Emergency Department Blood Gas Utilization and Changes in Ventilator Settings

BACKGROUND

Mechanically ventilated patients increasingly spend hours in emergency department beds before ICU admission. This study evaluated the performance of blood gases in mechanically ventilated subjects in the emergency department and subsequent changes to mechanical ventilation settings.

METHODS

This was a multi-center, prospective, observational study of subjects ventilated in the emergency department, conducted at 3 academic emergency departments from July 2011 to March 2013. We measured the rate of arterial blood gas (ABG) and venous blood gas (VBG) analysis, and we assessed the associations between the conditions of hypoxemia, hyperoxia, hypercapnia, or acidemia and changes to mechanical ventilator settings.

RESULTS

Of 292 ventilated subjects, 17.1% did not have a blood gas sent in the emergency department. Ventilator changes were made significantly more frequently for subjects who had an ABG as the initial blood gas sent in the emergency department (odds ratio 2.70, 95% CI 1.46-4.99, P = .002). However, findings of hypoxemia, hyperoxia, hypercapnia, or acidemia were not correlated with ventilator adjustments.

CONCLUSIONS

In this prospective observational study of subjects mechanically ventilated in the emergency department, the majority had a blood gas checked while in the emergency department. While ABGs were associated with having changes made to ventilator settings in the emergency department, clinical findings of hypoxemia, hyperoxia, hypercapnia, and acidemia were not. Inattention to blood gas results may lead to missed opportunities in guiding ventilator changes in the emergency department.

Duration of Mechanical Ventilation in the Emergency Department

INTRODUCTION

Due to hospital crowding, mechanically ventilated patients are increasingly spending hours boarding in emergency departments (ED) before intensive care unit (ICU) admission. This study aims to evaluate the association between time ventilated in the ED and in-hospital mortality, duration of mechanical ventilation, ICU and hospital length of stay (LOS).

METHODS

This was a multi-center, prospective, observational study of patients ventilated in the ED, conducted at three academic Level I Trauma Centers from July 2011 to March 2013. All consecutive adult patients on invasive mechanical ventilation were eligible for enrollment. We performed a Cox regression to assess for a mortality effect for mechanically ventilated patients with each hour of increasing LOS in the ED and multivariable regression analyses to assess for independently significant contributors to in-hospital mortality. Our primary outcome was in-hospital mortality, with secondary outcomes of ventilator days, ICU LOS and hospital LOS. We further commented on use of lung protective ventilation and frequency of ventilator changes made in this cohort.

RESULTS

We enrolled 535 patients, of whom 525 met all inclusion criteria. Altered mental status without respiratory pathology was the most common reason for intubation, followed by trauma and respiratory failure. Using iterated Cox regression, a mortality effect occurred at ED time of mechanical ventilation > 7 hours, and the longer ED stay was also associated with a longer total duration of intubation. However, adjusted multivariable regression analysis demonstrated only older age and admission to the neurosciences ICU as independently associated with increased mortality. Of interest, only 23.8% of patients ventilated in the ED for over seven hours had changes made to their ventilator.

CONCLUSION

In a prospective observational study of patients mechanically ventilated in the ED, there was a significant mortality benefit to expedited transfer of patients into an appropriate ICU setting.

Academic Emergency Medicine Physicians' Knowledge of Mechanical Ventilation

Introduction

Although emergency physicians frequently intubate patients, management of mechanical ventilation has not been emphasized in emergency medicine (EM) education or clinical practice. The objective of this study was to quantify EM attendings’ education, experience, and knowledge regarding mechanical ventilation in the emergency department.

Methods

We developed a survey of academic EM attendings’ educational experiences with ventilators and a knowledge assessment tool with nine clinical questions. EM attendings at key teaching hospitals for seven EM residency training programs in the northeastern United States were invited to participate in this survey study. We performed correlation and regression analyses to evaluate the relationship between attendings’ scores on the assessment instrument and their training, education, and comfort with ventilation.

Results

Of 394 EM attendings surveyed, 211 responded (53.6%). Of respondents, 74.5% reported receiving three or fewer hours of ventilation-related education from EM sources over the past year and 98 (46%) reported receiving between 0–1 hour of education. The overall correct response rate for the assessment tool was 73.4%, with a standard deviation of 19.9. The factors associated with a higher score were completion of an EM residency, prior emphasis on mechanical ventilation during one’s own residency, working in a setting where an emergency physician bears primary responsibility for ventilator management, and level of comfort with managing ventilated patients. Physicians’ comfort was associated with the frequency of ventilator changes and EM management of ventilation, as well as hours of education.

Conclusion

EM attendings report caring for mechanically ventilated patients frequently, but most receive fewer than three educational hours a year on mechanical ventilation, and nearly half receive 0–1 hour. Physicians’ performance on an assessment tool for mechanical ventilation is most strongly correlated with their self-reported comfort with mechanical ventilation.

Initial mechanical ventilator settings and lung protective ventilation in the ED

OBJECTIVE

Mechanical ventilation with low tidal volumes has been shown to improve outcomes for patients both with and without acute respiratory distress syndrome. This study aims to characterize mechanically ventilated patients in the emergency department (ED), describe the initial ED ventilator settings, and assess for associations between lung protective ventilation strategies in the ED and outcomes.

METHODS

This was a multicenter, prospective, observational study of mechanical ventilation at 3 academic EDs. We defined lung protective ventilation as a tidal volume of less than or equal to 8 mL/kg of predicted body weight and compared outcomes for patients ventilated with lung protective vs non-lung protective ventilation, including inhospital mortality, ventilator days, intensive care unit length of stay, and hospital length of stay.

RESULTS

Data from 433 patients were analyzed. Altered mental status without respiratory pathology was the most common reason for intubation, followed by trauma and respiratory failure. Two hundred sixty-one patients (60.3%) received lung protective ventilation, but most patients were ventilated with a low positive end-expiratory pressure, high fraction of inspired oxygen strategy. Patients were ventilated in the ED for a mean of 5 hours and 7 minutes but had few ventilator adjustments. Outcomes were not significantly different between patients receiving lung protective vs non-lung protective ventilation.

CONCLUSIONS

Nearly 40% of ED patients were ventilated with non-lung protective ventilation as well as with low positive end-expiratory pressure and high fraction of inspired oxygen. Despite a mean ED ventilation time of more than 5 hours, few patients had adjustments made to their ventilators.

Psychometric properties of a novel knowledge assessment tool of mechanical ventilation for emergency medicine residents in the northeastern United States

PURPOSE

Prior descriptions of the psychometric properties of validated knowledge assessment tools designed to determine Emergency medicine (EM) residents understanding of physiologic and clinical concepts related to mechanical ventilation are lacking. In this setting, we have performed this study to describe the psychometric and performance properties of a novel knowledge assessment tool that measures EM residents' knowledge of topics in mechanical ventilation.

METHODS

Results from a multicenter, prospective, survey study involving 219 EM residents from 8 academic hospitals in northeastern United States were analyzed to quantify reliability, item difficulty, and item discrimination of each of the 9 questions included in the knowledge assessment tool for 3 weeks, beginning in January 2013.

RESULTS

The response rate for residents completing the knowledge assessment tool was 68.6% (214 out of 312 EM residents). Reliability was assessed by both Cronbach's alpha coefficient (0.6293) and the Spearman-Brown coefficient (0.6437). Item difficulty ranged from 0.39 to 0.96, with a mean item difficulty of 0.75 for all 9 questions. Uncorrected item discrimination values ranged from 0.111 to 0.556. Corrected item-total correlations were determined by removing the question being assessed from analysis, resulting in a range of item discrimination from 0.139 to 0.498.

CONCLUSION

Reliability, item difficulty and item discrimination were within satisfactory ranges in this study, demonstrating acceptable psychometric properties of this knowledge assessment tool. This assessment indicates that this knowledge assessment tool is sufficiently rigorous for use in future research studies or for assessment of EM residents for evaluative purposes.

Pleth variability index and fluid responsiveness of hemodynamically stable patients after cardiothoracic surgery

BACKGROUND

Fluid responsiveness is a measure of preload dependence and is defined as an increase in cardiac output due to volume expansion. Recent publications have suggested that variation in amplitude of the pulse oximetry waveform may be predictive of fluid responsiveness. The pleth variability index (PVI) was developed as a noninvasive bedside measurement of this variation in the pulse oximetry waveform.

OBJECTIVES

To measure the discriminatory value of PVI for predicting fluid responsiveness as measured by pulmonary artery catheter thermodilution in patients after cardiothoracic surgery.

METHODS

A prospective observational study of hemodynamically stable postoperative cardiac surgery patients with pulmonary artery catheters. A fingertip sensor was used to measure PVI. Vital signs, PVI, and cardiac index were measured before, during, and after passive leg raise. Fluid responsiveness was defined by increase in cardiac index of greater than 15% during passive leg raise. The discriminatory value of PVI was assessed by using the Wilcoxon method to measure the area under the receiver operating curve.

RESULTS

In 13 months, 47 patients (24 receiving mechanical ventilation, 23 spontaneously breathing) were enrolled. Fluid responsiveness was noted in 42% of intubated patients and 48% of spontaneously breathing patients. PVI was not adequate to discriminate fluid responsiveness in intubated patients (area under curve, 0.63; P = .16) or spontaneously breathing patients (area under curve, 0.41; P = .75).

CONCLUSIONS

Among postoperative cardiac surgery patients, PVI is not reliable for predicting fluid responsiveness as measured by pulmonary artery catheter thermodilution, regardless of ventilatory status.

Abstract W P185: Clinical Predictors of Stroke, TIA and Mimic among Patients with Transient Neurological Dysfunction Admitted to an Emergency Department Observation Unit

Background: In patients with resolved symptoms, transient ischemic attack (TIA) is distinguished from ischemic stroke by neuroimaging evidence of acute infarction. DW-MRI has been shown to be more sensitive at detecting infarction than CT, but is not uniformly available in the acute setting.

Hypothesis: We sought to identify predictors of stroke diagnosis among a cohort of clinically suspected TIA patients undergoing an accelerated diagnostic protocol in an emergency department observation unit (EDOU).

Methods: We prospectively studied 189 patients treated in the EDOU of a single tertiary care academic medical center. Patients underwent DW-MRI of the brain (unless contraindicated), and bedside neurologist evaluation. A CT scan of the brain was considered optional prior to EDOU admission. We compared the odds of extremity weakness, sensory loss, facial droop, visual disturbance, slurred speech, aphasia, dizziness, and headache between patients with final diagnosis of stroke, TIA and mimic. This study was approved by the hospital IRB.

Results: Thirty-one patients (16%) were diagnosed with an acute ischemic stroke, 85 (45%) TIA, and 73 (39%) mimic. Mean age was 64.8 years (SD = 15.5; range = 30-90). DW-MRI was performed on 92% of patients. A CT scan was also performed in 80% of patients diagnosed with stroke and 0 were diagnostic. Median ABCD2 scores were 4 for stroke and TIA (IQR 3-5) and 3 for mimic (IQR 2-4). Only headache symptoms predicted lower odds of stroke (OR 0.22; 95% CI: 0.05-0.96). Both headache (OR 1.44; 95% CI: 1.03-2.03) and visual disturbance (OR 3.14; 95% CI: 1.49-6.65) increased the odds of mimic diagnosis, but were also present in 13% and 10% of stroke patients respectively. Slurred speech (OR 0.48: 95% CI: 0.25-0.93); aphasia (OR 0.34 95% CI: 0.15-0.76) and facial droop (OR 90.36: 95% CI: 0.14-0.94) significantly predicted lower odds of mimic diagnosis.

Conclusions: In our investigation of patients with transient neurologic dysfunction in an EDOU, stroke diagnosis was common and could not be predicted by clinical variables alone. Early DW-MRI should be considered in all TIA patients, especially those reporting slurred speech, aphasia or facial droop.

Abstract T MP74: The Diagnosis, Etiologic Classification, and Safe Discharge of Ischemic Strokes with Transient Symptoms from an Emergency Department Observation Unit

Background: Patients with a clinical diagnosis of transient ischemic attack (TIA) who have imaging evidence of infarction portend a high risk of short-term recurrent stroke. Emergency Department Observation Units (EDOU) offer an alternative to hospital admission and are becoming increasingly utilized for acute cerebrovascular emergencies.

Hypothesis: We sought to determine whether an EDOU protocol emphasizing etiologic determination and individualized secondary prevention could be a safe alternative to hospital admission for suspected TIA patients with and without brain infarction.

Methods: We prospectively studied 189 patients admitted to the TIA EDOU of a single tertiary care academic medical center. There was no ABCD2 cutoff for eligibility and exclusion criteria included persistent deficits or another diagnosis warranting hospitalization. Patients underwent DW-MRI/MRA of the head and neck unless contraindicated, transthoracic echocardiogram and bedside neurologist evaluation. Etiologic subtyping was determined using the Causative Classification System (CCS). 30-day follow-up was performed on all patients by telephone and/or review of medical records. This study was approved by the hospital IRB.

Results: Acute ischemic stroke was diagnosed in 31 (16%) of patients, including 30 with DWI lesions and 1 in whom MRI was contraindicated, but had clinical worsening while in the EDOU. An evident or probable etiology was determined by CCS subtyping in 38% of strokes and 32% of TIAs (17% atherosclerosis, 12% cardioembolism, 5% small vessel). Of the total cohort, 84% were discharged from the EDOU including 16 (52%) with confirmed stroke. Median LOS was 22 hours (IQR: 17-25). At 30 days, one patient was found to have a small recurrent stroke (0.7%). There was 1 non-stroke related death. Twenty (11%) overall returned to the ED, the vast majority (70%) from the non-stroke cohort.

Conclusions: Not all ischemic stroke patients require hospitalization. An EDOU is a safe and effective alternative for the complete diagnostic evaluation and management of patients with transient neurologic symptoms. Further study of cost and quality effectiveness in warranted.

Emergency medicine residents' knowledge of mechanical ventilation

BACKGROUND

Although Emergency physicians frequently intubate patients, management of mechanical ventilation has not been emphasized in emergency medicine (EM) residency curricula.

OBJECTIVES

The objective of this study was to quantify EM residents' education, experience, and knowledge regarding mechanical ventilation.

METHODS

We developed a survey of residents' educational experiences with ventilators and an assessment tool with nine clinical questions. Correlation and regression analyses were performed to evaluate the relationship between residents' scores on the assessment instrument and their training, education, and comfort with ventilation.

RESULTS

Of 312 EM residents, 218 responded (69.9%). The overall correct response rate for the assessment tool was 73.3%, standard deviation (SD) ± 22.3. Seventy-seven percent (n = 167) of respondents reported ≤ 3 h of mechanical ventilation education in their residency curricula over the past year. Residents reported frequently caring for ventilated patients in the ED, as 64% (n = 139) recalled caring for ≥ 4 ventilated patients per month. Fifty-three percent (n = 116) of residents endorsed feeling comfortable caring for mechanically ventilated ED patients. In multiregression analysis, the only significant predictor of total test score was residents' comfort with caring for mechanically ventilated patients (F = 10.963, p = 0.001).

CONCLUSIONS

EM residents report caring for mechanically ventilated patients frequently, but receive little education on mechanical ventilation. Furthermore, as residents' performance on the assessment tool is only correlated with their self-reported comfort with caring for ventilated patients, these results demonstrate an opportunity for increased educational focus on mechanical ventilation management in EM residency training.

Gender differences in neurologic emergencies part I: a consensus summary and research agenda on cerebrovascular disease

Cerebrovascular neurologic emergencies including ischemic and hemorrhagic stroke, subarachnoid hemorrhage (SAH), and migraine are leading causes of death and disability that are frequently diagnosed and treated in the emergency department (ED). Although sex and gender differences in neurologic emergencies are beginning to become clearer, there are many unanswered questions about how emergency physicians should incorporate sex and gender into their research initiatives, patient evaluations, and overall management plans for these conditions. After evaluating the existing gaps in the literature, a core group of ED researchers developed a draft of future research priorities. Participants in the 2014 Academic Emergency Medicine consensus conference neurologic emergencies working group then discussed and approved the recommended research agenda using a standardized nominal group technique. Recommendations for future research on the role of sex and gender in the diagnosis, treatment, and outcomes pertinent to ED providers are described for each of three diagnoses: stroke, SAH, and migraine. Recommended future research also includes investigation of the biologic and pathophysiologic differences between men and women with neurologic emergencies as they pertain to ED diagnoses and treatments.

Lack of gender disparities in emergency department triage of acute stroke patients

Introduction

Previous literature has shown gender disparities in the care of acute ischemic stroke. Compared to men, women wait longer for brain imaging and are less likely to receive intravenous (IV) tissue plasminogen activator (tPA). Emergency department (ED) triage is an important step in the rapid assessment of stroke patients and is a possible contributor to disparities. It is unknown whether gender differences exist in ED triage of acute stroke patients. Our primary objective was to determine whether gender disparities exist in the triage of acute stroke patients as defined by Emergency Severity Index (ESI) levels and use of ED critical care beds.

Methods

This was a retrospective, observational study of both ischemic and hemorrhagic stroke patients age ≥18 years presenting to a large, urban, academic ED within six hours of symptom onset between January 2010, and December 2012. Primary outcomes were triage to a non-critical ED bed and Emergency Severity Index (ESI) level. Primary outcome data were extracted from electronic medical records by a blinded data manager; secondary outcome data and covariates were abstracted by trained research assistants. We performed bivariate and multivariate analyses. Logistic regression was performed using age, race, insurance status, mode of and time to arrival, National Institutes of Health Stroke Scale, and presence of atypical symptoms as covariates.

Results

There were 537 patients included in this study. Women were older (75.6 vs. 69.5, p<0.001), and more women had a history of atrial fibrillation (39.8% vs. 25.3%, p<0.001). Compared to 9.5% of men, 10.3% of women were triaged to a non-critical care ED bed (p=0.77); 92.1% of women were triaged as ESI 1 or 2 vs. 93.6% of men (p=0.53). After adjustment, gender was not associated with triage location or ESI level, though atypical symptoms were associated with higher odds of being triaged to a non-critical care bed (aOR 1.98, 95%CI [1.03 – 3.81]) and 3.04 times higher odds of being triaged as ESI 3 vs. ESI 1 or 2 (95% CI [1.36 – 6.82]).

Conclusion

In a large, urban, academic ED at a primary stroke center, there were no gender differences in triage to critical care beds or ESI levels among acute stroke patients arriving within six hours of symptom onset. These findings suggest that ED triage protocols for stroke patients may be effective in minimizing gender disparities in care.

Abstract T P183: Transient Neurological Symptoms with MRI Evidence of Infarction in an Emergency Department Observation Unit

Background: Transient symptoms with infarction (TSI) is distinguished from transient ischemic attack (TIA) by the presence of diffusion-weighted imaging (DWI) lesions on MRI. We report a series of patients with DWI lesions identified in an emergency department observation unit (EDOU) TIA protocol.

Methods: Patients were treated in the ED of a tertiary care center (annual census=100,000 visits). In the first 3 months of operation of an EDOU for TIA, 50 patients met inclusion criteria of sudden/transient neurological deficit in the absence of known metabolic or cardiac source. Patients were excluded from the study if they demonstrated: persistent or stuttering deficits; fever >100.4F, heart rate <60 or >100 bpm; SBP >180 or <100mmHg; pulse ox <93%; or positive CT. Evaluation included: laboratory analyses; CT/MRI/MRA of the head/neck; EKG; echocardiogram, and neurology consult. MR data (ED 1.5T MR unit) include: DWI, ADC, GRE, and T2FLAIR images. Patients with TSI, were identified by the presence of 1 or more DWI bright and ADC dark lesions on ED MRI. All patients completed CT and MR imaging within < 24 hours of admission. This study was approved by the hospital IRB.

Results: TSI was identified in 10/50 EDOU patients (Mdn age 72.5, ABCD2 score 5). Two patients demonstrated infarcts in multiple vascular territories. Table 1 illustrates patient demographics, comorbidities, presentations, and outcome. Fifty percent of TSI patients were admitted from EDOU, 3 patients returned to the ED within 30 days , and no 30 day distinct recurrent events, such as recurrent stroke were identified.

Conclusions: EDOU for TIA including DW-MRI resulted in TSI diagnosis in 20% of patients. In the TSI cases identified, infarct locations were heterogeneous; vascular and/or cardiac etiology must be considered. Longitudinal analysis is recommended to further assess the relationship between presentation, TSI risk, risk of recurrent stroke, and need for hospital admission.

The role of lactate clearance in the resuscitation bundle

The sepsis resuscitation bundle is the result of an effort on behalf of the Surviving Sepsis Campaign and the Institute for Healthcare Improvement to translate individual guideline recommendations into standardized, achievable goals for physicians caring for the critically ill patient. Implementation of this bundle is associated with decreased mortality. Many of the bundle items reflect components of therapy shown to improve mortality in the seminal early goal-directed therapy trial for severe sepsis and septic shock, including an initial lactate measurement. Elevations in serum lactate are associated with increased mortality, and may result from either increased lactate production or impaired lactate clearance. Lactate clearance may be an important addition to the monitoring and management bundles of patients with severe sepsis and septic shock, However, specific mechanisms of lactate clearance, the relation of lactate clearance to traditional hemodynamic parameters, and the importance of lactate clearance as a therapeutic target or monitoring tool remain unclear.

Inadequacy of temperature and white blood cell count in predicting bacteremia in patients with suspected infection

BACKGROUND

Early treatment of sepsis in Emergency Department (ED) patients has lead to improved outcomes, making early identification of the disease essential. The presence of systemic inflammatory response criteria aids in recognition of infection, although the reliability of these markers is variable.

STUDY OBJECTIVE

This study aims to quantify the ability of abnormal temperature, white blood cell (WBC) count, and bandemia to identify bacteremia in ED patients with suspected infection.

METHODS

This was a post hoc analysis of data collected for a prospective, observational, cohort study. Consecutive adult (age ≥ 18 years) patients who presented to the ED of a tertiary care center between February 1, 2000 and February 1, 2001 and had blood cultures obtained in the ED or within 3 h of admission were enrolled. Patients with bacteremia were identified and charts were reviewed for presence of normal temperature (36.1-38°C/97-100.4°F), normal WBC (4-12 K/μL), and presence of bandemia (> 5% of WBC differential).

RESULTS

There were 3563 patients enrolled; 289 patients (8.1%) had positive blood cultures. Among patients with positive blood cultures, 33% had a normal body temperature and 52% had a normal WBC count. Bandemia was present in 80% of culture-positive patients with a normal temperature and 79% of culture-positive patients with a normal WBC count. Fifty-two (17.4%) patients with positive blood cultures had neither an abnormal temperature nor an abnormal WBC.

CONCLUSION

A significant percentage of ED patients with blood culture-proven bacteremia have a normal temperature and WBC count upon presentation. Bandemia may be a useful clue for identifying occult bacteremia.