement project, a peri-intubation checklist was instituted on November 1, 2011 to guide peri-intubation care. Using a predesign and postdesign, we compared peri-intubation parameters using parametric and nonparametric statistics when appropriate to evaluate the impact of a checklist on peri-intubation care. We also evaluated outcome measures including mortality and lengths of stay. RESULTS: During the 2-year study period, 187 trauma patients underwent ETI in the ED, 90 prechecklist and 97 postchecklist. Rapid sequence intubation (RSI) use was greater with the checklist than without (90.7% vs 75.6%, P=.005). No difference was found between the number of ETI attempts per patient, hemodynamic parameters (heart rate, blood pressure, and oxygen saturation), postintubation anxiolysis, median number of ventilator days, length of ED stay, length of ICU stay, or mortality. CONCLUSION: Peri-intubation checklists result in higher rates of RSI in ED trauma patients but do not alter other measured metrics of peri-intubation care.

High-resolution magnetic resonance flow imaging in a model of porous bone-implant interface.

PURPOSE: To evaluate the application of high-resolution MRI methodology for characterizing the fluid velocity field and evaluate fluid shear field within a simplified in vitro model of a bone-implant interface. MATERIALS AND METHODS: The study used a specific micromotion canine bone implant that has been used for over a decade in the experimental evaluation of anatomical, biomaterial, mechanical and surgical factors influencing the quality of the implant interface. To allow its implementation in an MR coil, a nonmagnetic model of the micromotion implant was fabricated. The model consisted of a cylinder of polymethylmethacrylate (PMMA) representing the implant, located within an annular controlled gap into a block of coralline-derived bulk porous hydroxyapatite (HA; Interpore Cross International, Irvine, CA, USA). The assembly was potted in a polycarbonate shell and connected to a gravity-feed flow system consisting of a water fluid reservoir and peristaltic pump. Cross-sectional fluid velocity images through the principal axis of the implant were generated using a phase-encoding MR imaging technique; axial fluid flow was derived, and fluid shear was evaluated using a Newtonian fluid model. RESULTS: Due to the nonuniform gap of the actual experimental construct, a highly nonuniform flow through the annular gap and a secondary flow through the porous HA block were observed. Axial velocity magnitudes in the range 0.04 to 14 mm/s were measured, and the flow velocities within the annular gap and the surrounding bone differed by nearly two orders of magnitude. Image analysis showed that 95% of total flow passed through the annular gap and 5% was transported through the porous HA block. Fluid shear was computed within the porous structure and the annular gap, and they differed by one order of magnitude. CONCLUSION: We demonstrated that high-resolution MR flow imaging has the resolution to measure fluid transport processes noninvasively through a nonmagnetic model bone implant. Gap fluid flow and fluid flow into the permeable skeleton (HA block) were quantified, and these data allowed the noninvasive determination of fluid shear. These promising results are encouraging for applications in biological tissue, artificial bone substitutes, tissue engineering and clinically relevant studies concerning implant fixation.

Ringer's ethyl pyruvate in hemorrhagic shock and resuscitation does not improve early hemodynamics or tissue energetics.

Reactive oxygen species (ROS) have been implicated in the pathogenesis of hemorrhagic shock. Ethyl pyruvate, a derivative of pyruvate and a proposed oxygen radical scavenger, is attractive as a possible resuscitation fluid. We investigated whether resuscitation with lactated Ringer's (LR) containing ethyl pyruvate (REP) had any hemodynamic or tissue energetic benefits compared with LR alone for hemorrhagic shock. Hemorrhagic shock was induced in splenectomized pigs via inferior vena cava cannula. After 90 min of shock, animals were resuscitated in a stepwise fashion with LR or REP (30 mg/kg/dose, given as 1.5 mg/mL in LR) at 20 cc/kg/step for four steps. Data collected during this experiment included physiologic and hemodynamic parameters, near-infrared reflectance spectroscopy measurements of tissue hemoglobin oxygen (StO(2)) of the stomach, liver, and hind limb, and nuclear magnetic resonance phosphorus spectra of the liver and hind limb at each time point. In both resuscitative groups, heart rate, and lactate and pyruvate values increased during shock and began to drop toward baseline values during resuscitation. Mean arterial pressure, oxygen delivery, and oxygen consumption decreased during shock and increased toward baseline levels during the resuscitative process. There were no significant changes in physiologic parameters between the LR- and REP-resuscitated animals. There was a significantly lower stomach StO(2) and hind limb cellular cytoplasmic pH during later resuscitative endpoints in REP-resuscitated animals. The clinical significance of these findings are unclear. There is no short-term hemodynamic or tissue energetic advantage to using REP as a resuscitation fluid when compared with LR. Long-term outcome studies are needed to further evaluate any potential benefits of use of REP in hemorrhagic shock.

Phosphomonoesters predict early mortality in porcine hemorrhagic shock.

BACKGROUND: Hemodynamic, laboratory, and tissue energetics were measured in a porcine model of hemorrhagic shock to evaluate variables as predictors of early mortality from shock. We hypothesized that elevated phosphomonoesters would predict early mortality in hemorrhagic shock. METHODS: Pigs (n = 36) were subjected to 35% hemorrhage for 90 minutes in a 1.5-T nuclear magnetic resonance (NMR) magnet. Measurements included base deficit (BD); lactate; oxygen consumption/delivery; near-infrared spectroscopy of liver, stomach, and skeletal muscle tissue oxyhemoglobin saturation; and NMR spectroscopic measurements of high-energy phosphates of liver and skeletal muscle. Variables were compared between nonsurvivors and survivors to resuscitation after 90-minute measurements. RESULTS: Ninety-minute mortality was 25%. Muscle phosphomonoesters (PMEs) and oxygen consumption differed significantly between survivors and nonsurvivors at baseline. Regression analysis identified baseline muscle PME levels, baseline BD, and 30-minute BD as early predictors of mortality before resuscitation (r2 = 0.304). CONCLUSION: Baseline elevation in muscle PME levels predicts mortality in an animal model of severe hemorrhagic shock.

Tissue energetics as measured by nuclear magnetic resonance spectroscopy during hemorrhagic shock.

The defect in energy production in an organism during shock states may be related to the impairment of mitochondrial respiration early in shock. The aim of this study was to investigate the timing and degree of cellular energetic changes during hemorrhagic shock in real time. Instrumented, splenectomized swine were randomized to undergo hemorrhagic shock, induced by a 35% blood volume bleed, for 90 min with (n = 10) or without (n = 9) subsequent resuscitation. Resuscitated animals received shed blood in two increments followed by two normal saline boluses (20 mL/kg/bolus). Throughout experimentation, tissue phosphoenergetics of liver and skeletal muscle were monitored using 31P nuclear magnetic resonance (NMR) spectroscopy via NMR coils on the liver and hindlimb. Near-infrared spectroscopy probes were used to measure liver, stomach, and skeletal muscle oxyhemoglobin saturation (StO2). Hemorrhagic shock induced an increase in phosphomonoesters in skeletal muscle (baseline: 7.09%, 90 min: 9.94% (P < 0.05); expressed as percent total phosphorus). This increase resolved in animals receiving resuscitation (n = 10) but remained elevated in those in unresuscitated shock (n = 9). Inorganic phosphate levels increased and betaATP levels decreased significantly in the liver of animals in shock as compared with baseline. StO2 in skeletal muscle, stomach, and liver correlated with whole organism oxygen delivery (r2 = 0.356, 0.368, and 0.432, respectively). We conclude that hemorrhagic shock induces early elevation of phosphomonoesters in skeletal muscle, which correlates with the severity of shock. This implies an early transition to anaerobic glycolysis during hemorrhagic shock, which may be indicative of early mitochondrial dysfunction.