Pulmonary artery compliance (PAC) is increasingly recognized as an important component of right ventricular (RV) afterload. Pulmonary artery compliance can be estimated as stroke volume divided by pulmonary artery (PA) pulse pressure and takes into account both pulsatile and resistive components of RV afterload.1–4 Lower PAC is associated with poor prognosis in patients with left-sided heart failure and isolated right heart failure from pulmonary arterial hypertension.5,6 The poor prognosis may reflect unfavorable coupling between the RV and the pulmonary circulation, leading to RV failure.7,8
The relationship between PAC and pulmonary vascular resistance (PVR), or resistor–capacitor (RC) time, is fixed in the general population and in patients with pulmonary arterial hypertension.3,9,10 Increased left-sided filling pressures, however, may shift this normally fixed relationship leading to a lower PAC for any given PVR. The shift results in increased pulsatile afterload for the RV.5,11 Reduction of left-sided filling pressures with medical therapy can improve PAC.1,5 The effect of mechanical left ventricular support on PAC, a rapid experiment in pulmonary capillary wedge pressure (PCWP) normalization, has not been previously described. Long-term PCWP normalization using a left ventricular assist device (LVAD) is associated with improved PVR.12 It remains unclear how quickly these changes in PVR occur and whether LVAD implantation leads to a similar acute or chronic improvement in PAC.
Understanding changes in RV afterload post-LVAD implantation are important because RV failure after LVAD placement remains a significant problem. In addition, understanding how rapid changes in left ventricular filling affect PAC and PVR contributes to our understanding of fixed versus dynamic components of World Health Organization (WHO) group II pulmonary hypertension (PH). This is the most common form of PH and is defined as a mean PA pressure ≥25 mm Hg in the setting of a PCWP or left ventricular end-diastolic pressure >15 mm Hg).13
The core aims of this study were to 1) determine whether PAC normalizes after LVAD placement and 2) to better understand the time course of changes in RV afterload after LVAD placement.
Materials and Methods
The cohort included 64 participants with end-stage systolic heart failure, who underwent LVAD implantation as bridge to heart transplantation at University of Minnesota between October 2001 and June 2007. Informed consent was obtained from all participants and the University of Minnesota Committee on Human Research approved the study. Patients with biventricular assist devices were not included in the analysis.
Right heart catheterization was performed by a cardiologist preoperatively, within 48–72 hours postoperatively via retained PA catheters, and >30 days postoperatively at a separate right heart catheterization. Right atrial pressure, PA pressures, systemic arterial pressures, and PCWP were measured and recorded. Mixed central venous blood gas was sampled from the PA and cardiac output was calculated using a Fick equation with assumed oxygen consumption.
Hemodynamic Variable Calculations
Pulmonary artery compliance was approximated using the stroke volume divided by PA pulse pressure (ml/mm Hg).3,4,14,15 This method assumes that the pulmonary arterial system acts as a two 2 element Windkessel (essentially a compliant bag with a resistance at the end.) This approximation works best when the arterial system is stiff and does not account for wave reflection which can alter pulse pressure.16 Using this method, we calculated the compliance of the entire pulmonary arterial system and not just the pulmonary “artery” compliance. Pulmonary vascular resistance was calculated as mean PA pressure minus PCWP divided by cardiac output (Wood units). The RC time, or pulmonary arterial time constant (τ), was calculated as the product of resistance and compliance (seconds). Transpulmonary gradient was calculated by subtracting the PCWP from the mean PA pressure (mm Hg). The diastolic pulmonary gradient was calculated by subtracting the PCWP from the PA diastolic pressure.
Baseline characteristics for study participants were presented. Given normal distribution of hemodynamic variables, we used a paired Student’s t-test to characterize changes in hemodynamic variables between right heart catheterizations (preoperative, within 72 hours of LVAD placement, and >30 days after LVAD placement). The change in hemodynamics was also compared between pulsatile and continuous flow (CF) pumps.
We explored relationships between changes in PCWP and the change in PAC using linear regression. Adjusted models evaluated whether associations between hemodynamic parameters were independent of age, device type, change in cardiac index, and change in heart rate. Statistics were performed with STATA 13 (College Station, TX) and p values less than 0.05 was considered significant.
The cohort included 64 participants, 37 (58%) of whom had ischemic cardiomyopathy. Forty participants (63%) had HeartMate XVEs, 17 (26%) had HeartMate IIs, and seven (11%) had VentrAssist devices. Baseline characteristics of the study sample are displayed in Table 1.
The median length of LVAD follow-up time (before death, transplant, or the end of follow-up) was 6.5 months (interquarticle range [IQR] 4.6–10.8 months). All participants had right heart catheterization measurements preoperatively and within 72 hours of LVAD placement and 44 participants (69%) had follow-up right heart catheterization measurements >30 days after LVAD implantation. The median length of time before the “late” follow-up right heart catheterization was 4 months (IQR 2–5 months). Participants who did not complete follow-up right heart catheterization after 30 days failed to do so for a number of reasons including cardiac transplantation (n = 14), death (n = 4), or performance of the “late” right heart catheterization before 30 days after LVAD implantation (n = 2).
Left Ventricular Assist Device Implantation Leads to a Rapid and Sustained Increase in Pulmonary Artery Compliance
Within 72 hours of LVAD placement, mean PAC increased from 2.0 to 3.7 ml/mm Hg (p < 0.0001), mean PCWP decreased from 23 to 15 mm Hg (p < 0.0001), PVR decreased from 3.5 to 1.7 Wood units (p < 0.0001; Table 1). For subjects who had a drop in PCWP of greater than or equal to 5 mm Hg within 72 hours, the RC time increased from 0.31 to 0.39 (p <0.05). The separation in the PAC and PVR relationship curves in these patients reflects that for any given level of PVR, the PAC was less rigid after LVAD placement (Figure 1).
In participants with follow-up right heart catheterization >30 days after LVAD placement the improvement in PAC, PCWP, and PVR persisted (Table 2, Figure 2). The average compliance for remaining participants at the end of follow-up was 3.6 ml/mm Hg. All the remaining participants had an improved PVR to <4 Wood units; however, there were still 12 participants (27%) with a suboptimal PAC below 2.5 ml/mm Hg and the majority of participants (31 participants; 70%) with a PAC below 3.8 L/mm Hg.
The increase in PAC immediately after LVAD implantation was rapid and was strongly associated with changes in PCWP. For each 5 mm Hg decrease in PCWP, there was an unadjusted increase in PAC by 0.52 L/mm Hg, p < 0.0001. The adjusted model suggested the strong relationship between change in PCWP and change in PAC was not explained by differences in participant age, device, improvement in cardiac output, or reduction in heart rate (PAC increased by 0.44 L/mm Hg with each 5 mm Hg decrease in PCWP in adjusted models p < 0.0001).
This immediate change in PAC was not different between those who received a pulsatile versus a CF pump. The PAC increased by a mean of 1.6 ml/mm Hg for those patients who received a pulsatile pump versus 2.1 ml/mm Hg for those who received a CF pump (p = 0.39). Improvements in PCWP, mean PA pressure, and cardiac index in the first 48 to 72 hours were also similar between the CF and pulsatile pump groups (Table 3).
We have shown that PAC increased rapidly after LVAD implantation in a single institution cohort with advanced heart failure. This change was strongly associated with the improvement in PCWP. The degree of improvement in those who had a rapid drop in PCWP was even greater than would be predicted by the normally fixed PAC–PVR relationship, as reflected by the increase in RC time in these subjects. This means that the pulmonary vascular characteristics changed quickly such that compliance was better for any given level of PVR.
Most of the PAC increase occurred within the 72 hours after LVAD placement with minimal additional change for most participants beyond the initial 72 hour window. This did not appear to differ between pump types. These data imply that a large component of the abnormal PAC is because of pulmonary vascular loading, pressure, and distention and not because of more permanent pulmonary vascular changes, such as vascular remodeling. Previous studies of PAC in left heart disease have suggested similar findings with long-term medical treatment to decrease PCWP1,5; however, these studies have demonstrated changes in PAC or PVR over a significantly longer period of time. The time course of these previous studies left open the possibility that improvement in PAC or PVR were related to reverse remodeling of the pulmonary vasculatures over time. The rapid decrease in PCWP in this study that occurs with LVAD placement makes this a unique dataset to understand the immediate effects on PAC with LV unloading and better contextualize these previous results with a strong suggestion of the operative mechanism to explain the current and prior studies.
While the PAC improved considerably in these subjects, it is not presently known what value represents a normal versus abnormal PAC. To establish normal values of PAC, prior studies have included matched control subjects with normal left ventricular function, no valvular disease, and no PH. In these studies, a PAC ≥ 3.8 ml/mm Hg has been suggested as normal,16,17 however, a bias cannot be excluded because PAC formulas have been derived from populations undergoing right heart catheterization for a “clinical indication.” It is possible that the average PAC of 3.6 ml/mm Hg in our population at longer term follow-up is lower than in “normal” subjects, likely related to persistent elevations in PCWP and PVR in some patients.
Better understanding changes in PAC and PVR is important because RV failure after LVAD implantation remains a significant cause of morbidity and mortality and remains difficult to predict based on existing preoperative clinical measures.18,19 It has been hypothesized that PAC, relative to PVR, is a stronger clinical predictor for poor outcomes because it integrates abnormalities of PCWP and PVR into one measure.16 Both PVR and PCWP are likely to be clinically significant and may contribute separately and additively to poor outcomes. In addition, the relationship between PVR and PAC is nonlinear such that early changes in the pulmonary vasculature can be better appreciated by differences in PAC relative to PVR (which is more sensitive to differences in advanced pulmonary vascular disease).3,4
This study may suggest that RV failure after LVAD placement is likely not explained by a persistently elevated RV load because in most cases, the reduction in resistive and pulsatile RV afterload occurs early after implantation. Careful characterization of RV afterload including PAC, PVR, and RC time may be valuable when assessing patients with heart failure and PH who may be considered for LVAD placement. In the setting of very high PCWP, an LVAD can significantly reduce the RV pulsatile afterload by increasing PAC.
Understanding the relationship of PAC and PVR in the end-stage systolic heart failure population is also relevant to help understand PH that occurs in patients with left heart failure (WHO group II PH). Despite initial optimism, therapies developed for PAH have enjoyed little success in PH with left heart failure.20,21 The rapid improvement in PVR and PAC in this study would suggest most group II PH (even those with a precapillary component and elevated PVR) is functional—the result of altered PAC from elevations in left heart filling pressures and pulmonary vascular “over-distention” or reversible pulmonary vasoconstriction, rather than arterial remodeling. This reinforces the recognition that WHO group II PH has a markedly distinct pathophysiology from pulmonary arterial hypertension, which is a disease of progressive pulmonary vascular remodeling and narrowing. Although there may me a minority of patients with WHO group II PH and homology to pulmonary arterial hypertension, the distinct physiologic differences we describe likely helps explain why medications developed for PAH are not effective for most patients with PH related to left heart failure.
This was a retrospective observational cohort and hemodynamic data for some participants over 30 days after LVAD placement was not available. We did not have data about inotropic support, which may have impacted the hemodynamic measurements. We also did not routinely perform vasodilator testing preimplant in this cohort. The LVADs in this study were older generation devices, many of which were prone to failure and the PCWP was not normal in all patients at longer term follow-up. That being said, the degree of LV unloading and improvements in PAC appeared to be similar between the pulsatile and CF pumps. The methods used for PAC calculation have limitations as outlined in the “Materials and Methods” section; however, these measures have been demonstrated to correlate with clinical outcomes previously and are the best approximation of PAC which is easily calculated from routine hemodynamic variables. In addition, we did not account for the impact of the change in mean PA pressure on compliance. Given the size of the dataset were not able to analyze whether preoperative patterns of PAC or PVR were predictive of increased mortality or postop RV failure.
Pulmonary artery compliance improves rapidly after LVAD implantation as a result of lowering left heart filling pressures and a reduction in PVR. This suggests that more permanent changes in the pulmonary vascular bed may not responsible for the abnormal PAC observed in WHO group II PH.
1. Dupont M, Mullens W, Skouri HN, et al. Prognostic role of pulmonary arterial capacitance in advanced heart failure. Circ Heart Fail. 2012.5: 778–785.
2. Segers P, Brimioulle S, Stergiopulos N, et al. Pulmonary arterial compliance
in dogs and pigs: The three-element windkessel model revisited. Am J Physiol. 1999.2772 Pt 2H725–H731.
3. Lankhaar JW, Westerhof N, Faes TJ, et al. Pulmonary vascular resistance and compliance stay inversely related during treatment of pulmonary hypertension. Eur Heart J. 2008.29: 1688–1695.
4. Lankhaar JW, Westerhof N, Faes TJ, et al. Quantification of right ventricular afterload in patients with and without pulmonary hypertension. Am J Physiol Heart Circ Physiol. 2006.291: H1731–H1737.
5. Tedford RJ, Hassoun PM, Mathai SC, et al. Pulmonary capillary wedge pressure augments right ventricular pulsatile loading. Circulation. 2012.125: 289–297.
6. Mahapatra S, Nishimura RA, Oh JK, McGoon MD. The prognostic value of pulmonary vascular capacitance determined by Doppler echocardiography in patients with pulmonary arterial hypertension. J Am Soc Echocardiogr. 2006.19: 1045–1050.
7. Guazzi M, Naeije R, Arena R, et al. Echocardiography of right ventriculoarterial coupling combined with cardiopulmonary exercise testing to predict outcome in heart failure. Chest. 2015.148: 226–234.
8. Pellegrini P, Rossi A, Pasotti M, et al. Prognostic relevance of pulmonary arterial compliance
in patients with chronic heart failure. Chest. 2013.145: 1064–1070.
9. Syyed R, Reeves JT, Welsh D, Raeside D, Johnson MK, Peacock AJ. The relationship between the components of pulmonary artery pressure remains constant under all conditions in both health and disease. Chest. 2008.133: 633–639.
10. Chemla D, Castelain V, Humbert M, et al. New formula for predicting mean pulmonary artery pressure using systolic pulmonary artery pressure. Chest. 2004.126: 1313–1317.
11. Guazzi M, Borlaug BA. Pulmonary hypertension due to left heart disease. Circulation. 2012.126: 975–990.
12. John R, Pagani FD, Naka Y, et al. Post-cardiac transplant survival after support with a continuous-flow left ventricular assist device
: Impact of duration of left ventricular assist device
support and other variables. J Thorac Cardiovasc Surg. 2010.140: 174–181.
13. Vachiery JL, Adir Y, Barbera JA, et al. Pulmonary hypertension due to left heart diseases. J Am College Cardiol. 2013.6225 supplD100–D108.
14. Stergiopulos N, Meister JJ, Westerhof N. Evaluation of methods for estimation of total arterial compliance. Am J Physiol. 1995.2684 Pt 2H1540–H1548.
15. Stergiopulos N, Segers P, Westerhof N. Use of pulse pressure method for estimating total arterial compliance in vivo. Am J Physiol. 1999.2762 Pt 2H424–H428.
16. Saouti N, Westerhof N, Postmus PE, Vonk-Noordegraaf A. The arterial load in pulmonary hypertension. Eur Respir Rev. 2010.19: 197–203.
17. de Perrot M, McRae K, Shargall Y, et al. Early postoperative pulmonary vascular compliance predicts outcome after pulmonary endarterectomy for chronic thromboembolic pulmonary hypertension. Chest. 2011.140: 34–41.
18. Lazar JF, Swartz MF, Schiralli MP, et al. Survival after left ventricular assist device
with and without temporary right ventricular support. Ann Thorac Surg. 2013.96: 2155–2159.
19. Takeda K, Naka Y, Yang JA, et al. Outcome of unplanned right ventricular assist device support for severe right heart failure after implantable left ventricular assist device
insertion. J Heart Lung Transplant. 2014.33: 141–148.
20. Kaluski E, Cotter G, Leitman M, et al. Clinical and hemodynamic effects of bosentan dose optimization in symptomatic heart failure patients with severe systolic dysfunction, associated with secondary pulmonary hypertension–a multi-center randomized study. Cardiology. 2008.109: 273–280.
21. Califf RM, Adams KF, McKenna WJ, et al. A randomized controlled trial of epoprostenol therapy for severe congestive heart failure: The Flolan International Randomized Survival Trial (FIRST). Am Heart J. 1997.134: 44–54.
Keywords:Copyright © 2017 by the American Society for Artificial Internal Organs
pulmonary arterial compliance; left heart failure; left ventricular assist device