ARTICLE IN PRESS

Current Anaesthesia & Critical Care (2006) 17, 237–244

0953-7112/$ - sdoi:10.1016/j.c

�Correspondifax: +44 208 38

E-mail addrd.hausenloy@u

www.elsevier.com/locate/cacc

MEDICINE, CARDIOLOGY AND ITU

Echocardiography and the critically ill patient

Refai Showkathali�, Derek Hausenloy

Cardiology SpR’s, Hammersmith Hospital, Du Cane Rd., London W12 0HS, UK

KEYWORDSEchocardiography;Transthoracic echo-cardiography;Intensive care unit;Critical care;LV function;RV function;Cardiac tamponade;Hand-held echocar-diography

ee front matter & 2006acc.2006.05.002

ng author. Tel.: +44 2073 4033.esses: refais@gmail.comcl.ac.uk (D. Hausenloy)

Summary Echocardiography provides valuable diagnostic information in theassessment and management of the critically ill patient on the intensive care unit(ICU). Transthoracic echocardiography (TTE), through the advent of portable hand-held echocardiography devices can provide the cardiologist, the intensivist or thesonographer with a rapid non-invasive and safe assessment of the cardiovascularstatus of the patient on ICU. Recent advances in technology such as harmonicimaging, digital image acquisition and contrast echocardiography have made forimprovements in image quality. This has extended the use of bedside echocardio-graphy from not only assessing the cardiac status, but also to diagnose or excludecomplications of procedures and therapies commonly used in ICU. However, incommon with other investigations, the information obtained from bedsideechocardiography should be interpreted in the context of other relevant investiga-tions and clinical findings. As more and more critical care physicians are gettingtrained in performing bedside echocardiography, they should be aware of differenttechniques that can be used while performing the study. Here, we review theapplication of echocardiography in various clinical diagnoses encountered in ICU andthe interpretation of findings in echocardiography.& 2006 Elsevier Ltd. All rights reserved.

Introduction

Echocardiography has emerged as an importantdiagnostic tool in the management of the criticallyill patient on the intensive care unit (ICU).1–4

Transthoracic echocardiography (TTE) is simplerand safer to perform in the ICU and is less timeconsuming when compared to transesophagealechocardiography (TEE), which was once consid-

Elsevier Ltd. All rights reserv

383 1000;

(R. Showkathali),.

ered the principle diagnostic tool for the assess-ment of the ICU patient. With new advances intechnology, such as harmonic imaging, digitalimage acquisition and the use of contrast agentsfor endocardial enhancement, TTE can provideimportant information that can aid in the diagnosisand management of the patient in the ICU.

With the advent of new portable hand-heldechocardiography devices, any suitably trainedperson (whether they be an intensivist, cardiologistor sonographer) are now able in a relatively shortperiod of time to undertake a point-of-care echocardiographic study. Obviously these bed-side echocardiographic studies have their own

ed.

ARTICLE IN PRESS

R. Showkathali, D. Hausenloy238

limitations in terms of image quality and functionallimitations of the hand-held machines, which thephysician should be aware of, when interpretingthe study. In addition, the ability to undertake anechocardiographic study in the average ICU patientis hampered by sub-optimal positioning of thepatient, the inability of the patient to cooperatewith the study, limited access to the traditionalechocardiography ‘windows’ (due to intravenouslines and chest drains), the use of inotropes and thefact that the patients is intubated. Also thesebedside studies should be employed as a means ofstabilizing the ICU patient such that furtherdefinitive information can be obtained by perform-ing a more detailed departmental echo.

However, in common with other investigations,the information obtained from bedside echocardio-graphy should be interpreted in the context ofother relevant investigations and clinical findings.This article will review the potential uses ofbedside echocardiography in the assessment and

Table 1 Echocardiography in the critically illpatient5.

Haemodynamic informationHypotensionAssess volume statusLeft ventricular (LV) systolic functionRegional wall motion abnormalityGlobal dysfunctionTransient dysfunction

Left ventricular diastolic functionRight ventricular functionOutflow tract obstructionValvular stenosis/regurgitationPericardial effusion/tamponade

HypoxiaRight ventricular functionRight ventricular pressureIntracardiac shuntsPulmonary embolus

Excluding infectionInfective endocarditis

Box 1 Calculating ejection fraction and FS from LV

Ejection fraction ¼ ðEnd�diastolic dimensionÞ3 � ðEndðEnd�diastolic dime

Fractional shortening ¼ End�diastolic dimension� EnEnd�diastolic dim

management of critically ill patients on the ICU(see Table 1).

Assessment of left ventricular (LV) function

The assessment of LV function is the most commonreason for performing bedside echocardiography inthe ICU. LV systolic function can be assessed usingechocardiography by measuring ejection fraction(EF�normal range 55–75%), fractional shortening(FS�normal range 30–42%) and cardiac output (CO).The relevant LV dimensions can be obtained from theparasternal long axis view and EF and FS can becalculated using the formula in Box 1. However,quantification of EF and FS has its limitationsparticularly, when accurate LV volumes and dimen-sions cannot be obtained due to suboptimal imaging.6

In this scenario, a visual estimate of global LV systolicfunction can be determined and graded normal LVsystolic function or mild, moderate and severeimpairment of LV systolic function.7

Estimation of stroke volume by the modifiedSimpson method (which requires accurate delinea-tion of the endocardium in systole and diastole) istechnically difficult in the critical care setting. It isfar more practical and amenable to estimate strokevolume and CO, using Doppler-derived instanta-neous blood flow velocity through a conduit with aknown cross-sectional area (CSA). Using the LVoutflow tract (LVOT) as the conduit is probably themost reliable and most commonly used applicationof this principle.8 The LV stroke volume is obtainedby measuring the CSA of the LVOT multiplied by thetransaortic flow velocity time integral (VTI) derivedby spectral doppler tracing (see Box 2). The LVOTdiameter is best measured in the parasternal longaxis view 1 cm below the aortic valve and thetransaortic VTI is obtained in the apical 5-chamberview.

The ability to assess the LV for regional wallmotion abnormalities (RWMA) in the criticallyill patient is essential particularly after a sus-pected myocardial ischemic event either in theperioperative patient, or in the patient who has

dimensions obtained from echocardiography.

�systolic dimensionÞ3

nsionÞ3,

d�systolic dimensionension

.

ARTICLE IN PRESS

Echocardiography and the critically ill patient 239

acutely deteriorated haemodynamically. This in-volves dividing the left ventricle into 16 segments(see Fig. 1) each of which is then graded accordingto their movement, as follows:9

Figseg

1 ¼ normokinesia,2 ¼ hypokinesia,3 ¼ akinesia,4 ¼ dyskinesia,5 ¼ aneurysmal.

From this, a wall motion score index (normalindex is 1) can be calculated as follows:

Wall motion score index ¼Sum of all segment score

Number of segments scored.

Box 2 Estimating stroke volume and cardiacoutput by measuring the cross-sectional areaof the LV outflow tract and the transaortic flowvelocity time integral.

Cross-sectional area (CSA) ¼ pr2 ¼ p(D/2)2

CSA ¼ D2� 0.785 (D is the measured LVOTdiameter in cm)LV stroke volume ¼ CSA�VTICardiac output (CO) ¼ Stroke volume�Heartrate

ParasternalLong axis

Pas

(m

Pash

Basal ant septal

Midant septal

Mid posterior

Basalposterior

(pa

Apical anterior

Mid anterior

Basalanterior

Basalinferior

Mid inferior

Apical inferior

Apical 2 chamber

Apical 4 chamber

Apical lateral

Midlateral

Basal lateral

Mid septum

Basal septum

Apical septum

ure 1 Assessing regional wall motion abnormalities by dividments the movement of which is graded on a scale of 1–5

LV diastolic function

LV diastolic dysfunction is commonly seen inpatients with hypertension, coronary artery dis-ease, cardiomyopathy and many forms of valvularheart disease, and is a potential cause of pulmon-ary edema in patients with documented normal LVsystolic function. Although there are number ofways for evaluating LV diastolic function, thecommonly used methods in routine practice includeDoppler assessment of mitral inflow, the pulmonaryvenous flow pattern and tissue Doppler imaging ofthe mitral annulus, using the apical 4-chamberview, with classical patterns portraying varyingdegree of LV diastolic dysfunction (see Fig. 2). Notethat the pseudonormal mitral inflow trace can bedistinguished from a normal mitral inflow trace, byexamining the pulmonary venous flow, which shoulddisplay atrial reversal.

Assessment of right ventricular (RV) function

RV dysfunction is common in critically ill patientsand its pathological role is underestimated in thesepatients. RV dysfunction can result from pressure orvolume overload of RV. The two common causes foracute RV dysfunction are massive pulmonaryembolus (PE) and acute respiratory distress syn-drome (ARDS).10 Adequate assessment of RV func-tion is needed in this condition, as the findings may

rasternal hort axis itral valve)

rasternal ort axis

(apex)

Basal septum

Basal inferior

Basal posterior

Basallateral

Basal anterior

Basal anterior septum

Mid anterior septum

Mid anterior

Mid lateral

Mid posterior

Midinferior

Mid septumParasternal

short axis pillary muscle)

Apical anterior

Apical lateralApical

inferior

Apical septum

ing the left ventricle viewed by echocardiography into 16(see text for details).9

ARTICLE IN PRESS

Figure 2 Assessing for LV diastolic dysfunction using echocardiography by analyzing the mitral valve inflow pattern andpulmonary venous flow.

R. Showkathali, D. Hausenloy240

alter therapy and is of prognostic value.11 Acute RVdysfunction may also be due to acute RV infarction,acute sickle cell crisis and sepsis.

Normal RV size is approximately two-thirds thatof the left ventricle, and so RV dilatation should beeasy to gauge. The qualitative assessment of RVsystolic function can be done by visualizing the RVin multiple views (parasternal long axis, RV inflowtract view, apical 4-chamber view and subcostal).Abnormal RV wall motion occurs in inferior myo-cardial infarction and pulmonary hypertension.Interventricular septal movement can be used toassess RV dysfunction and differentiate volumeoverload from pressure overload of the RV. Septalflattening is common in RV dysfunction and if theseptal distortion is only visualized during diastole,it is most likely due to volume overload, whereas inpressure overload, the septal flattening is usuallypresent in both systole and diastole. Quantitativeassessment of RV dysfunction such as RV wallthickness, RV fractional area change and long axisfunction by tissue Doppler imaging and myocardialperformance index (MPI) are difficult in the criticalcare setting.

It is possible to estimate the RV systolic pressure(RVSP) from the addition of right atrial pressure(RAP) to pulmonary artery systolic pressure (PASP).PASP can be quantified from the peak velocity ofthe tricuspid regurgitation (TR) jet velocity, in theapical 4-chamber (using the Bernoulli equation).RAP can be estimated from the diameter of theinferior vena cava (IVC), visualized in the subcostalview, and the degree to which it collapses withinspiration (Box 3).

Pulmonary embolus

TTE can be used to detect acute RV dilatation anddysfunction in massive PE. One study reported thatpatients presenting with acute PE displayed adistinct regional pattern of RV dysfunction withakinesia of the mid-free wall but with normalmotion at the apex.12 This finding of regional RVdysfunction was found to have a sensitivity of 77%and a specificity of 94% for the diagnosis of acutePE. It may also be possible to visualize thethrombus directly in the main pulmonary artery,

ARTICLE IN PRESS

Table 2 Echocardiographic signs used in thediagnosis of acute pulmonary thromboembolism.13

1. Direct visualization of thrombus in the rightsided chambers or the pulmonary artery

2. Right ventricular dilatation3. Reduced right ventricular function4. Reduced left ventricular cavity size5. Dilated pulmonary arteries6. Abnormal septal motion/systolic flattening of

the septum7. Significant (moderate to severe) TR8. Increased velocity of TR jet9. Dilatation of IVC

Box 3 Estimating the right ventricular sys-tolic pressure from the pulmonary arterysystolic pressure (PASP) and right atrial pres-sure (RAP). The PASP is estimated from thepeak velocity of the tricuspid regurgitant (TR)jet. The RAP is estimated from the diameter ofthe inferior vena cava (IVC), and its collapse ininspiration.

RVSP ¼ PASP+RAPPASP ¼ 4 (TR peak velocity)2

RAP(mmHg)

IVC diameter(cm)

Collapse oninspiration (%)

o5 o2 1005–10 o2 45010–15 42 25–5015–20 42 o25

Echocardiography and the critically ill patient 241

although TOE is more useful in visualizing thethrombi lodged in main and right pulmonary artery(see Table 2).

Pericardial effusion and cardiactamponade

Pericardial effusions are very common in ICUpatients particularly after cardiac surgery on thecardiothoracic ICU. The parasternal long and shortaxis view, apical view and subcostal view willusually reveal the effusion. In many critically illpatients, the subcostal view will often be the onlyadequate window to detect pericardial effusion. It

should be emphasized that cardiac tamponade is aclinical diagnosis and echocardiography may sug-gest a hemodynamic abnormality that may be thesubstrate for tamponade, but this alone will notestablish the diagnosis of tamponade. The signs oftamponade are directly due to actual elevation ofintrapericardial pressure. The RV free wall collapseis seen in early diastole in parasternal view andright atrial wall collapse is seen in late diastole inthe apical view.

Doppler evaluation of mitral and tricuspid inflowcan also be used to assess tamponade. In normalcircumstances, the peak velocity of mitral inflowvaries by no more than 15% with respiration andtricuspid inflow by no more than 25%. In cardiactamponade, respiratory variation in filling is ex-aggerated above these thresholds. The earliestfeature of tamponade to be noted is this exagger-ated respiratory variation of tricuspid inflow.14

Subsequent to this, exaggeration in mitral inflowpatterns can be noted. This is followed by rightatrial collapse and RV collapse is noted at a laterstage. Superior vena caval and hepatic vein flowscan also reflect the elevated intrapericardialpressure. Normally, venal caval flow is continuousand occurs in both systole and diastole. In thepresence of elevated intrapericardial pressure,flow during diastole is truncated and majority offlow into the heart occurs during ventricularsystole.

In some instances these changes may not beseen, the commonest example being in the patientwith significant RV hypertrophy, usually due topulmonary hypertension. The thick non-compliantRV wall is not compressed by the relatively modestelevation in pericardial pressure seen in earlydiastole and both clinical and echocardiographicsigns of compromise may be minimal or absent.Ventricular wall thickening due to malignantinfiltration, an overlying inflammatory response oran overlying thrombus in hemorrhagic pericarditismay also have the same effect.

Assessment of valvular function

Significant valvular abnormalities can be present inthe critically ill patient without being recognized.15

Precise evaluation of valvular pathologies needs tobe identified in ICU. The common indications forbedside echocardiography in this group of patientsare mitral and aortic valve disease, excludingnative infective endocarditis (IE) and for excludingprosthetic valve endocarditis.

In the critical care setting, TTE can providevaluable information concerning valvular integrity

ARTICLE IN PRESS

R. Showkathali, D. Hausenloy242

and function, but it may be suboptimal and notsensitive for detecting vegetations of endocarditis,or in the evaluation of a dysfunctional mitral valveand the assessment of most prosthetic valves. Inthis regard, TEE is superior for assessing valvularfunction.16 In some cases, TTE may provide betterimaging than TEE for evaluation of the anteriorstructures such as the aortic valve and also fordoing Doppler measurements.

Aortic stenosis

The aortic valve is first visualized in the parasternalview for thickening and shortening in case of aorticstenosis. All three leaflets can be visualized in shortaxis view and their mobility assessed. Dopplerassessment of flow through aortic valve can beobtained in apical view and peak and mean velocitymeasured across the valve with CW Doppler. Thepeak gradient across the valve can be derived fromthe equation below and thus the severity of aorticstenosis can be assessed as follows:

Peak pressure gradient ðmmHgÞ ¼ 4ðpeak velocityÞ2.

The aortic valve area can be assessed using thecontinuity equation, by measuring the LVOT area(as described before) and obtaining the meanvelocity of flow through LVOT with PW Doppler,and aortic mean velocity by CW Doppler. The aorticvalve area can be calculated by the equationbelow. In patients with impaired LV systolic func-tion, it is imperative to calculate the aortic valvearea when assessing the severity of aortic stenosis,as the peak pressure gradient across the stenoticaortic valve may well be reduced.

Aortic valve area ðcm2Þ ¼LVOT CSA� LVOT VTI

Aortic VTI.

Aortic regurgitation

Aortic regurgitation is detected and quantified byboth color flow and CW Doppler. The regurgitant jetcan be detected by color flow in the parasternallong axis view as a blue color coded jet. On theapical view, the direction of aortic regurgitationappears red, as the flow is towards the transducer.

AR can be quantified by four main methods:

1.

Color flow Doppler, 2. CW spectral Doppler trace of the regurgitant jet, 3. PW spectral Doppler trace of flow in descending

aorta,

4. LV volume overload.

Measurement of LV systolic and diastolic dimen-sions is important when assessing the severity of AR,

as one would expect LV dilatation if more thanmoderate AR is present. Pressure half-time measure-ment on the regurgitant jet, the slope of the jet andthe descending aorta diastolic flow reversal are otherimportant parameters for assessing severity.

Mitral valve

Assessment of the mitral valve by transthoracicecho in the ICU setting is difficult and mayunderestimate the severity of mitral valve disease.TEE provides better imaging for the assessment ofthe mitral valve in ICU patients, because of its closeanatomic proximity to the esophagus. Mitral valveleaflet opening and prolapse can be noted in aparasternal long axis view and in short axis view.M-mode across the mitral valve leaflet can show theopening of the valve clearly and can be used todiagnose mitral stenosis. Mitral valve area (MVA)can be assessed commonly by two methods:

1.

Valve planimetry can be done in a short axisview, by tracing around the inside edges of theMV leaflets when they are open to the maximum.

2.

It can also be assessed by measuring the pressurehalf time (PHT) on the E wave of the LV inflow jet,with PW Doppler along the inflow. MVA(cm2) ¼ 220/PHT. Measurement of LA size willgive additional information in mitral valve disease.

Assessment of MR by color flow doppler may notbe accurate and lead to underestimation of MRseverity, especially with the hand-held machines.17

TEE will provide a much more precise evaluation ofthe degree of MR and often provides information asto the etiology of MR. The diagnosis of acute severeMR is a surgical emergency, and so the threshold toperform a TEE if this is suspected should be low.

Infective endocarditis

IE is not uncommon in the ICU, either as apresenting diagnosis or a complication of concur-rent sepsis. IE has been reported to be the secondmost common indication for performance of echo-cardiography in ICU. As the consequence ofendocarditis may be potentially devastating andpossibly fatal, it is important to diagnose IE earlyand treat appropriately. The echocardiographicfeatures typical of IE are an oscillating intracardiacmass on a valve or supporting structure or aniatrogenic device, intracardiac abscesses, newpartial dehiscence of a prosthetic valve or new

ARTICLE IN PRESS

Echocardiography and the critically ill patient 243

valvular regurgitation.18 There is no single char-acter on the echocardiogram that will conclusivelyidentify a mass as vegetation. The ability to detecta vegetation depends on several factors includingvegetation size, location, the presence of under-lying heart disease, image quality and instrumentsettings. Different echo window should be used,and Doppler flow mapping should be performed toidentify any associated valvular regurgitation.Vegetations must be at least 3–6mm in size to bereliably seen by TTE, even though masses of 2mmhave been reported.19

The sensitivity of TTE for diagnosing endocarditishas been reported to be 58–62%.20 The sensitivityof TEE is superior to TTE and should be used in caseswhere clinical suspicion persists despite TTE beingnegative.

Intracardiac shunt

Intracardiac or intrapulmonary shunts should besuspected in cases of unexplained embolic stroke orrefractory hypoxemia. Common causes of right toleft shunts are atrial septal defects (ASD) and apatent foramen ovale (PFO). A PFO is present in25–30% of normal population.21 In cases where RApressure is greater than LA pressure such as ARDS,pulmonary embolism and RV infarction and severeTR, there will be increased right to left shuntingthrough the defect, causing significant hypoxemia.Ventricular septal defects can also complicatepatients presenting with an acute myocardialinfarction and leads to haemodynamic instability.

Color flow imaging can usually detect the largershunts, but contrast ‘bubble’ study is often neededto detect smaller ones. A contrast study shouldalways be performed when evaluating patients withunexplained embolic stroke or refractory hypox-emia. Approximately 0.5ml air is mixed with 10mlnormal saline solution, which is then agitated backand forth between two syringes connected by athree-way stopcock and injected into the venouscirculation, immediately following the releasephase of a Valsalva maneuvre. In the absence of ashunt, only minimal amount of contrast should beexpected to be seen in left sided cavities. If anintracardiac shunt is present, contrast in the leftsides of the heart will be observed immediatelyafter right side opacification and the contrast willbe seen going through the interatrial septum.

The application of echocardiography on theICU

The value of immediate bedside echocardiographyfor aiding in the diagnosis of acute haemodynamic

disturbances has been well demonstrated in boththe ICU and the emergency department.22 With theintroduction of new hand-held echocardiographicmachines, they can be used as an adjunct to thephysical examination. This should be directed toanswer a specific clinical question and should beshorter than the traditional echocardiogram, that isthe echocardiographic study is a point-of-careassessment. However, one should only view this asan extension of the physical examination and itshould not be compared with the quality of high-end machines that can be undertaken in thedepartment. A recent study by Gorcsan et al.23

investigated the utility of hand-held echo whenspecifically used as an extension of the physicalexamination. The hand-held echo demonstrated anexcellent close overall agreement (92–100%,r ¼ 0:91–0.96) for estimation of EF, LV hypertrophy,RWMA and pericardial effusion when compared toan echocardiographic study using a full-size echo-cardiographic system. The authors of this studyconcluded that use of ‘goal-directed hand-heldechocardiography has the potential to influencebedside patient treatment decisions and expeditehealthcare’.

In another study,24 it was reported that theaccuracy of the hand-held machines with respect totwo-dimensional imaging remains very good in thecritically ill patient when compared to standardechocardiography machine, but that informationprovided from hand-held color Doppler imagingshould be interpreted cautiously in this patientpopulation.

In most hospitals, particularly district generalhospitals, the provision of 24 h echocardiography byeither a cardiologist or sonographer is not feasible.In this regard, studies have reported the successfulundertaking of bedside echocardiography by suita-bly trained intensivists.25,26 However, adequatetraining is essential and this must be individualizedto the specific needs and application of the user.There is no specific accreditation available fromthe American or British society of echocardiographyin this scenario, but there are number of coursesavailable to equip the intensivist in performingbedside echocardiography on the ICU.

Conclusion

TTE has been accepted as one of the safest imagingmodalities to assess patients in ICU. This investiga-tion provides a powerful adjunctive tool to thephysical examination and can be used to guide inthe management of patients in the critical caresetting. It can also be used to exclude underlying

ARTICLE IN PRESS

R. Showkathali, D. Hausenloy244

cardiac abnormalities in patients admitted to theICU with a non-cardiac condition. Certain compli-cations of procedures and therapy given in ICU canalso be assessed by bedside TTE.

With hand-held echo machines readily available,bedside echocardiography will become an indis-pensable tool in the management of patients on theICU. There may be a graduated progression of theuse of hand-held device from cardiologist tointensivist in future. However, there should beadequate training and this should be individualizedto the specific needs of the operator. In addition,the limitations of bedside echocardiography shouldbe appreciated and the findings from a studyinterpreted in the context of other investigativefindings. The successful use and dissemination ofhand-held machine will not rest simply on the sizeof the instrument, but on the individual user andtheir understanding of and response to the in-formation imparted.

References

1. Stamos TD, Soble JS. The use of echocardiography in thecritical care setting. Crit Care Clin 2001;17:253–70.

2. Heidenreich PA. Transesophageal echocardiography (TEE) inthe critical care patient. Cardiol Clin 2000;18:789–805.

3. Cook CH, Praba AC, Beery PR, et al. Transthoracicechocardiography is not cost-effective in critically illsurgical patients. J Trauma 2002;52:280–4.

4. Beaulieu Y, Maric PE. Bedside ultrasonography in the ICU:part 1. Chest 2005;128:881–95.

5. Feigenbaum H, Armstrong WF, Ryan T. ICU and operative/perioperative application. In: Feigenbaum’s echocardiogra-phy. 6th ed. Philadelphia: Lippincott Williams & Wilkins;2005.

6. Vignon P, Mentec H, Terre S, et al. Diagnostic accuracy andtherapeutic impact of transthoracic and transesophagealechocardiography in mechanically ventilated patients in theICU. Chest 1994;106:1829–34.

7. Mueller X, Stauffer JC, Jaussi A, et al. Subjective visualechocardiographic estimate of left ventricular ejectionfraction as an alternative to conventional echocardiographicmethods: comparison with contrast angiography. Clin Car-diol 1991;14:898–902.

8. Ihlen H, Amlie JP, Dale J, et al. Determination of cardiacoutput by Doppler echocardiography. Br Heart J 1984;51:54–60.

9. Schiller NB, Shah PM, Crawford M, et al. Recommendationsfor quantitation of the left ventricle by two-dimensionalechocardiography. American Society of EchocardiographyCommittee on Standards, Subcommittee on Quantitation of

Two-Dimensional Echocardiograms. J Am Soc Echocardiogra-phy 1989;2:358–67.

10. Vieillard-Baron A, Schmitt JM, Augarde R, et al. Acute corpulmonale in acute respiratory distress syndrome submittedto protective ventilation: incidence, clinical implications,and prognosis. Crit Care Med 2001;29:1551–5.

11. Vieillard-Baron A, Page B, Augarde R, et al. Acute corpulmonale in massive pulmonary embolism: incidence,echocardiographic pattern, clinical implications and recov-ery rate. Intensive Care Med 2001;27:1481–6.

12. McConnell MV, Solomon SD, Rayan ME, et al. Regional rightventricular dysfunction detected by echocardiography inacute pulmonary embolism. Am J Cardiol 1996;78:469–73.

13. Leibowitz D. Echocardiographic signs used in diagnosis ofacute pulmonary thromboembolism. J Am Soc Echo 2001;14:921–6.

14. Feigenbaum H, William FA, Thomas R. Pericardial diseases.In: Feigenbaum’s echocardiography. 6th ed. Philadelphia:Lippincott Williams & Wilkins; 2005.

15. Bossone E, DiGiovine B, Watts S, et al. Range and prevalenceof cardiac abnormalities in patients hospitalized in amedical ICU. Chest 2002;122:1370–6.

16. Alam M. Transesophageal eehocardiography in critical careunits: Henry Ford Hospital experience and review of theliterature. Progr Cardiovase Dis 1996;38:315–28.

17. Smith MD, Cassidy JM, Gurley JC, et al. Echo Dopplerevaluation of patients with acute mitral regurgitation:superiority of transesophageal echocardiography with colorflow imaging. Am Heart J 1995;129:907–74.

18. Oh JK, Seward JB, Tajik AJ. The echo manual, 2nd ed.Philadelphia, PA: Lippincott, Raven; 1999.

19. Feigenbaum H, William FA, Thomas R. Infective endocardi-tis. In: Feigenbaum’s echocardiography. 6th ed. Philadel-phia: Lippincott Williams & Wilkins; 2005.

20. Shively BK, Gurule FT, Roldan CA, et al. Diagnostic value oftransesophageal compared with transthoracic echocardio-graphy in infective endocarditis. J Am Coll Cardiol1991;18:391–7.

21. Braunwald E, Zipes DP, Libby P. Heart disease: a textbook ofcardiovascular medicine, 6th ed. Philadelphia, PA: W.B.Saunders Company; 2001.

22. Counselman FL, Sanders A, Slovis CM, et al. The status ofbedside ultrasonography training in emergency medicineresidency programs. Acad Emerg Med 2003;10:37–42.

23. Gorcsan J, Pandey P, Sade LE. Influence of hand-carriedultrasound on bedside patient treatment decisions forconsultative cardiology. J Am Soc Echocardiogr 2004;17:50–5.

24. Vignon P, Chastagner C, Francois B, et al. Diagnostic abilityof hand-held echocardiography in ventilated critically illpatients. Crit Care 2003;7:R84–91.

25. Spevack- DM, Spevack DM, Tunick PA, et al. Hand carriedechocardiography in the critical care setting. Echocardio-graphy 2003;20:455–61.

26. Duvall WL, Croft LB, Goldman ME. Can hand-carriedultrasound devices be extended for use by the noncardiologymedical community? Echocardiography 2003;20:471–6.