Author + information
- Annelieke C.M.J. van Riel, MD,
- David M. Systrom, MD,
- Rudolf K.F. Oliveira, MD, PhD,
- Michael J. Landzberg, MD,
- Barbara J.M. Mulder, MD, PhD,
- Berto J. Bouma, MD, PhD,
- Bradley A. Maron, MD,
- Amil M. Shah, MD,
- Aaron B. Waxman, MD, PhD and
- Alexander R. Opotowsky, MD, MMSc∗ ()
- ↵∗Boston Children’s Hospital and Brigham and Women’s Hospital, 300 Longwood Avenue, Department of Cardiology, Boston, Massachusetts 02115
Fixed right ventricular outflow obstruction, such as pulmonary valve stenosis, causes a systolic pressure difference between the right ventricle (RV) and pulmonary artery (PA). This is an established confounder of PA pressure estimation using tricuspid regurgitation flow velocity and must be accounted for during routine echocardiography (1). It is expected that people without a resting RV-PA gradient do not develop such a gradient during exercise in the absence of specific types of congenital heart disease. We were led to question this assumption after observing several cases in which RV systolic pressure increased substantially more than PA systolic pressure during upright exercise performed with invasive hemodynamic monitoring.
To define the magnitude and frequency of RV-to-PA pressure differences during exercise, we studied patients referred for clinical invasive cardiopulmonary exercise testing between 2012 and 2015 who did not have a resting RV outflow tract (RVOT) pressure gradient during supine catheterization. Testing was performed as previously described (2). RVOT gradient was calculated as the difference between simultaneously transduced RV and PA systolic pressures. A peak exercise mean PA pressure (mPAP) >30 mm Hg was defined as abnormal. The Partners Human Research Committee approved this retrospective study and waived the requirement for informed consent (protocol #2011P000272).
The study included 294 patients without congenital heart disease (59.7 ± 15.5 years of age, 49.0% male). Body mass index averaged 29.0 ± 6.5 kg/m2. Hypertension (49%) and dyslipidemia (43%) were common; a small subset had previously undergone coronary artery bypass grafting (7%) or percutaneous coronary intervention (11%). Just under one-third of patients were prescribed a beta-adrenergic receptor antagonist. Average peak oxygen consumption (VO2) was 17.1 ± 7.8 mL/kg/min, or 74.8 ± 23.3% predicted.
Upright resting and peak RVOT gradients averaged 8.8 ± 5.5 mm Hg and 18.7 ± 11.2 mm Hg (Figure 1), respectively; the 90th percentile of peak RVOT gradient was 33 mm Hg. Peak RVOT gradient ≥33 mm Hg, present in 33 patients (11.2%), was associated with male sex (70 vs. 46% male; p = 0.01), younger age (43.6 ± 17.7 vs. 61.8 ± 13.9 years; p < 0.001), and higher peak VO2 (92.4 ± 33.8 vs. 72.6 ± 20.7% predicted; p = 0.002). These relationships were similar when RVOT gradient was analyzed as a continuous variable.
As expected, invasively measured systolic and mean PA pressures at peak exercise were strongly correlated (r2 = 0.88). Because of the variable RV-to-PA pressure difference during exercise, however, RV systolic pressure correlated less well with systolic PA pressure (r2 = 0.63). Invasively measured peak systolic PAP discriminated well between normal and abnormal mPAP response (area under the curve [AUC] = 0.97); however, peak RV systolic pressure was less predictive of abnormal mPAP response (AUC = 0.82). Peak RV systolic pressure was often elevated in patients with normal mPAP (e.g., 37% of patients with peak mPAP ≤30 mm Hg had RV systolic pressure >60 mm Hg).
Dynamic left ventricular outflow obstruction has been described extensively. Considered anatomically, the RVOT would seem more susceptible to dynamic obstruction given the presence of a circumferential, muscular, contractile infundibulum in contrast to the partly fibrous, noncontractile left ventricular outflow. Clinically significant dynamic RVOT obstruction has not been commonly reported, except in patients after cardiac surgery or after lung transplantation (3–5). Our observations suggest, however, that a substantial systolic RVOT pressure gradient frequently occurs during upright exercise in patients with structurally normal hearts. Development of such a pressure gradient was, surprisingly, associated with better exercise performance, which raises a question of whether this represents a normal, rather than pathologically abnormal, response to upright exercise. Either way, the presence of a variable pressure difference between the RV and PA threatens the accuracy of exercise echocardiography where PA systolic pressure is extrapolated from estimation of RV systolic pressure. The novelty of this finding likely reflects a lack of prior ascertainment. The physiologic mechanisms of upright exercise dynamic RVOT obstruction and the implications for clinical practice, including whether this phenomenon occurs during supine exercise, remain to be defined.
Please note: Dr. van Riel was supported by a fellowship grant provided by Netherlands Heart Institute; and a travel grant provided by ZON-MW. Drs. Opotowsky and Landzberg are supported by the Dunlevie Family Fund. Dr. Oliveira received funds from the São Paulo Research Foundation (FAPESP), grant #2014/12212-5 and from the Brazilian National Council for Scientific and Technological Development (CNPq), grant #232643/2014-8. Drs. Waxman and Systrom are supported by grant funding from the National Institutes of Health (U01HL125215). Dr. Shah is supported by the National Institutes of Health (K08HL116792) and the American Heart Association (14CRP20380422). Dr. Maron is supported by the NIH (1K08HL11207-01A1), American Heart Association (AHA 15GRNT25080016), Pulmonary Hypertension Association, and the Cardiovascular Medical Research and Education Fund; and receives research support from Novartis, Gilead, and Actelion; and consulting fees from Myocardia. Dr. Opotowsky has received research support from Actelion and Roche Diagnostics. Dr. Maron has received research support from Gilead Sciences to research pulmonary hypertension. Dr. Landzberg has received research support from Actelion. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose. The authors thank Abbey Karin, Julie Tracy, Jeff Voner, and Charlie Lee for their meticulous work performing these invasive clinical exercise tests; and Susana Mak, Ryan Tedford, James E. Lock, and John Granton for their insights into the physiology described in this paper.
- American College of Cardiology Foundation