- Email: [email protected]
114: A Comparison of Out-of-Hospital Rapid Sequence Intubation Success to Non-Paralyzed Patients
Research Forum Abstracts Morphine administration for pain occurred in 537 (2.3%) of the ⬎.25% group while it was administered to 487 (4.8%) of the ⬍.25% group. This is a difference of 2.5%. Conclusion: Ambulance services with UHU of ⬍25% have a higher rate of delivering at least one intervention to a patient, providing an ALS intervention and treating pain with morphine sulfate. Further study is necessary to determine what these differences have on patient outcomes.
111
Insurance Status as a Predictor of Mode of Arrival for Patients Who Present to the Emergency Department With Chest Pain
Wu JT, Bhatti P, Goetz JD, Weiner SG/Tufts University School of Medicine, Boston, MA; Tufts Medical Center, Boston, MA
Study Objectives: Previous studies have evaluated patient-specific factors for patients who choose to take an ambulance to the ED. We wished to determine if insurance status was a predictor for EMS arrival for the common, but potentially lifethreatening, complaint of chest pain. Methods: All adult patients ages 18 –99 who presented to an urban academic ED between 1/06 and 7/06 with a chief complaint that included “chest pain” were eligible for retrospective analysis. For patients with multiple visits, only the first visit for this complaint during the study period was included. Patients who were transferred, incarcerated or who left without being seen or against medical advice were excluded. Insurance status was documented by registration personnel on a computerized record. Results: There were 690 visits for chest pain during the study period, representing 4% of total ED census. A total of 42 visits met exclusion criteria, and 37 patients had 52 repeat visits, leaving 596 visits included for analysis. 22% (56/250) of patients with private insurance arrived via EMS. Using private insurance as a reference, 36% (46/129) of Medicare patients (OR 1.92, 95% CI 1.20 –3.06), 24% (36/152) of Medicaid patients (OR 1.08, 95% CI 0.67–1.73) and 25% (16/65) of Self Pay patients (OR 1.13, 95% CI 0.60 –2.14) arrived by ambulance. Only Medicare patients had a statistically significant increased likelihood of EMS transport (OR 1.84, 95% CI 1.21–2.80) when compared with private, Medicaid and Self Pay patients (p⫽0.005). Patients who arrived via ambulance were more likely to be admitted to the hospital (OR 1.75, 95% CI 1.21–2.55) but there was no difference in final diagnosis of “myocardial infarction” among the different insurance types (p⫽0.40). Conclusion: Of the four major types of insurance, only Medicare patients with a chief complaint of chest pain were more likely to utilize EMS. There was no significant difference among patients with private insurance, Medicaid or self-pay status.
112
Knowledge of Self-Injectable Epinephrine Technique Among Emergency Medical Services Providers
Davis JE, Churosh N, Borloz M, Howell J/Georgetown University Hospital & Washington Hospital Center, Washington, DC; Inova Fairfax Hospital, Fairfax, VA
Study Objectives: Emergency medical services (EMS) personnel may be the first to encounter a patient with an allergic emergency. Several studies have revealed that health care provider (physician, nurse) knowledge of the technique of self-injectable epinephrine (such as EpiPen® or EpiPen® Jr) administration is deficient in general. Studies focusing specifically on EMS personnel are lacking. We therefore sought to assess emergency medical technician (EMT) knowledge of self-injectable epinephrine use, and evaluate the efficacy of a brief, directed educational intervention. Methods: We assessed baseline knowledge of self-injectable epinephrine technique among EMT providers, then provided an educational intervention (we created an online training module lasting less than 5 minutes). Study subjects were retested immediately following training module completion, and again at 3-months. Proper technique was defined in 5 steps, per self-injectable epinephrine medication package insert instructions: (1) grasp device, (2) remove safety cap, (3) inject into lateral thigh, (4) hold in place for 10 seconds, and (5) rub injection site for 10 seconds following device removal. This study was approved by our institutional review board. Nominal data were analyzed using chi square and Fisher’s Exact tests. Alpha was set at 0.05 for all comparisons. Data were analyzed using GraphPad Prism version 5.00 for Macintosh, GraphPad Software, San Diego, California, USA, www.graphpad.com. Results: All participants were EMT basic certified providers from a single collegiate EMS system, with a mean of 1.8 years of experience (range: ⬍1 year to 3.5
S36 Annals of Emergency Medicine
years). At baseline, 4.6% of 22 participants correctly demonstrated all 5 steps compared with 73% (95% confidence interval: 50 – 89%) immediately post intervention, and 72% (95% confidence interval: 49 – 88%) at 3-month follow-up. Four participants were lost to 3-month follow-up. Baseline measurements were significantly different than immediate and 3-month post-intervention measurements. Conclusion: Similar to studies of other health care providers, EMT basic providers demonstrated poor baseline knowledge of proper self-injectable epinephrine technique. Knowledge improved significantly following a brief educational intervention, and was well retained at 3-month follow-up. Brief, focused educational interventions may assist health care providers in learning and retaining knowledge regarding the proper technique for self-injectable epinephrine administration.
113
The Treatment of Motion Sickness in the Out-ofHospital Setting: A Comparison of Metoclopramide and Diphenhydramine to Placebo
Weichenthal LA, Andrews J, Rubio S/UCSF-Fresno, Fresno, CA
Study Objectives: To determine the incidence of motion sickness in patients being transported via ambulance in a mountainous setting while comparing metoclopramide and diphenhydramine to placebo in the treatment of these patients. Methods: This was a prospective, randomized, double-blinded, placebo controlled study of patients transported by ambulance in the mountainous regions of Fresno County. Patients who met inclusion criteria and who agreed to participate in the study were asked to rate their motion sickness every 5 minutes on a visual analog scale (VAS) during transport. If they developed motion sickness, they were randomized to recieve metoclopramide (20 mg IV), diphenhydramine (50 mg IV), or placebo (saline). Symptoms then continued to be recorded every 5 minutes on a VAS. If subjects continued to have signs and symptoms of motion sickness after 15 minutes, a rescue dose of metoclopramide was offered. Results: Twenty six subjects were enrolled in the study. Twenty two (84.6 %) developed motion sickness during transport. These patients were randomized to the three different treatment arms: Eight receive metoclopramide, seven received diphenhydramine, and seven recieved placebo. The metoclopramide group showed a significant decrease in mean VAS score at 15 minutes compared to the dihenydramine and placebo groups (p⫽0.0226). Twelve of the twenty two patients required a rescue dose of metoclpramide after 15 minues. Eleven of these patients were from the diphenhydramine and placebo groups. At twenty five minutes, there was no difference in the VAS score between the three groups. Conclusion: There is a significant incidence of motion sickness in patients being transported by anbulance in a mountainous setting. Metoclopramide is superior to diphenhydramine and placebo in the treatment of motion sickness in this environment.
114
A Comparison of Out-of-Hospital Rapid Sequence Intubation Success to Non-Paralyzed Patients
Felderman H, Walsh B, Yasbin P/Morristown Memorial Hospital, Morristown, NJ; Atlantic Ambulance, Morristown, NJ
Study Objectives: Out-of-hospital intubation, and especially out-of-hospital rapid sequence intubation (RSI), is a controversial procedure that is frequently debated in the literature. Our paramedics intubate frequently, have regular educational updates, and are evaluated routinely in cadaver labs. In order to fine-tune our educational process, we sought to determine our paramedics’ baseline intubation skills and the impact of RSI on success rates. Methods: We retrospectively analyzed all patients in which intubation was attempted by our ground and air units over a 23-month period. In order to determine baseline procedural competence and the impact of RSI, we subdivided patients in to three groups: those in cardiac arrest (CA), those with a pulse who underwent RSI (RSI), and those with a pulse who did not receive RSI (I). We compared the group in terms of “successful” intubation (⬍⫽ 2 attempts) and “overall” intubation (⬍⫽4 attempts) using a Chi-Square test with a Marasciullo correction for multiple comparisons. Results: Of the 751 patients with intubation attempts, 330 were in cardiac arrest, 196 received RSI, 225 did not receive RSI. In terms of “successful” intubations: 88% of CA patients were intubated within 2 attempts, 90% of RSI patients were intubated within 2 attempts, and 82% of I patients were intubated within 2 attempts. The differences in “successful” intubation rates between these groups did not reach statistical significance. In terms of “overall” intubation rates, there were a total of 687
Volume , . : September
Research Forum Abstracts patients (91.5%) who were intubated out-of-hospitally: 94% of the CA group, 94% of the RSI group and 85% of the I group. Patients in the CA and RSI groups were significantly more likely to be intubated than those in the I group (p⬍0.05 for both comparisons). Conclusion: Although rates approached significance, we found no difference in rates of “successful” intubation in the three groups. In terms of “overall” intubation rates, paramedics have higher intubation rates in patients with a pulse when utilizing RSI, and the success rate of RSI approaches that of patients in CA. This suggests that RSI is an effective adjunct to intubation for patients with a pulse. Prospective, outcome-based studies are needed to determine the true impact of RSI in our group of paramedics.
115
The Predictive Value of Arrival With EMS
Felderman H, Walsh B, Shih S, Luk J, Sturm D/Morristown Memorial Hospital, Morristown, NJ
Study Objectives: Estimating the severity of illness is a crucial part of the initial triage in an emergency department. arrival with emergency medical services (EMS) is considered to be associated with increased severity, although it is unproven to date. We sought to determine the significance of mode of arrival in patients who present to the emergency department. Methods: A retrospective analysis of all patients seen in four emergency departments between 11/1/04 and 10/31/06 was conducted. Patients were evaluated with the following chief complaints: 1) All Diagnoses (AD), 2) Dyspnea (SOB), 3) Abdominal Pain (AP), 4) Mood Disorders (MD), 5) Palpitations (HR), 6) Syncope (S), and 7) Alcohol Abuse (AA). These patients were then subdivided into 20-year age groups (0 –20, 21– 40, 41– 60, 61– 80, ⬎80). We used admission to the hospital as a marker for severity of illness. We calculated the odds ratio (OR) and 95% confidence intervals for admission to the hospital for those who arrived with EMS compared with those who did not. Results: Of the 231,219 patients in our database, 222,619 patients had delineated modes of arrival; 50,700 arrived via EMS, 171,919 patients did not. Arrival by EMS was associated with an increased rate of admission for AD: OR 4.9[4.8 –5.0], SOB: OR 7.2[6.9 –7.7], AP: OR 2.4[2.1–2.6], and MD: OR 1.23[1.1–1.4]. These positive associations were significant in all age groups. For HR, positive associations were found in all age groups except 0 –20: OR 1.8[0.3–11.6], and ⬎80: OR 1.4[0.9 –2.2]. For S there were no significant associations when corrected for age. Interestingly, for patient arriving with AA, arrival by EMS is a negative predictor of admission: OR 0.4[0.3– 0.5]. Conclusion: While arrival by EMS, in general, is associated with an increased rate of admission, there are many important exceptions to this rule. This study suggests that arrival by EMS should not be used alone to make triage decisions.
116
Postural Hypotension in the Elderly: Predictors for Intervention
Chan W, Foo C/Tan Tock Seng Hospital, Singapore, Singapore, Singapore
Background: Postural hypotension in an elderly patient usually demands a search for, and management of, reversible causes. To date, however, there are no studies examining postural hypotension in elderly patients who present to the emergency department (ED). This study aims to identify predictors for intervention in patients with postural hypotension. Study Objectives: This study examines: 1. The characteristics of postural hypotension 2. The possible causes of postural hypotension 3. The factors that may predict the need for intervention Methods: This is a retrospective study evaluating elderly patients aged 65-andabove who were found to have postural hypotension prior to discharge from a 24hour ED short stay ward (Emergency Diagnostic and Treatment Centre; EDTC) from 1st April 2007 to 31st December 2008. Nursing home residents, patients with severe cognitive or functional impairment, patients already on follow-up with a geriatrician, and patients who refused geriatric assessment were excluded. Patient demographics, characteristics of postural hypotension and the likely causes were examined. Through review of case records by an independent reviewer, patients who required and benefited from intervention were identified and compared against those who did not, to determine if there were any variables that predicted the need for intervention.
Volume , . : September
Results: Of the 869 patients aged 65 years and above who were admitted to EDTC during the study period and received geriatric screening, 157 (18.1%) had postural hypotension, of which 92 (58.6%) were female and 140 (89.2%) required intervention. The mean age in the intervention group was 79.4, compared to 73.7 in the non-intervention group (p⫽0.01). The intervention group had a lower mean abbreviated mental test (AMT) score (7.4) when compared to the non-intervention group (8.7; p⫽0.02). With regards to the characteristics of postural hypotension, 145 (92.4%) had a systolic blood pressure (SBP) drop of 20mmHg, 149 (94.9%) were detected in the first minute of standing, 91 (63.6%) were reproducible, 75 (47.8%) were symptomatic and 64 (40.8%) had a history of falls or near falls. In 46.5% of cases, medications were found to be a contributor to postural hypotension. Dehydration (28.0%), sepsis (26.1%) and diabetic autonomic neuropathy (22.3%) were other common causes. 74 (47.1%) of postural hypotension cases were multifactorial. An etiology was not found in 29 (18.5%) of cases. We also found that postural hypotensive patients who were symptomatic (OR 3.3; 95% CI 1.0 to 10.8) and on 3 or more medications (OR 3.3; 95% CI 1.1 to 9.4) were more likely to have received intervention. Conclusions: Postural hypotension was found in 18.1% of elderly patients in the EDTC. Medications, dehydration, sepsis and diabetes were common causes. The majority required treatment and follow-up. Increased age, lower AMT score, polypharmacy and being symptomatic were predictors for need for intervention.
117
Do HIV-Positive Patients With Severe Sepsis Receive Adequate Initial Antibiotics in the Emergency Department When Compared With HIVNegative Patients?
McGrath ME, Bullock HN, Whitney D/Boston Medical Center, Boston, MA
Study Objectives: To determine if HIV-positive (HIV⫹) patients with severe sepsis received adequate initial antibiotics (abx) in the emergency department (ED) compared to HIV-negative (HIV⫺) patients. Methods: Retrospective observational study of HIV⫹ and HIV⫺ patients with severe sepsis (2⫹ SIRS criteria) admitted to the ICU from an urban academic ED over 18 months. HIV status was determined by review of ED and hospital records 6 months prior to presentation for CD4 counts and viral load. Patient characteristics, mortality, and length of stay was compared between groups using Chi-square, Fisher’s exact, and Wilcoxon rank sum tests. Cochran-Mantzel-Haenszel(CMH) test calculated relative risk (RR) of infection. Adequacy of coverage was determined by comparing initial abx ordered in the ED with sensitivities of pathogens cultured. Results: 325 patients were included: 39 HIV⫹ and 286 HIV⫺. HIV⫹ patients more often were younger (mean 47 yrs vs 62 yrs, p⬍0.001), black (59% vs 38%, p⫽0.02), used drugs (28% vs 7%, p⫽0.001) and smoked (41% vs 11%, p⬍0.001). No difference was found in length of hospital/ICU stay, MEDS score or mortality; 13% HIV⫹ patients died in hospital vs 18% HIV⫺ patients (p⫽0.41). Abx-resistant pathogens were common overall (20% MRSA, 7% VRE). In HIV⫹ patients, 21/ 24(88%) pathogens cultured were gram pos, 3/24(12%) gram neg. In HIV⫺ patients 122/208(59%) pathogens cultured were gram pos, 83/208(40%) gram neg, and 3/ 208(1%) fungal (p⬍0.001). HIV⫺ patients had 3.2 times RR of infection by gram neg pathogen than HIV⫹ patients (95%CI, 1.1–9.5). In HIV⫹ patients, 19/24(79%) pathogens were adequately covered by initial ED abx and 5/24(21%) were not. In HIV⫺ patients, 148/208(71%) pathogens were adequately covered and 60/208(29%) were not. No difference was found in adequacy of abx coverage between the groups (p⫽0.41). Conclusion: We found no difference in adequacy of initial ED abx coverage, length of stay, or hospital mortality of HIV⫹ patients with severe sepsis compared to HIV⫺ patients. HIV⫹ patients had more gram pos infections and there was a high prevalence of abx resistant pathogens overall.
118
External Site Testing of an Instrument to Predict Endocarditis in Injection Drug Users With Fever
Romero K, Rodriguez R, Chiang W, Fortman J, Colucci A/University of California, San Francisco, San Francisco, CA; Bellevue Hospital, New York City, NY
Study Objective: To externally test a previously derived decision instrument (100% sensitivity and 44% specificity in prior study) for endocarditis prediction in injection drug users (IDUs) admitted from the ED with fever. Methods: Blinded to the prior instrument, an investigator used the same chart
Annals of Emergency Medicine S37
111
Insurance Status as a Predictor of Mode of Arrival for Patients Who Present to the Emergency Department With Chest Pain
Wu JT, Bhatti P, Goetz JD, Weiner SG/Tufts University School of Medicine, Boston, MA; Tufts Medical Center, Boston, MA
Study Objectives: Previous studies have evaluated patient-specific factors for patients who choose to take an ambulance to the ED. We wished to determine if insurance status was a predictor for EMS arrival for the common, but potentially lifethreatening, complaint of chest pain. Methods: All adult patients ages 18 –99 who presented to an urban academic ED between 1/06 and 7/06 with a chief complaint that included “chest pain” were eligible for retrospective analysis. For patients with multiple visits, only the first visit for this complaint during the study period was included. Patients who were transferred, incarcerated or who left without being seen or against medical advice were excluded. Insurance status was documented by registration personnel on a computerized record. Results: There were 690 visits for chest pain during the study period, representing 4% of total ED census. A total of 42 visits met exclusion criteria, and 37 patients had 52 repeat visits, leaving 596 visits included for analysis. 22% (56/250) of patients with private insurance arrived via EMS. Using private insurance as a reference, 36% (46/129) of Medicare patients (OR 1.92, 95% CI 1.20 –3.06), 24% (36/152) of Medicaid patients (OR 1.08, 95% CI 0.67–1.73) and 25% (16/65) of Self Pay patients (OR 1.13, 95% CI 0.60 –2.14) arrived by ambulance. Only Medicare patients had a statistically significant increased likelihood of EMS transport (OR 1.84, 95% CI 1.21–2.80) when compared with private, Medicaid and Self Pay patients (p⫽0.005). Patients who arrived via ambulance were more likely to be admitted to the hospital (OR 1.75, 95% CI 1.21–2.55) but there was no difference in final diagnosis of “myocardial infarction” among the different insurance types (p⫽0.40). Conclusion: Of the four major types of insurance, only Medicare patients with a chief complaint of chest pain were more likely to utilize EMS. There was no significant difference among patients with private insurance, Medicaid or self-pay status.
112
Knowledge of Self-Injectable Epinephrine Technique Among Emergency Medical Services Providers
Davis JE, Churosh N, Borloz M, Howell J/Georgetown University Hospital & Washington Hospital Center, Washington, DC; Inova Fairfax Hospital, Fairfax, VA
Study Objectives: Emergency medical services (EMS) personnel may be the first to encounter a patient with an allergic emergency. Several studies have revealed that health care provider (physician, nurse) knowledge of the technique of self-injectable epinephrine (such as EpiPen® or EpiPen® Jr) administration is deficient in general. Studies focusing specifically on EMS personnel are lacking. We therefore sought to assess emergency medical technician (EMT) knowledge of self-injectable epinephrine use, and evaluate the efficacy of a brief, directed educational intervention. Methods: We assessed baseline knowledge of self-injectable epinephrine technique among EMT providers, then provided an educational intervention (we created an online training module lasting less than 5 minutes). Study subjects were retested immediately following training module completion, and again at 3-months. Proper technique was defined in 5 steps, per self-injectable epinephrine medication package insert instructions: (1) grasp device, (2) remove safety cap, (3) inject into lateral thigh, (4) hold in place for 10 seconds, and (5) rub injection site for 10 seconds following device removal. This study was approved by our institutional review board. Nominal data were analyzed using chi square and Fisher’s Exact tests. Alpha was set at 0.05 for all comparisons. Data were analyzed using GraphPad Prism version 5.00 for Macintosh, GraphPad Software, San Diego, California, USA, www.graphpad.com. Results: All participants were EMT basic certified providers from a single collegiate EMS system, with a mean of 1.8 years of experience (range: ⬍1 year to 3.5
S36 Annals of Emergency Medicine
years). At baseline, 4.6% of 22 participants correctly demonstrated all 5 steps compared with 73% (95% confidence interval: 50 – 89%) immediately post intervention, and 72% (95% confidence interval: 49 – 88%) at 3-month follow-up. Four participants were lost to 3-month follow-up. Baseline measurements were significantly different than immediate and 3-month post-intervention measurements. Conclusion: Similar to studies of other health care providers, EMT basic providers demonstrated poor baseline knowledge of proper self-injectable epinephrine technique. Knowledge improved significantly following a brief educational intervention, and was well retained at 3-month follow-up. Brief, focused educational interventions may assist health care providers in learning and retaining knowledge regarding the proper technique for self-injectable epinephrine administration.
113
The Treatment of Motion Sickness in the Out-ofHospital Setting: A Comparison of Metoclopramide and Diphenhydramine to Placebo
Weichenthal LA, Andrews J, Rubio S/UCSF-Fresno, Fresno, CA
Study Objectives: To determine the incidence of motion sickness in patients being transported via ambulance in a mountainous setting while comparing metoclopramide and diphenhydramine to placebo in the treatment of these patients. Methods: This was a prospective, randomized, double-blinded, placebo controlled study of patients transported by ambulance in the mountainous regions of Fresno County. Patients who met inclusion criteria and who agreed to participate in the study were asked to rate their motion sickness every 5 minutes on a visual analog scale (VAS) during transport. If they developed motion sickness, they were randomized to recieve metoclopramide (20 mg IV), diphenhydramine (50 mg IV), or placebo (saline). Symptoms then continued to be recorded every 5 minutes on a VAS. If subjects continued to have signs and symptoms of motion sickness after 15 minutes, a rescue dose of metoclopramide was offered. Results: Twenty six subjects were enrolled in the study. Twenty two (84.6 %) developed motion sickness during transport. These patients were randomized to the three different treatment arms: Eight receive metoclopramide, seven received diphenhydramine, and seven recieved placebo. The metoclopramide group showed a significant decrease in mean VAS score at 15 minutes compared to the dihenydramine and placebo groups (p⫽0.0226). Twelve of the twenty two patients required a rescue dose of metoclpramide after 15 minues. Eleven of these patients were from the diphenhydramine and placebo groups. At twenty five minutes, there was no difference in the VAS score between the three groups. Conclusion: There is a significant incidence of motion sickness in patients being transported by anbulance in a mountainous setting. Metoclopramide is superior to diphenhydramine and placebo in the treatment of motion sickness in this environment.
114
A Comparison of Out-of-Hospital Rapid Sequence Intubation Success to Non-Paralyzed Patients
Felderman H, Walsh B, Yasbin P/Morristown Memorial Hospital, Morristown, NJ; Atlantic Ambulance, Morristown, NJ
Study Objectives: Out-of-hospital intubation, and especially out-of-hospital rapid sequence intubation (RSI), is a controversial procedure that is frequently debated in the literature. Our paramedics intubate frequently, have regular educational updates, and are evaluated routinely in cadaver labs. In order to fine-tune our educational process, we sought to determine our paramedics’ baseline intubation skills and the impact of RSI on success rates. Methods: We retrospectively analyzed all patients in which intubation was attempted by our ground and air units over a 23-month period. In order to determine baseline procedural competence and the impact of RSI, we subdivided patients in to three groups: those in cardiac arrest (CA), those with a pulse who underwent RSI (RSI), and those with a pulse who did not receive RSI (I). We compared the group in terms of “successful” intubation (⬍⫽ 2 attempts) and “overall” intubation (⬍⫽4 attempts) using a Chi-Square test with a Marasciullo correction for multiple comparisons. Results: Of the 751 patients with intubation attempts, 330 were in cardiac arrest, 196 received RSI, 225 did not receive RSI. In terms of “successful” intubations: 88% of CA patients were intubated within 2 attempts, 90% of RSI patients were intubated within 2 attempts, and 82% of I patients were intubated within 2 attempts. The differences in “successful” intubation rates between these groups did not reach statistical significance. In terms of “overall” intubation rates, there were a total of 687
Volume , . : September
Research Forum Abstracts patients (91.5%) who were intubated out-of-hospitally: 94% of the CA group, 94% of the RSI group and 85% of the I group. Patients in the CA and RSI groups were significantly more likely to be intubated than those in the I group (p⬍0.05 for both comparisons). Conclusion: Although rates approached significance, we found no difference in rates of “successful” intubation in the three groups. In terms of “overall” intubation rates, paramedics have higher intubation rates in patients with a pulse when utilizing RSI, and the success rate of RSI approaches that of patients in CA. This suggests that RSI is an effective adjunct to intubation for patients with a pulse. Prospective, outcome-based studies are needed to determine the true impact of RSI in our group of paramedics.
115
The Predictive Value of Arrival With EMS
Felderman H, Walsh B, Shih S, Luk J, Sturm D/Morristown Memorial Hospital, Morristown, NJ
Study Objectives: Estimating the severity of illness is a crucial part of the initial triage in an emergency department. arrival with emergency medical services (EMS) is considered to be associated with increased severity, although it is unproven to date. We sought to determine the significance of mode of arrival in patients who present to the emergency department. Methods: A retrospective analysis of all patients seen in four emergency departments between 11/1/04 and 10/31/06 was conducted. Patients were evaluated with the following chief complaints: 1) All Diagnoses (AD), 2) Dyspnea (SOB), 3) Abdominal Pain (AP), 4) Mood Disorders (MD), 5) Palpitations (HR), 6) Syncope (S), and 7) Alcohol Abuse (AA). These patients were then subdivided into 20-year age groups (0 –20, 21– 40, 41– 60, 61– 80, ⬎80). We used admission to the hospital as a marker for severity of illness. We calculated the odds ratio (OR) and 95% confidence intervals for admission to the hospital for those who arrived with EMS compared with those who did not. Results: Of the 231,219 patients in our database, 222,619 patients had delineated modes of arrival; 50,700 arrived via EMS, 171,919 patients did not. Arrival by EMS was associated with an increased rate of admission for AD: OR 4.9[4.8 –5.0], SOB: OR 7.2[6.9 –7.7], AP: OR 2.4[2.1–2.6], and MD: OR 1.23[1.1–1.4]. These positive associations were significant in all age groups. For HR, positive associations were found in all age groups except 0 –20: OR 1.8[0.3–11.6], and ⬎80: OR 1.4[0.9 –2.2]. For S there were no significant associations when corrected for age. Interestingly, for patient arriving with AA, arrival by EMS is a negative predictor of admission: OR 0.4[0.3– 0.5]. Conclusion: While arrival by EMS, in general, is associated with an increased rate of admission, there are many important exceptions to this rule. This study suggests that arrival by EMS should not be used alone to make triage decisions.
116
Postural Hypotension in the Elderly: Predictors for Intervention
Chan W, Foo C/Tan Tock Seng Hospital, Singapore, Singapore, Singapore
Background: Postural hypotension in an elderly patient usually demands a search for, and management of, reversible causes. To date, however, there are no studies examining postural hypotension in elderly patients who present to the emergency department (ED). This study aims to identify predictors for intervention in patients with postural hypotension. Study Objectives: This study examines: 1. The characteristics of postural hypotension 2. The possible causes of postural hypotension 3. The factors that may predict the need for intervention Methods: This is a retrospective study evaluating elderly patients aged 65-andabove who were found to have postural hypotension prior to discharge from a 24hour ED short stay ward (Emergency Diagnostic and Treatment Centre; EDTC) from 1st April 2007 to 31st December 2008. Nursing home residents, patients with severe cognitive or functional impairment, patients already on follow-up with a geriatrician, and patients who refused geriatric assessment were excluded. Patient demographics, characteristics of postural hypotension and the likely causes were examined. Through review of case records by an independent reviewer, patients who required and benefited from intervention were identified and compared against those who did not, to determine if there were any variables that predicted the need for intervention.
Volume , . : September
Results: Of the 869 patients aged 65 years and above who were admitted to EDTC during the study period and received geriatric screening, 157 (18.1%) had postural hypotension, of which 92 (58.6%) were female and 140 (89.2%) required intervention. The mean age in the intervention group was 79.4, compared to 73.7 in the non-intervention group (p⫽0.01). The intervention group had a lower mean abbreviated mental test (AMT) score (7.4) when compared to the non-intervention group (8.7; p⫽0.02). With regards to the characteristics of postural hypotension, 145 (92.4%) had a systolic blood pressure (SBP) drop of 20mmHg, 149 (94.9%) were detected in the first minute of standing, 91 (63.6%) were reproducible, 75 (47.8%) were symptomatic and 64 (40.8%) had a history of falls or near falls. In 46.5% of cases, medications were found to be a contributor to postural hypotension. Dehydration (28.0%), sepsis (26.1%) and diabetic autonomic neuropathy (22.3%) were other common causes. 74 (47.1%) of postural hypotension cases were multifactorial. An etiology was not found in 29 (18.5%) of cases. We also found that postural hypotensive patients who were symptomatic (OR 3.3; 95% CI 1.0 to 10.8) and on 3 or more medications (OR 3.3; 95% CI 1.1 to 9.4) were more likely to have received intervention. Conclusions: Postural hypotension was found in 18.1% of elderly patients in the EDTC. Medications, dehydration, sepsis and diabetes were common causes. The majority required treatment and follow-up. Increased age, lower AMT score, polypharmacy and being symptomatic were predictors for need for intervention.
117
Do HIV-Positive Patients With Severe Sepsis Receive Adequate Initial Antibiotics in the Emergency Department When Compared With HIVNegative Patients?
McGrath ME, Bullock HN, Whitney D/Boston Medical Center, Boston, MA
Study Objectives: To determine if HIV-positive (HIV⫹) patients with severe sepsis received adequate initial antibiotics (abx) in the emergency department (ED) compared to HIV-negative (HIV⫺) patients. Methods: Retrospective observational study of HIV⫹ and HIV⫺ patients with severe sepsis (2⫹ SIRS criteria) admitted to the ICU from an urban academic ED over 18 months. HIV status was determined by review of ED and hospital records 6 months prior to presentation for CD4 counts and viral load. Patient characteristics, mortality, and length of stay was compared between groups using Chi-square, Fisher’s exact, and Wilcoxon rank sum tests. Cochran-Mantzel-Haenszel(CMH) test calculated relative risk (RR) of infection. Adequacy of coverage was determined by comparing initial abx ordered in the ED with sensitivities of pathogens cultured. Results: 325 patients were included: 39 HIV⫹ and 286 HIV⫺. HIV⫹ patients more often were younger (mean 47 yrs vs 62 yrs, p⬍0.001), black (59% vs 38%, p⫽0.02), used drugs (28% vs 7%, p⫽0.001) and smoked (41% vs 11%, p⬍0.001). No difference was found in length of hospital/ICU stay, MEDS score or mortality; 13% HIV⫹ patients died in hospital vs 18% HIV⫺ patients (p⫽0.41). Abx-resistant pathogens were common overall (20% MRSA, 7% VRE). In HIV⫹ patients, 21/ 24(88%) pathogens cultured were gram pos, 3/24(12%) gram neg. In HIV⫺ patients 122/208(59%) pathogens cultured were gram pos, 83/208(40%) gram neg, and 3/ 208(1%) fungal (p⬍0.001). HIV⫺ patients had 3.2 times RR of infection by gram neg pathogen than HIV⫹ patients (95%CI, 1.1–9.5). In HIV⫹ patients, 19/24(79%) pathogens were adequately covered by initial ED abx and 5/24(21%) were not. In HIV⫺ patients, 148/208(71%) pathogens were adequately covered and 60/208(29%) were not. No difference was found in adequacy of abx coverage between the groups (p⫽0.41). Conclusion: We found no difference in adequacy of initial ED abx coverage, length of stay, or hospital mortality of HIV⫹ patients with severe sepsis compared to HIV⫺ patients. HIV⫹ patients had more gram pos infections and there was a high prevalence of abx resistant pathogens overall.
118
External Site Testing of an Instrument to Predict Endocarditis in Injection Drug Users With Fever
Romero K, Rodriguez R, Chiang W, Fortman J, Colucci A/University of California, San Francisco, San Francisco, CA; Bellevue Hospital, New York City, NY
Study Objective: To externally test a previously derived decision instrument (100% sensitivity and 44% specificity in prior study) for endocarditis prediction in injection drug users (IDUs) admitted from the ED with fever. Methods: Blinded to the prior instrument, an investigator used the same chart
Annals of Emergency Medicine S37