American Journal of Emergency Medicine (2012) xx, xxxxxx

www.elsevier.com/locate/ajem

Original Contribution

Measuring cardiac index with a focused cardiac ultrasound examination in the ED,
Vi Am Dinh MD a , H. Samuel Ko MD, MBA a , Rajiv Rao BS d , Ramesh C. Bansal MD, FASE c , Dustin D. Smith MD a , Tae Eung Kim MD a , H. Bryant Nguyen MD, MS a,b,
a

Department of Emergency Medicine, Loma Linda University, Loma Linda, CA 92354, USA Department of Medicine, Division of Pulmonary and Critical Care, Loma Linda University, Loma Linda, CA 92354, USA c Department of Medicine, Division of Cardiology, Loma Linda University, Loma Linda, CA 92354, USA d School of Medicine, Loma Linda University, Loma Linda, CA 92354, USA
b

Received 28 February 2012; revised 26 March 2012; accepted 28 March 2012

Abstract Objectives: Noninvasive technology may assist the emergency department (ED) physician in determining the hemodynamic status in critically ill patients. The objective of our study was to show that ED physicians can accurately measure cardiac index (CI) by performing a bedside focused cardiac ultrasound examination. Methods: A convenience sample of adult subjects were prospectively enrolled. Cardiac index, left ventricular outflow tract (LVOT) diameter, velocity time integral (VTI), stroke volume index, and heart rate were obtained by trained ED physicians and a certified cardiac sonographer. The primary outcome was percent of optimal LVOT diameter and VTI measurements as verified by an expert cardiologist. Results: One hundred patients were enrolled, with obtainable CI measurements in 97 patients. Cardiac index, LVOT diameter, VTI, stroke volume index, and heart rate measurements by ED physician were 2.42 0.70 L min1 m2, 2.07 0.22 cm, 18.30 3.71 cm, 32.34 7.92 mL beat1 m2, and 75.32 13.45 beats/min, respectively. Measurements of LVOT diameter by ED physicians and sonographer were optimal in 90.0% (95% confidence interval, 82.6%-94.5) and 91.3% (73.2%-97.6%) of patients, respectively. Optimal VTI measurements were obtained in 78.4% (69.2%-85.4%) and 78.3% (58.1%90.3%) of patients, respectively. In 23 patients, the correlation (r) for CI between ED physician and sonographer was 0.82 (0.60-0.92), with bias and limits of agreement of 0.11 (1.06 to 0.83) L min1 m2 and percent difference of 12.4% 10.1%. Conclusions: Emergency department ED physicians can accurately measure CI using standard bedside ultrasound. A focused ultrasound cardiac examination to derive CI has potential use in the management of critical ill patients in the ED. 2012 Published by Elsevier Inc.
Funding source: This study was partially funded by the Emergency Medicine Research and Education Foundation. Presentations: This study was presented at the American College of Emergency Physicians, Scientic Assembly, October 2011, San Francisco, CA. Corresponding author. E-mail address: hbnguyen@llu.edu (H.B. Nguyen). 0735-6757/$ see front matter 2012 Published by Elsevier Inc. doi:10.1016/j.ajem.2012.03.025

1. Introduction
Hemodynamic monitoring can facilitate the resuscitation of critically ill patients in the emergency department (ED) [1]. Clinically, however, ED physicians are often inaccurate

2 in determining the hemodynamic status of acutely ill patients [2]. The use of advanced hemodynamic monitoring can change physician assessment of patient hemodynamics and treatment [3]. In applying any hemodynamic monitoring technique, cardiac index (CI) is considered the reference standard parameter for targeting organ perfusion and oxygen delivery in shock [4]. Insertion of a pulmonary artery catheter (Swan-Ganz) is traditionally required to calculate CI by thermodilution [5]. Besides being invasive, potential complications of pulmonary artery catheter placement include infection, arterial puncture, arrhythmias, pneumothorax, hemothorax, and pulmonary artery rupture [6,7]. Recently, there has been an increased interest in the development of noninvasive to minimally invasive technologies to measure CI, such as the use of pulse contour analysis, lithium dye dilution, electrical bioimpedance, and transesophageal and transthoracic echocardiography with pulsed or continuous wave Doppler ultrasound [8]. Many of these technologies are expensive and unavailable in most EDs. However, bedside transthoracic echocardiography with pulsed-wave Doppler ultrasound, or focused cardiac ultrasound, is noninvasive, commonly available, and emerging as a promising noninvasive technique for measuring hemodynamic parameters [9,10]. Emergency department physicians are able to effectively perform bedside focused cardiac ultrasound to assess pericardial effusion, left ventricular function, and cardiac standstill [11-13]. A review of the cardiology literature shows that echocardiography is a reliable method for measuring CI [14-17]. To our knowledge, no previous study has examined the ability of ED physicians in measuring CI using echocardiography. The objective of our study was to determine the accuracy of CI measurements by ED physicians performing a bedside focused cardiac ultrasound examination with technology available in the ED.

V.A. Dinh et al. CA). Training of each ED physician included 20 total hours of hands-on instructions by a certied cardiac sonographer. Training involved acquiring adequate images with parasternal long-axis and apical 5-chamber views. The ED physicians also learned how to use the cardiac software on the Z. One Ultra system to measure left ventricular outow tract (LVOT) diameter and velocity time integral (VTI). These are necessary parameters to calculate CI (see later).

2.3. Patient selection

A convenience sample of ED patients older than 18 years were enrolled. Enrollment occurred approximately 5 days per month when the study team was available. Patients were approached by an investigator, and the purpose of the study was explained. An informed consent was obtained, with written consent provided by a legal representative for patients unable to give consent, such as those with altered mental status, comatose, or sedated on mechanical ventilation. Patients with a medical history of congenital heart disease, aortic valvular disease, inability to lie supine, or inability to lie in the left lateral decubitus position throughout the duration of the study procedure were excluded.

2.4. Study protocol and data measurements

After patient enrollment, physiologic and clinical data were recorded, including weight, height, and chief concern. A set of hemodynamic parameters was then obtained by a trained ED physician using cardiac ultrasound. Parameters included were LVOT diameter, VTI, and heart rate (HR). For a selected number of patients, the sonographer obtained the same set of hemodynamic parameters directly following and

Table 1 Rating of ultrasound images to measure LVOT diameter and VTI: optimal, suboptimal, and unobtainable

2. Methods
2.1. Study design and setting
This was a prospective observational cohort study performed during a 5-month period, February 1 to June 30, 2011, at an academic ED with approximately 63 000 annual patient visits. The study was approved by the institutional review board at our institution and was considered to present minimal risk to the subjects.

2.2. Training
Before study enrollment, 2 ED physicians (V.A.D., H.S.K.) were trained to measure CI using the ultrasound system Z. One Ultra, equipped with the 4-MHz phased-array probe P4-1c (Zonare Medical Systems, Inc, Mountainview,

Horizontally lined image with clear denition of anterior and posterior limits of LVOT with correct placement of the caliper points Suboptimal Image may be angulated or anterior, and LVOT posterior limits of LVOT are not completely clear. However, measurements were still usable. Unobtainable Completely unobtainable images of parasternal LVOT long-axis view with unmeasurable LVOT Optimal Clear apical 5-chamber view with proper VTI alignment of the pulse Doppler sample volume parallel to the aortic ow with minimal spectral broadening Suboptimal Pulse Doppler sample volume not completely VTI parallel to ow but within 20-30 or with excessive spectral broadening Unobtainable Completely unobtainable images of apical 5VTI chamber view or extremely large pulsed-wave Doppler angle to ow Optimal LVOT

Cardiac index with focused cardiac ultrasound blinded to the ED physician measurements. All ultrasound images obtained by the ED physician and sonographer were stored and then veried by a cardiologist (R.C.B.) using a scale to rate the acceptability of the ultrasound measurements (Table 1).

Pulsed Wave Doppler Cursor

2.5. Cardiac index calculation by focused cardiac ultrasound

The 2 parameters required to calculate CI are the LVOT diameter and the VTI. The LVOT diameter is the diameter of the aortic outow tract, which can be calculated by obtaining a parasternal long-axis view (Fig. 1). The measurement is obtained by measuring the distance from the inner edge to inner edge, where the right aortic valve coronary cusp meets the interventricular septum to where the noncoronary cusp meets the anterior mitral valve leaet, in a line parallel to the aortic annulus. Velocity time integral is an estimation of the distance that a column of blood travels in 1 systolic stroke, or stroke distance. Using ultrasound, the VTI can be measured by obtaining an apical 5-chamber view and then placing a pulsed-wave Doppler cursor near the aortic valve annulus (Fig. 2). The Doppler signal is then traced using the cardiac software to calculate VTI. After obtaining the LVOT diameter and VTI, CI can be calculated from the following equation: CI = stroke volume index (SVI) HR, where SVI = stroke volume (SV)/body surface area, and SV = LVOT area stroke distance, or SV = (LVOT diameter/2)2 VTI. Heart rate is calculated by the ultrasound cardiac software during the VTI measurements, rather than from physical examination or bedside telemetry monitor.

Fig. 2 Measurement of VTI using the apical 5-chamber view. This view is obtained with the apical 4-chamber view and then angulating the probe slightly toward the patient's head until the LVOT can be visualized. The pulsed-wave Doppler cursor is placed at the LVOT, and the doppler signal is traced (dotted lines) to calculate the VTI.

2.6. Outcome
The primary outcome measured was percentage of optimal LVOT diameter and VTI measurements as veried by an expert cardiologist to conrm the technical adequacy of the calculated CI measurements. Table 1 illustrates our ratings of optimal, suboptimal, and unobtainable LVOT diameter and VTI measurements. The secondary outcome was interrater variability of CI between the ED physician and sonographer.

2.7. Statistical analysis

Descriptive analysis was performed for all subjects. The interrater reliability of ultrasound measurements by the emergency physicians and sonographer was calculated using percent difference, , Pearson correlation, and Bland-Altman analysis. For determining , normal CI was dened as 2.5 to 4.0 L min1 m2, with values outside this range being abnormal. Data are presented as mean SD or with 95% condence intervals. All analyses were performed using R-statistics (v2.13.1, R Development Core Team, R Foundation, Vienna, Austria).
1.98cm

Fig. 1 Measurement of LVOT diameter at aortic valve cusps. This is a parasternal long-axis view on zoom mode with calipers measuring the LVOT diameter at 1.98 cm. This view is obtained by placing the probe at the fourth intercostal space lateral to the left sternum.

3. Results
One hundred patients were enrolled in the study, with ages of 51 20 years and 50% being female. The most

4
Table 2 Patient characteristics 100 51 20 50:50 170 11 81 45 17.7 5.9 1.92 0.28 30, 30.0 (21.9-39.6) 18, 18.0 (11.7-26.7) 7, 7.0 (3.4-13.8) 9, 9.0 (4.8-16.2) 3, 3.0 (1.0-8.5) 16, 16.0 (10.1-24.4) 8, 8.0 (4.1-15.0) 3, 3.0 (1.0-8.5) 6, 6.0 (2.8-12.5)

V.A. Dinh et al. beat1 m2, and 75.32 13.45 beats/min, respectively (Table 4). The percent difference between ED physician and sonographer for CI measurements was 12.4% 10.1%, with = 0.56 and correlation coefcient of r = 0.82 (P b .01; Fig. 3). The bias and limits of agreement for CI between ED physician and sonographer was 0.11 (1.06 to 0.83) L min1 m2 (Fig. 4).

Total no. of subjects Age (y) Male/Female, no. Height (cm) Weight (kg) BMI (kg/m2) BSA (m2) Diagnostic category, no, % (95% CI) Cardiac Gastrointestinal Respiratory Neurologic Endocrine Musculoskeletal Infectious Hematologic Other

4. Discussion
Our study shows that CI can be measured by ED physicians using standard available ultrasound technology. After a training period of 20 hours, there was no statistical difference between the ability of the ED physicians to obtain LVOT diameter and VTI measurements as compared with a certied cardiac sonographer. Our training protocol is comparable with other studies examining focused cardiac ultrasound in the ED [11-13]. However, we specically showed that ED physicians with limited ultrasound training can accurately measure CI. We were unable to calculate CI in 3 patients due to unobtainable VTI measurements, resulting in an applicability of 97% with ultrasound CI measurements performed by ED physicians. The cardiologist (R.C.B.), who veried the measurements by both ED physicians and sonographer, has signicant expertise in echocardiography having developed one of the rst protocols for 2-dimensional echocardiographic examination [18]. Thus, we believe that our results are highly valid with respect to the accuracy of our measurements. Our study also showed that CI measurements between ED physicians and a sonographer had acceptable correlation. Measurements of LVOT diameter and VTI had similar interrater reliability compared with the reliability of cardiac dimensions measured by other trained sonographers [19]. The measurements by the ED physician slightly underestimated CI relative to sonographer measurements with no statistical difference. However, the percent difference in SVI measurements between ED physician and sonographer is similar to the variability in stroke volume obtained by 2

Data are presented as mean SD or as proportion with 95% CI. BMI, body mass index; BSA, body surface area; CI, condence interval.

common diagnostic categories, with respect to chief concern, were cardiac (30%), gastrointestinal (18%), musculoskeletal (16%), and neurologic (9%) (Table 2). All patients had LVOT diameter measurable by ED physician; however, VTI measurements were unobtainable in 3 patients due to body habitus or because the patient did not tolerate the examination. Twenty-three patients had LVOT diameter and VTI measurements performed by the sonographer within 10 minutes after the ED physician had completed the examination. Verication from the expert cardiologist showed optimal LVOT diameter measurements in 90.0% and 91.3% of patients by ED physician and sonographer, respectively (Table 3). Optimal VTI measurements by ED physician and sonographer were 78.4% and 78.3%, respectively. There was no statistical difference in cardiologist rating of LVOT diameter or VTI measurements between ED physician and sonographer. Cardiac index, LVOT diameter, VTI, SVI, and HR measurements by ED physician were 2.42 0.70 L min1 m2, 2.07 0.22 cm, 18.30 3.71 cm, 32.34 7.92 mL

Table 3 Expert cardiologist rating of cardiac ultrasound images, obtained by ED physician and sonographer, required to compute CI, including LVOT and VTI Operator Cardiologist rating No. of subjects, % (95% confidence interval) Optimal ED physician LVOT VTI Sonographer LVOT VTI 90, 90.0 (82.6-94.5) 76, 78.4 (69.2-85.4) 21, 91.3 (73.2-97.6) 18, 78.3 (58.1-90.3) Suboptimal 7, 7.0 (3.4-13.8) 15, 15.5 (9.6-24.0) 0, 0.0 (0.0-14.3) 4, 17.4 (7.0-37.1) Unobtainable 3, 3.0 (1.0-8.5) 6, 6.2 (2.9-12.8) 2, 8.7 (2.4-26.8) 1, 4.4 (0.8-21.0) 100 97 23 23 Total no. of subjects

Cardiac index with focused cardiac ultrasound

Table 4

Hemodynamic measurements of all patients obtained by ED physician and paired measurements obtained by both ED physician and sonographer All patients (N = 97), ED physician Patients with paired measurements by both ED physician and sonographer (n = 23) ED physician 2.58 0.60 2.04 0.20 19.17 3.38 34.16 6.33 76.00 13.15 Sonographer 2.69 0.83 2.09 0.22 19.35 4.00 35.81 8.47 75.44 13.4 % Difference 12.36 10.09 4.74 4.43 8.07 7.02 11.13 + 8.78 4.19 3.66 (95% confidence interval) 0.56 0.62 0.40 0.38 0.78 (0.23-0.90) (0.13-1.00) (0.04-0.76) (0.07 to 0.70) (0.36-1.00) Pearson correlation (95% confidence interval) 0.82 0.80 0.87 0.74 0.94 (0.60-0.92) (0.57-0.91) (0.70-0.94) (0.45-0.88) (0.86-0.97) Bias 0.11 0.05 0.17 1.65 0.57 Lower LOA 1.06 0.31 4.08 12.92 8.41 Upper LOA 0.83 0.22 18.01 9.62 9.54

CI (L min1 m2) LVOT (cm) VTI (cm) SVI (mL beat1 m2) HR (beats/min)
LOA. limits of agreement.

2.42 0.70 2.07 0.22 18.30 3.71 32.34 7.92 75.32 13.45

independent cardiologists [20]. Variability in CI measurements based on VTI has also been described because VTI measurement itself is dependent on the operator angle of the pulsed-wave Doppler [21-22]. In addition, CI variability may be affected by the physiologic uctuations in stroke volume, rather than solely by technical differences between 2 operators. A number of studies since the 1980s have shown that CI measurements by transthoracic echocardiography have a high correlation with thermodilution CI measurements obtained by pulmonary artery catheterization, with a correlation coefcient up to 0.96 [23-27]. Cardiac index determined by the LVOT

Fig. 4 Bland-Altman analysis for CI measured by ED physician compared with sonographer.

Fig. 3 Scattered plot for CI measured by ED physician compared with sonographer.

CIED Physician - CISonographer (L/min/m2)

-1.5 1 -1.0 -0.5 0.0 0.5 1.0 1.5

CIED Physician (L/min/m2)

1 1 2 3 4 5

Bias = -0.11 (-1.06 to 0.83)

CISonographer (L/min/m2)

2 3 4 5

CI Average (L/min/m2)

3 4 5

6 Doppler ultrasound method has been shown to be both sensitive and specic in identifying the cause of shock in critically ill patients [28]. We observed that the limiting factor in our ultrasound image acquisition was obtaining the apical 5-chamber view, required to measure VTI. Accordingly, our results showed a lower percentage of optimal VTI images compared with optimal LVOT images obtained by either ED physician or sonographer. Important to shock resuscitation is an end point that will reect uid responsiveness. This end point will guide the decision to administer more uids or, alternatively, initiate vasoactive therapy to maintain blood pressure and organ perfusion. Traditionally, an increase in CI of greater than 15% after a uid challenge has been considered the optimal end point reecting uid responsiveness. When CI does not increase in response to uid therapy aimed at increasing preload, cardiac contractility has reached a plateau and further uid resuscitation can be deleterious, resulting in volume overload and pulmonary edema. This fundamental tenet of shock resuscitation is based on Starling's [29] law of the heart. Thus, a change in CI after a uid bolus is often used as the reference standard when validating other indicators of uid responsiveness, such as central venous pressure, stroke volume variation, passive leg raise, and inferior vena caval index [30-33]. Among these, the caval index, or the change in the inferior vena cava diameter with inspiration, observed by ultrasound has been advocated as a noninvasive indicator of volume status [34]. However, the caval index is a surrogate for central venous pressure, which numerous studies already have shown not to be a reliable measure of uid responsiveness [30,35]. Other cardiac function parameters obtainable from ultrasound include ejection fraction, fractional shortening, and E-point septal separation; but none of these have been validated as a measure of uid responsiveness or end point of resuscitation [10]. The ability to measure CI as a guide to resuscitation is ideal because it is the reference standard that other volume indicators are based on [4]. Clinicians commonly use mean arterial pressure (MAP) as an end point for uid resuscitation. However, MAP is proportional to the product of cardiac output and systemic vascular resistance (SVR), MAP cardiac output SVR. In shock, MAP may be normal due to the increase in SVR by endogenous mechanisms to compensate for the decreased preload. Targeting a normal MAP during resuscitation may result in CI still being suboptimal and persistent organ hypoperfusion. For example, in a hemorrhagic shock model, blood loss up to 30% can maintain normal blood pressure. Beyond this point, the compensatory increase in SVR is exhausted and the fall in blood pressure is accelerated, resulting in rapid cardiovascular collapse and death [36]. Furthermore, knowing CI may assist the ED physician in determining the etiology of shock. For example, if CI is low and MAP is low, then SVR is most likely low, suggesting distributive (or vasodilatory) shock such as septic shock. However, if CI

V.A. Dinh et al. is low and MAP is normal or high, then cardiogenic shock may have occurred with an elevated vascular resistance. Many patients, of course, will fall into an indeterminate category in which trending CI measurements, in conjunction with other hemodynamic monitoring methods, may be additionally valuable in the diagnosis and resuscitation of shock. There are inherent disadvantages of using ultrasound to measure CI. The equation to calculate CI with ultrasound assumes that LVOT is a circle. Anatomically, however, LVOT varies in shape by each individual. In addition, VTI measurement is dependent on the angle at which the pulse Doppler waveform is obtained. Lastly, adequate cardiac views may be difcult to obtain in daily practice due to patient body habitus and position. Our results showed that despite these limitations, CI measurements were similar between ED physicians and sonographer, with a signicant number of measurements determined as optimal by the cardiologist. The availability of the sonographer was limited, and we were only able to compare ED physician and sonographer CI measurements in 24% of the patients. However, we believe that this number of patients is an adequate representation of the ED physicians' ability to reliably measure CI, especially with the cardiologist verifying the measurements performed in all patients by both the ED physicians and sonographer. Our study was performed in an academic institution with a formal ultrasound training curriculum; thus, the results may not yet be generalizable to other settings. We examined the ability to measure CI by 2 motivated ED physicians after 20 hours of training; thus, we do not know if our results in a convenience sample can be reproducible by other ED physicians in our group or at other institutions. However, our experience will be precedence for further studies examining ED physician measurements of CI by a focused cardiac ultrasound examination. As hospital overcrowding continues to worsen, critically ill patients will remain longer in the ED. Thus, it is becoming more important that the ED physician has the adequate tools to recognize and intervene early in those patients who are potentially hemodynamically compromised. Echocardiography has been used for hemodynamic monitoring for years by cardiologists. The use of bedside cardiac ultrasound as a hemodynamic monitoring device by noncardiologists is continuing to evolve [9]. Our study showed that CI can be measured by ED physicians with ultrasound technology available in the ED and can, therefore, potentially be used in the management of critically ill patients.

Acknowledgment
We thank Drs Kathleen Clem, MD, and Ellen Reibling, PhD, for their administrative support in the study. We thank Keiji Oda, MPH, for providing his expertise in the statistical

Cardiac index with focused cardiac ultrasound analysis for the study; Kimberly Ayers, RDCS, for providing cardiac ultrasound training; Jimmy Cunningham, BS, for data entry; and Sarah Pearl, BS, for administrative assistance. Finally, we are indebted to our patients who participated in the study without receiving any direct benet from our results during the course of their care in the ED.

7
[18] Bansal RC, Tajik AJ, Seward JB, et al. Feasibility of detailed twodimensional echocardiographic examination in adults. Prospective study of 200 patients. Mayo Clin Proc 1980;55:291-308. [19] Coletta C, De Marchis E, Lenoli M, et al. Reliability of cardiac dimensions and valvular regurgitation assessment by sonographers using hand-carried ultrasound devices. Eur J Echocardiogr 2006;7: 275-83. [20] Ihlen H, Endresen K, Myreng Y, et al. Reproducibility of cardiac stroke volume estimated by Doppler echocardiography. Am J Cardiol 1987;59:975-8. [21] Espersen K, Jensen EW, Rosenborg D, et al. Comparison of cardiac output measurement techniques: thermodilution, Doppler, CO2rebreathing and the direct Fick method. Acta Anaesthesiol Scand 1995;39:245-51. [22] Rawles J. How far is the cardiac output. Lancet 1985;1:461. [23] Darsee JR, Walter PF, Nutter DO. Transcutaneous Doppler method of measuring cardiac outputII. Noninvasive measurement by transcutaneous Doppler aortic blood velocity integration and M mode echocardiography. Am J Cardiol 1980;46:613-8. [24] Huntsman LL, Stewart DK, Barnes SR, et al. Noninvasive Doppler determination of cardiac output in man. Clinical validation. Circulation 1983;67:593-602. [25] Schuster AH, Nanda NC. Doppler echocardiographic measurement of cardiac output: comparison with a non-golden standard. Am J Cardiol 1984;53:257-9. [26] Lewis JF, Kuo LC, Nelson JG, et al. Pulsed Doppler echocardiographic determination of stroke volume and cardiac output: clinical validation of two new methods using the apical window. Circulation 1984;70:425-31. [27] Sjoberg BJ, Wranne B. Cardiac output determined by ultrasoundDoppler: clinical applications. Clin Physiol 1990;10:463-73. [28] Joseph MX, Disney PJ, Da Costa R, et al. Transthoracic echocardiography to identify or exclude cardiac cause of shock. Chest 2004;126: 1592-7. [29] Starling EH, editor. The Linacre lecture on the law of the heart. London, UK: Longmans, Green and Co.; 1918. [30] Marik PE, Baram M, Vahid B. Does central venous pressure predict fluid responsiveness? A systematic review of the literature and the tale of seven mares. Chest 2008;134:172-8. [31] Marik PE, Cavallazzi R, Vasu T, et al. Dynamic changes in arterial waveform derived variables and fluid responsiveness in mechanically ventilated patients: a systematic review of the literature. Crit Care Med 2009;37:2642-7. [32] Cavallaro F, Sandroni C, Marano C, et al. Diagnostic accuracy of passive leg raising for prediction of fluid responsiveness in adults: systematic review and meta-analysis of clinical studies. Intensive Care Med 2010;36:1475-83. [33] Moretti R, Pizzi B. Inferior vena cava distensibility as a predictor of fluid responsiveness in patients with subarachnoid hemorrhage. Neurocrit Care 2010;13:3-9. [34] Feissel M, Michard F, Faller JP, et al. The respiratory variation in inferior vena cava diameter as a guide to fluid therapy. Intensive Care Med 2004;30:1834-7. [35] Nagdev AD, Merchant RC, Tirado-Gonzalez A, et al. Emergency department bedside ultrasonographic measurement of the caval index for noninvasive determination of low central venous pressure. Ann Emerg Med 2010;55:290-5. [36] Schwaitzberg SD, Bergman KS, Harris BH. A pediatric trauma model of continuous hemorrhage. J Pediatr Surg 1988;23:605-9.

References
[1] Winters ME, McCurdy MT, Zilberstein J. Monitoring the critically ill emergency department patient. Emerg Med Clin North Am 2008;26: 741-57 ix. [2] Nowak RM, Sen A, Garcia AJ, et al. The inability of emergency physicians to adequately clinically estimate the underlying hemodynamic profiles of acutely ill patients. Am J Emerg Med 2011. [3] Rodriguez RM, Lum-Lung M, Dixon K, et al. A prospective study on esophageal Doppler hemodynamic assessment in the ED. Am J Emerg Med 2006;24:658-63. [4] Antonelli M, Levy M, Andrews PJ, et al. Hemodynamic monitoring in shock and implications for management. International Consensus Conference, Paris, France, 27-28 April 2006. Intensive Care Med 2007;33:575-90. [5] Swan HJ, Ganz W, Forrester J, et al. Catheterization of the heart in man with use of a flow-directed balloon-tipped catheter. N Engl J Med 1970;283:447-51. [6] Harvey S, Harrison DA, Singer M, et al. Assessment of the clinical effectiveness of pulmonary artery catheters in management of patients in intensive care (PAC-Man): a randomised controlled trial. Lancet 2005;366:472-7. [7] Binanay C, Califf RM, Hasselblad V, et al. Evaluation study of congestive heart failure and pulmonary artery catheterization effectiveness: the ESCAPE trial. JAMA 2005;294:1625-33. [8] Hofer CK, Ganter MT, Zollinger A. What technique should I use to measure cardiac output? Curr Opin Crit Care 2007;13:308-17. [9] Brown JM. Use of echocardiography for hemodynamic monitoring. Crit Care Med 2002;30:1361-4. [10] Weekes AJ, Quirke DP. Emergency echocardiography. Emerg Med Clin North Am 2011;29:759-87. [11] Mandavia DP, Hoffner RJ, Mahaney K, et al. Bedside echocardiography by emergency physicians. Ann Emerg Med 2001;38:377-82. [12] Moore CL, Rose GA, Tayal VS, et al. Determination of left ventricular function by emergency physician echocardiography of hypotensive patients. Acad Emerg Med 2002;9:186-93. [13] Blaivas M, Fox JC. Outcome in cardiac arrest patients found to have cardiac standstill on the bedside emergency department echocardiogram. Acad Emerg Med 2001;8:616-21. [14] Ihlen H, Amlie JP, Dale J, et al. Determination of cardiac output by Doppler echocardiography. Br Heart J 1984;51:54-60. [15] Seo H, Yamagishi M, Haque SA, et al. An enhanced method for measuring cardiac output using Doppler color flow echocardiography. Jpn Circ J 1997;61:905-11. [16] Fast JH, van den Merkhof L, Blans W, et al. Determination of cardiac output by single gated pulsed Doppler echocardiography. Int J Cardiol 1988;21:33-42. [17] Rowland T, Obert P. Doppler echocardiography for the estimation of cardiac output with exercise. Sports Med 2002;32:973-86.