Original Contribution

Interrater reliability of cardiac output measurements by transcutaneous Doppler ultrasound: implications for noninvasive hemodynamic monitoring in the ED^B,BB

H. Bryant Nguyen MD, MS*, Theodore Losey BS, Janet Rasmussen BA, Rebecca Oliver BS, Mindi Guptill BS,

William A. Wittlake MD, Stephen W. Corbett MD, PhD

Department of Emergency Medicine, Loma Linda University Medical Center, Loma Linda, CA 92354, USA

Received 1 April 2006; revised 8 May 2006; accepted 17 May 2006

Abstract

Introduction: Hemodynamic monitoring is an important aspect of caring for the critically ill patients boarding in the emergency department (ED). The purpose of this study is to investigate the Interrater agreement of noninvasive cardiac output measurements using transcutaneous Doppler Ultrasound technique.

Methods: This is a prospective observational cohort study performed in a 32-bed adult ED of an academic tertiary center with approximately 65000 annual patient visits. Patients were enrolled after verbal consent over a 7-month period. The raters were ED personnel involved in patient care. Paired measurements of cardiac index (CI) and stroke volume index (SVI) were obtained from a transcutaneous Doppler ultrasound cardiac output monitor.

Results: A convenience sample of 107 (50 women and 57 men) patients with a median age of 49 (32, 62) years was enrolled. One hundred two paired measurements were performed in 91 patients in whom adequate Doppler ultrasound signals were obtainable. The raters included 35 emergency medicine attending physicians, 31 emergency medicine residents, 80 medical students, 47 nurses, and 11 emergency medical technicians. Cardiac index range was 0.6 to 5.3 L/min per square meter, and SVI range was 7.7 to 63.0 mL/m². The correlation of CI measurements between 2 raters was good (r² = 0.87; 95% confidence interval, 0.86-1.00; P b .001). Likewise, SVI measurements between 2 raters also showed acceptable correlation (r² = 0.84; 95% confidence interval, 0.81-0.96; P b .001). Interrater reliability was strong for CI (j = 0.83 with 92.2% agreement) and SVI measurements (j = 0.72 with 88.2% agreement). Most patients had an interrater difference below 10% in CI and SVI measurements. Conclusions: Emergency department personnel, regardless of their role in patient care, are able to obtain

^B This study was supported by the Glen Helen Fund, which did not participate in any part of the study. This study was also supported by USCOM Pty Ltd, Australia, which provided funding and training on the use of the USCOM 1A monitor. USCOM Pty Ltd did not participate in the decision to submit the manuscript for publication.

^BB This study was presented at the American College of Emergency Physicians Annual Scientific Assembly, September 2005, Washington, DC.

* Corresponding author.

E-mail address: [email protected] (H.B. Nguyen).

0735-6757/$ - see front matter D 2006 doi:10.1016/j.ajem.2006.05.012

reliable cardiac output measurements in ED patients over a wide range of CI and SVI. Transcutaneous Doppler ultrasound technique may be an alternative to traditional invasive hemodynamic monitoring of critically ill patients presenting to the ED.

D 2006

Introduction

An important component in the assessment and treatment of critically ill patients in the emergency department (ED) is recognition of hemodynamic instability. As ED overcrowd- ing [1] becomes part of everyday practice for emergency physicians, so does the provision of intensive care to critically ill patients who board in the ED. Between 10 and 30 hours of critical care time is provided in the ED each day [2-4]. Similar to the golden hour in the care of trauma and acute myocardial infarction, early recognition and resuscitation of ED patients in septic shock within 6 hours has been shown to significantly decrease mortality [5]. The increased ED length of stay for these patients [2,3] require more advanced hemodynamic monitoring to be provided than has been done in the past.

The pulmonary artery catheter (PAC) has long been the mainstay for hemodynamic monitoring of critically ill patients in the intensive care unit (ICU) setting. However, the usefulness of this standard criterion device has been questioned with regard to its effect on outcome and potential complications [6]. Regardless of the debate surrounding the PAC, the widely accepted hemodynamic variable remains to be cardiac output (the product of heart rate and stroke volume) [7]. Several noninvasive technologies measuring cardiac output have become available to provide alternatives to pulmonary artery catheterization, including transcutane- ous Doppler ultrasonography, transesophageal Doppler, impedance cardiography, Pulse pressure variation, pulse contour analysis, and lithium indicator dilution [8,9].

In this study, we examined the interrater reliability of cardiac output measurements using transcutaneous Doppler ultrasound technique as a noninvasive hemodynamic monitoring alternative over a wide range of ED patients.

Materials and methods

Design

This was a prospective observational cohort study performed over a 7-month period, from February 1 to September 1, 2005, at an academic ED with approximately 65000 annual patient visits. The study was approved by the institutional review board at our institution.

Inclusion criteria

A convenience sample of ED patients older than 18 years were enrolled. Enrollment occurred 1 day per week when a member of the study team was available. Patients were arbitrarily approached by the investigator and explained the

purpose of the study. Verbal consent was obtained. Patients unable to give verbal consent, such as those with altered mental status, comatose, or sedated on mechanical ventila- tion, were enrolled with waiver of consent if an authorized representative was not available. Because we aimed to enroll patients with a wide range of clinical presentations and cardiac output, no other inclusion criteria were necessary.

Exclusion criteria

Patients unable to tolerate the supine position were excluded.

Study protocol

After patient enrollment, the investigator requested the ED personnel (rater) involved in the patient’s care to obtain cardiac index (CI) and stroke volume index (SVI) measurements, using a transcutaneous Doppler ultrasound cardiac output monitor (USCOM 1A, USCOM Ltd, Australia) (Fig. 1).

• • 1. Rater training

Before obtaining patient CI and SVI measurements, each rater was given a brief training on the usage of the Doppler ultrasound cardiac output monitor. The rater was presented with an example of bacceptableQ vs bnot acceptableQ signal tracing (Fig. 1). The rater was then instructed to recognize an acceptable signal tracing by performing Doppler Ultrasound measurements on himself or herself. Each measurement is obtained by the rater placing the Doppler ultrasound probe at the Suprasternal notch and optimizing the signal. Once the rater was satisfied with the signal, 3 cardiac profiles with the highest signal quality were selected by the rater. The USCOM 1A software then calculated the average CI and SVI from these 3 optimal cardiac profiles. The training session approximated 5 to 10 minutes and took place outside the patient area. After training, the rater approached the patient bedside to perform CI and SVI measurements on the enrolled patient. The investigator accompanied the rater as an observer and did not contribute to the rater’s recognition and selection of an acceptable signal tracing. None of the raters had experience in the use of the Doppler ultrasound cardiac output monitor before this study.

• • 1. Data measurements and collection

Paired measurements of CI and SVI were obtained by 2 blinded raters with a maximum time of 15 minutes between measurements. All measurements were obtained with the patient in supine position. Although the rater may not be unique for each patient enrollment, each rater-patient

Fig. 1 Measurements for CI and SVI are performed by the operator applying the Doppler ultrasound probe at the sternal notch. An acceptable ultrasound signal tracing includes cardiac profiles with triangular shape, highest and sharpest peak achievable, clear start and completion of systole, and filling in of profiles (A). An unacceptable tracing has incomplete filling in of cardiac profiles with irregular peaks and shapes (B).

measurement is considered unique. The rater category (emergency medicine attending, emergency medicine resi- dent, nurse, medical student, or emergency medical techni- cian) and experience with obtaining measurements were recorded. The experience of the rater was defined as the number of measurements performed before the current exam. Patient demographics and ED diagnosis category were also recorded.

Statistical analysis

All patients enrolled were included in the analysis. The applicability of the USCOM 1A monitor was defined as the percentage of patients with acceptable signal tracings obtained by both raters. Only patients with acceptable signals determined by both raters were included in the interrater agreement analysis. Pearson’s correlation and simple agreement were computed for paired measurements of CI and SVI. To examine the interrater reliability of cardiac output measurements defined as bnormalQ vs babnormal,Q j was also computed. Abnormal CI was defined as less than

2.5 L/min per square meter, and abnormal SVI was defined as less than 35 mL/m² [10]. Bland-Altman analysis was performed to determine the extent of deviation from the line of agreement between the raters. Interrater difference in CI and SVI measurements from the average CI and SVI,

respectively, were calculated. Data are presented as absolute value or median (25th, 75th percentile). Proportions are presented with 95% confidence interval. The software package, Intercooled Stata 8.0 (StataCorp, Tex), was used to perform the statistical analysis.

Results

During the study period, 107 patients were enrolled with a median age of 49 (32, 62) years. One hundred two paired measurements were obtained by 204 raters in 91 patients who had acceptable Doppler ultrasound signal tracings for CI and SVI measurements, with an applicability of 85.0% (95% confidence interval, 70.2-94.3). Reasons for unob- tainable Doppler ultrasound signal tracings from the remaining patients included tracheostomy in 4 patients, atrial fibrillation in 6, short neck in 4, and thick sternum in 2. The range of CI was 0.6 to 5.3 L/min per square meter, and the range of SVI was 7.7 to 63.0 mL/m². Most raters were medical students, emergency medicine residents, and attending physicians. Experience with the device varied from 1 to 5 examinations to more than 15 exams. The median interrater time (or time between paired measure- ments) was 6 (5, 10) minutes. Cardiac and trauma were the most common diagnosis categories (Table 1).

Table 1 Patient characteristics, rater categories, and interrater reliability analysis

Patient demographics

Patients enrolled (no.) 107

Male 57

Female 50

Age (y) 49 (32, 62)

body surface area (m²) 2.0 (1.7, 2.1)

CI range (L/min m²) 0.6-5.3

SVI range (mL/m²) 7.7-63.0

Diagnosis category (no. of patients)

Cardiac	22
Trauma	16
Gastrointestinal	14
Infectious	11
Neurologic	10
Respiratory	9
Metabolic	6
Other	19

Rater Category (no. of raters)

Attending 35

Resident 31

Medical student 80

Nurse 47

EMT 11

Rater experience before examination (no. of raters) 1-5 examinations 82

6-10 examinations 14

11-15 examinations 7

N 15 examinations 41

Interrater agreement

(or the mean difference between measures) of -0.02 L/min per square meter with a SD of 0.43 L/min per square meter and limits of agreement from -0.88 to 0.84 L/min per square meter. Stroke volume index showed a bias of -0.60 mL/m² with an SD of 5.28 mL/m² and limits of agreement from

-11.16 to 9.96 mL/m². The median difference between raters in CI and SVI measurements was 9.8% (4.3%, 20.3%) and 9.7% (4.8%, 20.3%), respectively, with most patients having interrater difference of less than 10% in both CI and SVI measurements (Table 1 and Fig. 2).

Discussion

A requisite in hemodynamic optimization is the usage of appropriate monitoring tools. The PAC has been tradition- ally the bstandard criterionQ in hemodynamic monitoring in the ICU. However, inserting a PAC in the ED setting is not practical nor feasible, especially when multiple studies have shown that the usage of PAC results in questionable outcome benefit [6,11-13]. Reasons may include lack of clinician knowledge and agreement in the Hemodynamic variables [14,15] and lack of a defined treatment protocol

[16] resulting in varying interventions based on the information available from pulmonary artery catheterization [17,18]. On the other hand, parameters traditionally available in the ED, such as vital signs and physical exam, poorly correlate with more acceptable signs of hypovolemia [19], cardiac dysfunction [20], organ failure, and shock

Applicability (%)

(95% confidence interval)

85.0 (70.2-94.3)

[21,22]. In fact, only 56% of intensivists can reliability

predict the hemodynamic profile of critically ill patients

Paired measurements (no.) 102

Interrater time (min) 6 (5, 10) Interrater difference (%)

CI 9.8 (4.3, 20.3)

SVI 9.7 (4.8, 20.3)

Pearson correlation (r²) (95% confidence interval)

CI 0.87

(0.86-1.00; P b .001)

SVI 0.84

(0.81-0.96; P b .001)

j

CI 0.83

SVI 0.72

Agreement (%)

CI 92.2

SVI 88.2

Applicability defines the number of patients with acceptable Doppler ultrasound signal. Age, body surface area, interrater time, and interrater difference are described as median (25th, 75th percentile). EMT, emergency medical technician.

Interrater reliability assessment demonstrated accept- able agreement in CI measurements (j = 0.83 with 92.2% agreement) and SVI measurements (j = 0.72 with 88.2% agreement). Bland-Altman analysis of CI showed a bias

before catheterization [18].

A feasible method of hemodynamic monitoring in the ED setting is the use of the central venous catheter to measure central venous pressure and central venous oxygenation [23]. However, such a procedure has been reported as a barrier when considering implementation of optimal therapy, such as early goal-directed therapy for severe sepsis and septic shock [24]. Reasons for this barrier may include the time required for the procedure in a busy ED; fear of complications such as thromboembo- lism, pneumothorax or bleeding; and limited experience in the procedure. Given the potential benefit for ED intervention and ED physicians playing a central role in the care of critically ill patients, these concerns would justify evaluation of alternative noninvasive hemodynamic monitoring techniques [9].

Our study showed that transcutaneous Doppler ultra- sound may be used to measure CI and SVI with acceptable interrater agreement by a varied group of ED personnel. The heterogeneous patient population with a wide range of CI and SVI also suggests that cardiac output may be measured noninvasively in the ED setting. Furthermore, Doppler ultrasound does not require user calibration, as needed by the central venous or PACs, adding to its potential ease of use.

Fig. 2 Scatter plot, Bland-Altman analysis, and interrater differences in CI and SVI measurements by Doppler ultrasound.

The clinical application for Noninvasive measurement of cardiac function in the ED is extensive. First, critically ill patients often present with a low-flow state, leading to shock. Although most of these patients will respond by increasing blood pressure after resuscitation, approximately 25% of patients may demonstrate undetected decrease in their CI [25]. Prompt recognition of these circulatory changes may prevent the Cardiovascular collapse that accompanies shock. Second, in addition to early detection of poor cardiac function, the use of noninvasive cardiac

output measurements could allow for titration of therapy in Disease states such as dyspnea [26], congestive heart failure [27], and hypovolemia [28]. Knowledge of cardiac output in the ED setting has been shown to result in a change in therapy 31% of the time [29]. Third, noninvasive monitor- ing may be useful to prognosticate outcome of ED patients undergoing resuscitation [30] because critically ill patients with high CI appear to have better outcome than those with lower CI [31,32]. Finally, while patients are awaiting intensive care bed transfer, noninvasive hemodynamic

assessment in the ED may serve as a bridge to more traditional invasive monitoring performed in the ICU, similar to the bridge that oxygen saturation monitoring has provided to arterial blood gas analysis.

Limitations

Because our study was performed in the ED setting, we did not examine the validity of CI and SVI measurements obtained from transcutaneous Doppler ultrasound when compared with a standard criterion. However, other studies have shown that Doppler ultrasound has an acceptable correlation with the PAC in measuring cardiac output [33-35]. Furthermore, in a dog model the Doppler ultra- sound measurements have correlation coefficient up to 0.98 (95% confidence interval, 0.97-1.0) with a flow probe placed on the aorta, reflecting true blood flow from the heart [36]. Although more studies are needed to validate Doppler ultrasound as a hemodynamic monitoring tool, we have focused on its reliability in this study.

We purposely examined CI and SVI measurements by raters who have varying roles in the patients’ care with the premise that noninvasive cardiac output can be obtainable by any ED personnel after a brief training session near the patient bedside. The presence of the investigator during the rater measurement conceivably may have added an inves- tigator bias to the results; however, the investigator did not contribute to the rater’s selection of the signal tracing. The raters themselves were also completely blinded from each other’s measurements.

The raters in this study have a wide range of experience in measuring cardiac output with Doppler ultrasound. Howev- er, most of our raters have less than 5 patient examinations as their experience before a current exam. This minimal experience may have contributed to unobtainable Doppler ultrasound signal tracings in 15% of patients, mainly in those patients with tracheostomy or atrial fibrillation. Because we used the sternal notch as a simple anatomical landmark to apply the Doppler ultrasound probe, cardiac output measure- ments were naturally unobtainable in patients with trache- ostomy. Patients with atrial fibrillation have beat-to-beat variability in left ventricular performance [37] and, thus, may not be candidates for cardiac output measurements by Doppler ultrasound. In those patients whom our raters could obtain acceptable signal tracings, we were able to show an acceptable j and a median difference below 10% in measurements between raters.

It is possible that the raters in our study have obtained measurements that agree in value but are physiologically inaccurate. However, the purpose of our study was to examine the reproducibility of noninvasive cardiac output measurements and not their accuracy. Furthermore, the level of variability found in our study is similar to the interrater reliability of vital sign measurements. Edmonds et al [38] found that the variability in obtaining vital signs in the ED

can be as high as 24%. In addition, it is difficult to determine if the variability in measurements is due to rater variation or to physiologic variation in cardiac output. In patients breathing spontaneously, cardiac output may have variability up to 10% over time without interventions such as Fluid loading [39]. In critically ill patients with sepsis having significant mortality, this variability may be as high as 14% [40]. These fluctuations may be due to complex interactions in the arterial and cardiopulmonary baroreflex and neurohormonal systems in controlling peripheral resistance [41,42]. Therefore, both rater and physiologic variations may affect the reliability of cardiac output measurements by Doppler ultrasound technique [43]. Given the wide range of cardiac output measurements in our study patients, we believe that the measurements are meaningful. Finally, static cardiac output measurements may vary up to 25% using the standard criterion of thermodilution with the PAC. However, a change in cardiac output of greater than 20% from a previous measurement will reflect true change in flow, such as in response to therapy or fluid resuscitation [7]. Thus, an interrater difference of less than 10% as observed in this study would suggest that the static measurements obtainable by Doppler ultrasound are reliable between different raters and of clinical relevance.

Conclusion

In this study, we showed that ED personnel could reliably measure a wide range of CI and SVI with the use of transcutaneous Doppler ultrasound. With the increasing necessity to provide intensive care to the critically ill patients in the ED setting and the potential outcome benefit with optimal resuscitation, further study is needed to examine noninvasive hemodynamic monitoring in the usual care of ED patients awaiting ICU admission. The potential of a resuscitation strategy, such as early goal-directed therapy, incorporating such technology is attractive and warrants further investigation [9]. However, the adaptation of new technology involves reliability and validity testing, assessment of therapeutic impact in clinical practice, and outcomes analysis. We believe that our study is the first step in these directions.

Acknowledgments

The authors thank the ED physicians, nurses, students and staff for their contribution in the data acquisition. Written consent was obtained from the patient in Fig. 1 for publication of photograph in the study.

References

1. Cowan RM, Trzeciak S. Clinical review: emergency department overcrowding and the potential impact on the critically ill. Crit Care 2005;9:291 - 5.
2. Fromm Jr RE, Gibbs LR, McCallum WG, Niziol C, Babcock JC, Gueler AC, et al. Critical care in the emergency department: a time- based study. Crit Care Med 1993;21:970 - 6.
3. McCaig LF, Burt CW. National hospital ambulatory medical care survey: 2001 emergency department summary. Adv Data 2003;1 - 29.
4. Nguyen HB, Rivers EP, Havstad S, Knoblich B, Ressler JA, Muzzin AM, et al. Critical care in the emergency department: a physiologic assessment and outcome evaluation. Acad Emerg Med 2000;7:1354 - 61.
5. Rivers E, Nguyen B, Havstad S, Ressler J, Muzzin A, Knoblich B, et al. Early goal-directed therapy in the treatment of severe sepsis and septic shock. N Engl J Med 2001;345:1368 - 77.
6. Connors Jr AF, Speroff T, Dawson NV, Thomas C, Harrell Jr FE, Wagner D, et al. The effectiveness of right heart catheterization in the initial care of critically ill patients. SUPPORT Investigators. JAMA 1996;276:889 - 97.
7. Pinsky MR. Why measure cardiac output? Crit Care 2003;7:114 - 6.
8. Chaney JC, Derdak S. Minimally invasive hemodynamic monitoring for the intensivist: current and emerging technology. Crit Care Med 2002;30:2338 - 45.
9. Gunn SR, Fink MP, Wallace B. Equipment review: the success of early goal-directed therapy for septic shock prompts evaluation of current approaches for monitoring the adequacy of resuscitation. Crit Care 2005;9:349 - 59.
10. Kruse JA. Hemodynamic monitoring. In: Kruse JA, Fink MP, Carlson RW, editors. Saunders manual of critical care. Philadelphia7 Elsevier Science, 2003. p. 774 - 7.
11. Shah MR, Hasselblad V, Stevenson LW, Binanay C, O’Connor CM, Sopko G, et al. Impact of the pulmonary artery catheter in critically ill patients: meta-analysis of randomized clinical trials. JAMA 2005;294:1664 - 70.
12. Richard C, Warszawski J, Anguel N, Deye N, Combes A, Barnoud D, et al. early use of the pulmonary artery catheter and outcomes in patients with shock and acute respiratory distress syndrome: a randomized controlled trial. JAMA 2003;290:2713 - 20.
13. Sandham JD, Hull RD, Brant RF, Knox L, Pineo GF, Doig CJ, et al. A randomized, controlled trial of the use of pulmonary-artery catheters in high-risk surgical patients. N Engl J Med 2003;348:5 - 14.
14. Iberti TJ, Fischer EP, Leibowitz AB, Panacek EA, Silverstein JH, Albertson TE. A multicenter study of physicians’ knowledge of the pulmonary artery catheter. Pulmonary artery catheter study group. JAMA 1990;264:2928 - 32.
15. Johnston IG, Jane R, Fraser JF, Kruger P, Hickling K. Survey of intensive care nurses’ knowledge relating to the pulmonary artery catheter. Anaesth Intensive Care 2004;32:564 - 8.
16. Pinsky MR, Vincent JL. Let us use the pulmonary artery catheter correctly and only when we need it. Crit Care Med 2005;33:1119 - 22.
17. Jain M, Canham M, Upadhyay D, Corbridge T. Variability in interventions with pulmonary artery catheter data. Intensive Care Med 2003;29:2059 - 62.
18. Mimoz O, Rauss A, Rekik N, Brun-Buisson C, Lemaire F, Brochard L. Pulmonary artery catheterization in critically ill patients: a Prospective analysis of Outcome changes associated with catheter-prompted changes in therapy. Crit Care Med 1994; 22:573 - 9.
19. McGee S, Abernethy III WB, Simel DL. The rational clinical examination. Is this patient hypovolemic? JAMA 1999;281:1022 - 9.
20. Wo CC, Shoemaker WC, Appel PL, Bishop MH, Kram HB, Hardin E. Unreliability of Blood pressure and heart rate to evaluate cardiac output in emergency resuscitation and critical illness. Crit Care Med 1993;21:218 - 23.
21. Rady MY, Smithline HA, Blake H, Nowak R, Rivers E. A comparison of the Shock Index and conventional vital signs to identify acute, critical illness in the emergency department. Ann Emerg Med 1994;24:685 - 90.
22. Rady MY, Rivers EP, Nowak RM. Resuscitation of the critically ill in the ED: responses of blood pressure, heart rate, shock index, Central venous oxygen saturation, and lactate. Am J Emerg Med 1996;14:218 - 25.
23. Rivers EP, Ander DS, Powell D. Central venous oxygen saturation monitoring in the critically ill patient. Curr Opin Crit Care 2001;7:204 - 11.
24. Jones AE, Kline JA. Use of goal-directed therapy for severe sepsis and septic shock in academic emergency departments. Crit Care Med 2005;33:1888 - 9 [author reply 1889-90].
25. Lopez-Saucedo A, Hirt M, Appel PL, Curtis DL, Harrier HD, Shoemaker WC. Feasibility of noninvasive physiologic monitoring in resuscitation of trauma patients in the emergency department. Crit Care Med 1989;17:567 - 8.
26. Peacock WF, Summers R, Vogel J, Emmerman C. Impedance cardiography changes therapy in dyspneic patients: results from the ED-IMPACT trial. J Card Fail 2004;10:119.
27. Milzman DP, Batiatta A, Zlidenny A, Janchar T. Noninvasive cardiac output monitoring improves the acute resuscitation of congestive heart failure in the ED beyond vital signs. Crit Care Med 1998;26:A62.
28. McKendry M, McGloin H, Saberi D, Caudwell L, Brady AR, Singer

M. Randomised controlled trial assessing the impact of a nurse delivered, flow monitored protocol for optimisation of circulatory status after cardiac surgery. BMJ 2004;329:258.

1. Urrunuga J, Rivers EP, Mullen MT, Karriem V, Nguyen HB, Knoblich B, et al. Hemodynamic evaluation of the critically ill in the emergency department: a comparison of Clinical impression versus transesophageal Doppler measurement. Ann Emerg Med 1999;34:S46.
2. Shoemaker WC, Wo CC, Chan L, Ramicone E, Kamel ES, Velmahos GC, et al. outcome prediction of emergency patients by noninvasive hemodynamic monitoring. Chest 2001;120:528 - 37.
3. Velmahos GC, Wo CC, Demetriades D, Shoemaker WC. Early continuous noninvasive haemodynamic monitoring after severe blunt trauma. Injury 1999;30:209 - 14.
4. Shoemaker WC, Wo CC, Yu S, Farjam F, Thangathurai D. Invasive and noninvasive haemodynamic monitoring of acutely ill sepsis and septic shock patients in the emergency department. Eur J Emerg Med 2000;7:169 - 75.
5. Tan HL, Pinder M, Parsons R, Roberts B, van Heerden PV. Clinical evaluation of USCOM ultrasonic cardiac output monitor in cardiac surgical patients in intensive care unit. Br J Anaesth 2005; 94:287 - 91.
6. Knobloch K, Lichtenberg A, Winterhalter M, Rossner D, Pichlmaier M, Phillips R. Non-invasive cardiac output determination by two- dimensional independent Doppler during and after cardiac surgery. Ann Thorac Surg 2005;80:1479 - 83.
7. Huntsman LL, Stewart DK, Barnes SR, Franklin SB, Colocousis JS, Hessel EA. Noninvasive Doppler determination of cardiac output in man. Clinical validation. Circulation 1983;67:593 - 602.
8. Critchley LA, Peng ZY, Fok BS, Lee A, Phillips RA. Testing the reliability of a new ultrasonic cardiac output monitor, the USCOM, by using aortic flowprobes in anesthetized dogs. Anesth Analg 2005;100:748 - 53.
9. Schneider J, Berger HJ, Sands MJ, Lachman AB, Zaret BL. Beat-to- beat left ventricular performance in atrial fibrillation: radionuclide assessment with the computerized nuclear probe. Am J Cardiol 1983;51:1189 - 95.
10. Edmonds ZV, Mower WR, Lovato LM, Lomeli R. The reliability of vital sign measurements. Ann Emerg Med 2002;39:233 - 7.
11. Sasse SA, Chen PA, Berry RB, Sassoon CS, Mahutte CK. Variability of cardiac output over time in medical intensive care unit patients. Crit Care Med 1994;22:225 - 32.
12. Seiver AJ, Szaflarski NL. Report of a case series of ultra low- frequency oscillations in cardiac output in critically ill adults with

sepsis, systemic inflammatory response syndrome, and Multiple organ dysfunction syndrome. Shock 2003;20:101 - 9.

1. Mukkamala R, Kim JK, Li Y, Sala-Mercado J, Hammond RL, Scislo TJ, et al. Estimation of arterial and cardiopulmonary total peripheral resistance baroreflex gain values: validation by chronic arterial baroreceptor denervation. Am J Physiol Heart Circ Physiol 2006;290:H1830-6.
2. Letienne R, Barres C, Cerutti C, Julien C. Short-term haemodynamic variability in the conscious areflexic rat. J Physiol 1998;506(Pt 1): 263 - 74.
3. Stern H, Haberhauer P, Emmrich P, Buhlmeyer K. Interobserver and spontaneous variability of Doppler echocardiography measured cardiac output in children with congenital heart defects. Z Kardiol 1990;79:89 - 95.