Prevalence of Acute Lung Injury Among Medical Patients in the Emergency Department

ORIGINAL RESEARCH CONTRIBUTION Prevalence of Acute Lung Injury Among Medical Patients in the Emergency Department Munish Goyal, MD, Daniel Houseman, MD, Nicholas J. Johnson, MD, Jason Christie, MD, Mark E. Mikkelsen, MD, MSCE, and David F. Gaieski, MD Abstract Background: Acute lung injury (ALI) affects an estimated 190,000 persons per year in U.S. intensive care units (ICUs), but little is known about its prevalence in the emergency department (ED). Objectives: The objective was to describe the prevalence of ALI among mechanically ventilated adult nontrauma patients in the ED. The hypothesis was that the prevalence of ALI in adult ED patients would be low. Methods: This was a retrospective cohort study of admitted nontrauma patients presenting to an academic ED. Two trained investigators abstracted data from patient records using a standardized form. The use of mechanical ventilation in the ED was identified in two phases. First, all ED patients were screened for the current procedural terminology (CPT) code for endotracheal intubation (CPT 31500) from January 1, 2003, to December 31, 2006. Second, each patient record was reviewed to verify the use of mechanical ventilation. ALI was defined in accordance with a modified version of the American-European Consensus Conference criteria as: 1) hypoxemia defined as PaO2 ⁄ FiO2 ratio £300 mm Hg on all arterial blood gases (ABGs) in the ED and the first 24 hours of admission, 2) the presence of bilateral infiltrates on chest radiograph, and 3) the absence of left atrial hypertension. Data are presented in absolute numbers and percentages. Interobserver agreement was evaluated using the kappa statistic. Results: Of the 552 patients who received mechanical ventilation in the ED and were subsequently admitted, a total of 134 (24.3%, 95% confidence interval [CI] = 20.8% to 28.0%) met hypoxemia criteria. Of these, 34 had evidence of left atrial hypertension, 52 did not have chest radiograph findings consistent with ALI, and two did not have a chest radiograph performed; the remaining 46 met ALI criteria. An additional two patients who died in the ED had clinical evidence of ALI. Thus, 48 of 552, or 8.7% (95% CI = 6.6% to 11.3%), met criteria for ALI. The kappa value for determination of ALI was 0.84 (95% CI = 0.54 to 1.0). Conclusions: The prevalence of ALI was nearly 9% in adult nontrauma patients receiving mechanical ventilation in the ED. Further study is required to determine which types of patients present to the ED with ALI, the extent to which lung protective ventilation is used, and the need for ED ventilator management algorithms. ACADEMIC EMERGENCY MEDICINE 2012; 19:1011–1018 ª 2012 by the Society for Academic Emergency Medicine From the Department of Emergency Medicine, Georgetown University Hospital, Washington Hospital Center (MG), Washington, DC; the Department of Emergency Medicine, St. Mary’s Medical Center (DH), Long Beach, CA; and the Department of Emergency Medicine (NJJ, DFG), the Center for Resuscitation Science (DFG), and the Department of Internal Medicine, Division of Pulmonary, Allergy, and Critical Care Medicine (JC, MEM), Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA. Received January 30, 2012; revision received April 6, 2012; accepted April 17, 2012. Presented at the Society of Critical Care Medicine Annual Congress, Houston, TX, February 2012. Dr. Christie is supported in part by an NHLBI RO1 grant and has consulted for Glaxo Smith Kline in the past. The rest of the authors have no disclosures or conflicts of interest to report. Supervising Editor: Daniel L. Theodoro, MD, MSCI. Address for correspondence and reprints: Nicholas J. Johnson, MD; e-mail: nicholas.johnson@uphs.upenn.edu. ª 2012 by the Society for Academic Emergency Medicine doi: 10.1111/j.1553-2712.2012.01429.x ISSN 1069-6563 PII ISSN 1069-6563583 1011 1012 Goyal et al. • ACUTE LUNG INJURY IN THE ED Prevalencia del Daño Pulmonar Agudo en Pacientes con Patología Médica en el Servicio de Urgencias Resumen Introducción: El daño pulmonar agudo (DPA) se estima que afecta a 190.000 personas por año en las unidades de cuidados intensivos de los Estados Unidos, pero su prevalencia es poco conocida en los servicios de urgencias (SU). Objetivos: Describir la prevalencia de DPA en los pacientes adultos no traumatológicos tratados con ventilación mecánica en el SU. La hipótesis fue que la prevalencia de DPA sería baja en los adultos del SU. Método: Es un estudio de cohorte retrospectivo de los pacientes no traumatológicos ingresados que acudieron a un SU universitario. Dos investigadores con formación previa obtuvieron los datos de las historias clínicas de los pacientes mediante un formulario estandarizado. El uso de ventilación mecánica en el SU se identificó en dos fases. En la primera, se realizó un despistaje en todos los pacientes del SU mediante el código de terminología actualizada de procedimientos médicos (TAP) para intubación endotraqueal (TAP 31500) desde el 1 de enero de 2003, hasta el 31 de diciembre de 2006. En la segunda, se revisó la historia clínica de cada paciente para verificar el uso de ventilación mecánica. El DPA se definió acorde a una versión modificada de los criterios de la Conferencia de Consenso Europea-Americana como: 1) hipoxemia definida como PaO2 ⁄ FiO2 £ 300 mmHg en la gasometría arterial en el SU y en las primeras 24 horas del ingreso, 2) presencia de infiltrados bilaterales en la radiografía de tórax, y 3) ausencia de hipertensión auricular izquierda. Los datos se presentaron en números absolutos y porcentajes. La concordancia interobservador se evaluó mediante el coeficiente kappa. Resultados: De los 552 pacientes que recibieron ventilación mecánica en el SU y que posteriormente ingresaron, un total de 134 (24,3%, IC 95% = 20,8%–28,0%) cumplió el criterio de hipoxemia. De éstos, 34 tenían hipertensión auricular izquierda, 52 no mostraban datos de DPA en la radiografía de tórax y a 2 no se les realizó una radiografía de tórax; los 46 restantes cumplieron los criterios de DPA. Otros dos pacientes que fallecieron en el SU mostraron evidencia clínica de DPA. Por tanto, 48 de 552 pacientes, esto es, un 8,7% (IC 95% = 6,6% a 11,3%), tuvieron criterios de DPA. El valor kappa para la determinación de DPA fue de 0,84 (IC 95% = 0,54 a 1,0). Conclusiones: La prevalencia de DPA fue casi del 9% en los pacientes adultos no traumatológicos que recibieron ventilación mecánica en el SU. Se requiere futuros estudios para determinar qué tipo de pacientes acuden al SU con DPA, el grado en que se usa ventilación protectora del pulmón y la necesidad de algoritmos para el manejo del ventilador en el SU. A cute lung injury (ALI), and its more severe form, the acute respiratory distress syndrome (ARDS), are characterized by diffuse inflammatory lung injury causing primarily nonhydrostatic pulmonary edema and hypoxemia. Diagnostic criteria include 1) acute onset; 2) presence of a predisposing condition; 3) bilateral infiltrates on frontal chest x-ray; 4) hypoxemia, defined as partial pressure of arterial oxygen (PaO2) to fractional inspired oxygen concentration (FiO2) ratio (P ⁄ F ratio) £300 mm Hg for ALI and £ 200 mm Hg for ARDS; and 5) pulmonary artery occlusion pressure < 18 mm Hg or no clinical evidence of left atrial hypertension.1 ARDS differs from ALI only in the degree of hypoxemia; henceforth these syndromes will be collectively referred to as ‘‘ALI’’ in this article. Recent population-based data estimate 190,000 cases and 74,000 deaths from ALI each year in the United States.2 The past decade has seen an increase in the understanding of the relationship between mechanical ventilation and pulmonary pathophysiology in ALI. A landmark study demonstrated a survival benefit in patients with ALI when low tidal volume ventilation strategies were employed.3 The use of such ‘‘lung protective’’ strategies, which minimize excessive tidal volumes and airway pressures, has become the standard of care in the intensive care unit (ICU) for patients with ALI. Numerous studies have described the epidemiology of ALI in ICU populations,2,4–9 but intensivists have also proposed that ALI occurs outside of the ICU.10,11 One study has attempted to prospectively identify ALI in patients with predisposing clinical conditions on regular hospital wards.12 Emergency physicians (EPs) often perform endotracheal intubation and initiate mechanical ventilation for the management of respiratory failure, yet a paucity of literature exists regarding mechanically ventilated patients in the emergency department (ED). Hospital crowding and ICU bed shortages often require EPs to ACADEMIC EMERGENCY MEDICINE • September 2012, Vol. 19, No. 9 • manage mechanically ventilated patients in the ED for extended periods of time.13 Studies have shown that increased ED length of stay correlates with poor outcomes in subsets of mechanically ventilated patients.14 The epidemiology of acute hypoxic respiratory failure, a component of ALI, has been described in a large pediatric ED population.15 Significant research over the past 25 years has been devoted to developing and evaluating tools to predict the development of ALI, but little is known about the baseline prevalence of ALI among adult patients presenting to the ED.16–22 The goal of this investigation was to describe the prevalence of ALI in nontrauma adult patients intubated in the ED. the institutional review board of the University of Pennsylvania. METHODS Study Design This was a retrospective chart review of all adult (age ‡18 years) nontrauma patients undergoing endotracheal intubation in a single ED between January 1, 2003, and December 31, 2006. This study was approved by www.aemj.org 1013 Study Setting and Population This investigation was conducted at The Hospital of the University of Pennsylvania (HUP). HUP is an urban, university-based, adult, tertiary referral center with 65,000 annual visits, a 24% admission rate, a Level I trauma center, and an emergency medicine residency program. We identified the use of mechanical ventilation in the ED in two phases. First, we screened all nontrauma ED patients for the current procedural terminology (CPT) code for endotracheal intubation (CPT 31500) during the specified time period. Second, we reviewed each patient record to verify the actual use of mechanical ventilation. We excluded trauma patients, as traumaassociated ALI differs from ALI due to an alternative predisposing condition, and outcomes are significantly better in trauma-associated ALI.23–27 In addition, the applicability of the 1994 American-European Consensus Conference (AECC) diagnostic criteria, which we All nontrauma ED patients with endotracheal intubation: 552 Expired in ED: 77 43 out-of-hospital cardiac arrest 24 ED cardiac arrest 8 misc. (no suggestion of ALI) 2 sepsis/multifocal pneumonia (treated as +ALI) No ABG performed, but no suggestion of ALI: 36 All pts with ABG in ED or in the first 24 hours of admission: 439 No ABGs with P/F ≤ 300 mm Hg: 236 At least one ABG in the ED or within 24 hours of admission with P/F ≤ 300 mm Hg but other ABG(s) with P/F > 300 mm Hg (transient hypoxemia): 69 All ABGs in ED and first 24 hours of admission with P/F ≤ 300 mm Hg (sustained hypoxemia): 134 + evidence of “left atrial hypertension”: 34 No evidence of “left atrial hypertension”: 100 No radiographic evidence of ALI on CXR: 52 No CXR available: 2 CXR consistent with ALI: 46 Patients who died in the ED or ICU in first 24 hours with findings concerning for ALI: 2 Nontrauma patients intubated in the ED with ALI: 48 Prevalence of ALI among intubated patients: 48/552 = 8.7% (95% CI = 6.6% to 11%) Figure 1. Patient selection flow diagram. ALI = acute lung injury; ABG = arterial blood gas; CXR = chest radiograph; ICU = intensive care unit; P ⁄ F = partial pressure of arterial oxygen (PaO2) to fractional inspired oxygen concentration (FiO2) ratio; pts = patients. 1014 Goyal et al. adapted to define cases of ALI, to trauma patients has been questioned.26 Study Protocol (Figure 1) A trained research assistant screened the records of all nontrauma patients who received endotracheal intubation in the ED during the study period. Patients were determined to have ALI if they met the 1994 AECC diagnostic criteria: 1) bilateral infiltrates on frontal chest x-ray, 2) P ⁄ F ratio £300 mm Hg, and 3) pulmonary artery occlusion pressure £18 mm Hg or ‘‘no clinical evidence of left atrial hypertension.’’1 Charts were initially screened to determine if patients met the hypoxemia criteria (P ⁄ F ratio £300 mm Hg) by reviewing all arterial blood gas (ABG) values during the ED stay and during the first 24 hours of hospital admission. To increase our specificity, only patients with all ABGs in the ED and the first day of admission with P ⁄ F ratio £ 300 mm Hg were considered to meet the hypoxemia criterion; we defined this as ‘‘sustained hypoxemia.’’ Although this is not a requirement in the AECC diagnostic criteria, this more rigorous requirement eliminates those patients who have an isolated ABG demonstrating hypoxemia, which can occur with periintubation atelectasis and other diseases. Prior studies have demonstrated that the median day of onset of ALI in the ICU is day 2 (interquartile range [IQR] = day 1 to day 4).17 We performed a sensitivity analysis of patients with transient hypoxemia (one or more ABGs with P ⁄ F ratio £300 mm Hg) to examine how the inclusion of these patients may have affected our results. Charts for patients meeting the hypoxemia criterion were then reviewed by a trained researcher to determine BNP >100 pg/mL No ALI Yes or unavailable CVP > 12 mm Hg No ALI Yes or unavailable Given furosemide or nitroglycerin No ALI Yes or unknown Charts independently reviewed by emergency physician for clinical evidence of heart failure. Figure 2. Criteria for heart failure or left atrial hypertension. ALI = acute lung injury; BNP = brain natriuretic peptide; CVP = central venous pressure. • ACUTE LUNG INJURY IN THE ED if there was any evidence of heart failure or ‘‘left atrial hypertension’’ using the algorithm depicted in Figure 2, which was developed by consensus among the authors. Data were recorded using a standardized form. Previous literature demonstrates that a brain natriuretic peptide (BNP) level of £200 to 250 pg ⁄ mL supports the diagnosis of ALI with a specificity of approximately 90%.28,29 One study showed improved specificity (95%) with a BNP level of <100 pg ⁄ mL, so this level was selected as our cutoff.30 Of note, the utility of BNP in the diagnosis of ALI has recently been questioned, and a normal value does not completely exclude cardiac dysfunction.28,30 Charts of patients not meeting heart failure ⁄ left atrial hypertension or ALI criteria were reviewed independently by an EP (DH) for clinical evidence of heart failure, which we defined as suspicion documented by the treating physician in the ED that congestive heart failure or hydrostatic pulmonary edema were the primary etiologies of the patient’s current symptoms. Those whose charts contained evidence of heart failure were excluded from further analysis. Because of inherent difficulty in determining left atrial hypertension on retrospective chart review and the limitations associated with BNP and because prior studies have shown that as many as one-third of patients with ALI will have high central venous pressure (CVP), we performed a sensitivity analysis to see how the inclusion of patients classified as having left atrial hypertension might affect our results.31 The radiologists’ reports of all ED chest radiographs of patients who met hypoxemia criteria and did not have evidence of heart failure were reviewed by a researcher (DH). Patients whose chest radiographs were read as normal or ‘‘no acute disease’’ were excluded from further analysis. The remaining chest radiograph radiology reports were reviewed by EPs (DG, MG) involved in routine clinical care. Those with bilateral infiltrates consistent with ALI were included. Patients who had all ABGs in the ED and in the first 24 hours after admission with P ⁄ F ratio of £300 mm Hg, no evidence of left atrial hypertension, and bilateral infiltrates on chest x-ray were considered to have ALI. Charts for patients who died in the ED and those who did not have an ABG performed in the first 24 hours were reviewed separately by EPs (DG, MG). Patients with a cause of death likely unrelated to ALI (e.g., out-of-hospital cardiac arrest or acute stroke) or reason for intubation likely unrelated to ALI (e.g., intoxication, seizure, or angioedema) were considered not to have ALI. Patients whose cause of death or reason for intubation might have been ALI (e.g., sepsis, multifocal pneumonia) underwent further review. First, the hypoxemia criteria were applied if ABGs were available. Second, the left atrial hypertension ⁄ congestive heart failure algorithm was applied if those data were available. If they were unavailable, these charts underwent consensus review for determination of ALI. To assess accuracy of chart extraction, we used a random number generator to select a sample of 15 charts (5% of total). A second trained reviewer (an emergency medicine resident [NJ]) abstracted data from these charts using a standardized form and the methodology described above while blinded of the ACADEMIC EMERGENCY MEDICINE • September 2012, Vol. 19, No. 9 • others’ results. Kappa values were calculated to determine agreement on manually abstracted variables (Kappa Calculator, http://faculty.vassar.edu/lowry/kappa.html). of 552 patients. The majority (105 of 134; 78%, 95% CI = 71% to 84%) of patients with sustained hypoxemia had hypoxemia on initial ABGs performed in the ED. The remaining patients had their initial ABGs performed in the ICU. The mean (±SD) age of 134 patients meeting hypoxemia criteria was 58.6 (±16.0) years, 68 were male (51%, 95% CI = 42% to 59%), 59% were African American, 28% were white, and 13% were other. Of these 134 patients, 97 were Emergency Severity Index (ESI) triage class 1–Emergent (72%, 95% CI = 64% to 79%), 36 were ESI triage class 2 (27%, 95% CI = 20% to 35%), and one was ESI triage class 3 (0.8%, 95% CI = 0.1% to 4%). The most common ED chief complaints were shortness of breath (37%, 95% CI = 30% to 46%), change in mental status (16%, 95% CI = 11% to 24%), and cerebral vascular accident (9%, 95% CI = 5.2% to 15%). Severe sepsis or septic shock as one of the primary ED diagnoses was documented in the notes for 28 (21%, 95% CI = 15% to 29%) of these patients. Data Analysis Microsoft Access and Excel (Microsoft Corp., Redmond, WA) were used to create a database that included all patients who were identified in the ED electronic medical record database search. Means and 95% confidence intervals (CIs) were calculated for normally distributed continuous variables. RESULTS Over 4 years, 576 nontrauma ED patients had the CPT code (CPT 31500) for endotracheal intubation. During chart review, it was found that 15 of these patients did not undergo endotracheal intubation in the ED; a total of eight patients were intubated in the prehospital setting, and one was a trauma patient managed in the ED. These 24 patients were excluded from further analysis, leaving 552 medical patients who were intubated in the ED to form our study population (Figure 1). The mean (±standard deviation [SD]) age of the cohort was 62.5 (±24.0) years and 48% were male. A total of 77 patients (14% of the total, 95% CI = 11% to 17%) expired in the ED. Of these patients, 43 (56%, 95% CI = 45% to 56%) suffered out-of-hospital cardiac arrest, 24 (31%, 95% CI = 22% to 42%) suffered cardiac arrest in the ED and had endotracheal intubation as part of their resuscitation, eight (10%, 95% CI = 5.3% to 19%) were intubated for other reasons (including intracranial hemorrhage ⁄ stroke, gastrointestinal bleeding, angioedema ⁄ allergic reaction, mental status change, and seizure) but had no suggestion of ALI, and two patients (2.6%, 95% CI = 0.7% to 9.0%) who died were intubated in the setting of fulminant sepsis. The two patients with sepsis had multifocal pneumonia and hypoxemia and died peri-intubation. We assumed that these patients had ALI and included them in the ALI positive group. Of the 475 remaining patients, 36 (7.6%, 95% CI = 5.5% to 10%) never had ABGs performed during their hospital stays. These patients had varied reasons for intubation (intracranial hemorrhage ⁄ stroke, alcohol or drug intoxication, mental status change, gastrointestinal bleeding, or angioedema ⁄ allergic reaction), but none had bilateral infiltrates on chest radiograph suggestive of ALI in their first 24 hours of hospitalization. Patients Meeting Hypoxemia Criteria We included patients with sustained hypoxemia, defined as all ABGs performed in the ED and the first 24 hours of hospitalization with P ⁄ F ratio £300 mm Hg. Of the 439 patients who had an ABG performed in the ED or the first 24 hours of admission, 236 (54%, 95% CI = 49% to 58%) had no ABGs with P ⁄ F ratio £300 mm Hg. A total of 69 patients (16%, 95% CI = 13% to 19%) had transient hypoxemia, defined as hypoxemia on an initial ABG that subsequently improved. Thus, 134 patients met hypoxemia criteria and had a P ⁄ F ratio £300 mm Hg on all ABGs performed in the ED or the first 24 hours of admission. These 134 patients comprised 24% (95% CI = 21% to 28%) of the initial cohort www.aemj.org 1015 Patients Meeting ALI Criteria Figure 1 shows the algorithm that was used to identify patients who met criteria for ALI. Of the 552 patients who were included for analysis, 134 (24%, 95% CI = 21% to 28%) met hypoxemia criteria. Of these, 34 of 134 (25%, 95% CI = 19% to 33%) had evidence of left atrial hypertension or heart failure as the etiology of respiratory failure. Of the patients meeting criteria for left atrial hypertension, four of 34 (12%, 95% CI = 4.6% to 26%) qualified due to elevated BNP, three of 34 (9% 95% CI = 3% to 23%) due to furosemide administration, six of 34 (18%, 95% CI = 8.3% to 34%) due to nitroglycerin administration, six of 34 (18%, 95% CI = 8.3% to 34%) due to administration of both furosemide and nitroglycerin, two of 34 (5.8%, 95% CI = 1.6% to 19%) due to both elevated BNP and furosemide administration, and three of 34 (9% 95% CI = 3% to 23%) due to elevated BNP and administration of both nitroglycerin and furosemide. The remaining 10 of 34 patients (29%, 95% CI = 17% to 46%) qualified based on suspicion of the treating EP documented on chart review. Of the remaining 100 patients, 52 did not have chest radiograph findings consistent with ALI, and two patients did not have chest radiographs performed. An additional two patients who died in the ED or ICU within the first 24 hours had clinical histories and chest radiograph findings concerning for ALI. Thus, of 552 patients who were intubated in the ED, 48 or 8.7% (95% CI = 6.5% to 11%) met criteria for ALI. The kappa value for determination of ALI was 0.84 (95% CI = 0.54 to 1.0) with agreement between reviewers on 14 of 15 charts independently classified. Sensitivity Analyses Two sensitivity analyses were performed to determine the effects on the prevalence of ALI after including patients with transient hypoxemia and left atrial hypertension. The first sensitivity analysis examined the effect of adding patients with transient hypoxemia to the total with ALI. Of the 69 patients with transient hypoxemia, 19 (28%, 95% CI = 18% to 39%) met criteria for left atrial hypertension and 28 (41%, 95% CI = 30% to 52%) 1016 lacked bilateral infiltrates on chest radiography consistent with ALI. This left 22 patients (32% of the patients with transient hypoxemia, 95% CI = 22% to 44%; and 3.9% of the total cohort of 552 patients, 95% CI = 2.7% to 6.0%) who were added to the 48 patients who previously met ALI criteria, resulting in a prevalence of ALI of 13% (95% CI = 10% to16%). The second sensitivity analysis examined the effect of adding patients classified as having left atrial hypertension or congestive heart failure to the total with ALI. Of the 34 patients classified as having left atrial hypertension or congestive heart failure, five (15%, 95% CI = 6.5% to 30%) lacked bilateral infiltrates consistent with ALI, leaving 29 patients (85%, 95% CI = 70% to 94%) for inclusion in the sensitivity analysis. Including these patients as having ALI would result in a prevalence of 14% (95% CI = 11% to 17%). Including both patients with transient hypoxemia and those classified as having left atrial hypertension would result in an additional 118 patients being added to the 48 patients who previously met ALI criteria, resulting in a prevalence of 21% (95% CI = 18% to 25%). DISCUSSION This study demonstrates that a small but substantive percentage of adult nontrauma patients who undergo mechanical ventilation in the ED meet criteria for ALI during the ED presentation or soon thereafter. The presence of hypoxemia on an initial ED ABG signals the potential for sustained hypoxemia and the development of ALI. These patients are older (approximately 60 years old) and mostly minorities, present critically ill, are most frequently triaged as emergent, and have a variety of chief complaints including shortness of breath, change in mental status, and neurologic emergencies. Further, 21% were assigned a diagnosis of severe sepsis or septic shock as one of their primary ED diagnoses. Estimates of the incidence of ALI vary widely. Early figures ranged from 1.5 to 75 cases per 100,000 persons. The development of the consensus definitions in 1994 reduced this variability somewhat.1 A large, multicenter prospective report in the United States estimated an incidence of ALI between 22 and 64 per 100,000 population per year.8 A more recent prospective, population-based cohort study in a single U.S. county found the incidence of ALI to be higher (78.9 per 100,000 persons).9 The majority of epidemiologic studies on patients with ALI include only those who have been admitted to an ICU; little is known about the baseline prevalence among patients presenting to the ED. A number of studies have attempted to identify patients at risk for the development of ALI.20,22,32,33 Ferguson et al.12 found that ALI occurred most commonly in the setting of sepsis, followed by undifferentiated shock and pneumonia. It is known that mechanical ventilation is a risk factor for the development of ALI. In a study of 332 ICU patients who did not have ALI at the onset of mechanical ventilation, nearly 25% developed ALI within 5 days.7 Gajic et al.17 derived a prediction score based on clinical variables to estimate the risk of ALI among at-risk patients presenting to the ED Goyal et al. • ACUTE LUNG INJURY IN THE ED or admitted for elective surgery. Of the 5,992 patients evaluated in this study with at least one predisposing condition for ALI (including sepsis, shock, pneumonia, aspiration, or high-risk trauma or surgery), 166 (3%) were excluded for already having ALI at the time of ED evaluation or upon hospital admission. Although this population differs from ours in that it included both intubated and nonintubated patients admitted for surgery and it required a predisposing condition, this may represent the only other estimate of the ED prevalence of ALI in the medical literature. Future research might identify risk factors for the development of ALI in ED patients. It is important for clinicians to understand the prevalence of ALI in the ED population for a number of reasons. First, ALI is an important cause of hypoxemia, and especially refractory hypoxemia, in mechanically ventilated patients. Second, a number of guidelines and experts recommend that lung protective strategies be used for all mechanically ventilated patients in the ED.34–39 While low-tidal-volume ventilation has been demonstrated to reduce mortality in patients with ALI, it has never been definitively shown to prevent or reduce morbidity or mortality in patients without ALI.3,40 There is, however, a causal link between high tidal volumes and the development of ALI.41 The low prevalence of ALI among ED patients in our study highlights the need for more data describing how lowtidal-volume ventilation might affect the 91% of ED patients who do not meet criteria for ALI at or near the onset of mechanical ventilation. We sought to describe the frequency with which ALI is encountered in one ED. By describing the epidemiology of ALI in this setting, we hope that EPs will recognize this important cause of acute hypoxemic respiratory failure, but realize that it is rarely present in ED patients at the time of endotracheal intubation. LIMITATIONS A total of 77 patients died in the ED. While we screened these charts to identify potential cases of ALI, it is possible that some were missed, as data were limited on these patients. Additionally, 36 patients never had ABGs performed. We reviewed these charts and determined that the vast majority of these patients were intubated for reasons other than respiratory failure and did not have chest radiograph findings suggestive of ALI, but it is possible that some cases of ALI were missed. In order to improve the specificity of the definition of ALI, we considered patients to meet the hypoxemia criterion for ALI if all ABGs in the ED and within the first 24 hours after admission had a P ⁄ F ratio £300 mm Hg. By ensuring that hypoxemia was not an isolated initial finding, we believe that the specificity of this criterion is improved. We may, however, underestimate the true frequency of disease. By extending the window to 24 hours, we also may not have the true prevalence of ALI in the ED, but rather the prevalence of ALI in the first 24 hours of hospital stay. The majority (78%) of our patients with sustained hypoxemia, however, had hypoxemia on initial ABGs performed in the ED; the remainder did not have ABGs performed until ACADEMIC EMERGENCY MEDICINE • September 2012, Vol. 19, No. 9 after ICU admission. The retrospective diagnosis of left atrial hypertension based on chart review represents another challenge, but we attempted to overcome this by using objective criteria in a rigorous algorithm (Figure 2). As mentioned previously, the utility of BNP in the diagnosis of ALI has recently been questioned, and a normal value does not completely exclude cardiac dysfunction.28,30 In addition, CVP is a proxy for right atrial pressure and may not accurately reflect volume status or left-sided pressures. It is also possible that patients with bilateral infiltrates on chest radiograph caused both by ALI and left atrial hypertension were missed. There are several other limitations to this study. As a single-center study in an urban teaching hospital, it may have limited applicability to other populations. Also, we only captured nontrauma patients who were intubated in the ED and did not include patients intubated in the prehospital setting, unless they were reintubated in the ED or intubated shortly after transfer from the ED to the ICU or other inpatient setting. Thus, this study does not capture the true frequency of ALI of all intubated patients in the ED, but rather the frequency of nontrauma patients who undergo intubation in the ED. We were unable to comment on risk factors associated with the presence of ALI in the ED, ventilator strategies employed in the ED, and what proportion of patients went on to develop ALI. Despite these limitations, we believe that these data are an important first step in benchmarking the epidemiology of ALI in the ED. • www.aemj.org 5. 6. 7. 8. 9. 10. 11. 12. 13. CONCLUSIONS In this single-center study, nearly 9% of nontrauma adults intubated in the ED met criteria for ALI. Based on our sensitivity analyses, the prevalence may be as high as 21%. These data could have implications for the management of ED patients who require endotracheal intubation and mechanical ventilation. Further study is required to determine which types of patients present to the ED with acute lung injury and the extent to which lung protective ventilation is used. References 1. Bernard GR, Artigas A, Brigham KL, et al. Report of the American-European consensus conference on ARDS: definitions, mechanisms, relevant outcomes and clinical trial coordination. The Consensus Committee. Intensive Care Med. 1994; 20: 225–32. 2. Rubenfeld GD, Herridge MS. Epidemiology and outcomes of acute lung injury. Chest. 2007; 131: 554–62. 3. The Acute Respiratory Distress Syndrome Network. Ventilation with lower tidal volumes as compared with traditional tidal volumes for acute lung injury and the acute respiratory distress syndrome. N Engl J Med. 2000; 342:1301–8. 4. Bersten AD, Edibam C, Hunt T, Moran J, Australian and New Zealand Instensive Care Society Clinical Trials Group. Incidence and mortality of acute lung injury and the acute respiratory distress syndrome 14. 15. 16. 17. 18. 19. 20. 21. 1017 in three Australian States. Am J Respir Crit Care Med. 2002; 165:443–8. Baumann WR, Jung RC, Koss M, Boylen CT, Navarro L, Sharma OP. Incidence and mortality of adult respiratory distress syndrome: a prospective analysis from a large metropolitan hospital. Crit Care Med. 1986; 14:1–4. Bice T, Li G, Malinchoc M, Lee AS, Gajic O. Incidence and risk factors of recurrent acute lung injury. Crit Care Med. 2011; 39:1069–73. Gajic O, Dara SI, Mendez JL, et al. Ventilator-associated lung injury in patients without acute lung injury at the onset of mechanical ventilation. Crit Care Med. 2004; 32:1817–24. Goss CH, Brower RG, Hudson LD, Rubenfeld GD; ARDS Network. Incidence of acute lung injury in the United States. Crit Care Med. 2003; 31:1607–11. Rubenfeld GD, Caldwell E, Peabody E, et al. Incidence and outcomes of acute lung injury. N Engl J Med. 2005; 353:1685–93. Quartin AA, H Schein RM. Acute lung injury: is the intensive care unit the tip of the iceberg? Crit Care Med. 2003; 31:1860–1. Finney SJ, Evans TW. Acute lung injury outside the ICU: a significant problem. Crit Care. 2007; 11:169. Ferguson ND, Frutos-Vivar F, Esteban A, et al. Clinical risk conditions for acute lung injury in the intensive care unit and hospital ward: a prospective observational study. Crit Care. 2007; 11:R96. Carr BG, Hollander JE, Baxt WG, Datner EM, Pines JM. Trends in boarding of admitted patients in US emergency departments 2003–2005. J Emerg Med. 2009; 39:506–11. Carr BG, Kaye AJ, Wiebe DJ, Gracias VH, Schwab CW, Reilly PM. Emergency department length of stay: a major risk factor for pneumonia in intubated blunt trauma patients. J Trauma. 2007; 63:9–12. Freishtat RJ, Mojgani B, Mathison DJ, Chamberlain JM. Toward early identification of acute lung injury in the emergency department. J Investig Med. 2007; 55:423–9. Gajic O, Afessa B, Thompson BT, et al. Prediction of death and prolonged mechanical ventilation in acute lung injury. Crit Care. 2007; 11:R53. Gajic O, Dabbagh O, Park PK, et al. Early identification of patients at risk of acute lung injury: evaluation of lung injury prediction score in a multicenter cohort study. Am J Respir Crit Care Med. 2010; 183:462–70. Hudson LD, Milberg JA, Anardi D, Maunder RJ. Clinical risks for development of the acute respiratory distress syndrome. Am J Respir Crit Care Med. 1995; 151(2 Pt 1):293–301. Ahmed A, Kojicic M, Herasevich V, Gajic O. Early identification of patients with or at risk of acute lung injury. Neth J Med. 2009; 67:268–71. Pepe PE, Potkin RT, Reus DH, Hudson LD, Carrico JC. Clinical predictors of the adult respiratory distress syndrome. Am J Surg. 1982; 144:124–30. Maunder RJ, Harlan JM, Pepe PE, Paskell S, Carrico CJ, Hudson LD. Measurement of plasma fibronectin in patients who develop the adult respiratory 1018 22. 23. 24. 25. 26. 27. 28. 29. 30. distress syndrome. J Lab Clin Med. 1984; 104: 583–90. Trillo-Alvarez C, Cartin-Ceba R, Kor DJ, et al. Acute lung injury prediction score: derivation and validation in a population-based sample. Eur Respir J. 2011; 37:604–9. Calfee CS, Eisner MD, Ware LB, et al. Trauma-associated lung injury differs clinically and biologically from acute lung injury due to other clinical disorders. Crit Care Med. 2007; 35:2243–50. Miller PR, Croce MA, Bee TK, et al. ARDS after pulmonary contusion: accurate measurement of contusion volume identifies high-risk patients. J Trauma. 2001; 51:223–8. Croce MA, Fabian TC, Davis KA, Gavin TJ. Early and late acute respiratory distress syndrome: two distinct clinical entities. J Trauma. 1999; 46:361–6. Dicker RA, Morabito DJ, Pittet JF, Campbell AR, Mackersie RC. Acute respiratory distress syndrome criteria in trauma patients: why the definitions do not work. J Trauma. 2004; 57:522–6. Silverboard H, Aisiku I, Martin GS, Adams M, Rozycki G, Moss M. The role of acute blood transfusion in the development of acute respiratory distress syndrome in patients with severe trauma. J Trauma. 2005; 59:717–23. Rana R, Vlahakis NE, Daniels CE, et al. B-type natriuretic peptide in the assessment of acute lung injury and cardiogenic pulmonary edema. Crit Care Med. 2006; 34:1941–6. Karmpaliotis D, Kirtane AJ, Ruisi CP, et al. Diagnostic and prognostic utility of brain natriuretic peptide in subjects admitted to the ICU with hypoxic respiratory failure due to noncardiogenic and cardiogenic pulmonary edema. Chest. 2007; 131:964–71. Levitt JE, Vinayak AG, Gehlbach BK, et al. Diagnostic utility of B-type natriuretic peptide in critically ill patients with pulmonary edema: a prospective cohort study. Crit Care. 2008; 12:R3. Goyal et al. • ACUTE LUNG INJURY IN THE ED 31. Wheeler AP, Bernard GR, Thompson BT, et al. Pulmonary-artery versus central venous catheter to guide treatment of acute lung injury. N Engl J Med. 2006; 354:2213–24. 32. Pepe PE, Thomas RG, Stager MA, Hudson LD, Carrico CJ. Early prediction of the adult respiratory distress syndrome by a simple scoring method. Ann Emerg Med. 1983; 12:749–55. 33. Iscimen R, Cartin-Ceba R, Yilmaz M, et al. Risk factors for the development of acute lung injury in patients with septic shock: an observational cohort study. Crit Care Med. 2008; 36:1518–22. 34. Amato MB, Barbas CS, Medeiros DM, et al. Effect of a protective-ventilation strategy on mortality in the acute respiratory distress syndrome. N Engl J Med. 1998; 338:347–54. 35. Hickling KG, Walsh J, Henderson S, Jackson R. Low mortality rate in adult respiratory distress syndrome using low-volume, pressure-limited ventilation with permissive hypercapnia: a prospective study. Crit Care Med. 1994; 22:1568–78. 36. Ware LB, Matthay MA. The acute respiratory distress syndrome. N Engl J Med. 2000; 342:1334–49. 37. Brower RG, Ware LB, Berthiaume Y, Matthay MA. Treatment of ARDS. Chest. 2001; 120:1347–67. 38. Dellinger RP, Levy MM, Carlet JM, et al. Surviving Sepsis Campaign: international guidelines for management of severe sepsis and septic shock: 2008. Crit Care Med. 2008; 36:296–327. 39. Wood S, Winters ME. Care of the intubated emergency department patient. J Emerg Med. 2010; 40:419–27. 40. Yilmaz M, Gajic O. Optimal ventilator settings in acute lung injury and acute respiratory distress syndrome. Eur J Anaesthesiol. 2008; 25:89–96. 41. Schultz MJ. Lung-protective mechanical ventilation with lower tidal volumes in patients not suffering from acute lung injury: a review of clinical studies. Med Sci Monit. 2008; 14:RA22–6.