Optimization of Right Ventricular Function Preoperatively for LVAD Implantation

Fig. 3.1
(a) The inlet, trabeculated apical myocardium and infundibulum of the RV. The tricuspid and pulmonary valves are separated by the ventriculoinfundibular fold (VIF). (b) Short-axis plane of the RV demonstrating its crescentic shape. (c) The four-chamber anatomic plane of the heart showing the moderator band (MB) and the more apical insertion of the tricuspid valve. (d) Superficial muscle layer of the RV (dissection by Damian Sanchez-Quintana, University of Extremadura, Spain). SMT indicates septomarginal trabeculation with its anterior (a) and posterior (p) arm; A-S anterosuperior leaflet of the tricuspid valve, PT pulmonary trunk, Ao aorta, RA right atrium, LA left atrium (Reproduced with permission from Ho SY, Nihoyannopoulos P. Anatomy, echocardiography, and normal right ventricular dimensions. Heart. Copyright © 2006, BMJ Publishing Group Ltd)

Table 3.1
Structural findings of abnormal RV

Structural characteristics of RV





Volume > 101 mL/m2

Volume overload

RV max SAX >43 mm

Pressure overload


Intrinsic myocardial disease

D-shaped LV

aEccentricity index >1

RV pressure or volume overload

Diastolic D-shaped LV suggests RV volume overload

Systolic D-shaped LV is RV pressure overload


Mass > 35 g/m

Pressure-overloaded RV

RV infarction wall >5 mm

Hypertrophic cardiomyopathy infiltrative disease, exclude double-chambered RV


Localized RV dilation

AVRD, RVMI; localized absence of pericardium

TV septal insertion

Septal insertion > 1 cm or 8 mm/m

Consider Ebstein’s anomaly

Delayed hyperenhancement

Area of delayed contrast uptake and washout in MRI

Suggests myocardial fibrosis

Fatty infiltration

High-intensity signal on MRI

Consider ARVD

Haddad, F., et al., Circulation, 2008. 117 (11): p. 1436–48 [7]

RV max SAX indicates RV maximal short-axis diameter, RVEDA/LVEDA ratio of RV to LV end-diastolic area, ARVD arrhythmogenic RV dysplasia, RVMI RV myocardial infarction, TV tricuspid valve

aThe eccentricity index measures the degree of septal displacement and is defined as the ratio of the minor axis diameter of the LV parallel to the septum to that perpendicular to it

RV myocardium is composed of two layers: a superficial layer with circumferential fibers running parallel to the atrioventricular groove and a deep layer with fibers aligned longitudinally from base to apex. This differs from the LV, where oblique fibers are superficial, longitudinal fibers are in the subendocardium, and circumferential fibers are in between. The septum is shared and structurally similar to the LV, providing an intimate anatomic and functional relationship between the two, a basis for ventricular interdependence. Multiples studies have indicated that septal contribution to RV cardiac output is significant, ranging between 20 and 60% [8, 9]. Hoffman demonstrated that in a non-dilated RV, even after replacing the free wall with a noncontractile material (Dacron patch), the septum was able to maintain sufficient RV cardiac output for hemodynamic stability [10].

Normal RV Physiology

The RV contracts by three separate mechanisms, the first of which is an inward movement of the free wall followed by contraction of deeper longitudinal muscle fibers that draw the tricuspid annulus toward the apex and finally traction of the free wall from LV contraction generating a forward stroke volume. Normal baseline RV ejection fraction (RVEF) ranges from 40 to 76% and varies based on loading conditions [7].

RV function, like the LV, is affected by preload, afterload, and contractility. This complex relationship is best illustrated by differences in pressure-volume curves. The RV pressure-volume loop is more triangular in shape with the pulmonary valve opening early in systole once RV pressure reaches the low pulmonary pressure. As there is very little time spent in isovolumetric contraction, the loop assumes a triangular configuration in contrast to a square LV loop. This also hints to the fact that the RV performs primarily volume work [11] (Fig. 3.2).


Fig. 3.2
The solid line depicts the square-shaped pressure-volume loop for the LV (solid line) and the triangular-shaped pressure-volume loop for the RV (broken line)

The slope of the end-systolic pressure-volume relationship is called ventricular elastance and is a relatively load-independent measure of ventricular contractility. Dell’Italia demonstrated that the normal maximal RV elastance was 1.3 ± 0.84 mmHg/mL, four times lower than the LV. Consequently, the RV is more afterload sensitive than the LV [12]. This is demonstrated in the acute setting (i.e., massive pulmonary embolism), where RV stroke volume decreases significantly with sudden increase in pulmonary artery pressure.

Similar to the LV, in a normal RV, based on the Frank-Starling principle, an increase in preload improves contraction. However, in RV failure, the curves flatten and move down and to the right depict a drop in RV output with increase in preload (Fig. 3.3).


Fig. 3.3
Comparison of RV and LV Starling curves. LV requires higher atrial filling pressures (AP) to produce equivalent cardiac output (CO). In RV failure (RVF), the curve moves downward and to the right.

The pericardium encases both ventricles and under normal circumstances provides an important protective barrier from overdistention and direct pathogen invasion. In addition, it also imparts a diastolic interdependence effect on both ventricles [13, 14]. Bernheim first hypothesized the importance of the interventricular relationship [15]. Henderson and Prince later demonstrated that volume and pressure loading of one ventricle decreased the output of the other [16]. This was clinically demonstrated in 1956 by Dexter as deterioration of LV function in patients with atrial septal defects who developed RV pressure and volume overload, a phenomenon called “reverse Bernheim effect” in which he postulated that the leftward septal shift resulted in impaired LV filling [17] (Fig. 3.4).


Fig. 3.4
RV size and function. Two perpendicular sections of a 3D TEE reconstruction of the right ventricle from tricuspid valve [TV] to pulmonary [PV] valve are shown. The cross section [a] demonstrates the crescent shape and the sagittal section and [b] the triangular shape of the RV. Ventricular interdependence between the left ventricle [LV] and RV during systole relies on interventricular septum position as shown in cross section for different clinical scenarios (Reproduced with permission from Meineri M, Van Rensburg AE, Vegas A. Right ventricular failure after LVAD implantation: prevention and treatment. Best Pract Res Clin Anaesthesiol. 2012)

Measuring Normal RV Function

Multiple methods have been used to measure RV function clinically. Cardiac MRI is the most accurate tool to assess RV diastolic and systolic volumes as well as the RVEF [7]. By MRI, RVEF ranges 47–76%. A technique not used frequently is radionuclide angiography by which RVEF is usually 40–45%. Echo is least accurate in assessing RVEF but the most clinically used. Two-dimensional echo assessment by Simpson’s rule can be used and correlates well with MRI; however, this is dependent on the quality of the images. RV fractional change area can be measured in four-chamber views and easily incorporated into most echo reports. Tricuspid annular plane systolic excursion (TAPSE) is another useful quantitative measurement of RV systolic performance [18, 19]. RV myocardial performance index, a ratio of isovolumic time interval to ventricular ejection time, doesn’t involve ventricular geometry and is a load-independent measure of RV function. Finally, tissue Doppler imaging also allows for quantitative RV assessment. Finally, myocardial strain and speckle-tracking analysis can also be used to define RV function [20, 21] (Table 3.2).

Table 3.2
RV function echocardiography indices, Haddad, F., et al., Circulation, 2008. 117(11): p. 1436–48 [7]

RV contractility indices

Functional parameters

Normal value

Load dependencea

Clinical use


61 ± 7% (47–76%)


Clinical validation, wide acceptance


Prognostic value in cardiopulmonary disorders




Good correlates with RVEF

Prognostic value in MI and bypass surgery




Simple measure, not limited by endocardial border recognition: Good correlation well with RVEF

Sm annular, cm/s



Good sensitivity and specificity for RVEF <50%


Basal: 19 ± 6


Correlates with stroke volume

Mid: 27 ± 6

Apical: 32 ± 6

Strain rate, s−1

Basal: 1.50 ± 0.41


Correlates with contractility

Mid: 1.72 ± 0.27

Apical: 2.04 ± 0.41


0.28 ± 0.04


Global nongeometric index, index of systolic and diastolic function, prognostic value in PH and CHD

dP/dt max, mmHg/s



Not a reliable index of contractility

More useful in assessing directional change when preload accounted for

IVA, m/s2

1.4 ± 0.5


Promising new noninvasive index of contractility, studies in CHD

Maximal RV elastance mmHg/mL

1.30 ± 0.84


Most reliable index of contractility

RVFAC indicates RV fractional area change, MI myocardial infarction, TAPSE tricuspid annular plane systolic excursion, Sm tissue Doppler maximal systolic velocity at the tricuspid annulus, RVMPI RV myocardial perfusion index, PH pulmonary hypertension, CHD congenital heart disease

aShould be viewed as a general indication of load dependence

Invasive hemodynamic assessment is the gold standard to assess RV function and can often tease out acute RVF from chronic RV dysfunction. Direct measurement of RA, RV, and PA pressures can clarify inconclusive noninvasive data. RV stroke work index (RVSWI) , pulmonary artery pulsatility index (PAPi) , and a RA to pulmonary capillary wedge pressure ratio have all been used to measure RV function [22] and predict RV failure post-LVAD implantation. At the Texas Heart Institute, we routinely use noninvasive parameters including TAPSE, tissue Doppler imaging (TDI) , PAPi, and RA/PCWP ratio to aid in clinical decision-making (Table 3.3).

Table 3.3
Assessing normal RV performance using hemodynamic parameters



Desirable value

RV size


RVEDV <200 mL, RVESV <177 mL

Central venous pressure (CVP)


<15 mmHg, 5 mmHg < PCWP

Transpulmonary gradient (TPG)


<15 mmHg

Pulmonary vascular resistance (PVR)



RV stroke work index (RVSWI)


>300–600 mmHg mL/m2

Pulmonary artery pressure index (PAPi)



Right atrial pressure to pulmonary capillary wedge pressure ratio



MPA mean pulmonary artery pressure, CO cardiac output, RVEDV RV end-diastolic volume, RVESV RV end-systolic volume, PASP pulmonary artery systolic pressure, PADP pulmonary artery diastolic pressure

Right Heart Failure Definition

The definition of RVF remains nebulous due to the use of inconsistent criteria in different publications. Some authors have described RVF as a need for intravenous inotrope or pulmonary vasodilator therapy for 14 days postoperatively and/or need for RVAD, while others defined it as simply a requirement for RVAD. Others still have used two or more of the following hemodynamic parameters to define RVF: central venous pressure greater than 16 mmHg, mean arterial pressure lower than 55 mmHg, cardiac index less than 2.0 L/min/m2, inotrope support >20 units, and mixed venous saturation lower than 55%, all in the absence of cardiac tamponade [2328].

According to the INTERMACS registry , RVF is present if symptoms or findings characterized by both elevation of central venous pressure (right atrial pressure > 16 mmHg on right heart catheterization, significantly dilated inferior vena cava with no inspiratory variation on echocardiography, and elevated jugular venous pressure) and manifestations of elevated CVP (peripheral edema, ascites, or hepatomegaly on exam or diagnostic imaging and laboratory evidence of worsening hepatic total bilirubin >2.0 mg/dL and renal dysfunction creatinine >2.0 mg/dL) are present [29] as illustrated in Table 3.4.

Table 3.4
Interagency Registry for Mechanically Assisted Circulatory Support definition of right ventricular failure

Interagency Registry for mechanically assisted circulatory support definition of right ventricular failure

RVF definition

Symptoms or findings of persistent RVF characterized by both of the following:

 – Right atrial pressure > 16 mmHg on right heart catheterization

 – Significantly dilated inferior vena cava with no inspiratory variation on echocardiography

 – Elevated jugular venous pressure

Manifestations of elevated CVP characterized by:

 – Peripheral edema (>2+)

 – Ascites or hepatomegaly on exam or diagnostic imaging

 – Laboratory evidence of worsening hepatic (total bilirubin >2.0 mg/dL) or renal dysfunction (creatinine >2.0 mg/dL)

Severity scale


Patient meets both criteria for RVF plus:

 – Post-implant inotropes, inhaled nitric oxide, or intravenous vasodilators not continued beyond post-op day 7 after VAD implant

 – No inotropes continued beyond post-op day 7 after VAD implant


Patient meets both criteria for RVF plus:

 – Post-implant inotropes, inhaled nitric oxide, or intravenous vasodilators continued beyond post-op day 7 and up to post-op day 14 after VAD implant patient meets both criteria for RVF plus:

 – CVP or right atrial pressure > 16 mmHg

 – Prolonged post-implant inotropes, inhaled nitric oxide, or intravenous vasodilators continued beyond post-op day 14 after VAD implant


Patient meets both criteria for RVF plus:

 – CVP or right atrial pressure > 16 mmHg

 – Need for right ventricular assist device at any time after VAD implant

Severe acute

Death during VAD implant hospitalization with RVF as primary cause

CVP central venous pressure, RVF right ventricular failure, VAD ventricular assist device (Reproduced with permission [29, 47])

RV failure results from a number of reasons and similar to the LV can be both systolic and diastolic in nature. While the RV can accommodate increased preload, it is sensitive to increased afterload and elevations in PA pressures [29, 30]. As PA pressures increase due to worsening left-sided function, there is delayed pulmonary valve opening, leading to increased RV work and O2 consumption. Moreover, this leads to progressive RV dilation, wall stress, and impaired coronary perfusion pressure. As the dilation progresses, geometry changes lead to tricuspid annular dilation and functional tricuspid regurgitation due to non-coaptation of leaflets. Abnormal septal activation also disrupts normal IVS function [29, 30]. Over time, if the heart failure remains untreated, cardiomyocyte stress and hypertrophy lead to irreversible apoptosis [29].

RV infarction due to obstructive coronary disease is another mechanism, which can lead to RVF although the incidence of RV infarct post-MI is low and is usually due to an isolated inferior wall MI. The likely reason for the lower incidence of ischemic RV failure compared to ischemic LV failure is probably lower RV wall stress and stroke work in addition to smaller mass requiring lower resting coronary flow and O2 extraction [31].

Pathophysiology of Right Heart Failure After LVAD Implantation

RV failure after LVAD implantation results from a complex sequence of events in the setting of underlying risk factors. Multiple mechanisms have been suggested. LV decompression and increased cardiac output increase venous return to the RV. Intraoperative volume resuscitation including blood transfusions also contributes to this increase in RV preload and can aggravate a decompensated RV [32]. Abnormal interventricular septum (IVS) geometry due to excessive leftward shift at high LVAD speeds can also worsen RV function due to loss of IVS contribution to RV output [33].

Ischemic injury is another mechanism seen after prolonged bypass times, coronary ischemia, and/or loss of coronary bypass grafts or coronary embolism. Between 30 and 64% of patients with advanced HF will have associated tricuspid regurgitation, which improves after LV decompression with a LVAD [34, 35]. However, a dilated tricuspid annulus or an incompetent valve generally worsens TR after LV decompression and increased preload. Severe TR further contributes to RV failure through the development of right-sided volume overload and reduced RV ejection. Although pulmonary artery pressures improve after LV decompression, perioperative ischemia-related pulmonary endothelial injury and transfusion-related lung injury often conversely increase pulmonary vascular resistance and can result in RV failure.

Supraventricular arrhythmias are seen in more than 20% of patients after LVAD implantation and have been associated with twice the risk of RV failure [36]. More sinister rhythms like ventricular fibrillation have been associated with a 32% drop in cardiac output. Incessant postoperative VT can therefore adversely affect RV function in LVAD patients and should be avoided as far as possible [37].

Predicting Right Heart Failure After LVAD Implantation

Over the last three decades, many centers have worked to develop algorithms and risk scores to predict RVF after LVAD implantation . Early identification of high-risk patients remains important as it allows for the formulation of strategies to avoid RV failure. Unfortunately, most risk scores devised from retrospective, small single-center experiences provide a variable spectrum of predictors including hemodynamic, echocardiographic, biochemical, and intra- and postoperative parameters with no single model dependably forecasting RVF. Many early studies incorporated pulsatile devices in BTT cohorts and therefore did not accurately reflect outcomes in the current CF-VAD era. In addition, validation of many of these scores has demonstrated the modest real-world application [38]. In our center, we have noted that a preoperative, systemic inflammatory syndrome associated with a leukocytosis and thrombocytopenia may prime the RV for failure.

Hemodynamic Models

Fukamachi et al. reported RVAD support requirement for 11 out of 100 patients after HeartMate XVE pulsatile LVAD implantation. RVAD use was significantly higher in young, female patients with small BSA and those with myocarditis. There was no significant difference in the cardiac index, RV ejection fraction, or right atrial pressure between groups preoperatively. Low preoperative mean pulmonary arterial pressure (PAP) and RV stroke work index (RVSWI) were associated with the need for post-op RVAD. Survival to transplant was poor in the RVAD group, 27% vs. 83% in the no-RVAD group. However, the incidence and underlying mechanisms of RV failure changed after the introduction of continuous-flow LVADs (CF-LVAD) [39].

The right ventricular failure risk score (RVFRS) evaluated 197 patients undergoing HM II CF-LVAD implantations. Sixty-eight cases (35%) were complicated by postoperative RV failure. Points were given for need for vasopressors, elevation in aspartate aminotransferase (>80 IU/L), bilirubin (>2.0 mg/dL), and creatinine (>2.3 mg/dL). All were found to be independent predictors of RV failure. The odds ratios for RV failure for patients with an RVFRS of 3.0, 4.0–5.0, and 5.5 were 0.49 (95% confidence interval [CI], 0.37–0.64), 2.8 (95% CI, 1.4–5.9), and 7.6 (95% CI, 3.4–17.1), respectively, and 180-day survival of 90 ± 3%, 80 ± 8%, and 66 ± 9%, respectively (P < 0.0045) [40]. The different studies and RV failure risk models are listed in Table 3.5.

Table 3.5
Clinical trials evaluating hemodynamic parameters for RV failure post-LVAD implantation (Reproduced with permission [47])



VAD type

RVF definition


Risk factors/scores


Ochai [27]


100% pulsatile VAD

Need for RVAD


Pre-op circulatory support (OR 5.3)


Female gender (OR 4.4)

BTT 98%

Nonischemic etiology (OR 3.3)

Drakos [40]


86% pulsatile VAD

Need for RVAD


(1 point for each)

365-day post-LVAD survival:


14% CF-VADs

≥14 days inotropes

Destination therapy (OR 3.31)

≤5.0 = 83%

iNO ≥ 48 h

Inotrope dependency (OR 2.47)

5.5–8.0 = 77%

Retrospective analysis
Obesity (BMI ≥ 30 kg/m2) (OR 1.99)

8.5–12.0 = 71%

BTT 58%
IABP (or 3.88)

≥12.5 = 61%

RVF % for risk score categories:
 1.8–2.7 Wu (or 1.95)

≤5.0 = 11%
 2.8–4.2 Wu (or 3.01)

5.5–8.0 = 37%

Only gold members can continue reading. Log In or Register to continue

Feb 24, 2018 | Posted by in CARDIOLOGY | Comments Off on Optimization of Right Ventricular Function Preoperatively for LVAD Implantation
Premium Wordpress Themes by UFO Themes