Author + information
- Published online October 30, 2017.
- Masaru Obokata, MD, PhD and
- Barry A. Borlaug, MD∗ ()
- ↵∗Address for correspondence:
Dr. Barry A. Borlaug, Division of Cardiovascular Diseases, Mayo Clinic and Foundation, 200 First Street Southwest, Rochester, Minnesota 55905.
- diffusing capacity of carbon monoxide
- filling pressures
- heart failure with preserved ejection fraction
- pulmonary hypertension
In 1970, Jeremy Swan and William Ganz introduced a new, flexible, balloon-tipped catheter that enabled easier access to pulmonary circulation (1). This was a valuable innovation because, prior to that time, catheterization of the right heart and pulmonary artery (PA) required larger, semirigid catheters that were cumbersome to use for all but the most skilled operators. The balloon-tipped Swan-Ganz catheter allowed for more routine bedside use, without the need for fluoroscopy, opening the door for direct assessment of cardiac hemodynamics in a much larger cohort of patients. Right heart catheterization has been used since that time to guide clinical decision making for patient care and to better understand the central circulation in a variety of cardiovascular diseases.
In the absence of lesions in the pulmonary venules, veins, left atrium, and mitral valve, the pulmonary capillary wedge pressure (PCWP) obtained by occluding the PA with these catheters provides an accurate measurement of left atrial and left ventricular end diastolic pressure (LVEDP) (2). PCWP is also the most direct measurement of the hydrostatic forces that govern fluid transport across the alveolar–pulmonary capillary interface. Thus, changes in PCWP play a key role in determining lung gas diffusion, and when elevated, they contribute to the development of dyspnea (3).
Elevated PCWP at rest or during exercise is used as a gold standard metric to definitively establish the diagnosis of heart failure with preserved ejection fraction (HFpEF) when the diagnosis is not apparent from clinical indicators (4,5). In addition to its role in diagnosis, elevated PCWP identifies HFpEF patients at increased risk of death (Online Ref. 1), and lowering PCWP and PA pressures reduces HF hospitalizations in patients with HFpEF (Online Ref. 2). As such, a number of interventions are being tested that target PCWP elevation in people with HFpEF (Online Refs. 3–5). Despite these compelling data supporting the importance of PCWP, there is a widespread perception that it is somehow inferior to direct measurement of LVEDP. In other words, the conventional wisdom is that using PCWP to estimate LV filling pressure is like getting information from looking at the tail rather than the whole dog. However, data to support this perception are sparse.
In this issue of JACC: Heart Failure, Mascherbauer et al. (6) present intriguing new data regarding the prognostic and mechanistic differences between LVEDP and PCWP in HFpEF. The authors conducted a prospective cohort to evaluate how these 2 measurements of LV filling pressures relate to outcome in HFpEF and how discordances between them may be related to pulmonary capillary function, assessed by the diffusing capacity of carbon monoxide (DLCO) (3). The authors found that PCWP was slightly higher than LVEDP, with a mean difference of 2 mm Hg. Over 2 years of follow-up, there were 51 primary endpoints, defined by a composite of hospitalization for HF and cardiac death (33.6%). Intriguingly, patients with LVEDP above the median value (>20 mm Hg) had similar outcome to those with values below the median (p = 0.26), arguing against the prognostic significance of LVEDP in HFpEF. Conversely, elevated PCWP (>20 mm Hg) was associated with increased risk of adverse outcome in HFpEF (p = 0.01), outperforming the more historically revered LVEDP to predict risk.
The PCWP remained an independent predictor of outcome after adjusting for LVEDP and the PCWP-LVEDP pressure difference (hazard ratio: 1.055; 95% confidential interval [CI]: 1.003 to 1.110). Multivariate regression analysis identified DLCO as the only parameter that was related to this pressure difference. Patients with low DLCO (≤45%) displayed higher PCWP-LVEDP pressure gradient and greater HFpEF severity, as shown by higher natriuretic peptide concentrations, PA pressures, and poorer exercise capacity, than patients with preserved DLCO. The authors concluded that PCWP measurements are more closely related to adverse outcome in HFpEF than LVEDP and that a high PCWP-LVEDP pressure gradient may be associated with greater pulmonary capillary and vascular remodeling in HFpEF (6).
The authors are to be commended for this important contribution (6). Despite the general assumption that LVEDP is the most robust parameter reflecting disease severity in HFpEF (because it is directly measured in the LV), PCWP but not LVEDP was found to be associated with event-free survival. Why might this be? To answer this question, we must first ask why PCWP may exceed LVEDP. Any lesions that lie between the pulmonary capillaries and left ventricle could influence the relationship between PCWP and LVEDP (2). It has recently become clear that many people with HFpEF display pathology in these domains, including structural and functional changes in the left atrium, the lung parenchyma, and the pulmonary vasculature (3,7) (Online Refs. 6–10). Thus, part of the explanation for the differential relationships between PCWP and LVEDP and outcome may be that these impairments importantly contribute to adverse outcome, above and beyond the deleterious effects of LV diastolic dysfunction alone.
The authors used DLCO as a marker of pulmonary capillary function and speculated that the increased PCWP-LVEDP pressure gradient observed was attributable to pulmonary capillary and/or post-capillary vascular remodeling, as DLCO was impaired in this group (6). To better frame the interpretation of their findings, it is worthwhile reviewing the determinants of pulmonary gas transfer. The DLCO is determined by 2 components, the volume of blood in the pulmonary capillaries that is available for gas transfer (VC) and the alveolar–capillary membrane conductance (DM). Each of these components is impaired and contributes to low DLCO in patients with HFpEF (3). Impairment in right ventricular (RV) ejection may depress VC, but RV systolic function was similar in patients with low and normal DLCO (6). This would suggest that any ostensible reduction in VC might be related more to pulmonary vascular remodeling, as speculated by the authors. However, markers of congestion, including right atrial pressure, pulmonary artery pressures, and N-terminal pro–B-type natriuretic peptide concentrations, were also elevated in patients with low DLCO (6). As such, one cannot exclude the possibility that greater congestion of the alveolar capillary interface might have caused low DM and thus depressed DLCO, independent of any structural remodeling (Online Ref. 11). Because the authors did not directly measure VC and DM in this study, we do not know how these components relate to hemodynamics or vascular structure. This represents an important avenue for further study, ideally with imaging and histopathologic correlations. Indeed, it is ironic that so little is known about the lungs, the organs most affected by left heart congestion in HFpEF, particularly at the microvascular level, making this an enormous unmet need in our understanding of this disorder (Online Refs. 9,10).
The PCWP is optimally determined visually at mid-A wave, during end-expiration, where intrapleural pressure is equal to atmospheric pressure. Instead, the authors used digitized mean PCWP averaged over multiple respiratory cycles (6). Reduction in intrathoracic pressure during inspiration lowers all pressures in the heart and lungs, so the values obtained in this way will (by definition) be lower. Although the respiratory swings in PCWP and LVEDP may offset one another, the measurements in this study were not obtained simultaneously, and this likely contributed to some of the variability.
There is another key potential contributor to a positive PCWP to LVEDP gradient that the authors did not explore, and that is due to abnormalities in the left atrium (7). The area under the PCWP pressure-time curve will be higher relative to LVEDP in patients with left atrial dysfunction, because these patients display prominent V waves due to increased left atrial stiffness, which would have driven up the PCWP to LVEDP gradient even in the absence of pulmonary vascular disease (7). It would have been interesting to examine whether patients with more severe left atrial dysfunction displayed higher PCWP to LVEDP gradient, particularly since left atrial dysfunction is known to contribute to increased burden of pulmonary vascular disease as well as increased risk of death in HFpEF (7) (Online Ref. 6).
More than 60% of patients enrolled in this study were in atrial fibrillation (6), and this limits the accuracy for identifying LVEDP because of tachycardia, cycle length variability, and the absence of an atrial pressure deflection on the LV tracing. This might have influenced the ability of estimated LVEDP to predict risk. Other predictors of outcome, including age, exercise capacity, renal function, and natriuretic peptide levels were not included in the model, but PCWP (and LVEDP) would be related to several of these covariates as well, so even that adjustment would have been problematic.
In summary, Mascherbauer et al. (6) have provided intriguing and provocative new data that challenge conventional thinking, leaving us to question which is the tail and which is the dog when it comes to LV filling pressures and outcomes in HFpEF. We know that rest or exercise elevation in PCWP is a universal finding in people suffering with HFpEF, and we know that this high PCWP causes dyspnea, interferes with activity levels, and impairs quality of life. Mascherbauer et al. (6) have shown us that elevated PCWP is also associated with increased risk of adverse outcomes, even when directly measured LVEDP is not. The next steps are to replicate these findings in other cohorts and then to better understand why PCWP is increased in some patients in excess of LVEDP, so that we can more effectively treat it and improve the health for millions of people worldwide burdened with this growing cardiac disorder.
↵∗ Editorials published in JACC: Heart Failure reflect the views of the authors and do not necessarily represent the views of JACC: Heart Failure or the American College of Cardiology.
Dr. Borlaug is supported by U.S. National Heart, Lung, and Blood Institute awards RO1 HL128526, R01 HL126638, U01 HL125205, and U10 HL110262; has received research support from Mast Therapeutics, Medtronic, GlaxoSmithKline, Teva, and Novartis; and sits on the advisory boards of and consults for Actelion, Amgen, AstraZeneca, Merck, and Cardiokinetix. Dr. Obokata has reported he has no relationships relevant to the contents of this paper to disclose.
- 2017 American College of Cardiology Foundation
- Luchsinger P.C.,
- Seipp H.W. Jr..,
- Patel D.J.
- Olson T.P.,
- Johnson B.D.,
- Borlaug B.A.
- Borlaug B.A.,
- Nishimura R.A.,
- Sorajja P.,
- Lam C.S.,
- Redfield M.M.
- Obokata M.,
- Kane G.C.,
- Reddy Y.N.,
- Olson T.P.,
- Melenovsky V.,
- Borlaug B.A.
- Mascherbauer J.,
- Zotter-Tufaro C.,
- Duca F.,
- et al.
- Melenovsky V.,
- Hwang S.J.,
- Redfield M.M.,
- Zakeri R.,
- Lin G.,
- Borlaug B.A.