Venous dysfunction in obstructive hypertrophic cardiomyopathies: Results from the “HCM-Vein” pilot study

Abstract

Background

Hypertrophic cardiomyopathies (HCM) can be complicated by left ventricular outflow-tract obstruction (LVOTO) responsible for disabling exercise symptoms, a phenomenon influenced by hemodynamic factors including venous return.

Methods

We aimed to evaluate venous dysfunction in obstructive HCM patients compared to healthy controls, and to investigate the relationship between venous dysfunction parameters and LVOTO in HCM. This is a clinical, monocentric, prospective, pilot study, in a tertiary care center. We investigated venous function using venous air plethysmography, and endothelial function.

Results

Among the 30 symptomatic obstructive HCM patients, 30% (n = 9) presented abnormal venous residual volume fraction (RVFv) which translates in elevated ambulatory venous pressure vs. 0% in the 10 healthy controls (p < 0.05). Comparing obstructive HCM patients with abnormal RVFv (n = 9) to other obstructive HCM patients with normal RVFv (n = 21), there were no significant differences in terms of age, sex (67% male), and classical echocardiographic parameters both at rest and during exercise, except for left ventricular end-diastolic volume index which was significantly lower in the group with abnormal RVFv compared to the other HCM patients (40.1 ± 9.0 ml/m²vs. 50.2 ± 10.6 ml/m², p = 0.01). Fifty six percent of obstructive HCM patients with abnormal RVFv had an absolute increase in Willebrand factor (vs. 26% of other obstructive HCM patients, p < 0.05).

Conclusions

In this pilot monocentric study, venous insufficiency was observed in about 30% of symptomatic obstructive HCM patients. Patients with venous insufficiency had more frequently a smaller LV cavity volume. Due to the small sample size, this study is only hypothesis-generating, and further investigations are needed.

1. Introduction

Hypertrophic cardiomyopathies (HCM) can be complicated by left ventricular (LV) outflow-tract obstruction (LVOTO) responsible for disabling exercise symptoms. LVOTO is a complex, multifactorial and dynamic phenomenon influenced by the degree of LV hypertrophy but also by mitral valve and apparatus abnormalities, and by hemodynamic factors including venous return (LV preload). Our team has previously demonstrated that LVOTO can also be influenced by the conditions of realization of exercise echocardiography tests (upright position vs. semi-supine position, bicycle vs. treadmill), and by an abnormal venous return capacity. [1], [2], [3], [4] In parallel, it has been demonstrated, by other research teams, that HCM can be associated with endothelial and microvascular peripheral dysfunction. [5], [6], [7] Systemic microcirculatory involvement may suggest more global venous dysfunction. To our knowledge, no studies have investigated the venous circulation in the context of HCM.

Indeed, to date, we have no information in the literature regarding:

• –the prevalence of venous dysfunction in obstructive HCM compared to healthy controls;
• –the relation between venous and endothelial dysfunction in HCM;
• –the relation between venous dysfunction and the level of obstruction in HCM.

Consequently, we aimed to investigate these unresolved points in a pilot study comparing symptomatic obstructive HCM patients to healthy controls.

2. Methods

This monocentric, prospective, non-randomized pilot study was designed to include 30 symptomatic adult patients with a diagnosis of symptomatic obstructive HCM of sarcomeric origin or secondary to Fabry’s disease, followed or referred in our center with expertise in hereditary cardiomyopathies at Bordeaux University Hospital, between October 2018 and December 2021, and to compare them to 10 healthy controls. Information concerning the study and data collection was provided to all patients, and the study protocol was approved by the institutional review board. Patients and healthy control subjects were included after providing informed consent.

The inclusion criteria were: age ≥ 18 years-old with echocardiographic documentation of LVOTO combined to LV hypertrophy (maximal wall thickness ≥ 15 mm) with the diagnosis of HCM of sarcomeric origin or secondary to Fabry disease; New York Heart Association (NYHA) functional Class ≥ 2; sinus rhythm; and ability to perform exercise testing. As per current guidelines, [8], [9] significant obstruction at rest in the supine position was defined as LVOT maximal gradient ≥ 30 mmHg, and during provocation maneuvers as LVOT maximal gradient ≥ 50 mmHg, as this is the threshold considered clinically relevant when deciding the need for invasive septal reduction therapy. [2], [10], [11].

The inclusion criteria for healthy controls were: subjects ≥ 18 years-old, without known cardiac disease.

The exclusion criteria were: poor quality echocardiographic window on supine evaluation; recent history of syncope or severe arrhythmia; apical LV hypertrophy; contraindication to exercise; pregnancy or breastfeeding; uncontrolled arterial hypertension (systolic > 160 mmHg and/or diastolic > 120 mmHg).

For all subjects, past medical history and clinical characteristics were obtained from the hospital database and following clinical examination.

HCM patients underwent the same day: clinical evaluation, 12-lead electrocardiography, transthoracic echocardiography at rest and during treadmill exercise, a blood sample test (measuring endothelial function biomarkers), venous air plethysmography, and upper member arterial Doppler echography with analysis of Flow Mediated Dilatation (FMD).

Healthy controls underwent the same day: clinical evaluation, 12-lead electrocardiography, a blood sample test (endothelial function biomarkers), venous air plethysmography, and upper member arterial Doppler echography with analysis of FMD.

Biomarkers evaluated were Brain Natriuretic Peptide (for HCM patients only) and a selection of endothelial function biomarkers (see below).

2.1. Rest echocardiography

Echocardiography was performed using M3S probe, Vivid 9 or E95 (GE® Vingmed Ultrasound AS, Horten, Norway). All images were digitally stored for offline analysis using EchoPAC® version 203 (GE Vingmed Ultrasound®). Complete two dimensional and Doppler images were acquired according to standard techniques, and two-dimensional measurements were obtained in conventional views over 3 separate cardiac cycles. [12] Evaluation of diastolic function included conventional Doppler-based measurements. Pulsed-wave Doppler velocities were measured from the apical four-chamber view by positioning a 2-mm sample volume at the extremity of the mitral leaflets. Tissue Doppler was applied in the pulsed Doppler mode to record E’ mitral annular velocities at the septal and lateral corners. Tricuspid regurgitation continuous Doppler flow was acquired as well as vena cava maximal diameter and excursion. [13] To study longitudinal LV systolic function, two-dimensional recordings of apical four-chamber, two-chamber, and three-chamber views were obtained at a frame rate > 50 Hz.

2.2. Exercise treadmill echocardiography

Exercise echocardiography (EE) evaluations were performed during treadmill exertion using the T2100 machine and the same echocardiographic machine used for rest evaluation. The modified Bruce protocol was applied. Starting at 2.2 METS, the workload was increased every 3  min up to the maximal tolerated effort. At each stage from rest to the 6-min recovery period, conventional recordings were acquired in two dimensional views and continuous and color Doppler modalities, and stored for offline analysis. In addition to ECG tracings, systolic and diastolic blood pressures were recorded at each stage.

2.3. Echocardiographic measurements

An independent observer, blinded to patients' history, performed a retrospective analysis on all cases, applying standard measurements according to ASE/EAE guidelines [12] and using the echography’s internal quantitation package. Maximal end-diastolic wall thickness was measured on two dimensions. LV volumes and LV ejection fraction were calculated from apical views using Simpson’s rule. [8] Biplane maximal left atrial volume (calculated by area-length method) was indexed to body surface area. Transmitral E and A diastolic velocities and E-wave deceleration time were measured. The mean E/e’ ratio was calculated. Systolic pulmonary artery pressure was calculated based on the measured tricuspid gradient and estimated right atrial pressure from inferior vena cava size and degree of collapse. [13] Mitral regurgitation was graded (none, mild, moderate, severe) using the Proximal Isovelocity Surface Area method. In the event of non-central jets, color jet extension in the three main apical views and the peak mitral E-wave velocity were also considered. Longitudinal LV deformation was measured using the two-dimensional speckle-tracking method. [11], [12], [13], [14] Global longitudinal strain was obtained from apical views. The software package (EchoPAC® version 203 GE Medical systems) automatically tracks and accepts segments of good tracking quality, and the region of interest can be manually modified if necessary to ensure optimal tracking of LV hypertrophic endocardium. It was expressed in absolute value in order to make the interpretation easier. Systolic anterior motion was considered present only in the event of complete systolic apposition of the mitral valve on the septum. LV outflow velocities were measured using continuous-wave Doppler (in the supine position, during Valsalva maneuver, and in standing position). LV outflow gradients were measured and automatically calculated. Specific attention was given not to confuse them with mitral regurgitation flow when present.

During exercise, mitral regurgitation, if present, was graded (none, mild, moderate, severe) according to the same method as during rest. During both exercise and post-exercise evaluation, LV outflow velocities were measured using continuous-wave Doppler, with the same direction and angle as recorded at rest. Specific attention was paid not to confuse them with mitral regurgitation flow when present.

2.4. Venous air plethysmography

An independent observer, blinded to patients' history performed venous air plethysmography and FMD. Venous plethysmography is a non-invasive, and non-painful test evaluating different parameters of venous filling by inflation of an armband around the leg, upright positioning, and flexion–extension of the leg. [15] Total duration of the test was estimated at 45 min. Were obtained for each subject: venous outflow fraction (normal value > 40%), venous volumes (in ml), venous filling index (normal value < 2 ml/s), venous ejection volume (in ml), venous ejection fraction (normal value > 60%), and venous residual volume fraction (RVFv) (normal value < 35%).

2.5. Upper member arterial duplex echography with analysis of flow mediated dilatation

The measurement of the evolution of brachial artery diameter before and after inflation of an armband for 5 min allowed us to evaluate FMD. [16] It is calculated as the percent change of brachial artery diameter from baseline to the maximum diameter reached after cuff release. A percent change < 7% is considered abnormal. This is a non-invasive and non-painful test of endothelial function, and mean duration of realization was of 30 min.

2.6. Endothelial function biomarkers

A blood sample test was taken on all subjects on the day of inclusion, for deferred grouped analyzes at the end of study inclusion. Several endothelial function biomarkers were measured: Von Willebrand rag (normal value considered between 50 and 150), VCAM-1, e-selectin, thrombomodulin, endothelin 1, sEPCR, and nitrate + nitrite.

3. Statistical analysis

Statistical analyses were performed on the version 17.0 of SPSS software for Windows (SPSS® Inc, Chicago, IL, USA). Continuous variables were expressed in mean ± standard deviation and categorical variables in frequencies and percentages. Comparison of these variables between groups (and subgroups) of subjects were performed using the Mann-Whitney U test or Pearson Chi-square test, respectively.

The correlations between the data of plethysmography, FMD, biomarkers of endothelial function, and intra-LVOTO were performed using Fisher test. A p-value < 0.05 was considered statistically significant.

4. Results

4.1. Studied populations

The mean age was 61 ± 12 years in the HCM group (n = 30) vs. 56 ± 13 years (p = 0.31) in the control group (n = 10). Sixty seven percent of HCM patients were men vs. 50% of the healthy controls (p = 0.30) (Table 1). Twenty eight patients had sarcomeric OHCM and 2 patients had OHCM secondary to Fabry disease. Among the 30 OHCM patients, 8 patients had obstruction at baseline, and 22 patients had obstruction only after provocation.

Table 1.

Comparison between HCM subjects and controls.

	HCM subjects (n = 30)	Controls (n = 10)	p
Men	20 (67%)	5 (50%)	0.35
Age, years	61 ± 12	56 ± 13	0.30
Body mass index	29.3 ± 4.9	25.3 ± 3.6	0.052
Height, cm	168 ± 10	167 ± 11	0.84
Systolic blood pressure, mmHg	136 ± 15	133 ± 18	0.56
Von Willebrand factor rag	152 ± 63	134 ± 35	0.96
Abnormal flow mediated dilatation (value < 7%)	20 (67%)	4 (40%)	0.14
Abnormal venous filling index (value > 2)	5 (17%)	1 (10%)	0.61
Abnormal venous residual volume fraction (value > 35%)	9 (30%)	0 (0%)	0.049
Venous ejection fraction, %	59 ± 15	63 ± 19	0.68

Data are expressed in mean ± SD or n (%), HCM = hypertrophic cardiomyopathy.

4.2. Prevalence of venous dysfunction in obstructive HCM compared to healthy controls

Among the 30 symptomatic obstructive HCM patients, 30% (n = 9, 8 sarcomeric, 1 Fabry) presented abnormal venous residual volume fraction (RVFv) suggesting elevated ambulatory venous pressure vs. 0% of the 10 healthy controls (p < 0.05) (Table 1). From the 28 sarcomeric OHCM patients, 29% (8/28) presented abnormal RVFv.

4.3. Comparison of obstructive HCM patients with vs. without abnormal RVFv

We compared obstructive HCM patients with abnormal RVFv (n = 9) to other obstructive HCM patients with normal RVFv (n = 21). There were no significant differences with regards to age, sex (67% male), body mass index, and height (Table 2), as well as classic echocardiographic parameters at rest and during exercise except for left ventricular end-diastolic volume index which was significantly lower in abnormal RVFv group compared to other HCM patients (40.1 ± 9.0 ml/m² vs. 50.2 ± 10.6 ml/m², p = 0.013) (Table 2).

Table 2.

Comparison of data between HCM patients with or without abnormal venous residual volume fraction.

	Abnormal RVFv (n = 9)	Normal RVFv (n = 21)	p
Men	67%	67%	0.89
Age	64 ± 9	60 ± 13	0.29
Body mass index	31.2 ± 4.6	28.5 ± 4.9	0.17
Height, cm	169 ± 11	166 ± 8	0.41
Systolic blood pressure, mmHg	135 ± 13	137 ± 21	0.85
NYHA functional class	2.3 ± 0.5	2.3 ± 0.5	0.81
Brain Natriuretic Peptide	377 ± 388	312 ± 507	0.74
Elevated Von Willebrand factor	56%	26%	< 0.05
Maximal LV wall thickness, mm)	17.3 ± 2.0	18.1 ± 2.5	0.36
Left ventricular ejection fraction, %	68.9 ± 7.3	70.1 ± 7.8	0.68
Global longitudinal strain, %	15.3 ± 4.5	15.6 ± 4.2	0.89
Left ventricular end-diastolic volume index, ml/m2	40.1 ± 9.0	50.2 ± 10.6	0.02
Max LVOTO gradient, mmHg	59.4 ± 31.0	65.9 ± 31.4	0.61
Mean E/E’	12.0 ± 2.6	11.5 ± 4.7	0.67
Left atrial volume index, mm/m2	49.8 ± 16.5	53.9 ± 18.3	0.55
Vena cava expirium diameter, mm	16.3 ± 6.2	15.8 ± 6.0	0.82
Systolic pulmonary artery pressure, mmHg	34.4 ± 12.8	33.2 ± 7.9	0.78

Data are expressed in mean ± SD or n (%), RVFv = venous residual volume fraction.

4.4. Relation between venous dysfunction and endothelial dysfunction in HCM

Comparing obstructive HCM patients with abnormal RVFv (n = 9) to other obstructive HCM patients with normal RVFv (n = 21), there was no significant difference in FMD between the 2 groups. There was also no obvious relation between RVFv and FMD as depicted in Fig. 1A (r = 0.02).

Fig. 1.

Relations between RVFv and FMD (A), RVFv and Willebrand factor (B), RVFv and the maximal peak of LVOTO (C), and RVFv and left ventricular end-diastolic volume index (D).

However, it appears that obstructive HCM patients with abnormal RVFv had more frequently an elevated Willebrand factor (56% vs. 26% of other obstructive HCM patients, p < 0.05) (Table 2). Fig. 1B depicts the relation, relatively weak, between RVFv and Willebrand factor in the HCM group (r = 0.27).

Other measured biomarkers did not appear to be different between the 2 groups of patients (VCAM-1, e-selectin, thrombomodulin, endothelin 1, sEPCR, and nitrate + nitrite).

4.5. Relation between venous dysfunction and the level of obstruction in HCM

There was no linear relation between the RVFv and the maximal peak of LVOTO (Fig. 1C).

However, RVFv seemed to be related to left ventricular end-diastolic volume index (r = 0.24) (Fig. 1D).

Fig. 2 displays an example: A is an obstructive HCM patient with normal RVFv and greater left ventricular end-diastolic volume index compared to B HCM patient which presents an abnormal RVFv and smaller left ventricular end-diastolic volume index as well as a higher LVOTO gradient.

Fig. 2.

Example: A: obstructive HCM patient with normal RVFv (23%), LVEDVi = 55 ml/m², and maximal LVOTO gradient = 47 mmHg B: obstructive HCM patient with abnormal RVFv (40%), LVEDVi = 43 ml/m², and maximal LVOTO gradient = 153 mmHg.

5. Discussion

This pilot study shows a prevalence of 30% of venous dysfunction in symptomatic obstructive HCM patients (29% of sarcomeric OHCM patient), which is not observed in our control population. This phenomenon surely participates in the multifactorial phenomenon that is LVOT obstruction. Moreover, this addresses the question of the vascular component of this genetic disease which is already known to be associated with endothelial dysfunction, as well as its prognostic impact.

5.1. Venous dysfunction in HCM

LVOT obstruction is a multifactorial phenomenon induced by a genetic defect responsible for LV asymmetrical hypertrophy and hypercontractility, attraction and elongation of the mitral valve to the septal wall, and aggravated by afterload and preload conditions. Indeed, dehydration can increase LVOT gradient and its related symptoms. This study shows that a significant proportion of patients with symptomatic LVOT obstruction can also be impacted by venous dysfunction aggravating their situation, and this could be taken into account in their management. Evaluation of venous dysfunction is not routinely performed, and its treatment is not part of the current management of HCM. However, a susceptibility to venous dysfunction seems to exist in this population. Thirty percent of symptomatic obstructive HCM patients had inadequate leg venous function in this study. Of note, only one parameter of venous dysfunction was significantly different between HCM patients and controls with venous filling index tending to be different without reaching statistical significance. It would be interesting to evaluate the pathophysiological mechanisms of this potential venous dysfunction in sarcomeric HCM: is the venous component part of a more global endothelial vascular dysfunction in this genetic disease classically involving mainly the heart?

5.2. Relation between venous dysfunction and endothelial dysfunction

What is the cause and what is the consequence? It was previously demonstrated that chronic venous dysfunction with venous hypertension is responsible for venous dilatation, valve distortion, altered shear stress ultimately leading to inflammation, leucocyte activation and endothelial dysfunction. [17], [18], [19] So it seems that the primum movens is more venous dysfunction. We don’t know all pathophysiologic phenomena involved in sarcomeric HCM but it appears that it is not an isolated cardiac myopathy but a more general disease affecting the systemic vasculature as well (peripheric veins and microvessels in particular).

5.3. Clinical perspective

It is probably unrealistic in routine practice to test for venous dysfunction in all HCM or even obstructive HCM patients. However, if venous dysfunction is suspected in obstructive HCM patient, practicians should try to document it. Indeed treating and preserving venous function could be important in the evolution of the disease, and LVOT obstruction could be partly improved by measures improving venous return (such as venous compression and various therapies). [2] This hypothesis needs further investigation. Air plethysmography is not routinely performed but may be done in some select cases. The diagnosis of chronic venous disease may simply rely on clinical examination (malleolar dermatitis, telangiectasis, functional symptoms of leg swelling in the evenings) and on duplex echography, although this latter method can only show valvular incompetence or vein obstruction and not muscular efficiency which may result in purely functional venous insufficiency.

5.4. Limitations

As this is a pilot study, the small sample size does not allow us to conclude about the entire spectrum of HCM population. This is more a collection of clinical cases rather than an observational study. However, it appears that HCM patients are more susceptible to be impacted in term of obstruction when they have an associated venous dysfunction. Another limitation is the lack of inclusion of asymptomatic HCM patients and non-obstructive HCM patients as this could have shown differences in venous status across different subgroups.

6. Conclusions

In this pilot monocentric study, venous insufficiency was observed in about 30% of symptomatic obstructive HCM patients while absent in controls. Patients with venous insufficiency tend to have an associated endothelial dysfunction. However, there was no linear relation between the degree of venous dysfunction and the maximal gradient of LVOTO. Due to the small sample size of our population, this study is only hypothesis-generating, and these findings need to be confirmed in larger HCM cohorts that include both obstructive and non-obstructive HCM patients.

Declaration of Competing Interest

The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: This study was partially granted by Amicus Therapeutics.