Echocardiography Diagnosis

Eur J Echocardiography (2007) 8, 4e14

INVITED PAPER

Echocardiography in heart failure: Beyonddiagnosis

J.K. Oh*

Mayo Clinic College of Medicine, Division of Cardiovascular Diseases, 200 First Street SW,Rochester, MN 55905-0001, USA

Received 11 September 2006; accepted 13 September 2006Available online 1 November 2006

by guest on April 19, 2016

Dow

nloaded from

It is a great honor for me to deliver this year’sEuro Echo Lecture in this beautiful city of Florenceand I appreciate the European Association ofEchocardiography for the invitation and thisopportunity to discuss the role of echocardiographyin systolic and diastolic heart failure as the singlemost useful test, not only in diagnosis, but also asa tool to provide insights into the pathophysiolog-ical mechanism of various etiologies of heartfailure, as a tool to monitor patient’s response tovarious treatment options, and as a tool to help usdevelop innovative new therapies for heart failure.

According to the ESC guidelines published in2005, heart failure is defined as ‘‘a complexclinical syndrome that can result from any struc-tural or functional cardiac disorder that impairsthe ability of ventricles to fill with or eject blood’’which includes diastolic as well as systolic heartfailure.1 ACC/AHA guidelines as well as ESC guide-lines state that echocardiography is the singlemost useful test in the diagnosis of heart failuresince structural abnormality, systolic dysfunction,diastolic dysfunction, or a combination of theseabnormalities needs to be documented in patientswho present with resting or/and exertional symp-toms of heart failure to establish a definitive

* Tel.: þ1 507 284 3685; fax: þ1 507 284 3968.E-mail address: [email protected]

1525-2167/$32 ª 2006 Published by Elsevier Ltd on behalf of Thedoi:10.1016/j.euje.2006.09.002

diagnosis of heart failure.1,2 It is important to dem-onstrate an objective evidence of structural orfunctional abnormalities to explain patient’ssymptoms of heart failure since symptoms of heartfailure are not specific and more than a third of pa-tients with a clinical diagnosis of heart failure maynot actually have heart failure. It is also interest-ing to note from the SHAPE study that only 3% ofmore than 7000 subjects surveyed from nine Euro-pean countries recognized breathlessness, fatigue,and edema as symptoms of heart failure.3 More-over, even when someone presents with typicalsymptoms of heart failure, a diagnostic test,most frequently an echocardiography examina-tion, is required to establish the underlying etiol-ogy for an optimal management strategy. In thislecture, I like to review with you how echocardiog-raphy is used for detection of structural abnormal-ity, diagnosis and management of systolic heartfailure, and evaluation of diastolic function inpatients with symptoms of heart failure.

When a cardiac structural abnormality is respon-sible for patient’s heart failure, it is usually obviouson 2-dimensional echocardiogram. Fig. 1 illustratesan echocardiographic image of a middle age womanwith increasing dyspnea. Parasternal long axis viewdemonstrated an enlarged right ventricle consistentwith right ventricular volume or pressure overloadwhich provides a list of differential diagnoses such

European Society of Cardiology.

mailto:[email protected]

Echocardiography in heart failure 5


Dow

nloaded from

as severe tricuspid regurgitation, pulmonary hyper-tension, or intracardiac shunt to be investigated. Asubcostal view subsequently showed a large secun-dum atrial septal defect. Echocardiographic detec-tion of most structural abnormalities responsible forheart failure is straight forward, but some structuralabnormalities demands a more detective work andechocardiographer’s knowledge as illustrated ina 55-year-old woman with increasing shortness ofbreath and history of radiation to the chest. Becauseof her radiation history, restrictive myopathy, peri-cardial disease, coronary disease, and valvularheart disease are considered and evaluated.4 How-ever, echocardiography showed an unusual flow di-rected to the right atrium and the tricuspid valvewhich was initially thought of as an atrial septal de-fect, but there was no evidence of right ventricularenlargement or volume overload. Further investiga-tion with color-flow imaging and Doppler echocardi-ography demonstrates that the flow was originatedfrom the coronary sinus with a peak velocity ofgreater than 2 m/sec. Because of a concern forcoronary sinus stenosis, transesophageal echocardio-graphy was performed and it confirmed the pres-ence of high velocity jet originating from thestenotic coronary sinus ostium. That was followedby the cardiac catheterization, and coronary sinusvenography demonstrated a tight ostium whichwas stented successfully. Her LV end-diastolic pres-sure was increased (29 mmHg) due to the coronarysinus stenosis. Three months after the stent, the pa-tient’s symptoms markedly improved. This is a good

Figure 1 Parasternal long axis view from 54-year-oldwoman with increasing shortness of breath. The right ven-tricle (RV) is enlarged. In real time 2-D echocardiographicimage, the left ventricle (LV) was of normal size and func-tion. Subsequent views from subcostal window showeda large secundum atrial septal defect with left-to-rightshunt resulting in right ventricular volume overload.Ao ¼ Aorta, LA ¼ left atrium.

example of identifying an uncommon but clinicallyimportant structural abnormality responsible forpatient’s heart failure. I believe it would have beendifficult for other diagnostic imaging tests to be ableto identify such an uncommon abnormality so readily.

Heart failure due to systolic dysfunction isrelatively easy to diagnose by echocardiographywhich demonstrates a dilated left ventricle witha reduced ejection fraction. In systolic heartfailure, however, echocardiography has manyother roles beyond the recognition of systolic heartfailure since dilatation of the LV results in alter-ation of intracardiac geometry and hemodynamicsleading to increased morbidity and mortality. Newmedication strategies, devices, and surgical ther-apies are being developed to manage systolic heartfailure. Echocardiography has been essential inidentifying structural and hemodynamic abnormal-ities associated with systolic heart failure and tomonitor patient’s response to new and noveltherapies. It has been well documented for thelast 20 years that the left ventricular volume is oneof the most important prognostic indicators for thepatients with ischemic or dilated cardiomyopa-thy.4,5 White and his associates demonstratedthat worsening of patient’s survival was associatedwith increased end systolic volume after an acutemyocardial infarction. With LV remodeling anda progressive dilatation of the heart, although ini-tially an adaptive process to the myocardial dam-age, the left ventricle becomes more spherical,and the mitral annulus is dilated with apical dis-placement of the mitral leaflets causing functionalmitral valve regurgitation.6e8 A combination of LVdilatation and mitral valve regurgitation resultsin increased filling pressure, symptoms of heartfailure, and death. Grayburn and his collaboratorsin the BEST trial demonstrated that among variousechocardiography parameters, LV end-diastolicvolume, deceleration time of early diastolic mitralinflow velocity which is a surrogate for pulmonarycapillary wedge pressure, and severity of mitralvalve regurgitation were the strongest predictorsof survival in patients with ejection fraction of35% or less5 (Fig. 2). Since a smaller LV with lessamount of mitral regurgitation is associated witha better functional capacity and survival, thereare many management attempts to reduce theleft ventricular volume and MR with the use ofsurgical, pacing, and other percutaneous devices,as well as medications and surgical revasculariza-tion. In this context, echocardiography is a valu-able tool since it can measure or assess LVdimensions, volumes, sphericity index, severity ofmitral regurgitation, diastolic filling parameters,and pulmonary artery systolic pressure.

6 J.K. Oh


Dow

nloaded from

SAVE trial was one of the first studies whereechocardiography was used in a systolic heartfailure trial to show that adverse cardiac eventswere associated with increased left ventricularsize after acute myocardial infarction.8 Biventricu-lar pacing has been found to improve functionalstatus and survival of patients with reduced LVEFand ventricular conduction delay.9e11 Echocardi-ography has been used to assess the success ofthe device therapy by documenting smaller LV vol-ume, increased LVEF, increased dP/dT, and lessamount of mitral regurgitation which are all signsof reverse remodeling. Another potentially criticalrole of echocardiography in biventricular pacing isto identify the patients who respond positively tothe therapy since 30e50% of patients who receivebiventricular pacing may not improve. Currently,mechanical dyssnchrony measures obtained bytissue Doppler imaging velocities are considered tobe best in that role. Time intervals from the onsetof QRS to the time of peak systolic velocity of 2 to12 segments of LV are used to determine themaximum or standard deviation interval.12,13 Themaximum interval of 100 msec or standard

Figure 2 Survival curve of patients with reduced leftventricular ejection fraction in the Best Trial. LVEDVI ¼left ventricular end diastolic volume index, DT ¼ decel-eration time, and VCW ¼ vena contracta width (Repro-duced with permission from Ref. 5).

deviation interval of 34 msec were found to havethe highest predictive power for identifying positiveresponder. However, we have found that suchintervals by TDI are present in normal subjects andpatients without reduced ejection fraction orconduction delay.14 Instead, time intervals fromthe QRS to the peak negative strain from 12 seg-ments correlated best with LV remodeling parame-ters and hemodynamic improvement after cardiacresynchronization therapy.15 It is, however, timeconsuming and challenging to obtain tissue Doppleror strain imaging from 12 segments. Further clinicalexperience and simplification of current strain im-aging procedure are required to use echocardiogra-phy in identification of positive responders to CRTroutinely in clinical practice.

Another innovative device for reducing LV vol-ume or achieving reverse LV remodeling is a passiveconstraint device called ACORN.16 Mayo ClinicEchocardiography Core Lab has been working forthe US clinical trial of this device which was re-cently completed. I would like also to acknowledgethat many European centers were the initial clini-cal sites for development and testing the safety ofthis device. There was a progressive and significantreduction in left ventricular volume measured byechocardiography in patients treated with thisdevice compared to a control group. The leftventricular volume reduction was associated withsignificant reduction in major cardiac proceduresincluding cardiac transplantation.

Surgical Ventricular Restoration or Reconstruc-tion (SVR) procedure which was pioneered byDr. Dor in 1984 is a surgical attempt to restore LVsize and shape by excluding the scarred and akineticapical segments.17 The results from SVR procedurein more than 1000 patients were reported with anoverall 30 day mortality of 5.3%.18 LVEF improvedfrom 29% to 39% and LV end-systolic volume indexdecreased from 80 mL/m2 to 56 mL/m2. Patientswith a larger ventricle and worse functional classpreoperatively had worse outcome after SVR. How-ever, this report did not come from a randomizedtrial, and there is no way to compare this resultwith the intensive medical therapy, other devicetherapy, or revascularization therapy alone inpatients with reduced ejection fraction and aki-netic apex. If SVR procedure has an incrementalbeneficial effect, echocardiography should behelpful in identifying ideal patients with the largeleft ventricle, akinetic scared apex, reduced LVEFas well as normally functioning inferior-lateralwall. Structural and hemodynamic response tothe SVR can be readily assessed by comprehensiveechocardiogram. However, the long-term struc-tural and functional result of SVR procedure is not



Dow

nloaded from

yet clearly understood and further clinical trial isrequired to demonstrate beneficial effects of SVRcompared to other device and medical/surgicaltherapy. Also, the long axis dimension becomesshorter by SVR, but the short axis dimension andmitral annulus may continue to dilate in some pa-tients which may lead to more severe mitral valveregurgitation and higher diastolic filling pressure.Hence, echocardiography should be able to providean insight and understanding of the long-termstructural and hemodynamic outcome of the SVR.

In patients with severe coronary artery disease,reduced LVEF, and heart failure, there has notbeen a clinical trial data to guide our treatment. Issurgical revascularization better than medicaltherapy? What if the patient has akinetic apex?Does additional SVR procedure improve functionalclass and survival of the patients with ischemiccardiomyopathy? These important clinical ques-tions are being investigated by STICH (SurgicalTreatment In patients with Coronary artery diseaseand Heart failure) trial in which patients withischemic cardiomyopathy and revascularizablecoronary disease are randomized to intensivemedical therapy, coronary artery bypass surgery,and bypass plus SVR. The first phase of STICH trialof comparing two surgical strategies was com-pleted in the early 2006 with 1000 patientsenrolled. The comparison of medical therapyversus surgical therapy will be completed in early2007. STICH trial is truly an international trial withmany European clinical centers participating. Infact, STICH investigators in Poland has enrolled thelargest number of patients in the study and shouldbe congratulated on their fantastic effort. Thisstudy will provide a very rich imaging data since allpatients in the trial have baseline, four month, andtwo year follow-up echocardiograms and selectedpatients will have cardiac magnetic resonanceimaging and nuclear studies. Mayo Clinic EchoCore Lab is analyzing all echocardiography datafor the STCIH to address the following two aims;1) To determine if the use of echocardiography atbaseline to assess LV systolic and diastolic dys-function, cardiac hemodynamics, and valvular re-gurgitation in the STICH Trial population canidentify a subgroup that derives a substantialbenefit or incurs a significant harm from surgicalintervention as determined by all-cause mortality.2) To assess the effects of the different random-ized treatment strategies on echocardiographi-cally-derived parameters of cardiac systolic anddiastolic dysfunction, hemodynamics, and theseverity of valvular regurgitation at 4 months and2 years as compared to baseline as they relate tochanges in HF functional class and the clinical

endpoints of death and HF hospitalization. Echo-cardiography data from 2000 patients from allover the world with ischemic cardiomyopathy willbe analyzed independently and correlated withclinical and other laboratory data. We are lookingforward to sharing many interesting results fromthis rich STICH echocardiographic data.

Now, I like to turn my discussion to assessmentof diastolic function in patients who present withsymptoms of heart failure. It has been wellestablished that about 50% of patients with newonset of heart failure do have a normal ejectionfraction.19,20 Some have suggested that this condi-tion be called heart failure with preserved ejec-tion fraction (HFPEF) since intrinsic diastolicdysfunction may not be a unifying underlyingmechanism for heart failure.21,22 However, the ter-minology is also troublesome since there are manyconditions other than a myocardial disease whichcan cause heart failure with preserved ejectionfraction such as constrictive pericarditis, severevalvular heart disease, congenital heart disease,and intracardiac tumor. Since the most commonetiology for heart failure with preserved ejectionfraction and no other structural abnormalities isdue to diastolic dysfunction, diastolic heart failureappears to be a better terminology to describe thisdisease entity.20 This condition is often difficult to bediagnosed by 2-D echocardiography alone althoughthere are diagnostic clues in two-dimensional echo-cardiography if carefully looked for. Comparedto a normal heart, the heart with diastolic dys-function has less vigorous mitral annulus motionduring systole and early diastole corresponding tothe period of active myocardial relaxation, theleft ventricular wall tends to be thicker thannormal, and the left atrial size is increased. Theseabnormalities suggest that the patient’s diastolicdysfunction is abnormal and a more detailedDoppler assessment is required. Doppler evalua-tion of diastolic function requires the following:Early (E) and late (A) diastolic velocities of themitral inflow, Deceleration time (DT) of mitral Evelocity, early diastolic mitral annulus velocity(Ea), pulmonary vein velocities, hepatic veinvelocities, and mitral inflow propagation velocityobtained by color M-mode (Fig. 3). It usually beginswith recording of mitral inflow velocities which re-flect the relationship between LA and LV pres-sures. This relationship, hence the configurationof mitral inflow velocities, alters predictably withimpaired myocardial relaxation and increased fill-ing pressure. Mitral inflow velocity pattern of nor-mal subjects is characterized by predominantearly diastolic filling with DT of 200 � 40 msec.Impaired myocardial relaxation which is usually an

8 J.K. Oh


Dow

nloaded from

initial diastolic abnormality decreases E velocity,increases A velocity, and lengthens DT. On theother hand, increased filling pressure has anopposite effects on these Doppler parameters.Therefore, during an early stage of diastolicdysfunction with impaired myocardial relaxationand normal filling pressure, mitral inflow velocityrecording alone is usually sufficient for diastolicfunction assessment. As filling pressure increaseswith worsening of diastolic function, mitral E ve-locity increases with shorten DT and A velocity de-creases with shorter flow duration. When diastolicdysfunction is far advanced and/or filling pressureis markedly elevated, mitral inflow velocity isagain sufficient. However, mitral inflow velocitypattern may appear normal (so called pseudonor-malized mitral inflow) in patients with moderateelevation of filling pressure. Even in this situation,a careful attention to isovolumic relaxation timeand atrial flow duration is helpful in distinguishingpseudonormalized from truly normal mitral inflowvelocities. It is important to keep in mind thatpseudonormalized mitral inflow is a product of im-paired myocardial relaxation and moderate eleva-tion of filling pressure. In this context, Dopplerparameter to evaluate myocardial relaxation iscritical in our evaluation of diastolic function. Tis-sue Doppler Imaging allows recording of myocar-dial tissue velocities and it turned out that earlydiastolic velocity of the mitral annulus (Ea) corre-lates well with tau or the status of LV myocardial

Figure 3 Schematic of pulse-wave Doppler of mitralinflow, color M-mode demonstrating mitral inflow propa-gation velocity, and tissue Doppler of the septal mitralannulus in normal diastolic function and various stagesof diastolic dysfunction. (Reproduced with permissionfrom reference 28.)

relaxation.23,24 Ea is lower at the septal annulus(normal� 10 cm/sec) compared to the lateral an-nulus (normal� 15 cm/sec) and at our institutionwe use the septal annulus in all patients exceptwhen there is regional wall motion abnormality inthe basal inferior septum. Pulmonary vein, hepaticvein, and the color M-mode of mitral inflow may bebeneficial in certain cases to evaluate diastolicfunction and estimated filling pressures. Pulmo-nary venous flow velocity shows progressive reduc-tion of systolic forward flow velocity compared tothe diastolic flow velocity with increased fillingpressure (Fig. 4). In all phases of diastolic dysfunc-tion, early diastolic velocity of the mitral annulus(Ea) is reduced and it does not increase with highfilling pressure. Since mitral E velocity increaseswith a higher filling pressure and tissue DopplerEa remains reduced, the ratio of the mitral E ve-locity and the mitral annulus velocity of Ea (E/Ea) has a good correlation with the pulmonary cap-illary wedge pressure or left ventricular fillingpressure.25,26 E/Ea ratio >15 usually indicates pul-monary capillary wedge pressure of greater than20 mmHg when septal Ea is used; and if the ratio<8, the filling pressure is usually normal and forthe ratio between 8 and 15, we may require fur-ther diastolic parameters to estimate diastolic fill-ing pressures.25 As mentioned earlier, motion orvelocity of the mitral annulus can also be appreci-ated from parasternal long axis and apical fourchamber views of 2-dimensional echocardiogra-phy. Assessment of filling pressure by 2-dimen-sional and Doppler echocardiography is useful inpatients with systolic dysfunction as well as in pa-tients with normal LVEF. Numerous studies havedemonstrated a good prognostic value of Dopplerechocardiographic assessment of filling pressures.We also have demonstrated that E/Ea >15 was oneof the strongest predictors for survival after acutemyocardial infarction whether LVEF was normal orabnormal.27,28 This work of retrospective reviewwas carried out by G. Hillis who was a research fel-low from the United Kingdom, and he has led a pro-spective study upon return to his University ofassessing prognostic power of clinical, laboratory,and echocardiographic parameters in patientswith acute myocardial infarction. His group pre-sented at the last AHA meeting that Plasma BNP,LVEF, and septal E/Ea were strong independent pre-dictors of morbidity following acute myocardial in-farction. Of them, septal E/Ea ratio was found tobe the single most powerful predictor of death.29

In our retrospective review, LA volume was foundto be also prognostic for death after acute MI.30

Therefore, estimation of filling pressure shouldalso be an integral part of echocardiographic



Dow

nloaded from

Figure 4 A: Parasternal long axis view from a middle-aged man with heart failure. It shows normal left ventricularcavity dimension with increased wall thickness and enlarged left atrium. The left ventricular systolic function waswell preserved. There was a small amount of pericardial effusion (*) circumferentially and large pleural effusion(PL). LA ¼ left atrium, RV ¼ right ventricle, VS ¼ ventricular septum. B: A composite of mitral inflow (left upper), pul-monary vein (right upper), tissue Doppler velocity of the mitral annulus (left lower), and color M-mode of mitral inflowdemonstrating flow propagation velocity (right lower). Mitral inflow velocity shows E/A ratio of 2 and decelerationtime of 160 msec which is in the low normal range. Pulmonary venous flow velocity, however, shows marked reductionin systolic forward flow velocity (S) and decreased deceleration time of diastolic forward flow (D) velocity consistentwith increased filling pressure. Mitral annulus velocity shows reduction in early diastolic velocity (E0) to 6 cm/secwhich represents marked impairment of myocardial relaxation. Flow propagation is slightly reduced (slope in whitecolor of the highest velocity in the center of mitral inflow shown).

evaluation in patients with reduced LVEF. Let methen share with you how echocardiographic evalua-tion of diastolic function should be used in patientswith symptoms of heart failure and normal LVEF. Iwill start with an example of a 70-year-old manwho presented with sudden onset of dyspnea to

the emergency room. Chest X-ray suggested pulmo-nary edema, but serum BNP was within normallimits, and the echocardiography showed normalejection fraction and mild increase in LV wall thick-ness. Because of the normal BNP and ejection frac-tion, the patient’s clinical and chest X-ray findings

10 J.K. Oh


Dow

nloaded from

were thought to be of non-cardiac etiology initially.However, when his echocardiogram was reviewedmore carefully, it was obvious that LV myocardial re-laxation was markedly impaired (mitral annulus mo-tion was reduced) and diastolic evaluation showedE/Ea ratio was 22, much higher than the 15 cutoffvalue. Subsequent evaluation showed a criticalleft main coronary artery disease and LV end dia-stolic pressure of 28 mmHg. When plasma BNP andmitral E/Ea0 ratio were correlated with pulmonarycapillary wedge pressure, correlation was betterfor the mitral E/Ea ratio compared to the BNP.31

BNP has been shown to be helpful in triaging pa-tients with dyspnea in the emergency room, butwe should be mindful that the sensitivity and thespecificity of BNP for increased filling pressure isless than optimal, and probably lower than echocar-diographic estimation of filling pressure. Even whenhigh filling pressure is detected by BNP in patientswith heart failure, echocardiography will be neces-sary to establish an etiology and to provide an opti-mal treatment. In patients with heart failure andnormal LVEF, optimal therapy has not been estab-lished although there are several studies demon-strating some benefits in this group of patients. Ianticipate several large clinical trials to identifythe most efficacious management strategy in dia-stolic heart failure and echocardiography will bean integral part of those trials for identification ofideal patients and monitoring the response of vari-ous therapies.

Constrictive pericarditis is an elusive entitywhich causes heart failure in patients with normalLVEF and is potentially curable. Its hemodynamicmanifestations are similar to those of myocardialdiseases and it is often challenging to differentiateconstrictive pericarditis from restrictive myopathyor other myocardial diseases. With comprehensiveechocardiography, we now should be able todiagnose constrictive pericarditis in almost allcases non-invasively. There are two unique fea-tures of constriction which are not present inmyocardial diseases: 1. Dissociation between in-trathoracic and intracardiac pressures; and 2.Exaggerated interventricular dependence.32 Thesetwo features in constriction result in characteristicdiastolic filling pattern which can be reliably re-corded by Doppler echocardiography. Typically,mitral inflow early diastolic velocity decreaseswith inspiration and hepatic vein diastolic reversalflow increases with expiration. However, the respi-ratory variation of mitral inflow velocities may notbe present in some patients.33 Tissue Doppler re-cording of mitral annulus velocity has added an im-portant diagnostic value in constrictive pericarditissince Ea velocity which is reduced in almost all

myocardial diseases is increased in constrictivepericarditis34,35 (Fig. 5). Other subtle findingssuch as ventricular septal motion changes with res-piration and septal shudder in constriction can beappreciated from careful evaluation of 2-D echo-cardiography. At our institution, more careful at-tention to these echocardiographic features ofconstrictive pericarditis allowed increased detec-tion of this entity in patients with heart failurewhich has been partly responsible for higher num-ber of pericardiectomies (from 15 in 1990 to morethan 60 in 2005). More reliable detection of con-strictive pericarditis has been a valuable productof assessing diastolic function by 2-D, Doppler,and myocardial tissue imaging.

Let me summarize what I have discussed so farregarding the role of echocardiography in patientswith symptoms of heart failure. When heart failureis clinically suspected, echocardiography is thesingle most important diagnostic imaging tech-nique. In a half of patients with a final diagnosisof heart failure, transthoracic and/or transesopha-geal echocardiography demonstrates a majorstructural and/or functional abnormality such asin systolic heart failure, severe valvular heartdisease, congenital heart disease, or cardiomyop-athies. It is more difficult to diagnose heart failurewhen there are no obvious structural and func-tional abnormalities in 2-dimensional echocardiog-raphy. In the other half of patients with heartfailure, most likely diastolic heart failure, theremay be subtle abnormalities in 2-dimensionalechocardiography, but a firm diagnosis of diastolicheart failure requires Doppler and myocardial(or tissue Doppler) imaging. If mitral inflow velocitydemonstrates clearly a restrictive filling patternand/or high filling pressure at resting stage, thediagnosis of diastolic heart failure or constrictivepericarditis is usually secured. Tissue Dopplerimaging of the septal mitral annulus is helpful indifferentiating pseudonormal from true normalmitral inflow velocities, and differentiating con-strictive pericarditis from myocardial diseases. It ismore problematic when diastolic function evalua-tion shows relatively normal filling pressure orpredominantly impaired myocardial relaxation atrest. In this situation, right heart failure, pulmo-nary process, and non-cardiopulmonary processshould be considered. Another important conditionwe need to consider is exercise-induced increasedfilling pressure responsible for patient’s symptomsof shortness of breath with exertion. Therefore,we need to assess filling pressure with exercise aswell as at resting stage.

The mitral annulus Ea velocity which reflects thestatus of mitral relaxation does actually increases



Dow

nloaded from

Figure 5 Three patients with a similar degree of shortness of breath. Upper panel shows the mitral inflow velocity.Middle panel shows tissue Doppler velocity from the mitral septal annulus. The bottom row shows color M-mode of themitral inflow velocity. The left column is from a normal subject. Middle column from a patient with a myocardial dis-ease (restrictive cardiomyopathy). The right column is from a patient with constrictive pericarditis. The mitral inflowvelocity appeared the same in all three examples with E/A ratio of 2. However, early diastolic velocity of the mitralannulus is markedly reduced (5 cm/sec) in the patient with myocardial disease (in the center column) whereas earlydiastolic velocity (E0) of the mitral annulus is normal in normal subject and exaggerated in the patient with constric-tive pericarditis (right column, middle row). Color M-mode of mitral inflow demonstrates normal flow propagation inthe normal subject and the patient with pericardial disease. However, it is usually reduced (in the center column,bottom row, left color M-mode) in patients with myocardial disease but may be falsely normal when LV cavity is rel-atively small such as in hypertrophic cardiomyopathy for early stage of cardiac amyloid (in the center column, bottomrow, right color M-mode).

with exercise or increased preload when myocar-dial relaxation is normal, but Ea velocity does notaugment as much with higher transmitral pressuregradient or with exercise when relaxation is im-paired with increased tau. Therefore, we hypoth-esized several years ago that in normal subjectsmitral inflow E velocity and mitral annulus Eavelocity increase proportionately with a resultthat E/Ea remains less than 15 at resting stageand exercise.36,37 However, in patients with dia-stolic dysfunction and exertional dyspnea, E veloc-ity increases with exercise but the Ea velocityremains reduced or increases less with exercise.Therefore, E/Ea ratio increases with exercise com-pared with the resting stage in patients with

diastolic dysfunction. We used a supine bike ortreadmill exercise test for assessment of fillingpressure with exercise, and we analyzed diastolicparameters of mitral inflow and mitral annulus ve-locities which provided E/Ea ratio. Initially we per-formed diastolic stress test in healthy individuals.This work was initially carried out by Dr. F. Lulicfrom Croatia, and then completed by Dr. J. Hafrom Korea working in our laboratory.36,37 It wasfound that in normal subjects, mitral E velocity in-creased about 25% with exercise from baseline dataand deceleration time decreased an average of16 msec (from 192 to 176 msec) and Ea velocity in-creased also about 25%. Therefore, E/Ea ratio re-mained the same (mean of 6.6) at baseline and

12 J.K. Oh

by guest on April 19,

Dow

nloaded from

exercise in individuals with normal diastolic func-tion. Subsequently, we performed supine bike dia-stology stress test in patients with exertionaldyspnea and were able to identify three groups ofpatients. The first group of patients was found tohave E/Ea >15 (increased filling pressure) at restand the ratio remained elevated with exercise inall. The second group of patients had relativelynormal filling pressure and impaired myocardial re-laxation pattern on mitral inflow velocity at restwhich turned to a restrictive filling pattern withE/Ea >15 with exercise (Fig. 6). The third groupof patients had normal filling pressure and E/Ea<15 at rest and with exercise. The patients with in-creased filling pressure with exercise whether ornot they had normal filling pressure at rest (the firstand second groups), had shorter exercise durationcompared to the patients (the third group) withnormal filling pressure at rest and with exercise.Subsequently, we have performed simultaneouscardiac catheterization and Doppler echocardiog-raphy in patients who were undergoing exercisetesting in the cardiac catheterization laboratoryfor an evaluation of exertional dyspnea.38 In thispatient population, E/Ea >15 correlated well withpulmonary capillary wedge pressure >20 mmHg,not only at resting stage but also with exercise.Therefore, we believe that we now have a very re-liable non-invasive diagnostic technique to esti-mate diastolic filling pressures both at restingstage, and with exercise. Marwick and his group

2

have also demonstrated a similarly good correla-tion between E/Ea and PCWP at rest and exer-cise.39 Several investigators also have exploreduse of the duration between the onset of mitral Evelocity and the onset of mitral annulus Ea velocity.This duration or time interval has been reasonablywell correlated with the filling pressure and alsothe status of myocardial relaxation. I had an oppor-tunity to write an editorial to summarize thecurrent status of using diastolic parameters to esti-mate diastolic filling pressures, and I concluded inmy editorial from numerous investigations in thisarea that ‘‘now echocardiography is able to esti-mate left ventricular filling pressure under variousclinical conditions’’. When I submitted this edito-rial, the Journal suggested that the phrase, ‘‘nowechocardiography is able,’’ be changed to ‘‘echo-cardiography is now able to’’ which I did not thinkwas a big issue. I agreed to the change, but whenthe paper was published, I was horrified to seemy conclusion. Instead of ‘‘echocardiography isnow able to estimate left ventricular filling pres-sures’’, it read ‘‘echocardiography is not able toestimate left ventricular filling pressures’’ whichcompletely changed the conclusion of my edito-rial,40 but I hope that some of you who read theeditorial were able to see the obvious error.

Ladies and gentleman, I tried to summarize foryou today in my lecture the current clinical use ofechocardiography in patients with heart failure,not only systolic but also diastolic heart failure.

016

Figure 6 (Left) Baseline mitral inflow (top) and tissue Doppler (bottom) demonstrate grade 1 diastolic dysfunction(E velocity of 50 cm/sec and deceleration time of 250 msec) with impaired myocardial relaxation but normal fillingpressure with E/E0 ratio of 7, and tricuspid regurgitation velocity was also normal indicating normal pulmonary arterysystolic pressure. (Right) With 50 watts of supine bike exercise, E velocity increased to 85 cm/sec and decelerationtime shortened from 250 msec to 140 msec. E0 velocity, however, increased slightly from 7 to 8 cm/sec. Therefore,E/E0 ratio increased from 7 to 11. TR velocity increased from 2.4 m/sec to 3.8 m/sec (Reproduced with permissionfrom Ref. 37).



Dow

nloaded from

There are many exciting new technologies andapplications of echocardiography such as 3-dimensional echocardiography, strain imaging,and myocardial perfusion imaging. I am certainthat these new applications will strengthen andimprove the role of echocardiography in general,but more so in patients with symptoms of heartfailure. Undoubtedly, echocardiography is thesingle most useful test in patients with symptomsof heart failure. It is essential in the diagnosis andidentification of underlying etiology of heart fail-ure. However, we should move beyond the con-cept of echocardiography as only a diagnostic toolin patients with heart failure, hence, use thisversatile and easily available technology to gainpathophysiologic insights of various forms of heartfailure, to help identify or establish an optimaltherapy for the patients with heart failure, tomonitor treatment response, to prognosticate, andto be an important tool in clinical heart failuretrials.

Thanks again for the invitation to Euro EchoLecture 2005 and thank you all for listening.

References

1. Swedberg K, Cleland J, Dargie H, Drexler H, Follath F,Komajola M, et al. Guidelines for the diagnosis and treat-ment of chronic heart failure: executive summary (update2005). Eur Heart J 2005;26:1115e40.

2. Hunt S, Baker D, Chin M, Cinquegrani M, Feldmanmd A,Francis G, et al. ACC/AHA guidelines for the evaluationand management of chronic heart failure in the adult: exec-utive summary a report of the American College of Cardio-logy/American Heart Association Task Force on PracticeGuidelines. Circulation 2001;104:2996.

3. Remme W, McMurray J, Rauch B, Zannad F, Keukelaar K,Cohen-Solal A, et al. Public awareness of heart failure inEurope: first results from SHAPE. Eur Heart J 2005;26:2413e21.

4. Wong S, French J, Lydon A, Manda SO, Gao W, Ashton NG,et al. Relation of left ventricular sphericity to 10-year sur-vival after acute myocardial infarction. Am J Cardiol2004;94:1270e5.

5. Grayburn P, Appleton C, DeMaria A, Greenberg B, Lowes B,Oh J, et al. Echocardiographic predictors of morbidity andmortality in patients with advanced heart failure. TheBeta-blocker Evaluation of Survival Trial (BEST). J Am CollCardiol 2005;45:1064e71.

6. Yiu S, Sarano M, Tribouilloy C, Seward J, Tajik A. Determi-nants of the degree of functional mitral regurgitation inpatients with systolic left ventricular dysfunction. A quanti-tative clinical study. Circulation 2000;102:1400e6.

7. Lamas G, Mitchell G, Flaker G, Smith Jr SC, Gersh BJ,Basta L, et al. Clinical significance of mitral regurgitationafter acute myocardial infarction. Circulation 1997;96:827e33.

8. Sutton M, Pfeller M, Palppert T, Rouleau JL, Moye LA,Dagenais GR, et al. Quantitative two-dimensional echocar-diographic measurements are major predictors of adversecardiovascular events after acute myocardial infarction.

The protective effects of Captopril. Circulation 1994;89:68e75.

9. Yu C, Chau E, Sanderson J, Fan K, Tang M, Fung W, et al. TissueDoppler echocardiographic evidence of reverse remodelingand improved synchronicity by simultaneously delayingregional contraction after biventricular pacing therapy inheart failure. Circulation 2002;105:438e45.

10. Abraham W, Hayes D. Cardiac resynchronization therapy forheart failure. Circulation 2003;108:2596e603.

11. Bristow M, Saxon L, Boehmer J, Krueger S, Kass DA,DeMarco T, et al. Cardiac-resynchronization therapy withor without an implantable defibrillator in advanced chronicheart failure. N Engl J Med 2004;350:2140e50.

12. Yu C, Wing-Hong Fung J, Zhang Q, Chan C, Chan Y, Lin H,et al. Tissue Doppler imaging is superior to strain rate imag-ing and postsystolic shortening on the prediction of reverseremodeling in both ischemic and non-ischemic heart failureafter cardiac resynchronization therapy. Circulation 2004;110:66e73.

13. Gorscan J, Kanzaki H, Bazaz R, Dohi K, Schwartzman D.Usefulness of echocardiographic tissue synchronization im-aging to predict acute response to cardiac resynchroniza-tion therapy. Am J Cardiol 2004;93:1178e81.

14. Miyazaki C, Miller F, Karon B, Espinosa R, Khandheria B,Oh J. Measurement of intraventricular mechanical dyssyn-chrony by tissue Doppler imaging: is it time to peak velocitythe best parameter for dyssynchrony? J Am Coll Cardiol2005;45:1001e58.

15. Miyazaki C, Cha Y, Espinosa R, et al. Correlation betweenacute effect of cardiac resynchronization therapy and me-chanical dyssynchrony measured by tissue velocity andstrain parameters. J Am Coll Cardiol 2006;47:95A.

16. Mann D, Acker M, Jessup M, Sabbah H, Starling R, Kubo S.Rationale, design, and methods for a pivotal randomizedclinical trial for the assessment of a cardiac support devicein patients with New York Health Association class IIIeIVheart failure. J Cardiac Failure 2004;10:185e92.

17. Dor V, Di Donato M, Sabatier M, Montiglio F, Civaia F. Leftventricular reconstruction by endoventricular circularpatch plasty repair: a 17-year experience. Semin ThoracCardiovasc Surg 2001;13:435e47.

18. Athanasuleas C, Buckberg G, Stanley A, Siler W, Dor V,Donato M, et al. Surgical ventricular restoration in thetreatment of congestive heart failure due to post-infarctionventricular dilation. J Am Coll Cardiol 2004;44:1439e45.

19. Senni M, Rodeheffer R, Tribouilloy C, Evans J, Jacobsen S,Bailey K, et al. Use of echocardiography in the managementof congestive heart failure in the community. J Am Coll Car-diol 1999;33:164e70.

20. Oh J, Hatle L, Tajik A, Little W. Diastolic heart failure canbe diagnosed by comprehensive two-dimensional and Dopp-ler echocardiography. J Am Coll Cardiol 2006;47:500e6.

21. Mauer M, Spevack D, Burkhoff D, Kronzon I. Diastolic dys-function: can it be diagnosed by Doppler echocardiography?J Am Coll Cardiol 2004;44:1543e9.

22. Burkhoff D, Mauer M, Packer M. Heart failure with a normalejection fraction: Is it really a disorder of diastolic function?Circulation 2003;107:656e8.

23. Sohn D, Chai I, Lee D, Kim HC, Kim HS, Oh BH, et al. Assess-ment of mitral annulus velocity by Doppler tissue imaging inthe evaluation of left ventricular diastolic function. J AmColl Cardiol 1997;30:474e80.

24. Nagueh S, Sun H, Kopelen H, Middleton K, Khoury D. Hemo-dynamic determinants of the mitral annulus diastolic veloc-ities by tissue Doppler. J Am Coll Cardiol 2001;37:278e85.

25. Ommen S, Nishimura R, Appleton C, Miller FA, Oh JK,Redfield MM, et al. Clinical utility of Doppler

14 J.K. Oh

by guD

ownloaded from

echocardiography and tissue Doppler imaging in the estima-tion of left ventricular filling pressures. Circulation 2000;102:1788e94.

26. Nagueh S, Middleton K, Kopelen H, Zoghbi W, Quinones M.Doppler tissue imaging: a non-invasive technique for evalu-ation of left ventricular relaxation and estimation of fillingpressures. J Am Coll Cardiol 1997;30:1527e33.

27. Hillis G, Moller J, Pellikka P, Gersh B, Wright R, Ommen S,et al. Non-invasive estimation of left ventricular filling pres-sure by E/e0 is a powerful predictor of survival after acutemyocardial infarction. J Am Coll Cardiol 2004;43:360e7.

28. Moller J, Pellikka P, Hillis G, Oh J. Prognostic importance ofdiastolic function and filling pressure in patients with acutemyocardial infarction. Circulation 2006;114:438e44.

29. Kruszewski K, Barclay J, Scott A, Croal B, Oh J, Hillis G.Non-invasive assessment of left ventricular filling pressurefollowing acute myocardial infarction: a prospective studyof the relative prognostic utility of clinical, biochemical,and echocardiographic parameters. Circulation 2005;112:II-499.

30. Moller J, Hillis G, Oh J, Seward JB, Reeder GS, Wright RS,et al. Left atrial volume. A powerful predictor of survivalafter acute myocardial infarction. Circulation 2003;107:2207e12.

31. Dokainish H, Zoghbi W, Lakkis N, Al-bakshy F, Dhir M,Quinones M, et al. Optimal non-invasive assessment of leftventricular filling pressures. A comparison of tissue Dopplerechocardiography and B-type natriuretic peptide in patientswith pulmonary artery catheters. Circulation 2004;109:2432e9.

32. Hatle L, Appleton C, Popp R. Differentiation of constrictivepericarditis and restrictive cardiomyopathy by Dopplerechocardiography. Circulation 1989;79:357e70.

es

33. Ha J, Oh J, Ommen S, Ling L, Tajik A. Diagnostic value ofmitral annular velocity for constrictive pericarditis in theabsence of respiratory variation in mitral inflow velocity.J Am Soc Echocardiogr 2002;15:1468e71.

34. Garcia M, Rodriguez L, Ares M, Griffin B, Thomas J, Klein A.Differentiation of constrictive pericarditis from restrictivecardiomyopathy: assessment of left ventricular diastolicvelocities in longitudinal axis by Doppler tissue imaging.J Am Coll Cardiol 1996;27:108e14.

35. Ha J, Ommen S, Tajik A, Barnes M, Ammash N, Gertz M,et al. Differentiation of constrictive pericarditis fromrestrictive cardiomyopathy using mitral annular velocityby tissue Doppler echocardiography. Am J Cardiol 2004;94:316e9.

36. Ha J, Lulic F, Bailey K, Pellikka Pa, Seward JB, Tajik AJ,et al. Effects of treadmill exercise on mitral inflow andannular velocities in healthy adults. Am J Cardiol 2003;91:114e5.

37. Ha J, Oh J, Pellikka P, Ommen S, Stussy V, Bailey K, et al.Diastolic stress echocardiography: a novel invasive diagnos-tic test for diastolic dysfunction using supine bicycleexercise Doppler echocardiography. J Am Soc Echocardiogr2005;18:63e8.

38. Talreja D, Nishimura R, Oh J. Non-invasive parameters of di-astolic function reflect invasively measured filling pressuresduring exercise. Circulation 2004;110:II-474.

39. Burgess M, Jenkins C, Sharman J, Marwick T. Diastolicstress echocardiography: hemodynamic validation andclinical significance of estimation of ventricular fillingpressure with exercise. J Am Coll Cardiol 2006;47:1891e

900.40. Oh J. Echocardiography as a non-invasive swan-ganz cathe-

ter. Circulation 2005;111:3192e4.

t on April 19, 2016

Documents

Echocardiography Diagnosis