Implantation of a continuous flow left ventricular assist devices (CF-LVAD or LVAD) in patients with advanced heart failure (HF) increases survival and improves quality of life.1 Functional capacity is also improved postimplantation,1–4 but peak oxygen uptake (pVO2) remains on average reduced to 50% of the expected value.1 Right ventricular (RV) function is of importance for circulatory function after LVAD implantation, and the association between RV failure (RVF) and outcome early after implantation has been repeatedly documented.5–7 However, the importance of RV function for exercise capacity post-LVAD implantation has not been extensively studied. One reason for this is the difficulty related to imaging of the RV in patients with LVADs, especially current models, such as the HeartMate 3 (HM 3). Echocardiography will often provide suboptimal apical acoustic windows because of the position of the pump, and magnetic resonance imaging (MRI) is contraindicated in this patient group. Emerging evidence suggests that gated cardiac computed tomography (CCT) scans can provide high-resolution images with a frame rate allowing for dynamic evaluation of RV function.5,8 Studies using this technique in LVAD recipients are sparse,6 and the ability to evaluate RV function during stress has not been explored.
Using 4D CCT scans (to minimize motion-related artifacts) at rest and during physical stress, the purpose of this study was to examine right heart systolic function and the association to exercise capacity in LVAD recipients.
Stable LVAD recipients aged > 18 years without ongoing infection (HM II or 3, [Abbott, Pleasanton, CA]) at the outpatient clinic of advanced Heart failure at Rigshospitalet, Copenhagen, Denmark, were consecutively offered inclusion in the study, between April 2016 and June 2017. Both bridge to transplantation (BTT) and destination therapy (DT) patients could be included. Implantation of CF-LVAD less than 1 month before the study was an exclusion criterion as were contrast allergy or estimated glomerular filtration rate lower than 45 ml/min. The trial was approved by the Ethics Committee of Copenhagen in accordance with the Helsinki declaration (Project no. H-150-21-598), and the trial was registered at www.clinicaltrials.gov (NCT02658136). All participants provided informed, signed consent before inclusion.
Medical history and current list of medications were obtained from patient medical records.
Setup and Equipment
All tests (i.e., blood sample, echocardiography, pVO2, regular-CCT, and postexercise CCT) were obtained on the same day. All protocols are explained in the following section.
First, blood samples were drawn, followed by regular resting transthoracic echocardiography (Philips iE33 cardiac ultrasound system [Philips Healthcare, Best, Netherlands]) and cardiopulmonary exercise testing (CPET) using an upright ergometer bicycle (Schiller CS- 200; Schiller AG, Bar, Switzerland). After minimum two hours of rest, patients underwent a regular, resting CCT (320-slice MDCT Toshiba VISION Edition Aquilion One Scanner). A bedside cycle ergometer (Fabrication Enterprises, New York, NY) was then placed inside the CCT scanner, and the patient was asked to pedal zealously for 2 minutes. After exactly two minutes, the second CCT was initiated and then the patient was told to stop cycling while the bedside cycle remained inside the scanner.
All CCTs were enhanced using intravenous Visipaque 320-contrast infusion.
All blood pressures were measured noninvasively (SunTech Medical, Inc, Morrisville, NC).
Protocols and Measurements
Echocardiography (Philips Healthcare, Best, Netherlands) was performed in accordance with current guidelines.7,9
The examination included two-dimensional (2D), tissue color Doppler imaging (to evaluate valve pathology) and continuous wave Doppler images acquired in the views: parasternal long-axis, apical four-chamber, and five-chamber.
Echocardiograms were analyzed and documented by a reader blinded to other study measures.
Before each CPET, calibration was performed in accordance with manufacturer’s instructions (for gas, ambient conditions, and flow). Breath-by-breath respiratory gas analysis measuring oxygen consumption (VO2), carbon dioxide excretion (VCO2), and expiratory minute ventilation (VE) was used. Peak oxygen uptake was expressed as milliliters of oxygen per kilogram of body mass per minute. The participants were encouraged to keep cycling until exhaustion and were monitored with pulse oximetry and continuous 12-lead electrocardiogram during exercise. Peak heart rate and blood pressure were measured continuously during the test.
Starting load was 25 Watt, followed by an increase of 10 Watt per minute. The pedaling rotation was aimed at 60–70 rpm.
Two sequential CCTs with contrast were performed as previously mentioned; one at rest and one immediately after two minutes of zealous exercise on a bedside, supine 25 Watt ergometer bike (Fabrication Enterprises, New York, NY). The postexercise CCT protocol was initiated within 1 second after the exercise bout; the CCTs were undertaken with intravenous Visipaque 320-contrast infusion of 5–6 ml/second.
All scans were performed with sole focus on the RV. To limit contrast use, scans were initiated manually ensuring maximal contrast in RV, which subsequently led to unattainable LV measures because of limited contrast in the LV.
Using ECG-gated CCT (320-slice MDCT Toshiba VISION Edition Aquilion One Scanner), cardiac scans were obtained. Using Vitrea software (Vital images, Minnetonka, MN), the following measurements were obtained: right atrial (RA) end-diastolic and end-systolic volumes, RV end-diastolic and end-systolic volumes. Volumes were indexed for body surface area using the DuBois formula.10
Borg Rating of Perceived Exertion Scale (Borg)11 was recorded directly after each exercise test terminated. Furthermore, Borg was recorded two minutes post CPET initiation, that is, corresponding to the duration of the exercise test in the CCT scanner, although this Borg value was missing for four patients.
Betablocker (BB) equivalent doses were calculated as metoprolol 100 mg ~ Bisoprolol 5 mg ~ Carvedilol 25 mg ~ Labetalol 200 mg ~ Propranolol 80 mg.
Furosemide equivalent doses were calculated as Bumetanide 1 mg ~ Furosemide 40 mg.
The %Predicted pVO2 (%Pred pVO2) was calculated using the equation12
The primary outcome variable was CCT-derived RVEF and pVO2 during exercise. Previous analyses have estimated a 7.7% standard deviation (SD) for RVEF measurements (using echocardiography3 and CCT6) and 6.3 ml/kg/min for pVO2 in CF-LVAD patients. As the current study was explorative, a meaningful power calculation could not be performed.
Analyses were performed using Statistical Package for the Social Sciences (version 22.0; SPSS Inc., Chicago, IL). Statistical significance was defined as a two-tailed p value below 0.05.
Differences between groups were assessed using paired or unpaired t-tests as appropriate. Continuous variables were expressed as mean ± SD if parametric, and if nonparametric, as median and interquartile range (IQR). Categorical variables were described with frequency and percentage. Relations between continuous variables were calculated as correlation coefficients, if data were parametric distributed using Pearson’s coefficient and otherwise Spearman’s rank correlation coefficient.
Fifteen patients (age range 32–72 years) were included in this study, of whom 13% were female (Table 1). Four patients had an HM II, and 11 had an HM 3. Pump speed settings for HM II were (mean ± SD) 9500 ± 258 and 5518 ± 388 rpm for HM 3.
Baseline characteristics were obtained at the time of pVO2 test and are shown in Table 1.
In most cases, the LVAD had been implanted as DT or BTT; however, given the long waiting time for transplantation, support duration was longer than one year (median [IQR] 313  days).
All were outpatients and on average 377 (range 43–1498) days postimplantation.
Mean pVO2 value was 15 ± 5 ml/kg/min, Table 2. Patients were adequately stressed as evidenced by a respiratory exchange rate > 1.05 in 87% and limitation stated as fatigue or breathlessness.
CCT revealed a resting RA ejection fraction (RAEF) of 18.1% ± 9% and resting RV ejection fraction (RVEF) of 36% ± 8%. Stress RAEF was 19% ± 10%, and stress RVEF was 37% ± 8%.
Correlation analyses revealed no significant correlations between pVO2 and RAEF or RVEF at rest or during stress.
Resting-RAEF correlated significantly with and stress RAEF: r = 0.873, p < 0.0001. The same association was seen between resting-RVEF and stress RVEF: r = 0.76, p = 0.001; Figure 2.
RVEF was < 35% in N = 6 (40%) and at rest and N = 4 (27%) during stress. When comparing pVO2 in patients with RVEF > 35% vs. RVEF < 35% no difference was found at rest (p = 0.86) or during stress test (p = 0.43).
Furthermore, peak HR was significantly associated with pVO2 (r = 0.558, p = 0.031), but not %Pred pVO2 (r = 0.446, p = 0.096).
Lastly, the RVEF versus pVO2 plot is shown in Figure 3 and RVEF versus VE/VCO2 is shown in Figure 4.
Patients with atrial fibrillation (AF) had lower atrial EF values at rest and during stress than those with sinus rhythm (SR); resting-RAEFAFversus resting-RAEFSR 10.4% ± 2.1% vs. 21.2% ± 8.6% (p = 0.02), stress RAEFAFversus stress RAEFSR 7.3% ± 1.4% vs. 23% ± 8.3% (p = 0.004), whereas neither resting-RVEFAF nor stress RVEFAF were different from values obtained in patients in SR (32.7% ± 3.5% vs. 36.4% ± 8.7% [p = 0.3] and 36% ± 5.8% vs. 36.4% ± 9% [p = 0.8], respectively). Patients with AF did not exhibit lower pVO2 values than patients with SR (p = 0.19).
Heart Failure Etiology
Of the 15 patients, 60% had a nonischemic etiology. No significant difference was seen in pVO2 values between patients with nonischemic cardiomyopathy (N-ICM) and patients with ICM, respectively, 16.23 ± 5.17 and 12.12 ± 3.49 ml/kg/min, p = 0.087.
Furthermore, RAEF and RVEF did not differ between HF etiology subgroup (ICM versus N-ICM) during stress or at rest.
Left ventricular ejection fraction (LVEF) was 15% ± 9% and did not significantly differ between N-ICM versus ICM patients (p = 0.75).
Because of poor acoustic windows, a reliable TAPSE (1.1 [0.54]) and RVEDD (3.45 [0.6]) was acquired from only 6 (40%) patients.
The aortic valve was closed in 5 patients, intermittently open in 5 patients, continuously open in 2 patients, and assessable in 3 patients. Moderate aortic regurgitation was seen in a single patient. No other patient suffered from moderate or severe valve pathology of any kind. No correlations were found between the mentioned aortic valve status and pVO2.
Betablockers were prescribed to 80% of the study population, with a (mean [SD]) dose of 86 ± 67 mg (metoprolol equivalents). Ejection fractions were not found significantly different between patients receiving BBs and those not receive BBs.
Study Completion, Radiation, and Contrast Use
There were no drop outs, and no adverse events (such as LV suction or arrhythmias etc.) were recorded during the study.
Mean of total radiation was 23.3 ± 7.5 mSv, radiation during rest-CCT was 13.2 ± 6.2 and during stress-CCT was 10.1 ± 3.2. Total contrast volume did not exceed 140 ml.
Right ventricular and atrial volume curves are depicted in Figure 1. The curves A–D are shown during a single cardiac cycle during rest and stress.
The current study is the first to present data of RV and atrial ejection fractions during rest and immediately after exercise using CCT imaging in LVAD recipients. Our main finding was that pVO2 in LVAD recipients was not predicted by RVEF. Also, resting EF values predicted EF during stress for both RA and RV.
In LVAD-supported patients, the relation between postimplantation RVF and increased morbidity and mortality is well established.13–15 RVF can be present before the LVAD implantation and/or develop after the implantation. Development of RVF after LVAD implantation depends on a number of factors,14 including hemodynamic variables, such as RV preload16 and afterload.17 In most patients, RV function improves1 postimplantation, and earlier single-center studies have shown post-LVAD improvement of RVF17–19 generally because of the reduced RV afterload.17
Right ventricular function has been shown to be of importance for exercise capacity in HF patients not receiving mechanical circulatory support. Based on the clinical observation that in patients with chronic LV failure pVO2 was only increased by venodilating agents and not arterial vasodilators, Baker et al.20 conducted a study that showed RVEF was indeed significantly correlated with pVO2 in patients with chronic LV failure (r = 0.7, p < 0.001). The study further showed that LVEF did not correlate with pVO2 in the same patient cohort.
However, to our knowledge, no previous studies have directly addressed the relation between RV function and pVO2 after LVAD implantation.
Although our data show lack of an association between resting RVEF and pVO2, and appear conclusive, we cannot entirely rule out that longer duration or greater intensity of exercise might have yielded different results regarding exercise RVEF. Indeed, to avoid motion artifacts, exercise CCT scans were performed immediately after, rather than during maximal exercise, and the exercise intensity was low owing to the experimental circumstances. The strong association between resting and stress RAEF/RVEF might indicate that the relatively low level of physical exercise or the elapsed time might not enable us to unmask stress-induced changes in RVEF with the protocol applied. Further studies are required to evaluate these factors.
Hypothetically, it cannot be excluded that a correlation between RVEF and pVO2 exists in patients with more advanced RV dysfunction than that observed in this study. However, RVEF was reduced, mean N-terminal pro-B-type natriuretic peptide (NT-proBNP) was significantly elevated (> 2000 pg/ml), and the majority of patients required diuretic therapy. As such, they compare well with other LVAD populations.
It is also evident that a large number of other factors play a role for exercise capacity in these patients, such as LV recovery, pump speed settings, chronotropic incompetence, and peripheral factors, body position during exercise, and as such, RV function may not be expected to be the decisive limiting factor.1
Elshazly et al.21 (N = 1744) showed in patients with HF with preserved EF (HFpEF) that AF is associated with work intolerance. As HFpEF may to some extent be comparable to post-LVAD HF, and because AF is associated with HF admissions in LVAD recipients,22 we speculated if AF played a similar role in this cohort. However, even though AF was associated with lower RAEF, we did not see an association with RVEF nor did we observe a correlation with pVO2. Thus, our findings are in accord with the earlier study by Enriquez et al.23 that found no difference in pVO2 in patients with AF (permanent/paroxysmal) and SR.
The current study clearly showed that CCT is a valid method for measuring RV systolic function in LVAD recipients in accordance with findings previously reported by Garcia-Alvarez et al.6 We have extended the methodological investigation and shown that it is possible to examine the patients in close relation to physical exercise without losing information about right heart chamber function. This method has great promise for future studies of RV function in LVAD patients complaining of symptoms compatible with recurrence of HF, as well as for studies aiming to understand the factors underlying limitation of exercise in this patient group. The use of contrast could be a limitation for the use of this technique in LVAD recipients, and in the current study, a reasonable renal function was required. However, we did not see any adverse events in the study. Potential development of better protocols allowing for reduced contrast use and minimization of radiation will allow for broader use in this population.
Strengths and Limitations
The foremost strength of our study is the prospective design and execution of all examinations in one day in stable patients minimizing clinical variability. Main limitations include the small sample size (preventing subgroup analyses of for instance correlations in axial versus centrifugal pumps). Furthermore, the length of support varied among the LVAD recipients; however, the length of support did not influence exercise capacity in this cohort.
Lastly, we cannot exclude that RV performance may be better evaluated by other measures not investigated in the current study, such as invasive measures of RV-pulmonary artery coupling or pulmonary artery compliance.
Resting-EF predicts EF during stress for both RA and RV, but RVEF and RAEF do not determine pVO2 after LVAD implantation. CCT is a novel, promising method for studying right function at rest and during stress in LVAD recipients.
1. Jung MH, Gustafsson F. Exercise in heart failure patients supported with a left ventricular assist device
. J Heart Lung Transplant 2015.34: 489–496
2. Jung MH, Houston B, Russell SD, Gustafsson F. Pump speed modulations and sub-maximal exercise tolerance in left ventricular assist device
recipients: A double-blind, randomized trial. J Heart Lung Transplant 2017.36: 36–41
3. Jung MH, Hansen PB, Sander K, et al. Effect of increasing pump speed during exercise on peak oxygen uptake
in heart failure patients supported with a continuous-flow left ventricular assist device
. A double-blind randomized study. Eur J Heart Fail 2014.16:403–408
4. Gustafsson F, Rogers JG. Left ventricular assist device
therapy in advanced heart failure: Patient selection and outcomes. Eur J Heart Fail 2017.19: 595–602
5. Sigvardsen PE, Larsen LH, Carstensen HG, et al. Prognostic implications of left ventricular asymmetry in patients with asymptomatic aortic valve stenosis. Eur Heart J Cardiovasc Imaging 2018.19: 168–175
6. Garcia-Alvarez A, Fernandez-Friera L, Lau JF, et al. Evaluation of right ventricular function and post-operative findings using cardiac computed tomography
in patients with left ventricular assist devices. J Heart Lung Transplant 2011.30: 896–903
7. Hayek S, Sims DB, Markham DW, Butler J, Kalogeropoulos AP. Assessment of right ventricular function in left ventricular assist device
candidates. Circ Cardiovasc Imaging 2014.7: 379–389
8. Hindsø L, Fuchs A, Kühl JT, et al. Normal values of regional left ventricular myocardial thickness, mass and distribution-assessed by 320-detector computed tomography angiography in the Copenhagen General Population Study. Int J Cardiovasc Imaging 2017.33: 421–429
9. Rudski LG, Lai WW, Afilalo J, et al. Guidelines for the echocardiographic assessment of the right heart in adults: A report from the American Society of Echocardiography endorsed by the European Association of Echocardiography, a registered branch of the European Society of Cardiology, and the Canadian Society of Echocardiography. J Am Soc Echocardiogr 2010.23: 685–713; quiz 786
10. Du Bois D, Du Bois EF. Clinical calorimetry: Tenth paper a formula to estimate the approximate surface area if height and weight be known. Arch Intern Med. 1916.XVII:863–871
11. Borg G. Psychophysical scaling with applications in physical work and the perception of exertion. Scand J Work Environ Health 1990.16(suppl 1): 55–58
12. Hansen JE, Sue DY, Wasserman K. Predicted values for clinical exercise testing. Am Rev Respir Dis 1984.129:S49–S55
13. Matthews JC, Koelling TM, Pagani FD, Aaronson KD. The right ventricular failure risk score a pre-operative tool for assessing the risk of right ventricular failure in left ventricular assist device
candidates. J Am Coll Cardiol 2008.51: 2163–2172
14. Dang NC, Topkara VK, Mercando M, et al. Right heart failure after left ventricular assist device
implantation in patients with chronic congestive heart failure. J Heart Lung Transplant 2006.25: 1–6
15. Sabashnikov A, Mohite PN, Zych B, et al. Outcomes and predictors of early mortality after continuous-flow left ventricular assist device
implantation as a bridge to transplantation. ASAIO J 2014.60: 162–169
16. Loghmanpour NA, Kormos RL, Kanwar MK, Teuteberg JJ, Murali S, Antaki JF. A Bayesian model to predict right ventricular failure following left ventricular assist device
therapy. JACC Heart Fail 2016.4: 711–721
17. Morgan JA, Paone G, Nemeh HW, et al. Impact of continuous-flow left ventricular assist device
support on right ventricular function. J Heart Lung Transplant 2013.32: 398–403
18. Kavarana MN, Pessin-Minsley MS, Urtecho J, et al. Right ventricular dysfunction and organ failure in left ventricular assist device
recipients: A continuing problem. Ann Thorac Surg 2002.73: 745–750
19. Furukawa K, Motomura T, Nosé Y. Right ventricular failure after left ventricular assist device
implantation: The need for an implantable right ventricular assist device. Artif Organs 2005.29: 369–377
20. Baker BJ, Wilen MM, Boyd CM, Dinh H, Franciosa JA. Relation of right ventricular ejection fraction
to exercise capacity in chronic left ventricular failure. Am J Cardiol 1984.54:596–599
21. Elshazly MB, Senn T, Wu Y, et al. Impact of atrial fibrillation on exercise capacity and mortality in heart failure with preserved ejection fraction: Insights from cardiopulmonary stress testing. J Am Heart Assoc 2017.6: e006662
22. Rossing K, Jung MH, Sander K, et al. Outcomes and hospital admissions during long-term support with a HeartMate II. Scand Cardiovasc J 2015.49: 367–375
23. Enriquez AD, Calenda B, Gandhi PU, Nair AP, Anyanwu AC, Pinney SP. Clinical impact of atrial fibrillation in patients with the HeartMate II left ventricular assist device
. J Am Coll Cardiol 2014.64: 1883–1890
24. Di Salvo TG, Mathier M, Semigran MJ, Dec GW. Preserved right ventricular ejection fraction
predicts exercise capacity and survival in advanced heart failure. J Am Coll Cardiol 1995.25: 1143–1153
25. Murninkas D, Alba AC, Delgado D, et al. Right ventricular function and prognosis in stable heart failure patients. J Card Fail 2014.20: 343–349