Determinants of altered left ventricular suction in pre-capillary pulmonary hypertension 

Abstract

Aims

Although the left ventricular (LV) dysfunction in pre-capillary pulmonary hypertension (PH) has been recently recognized, the mechanism of LV dysfunction in this entity is not completely understood. We thus aimed to elucidate the determinants of intraventricular pressure difference (IVPD), a measure of LV suction, in pre-capillary PH.

Methods and results

Right heart catheterization and echocardiography were performed in 86 consecutive patients with pre-capillary PH (57 ± 18 years, 85% female). IVPD was determined using colour M-mode Doppler to integrate the Euler equation. In overall, IVPD was reduced compared to previously reported value in normal subjects. In univariable analyses, QRS duration (P = 0.028), LV ejection fraction (P = 0.006), right ventricular (RV) end-diastolic area (P < 0.001), tricuspid annular plane systolic excursion (P = 0.004), and LV early-diastolic eccentricity index (P = 0.009) were associated with IVPD. In the multivariable analyses, RV end-diastolic area and LV eccentricity index independently determined the IVPD.

Conclusion

Aberrant ventricular interdependence caused by RV enlargement could impair the LV suction. This study first applied echocardiographic IVPD, a reliable marker of LV diastolic suction, to investigate the mechanism of LV diastolic dysfunction in pre-capillary PH.

Graphical Abstract

Open in new tabDownload slide

Introduction

Pulmonary hypertension (PH) is an intractable chronic disease that threatens multiple organs and lives.^1,2 Haemodynamically, it is classified as post- and pre-capillary PH based on the elevation of pulmonary artery wedge pressure (PAWP).^2,3 Although the primary lesion of pre-capillary PH is at the pulmonary vasculature, the coexistence of impaired left ventricular (LV) function during follow-up has been recognized.^4–8 While earlier studies have suggested that the possible contributing factors are impaired LV filling owing to abnormal interventricular septal motion and reduced right ventricular (RV) output, as well as intrinsic diastolic abnormalities of the LV myocardium,^5,6,9,10 the mechanism of LV dysfunction is not completely understood.

In early diastole, rapid myocardial relaxation and recoil of elastic elements compressed during ejection cause LV pressure to fall below the left atrial (LA) pressure.¹¹ Since this progressive intraventricular pressure difference (IVPD) drives brisk early diastolic filling, IVPD is regarded as a measure for LV suction.¹¹ The key mechanism of LV dysfunction in this population could be addressed by elucidating the determinants of IVPD in pre-capillary PH. Therefore, we investigated factors influencing IVPD in patients with pre-capillary PH using echocardiographic colour M-mode Doppler (CMMD) imaging performed simultaneously with invasive haemodynamic measurements.

Methods

Study protocol and population

This study prospectively enrolled 129 consecutive patients who underwent right heart catheterization (RHC) for the initial diagnosis or a follow-up of pre-capillary PH at the Hokkaido University Hospital.

Pre-capillary PH was defined as mean pulmonary arterial pressure (PAP) ≥25 mmHg, PAWP ≤15 mmHg, and pulmonary vascular resistance (PVR) >3 Wood units at rest.² Echocardiography was performed twice, within 2 weeks for chamber quantification and simultaneously with the scheduled catheterization to obtain haemodynamic Doppler parameters and CMMD images. Patients who did not consent to the study (n = 4), those with inadequate echocardiographic image quality (n = 37), or those with post-capillary PH revealed by invasive PAWP (n = 2) were excluded. Accordingly, 86 patients were included in the final analysis (Figure 1). The study was performed according to the Declaration of Helsinki and the ethical standards of the responsible committee on human experimentation (institutional and national). The study protocol was approved by the Institutional Review Board of the Hokkaido University Hospital (No. 019-0064).

Cardiac catheterization

RHC was performed by trained physicians as previously described.¹² Briefly, pressure waveforms were obtained using a 7-F, fluid-filled, balloon-tipped catheter during a breath hold at shallow expiration or the intermediate expiratory position under quiet respiration. From pressure waveforms, the mean PAWP, PAP, and right atrial pressure were determined. Cardiac output (CO) was measured using the thermodilution method and corrected for body surface area to determine the cardiac index. PVR was calculated using the following formula: PVR (Wood units) = (mean PAP − PAWP)/CO.

Echocardiography

Conventional transthoracic echocardiography was performed at the left lateral decubitus position using commercially available ultrasound systems within 2 weeks of RHC. LV ejection fraction (EF) and LA volume were measured in the apical four- and two-chamber views using the biplane method of disks.¹³ LV mass was calculated using the Devereux formula.¹⁴ The RV end-diastolic area (RVEDA) and tricuspid annular plane systolic excursion (TAPSE) were measured according to current guidelines.¹³ The early-diastolic LV eccentricity index, which reflects the degree of direct ventricular interaction owing to the RV pressure overload, was calculated as a ratio of major axis to minor axis of the basal short-axis view at the phase when the ratio was highest from end-systole to early diastole according to a previous report.^12,15 At the time of RHC, echocardiographic images were obtained at the supine position. Transmitral Doppler flow was recorded, and peak early diastolic velocity (E), late diastolic velocity (A), and E/A ratio were measured. Septal and lateral early diastolic mitral annular velocities (e′) were measured from the apical four-chamber view using pulsed-wave tissue Doppler imaging, and the E/e′ ratio was calculated from the average of the septal and lateral e′ values.¹⁶

Speckle tracking method

Two-dimensional echocardiographic images were analysed offline using vendor-independent two-dimensional speckle-tracking software (2D Strain Analysis software version TTA2.4, TomTec Imaging Systems, Munich, Germany) to obtain LV global longitudinal strain (GLS) and LA peak longitudinal strain. The LV endocardial border was traced in the apical four-, two-, and three-chamber views to determine LV GLS.^17,18 In addition, the LA peak strain was obtained from the apical four- and two-chamber views as previously described.^19–21 Strains from each view were averaged and used for the final analysis.

IVPD estimation

To estimate early diastolic IVPD, CMMD images of transmitral flow were obtained from the apical view on the catheterization table and analysed with an in-house-developed automated analysis algorithm based on MATLAB (The MathWorks, Natick, MA, USA) (Figure 2).^22–24 The early diastolic IVPD from the mitral annulus to the LV apex was determined using CMMD data to integrate the one-dimensional Euler equation as previously described.^22–24 And from the temporal profile of the IVPD, the early diastolic peak IVPD from the mitral annulus to the LV apex was calculated. This method has been previously validated by comparing with direct measurements using micromanometers.^25,26

Cardiac magnetic resonance imaging

Cardiac magnetic resonance imaging (MRI) was performed using a 1.5-T Philips Achieva MRI system or Philips Achieva dStream MRI system (Philips Medical Systems, Best, The Netherlands). Gd-BT-DO3A (0.1 mmol/kg, Gadovist; Bayer Pharma AG, Berlin, Germany) was intravenously administered. After 10 min of injection, a breath-holding inversion recovery-prepared three-dimensional turbo field echo pulse sequence with electrocardiogram gating was performed to obtain a delayed-enhancement image with fat saturation of the spectral pre-saturation with inversion recovery. The imaging parameters were as follows: slice thickness, 5 mm; FOV,  400 mm; matrix size, 157 × 256 or 256 × 192; TR/TE, 3.8/1.2 ms or 3.4/1.01 ms; flip angle, 15°. For each subject, the inversion time was adjusted to null the signal from the normal myocardium using the Look-Locker sequence. Hyper-enhanced [late gadolinium enhancement (LGE)] lesions at the ventricular insertion point were visually evaluated.

Statistical analyses

The distribution of continuous variables was visually assessed for normality and expressed as mean ± standard deviation or as median and interquartile ranges, wherever appropriate, and compared using one-way analysis of variance among the groups. Categorical variables are expressed as numbers (percentages) and compared between groups using the χ² test. Pearson’s linear regression analysis was used to detect correlations between two continuous variables. Multivariable linear regression analyses were performed incorporating the significant variables in univariable analyses together with clinically relevant parameters to investigate the independent determinants of IVPD. All tests with P-values <0.05 were considered significant. Statistical analyses were performed using JMP software (version 14.0; SAS Institute Inc., Cary, NC, USA).

Results

Patient characteristics

Among 86 patients enrolled in the final analysis, 37 had substantial PH symptoms (Figure 1). Table 1 summarizes patients’ characteristics according to the severity of the World Health Organization (WHO) functional class. Overall, the mean age of patients was 57 ± 18 years, and 73 (85%) patients were women. Most patients (43%) were classified as Group 1 PH (pulmonary arterial hypertension), followed by 24% as Group 3 (PH due to lung diseases and/or hypoxia), 19% as Group 4 (chronic thromboembolic PH), and 14% as normalized PH during the follow-up. The 6-min walk distance was shortened, and the diffusion capacity of the lungs was lowered according to the WHO functional class. However, the plasma brain natriuretic peptide levels were comparable among the groups. Haemodynamic parameters revealed PVR elevation with WHO functional class, whereas PAWP was comparable among the groups. Although LV EF and RVEDA were comparable among the groups, both septal and lateral e′, GLS, and TAPSE were reduced, and the eccentricity index was increased based on worsened symptoms. IVPD was reduced in patients with WHO functional classes II and III/IV compared to that in patients with class I (not statistically significant). Notably, IVPD in patients with WHO functional class I was reduced compared to the previously reported normal range in healthy subjects (average, 3.16 mmHg),²³ suggesting impaired LV suction in the early stage of pre-capillary PH. Furthermore, the reduced IVPD was associated with reduced inertial component, but not with increased convective component (Table 1).

Determinants of IVPD

In the univariate analyses, prolonged QRS duration, reduced LV EF, enlarged RVEDA, reduced TAPSE, and increased eccentricity index were associated with reduced IVPD (Table 2, Supplementary data online, Figure S1). When these parameters were adjusted by clinically relevant parameters (age, sex, comorbidities, systolic blood pressure, 6-min walk distance, diffusion capacity of the lungs for carbon monoxide, and plasma brain natriuretic peptide level), RVEDA, early-diastolic eccentricity index, and PAWP remained independent determinants of IVPD.

IVPD and LGE

Among the 62 patients who underwent cardiac MRI, LGE was observed at the ventricular insertion point in 39 (63%) patients. Figure 3 shows the comparison of IVPD based on the presence or absence of LGE on cardiac MRI. Interestingly, IVPD was significantly reduced in patients exhibiting LGE compared to that in those not exhibiting LGE (2.1 ± 0.9 vs. 3.0 ± 1.1 mmHg, P = 0.011).

Reproducibility analysis

Intra- and interobserver variabilities of the IVPD were assessed in 20 randomly selected study subjects. Two independent observers (Y.C. and S.I.) analysed the same CMMD images and one blinded observer (Y.C.) repeated the analysis on a separate day. Intra- and interobserver intraclass correlation coefficients for IVPD were 0.997 (95% CI, 0.993–0.999) and 0.878 (95% CI 0.723–0.950), respectively.

Discussion

In the present study, we evaluated the determinants of reduced IVPD, a marker for LV suction, in patients with pre-capillary PH. The main findings of this study are as follows: (i) IVPD decreased with an increase in QRS duration, enlargement of RV, and subsequent flattening of the LV; (ii) the RV size and eccentricity index were independent determinants of IVPD; and (iii) IVPD was reduced in patients exhibiting LGE on cardiac MRI. Of note, this is the first study to evaluate non-invasive early-diastolic IVPD, a reliable measure of LV suction, in patients with pre-capillary PH.

In contrast to the past recognition in which the left heart system should be healthy in pre-capillary PH, increasing attention is being paid to the existence of left heart disorder in this population. Abnormalities of the left heart in pre-capillary PH have been revealed via impaired early diastolic LV filling detected using Doppler echocardiography,²⁷ abnormal elevation of LV end-diastolic pressure after fluid challenge,⁴ reduced LV-free wall strain,⁵ and worsened outcomes associated with reduced LV GLS.⁶ From a pathological view, atrophied cardiomyocytes, attenuated myocardial contractility associated with a reduction of available myosin-based cross-bridges, and reduced phosphorylation levels of sarcomeric proteins have been suggested as one of the mechanisms of LV disorder in patients with pulmonary arterial hypertension.⁷ In addition, neuroendocrine mediators have been implicated in causing LV dysfunction without LV hypertrophy or fibrosis,²⁸ via local LV autocrine/paracrine system activation in experimental studies.²⁹ However, from a physiological view, the leftward shift of the interventricular septum augmented by ventricular interdependence is considered to cause diminished rapid filling in early diastole and subsequent elevation of LV end-diastolic pressure.^12,30,31 Although the relevance and mechanism of LV dysfunction in pre-capillary PH have been elucidated over time as aforementioned, no studies have tested the alteration of the non-invasively assessed IVPD, reliable marker of the LV suction, in this population.

During ejection, the mitral annulus is pulled toward the apex which compresses elastic elements in the LV wall, allowing the annulus to recoil away from the apex in early diastole, resulting in pressure fall in the LV apex relative to the basal LV.^32–34 IVPD is considered a measure of the LV suction strength because of the LV fills as a result of this early diastolic IVPD.²² In the early stage of diastolic dysfunction, IVPD diminishes owing to the slowed rate of LV relaxation.^11,35 On the other hand, early LV filling is consisted of not only diastolic suction but also lengthening load forced by elevated LA pressure.³⁶ Although IVPD from mitral annulus to the LV apex is a different measure from pressure difference between the left atrium and LV apex because of the presence of trans-mitral valvular pressure difference,²⁵ elevated LA pressure could influence the maintained basal IVPD in heart failure patients.²³ Therefore, we need to interpretate the IVPD with a caution in patients with elevated LV filling pressure. Nevertheless, reduced IVPD can be considered a sensitive marker for LV diastolic dysfunction in patients without heart failure as in the present population.¹¹ In the present study, we first demonstrated the determinants of IVPD in pre-capillary PH. Notably, reduced IVPD was determined by an enlarged RV and subsequently increased eccentricity index but not by either e′ or GLS, suggesting that impaired LV suction was mainly caused by augmented ventricular interdependence instead of intrinsic LV myocardial damage. These findings are partially in line with previous findings.^5,6 It is quite reasonable that mechanical compression from the RV impairs early diastolic relaxation of the LV^9,10 before the occurrence of intrinsic myocardial damage of the LV. The decreased inertial component of the IVPD might be associated with the subsequently disturbed early-diastolic wall expansion as in heart failure with preserved EF.²² The negative impact of prolonged QRS duration on IVPD could also support the alteration of IVPD with disease progression. Our study underlines the importance of controlling RV remodelling to prevent impairment of LV diastolic function in pre-capillary PH.

In contrast, IVPD was also found to be reduced in patients exhibiting LGE on cardiac MRI. In the pressure-overloaded RV, myocardial fibrosis occurs owing to a trigger of mechanical stress/stretch.³⁷ The LGE on cardiac MRI detects myocardial fibrosis typically at the ventricular insertion points in PH.³⁷ Freed et al.³⁸ reported an association between LGE and disease progression or poor prognosis in PH patients. Altogether with the relationship between e′ or GLS and IVPD, reduced IVPD in the LGE positive population might occur owing to advanced disease severity and not owing to reduced LV function itself. Nevertheless, reduced LV suction in patients exhibiting LGE is noteworthy, suggesting that IVPD becomes a marker for LGE and subsequent worse prognosis in pre-capillary PH.

The present study had some limitations. Firstly, because the IVPD around the basal LV could be maintained by elevated LA pressure in patients with heart failure as mentioned above, subdivided IVPD around the LV apex might be preferred to express LV suction.²³ Although we tried to analyse the subdivided IVPD in the present population, we could not obtain stable data of apical and basal IVPDs, which might be due to relatively low image quality acquired on the catheter table. In contrast, reproducibility of the total IVPD from the mitral annulus to the apex, which can be obtained through more simple calculation process than the subdivided IVPDs, was acceptable. While the present population consisted of relatively low LA pressure, the results thus need to be understood with a caution. Second, this was a single-centre study with a relatively small population; thus, larger studies are needed to confirm our results. Third, patients with various subtypes of pre-capillary PH were enrolled in the study, resulting in the application of our results for a specific aetiology. However, the population excluding left heart disease could manifest as a clinical condition caused by abnormal pulmonary vasculature. Fourth, since some patients with severe lung diseases were included in this study, there could be a discrepancy between the severity of pulmonary vascular lesions and clinical symptoms. Finally, echocardiographic haemodynamic parameters were acquired simultaneously with RHC, whereas chamber measurements were taken on the echocardiography performed within 2 weeks of the RHC for the reason of image quality, and also, because of the inadequate position of the image acquisition of CMMD of the LV inflow during RHC, not a few cases were excluded due to insufficient image quality.

In conclusion, aberrant ventricular interdependence caused by RV enlargement could impair LV suction in patients with pre-capillary PH. Echocardiographic IVPD, a reliable marker of LV diastolic suction, underlines the importance of the early recognition of LV dysfunction in pre-capillary PH.

Supplementary data

Supplementary data are available at European Heart Journal - Cardiovascular Imaging online.

Acknowledgements

We would like to thank Honyaku Center Inc. (www.honyakuctren.com) for English language editing.

References

Author notes