Pulmonary Vascular Distensibility and Early Pulmonary Vascular Remodeling in Pulmonary Hypertension

[

Original Research

]

Pulmonary Vascular Distensibility and Early Pulmonary Vascular Remodeling in Pulmonary Hypertension Inderjit Singh, MD; Rudolf K. F. Oliveira, MD, PhD; Robert Naeije, MD, PhD; Farbod N. Rahaghi, MD, PhD; William M. Oldham, MD, PhD; David M. Systrom, MD; and Aaron B. Waxman, MD, PhD

Exercise stress testing of the pulmonary circulation may uncover decreased pulmonary vascular (PV) distensibility as a cause of impaired aerobic exercise capacity and right ventricular (RV)-pulmonary arterial (PA) uncoupling. As such, it may help in the differential diagnosis of unexplained dyspnea, including pulmonary hypertension (PH) and/ or heart failure with preserved ejection fraction (HFpEF). We investigated rest and exercise invasive pulmonary hemodynamics, ventilation, and gas exchange in patients with unexplained dyspnea, including 44 patients with HFpEF (of whom 20 had a normal pulmonary vascular resistance [PVR] during exercise [ie, passive HFpEF] and 24 had a higher than normal exercise PVR), 22 patients with exercise PH, 19 patients with pulmonary arterial hypertension (PAH), and 24 age- and sex-matched normal control subjects.

BACKGROUND:

A PV distensibility coefficient a (%/mm Hg) was determined from multipoint PV pressure-flow plots. RV-PA coupling was quantified from the analysis of RV pressure curves to determine ratios of end-systolic to arterial elastances (Ees/Ea). Aerobic exercise capacity was estimated by peak oxygen consumption.

METHODS:

The a coefficient decreased from 1.35
0.58%/mm Hg in control subjects and 1.1
0.48%/mm Hg in patients with passive HFpEF to 0.62
0.32%/mm Hg in exercise PH, 0.54
0.27%/mm Hg in HFpEF with high exercise PVR, and 0.18
0.16%/mm Hg in PAH. On multivariate analysis, PV distensibility was associated with decreased Ees/Ea and maximal volume of oxygen consumed. RESULTS:

PV distensibility is an early and sensitive hemodynamic marker of PV disease that is associated with RV-PA uncoupling and decreased aerobic exercise capacity.

CONCLUSIONS:

CHEST 2019;

-(-):---

KEY WORDS: exercise capacity; heart failure; pulmonary arterial hypertension; pulmonary vascular distensibility; RV-PA coupling

ABBREVIATIONS: AUC = area under the curve; CO = cardiac output; Ea = pulmonary arterial elastance; Ees = end-systolic elastance; ePH = exercise PH; HFpEF = heart failure with preserved ejection fraction; HFpEFþPVR = HFpEF with associated abnormal increases in PVR during exercise; HR = heart rate; iCPET = invasive cardiopulmonary exercise test; mPAP = mean pulmonary artery pressure; PA = pulmonary artery; PAC = pulmonary artery compliance; PAH = pulmonary arterial hypertension; PAWP = pulmonary artery wedge pressure; PH = pulmonary hypertension; PV = pulmonary vascular; PVR =

pulmonary vascular resistance; RHC = right heart catheterization; RV = right ventricular; RV-PA = right ventricular-pulmonary artery; SV = stroke volume; TPR = total pulmonary resistance; VO2 = oxygen consumption; WU = Wood units AFFILIATIONS: From the Division of Pulmonary, Critical Care, and Sleep Medicine (Dr Singh), Department of Medicine, Yale New Haven Hospital and Yale School of Medicine, New Haven, CT; the Division of Respiratory Medicine (Dr Oliveira), Federal University of São

1

chestjournal.org

FLA 5.6.0 DTD
CHEST2349_proof
12 July 2019
7:20 pm
EO: CHEST-19-0350

Pulmonary hypertension (PH) is defined by an increase in mean pulmonary artery pressure (mPAP) $ 25 mm Hg, whereas the diagnosis of pulmonary vascular (PV) disease reflects the concomitant increase in pulmonary vascular resistance (PVR) of > 3 Wood units (WU).1 Therefore, the diagnosis of PH with pathologic pulmonary arterial remodeling relies on determination of PVR. However, there is considerable uncertainty in the PVR cutoffs because the upper limit of normal increases with aging from 1.5 WU in young adults to a maximum of 2 WU in individuals > 70 years of age.2

(mPAP/CO) at rest, and a is the rate of increase of resistive vessel diameter per millimeter of mercury of transmural vascular pressure.

The PVR equation assumes a linear relationship between the variables and a zero crossing of PV pressure difference-flow relationships. Accordingly, mPAP is exclusively determined by a dynamic interplay between PVR, cardiac output (CO), and pulmonary artery wedge pressure (PAWP):

The improved PVR equation was used to demonstrate impaired resistive vessel distensibility in early or latent PV disease7 and in heart failure with reduced ejection fraction or heart failure with preserved ejection fraction (HFpEF).8 In the latter study, the a coefficient was positively correlated to right ventricular (RV) ejection fraction and independently predicted peak oxygen uptake (VO2) and cardiovascular mortality. However, the impact of impaired PV distensibility on right ventricular-pulmonary artery (RV-PA) coupling and its subsequent impact on aerobic exercise capacity in the setting of unexplained dyspnea remains incompletely characterized.

mPAP ¼ PVR  CO þ PAWP

(1)

This reasoning overlooks the natural distensibility of pulmonary vessels as a cause of nonlinear mPAP-CO relationships and the decrease in PVR seen with increasing flow.3 Linehan et al4 had addressed this shortcoming with a simple distensibility model of the pulmonary circulation and improved the PVR equation by incorporating a resistive vessel distensibility coefficient a: 1  ð1 þ aPAWPÞ5 þ 5aR0 CO 5  1 ; mPAP ¼

a

(2)

where R0 is the total pulmonary resistance (TPR)

Methods Study Population and Design We retrospectively identified patients from the Brigham and Women’s Hospital Dyspnea Clinic (Boston, MA) between March 2011 and

Paulo - UNIFESP, São Paulo, Brazil; the Department of Pathophysiology (Dr Naeije), Faculty of Medicine, Erasme Campus, Université Libre de Bruxelles, Brussels, Belgium; and the Division of Pulmonary and Critical Care (Drs Rahaghi, Oldham, Systrom, and Waxman), Department of Medicine, Brigham and Women’s Hospital and Harvard Medical School, Boston, MA. FUNDING/SUPPORT: The authors have reported to CHEST that no funding was received for this study. CORRESPONDENCE TO: Aaron B. Waxman, MD, PhD, Pulmonary and Critical Care Medicine, Department of Medicine, Brigham and Women’s Hospital, Clinics 3, 75 Francis St, Boston, MA 02115; e-mail: [email protected] Copyright Ó 2019 American College of Chest Physicians. Published by Elsevier Inc. All rights reserved. DOI: https://doi.org/10.1016/j.chest.2019.04.111

2 Original Research

An interesting application of this equation is that a can be calculated from the best fit of mPAP, PAWP, and CO measurements obtained during exercise to increase CO.5 Invasive and noninvasive studies have shown that calculated a is normally between 1 and 2%/mm Hg, higher in young healthy women, and decreased with aging or chronic hypoxic exposure.3,6

In this study, we tested the hypothesis that exercise stress testing of the pulmonary circulation could unmask impaired PV distensibility in patients with unexplained dyspnea or with HFpEF, and in both circumstances, the resulting impaired PV distensibility would decrease peak exercise aerobic capacity through RV-PA uncoupling.

September 2017 who underwent resting supine right heart catheterization (RHC) followed by symptom-limited upright invasive cardiopulmonary exercise testing (iCPET) as part of their clinically indicated evaluation for unexplained exercise intolerance.9 The study protocol was approved by Partners Healthcare Human Research Committee (No. 2011P000272). Included patients had normal left ventricular ejection fraction (ie, > 50%) with no significant valvular abnormalities by resting echocardiography. According to the maximum exercise hemodynamics, we identified five groups: (1) passive HFpEF defined by peak PAWP > 17 mm Hg for ages > 50 years or peak PAWP > 19 mm Hg for ages # 50 years; (2) HFpEF with associated abnormal increases in PVR during exercise (HFpEFþPVR), defined additionally by a peak PVR > 1.34 WU for ages # 50 years or a peak PVR > 2.10 WU for ages > 50 years; (3) exercise PH (ePH), defined by normal supine resting RHC hemodynamics, but with peak mPAP > 30 mm Hg and peak PVR > 1.34 WU for patients # 50 years of age or a peak mPAP > 33 mm Hg and peak PVR > 2.10 WU for patients >50 years of age9; (4) pulmonary arterial hypertension (PAH), defined by resting supine RHC of $ 25 mm Hg and PAWP # 15 mm Hg along with PVR > 3 WU in

[

-#- CHEST - 2019

FLA 5.6.0 DTD
CHEST2349_proof
12 July 2019
7:20 pm
EO: CHEST-19-0350

]

the absence of identifiable causal cardiac or pulmonary disease, including pulmonary thromboembolism; and (5) control subjects, who exhibited a normal physiologic limit to exercise defined by a peak VO2 and peak CO of $ 80% predicted. Control subjects were matched for age and sex to the other groups. Exclusion criteria included the following: (1) relevant lung disease defined as FEV1/FVC < 70% predicted or a radiologic diagnosis of lung fibrosis; (2) acute coronary syndrome defined by ST-segment elevation myocardial infarction, non-ST-segment elevation myocardial infarction, or unstable angina during iCPET; (3) submaximal iCPET defined by peak respiratory exchange ratio < 1.05 and peak heart rate (HR) < 85% predicted along with a peak mixed venous partial pressure of oxygen > 27 mm Hg10; (4) incomplete exercise hemodynamics; and (5) perfusion scan suggestive of pulmonary thromboembolism. Measurements Our RHC and iCPET protocol have been described previously.9,11,12 iCPET techniques and conventional hemodynamic measurements are detailed in e-Appendix 1, e-Figures 1 and 2, and e-Tables 1-5. PV Distensibility: PV distensibility, a at constant hematocrit, was determined by nonlinear squares fitting (ie, least squares fitting) of equation 2. In our study, we measured PAWP, mPAP, and CO at rest and at peak exercise. We integrated the resting and peak exercise values of PAWP and CO and R0 to determine the dynamic a value that best predicted observed mPAP values during maximum incremental exercise.

Results Demographic and Clinical Characteristics

Consecutive iCPET reports were analyzed to identify 44 patients with HFpEF, 22 patients with ePH, 19 patients with PAH, and 24 control subjects based on the aforementioned inclusion and exclusion criteria (e-Fig 2). Therefore, the study sample constituted 109 subjects. Among the patients with HFpEF, 20 patients had a normal PVR during exercise (passive HFpEF) and 24 patients had HFpEFþPVR. The diagnosis of patients with ePH and PAH is described in e-Table 1.

RV-PA Coupling: RV-PA coupling assessment was performed only in subjects who had RV pressure waveform tracings available during iCPET. The end-systolic elastance (Ees) was determined at rest and at peak exercise, using the single beat method13 as validated for the right ventricle,14 but with RV end-systolic pressure estimated by mPAP. Arterial elastance (Ea) was determined by mPAP divided by stroke volume (SV) (e-Fig 1).15 The exertional contractile reserve was determined by the difference between rest and peak exercise Ees. RV-PA coupling was subsequently determined by the ratio between Ees and Ea (Ees/Ea).16 Statistical Analysis Unless otherwise stated, values are presented as mean
SD. Comparisons of baseline characteristics, resting hemodynamics, iCPET parameters, a values, and RV-PA coupling between the five study groups were performed using one-way analysis of variance with Bonferroni post hoc correction. Receiver operating characteristic curve analysis was used to assess the ability of a to distinguish control subjects from ePH. The relationships of peak exercise RV-PA coupling and peak VO2 (% predicted) with PV distensibility were examined using linear regression. To identify if a predicts peak RV-PA coupling and peak VO2 (% predicted), Pearson correlation was performed. Univariable analysis was performed to determine the predictors of peak RV-PA coupling and peak VO2 (% predicted) for HFpEF, ePH, and PAH groups. Noncolinear variables (ie, Pearson correlation r < 0.6) with a significant P value (P < .05) on univariable analysis were incorporated into multivariable models to identify independent predictors of peak VO2 (% predicted). A probability value of < 0.05 was considered significant. Statistical analyses were performed using GraphPad Prism 7 (GraphPad Software, LLC) and SAS 9.4 (SAS Institute Inc).

mPAP, TPR, and PVR were greatest in patients with PAH. Patients with HFpEFþPVR showed higher resting mPAP compared with patients with ePH, patients with passive HFpEF, and control subjects. Patients with HFpEFþPVR also demonstrated the highest resting PAWP among the five groups. The resting pulmonary artery compliance (PAC) was reduced in patients with PAH compared with control subjects and patients with passive HFpEF but was not different than patients with HFpEFþPVR and ePH. The resting RHC data are summarized in Table 1. Peak Exercise Hemodynamics

There was no statistical difference in age, sex, and resting mean hemoglobin concentration between the groups. Patients with HFpEFþPVR had higher BMI than control subjects and patients with PAH. The baseline characteristics and comorbidities are summarized in Table 1. Resting RHC

There was no difference between the resting values of HR, CO, cardiac index, and SV index among the five groups. The resting mean right atrial pressure was greater in patients with HFpEFþPVR compared with control subjects and patients with ePH. The resting

Both patients with PAH and patients with HFpEFþPVR had a reduced peak VO2 (% predicted) than patients with passive HFpEF and control subjects. Patients with HFpEFþPVR and patients with passive HFpEF demonstrated similarly decreased maximum HR (% predicted) than control subjects. Patients with PAH demonstrated a greater degree of arterial oxygen desaturation at peak exercise. Patients with PAH also demonstrated the greatest degree of ventilatory inefficiency as indicated by a markedly abnormal ventilatory efficiency slope. The maximum arterial-mixed venous oxygen content difference was

3

chestjournal.org

FLA 5.6.0 DTD
CHEST2349_proof
12 July 2019
7:20 pm
EO: CHEST-19-0350

TABLE 1

] Baseline Characteristics and Resting Hemodynamic Data Control Subjects (n ¼ 24)

Variable

Passive HFpEF Group (n ¼ 20)

ePH Group (n ¼ 22)

HFpEFþPVR Group (n ¼ 24)

PAH Group (n ¼ 19)

Characteristics Age, y

65
9

67
9

68
9

Female

14 (58)

7 (35)

9 (41)

28
5

32
8

30
6

34
8

14.6
1.8

13.2
1.5

13.6
1.3

13.6
1.5

2

BMI, kg/m

Hemoglobin, g/dL

71
9

63
12

15 (63)

6 (32) 27
6

a, b

14.7
2.2

Comorbidities Systemic hypertension

8 (33)

6 (30)

10 (45)

11 (46)

7 (37)

Hyperlipidemia

7 (29)

7 (35)

8 (36)

10 (42)

8 (42)

Diabetes

0 (0)

3 (15)

6 (27)a

3 (13)

4 (21)a

Coronary artery disease

2 (8)

2 (10)

5 (23)

5 (21)

2 (11)

Beta adrenergic receptor antagonist

4 (17)

9 (45)

8 (37)

13 (54)

7 (37)

Calcium channel antagonist

3 (13)

4 (20)

4 (18)

11 (46)a,

Medications

ACE inhibitor or ARB Diuretics

b

1 (5)

10 (42)

4 (20)

6 (27)

8 (33)

3 (16)

8 (33)

11 (55)

5 (23)

13 (54)

10 (53)

100
14

85
19

Pulmonary function FEV1, % predicted FVC, % predicted

100
15

87
17

FEV1/FVC, % predicted

100
12

a

79
21a

70
20a

a

a

73
24a 74
18a,c

82
21

70
18

97
11

95
11

99
12

97
14

97
1

95
3a

93
5a,d,e 79
13

Resting upright hemodynamics SaO2, %

98
1

97
1

Heart rate, beats/min

72
10

71
11

75
14

74
15

3
3

5
4

4
4

8
4

RAP, mm Hg

34.7
7.3

SVI, mL/m2 CO, L/min

a

39.3
15.3

31.3
9.6

5
5

a, d

38.4
15.6

37.3
9.3

5.1
1.4

5.5
2.1

4.6
1.4

5.8
2.4

5.2
1.6

2.5
0.6

2.7
1.0

2.9
0.8

2.7
0.9

2.3
0.6

mPAP, mm Hg

13
3

18
4a

18
4a

PAWP, mm Hg

5
3

9
5a

TPG, mm Hg

7.8
2.3

8.7
2.9

10.2
3.2

11.4
4.5

26.1
8.6a,c,d,e

PVR, Wood units

1.6
0.5

1.8
1.0

2.4
0.8

2.1
0.9

5.2
2.9a,c,d,e

TPR, Wood units

2.8
1.2

3.7
1.8

4.2
1.2

5.0
1.9a

7.3
3.4a,c,d,e

CI, L/min/m

2

PAC, mL/mm Hg

7.3
3.1

6.1
3.0

26
8a,d,e

8
4

34
10a,c,d,e

15
6a,b,d,e

*

4.1
1.5

8
4

a,e

3.9
1.7

2.6
1.7a,e

Data presented as No. (%) or mean
SD. ACE ¼ angiotensin converting enzyme; ARB ¼ angiotensin receptor blocker; CI ¼ cardiac index; CO ¼ cardiac output; ePH ¼ exercise pulmonary hypertension; HFpEF ¼ heart failure with preserved ejection fraction; HFpEFþPVR ¼ HFpEF with associated abnormal increases in PVR during exercise; mPAP ¼ mean pulmonary artery pressure; PAC ¼ pulmonary artery compliance; PAH ¼ resting pulmonary arterial hypertension; PAWP ¼ pulmonary artery wedge pressure; PVR ¼ pulmonary vascular resistance; RAP ¼ right atrial pressure; SaO2 ¼ arterial oxygen saturation; SVI ¼ stroke volume index; TPG ¼ transpulmonary gradient; TPR ¼ total pulmonary resistance. a P < .05 compared with control subjects. b P < .05 compared with patients with PAH. c P < .05 compared with patients with HFpEFþPVR. d P < .05 compared with patients with ePH. e P < .05 compared with patients with HFpEF.

decreased in all study groups compared with control subjects, but there was no difference in systemic oxygen extraction between the groups.

4 Original Research

By study design, peak PAWP was higher in the HFpEF groups. Similarly, peak PVR and TPR were greatest in the PAH group compared with other groups. All

[

-#- CHEST - 2019

FLA 5.6.0 DTD
CHEST2349_proof
12 July 2019
7:20 pm
EO: CHEST-19-0350

]

progressive decline in PV distensibility from patients without precapillary PV disease (control subjects and patients with HFpEF) to patients with precapillary PV disease (patients with ePH, patients with HFpEFþPVR, and patients with PAH) at maximum exercise. Overall, patients with PAH demonstrated the lowest PV distensibility at maximum exercise as demonstrated in Figure 1.

studied groups, including the ePH group, had mean peak TPR that was > 3 mm Hg.min.L1. The peak PAC was decreased in patients with PAH compared with control subjects, patients with passive HFpEF, and patients with ePH but was not different in relation to patients with HFpEFþPVR. The maximum iCPET and peak exercise hemodynamic data are summarized in Table 2.

Using dynamic PV distensibility to distinguish patients with ePH from control subjects, the area under the curve (AUC) was 0.91
0.04, with a sensitivity of 86% and a specificity of 96% at an

Dynamic PV Distensibility

Among control subjects, the mean dynamic PV distensibility was 1.35
0.58%/mm Hg. There was a TABLE 2

] Maximum Cardiopulmonary Exercise Data and Peak Exercise Hemodynamics

Variable

Control Subjects (n ¼ 24)

Passive HFpEF Group (n ¼ 20)

ePH Group (n ¼ 22)

HFpEFþPVR Group (n ¼ 24)

102
16

83
23a

73
17a

62
18a,b

PAH Group (n ¼ 19)

Maximum CPET data VO2max, % predicted

61
25a,

22.7
6.9

14.5
4.7a

13.7
3.5a

10.7
2.7a

Work, W

154
60

82
33

90
36

51
30

Heart rate, beats/ min

139
19

121
26

126
19

108
19a

122
27

VO2, mL/kg/min

a

a

a

b

13.2
5.7a 57
44a

Heart rate, % predicted

90
10

79
15a

82
11

72
14a

77
15

SaO2, %

98
1

97
2

95
3a,b

93
5a,b

87
8a,b,c,d

102
14

98
24

95
17

80
18a,b

72
24a,b,d

33
7

37
10

35
7

40
11

60
28a,b,c,d

CO, % predicted VE/VCO2 slope

20.0
2.6

CaO2, mL/dL

269
86

DO2, mL/min

a

17.9
2.1

18.1
2.1

17.7
2.6

a

a

a

183
47

a

174
43

Ca-vO2, mL/100 mL

14.2
2.2

12.0
1.6

11.9
2.2

Oxygen extraction, Ca-vO2/CaO2

0.71
0.1

0.67
0.1

0.65
0.1

a

156
41

11.5
2.1

a

0.64
0.1

17.8
3.0 145
76a 11.8
2.5a 0.67
0.1

Peak exercise hemodynamics 5
3

RAP, mm Hg

12
6a,d

8.0
3.5a

5.1
1.0

4.9
1.1

4.2
0.9

4.4
1.8a

46.6
7.1

46.3
9.4

40.2
8.3

40.2
9.9

mPAP, mm Hg

25
5

38
4

38
7

PAWP, mm Hg

11
4

25
4

12
3

CI, L/min/m SVI, mL/m2

6.5
1.3

11
6a

8.9
2.5a

2

10.2
2.3

14
6a,d

9.6
2.1a

CO, L/min

13.3
3.4

7
4

a

a a,d,e

a

a

PVR, Wood units

1.2
0.4

1.3
0.5

2.7
0.6

a,b

TPR, Wood units

2.0
0.7

4.0
0.8

3.8
0.9

a

PAC, mL/mm Hg

3.2
1.2

2.7
1.0

2.5
1.1

a

a,b

36.0
10.4a

50
7

a,b,d

56
15a,b,d

26
4

a,d,e

13
5

2.8
0.6

a,b

6.4
3.6a,b,d,e

5.8
1.4

a,b,d

8.3
4.5a,b,d,e

1.9
0.7a,b

1.2
0.5a,b,d

Data are presented mean
SD. CaO2 ¼ arterial oxygen content; Ca-vO2 ¼ arterial-mixed venous oxygen content difference; CPET ¼ cardiopulmonary exercise test; DO2 ¼ oxygen delivery; VE/VCO2 ¼ ventilatory efficiency; VO2 ¼ oxygen consumption; VO2max ¼ maximal volume of oxygen consumed. See Table 1 legend for expansion of other abbreviations. a P < .05 compared with control subjects. b P < .05 compared with patients with HFpEF. c P < .05 compared with patients with HFpEFþPVR. d P < .05 compared with patients with ePH. e P < .05 compared with patients with PAH.

5

chestjournal.org

FLA 5.6.0 DTD
CHEST2349_proof
12 July 2019
7:20 pm
EO: CHEST-19-0350

Pulmonary Vascular Distensibility (% per mm Hg)

4.0

3.0 2.5 2.0 1.5 1.0 0.5 0.0

Distensibility (% per mm Hg)

Control

HFpEF

ePH

HFpEF+PVR

PAH

Controls (n = 24)

HFpEF (n = 25)

ePH (n = 27)

HFpEF+PVR (n = 24)

PAH (n = 20)

1.35 ± 0.58

1.10 ± 0.48

0.62 ± 0.32*+

0.54 ± 0.27*+ 0.18 ± 0.16*+^#

Figure 1 – Dynamic pulmonary vascular distensibility in exercise intolerant phenotypes. Participants with heart failure with preserved ejection fraction (HFpEF) with associated abnormal increases in pulmonary vascular resistance during exercise (*þ) and participants with exercise pulmonary hypertension (*þ) both demonstrated reduced distensibility compared with control subjects and participants with passive HFpEF. Pulmonary arterial hypertension group (*þ #) demonstrated greatest impairment in distensibility compared with the other groups (P < .05 for all group comparisons). Data presented as mean
SD. ePH ¼ exercise pulmonary hypertension; HFpEF ¼ heart failure with preserved ejection fraction; HFpEFþPVR ¼ HFpEF with associated abnormal increases in pulmonary vascular resistance; PAH ¼ pulmonary arterial hypertension. ˇ

print & web 4C=FPO

Control HFpEF ePH HFpEF+PVR PAH

3.5

optimal cutoff point of 0.79%/mm Hg. Other hemodynamic parameters including resting PAC (AUC, 0.84
0.06), resting PVR (AUC, 0.82
0.07), delta PAC (AUC, 0.76
0.07), and peak PAC (AUC, 0.74
0.07) provided less discrimination between control subjects and patients with ePH, as depicted in Figure 2.

The univariable analysis for predicting RV-PA coupling and peak VO2 (% predicted) is presented in e-Tables 3 and 4. On multivariate analysis, PV distensibility emerged as an independent predictor of peak RV-PA coupling (e-Table 5), whereas peak cardiac index, PV distensibility, and maximum arterial-mixed venous oxygen content difference emerged as independent predictors of peak VO2 (% predicted) (Table 3).

Independent Predictors of RV-PA Coupling and Peak VO2

The determination of peak VO2 predictors was performed for the entire study sample excluding control subjects (85/109). RV-PA coupling assessment was performed only in subjects who had RV pressure waveform tracings available during incremental exercise testing (38/85). This cohort constituted 8 patients with passive HFpEF, 13 patients with HFpEFþPVR, 8 patients with ePH, 9 patients with PAH, and 24 control subjects (e-Table 2).

6 Original Research

Discussion The results show that decreased PV distensibility is associated with RV-PA uncoupling and decreased aerobic exercise capacity in patients with ePH and patients with HFpEF with an excessive exercise-induced increase in PVR. Therefore, exercise testing of the pulmonary circulation and derived PV distensibility estimations may help earlier diagnosis of PV disease in patients with unexplained dyspnea and/or HFpEF.

[

-#- CHEST - 2019

FLA 5.6.0 DTD
CHEST2349_proof
12 July 2019
7:20 pm
EO: CHEST-19-0350

]

Controls vs. ePH

Sensitivity%

100

Distensibility α (AUC 0.95, CI 0.9 - 1.0) Rest compliance (AUC 0.84, CI 0.7 - 1.0)

50

Rest PVR (AUC 0.82, CI 0.7 - 1.0)

print & web 4C=FPO

Delta compliance (AUC 0.76, CI 0.6 - 0.9)

0

Peak compliance (AUC 0.74, CI 0.6 - 0.9)

0

100 50 100% - Specificity%

150

Figure 2 – Receiver operating characteristic curves for pulmonary vascular distensibility index (a) for classification of participants with exercise pulmonary hypertension (ePH) compared with control subjects. Distensibility demonstrated excellent discrimination between control subjects and those with ePH compared with rest and delta compliance and rest pulmonary vascular resistance.

In this study, patients with HFpEF had PVR that was on average below the pathologic cutoff value of 3 WU. However, patients with HFpEFþPVR, as defined from previous studies in control subjects from this laboratory,7 had a significant decrease in a, suggesting early PV disease. These data are in keeping with reported direct correlations between a and peak VO2, RV ejection fraction, and survival in a mixed cohort of patients with heart failure.8 In the current study, functional relevance of this finding was additionally supported by associated decreased RV-PA coupling and lower peak VO2. PAC was reduced in patients with ePH, HFpEFþPVR, and PAH at rest. At peak exercise, PAC was reduced only in patients with HFpEFþPVR and PAH. Furthermore, PAC did not emerge as an independent predictor of peak VO2 at multivariable analysis. This apparent discrepancy between PAC and the distensibility factor a is explained by the fact that PAC calculated as SV divided by pulse pressure is a global parameter of PA distensibility in pulsatile flow TABLE 3

conditions,17 whereas a refers specifically to resistive vessel distensibility in steady flow conditions at the periphery of PV tree, which is the site of remodeling of pulmonary vessels.18,19 Therefore, from a pathophysiologic standpoint, the distensibility factor a more accurately identifies PV disease than PAC. Our results were obtained with exercise stress testing in the upright position. This could have exaggerated exercise-induced decrease in PVR because of initial high resting value because of low venous return, CO, and PAWP with unchanged mPAP.2 With multiple measurements in a fully recruited pulmonary circulation during exercise, the impact of this high resting PVR becomes progressively negligible; therefore, mPAP-CO plots are body positionindependent.20 Therefore, given that PV pressures and flow are measured at multiple levels of exercise, it is less likely that body position affected distensibility calculations.21 The distensibility factor a in the control subjects was 1.35
0.58%/mm Hg, which is similar to 1.39%/ mm Hg and 1.40%/mm Hg reported in invasive studies on patients in the same age range without PH.7,8 Higher values around 2%/mm Hg have been reported in young adults and in premenopausal women.3,6 Our results also are in keeping with previously reported a values in the range of 0.25 to 0.4%/mm Hg in PAH, and around 0.8 to 0.9%/mm Hg in heart failure with either reduced or preserved ejection fraction.8 Distensibility factor calculations from multipoint PVpressure relationships appear to be consistent and reproducible across studies. During exercise, PV distensibility decreases with increasing pulmonary blood flow.22 Hence, by integrating rest to exercise changes of the different variables, a can be considered a measure of distensibility reserve (ie, the ability of the pulmonary circulation to dynamically accommodate flow during exercise). The measurement of a implicates the distensibility of the small resistive vessels (vessels that

] Multivariable Model for Predicting Peak VO2 (% predicted) in All Patients With HFpEF ePH, and PAH (n ¼ 85)

Variable Peak cardiac index, L/min/m

2

ß Coefficient

P Value

13.51

< .0001

PV distensibility, %/mm Hg

3.69

.04

Ca-vO2, corrected for Hgb

8.34

< .0001

95% CI 10.00–17.02 0.16–7.23 4.78–11.91

Hgb ¼ hemoglobin; PV ¼ pulmonary vascular. See Table 1 and 2 legends for expansion of other abbreviations.

7

chestjournal.org

FLA 5.6.0 DTD
CHEST2349_proof
12 July 2019
7:20 pm
EO: CHEST-19-0350

are primarily implicated in PV disease). This is distinct from PAC where approximately 15% to 20% of the PAC is stored in the main pulmonary artery (PA) and proximal left and right PAs.23 Hence, the site of measurement between PAC and a are anatomically discordant. Similar to PAC, Ea is a global PV stiffness parameter. However, unlike PA compliance, Ea is determined from RV pressure waveform measurement at the point of maximal elastance to calculate RV-PA coupling. Therefore, although Ea and PAC reflect the same physiologic phenomenon, the absolute values differ. A lot of directly measured or calculated invasive hemodynamic and iCPET variables, such as PAP, PAC, RV ejection fraction, peak VO2, and ventilatory efficiency, have been shown to be of prognostic relevance in heart failure and/or PAH.16,18,24,25 The distensibility coefficient a has been added to the list for patients with heart failure.8 The present study combining iCPET with invasive hemodynamic measurements, adds further insight into the physiologic interrelationships, and demonstrates the functional relevance of resistive vessel distensibility in the setting of unexplained dyspnea and HFpEF. At our institution, the iCPET program is a collaborative effort between cardiologists, pulmonologists, and exercise physiologists. A supervising physician, a physician assistant, and two exercise physiologists monitor the patient’s clinical status throughout the duration of the test. We only perform iCPET in patients who are able to tolerate exercise without the need of oxygen supplementation. The safety profile and contraindications to iCPET are similar to those reported for noninvasive testing12,26; however, risks associated with radial and pulmonary arterial catheter placement are also to be considered prior to patient enrollment. We have previously shown that a multidisciplinary approach using iCPET to diagnose potential causes of unexplained dyspnea dramatically reduces the time to diagnosis and therefore cost compared with traditional treatment and testing methods.27 The techniques we use to obtain invasive hemodynamic measurements are detailed in eAppendix 1. There are several limitations to the present study. First, the subjects were derived from iCPET evaluation at a tertiary referral center for unexplained exertional

8 Original Research

dyspnea; therefore, control subjects may not be representative of a completely healthy population. However, control subjects were selected based on a preserved peak exercise capacity defined by a normal cardiac limit to exercise (peak VO2 and peak CO $ 80% predicted). Therefore, they represent a studied population with a normal physiologic response to exercise and reflect symptomatic normal subjects. Second, exercise hemodynamics in healthy subjects vary considerably according to age.2 Hence, for the ePH, passive HFpEF, and HFpEFþPVR definitions, we used age-related mPAP, PAWP, and PVR thresholds.9 Using age-specific criteria for upright exercise pulmonary hemodynamics, we likely decreased the number of false-positive/false-negative diagnoses of ePH, passive HFpEF, and HFpEFþPVR as a function of normal aging, as could be derived from the current definition of ePH based on a mPAPCO slope and total PVR > 3 WU.21,28 Third, the single beat method rests on several assumptions which still require validation against criterion standard multiple beat measurements. Fourth, the current calculation of PVR does not account for changes in blood viscosity. Some degree of hemoconcentration does occur with exercise, and because of the exponential relationship between PVR and hematocrit, the impact of hematocrit on PVR is greater when there is a greater degree of hemoconcentration. Furthermore, in the setting of HFpEF, the increased PAWP makes the PVR vs hematocrit relationship even steeper.29 However, given that the resting mean hemoglobin concentrations were similar across all groups and that all patients in this study achieved a predefined physiologic criteria of maximum exercise testing, the degree of hemoconcentration is assumed to be similar across groups. In conclusion, PV distensibility is a sensitive hemodynamic marker for early detection and progression of PV disease in patients with unexplained dyspnea. Reduced PV distensibility is associated with dynamic RV-PA uncoupling during exercise and impaired peak exercise capacity in PAH, ePH, and HFpEFþPVR. These findings suggest PV distensibility might be useful as an adjunct diagnostic hemodynamic marker and potential therapeutic target for patients with exertional intolerance and abnormal PV response to exercise.

[

-#- CHEST - 2019

FLA 5.6.0 DTD
CHEST2349_proof
12 July 2019
7:20 pm
EO: CHEST-19-0350

]

Acknowledgments Author contributions: A. B. W. and I. S. had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. I. S. and A. B. W. together conceived the original study design, performed the primary data analysis and interpretation, and wrote the manuscript. R. N., R. K. F. O., F. N. R., and D. M. S. contributed substantially to the study design, data analysis and interpretation, and writing of the manuscript. W. M. O. assisted with data analysis. R. N. and R. K. F. O. contributed to study design and critically reviewed the manuscript. Financial/nonfinancial disclosures: None declared. Additional information: The e-Appendix, e-Figures, and e-Tables can be found in the Supplemental Materials section of the online article.

References

8. Malhotra R, Dhakal BP, Eisman AS, et al. Pulmonary vascular distensibility predicts pulmonary hypertension severity, exercise capacity, and survival in heart failure. Circ Heart Fail. 2016;9(6). pii: e003011.

20. Forton K, Motoji Y, Deboeck G, Faoro V, Naeije R. Effects of body position on exercise capacity and pulmonary vascular pressure-flow relationships. J Appl Physiol (1985). 2016;121(5):1145-1150.

9. Oliveira RK, Agarwal M, Tracy JA, et al. Age-related upper limits of normal for maximum upright exercise pulmonary haemodynamics. Eur Respir J. 2016;47(4): 1179-1188.

21. Naeije R, Saggar R, Badesch D, et al. Exercise-induced pulmonary hypertension: translating pathophysiological concepts into clinical practice. Chest. 2018;154(1):10-15.

10. Edwards RH, Denison DM, Jones G, Davies CT, Campbell EJ. Changes in mixed venous gas tensions at start of exercise in man. J Appl Physiol. 1972;32(2):165-169.

22. Argiento P, Vanderpool RR, Mule M, et al. Exercise stress echocardiography of the pulmonary circulation: limits of normal and sex differences. Chest. 2012;142(5): 1158-1165.

11. Maron BA, Cockrill BA, Waxman AB, Systrom DM. The invasive cardiopulmonary exercise test. Circulation. 2013;127(10):1157-1164.

23. Saouti N, Westerhof N, Helderman F, et al. RC time constant of single lung equals that of both lungs together: a study in chronic thromboembolic pulmonary hypertension. Am J Physiol Heart Circ Physiol. 2009;297(6):H2154H2160.

12. Berry NC, Manyoo A, Oldham WM, et al. Protocol for exercise hemodynamic assessment: performing an invasive cardiopulmonary exercise test in clinical practice. Pulm Circ. 2015;5(4):610-618.

1. Hoeper MM, Bogaard HJ, Condliffe R, et al. Definitions and diagnosis of pulmonary hypertension. J Am Coll Cardiol. 2013;62(25 suppl):D42-D50.

13. Sunagawa K, Yamada A, Senda Y, et al. Estimation of the hydromotive source pressure from ejecting beats of the left ventricle. IEEE Trans Biomed Eng. 1980;27(6):299-305.

2. Kovacs G, Olschewski A, Berghold A, Olschewski H. Pulmonary vascular resistances during exercise in normal subjects: a systematic review. Eur Respir J. 2012;39(2):319-328.

14. Brimioulle S, Wauthy P, Ewalenko P, et al. Single-beat estimation of right ventricular end-systolic pressure-volume relationship. Am J Physiol Heart Circ Physiol. 2003;284(5):H1625-H1630.

3. Naeije R, Vanderpool R, Dhakal BP, et al. Exercise-induced pulmonary hypertension: physiological basis and methodological concerns. Am J Respir Crit Care Med. 2013;187(6):576-583.

15. Trip P, Kind T, van de Veerdonk MC, et al. Accurate assessment of loadindependent right ventricular systolic function in patients with pulmonary hypertension. J Heart Lung Transplant. 2013;32(1):50-55.

4. Linehan JH, Haworth ST, Nelin LD, Krenz GS, Dawson CA. A simple distensible vessel model for interpreting pulmonary vascular pressure-flow curves. J Appl Physiol (1985). 1992;73(3):987-994. 5. Reeves JT, Linehan JH, Stenmark KR. Distensibility of the normal human lung circulation during exercise. Am J Physiol Lung Cell Mol Physiol. 2005;288(3):L419L425.

16. Vanderpool RR, Pinsky MR, Naeije R, et al. RV-pulmonary arterial coupling predicts outcome in patients referred for pulmonary hypertension. Heart. 2015;101(1):37-43. 17. Galie N, Manes A, Palazzini M. The difficult diagnosis of pulmonary vascular disease in heart failure. Eur Respir J. 2016;48(2):311-314.

6. Lewis GD, Bossone E, Naeije R, et al. Pulmonary vascular hemodynamic response to exercise in cardiopulmonary diseases. Circulation. 2013;128(13):14701479.

18. Guazzi M, Naeije R. Pulmonary hypertension in heart failure: pathophysiology, pathobiology, and emerging clinical perspectives. J Am Coll Cardiol. 2017;69(13):1718-1734.

7. Lau EMT, Chemla D, Godinas L, et al. Loss of vascular distensibility during exercise is an early hemodynamic marker of pulmonary vascular disease. Chest. 2016;149(2):353-361.

19. Naeije R, Gerges M, Vachiery JL, Caravita S, Gerges C, Lang IM. Hemodynamic phenotyping of pulmonary hypertension in left heart failure. Circ Heart Fail. 2017;10:e004082.

24. Grunig E, Tiede H, Enyimayew EO, et al. Assessment and prognostic relevance of right ventricular contractile reserve in patients with severe pulmonary hypertension. Circulation. 2013;128(18): 2005-2015. 25. Wensel R, Opitz CF, Anker SD, et al. Assessment of survival in patients with primary pulmonary hypertension: importance of cardiopulmonary exercise testing. Circulation. 2002;106(3): 319-324. 26. Gibbons RJ, Balady GJ, Bricker JT, et al. ACC/AHA 2002 guideline update for exercise testing: summary article. A report of the American College of Cardiology/ American Heart Association Task Force on Practice Guidelines (Committee to Update the 1997 Exercise Testing Guidelines). J Am Coll Cardiol. 2002;40(8):1531-1540. 27. Huang W, Resch S, Oliveira RK, Cockrill BA, Systrom DM, Waxman AB. Invasive cardiopulmonary exercise testing in the evaluation of unexplained dyspnea: insights from a multidisciplinary dyspnea center. Eur J Prev Cardiol. 2017;24(11): 1190-1199. 28. Kovacs G, Herve P, Barbera JA, et al. An official European Respiratory Society statement: pulmonary haemodynamics during exercise. Eur Respir J. 2017;50(5). pii: 1700578. 29. Vanderpool RR, Naeije R. Hematocritcorrected pulmonary vascular resistance. Am J Respir Crit Care Med. 2018;198(3): 305-309.

9

chestjournal.org

FLA 5.6.0 DTD
CHEST2349_proof
12 July 2019
7:20 pm
EO: CHEST-19-0350