Performance of Prehospital Use of Chest Pain Risk Stratification Tools: The RESCUE Study.

BACKGROUND: Emergency medical services (EMS) assesses millions of patients with chest pain each year. However, tools validated to risk stratify patients for acute coronary syndrome (ACS) and pulmonary embolism (PE) have not been translated to the prehospital setting. The objective of this study is to assess the prehospital performance of risk stratification scores for 30-day major adverse cardiac events (MACE) and PE. METHODS: A prospective observational cohort study of patients ≥21 years of age with acute chest pain who were transported by EMS in two North Carolina (NC) counties was conducted from 18 April 2018-2 January 2019. In this convenience sample, paramedics completed HEAR (history, electrocardiogram, age, risk factor), ED Assessment of Chest Pain Score (EDACS), Revised Geneva Score (RGS), and pulmonary embolism rule-out criteria (PERC) assessments on each patient. MACE (all-cause death, myocardial infarction, and revascularization) and PE at 30 days were determined by hospital records and NC Death Index. The positive (+LR) and negative likelihood ratios (-LR) of the risk scores for 30-day MACE and PE were calculated. RESULTS: During the study period, 82.1% (687/837) patients had all four risk score assessments. The cohort was 51.1% (351/687) female, 49.5% (340/687) African American, and had a mean age of 55.0 years (SD 16.0). At 30 days, MACE occurred in 7.4% (51/687), PE occurred in 0.9% (6/687), and the combined outcome occurred in 8.2% (56/687). The HEAR score had a - LR of 0.46 (95% CI 0.27-0.78) and + LR of 1.48 (95% CI 1.26-1.74) for 30-day MACE. EDACS had a - LR of 0.61 (95% CI 0.46-0.81) and + LR of 2.53 (95% CI 1.86-3.46) for 30-day MACE. The PERC score had a - LR of 0 (95% CI 0.0-1.4) and a + LR of 1.38 (95% CI 1.32-1.45) for 30-day PE. The RGS score had a - LR of 0 (95% CI 0.0-0.65) and a + LR of 2.36 (95% CI 2.16-2.57) for 30-day PE. The combination of a low-risk HEAR score and negative PERC evaluation had a - LR of 0.25 (95% CI 0.08-0.76) and a + LR of 1.21 (95% CI 1.21-1.30) for 30-day MACE or PE. CONCLUSION: The combination of a paramedic-obtained HEAR score and PERC evaluation performed best to exclude 30-day MACE and PE but was not sufficient for directing prehospital decision making.

Prehospital Translation of Chest Pain Tools (RESCUE Study): Completion Rate and Inter-rater Reliability.

INTRODUCTION: Chest pain is a common reason for ambulance transport. Acute coronary syndrome (ACS) and pulmonary embolism (PE) risk assessments, such as history, electrocardiogram, age, risk factors (HEAR); Emergency Department Assessment of Chest Pain Score (EDACS); Pulmonary Embolism Rule-out Criteria (PERC); and revised Geneva score, are well validated for emergency department (ED) use but have not been translated to the prehospital setting. The objectives of this study were to evaluate the 1) prehospital completion rate and 2) inter-rater reliability of chest pain risk assessments. METHODS: We conducted a prospective observational cohort study in two emergency medical services (EMS) agencies (April 18, 2018 - January 2, 2019). Adults with acute, non-traumatic chest pain without ST-elevation myocardial infarction or unstable vital signs were accrued. Paramedics were trained to use the HEAR, EDACS, PERC, and revised Geneva score assessments. A subset of patients (a priori goal of N = 250) also had the four risk assessments completed by their treating clinicians in the ED, who were blinded to the EMS risk assessments. Outcomes were 1) risk assessments completion rate and 2) inter-rater reliability between EMS and ED assessments. An a priori goal for completion rate was set as >75%. We computed kappa with corresponding 95% confidence intervals (CI) for each risk assessment as a measure of inter-rater reliability. Acceptable agreement was defined a priori as kappa ≥ 0.60. RESULTS: During the study period, 837 patients with acute chest pain were accrued. The median age was 54 years, interquartile range 43-66, with 53% female and 51% Black. Completion rates for each risk assessment were above goal: the HEAR score was completed on 95.1% (796/837), EDACS on 92.0% (770/837), PERC on 89.4% (748/837), and revised Geneva score on 90.7% (759/837) of patients. We assessed agreement in a subgroup of 260 patients. The HEAR score had a kappa of 0.51 (95% CI, 0.41-0.61); EDACS was 0.60 (95% CI, 0.49-0.72); PERC was 0.71 (95% CI, 0.61-0.81); and revised Geneva score was 0.51 (95% CI, 0.39-0.62). CONCLUSION: The completion rate of risk assessments for ACS and PE was high for prehospital field personnel. The PERC and EDACS both demonstrated acceptable agreement between paramedics and clinicians in the ED, although assessments with better agreement are likely needed.

The ethical dilemma of emergency department patients with low-risk chest pain.

Millions of patients present to US EDs each year with symptoms concerning for acute coronary syndrome (ACS), but fewer than 10% are ultimately diagnosed with ACS. Well-tested and externally validated accelerated diagnostic protocols were developed to aid providers in risk stratifying patients with possible ACS and have become central components of current ED practice guidelines. Nevertheless, the fear of missing ACS continues to be a strong motivator for ED providers to pursue further testing for their patients. An ethical dilemma arises when the provider must balance the risk of ACS if the patient is discharged compared with the potential harms caused by a cardiac workup. Providers should be familiar with the ethical principles relevant to this dilemma in order to determine what is in the best interests of the patient.

The disutility of stress testing in low-risk HEART Pathway patients.

BACKGROUND: The HEART Pathway identifies low-risk chest pain patients for discharge from the Emergency Department without stress testing. However, HEART Pathway recommendations are not always followed. The objective of this study is to determine the frequency and diagnostic yield of stress testing among low-risk patients. METHODS: An academic hospital's chest pain registry was analyzed for low-risk HEART Pathway patients (HEAR score ≤ 3 with non-elevated troponins) from 1/2017 to 7/2018. Stress tests were reviewed for inducible ischemia. Diagnostic yield was defined as the rate of obstructive CAD among patients with positive stress testing. T-test or Fisher's exact test was used to test the univariate association of age, sex, race/ethnicity, and HEAR score with stress testing. Multivariate logistic regression was used to determine the association of age, sex, race/ethnicity, and HEAR score with stress testing. RESULTS: There were 4743 HEART Pathway assessments, with 43.7% (2074/4743) being low-risk. Stress testing was performed on 4.1% (84/2074). Of the 84 low-risk patients who underwent testing, 8.3% (7/84) had non-diagnostic studies and 2.6% (2/84) had positive studies. Among the 2 patients with positive studies, angiography revealed that 1 had widely patent coronary arteries and the other had multivessel obstructive coronary artery disease, making the diagnostic yield of stress testing 1.2% (1/84). Each one-point increase in HEAR score (aOR 2.17, 95% CI 1.45-3.24) and being male (aOR 1.59, 95% CI 1.02-2.49) were associated with testing. CONCLUSIONS: Stress testing among low-risk HEART Pathway patients was uncommon, low yield, and more likely in males and those with a higher HEAR score.

Prehospital use of a modified HEART Pathway and point-of-care troponin to predict cardiovascular events.

CLINICAL TRIAL REGISTRATION: clinicaltrials.gov (NCT02709135).

Prehospital Trauma Scene and Transport Times for Pediatric and Adult Patients.

INTRODUCTION: Increased out-of-hospital time is associated with worse outcomes in trauma. Sparse literature exists comparing prehospital scene and transport time management intervals between adult and pediatric trauma patients. National Emergency Medical Services guidelines recommend that trauma scene time be less than 10 minutes. The objective of this study was to examine prehospital time intervals in adult and pediatric trauma patients. METHODS: We performed a retrospective cohort study of blunt and penetrating trauma patients in a five-county region in North Carolina using prehospital records. We included patients who were transported emergency traffic directly from the scene by ground ambulance to a Level I or Level II trauma center between 2013-2018. We defined pediatric patients as those less than 16 years old. Urbanicity was controlled for using the Centers for Medicare and Medicaid's Ambulance Fee Schedule. We performed descriptive statistics and linear mixed-effects regression modeling. RESULTS: A total of 2179 records met the study criteria, of which 2077 were used in the analysis. Mean scene time was 14.2 minutes (95% confidence interval [CI], 13.9-14.5) and 35.3% (n = 733) of encounters had a scene time of 10 minutes or less. Mean transport time was 17.5 minutes (95% CI, 17.0-17.9). Linear mixed-effects regression revealed that scene times were shorter for pediatric patients (p<0.0001), males (p=0.0016), penetrating injury (p<0.0001), and patients with blunt trauma in rural settings (p=0.005), and that transport times were shorter for males (p = 0.02), non-White patients (p<0.0001), and patients in urban areas (p<0.0001). CONCLUSION: This study population largely missed the 10-minute scene time goal. Demographic and patient factors were associated with scene and transport times. Shorter scene times occurred with pediatric patients, males, and among those with penetrating trauma. Additionally, suffering blunt trauma while in a rural environment was associated with shorter scene time. Males, non-White patients, and patients in urban environments tended to have shorter transport times. Future studies with outcomes data are needed to identify factors that prolong out-of-hospital time and to assess the impact of out-of-hospital time on patient outcomes.

Point-of-Care Troponin Testing during Ambulance Transport to Detect Acute Myocardial Infarction.

Objective: Use of point-of-care (POC) troponin (cTn) testing in the Emergency Department (ED) is well established. However, data examining POC cTn measurement in the prehospital setting, during ambulance transport, are limited. The objective of this study was to prospectively test the performance of POC cTn measurement by paramedics to detect myocardial infarction (MI) among patients transported to the ED for acute chest pain. Methods: A prospective cohort study of adults with non-traumatic chest pain was conducted in three Emergency Medical Services agencies (December 2016 to January 2018). Patients with ST-elevation MI on ECG were excluded. During ambulance transport paramedics initiated intravenous access, collected blood, and used a POC device (i-STAT; Abbott Laboratories) to measure cTn. Following ED arrival, participants received standard evaluations including clinical blood draws for cTn measurement in the hospital central lab (AccuTnI +3 assay; Beckman Coulter, or cTnI-Ultra assay; Siemens). Blood collected during ambulance transport was also analyzed for cTn in the central lab. Index visit MI was adjudicated by 3 experts using central lab cTn measures from the patient's clinical blood draws. Test characteristics (sensitivity, specificity, and predictive values) for detection of MI were calculated for POC and central lab cTn measurement of prehospital blood and compared with McNemar's test. Results: During the study period prehospital POC cTn results were obtained on 421 patients, of which 5.0% (21/421) had results >99th percentile upper reference limit. MI was adjudicated in 16.2% (68/421) during the index visit. The specificity and positive predictive value of the POC cTn measurement were 99.2% (95% CI 97.5-99.8%) and 85.7% (95% CI 63.7-97.0%) for MI. However, the sensitivity and NPV of prehospital POC cTn were 26.5% (95% CI 16.5-38.6%) and 87.5% (95% CI 83.9-90.6%). Compared to POC cTn, the central lab cTn measurement of prehospital blood resulted in a higher sensitivity of 67.9% (95% CI 53.7-80.1%, p < 0.0001), but lower specificity of 92.4% (95% CI 88.4-95.4%, p = 0.0001). Conclusions: Prehospital POC i-STAT cTn measurement in patients transported with acute chest pain was highly specific for MI but had low sensitivity. This suggests that prehospital i-STAT POC cTn could be useful to rule-in MI, but should not be used to exclude MI.