Evaluation of Right and Left Ventricular Diastolic Filling

Evaluation of Right and Left Ventricular Diastolic Filling

Ares Pasipoularides

Received: 7 January 2013 /Accepted: 27 March 2013 /Published online: 13 April 2013# Springer Science+Business Media New York 2013

Abstract A conceptual fluid–dynamics framework for dia-stolic filling is developed. The convective deceleration load(CDL) is identified as an important determinant of ventric-ular inflow during the E wave (A wave) upstroke.Convective deceleration occurs as blood moves from theinflow anulus through larger-area cross-sections toward theexpanding walls. Chamber dilatation underlies previouslyunrecognized alterations in intraventricular flow dynamics.The larger the chamber, the larger becomes the endocardialsurface and the CDL. CDL magnitude affects strongly theattainable E wave (Awave) peak. This underlies the conceptof diastolic ventriculoannular disproportion. Large vortices,whose strength decreases with chamber dilatation, ensueafter the E wave peak and impound inflow kinetic energy,averting an inflow-impeding, convective Bernoulli pressurerise. This reduces the CDL by a variable extent dependingon vortical intensity. Accordingly, the filling vortex facili-tates filling to varying degrees, depending on chamber vol-ume. The new framework provides stimulus for functionalgenomics research, aimed at new insights into ventricularremodeling.

Keywords Diastolic filling vortex . Ventricular chambervolumes . Diastolic function . Heart failure . Ventricularremodeling . Convective deceleration load . Diastolicventriculoannular disproportion . Functional genomicsof intracardiac blood flow

“Cardiology is flow.” Richter Y. & Edelman E.R.(2006). Circulation 113, 2679.

Diastolic cardiac dysfunction as a component of heart failureis nowadays sufficiently recognized as to be a part of coding forcongestive heart failure (CHF) in the International Classificationof Diseases (ICD-10, codes I50.30–33). This WHO recognitionof diastole has been accompanied by major strides in the ap-praisal of diastolic function [1], made possible by technologyencompassing multisensor cardiac catheterization [1–4] andnew digital imaging modalities [1, 4]. It is now widely appreci-ated that enhancing diastolic filling has clinical merit and thatthe significance of diastolic dysfunction is far reaching [5].Nonetheless, the fact that in health and disease, diastolic dy-namics are dependent on a large number of factors and theirinteractions [1, 6] has complicated evaluation of the multifacto-rial causes of the observed ventricular filling abnormalities.

Remodeling with ventricular enlargement has surfaced as acentral mechanism, relevant to heart failure progression and toprediction of outcomes, as pathophysiologic understanding ofheart failure has progressed in the past three to four decades[1]. In primary systolic myocardial dysfunction and failure,ventricular enlargement occurs and entails a greater increasein transverse than in long chamber axis (sphericalization). Inprimary diastolic myocardial dysfunction, the ejection fractionis normal or higher than normal and chamber size isunchanged or decreased. Most intriguing is that ventriculardiastolic function is frequently abnormal in patients withprimary systolic dysfunction and failure. The present surveyconsiders evidence that besides leading to increased wallstresses and impaired systolic function [7, 8], ventricularchamber enlargement per se hampers diastolic filling, thusimpairing ventricular diastolic function.

This viewpoint stems from integrative research findingsand concepts, which pertain to interactions of cardiac fluiddynamic mechanisms with atrioventricular (AV) dynamic ge-ometry and inflow patterns, and from diastolic function alter-ations accompanying chamber enlargement in animal models

A. Pasipoularides (*)Department of Surgery, Duke University School of Medicine,HAFS-7th floor, DUMC 3704,Durham, NC 27710, USAe-mail: [email protected]

J. of Cardiovasc. Trans. Res. (2013) 6:623–639DOI 10.1007/s12265-013-9461-4

of heart disease and failure [4, 9, 10]. The clinical studies thatwe will review, which confirm the underlying fluid dynamicsprinciples, also demonstrate the value of these simple butuseful principles in managing heart failure with remodelingand chamber dilatation.

Ventricular Enlargement Does Not Impair MyocardialDiastolic Function

Ventricular diastole encompasses two main features determin-ing dynamics: relaxation and compliance. The interplay be-tween all the factors determining ventricular filling is complexand quite protean, reflecting continual adaptation to shifts in theoperating environment (see Fig. 1). The inflow velocity timecourse, in particular, may be affected by changes in myocardialrelaxation and ventricular compliance properties, atrial pressureand compliance, and atrioventricular valvular morphology.Inflow valve opening in the physical sense and the onset ofrapid ventricular filling coincide well with the instant of rever-sal of the atrioventricular pressure difference across the atrio-ventricular valve cusps [1]: blood rushes into each ventricle,

which is at its end systolic volume (ESV). Early diastolic fillingis rapid because of continuing ventricular relaxation with de-clining RV/LV chamber pressures. At normal or abnormallylow heart rates, the rapid ventricular filling phase evolves intothe slow filling phase; at high normal or abnormally fast rates, itmerges with the active filling that is caused by atrial systole.The initial phase of the translational research, which led to theintegrative diastolic function framework that is depicted inFig. 1, entailed baseline and longitudinal studies on subacute-to-chronic canine surgical models of RV volume overload(VO), pressure overload (PO), and myocardial ischemia (IS).We utilized right-heart multisensorMillar catheters, real-time 3-D echocardiography, and specially designed pulse-transit ultra-sonic dimension transducers [9, 11, 12]. Only chronicallyinstrumented awake dogs were studied, because of importantlimitations of acute studies under conditions of anesthesia,recent surgery, and open chest [11, 13, 14].

Changes from control ensuing in myocardial and ventric-ular diastolic active and passive wall mechanics in theanimal disease models are presented and discussed in detailin a companion paper in JCTR [15]. Here, I will summarizethe relevant findings regarding myocardial relaxation

Fig. 1 The determinants ofventricular inflow patterns anddiastolic filling include factorsintrinsic and extrinsic to the rightand left ventricles. The newdiastolic function paradigm(right) complements and acts inparallel with the traditional one(left): changes in RV/LV size,diastolic myocardial dynamics,and wall motion patterns causeconcomitant alterations in themagnitude of the convectivedeceleration load (CDL) and theproperties of the diastolic vortex,which affect inflow patterns andventricular filling. A atrial,V ventricular, P pressure, t time,DVAD diastolicventriculoannular disproportion,Grad P gradient (adapted fromPasipoularides [1], bypermission of PMPH-USA)

624 J. of Cardiovasc. Trans. Res. (2013) 6:623–639

dynamics and passive filling pressure vs. volume relationsand myocardial compliance. First, I must note that earlydiastolic filling is normally dominant (E wave), with a sup-plementary increment associated with atrial contraction (Awave). The main influences on early ventricular inflow rateand volume pertain to myocardial relaxation dynamics and aredifferent from those affecting late filling which commonlyrelate to passive ventricular and myocardial compliance.

Myocardial Relaxation Dynamics

The processes going on during the early filling period areneither purely passive nor purely active. This led me by1980 to postulate that total measured left ventricular diastolicpressure (PM) is determined by two overlapping processes,namely, the myocardial relaxation-induced decay of activelydeveloped pressure (PR) and the concurrent buildup of passivefilling pressure (P*), and to develop a comprehensive mathe-matical model for diastolic dynamics, as I have detailed in arecent survey [6]. Further application of this model hasallowed comprehensive evaluations of the role of active andpassive dynamic factors in diastolic ventricular mechanics inhealth and disease [1, 2, 6, 15].

PR ¼ P0 � exp �t=tð Þ þ Pb ð1Þ

An exponential model (Eq. 1) characterized accurately thedecay of isovolumic RV pressure, PR, with three parameters:P0, τ, and Pb. The time constant of relaxation, τ (tau), is themost important among them; it quantifies the rate of myocar-dial relaxation and pressure decay with the reuptake of Ca2+

(which when bound to troponin C allows the cross-bridgemovement of the myofibrils resulting in force developmentand contraction) by the sarcoplasmic reticulum [16]. Pb is alsoimportant, because it represents the asymptotic level to which

the pressure would tend, if the decaying process were allowedto proceed indefinitely. Finally, when the pleural pressure doesnot deviate significantly from zero, the parameter P0 is indica-tive of the beginning level of the exponential isovolumic pres-sure decay. As is shown in Table 1, in contrast to pressureoverload and myocardial ischemia, no significant change fromcontrol was found with VO and the resultant chamber enlarge-ment [11, 15] in the RV time constant of relaxation, τ [11, 15].The only significant change in VO was a raised RV diastolicpressure asymptote, Pb, reflecting increased diastolic constraintfrom elevated right heart volumes [1, 2, 11, 14, 16]. Thesefindings suggest that the relaxationmechanism is unimpaired insubacute-to-chronic RV volume overload with chamber dilata-tion and does not adversely impact ventricular filling [11]. Theyare consistent with earlier LV findings by Zile et al. on acomparable canine VO model [17], and with subsequent clin-ical findings in children with diverse conditions producing RVvolume overload and chamber enlargement [18].

Filling Pressure vs. Volume Relationships

A sigmoidal model [1, 12, 15] for passive filling pressure vs.volume relations and the resultant myocardial complianceformulations showed that the maximum RV myocardial com-pliance, which is attained during early filling, was reducedsignificantly from control with pressure overload and ische-mia but not with VO. Statistical analyses, including ANOVAand Bonferroni–Holm statistics, were performed on the im-pact of each disease modality on model parameters and in-dexes of RV diastolic function. Important changes in passivefilling pressure vs. volume relations and diastolic propertiesare induced by the three disease modalities (see Table 2). As isshown in Table 2, PO alters myocardial compliance propertiessubstantially. It caused a decrease in maximum RV myocar-dial compliance as well as an increase in passive filling

Table 1 Isovolumic relaxation dynamics

P0 (mmHg) Pb (mmHg) τ (ms)

Condition (mean ± SD) Control 13.1±5.1 0.28±1.8 27.8±3.9

PO 56.2±19.1 −7.9±1.5 57.1±2.8

VO 11.3±5.1 5.3±2.9 26.8±6.1

IS 9.0±3.5 −1.95±1.1 41.4±13.0

ANOVA F statistic 43.33 51.62 31.81

P value 0.0001 0.0001 0.0001

PBONFa PO vs. Cl 0.0001 0.0001 0.0001

=0.0167 VO vs. Cl 0.2295 0.0001 0.3140

IS vs. Cl 0.0355 0.0038 0.0004

Values significantly different from control are presented in italics. For the three comparisons, a conservative critical value for the modified tstatistics is given using a significance level, PBONF, of (P/3)=(0.05/3)=0.0167 (reproduced from Pasipoularides et al. [11] by permission of TheAmerican Association for Thoracic Surgery)

SD standard deviation, PO pressure overload, VO volume overload, IS ischemiaa ANOVA with Bonferroni’s inequality

J. of Cardiovasc. Trans. Res. (2013) 6:623–639 625

pressure levels at both maximum and end-diastolic myocardi-al compliance. IS also decreased considerably the maximumcompliance that was attained during the early filling process.Therefore, both PO and IS impaired myocardial compliance[12, 15] and diastolic filling. On the contrary, VOwith chamberenlargement exhibited only a slightly decreased maximal com-pliance and a strikingly strong increase in end-diastolic myo-cardial compliance [12, 15] compared to control (see Table 2),which implies no impairment of diastolic ventricular function[2, 12, 15]. In contrast to pressure overload and ischemia, inVO the pressure–volume relationship shifts far to the right anddownward over much of the filling process [1, 12, 15]; thus,end-diastolic myocardial compliance actually increased in VOwith chamber enlargement compared to control, while end-diastolic pressure was unchanged. These RV results are againcongruous with LV data obtained on a similar VO caninemodel, by Zile et al. [17].

Fluid Dynamic Underpinnings for Filling Impairmentin Dilated Ventricles

The intriguing feature of the foregoing findings was thateccentric hypertrophy (chamber enlargement) consequent toVO resulted in minimal abnormalities relative to control inmyocardial properties. Since the RV chamber size in VO(EDV, 60±29 ml) increased markedly (P<0.05) from con-trol (45±21 ml), reflecting elongation and “slippage” ofmyocytes within the wall contributing to myocardial creepand remodeling [1, 12, 15], it was hypothesized that it mightbe responsible for unrecognized functional filling dynamicschanges relative to control. Therefore, it became apparentthat a fuller understanding of integrative diastolic functionin chamber enlargement would be forthcoming only byexpanding the scope of the investigations beyond myocar-dial mechanics, which consider exclusively myocardial

relaxation and wall compliance changes affecting diastolicventricular function.

A comprehensive examination of ventricular filling wastherefore undertaken, utilizing combined sonomicrometric,digital imaging, and computational fluid dynamics methods,to look for fluid dynamic underpinnings of diastolic dysfunc-tion in enlarged ventricles [9, 10, 19, 20]. Such mechanismswould significantly alter the prevalent understanding of fac-tors contributing to diastolic filling impairment in heart failurewith remodeling.

The Functional Imaging Method for VisualizingIntracardiac Flow

The ensuing fluid dynamics studies [9, 10, 19] revealed im-portant but previously unrecognized mechanisms responsiblefor the filling impairment in the volume overloaded, dilatedventricles. These underlying mechanisms were revealed andinvestigated by the functional imaging (FI) method [1, 9]. Themethod comprises real-time, 3-D echocardiographic (RT3D)and sonomicrometric measurements, combined with compu-tational fluid dynamics simulations of the intracardiac flowfield [1, 4, 20, 21]. It was applied to ascertain diastolic fillingdynamics in individual dogs, studied at control and withchamber enlargement induced by RV subacute-to-chronicVO.With the use of functional imaging [9], our investigations[10, 19] were the first to show in detail the kinematic anddynamic characteristics of the diastolic filling vortex insidethe right ventricle. Its LV counterpart had already been shownin the pioneering investigations of Taylor and Wade andBellhouse (see below).

Although RT3D images [1, 9] were used in these studies,the pertinent techniques are applicable generally to moderndigital imaging methods [22, 23]. Robert Hooke, the EnglishLeonardo Da Vinci, had argued in his Micrographia thatscientific progress would depend on replacing or reinforcing

Table 2 Myocardial compliance

MCmax (kPa−1) PMCmax (Pa) MCed (kPa

−1) PMCed (Pa)

Condition (mean ± SD) Cl 0.7276±0.217 315±226 0.0466±0.0633 773±346

PO 0.2850±0.382 1,507±785 0.0235±0.0254 2,427±973

VO 0.502±0.263 137±168 0.1770±0.1935 760±453

IS 0.502±0.120 352±227 0.0563±0.0689 680±187

ANOVA F statistic 5.02 18.73 3.38 18.52

P value 0.0060 0.0001 0.0306 0.0001

PBONF=0.0167 PO vs. Cl 0.0016 0.0001 0.2020 0.0001

VO vs. Cl 0.0232 0.0407 0.0131 0.4768

PHOLM=0.0250 IS vs. Cl 0.0102 0.3619 0.3731 0.2714

Values are means ± SD. Italicized values are significantly changed from CL (reproduced from Pasipoularides et al. [12] by permission of TheAmerican Physiological Society)

MCmax maximum right ventricular (RV) myocardial compliance, PMCmax RV passive filling pressure at MCmax, MCed end-diastolic RV myocardialcompliance, PMCed RV passive filling pressure at MCed (end diastolic)


weak human senses with mechanical or optical devices [24].Paraphrasing today, we see that translational scientific prog-ress depends, in large measure, on state-of-the-art visualiza-tion methods. The FI method yields values of velocity andpressure at literally thousands of discrete points in space andtime within each filling chamber simulated. The acquired vastamounts of qualitative and quantitative spatiotemporal infor-mation can be envisaged effectively only when transformedinto graphic representations (cf. the example in Fig. 4, bottompanels). Accordingly, high-resolution plots of instantaneousintraventricular velocity and pressure distributions wereextracted from the FI simulation datasets, using interactivegraphics modules [1, 4, 9, 10, 19–21, 25]. They reveal time-dependent, subtle interactions (see below) between the localacceleration and the convective acceleration components ofthe total pressure gradient [1, 4, 19]. Not only does the small-ness of the directly measurable total early diastolic intraven-tricular pressure gradients render most direct measurements—even by solid-state multisensor catheter—unreliable but,until the comprehensive studies surveyed here [4, 9, 10,19–21] this smallness also concealed the clinically noteworthyfluid dynamics mechanisms underlying these gradients [1, 4].

Local and Convective Acceleration Componentsof Instantaneous Flow Velocity

Just as during ejection [3], in the intraventricular diastolic flowfield, the velocity is a function of both time and space [4].Generally, there can be acceleration of fluid passing through apoint even when the velocity at the given point is constant, i.e.,even if the local acceleration is zero [1, 3, 4, 26]. Streamlinesconnect velocity vectors in a flow field and show the directiona fluid element is traveling in at a given instant; a tangent to astreamline at any point is parallel to the fluid’s instantaneousvelocity at that point. When streamlines are plotted so that thedistances between them correspond to equal volumetric flux(integration of the flux over the distance between adjacentstreamlines gives the volumetric flow rate), the resulting plotgives information about regions of high and low velocities [27,28]. Simply put, closely spaced streamlines indicate relativelyhigh linear velocities and vice versa. Converging streamlines ina flow region imply convective (flow associated) acceleration;diverging streamlines imply convective deceleration along thestream (see Figs. 1 and 2) [1, 3, 10, 19, 27, 29].

Atrioventricular Transvalvular and Intraventricular DiastolicPressure Gradients

Tricuspid transvalvular phasic pressure differences were mea-sured by multisensor right-heart catheters under experimentalhyperdynamic conditions, which accentuate them, and werefound [1, 4, 19] to possess dynamic characteristics, includingtheir timing relative to the inflow velocity, similar to those

derived and validated for the mitral transvalvular pressure dropusing clinical human data byMirsky and Pasipoularides [2] andIsaaz [30]. Instantaneous RV intraventricular diastolic pressuregradients are smaller [1, 4, 19] than their LV counterparts andare not generally amenable to reliable direct measurement bycatheter, as noted above.

The FI method has shown [1, 9, 10, 19–21, 25] that up tothe E wave peak, instantaneous inflow streamlines extendfrom the tricuspid orifice to the RV endocardial surface in anexpanding fan-like pattern (Fig. 1, right). Consequently, theRV inflow is undergoing a convective deceleration, similar tothe early diastolic LV inflow deceleration found in humans by

Fig. 2 During the E wave upstroke, flow is confluent between atrialendocardium and atrioventricular valve orifice and diffluent betweenthe latter and the ventricular chamber walls. There is convectiveacceleration up to the orifice and convective deceleration beyond it.Transvalvular pressure drops embody convective acceleration, whereasintraventricular gradients convective deceleration, which counter-balances in part the simultaneous local acceleration gradient dur-ing the E wave upstroke. These convective pressure gradients aremuch greater (depend on r2, where r is an effective radial coor-dinate) in 3-D space than what might be anticipated on the basisof 2-D projections (dependence on r)


Houlind et al. [31] in meticulous investigations using three-directional velocity-encoded MRI. During the E wave up-stroke, the total instantaneous intraventricular pressure gradi-ent along the inflow stream is the algebraic sum of astreamwise (i.e., in the flow direction) pressure fall contribut-ed by the local acceleration, and a Bernoulli pressure-risecontributed by the convective deceleration. This streamwiseBernoulli pressure-rise counterbalances partially the pressurefall engendered by the simultaneously applying local acceler-ation of the inflow [1, 4, 19].

Convective Deceleration Load

The preceding mutually offsetting action underlies the pe-culiar smallness of the total early diastolic intraventricularpressure gradients in animals and humans [1, 2, 4, 19, 32].During ejection, on the other hand, throughout the upstrokeof the ejection waveform, both the convective and the localintraventricular acceleration effects act in the same sense[1, 3, 4, 8, 24, 29], actually reinforcing each other. Thesecontrasting dynamics of its components under normal con-ditions underscore the much less prominent total (measuredby multimicromanometric catheter [1–4]) intraventriculardiastolic pressure gradient during the upstroke of the Ewave, even under hyperdynamic conditions, such as duringexercise, when both convective and local acceleration com-ponents are augmented and do accentuate strongly the totalejection gradient [1, 3, 26].

At peak volumetric inflow, the local acceleration vanishesand the total intraventricular gradient is represented by theconvective component which is adverse; i.e., the flow encoun-ters higher pressures as it moves downstream. Flow persistsunder its previously built-up momentum [19], just as forwardvehicle motion persists transiently after applying the brakes.These considerations led to the formulation of a new mecha-nism, the “convective deceleration load” (CDL) [1, 4, 9, 10, 19,20, 25], as an important determinant of diastolic inflow and itshemodynamic impairment in presence of chamber enlarge-ment, as is detailed below.

Diastolic Ventriculoannular Disproportion and ClinicalIndicators of Elevated CDL

Clinically, the magnitude of CDL affects strongly the attain-able peak E wave velocities while, simultaneously, CDL itselfis being affected strongly by them. Furthermore, the pressure-rise representing CDL is proportional to the square of theapplying inflow velocities. This is simply a consequence ofthe familiar Bernoulli principle, which shows that the convec-tive acceleration- or deceleration-induced changes in pressureare proportional to the square of the applying reference veloc-ity [1, 3]. Consequently, diastolic filling difficulties attribut-able to an augmented CDL will be exacerbated under any

(patho) physiologic states that are characterized by increasedventricular inflow velocities; e.g., during exercise or stress.

Moreover, the bigger the ventricle, the larger becomes theratio of the inner surface area of the chamber to the area ofthe inflow annulus, and the higher the convective decelera-tion and the CDL (Fig. 1, right). It is necessary to bear inmind that the discrepancy in the two areas involved isgreatly underestimated by envisaging 2-D axial sectionsthrough the actually three-dimensional flow field. To ascer-tain this generally unappreciated fact, please refer to Fig. 2.Accordingly, an enlarged (e.g., dilatation in failure) ventri-cle at end-systole/early diastole provokes a disproportionateincrease of CDL and more difficult diastolic inflow [1, 4, 9,10, 19, 20, 25]. More specifically, it is abnormally raisedend-systolic or, equivalently, early-diastolic volumes thatmatter; this is considered further later on, in the sectiontitled “Clinical Implications for Diastolic Filling in DilatedVentricles.” Difficulties accruing from pathologic ESV ele-vations can be exacerbated by tachycardia, which common-ly accompanies systolic dysfunction and failure and causesdiastolic filling time to wane. Of course, in normal subjectstachycardia elicited by adrenergic mechanisms decreasesESV. Cardiac resynchronization therapy [33–35] can reduceESV in patients with primary systolic dysfunction and fail-ure; thus, it should reduce the CDL and mitigate the sec-ondary impairment in diastolic filling and function.

Clinically, augmentation of CDL may contribute to de-pressed E wave amplitudes and E/A ratio abnormalities, andto elevated atrial pressures and upstream congestion in acuteor chronic heart failure with chamber dilatation. Ensuing atrialoverload (active compensation for difficult ventricular fillingby strong atrial myocardial contraction) may account for theassociation of atrial fibrillation (AF) with clinical ventricularenlargement in absence of coronary disease, especially inelderly subjects [19]. This is especially relevant today, withthe rising proportion of the elderly in the population. The lossof the “atrial kick” in AF can have severe repercussions underconditions of increased CDLwith diastolic filling impairment.

The greater the discrepancy between the sizes of theRV/LVendocardial surface and the AV valve anulus, the largerbecomes the CDL (Fig. 2). This consideration underlay theformulation [1, 9, 10, 19, 20, 25] of the clinically importantconcept of a “diastolic ventriculoannular (inflow valve) dis-proportion” (see DVAD in Fig. 1, right), which is the coun-terpart, or diastolic analog, of the “systolic ventriculoannular(outflow valve) disproportion,” formulated in a systolic clini-cal fluid dynamics survey in JACC [3].

Diastolic Large-Scale Intraventricular Vortical Motions

Early on during the downstroke of the E wave, theintraventricular total pressure gradient becomes strongly


adverse: it now embodies pressure augmentations in thedownstream flow direction accruing from both local andconvective decelerations [1, 4, 10, 19]. This inducesflow instability and large-scale vortical motions(Fig. 1, right). Details on the formation and evolutionof the intraventricular diastolic filling vortices arepresented elsewhere [1]. Figure 3 epitomizes graphicallythe clinically important concepts. Relevant here are theclassic discoveries of Taylor and Wade, in dogs andsheep; using cineradiography with fine stream dye in-jection, and endoscopic cinephotography, these pioneersdemonstrated vortex formation in both the right and leftventricles during diastolic filling [36]. They agree alsowith contemporary systematic clinical measurements ofRodevand et al. [37] in the left ventricle: using Dopplerechocardiography techniques, they showed that duringearly transmitral flow acceleration, all intraventricularvelocities are directed toward the chamber walls; how-ever, after the peak of the E wave retrograde velocities

ensue around the central inflow, reflecting large-scalevortical flow [1, 37].

Large-scale (of a size comparable to that of the RV or theLV chamber) vortices dissipate little flow energy by viscous(frictional) effects. However, interactions of these vorticeswith each other generate intermediate- and small-scale eddies,which are strongly dissipative. Thus, there is a cascade ofenergy from large, through intermediate, to small eddies,where rotational kinetic energy is dissipated as heat [1, 6, 10].

Spanish investigators, in collaboration with scientists atthe University of California San Diego, have recently beenstudying how heart rate and AV synchronicity influencevortex kinematics, using 2-D, B-mode and conventionalcolor Doppler sequences of the full LV, obtained from anapical long-axis view [38, 39]. Initial results, in patientswith an implantable cardiac resynchronization device, indi-cate that the duration of diastole, as modulated by heart rateand AV delay, modifies significantly intraventricular vortexkinematics.

Fig. 3 As blood fans away from the central stream toward the endo-cardial walls, the convective deceleration effect tends, by the Bernoullimechanism, to elevate downstream pressure opposing ventricular in-flow (top right). There is an additional convective pressure decreasenormal to the curved fanning streamlines (top left), which tends todecrease pressure in the direction of the chamber’s base and away fromthe endocardial walls (long arrows crossing streamlines, lower left).During the upstroke of the E wave, these combined adverse convectivepressure effects are opposed by the local acceleration gradient, whichfavors forward flow. During the E wave downstroke, they are joined in

sense and augmented by the local deceleration gradient and can nowreverse the flow. This leads to disruption of the boundary betweenoncoming blood and endocardial walls, or “flow separation,” and to theformation of a toroidal vortex that surrounds the central core (lowerright). Center inset volumetric filling rate calculated by numericaldifferentiation of RV chamber volume obtained from dynamic repre-sentations using combined real-time 3-D echocardiographic andsonomicrometric measurements; small arrows point to representativebeginning and ending points of the FI simulations (adapted, somewhatmodified, from Pasipoularides [1], by permission of PMPH-USA)


Visualization of the Diastolic Filling Vortex

The diastolic vortex is considerably more intense in thenormal-sized than in the dilated chamber [1, 10, 19, 20], asis nicely demonstrated in the color mappings offered in Fig. 4(lower panels), which were obtained by the FI method [9, 10].Comparable color mappings are familiar to echocardiog-raphers and are appropriate for revealing the global organiza-tion of intracardiac flows. The regions with red and orangecolors represent blood flow toward the base of the ventricle,whereas regions with blue and green colors represent bloodflow toward the apex. The intense deceleration of the flow inthe proximate vicinity of the inflow anulus is striking; it will bediscussed further in conjunction with the plots of velocity vs.axial distance from the inflow anulus in Fig. 5. At this junc-ture, it is interesting to note that, contrariwise during ejection,in the close vicinity of the outflow valve anulus, there isinvariably intense convective acceleration of the flow [1, 3,26, 27, 29, 40]. These concentrated convective velocitychanges in the vicinity of circular 2-D inflow/outflow anuliare related to the 3-D ventricular flow field geometry (seeFig. 2)—as is also the “inverse-square law” of physics for,e.g., sound intensity vs. distance from its source.

Vortices represent rotatory motions of a multitude of fluidparticles around a common center and rotating flows havefascinated people for centuries. Leonardo da Vinci studied,drew, and described accurately vortices within the sinuses ofValsalva in his Quaderni d’ Anatomia, which was publishedin 1513 [1, 3], a century before Harvey’s discovery of thecirculation of blood.

Bellhouse’s Model Studies of the Effect of Chamber Sizeon Vortex Strength and AV Valve Closure

Diastolic LV vortices are well visualized using cardiac MRIand color Doppler-based echocardiographic vector flow map-ping [1, 23, 41–44]. Their fluid dynamics were investigated inpioneering studies on mechanical analogs of the left heart byBellhouse [45]. He observed that the mitral valve leafletsopenedwide early in diastole, thenmoved increasingly towardclosure, and were almost fully apposed by end diastole,whether atrial systole was present or not. Atrial systole causedpartial AV valve reopening before resumption of closure.

Of special interest in the present context is that he com-pared AV valve closure rates for two different end-systolicvolumes, while maintaining all other parameters identical.When the operating end-systolic LV volume was small, astrong “ring vortex” (akin to a confined smoke ring) formedwithin the chamber. The thrust of this vortex behind theminitiated mid-diastolic leaflet movements toward closure, andthe valve was nearly closed by end-diastole. In contrast, whenthe end-systolic volume was abnormally large, as in ventric-ular failure, no vigorous ring vortex formed, and the valve

moved sluggishly toward closure, so that it was only 25 %closed at end-diastole, and reversed flowwas required in orderto seal it [34]. This vortical recirculation pattern and its di-minished intensity with chamber enlargement are unambigu-ously illustrated in Fig. 4 by the intraventricular flow colormappings that were discussed above.

Bellhouse concluded that the vortex plays an important rolein the normal mechanism of closure of the mitral valve,ensuring minimal regurgitation by pushing the mitralanteromedial and posterolateral leaflets toward each otherprior to LV myocardial contraction [34]. However, no otherplausible useful hemodynamic function had been ascertainedfor the diastolic vortex until the integrative diastolic functionstudies surveyed here and in the companion JCTR paper [15].

The Facilitatory Role of the Diastolic Vortex in VentricularFilling

Out of these comprehensive diastolic function studies, whichintegrated intracardiac flow phenomena with myocardial re-laxation and compliance dynamics [4, 9–12, 15, 19], thehypothesis of an assisting role in diastolic filling was ad-vanced for the intraventricular diastolic vortex. The key tothis useful physiological role [10] of the RVand LV diastolicvortices lies in their shunting and “trapping” a certain amountof flow energy into the rotational kinetic energy of theirgyratory motion, thereby preventing its Bernoulli conversioninto increased pressure with the strong flow deceleration alongfanning-out streamlines (see Fig. 1, right); this becomes man-ifest as a decrease in the pressure energy of the inflowingblood.

This essential energy-shunting role of the filling vortex wasdemonstrated first in translational animal research [10], andwas subsequently confirmed [22] by meticulous clinical im-aging examinations using 4-D phase contrast flow MRI, atechnique for assessing over time the flow velocity compo-nents in the three spatial dimensions within a flow region. Bypre-empting an inflow-impeding Bernoulli pressure-rise alongthe fanning-out flow streamlines between AV valve anulusand the expanding endocardial surface of the chamber (seeFig. 2 and Fig. 1, right), the diastolic vortex facilitates higherdiastolic filling and stroke volumemaintenance. Recapitulating,the diastolic vortex shunts the inflow kinetic energy, whichwould otherwise contribute to the adverse convectiveBernoulli pressure rise and the CDL, into the kinetic energyof its gyratory motion. This rotational kinetic energy is ulti-mately dissipated as heat [1, 6, 10].

The Question of a Role for the Diastolic Vortex in LVEjection

Some medical investigators [46, 47] have suggested that thepresence of chirally asymmetric LV diastolic vortical


Fig. 4 Top panels Under control conditions with normal wall mo-tion, diastolic filling entails anterior-directed motion of both thefree RV wall and the septum (black arrows, a). In RV chamberdilatation with paradoxical septal motion, the septum moves to-ward the left ventricle (back-pointing black arrow, b). Interposed isthe pattern of RV dilatation with normal septal motion (grayarrows). By increasing both intraventricular mass and effectiverotation radii, increased chamber size yields smaller recirculatingvelocities and vortex strength, as the “whirling dervish” tops sug-gest: with arms extended and wider girth, spinning is slower in bthan in a. The stronger vortex ring in the normal-sized chamber (a)encroaches more on the central core than the weaker vortex in theenlarged chamber (b). Consequently, although instantaneous volu-metric inflow rates at control were smaller than with volume

overload, after vortex development higher linear core velocitieswere present at control than with chamber dilatation attendant tovolume overload. The width of each white arrow in the top panelsis proportional to central core area; the length, to linear inflowvelocity. Bottom panels Simultaneous frontal-plane and transverse(at a level 2.5 cm below the inflow orifice) sections showing RVcolor flow maps, close to the end of the E wave. The instantaneousvolumetric inflow velocities were 39 cm3/s in the normal-sized(left) and 71 cm3/s in the dilated (right) chamber, respectively.These functional imaging mappings show expediently the higheraxial core velocities in the normal and the larger core cross-sectionin the enlarged chamber (color maps reproduced fromPasipoularides et al. [10] by permission of The American Physio-logical Society)


motions helps in the ensuing process of ejection, by conver-sion of vortex kinetic energy into kinetic energy of LVoutflowafter aortic valve opening. According to this intuitively ap-pealing mentalistic notion, the filling vortex persistingthrough end-diastole would serve as a storage means for thekinetic energy of the incoming blood filling the ventricle,reducing the work necessary for the acceleration of blood intothe LVoutflow tract. The rotating stream of blood in the largerlobe of the persisting chirally asymmetric LV vortex wouldthen leave the chamber in a tangential direction when ventric-ular contraction started. However, in dealing with fluid dy-namics phenomena, intuition is treacherous.

This is an admittedly controversial topic with influentialproponents on either side of the issue. It is well known in fluidmechanics that large vortices and rotational fluid flow struc-tures never unwind smoothly [1, 48]. Instead, as was pointedout above, they break up into smaller eddies, and this processis continued, until the vortices are reduced to micron-size

eddies and at that level (Kolmogorov “microscale”) they dis-sipate under the action of viscous forces [49]. The process isknown as the vortical cascade mechanism [1]. This impliesthat vortices are essentially “traps” or “sinks” of energy[1, 10]. Whatever kinetic energy is trapped in the recirculatingmotion of a vortex is bound to be converted into heat; it isbound to be lost from the motion and therefore cannot, in fact,aid ejection [1] despite the, on the face of it, intellectual appealof a “sling-like,” “morphodynamic” mode of action [46].Vortical energy dissipation can occur both within the ventri-cles and within their outflow trunks and beyond.

To analyze the actual energetics of such a sling-like pro-cess, Japanese investigators [50] developed a multiscale,metaphysics heart simulation model, based on the finite ele-ment method. Taking advantage of this highly sophisticatedsimulation model, which encompasses electromechanicalmyocardial events and fluid dynamics processes, they com-pared the hemodynamics of ventricles with physiological and

Fig. 5 Velocity distributionsalong the RV inflow axis fromfunctional imaging simulations;a at control and b with chamberenlargement. At control,although the volumetric inflowrate is then much lower, inflowvelocities along the axis in thebasal region of the chamber aresubstantially higher late in theE wave than their levels at thepeak volumetric inflow rate,Epk. This unexpectedphenomenon is a corollary ofthe intense squeeze of theinflow core at control by theencircling strong ring vortex, ashinted by the graphic insetsaccompanying the velocitydistribution plots. Arrow lengthis proportional to linearvelocity; arrow width tovolumetric inflow rate. cSchematic representation of thering vortex surrounding theinflowing core stream androtating poloidally, i.e., in thesense of the curved arrow. Thispoloidal rotation pattern is akinto that of the magnetic field in amagnet, which flows from thenorth pole to the south pole;hence, the name (reproduced,slightly modified, fromPasipoularideset al. [19] by permission of TheAmerican Physiological Society)


nonphysiological diastolic vortical flow paths. They examinedthe effects of the inflow direction in the left ventricle on itsperformance under conditions where other experimental vari-ables, including ventricular morphology, myocardial proper-ties, and pre- and afterloads, were completely matched. Theyfound that the physiological vortical flow path did not haveany energy-saving effect.

The stroke work performed by the LVwall was comparableat both slower and faster heart rates (physiological vorticalflow path vs. nonphysiological, 0.864 vs. 0.874 J, at a heartrate of ≈60 beats/min; and 0.599 vs. 0.590 J, at a heart rate of≈100 beats/min), indicating that vortex-associated chiralasymmetry of the flow paths in the mammalian heart hasminimal, if any at all, energetic functional merit [50]. Thisthorough and meticulously documented investigation contra-dicts the notion [46, 47] of an energy-saving fluidic effect ofcardiac looping and chirally asymmetric LV diastolic vorticalmotions on ejection. Moreover, as the Japanese investigatorshave emphasized [50], meticulous clinical assessments [51]and outcomes of animal experiments [52, 53] are in harmonywith their results [1].

The Atrial Filling Vortex

Every movement in a finite, closed space must self-evidentlylead to gyration, unless a fan-like inflow pattern transientlyleads to a volumetrically matched chamber expansion(cf. Fig. 1). This ensues from the law of mass conservationand the fluid’s incompressibility. Thus, if you stir coffee in acup, rotation in some form is always generated; the usualstirring movement to mix cream is circular, but blowing intothe cup toward the center produces more or less symmetricvortices, which at the surface appear akin to the pattern shownin the lower right corner of Fig. 1. It does not come as asurprise, therefore, that filling vortices have been describedboth in the right atrium, associated with vena caval inflow ofblood [54], and in the left atrium, associated with pulmonaryvenous inflow [1, 55, 56].

Clinical Implications for Diastolic Filling in DilatedVentricles

The FI method has shown the strength of the toroidal (dough-nut shaped) ring vortex surrounding the main inflow streamand rotating poloidally (in the direction of the arrow in inset c,Fig. 5) to diminish with chamber dilatation [1, 9, 10, 19–21,25], confirming Bellhouse’s ingenious mechanical models[34] demonstrating reduced diastolic vortex strength withincreased chamber volume. Qualitatively, the strength of thevortex is proportional to the intensity (rotation rate) of thepoloidal rotatory motion. According to its facilitatory role forventricular inflow that was discussed above, the diastolic

vortex can assist filling by eliminating CDL to a variabledegree, which depends on the vortex strength [1, 10, 19].

In the normal-sized chamber, the strong toroidal vortexsurrounding the central inflow core encroaches on the areaavailable for flow toward the apex. This encroachment in-creases the central jet velocities and causes higher linearvelocities (in meters per second) to actually occur later—during the downstroke of the E wave—than at peak volu-metric (in cubic meters per second) inflow rate [19], as isshown in Fig. 5. In the enlarged ventricles, this effect isblunted so that linear and volumetric peak velocities tend tocoincide [19] (see Fig. 5).

These translational animal research findings are consistentwith meticulous clinical results published by Yamamoto et al.[57] who compared normal intraventricular velocity patternsto those prevailing with LVenlargement, in dilated cardiomy-opathy and in hypertensive heart disease. The regional dia-stolic velocity patterns at 1, 2, and 3 cm from the mitral tiptoward the apex were recorded simultaneously with the mitralinflow velocity pattern, using multigated pulsed Doppler. Inthe abnormally dilated left ventricles, linear velocities de-creased strikingly from the mitral tip toward the apex [35].In the new fluid dynamics framework for diastolic filling [1,10, 19, 20], this abnormal velocity distribution accrues fromthe diffluent (fanning out) streamline pattern and a strongconvective deceleration that in the dilated chamber persists,unabated by a strong toroidal vortex (Fig. 1, right).

In view of the integrative diastolic function framework ofFig. 1, Yamamoto et al.’s meticulous clinical findings [35] arenow clearly understood to be a manifestation of the strongconvective deceleration occurring as the inflow fans out—seeFigs. 1 and 2—in the absence of a strong, encroaching dia-stolic vortex [1, 9, 10, 19, 20]. On the contrary, in subjectswith normal LV function and size, the simultaneouslyrecorded velocities at these successive sites were remarkablyuniform [35]. This reflects the presence of the filling ring vortexthat surrounds the central inflowing stream [10, 19, 20]. Thestrong vortex gets in the way of a fanning-out streamline patternand prevents the ensuing strong convective deceleration fromtaking place—see Figs. 1, 2, and 4.

Using sophisticated digital processing of color DopplerM-mode recordings, Yotti et al., at the University HospitalGregorio Marañón, in Madrid, Spain [58], obtained mea-surements of the spatiotemporal distributions of LV diastolicintraventricular pressure gradients. Their elegant findings inpatients with dilated LV cardiomyopathy and their conclu-sions are in perfect agreement [4] with our previously pub-lished [9, 10, 19] fluid dynamic analyses and propositionsregarding the convective deceleration load, the diastolicventriculoannular disproportion, and the impairment of Ewave peak velocities and E/A wave ratios in dilated ventri-cles. The Editorial accompanying Yotti et al.’s elegant studyin Circulation underscored this agreement [59].


Medical/Surgical Corollaries for Management of RV/LVChamber Dilatation

Conditions leading to eccentric hypertrophy and remodelingwith chamber enlargement as seen, e.g., in CHF or aftermyocardial infarction, should increase the Bernoulli-induced diastolic convective deceleration load and simulta-neously reduce the intensity of the filling vortex. Both ofthese occurrences have been shown to contribute to thefilling impairment attendant to a chamber enlargement [1,4, 9, 10, 19–21, 25]. The converse applies for surgicalinterventions, such as the Batista procedure (partial leftventriculectomy), and medical treatments attenuating or re-versing pathological ventricular remodeling and chamberenlargement [1, 7]. Medical interventions include cardiacresynchronization therapy (atrial-synchronized biventricularpacing) to reverse the remodeling seen in CHF and improvecardiac structure and function [34, 60, 61], and the attenu-ation of ventricular remodeling and LV operating volumeelevations after myocardial infarction by the administrationof cell therapy or ACE inhibitors [62]. Such interventionsshould act, in part, by decreasing ESV (see discussion insubsection “Diastolic Ventriculoannular Disproportion andClinical Indicators of Elevated CDL”) and, thus, the diastol-ic CDL, and by diminishing the diastolic ventriculoannulardisproportion [19, 20] as well as promoting stronger diastol-ic vortices within the less-dilated chambers [10].

Conversely, diastolic ventriculoannular disproportion andthe convective deceleration load should be augmented by acardiac procedure-associated pericardiotomy when chamberdilatation is no longer curtailed by pericardial restraint, as incases of acute or subacute systolic LV dysfunction, beforegradual compensatory pericardial expansion that ensues withchronic cardiac enlargement [15]. More importantly, DVADwith augmentation of CDL occurs initially after surgical in-terventions, i.e., tricuspid or mitral ring annuloplasties, whichdecrease the effective tricuspid or mitral orifice cross-sectionrelative to RV or LV chamber volume and inner surface area[63]. However, the subsequent regression of abnormal remod-eling—reversal of chamber dilatation and eventual chambersize normalization—remedies this transitory situation. Suchconsiderations point toward innovative translational researchdirections.

Fluid Dynamic Interactions of Systolic and DiastolicDysfunction

In primary systolic dysfunction and failure, impaired contrac-tile performance is the principal functional derangement; it isthe major mechanism for the right and downward shift of theend-systolic pressure–volume line (see Fig. 6, lower right) andthe diminished ejection fraction. Increased passive stiffnessand impaired ventricular relaxation are the traditionally

recognized [2, 15] functional derangements in primary dia-stolic dysfunction and failure. The diastolic pressure–volumerelation shifts upward and to the left (Fig. 6, lower left);consequently, there is a disproportionately large diastolic pres-sure increase as volume rises during diastolic filling, leadingto raised atrial and upstream venous pressures.

In primary systolic dysfunction and failure, ensuing dia-stolic dysfunction as gauged by a ventricular filling impair-ment is common, particularly with a substantially raised end-systolic ventricular volume (see Fig. 6, top). In the innovativeframework surveyed here [1, 4, 9, 10, 19, 20, 25], the in-creased end-systolic/early-diastolic chamber size inducesventriculoannular disproportion with a disproportionate in-crease of the CDL and a weakening of the filling vortex thatnormally facilitates filling by impounding inflow kinetic en-ergy and averting an inflow impeding, convective Bernoullipressure rise [10]; elevated intraventricular diastolic fillingpressure levels ensue (Fig. 6, lower left). More difficult ven-tricular diastolic inflow and filling impairment can accruefrom these changes. This sequence of events explains theintriguing observation (alluded to in the first page of thisSurvey) that diastolic function is frequently abnormal in pa-tients with primary systolic dysfunction and failure.

The new framework complements conventional assess-ments of diastolic function in failure and remodeling, byencompassing heretofore overlooked vital fluid dynamics as-pects. Thus (Fig. 6, top), through the increased end-systolicventricular volume and CDL, with diminished vortex strengthand superimposed filling impairment, an underlying systolicdysfunction and failure can beget diastolic dysfunction.Conversely, diastolic dysfunction with impaired chamber fill-ing in absence of effective “atrial kick” compensation (e.g., inatrial fibrillation) can lead directly to a preload-dependentstroke volume curtailment: the stroke volume decline accruesfrom decreased EDV, while systolic myocardial function andcontractility may remain normal.

Epigenetic Propositions and Directions for FunctionalGenomics Research

Epigenetics affects the interpretation of DNA genetic se-quences in cells; they help drive cell processes by regulatingwhen and how genes are activated. The epigenome [64] is animportant target of “environmental” forces. One way in whichDNA expression can be controlled in response to such forcesis by the methylation (addition of a bulky methyl group) toone of the DNA bases [64–66], rendering this “tagged” area ofthe DNA less active and perhaps suppressing the productionof a particular protein. Methylation modifies only the C(cytosine) of the four-letter DNA “alphabet” (A, G, C, andT). The abnormal methylation of genes can ultimately lead toalterations of normal cardiomyocyte functions and to the


development of heart disease. Another way that gene expres-sion can be modified involves the histones, the chief proteincomponents of chromatin, which act as spools around whichDNA winds. Histone proteins are subject to a variety ofmodifications, the combination of which creates a code (his-tone code) that influences the gene transcriptional status andthe expression of genetic information, by modulating thechromatin structure and DNA accessibility [64, 67–70].

Epigenetic mechanisms acting as switches in gene expres-sion in response to various stimuli are fundamental in cardiacadaptations, remodeling, reverse remodeling, and disease.They underlie “phenotypic plasticity” (or “plastic adaptation”)and can lead the same combination of genes to produce differ-ent structural/functional outcomes, depending on “environ-mental” variation or stimuli [71, 72] to which they areexposed; i.e., the environment can have a direct impact onthe phenotype without any genetic change [73].

In a contemporaneous survey [74], I propose thatchanging “environmental” forces associated with diastol-ic RV and LV rotatory flows during filling exert impor-tant, albeit still unappreciated, epigenetic actionsinfluencing functional and morphological cardiac adapta-tions. Mechanisms analogous to Murray’s law of hydro-dynamic shear-induced endothelial cell modulation ofvascular geometry are likely to link to RV/LV adaptations[1, 74] the variable laminar vortical shear and centrifugal

“squeeze” forces, which are exerted by the poloidallyrotating blood (cf. inset c, Fig. 5) on the endocardiumand myocardial walls. By Newton’s Third Law, the cen-trifugal force is directed away from the center of vorticalrotation and is exerted by the revolving blood on theventricular walls that provide its centripetal acceleration—seeFig. 7.

The time has come to explore a new paradigm in whichintraventricular diastolic filling vortex-associated forcesplay a fundamental epigenetic role, and to work out howheart cells react to these forces and to changes in theirintensity [74], connected to ventricular chamber enlarge-ment and remodeling of manifold etiologies. In this context,the epigenome can be interpreted as a template placed overthe potential blueprint of the genome, with variously posi-tioned openings depending on the phenotype that is to arisein a given “environment.” Exposed are the blueprint genesthat are actually expressed, whereas covered are genes thatare not expressed.

In future research, ventricular tissue samples from dilatedventricles having reduced vortex strength can be comparedagainst samples from control ventricles having diastolicfilling vortices of normal strength. The myocardial samplescan be individually analyzed for DNA methylation/histonemodifications and expressional changes using expressionmicroarrays and gene promoter microarrays to identify

Fig. 6 Bottom panels in diastolic dysfunction/failure, the diastolicpressure–volume relation (dashed curve) shifts upward-and-leftward(counter-clockwise rotation), indicative of disproportionately largepressure increments as volume rises during filling; if there ensues anEDV decrease, e.g., with loss of atrial kick in atrial fibrillation, then SVfalls too. In systolic dysfunction/failure, a clockwise rotation of theend-systolic pressure–volume line imputes decreased myocardial con-tractility for the reduced ejection fraction and increased ESV. Toppanels in systolic dysfunction, the increased ESV augments the CDLand depresses the diastolic filling vortex strength, leading to filling

impairment. Through the increased ESV and superimposed fillingimpairment, a primary systolic dysfunction/failure can beget supple-mentary diastolic filling impairment/dysfunction, increasing the needfor compensatory mechanisms for maintenance of SV and/or CO.Conversely, diastolic dysfunction with impaired chamber filling canlead directly to a preload-dependent stroke volume curtailment. EDVend-diastolic volume, ESV end-systolic volume, SV stroke volume, COcardiac output; CDL convective deceleration load, VXS diastolic vortexstrength


differentially methylated gene promoters between the twogroups [75–78].

Elucidating the role of epigenetic mechanisms in ventric-ular dilatation and failure could provide a basis for thedevelopment of new therapeutic approaches to ventriculardilatation and failure associated with diminished vortexstrength and ventricular filling [74]; appropriate modulationof epigenetic structure could correct gene expression. Thisnew frontier in contemporary cardiac research should un-cover versatile mechanistic insights linking filling vortexpatterns and attendant forces to variable expressions of generegulation in RV/LV myocardium. In due course, it shouldreveal intrinsic homeostatic arrangements that support ven-tricular myocardial function and adaptability, and their dis-turbances. Such advances should lead to improved (patho)physiologic understanding and to innovative and more ef-fective treatments of heart failure-related structural andfunctional remodeling and of ventricular dilatation of di-verse origins. Indeed, appropriate modulation of epigeneticstructure could correct gene expression of myocardialcells impacted by abnormal diastolic vortex dynamics. AsDr. Victor Dzau, currently Chancellor of Health Affairs atDuke has summed up in an Editorial [79], Dr. DanielTosteson of Harvard arrived more than 15 years ago at theastute conclusion that: “In the past we have had functions insearch of sequences. In the future, pathology and physiologywill become functionators for the sequences.” It rings sotantalizing true today for those engaged in cardiac transla-tional research!

Conclusions

This article has summarized concepts regarding normal dia-stolic function and its disturbances in ventricular dilatationfrom multiple etiologies, including primary systolic dysfunc-tion and heart failure, with central emphasis on developing auseful innovative framework for approaching ventricular fill-ing dynamics in RV/LV failure. In the setting of ventricularchamber enlargement, fundamental cardiac fluid dynamicmechanisms can impair diastolic filling even in the absenceof inflow valvular stenosis or of abnormalities of myocardialrelaxation and ventricular diastolic compliance.

The diastolic vortex facilitates filling. The CDL is an im-portant determinant of RVor LV inflow during the E wave (Awave) upstroke. The larger the chamber, the larger the CDL; itsmagnitude affects strongly peak E (and A wave) velocities.This underlies the concept of diastolic ventriculoannular dis-proportion. Large-scale vortical motions after the E wave peakimpound inflow kinetic energy, preventing an inflow-impeding Bernoulli pressure rise between inflow orifice andthe expanding endocardial surface. This reduces the CDL by avariable amount, depending on vortical intensity, which de-creases with chamber dilatation.

The new fluid dynamic framework complements myocar-dial and ventricular mechanics [2, 15] assessments of diastolicdysfunction in failure/remodeling, by encompassing hereto-fore overlooked but vital dynamic aspects of diastolic intra-ventricular flow. It also provides stimulus for functionalgenomics research, which promises to rapidly narrow the

Fig. 7 Epigenetic dynamic actions of the RV/LV diastolic toroidalvortex: cellular components of the myocardial walls react to theirphysical 3-D “environment,” including variable diastolic vortical shearand “squeeze” forces, by transducing mechanical deformations andforces into differentiated transcription and translation signals, whichcan adjust myocardial cellular and extracellular tissue and ventricular

structure. Mechanosensitive controls modulate cardiomyocyte shapeand intracellular architecture, and processes as diverse as proliferation,hypertrophy, and apoptosis, involved in cardiac homeostasis and ad-aptations (reproduced, slightly modified, from Pasipoularides [74] bypermission of The Hellenic Cardiological Society)


gap between gene sequence and myocardial function and toyield new insights into ventricular adaptive and maladaptiveremodeling.

Summing up, irrespective of any other possible coexistingmyocardial and ventricular abnormalities, chamber enlarge-ment—a hallmark of ventricular systolic dysfunction andfailure—in and by itself impairs diastolic filling, by enhancingthe convective deceleration load and by diminishing thestrength of the inflow-facilitating, beneficial diastolic intra-ventricular vortex. Contemporary cardiac research should un-cover versatile mechanistic insights linking filling vortexpatterns and attendant forces to variable expressions of myo-cardial gene regulation.

Acknowledgments Research support, for work in the author’sLaboratory surveyed in this Review, was provided by the NationalHeart, Lung, and Blood Institute [grant number R01-HL-050446]; theNational Science Foundation [grant number CDR 8622201]; and theNorth Carolina Supercomputing Center and Cray Research. I thank theco-Editor-in-Chief and the reviewers for their constructive remarks.The final product has benefited greatly from Dr. Hall’s interest andpersonal editorial attention.

References

1. Pasipoularides, A. (2010). Heart’s vortex: Intracardiac blood flowphenomena (p. 960). Shelton: People’s Medical Publishing House.

2. Mirsky, I., & Pasipoularides, A. (1990). Clinical assessment ofdiastolic function. Prog Cardiovasc Dis, 32, 291–318.

3. Pasipoularides, A. (1990). Clinical assessment of ventricular ejec-tion dynamics with and without outflow obstruction. [Review].Journal of the American College of Cardiology, 15, 859–882.

4. Pasipoularides, A. (2012). Right and left ventricular diastolic flowfield: why are measured intraventricular pressure gradients small?[Flujo diastólico ventricular derecho e izquierdo: ¿por qué sonbajos los gradientes de presión intraventricular medidos?] RevEsp Cardiol. 10.1016/j.recesp.2012.07.019

5. Borlaug, B. A., & Redfield, M. M. (2011). Diastolic and systolicheart failure are distinct phenotypes within the heart failure spec-trum. Circulation, 123, 2006–2013.

6. Pasipoularides, A. (2011). LV twisting-and-untwisting in HCM:Ejection begets filling. Diastolic functional aspects of HCM.[Progress in Cardiology]. American Heart Journal, 162, 798–810.

7. Tønnessen, T., & Knudsen, C. W. (2005). Surgical left ventricularremodelling in heart failure. European Journal of Heart Failure, 7,704–709.

8. Isaaz, K., & Pasipoularides, A. (1991). Noninvasive assessment ofintrinsic ventricular load dynamics in dilated cardiomyopathy.Journal of the American College of Cardiology, 17, 112–121.

9. Pasipoularides, A., Shu, M., Womack, M. S., Shah, A., von Ramm,O., &Glower, D. D. (2003). RV functional imaging: 3-D echo-deriveddynamic geometry and flow field simulations. American Journal ofPhysiology. Heart and Circulatory Physiology, 284, H56–H65.

10. Pasipoularides, A., Shu, M., Shah, A., Womack, M. S., & Glower,D. D. (2003). Diastolic right ventricular filling vortex in normaland volume overload states. American Journal of Physiology.Heart and Circulatory Physiology, 284, H1064–H1072.

11. Pasipoularides, A. D., Shu, M., Shah, A., & Glower, D. D. (2002).Right ventricular diastolic relaxation in conscious dog models of

pressure overload, volume overload and ischemia. The Journal ofThoracic and Cardiovascular Surgery, 124, 964–972.

12. Pasipoularides, A. D., Shu, M., Shah, A., Silvestry, S., & Glower,D. D. (2002). Right ventricular diastolic function in canine modelsof pressure overload, volume overload and ischemia. AmericanJournal of Physiology. Heart and Circulatory Physiology, 283,H2140–H2150.

13. Ihara, T., Shannon, R. P., Komamura, K., Pasipoularides, A.,Patrick, T., Shen, Y. T., et al. (1994). Effects of anesthesia andrecent surgery on diastolic function. Cardiovascular Research, 28,325–336.

14. Condos, W. R. J., Latham, R. D., Hoadley, S. D., & Pasipoularides,A. (1987). Hemodynamics of the Mueller maneuver in man: Rightand left heart micromanometry and Doppler echocardiography.Circulation, 76, 1020–1028.

15. Pasipoularides, A. (2013). Right and left ventricular diastolic pres-sure–volume relations: A comprehensive review. Journal ofCardiovascular Translational Research, 6, 239–252.

16. Craig, W. E., Murgo, J. P., & Pasipoularides, A. (1987).Calculation of the time constant of relaxation. In W. Grossman& B. Lorell (Eds.), Diastolic relaxation of the heart (pp. 125–132).The Hague: Martinus Nijhoff.

17. Zile, M. R., Tomita, M., Ishihara, K., Nakano, K., Lindroth, J., Spinale,F., et al. (1993). Changes in diastolic function LV volume overloadproduced by mitral regurgitation. Circulation, 87, 1378–1388.

18. Pacileo, G., Limongelli, G., Verrengia, M., Gimeno, J., Di Salvo,G., & Calabrò, R. (2004). Backscatter evaluation of myocardialfunctional and textural findings in children with right ventricularpressure and/or volume overload. The American Journal ofCardiology, 93, 594–597.

19. Pasipoularides, A., Shu, M., Shah, A., Tucconi, A., & Glower, D.D. (2003). RV instantaneous intraventricular diastolic pressure andvelocity distributions in normal and volume overload awake dogdisease models. American Journal of Physiology. Heart andCirculatory Physiology, 285, H1956–H1968.

20. Pasipoularides A. (2011). Analysis of vortex flow imaging innormal and dysfunctional RV’s. American Society ofEchocardiography 22nd Annual Scientific Sessions, Montreal,EE02d - Flow Vortex Imaging; PROLibraries.com. http://www.prolibraries.com/ase/?select=session&sessionID=3049.

21. Pasipoularides, A. (2008). Invited commentary: Functional imaging(FI) combines imaging datasets and computational fluid dynamics tosimulate cardiac flows. Journal of Applied Physiology, 105, 1015.

22. Fredriksson, A. G., Zajac, J., Eriksson, J., Dyverfeldt, P., Bolger,A. F., Ebbers, T., et al. (2011). 4-D blood flow in the human rightventricle. American Journal of Physiology. Heart and CirculatoryPhysiology, 301, H2344–H2350.

23. Wang, X. F., Deng, Y. B., Nanda, N. C., Deng, J., Miller, A. R., &Xie, M. X. (2003). Live three-dimensional echocardiography:Imaging principles and clinical application. Echocardiography,20, 593–604.

24. Inwood, S. (2003). The forgotten genius: The biography of RobertHooke 1635–1703 (p. 105). San Francisco: MacAdam.

25. Pasipoularides, A. (2007). Complementarity and competitivenessof the intrinsic and extrinsic components of the total ventricularload: Demonstration after valve replacement in aortic stenosis[Editorial]. American Heart Journal, 153, 4–6.

26. Pasipoularides, A., Murgo, J. P., Miller, J. W., & Craig, W. E.(1987). Nonobstructive left ventricular ejection pressure gradientsin man. Circulation Research, 61, 220–227.

27. Georgiadis, J. G., Wang, M., & Pasipoularides, A. (1992).Computational fluid dynamics of left ventricular ejection. Annalsof Biomedical Engineering, 20, 81–97.

28. Pasipoularides, A. (2011). Fluid dynamic aspects of ejection inhypertrophic cardiomyopathy. [Review]. Hellenic Journal ofCardiology, 52, 416–426.


http://dx.doi.org/10.1016/j.recesp.2012.07.019

http://prolibraries.com/

http://www.prolibraries.com/ase/?select=session&sessionID=3049

http://www.prolibraries.com/ase/?select=session&sessionID=3049

29. Pasipoularides, A., Murgo, J. P., Bird, J. J., & Craig, W. E. (1984).Fluid dynamics of aortic stenosis: Mechanisms for the presence ofsubvalvular pressure gradients. American Journal of Physiology,246, H542–H550.

30. Isaaz, K. (2000). Expanding the frontiers of Doppler echocardiog-raphy for the noninvasive assessment of diastolic hemodynamics.Journal of the American College of Cardiology, 36, 1950–1952.

31. Houlind, K., Schroeder, A. P., Egeblad, H., & Pedersen, E. M.(1999). Age-dependent changes in spatial and temporal bloodvelocity distribution of early left ventricular filling. MagneticResonance Imaging, 17, 859–868.

32. Cortina, C., Bermejo, J., Yotti, R., Desco, M. M., Rodríguez-Pérez,D., Antoranz, J. C., et al. (2007). Noninvasive assessment of theright ventricular filling pressure gradient. Circulation, 116, 1015–1023.

33. Yu, C. M., Sanderson, J. E., & Gorcsan, J., 3rd. (2010).Echocardiography, dyssynchrony, and the response to cardiacresynchronization therapy. European Heart Journal, 31, 2326–2337.

34. Yu, C. M., Fung, W. H., Lin, H., Zhang, Q., Sanderson, J. E., &Lau, C. P. (2003). Predictors of left ventricular reverse remodelingafter cardiac resynchronization therapy for heart failure secondaryto idiopathic dilated or ischemic cardiomyopathy. The AmericanJournal of Cardiology, 91, 684–688.

35. Khaykin, Y., Exner, D., Birnie, D., Sapp, J., Aggarwal, S., &Sambelashvili, A. (2011). Adjusting the timing of left-ventricularpacing using electrocardiogram and device electrograms.Europace, 13, 1464–1470.

36. Taylor, D. E. M., & Wade, J. D. (1973). Pattern of blood flowwithin the heart: A stable system. Cardiovascular Research, 7, 14–21.

37. Rodevand, O., Bjornerheim, R., Edvardsen, T., Smiseth, O. A., &Ihlen, H. (1999). Diastolic flow pattern in the normal left ventricle.Journal of the American Society of Echocardiography, 12, 500–507.

38. Alhama, M., Bermejo, J., Yotti, R., Perez-David, E., Benito, Y.,Gonzalez-Mansilla, A., et al. (2012). Quantitative assessment ofintraventricular vorticity using conventional color-Doppler ultra-sound. Head to head clinical validation against phase-contrastmagnetic resonance imaging. Journal of the American College ofCardiology, 59(13s1), E1128–E1128.

39. Benito, Y., Bermejo, J., Alhama, M., Yotti, R., Perez del Villar, C.,Martínez-Legazpi, P., et al. (2012). Heart rate and AV delay modifyleft ventricular filling vortex properties. Circulation, 126, A18099.

40. Bird, J. J., Murgo, J. P., & Pasipoularides, A. (1982). Fluid dy-namics of aortic stenosis: Subvalvular gradients withoutsubvalvular obstruction. Circulation, 66, 835–840.

41. Bazilevs, Y., del Alamo, J. C., & Humphrey, J. D. (2010). Fromimaging to prediction: Emerging non-invasive methods in pediatriccardiology. Prog Ped Cardiol, 30, 81–89.

42. Ebbers, T., Wigstrom, L., Bolger, A. F., Engvall, J., & Karlsson,M. (2001). Estimation of relative cardiovascular pressures usingtime-resolved three-dimensional phase contrast MRI. MagneticResonance in Medicine, 45, 872–879.

43. Bolger, A. F., Heiberg, E., Karlsson, M., Wigström, L., Engvall, J.,Sigfridsson, A., et al. (2007). Transit of blood flow through thehuman left ventricle mapped by cardiovascular magnetic resonance.Journal of Cardiovascular Magnetic Resonance, 9, 741–747.

44. Lu, J., Li, W., Zhong, Y., Luo, A., Xie, S., & Yin, L. (2012).Intuitive visualization and quantification of intraventricular con-vection in acute ischemic left ventricular failure during earlydiastole using color Doppler-based echocardiographic vector flowmapping. The International Journal of Cardiovascular Imaging,28, 1035–1047.

45. Bellhouse, B. J. (1972). Fluid mechanics of a model mitral valveand left ventricle. Cardiovascular Research, 6, 199–210.

46. Kilner, P. J., Yang, G. Z., Wilkes, A. J., Mohiaddin, R. H., Firmin,D. N., & Yacoub, M. H. (2000). Asymmetric redirection of flowthrough the heart. Nature, 404, 759–761.

47. Taylor, T. W., & Yamaguchi, T. (1995). Flow patterns in three-dimensional left ventricular systolic and diastolic flows determinedfrom computational fluid dynamics. Biorheology, 32, 61–71.

48. She, Z. S., & Waymire, E. C. (1995). Quantized energy cascadeand log-Poisson statistics in fully developed turbulence. PhysicalReview Letters, 74, 262–265.

49. She, Z. S., & Waymire, E. C. (1995). Quantized energy cascadeand log-Poisson statistics in fully developed turbulence. PhysicalReview Letters, 74, 262–265.

50. Watanabe, H., Sugiura, S., & Hisada, T. (2008). The looped heartdoes not save energy by maintaining the momentum of bloodflowing in the ventricle. American Journal of Physiology. Heartand Circulatory Physiology, 294, H2191–H2196.

51. Maire, R., Ikram, S., Odemuyiwa, O., Groves, P. H., Lo, S. V.,Banning, A. P., et al. (1994). Abnormalities of left ventricular flowfollowing mitral valve replacement: A colour flow Doppler study.European Heart Journal, 15, 293–302.

52. Elbeery, J. R., Lucke, J. C., Feneley, M. P., Maier, G. W., Owen, C.H., Savitt, M. A., et al. (1995). The mechanical determinants ofmyocardial oxygen consumption in the conscious dog. AmericanJournal of Physiology. Heart and Circulatory Physiology, 269,H609–H620.

53. Glower, D. D., Spratt, J. A., Snow, N. D., Kabas, J. S., Davis, J.W., Olsen, C. O., et al. (1985). Linearity of the Frank–Starlingrelationship in the intact heart: The concept of preload recruitablestroke work. Circulation, 71, 994–1009.

54. Rushmer, R. F. (1970). Cardiovascular dynamics (3rd ed., p. 50).Philadelphia: Saunders.

55. Fyrenius, A., Wigström, L., Ebbers, T., Karlsson, M., Engvall, J.,& Bolger, A. F. (2001). Three dimensional flow in the human leftatrium. Heart, 86, 448–455.

56. Park, K. H., Son, J. W., Park, W. J., Lee, S. H., Kim, U., Park, J. S.,et al. (2013). Characterization of the left atrial vortex flow by two-dimensional transesophageal contrast echocardiography using particleimage velocimetry. Ultrasound in Medicine and Biology, 39, 62–71.

57. Yamamoto, K., Masuyama, T., Tanouchi, J., Naito, J., Mano, T.,Kondo, H., et al. (1995). Intraventricular dispersion of early dia-stolic filling: A new marker of left ventricular diastolic dysfunc-tion. American Heart Journal, 129, 291–299.

58. Yotti, R., Bermejo, J., Antoranz, J. C., Desco, M. M., Cortina, C.,Rojo-Alvarez, J. L., et al. (2005). A noninvasive method forassessing impaired diastolic suction in patients with dilated cardio-myopathy. Circulation, 112, 2921–2929.

59. Little, W. C. (2005). Diastolic dysfunction beyond distensibility;adverse effects of ventricular dilatation. Circulation, 112, 2888–2890.

60. Abraham, W. T., & Hayes, D. L. (2003). Cardiac resynchronizationtherapy for heart failure. Circulation, 108, 2596–2603.

61. St John Sutton, M., Ghio, S., Plappert, T., Tavazzi, L., Scelsi, L.,Daubert, C., et al. (2009). REsynchronization reVErses Remodelingin Systolic left vEntricular dysfunction (REVERSE) Study Group.Cardiac resynchronization induces major structural and functionalreverse remodeling in patients with New York Heart Associationclass I/II heart failure. Circulation, 120, 1858–1865.

62. Sutton, M. G., & Sharpe, N. (2000). Left ventricular remodelingafter myocardial infarction: Pathophysiology and therapy.Circulation, 101, 2981–2988.

63. Glower, D. D., Tuttle, R. H., Shaw, L. K., Orozco, R. E., &Rankin, J. S. (2005). Patient survival characteristics after routinemitral valve repair for ischemic mitral regurgitation. The Journalof Thoracic and Cardiovascular Surgery, 129, 860–868.

64. Bernstein, B. E., Meissner, A., & Lander, E. S. (2007). Themammalian epigenome. Cell, 128, 669–681.


65. Klose, R. J., & Bird, A. P. (2006). Genomic DNA methylation: Themark and its mediators. Trends in Biochemical Sciences, 31, 89–97.

66. Weber, M., & Schübeler, D. (2007). Genomic patterns of DNAmethylation: Targets and function of an epigenetic mark. CurrentOpinion in Cell Biology, 19, 273–280.

67. Kouzarides, T. (2007). Chromatin modifications and their function.Cell, 128, 693–705.

68. Chang, C. P., & Bruneau, B. G. (2012). Epigenetics andcardiovascular development. Annual Review of Physiology,74, 41–68.

69. Rusk N. (2012). Epigenetics: Writing the histone code. NatureMethods, 9, 777.

70. Lorenzen, J. M., Martino, F., & Thum, T. (2012). Epigenetic modi-fications in cardiovascular disease. Basic Research in Cardiology,107, 245. doi:10.1007/s00395-012-0245-9.

71. Sollars, V., Lu, X., Xiao, L., Wang, X., Garfinkel, M. D., & Ruden,D. M. (2003). Evidence for an epigenetic mechanism by whichHsp90 acts as a capacitor for morphological evolution. NatureGenetics, 33, 70–74.

72. Feinberg, A. P. (2007). Phenotypic plasticity and the epigenetics ofhuman disease. Nature, 447, 433–440.

73. Réale, D., McAdam, A. G., Boutin, S., & Berteaux, D. (2003).Genetic and plastic responses of a northern mammal to climatechange. Proc R Soc Lond B, 270, 591–596.

74. Pasipoularides, A. (2012). Diastolic filling vortex forces and car-diac adaptations: Probing the epigenetic nexus. Hellenic Journal ofCardiology, 53, 458–469.

75. Dahl, C., & Guldberg, P. (2003). DNA methylation analysis tech-niques. Biogerontology, 4, 233–250.

76. Koga, Y., Pelizzola, M., Cheng, E., Krauthammer, M., Sznol, M.,Ariyan, S., et al. (2009). Genome-wide screen of promoter meth-ylation identifies novel markers in melanoma. Genome Research,19, 1462–1470.

77. Lee, E. K. (2007). Large-scale optimization-based classificationmodels in medicine and biology. Annals of Biomedical Engineering,35, 1095–1109.

78. da Huang, W., Sherman, B. T., & Lempicki, R. A. (2009).Systematic and integrative analysis of large gene lists usingDAVID bioinformatics resources. Nature Protocols, 4, 44–57.

79. Dzau, V. J., Austin, M. J., Brown, P., Cowley, A., Housman, D.,Mulligan, R., et al. (1999). Revolution and renaissance.Physiological Genomics, 1, 1–2.


http://dx.doi.org/10.1007/s00395-012-0245-9

Documents

Evaluation of Right and Left Ventricular Diastolic Filling