The Influence of Genotype on the Cardiopulmonary Test Response in Patients Affected by Hypertrophic Cardiomyopathy

Abstract

In hypertrophic cardiomyopathy (HCM), the presence of pathogenic/likely pathogenic (P/LP) disease-causing genetic variants may indicate a worse prognosis. Few data exist on the effects of these genetic variants on cardiopulmonary exercise test (CPET) performance in HCM patients. We analysed asymptomatic and slightly symptomatic HCM patients (NYHA I-II) whose genetic analysis and CPET were available; at baseline, left ventricular function was normal and severe left ventricular outflow trait obstruction was excluded. Out of 120 HCM patients, we excluded 13 carrying variants of uncertain significance; of the remaining 107 patients, 54 were genotype negative [gene (−)], and 53 had a P/LP variant in sarcomeric genes [gene (+)]. Patients in the two groups had similar NYHA class, cardiovascular risk factors and echocardiographic characteristics. Gene (+) patients showed a lower peak VO₂% and O₂ pulse % (p < 0.05). Moreover, among gene (+), patients with P/LP variants in the so called “thin-filament” genes (TNNT2, TPM1 and MYL3) had the poorest CPET results. In asymptomatic or slightly symptomatic HCM patients with similar echocardiographic characteristics, exercise tolerance is affected by the genetic background. Indeed, exercise capacity is poorer in gene (+) compared to gene (−) patients and those carrying P/LP variants in “thin-filament” genes show the worst performance.

1. Introduction

Hypertrophic cardiomyopathy is a primary cardiac disorder, defined as the presence of increased left ventricular wall thickness (≥15 mm) with or without right ventricular hypertrophy, that is not solely explained by abnormal loading conditions [1,2]; lesser degrees of wall thickening (≥13 mm) are required for the diagnosis in family members [2]. HCM is the most common among inherited cardiac conditions, it is mainly transmitted in an autosomal dominant mode and genetic analysis is recommended in all patients fulfilling diagnostic criteria for HCM. The identification of a P/LP variant in the proband confirms the diagnosis of sarcomeric HCM and excludes the presence of phenocopies that could require a specific treatment. Family members should be screened whenever a P/LP variant is identified in the proband [2,3,4,5,6,7,8]. The screening should include genes with definite or strong evidence for being disease causing, such as the eight sarcomere genes MYH7, MYBPC3, TNNI3, TNNT2, TPM1, MYL2, MYL3 and ACTC1 [2,6,7,8,9] and usually genes associated with phenocopies (i.e., LAMP2, GLA, PRKAG2 and TTR). Most genotyped-positive HCM patients carry defects in the thick-filament genes, myosin heavy chain (MYH7) and myosin-binding protein C (MYBPC3); less frequently in thin-filament genes, including cardiac troponin T (TNNT2) and I (TNNI3), a-tropomyosin (TPM1) and cardiac actin (ACTC1) [2,7,8,9]. Overall, around 50–60% of HCM patients carry P/LP variants in these sarcomeric genes. Furthermore, other genes have been associated with the disease, encoding proteins of the sarcomere apparatus (i.e., MYH6, TNNC1, TTN), of the Z-disc (i.e., ACTN2, FLNC, DES) or involved in calcium homeostasis pathways (i.e., PLN, JPH2) [10], but the strength of association with the disease is not as strong as for the eight sarcomeric genes [6]. HCM has an age-related penetrance and expressivity can be heterogenous and influenced by additional genetic [2,9,10,11,12] and acquired factors (e.g., hypertension, obesity, exercise, environmental factors) [2,3,13]. Data from the SHaRe Registry showed that being gene (+) might be associated with a worse outcome [5] and although the discussion about this finding is still ongoing [13], there is general agreement that patients with more than one sarcomere P/LP variant are more likely to progress to a dilated phenotype [3,4,5]. Finally, it has been suggested that P/LP variants in the so-called thin-filament genes (TNNT2, TPM1, MYL2 and MYL3) are at higher risk of evolution toward heart failure [14,15] and possibly malignant arrhythmias [16].

Fatigue and dyspnoea during exercise can often occur in HCM patients [1,2,7,12]. The limited exercise performance has been attributed to several mechanisms, such as diastolic dysfunction, dynamic obstruction of the left ventricular outflow trait, left ventricular systolic dysfunction and chronotropic incompetence [2,7,12]. Objective assessment of functional capacity through the cardiopulmonary exercise test (CPET) may be useful in HCM patients, but its role in patients’ risk stratification and management is still a matter of debate [17,18,19,20,21]. In our population, we specifically focused on the impact of genotype, carefully evaluated according to ACGM criteria [22], on CPET performance.

2. Materials and Methods

2.1. Study Population

The study was approved by our Ethical Committee. All patients signed an informed consent (2021_01_26_06 CARDGEN-REG approved by the Istituto Auxologico Italiano local Ethical Committee). We considered HCM patients followed up in our Cardiomyopathy Outpatient Clinic whose genetic analysis results were available. We included in this analysis patients in NYHA class I-II without a history of heart failure, already receiving a personalised optimal treatment as recommended by European Guidelines [2], and capable of performing a CPET within three months from an echocardiogram. Regarding echocardiographic data, we excluded patients with left ventricular (LV) dysfunction, i.e., LV ejection fraction less than 50% [2,12]. Moreover, due to the known influence on exercise performance of a significant left ventricular obstruction [23,24], we excluded from the study population patients showing, at rest or after Valsalva manoeuvre at echocardiography, a clinically significant gradient (see below). All clinical data were collected and stored in an internal database.

2.2. Echocardiography

Echocardiograms were collected by an experienced echocardiography operator using a Vivid 9 GE echocardiographic system, using two-dimensional parasternal long-axis and short-axis views, two-chamber, three-chamber, and four-chamber apical views. Three consecutive cardiac cycles of each view were stored digitally (all patients were in sinus rhythm). LV hypertrophy was assessed with 2-dimensional echocardiography, and the site and extent of maximal wall thickness were identified. Maximal end-diastolic LV wall thickness (MWT) was used as the dimension of greatest magnitude at any site within the LV chamber [25,26]. Peak instantaneous LV outflow gradient was estimated with continuous wave Doppler under basal conditions and after Valsalva manoeuvre; LV outflow obstruction (LVOTO), due to mitral valve systolic anterior motion and mitral-septal contact, was identified either at rest or after Valsalva manoeuvre in the semi supine position [2,3,4,25,26]. Patients showing peak instantaneous outflow gradient ≥ 30 mmHg at rest or ≥50 mmHg after Valsalva manoeuvre, i.e., a relevant obstruction, were excluded from the study population. The following echocardiographic measurements were also collected according to guidelines: the left atrium indexed volume (LAVI), the LV ejection fraction with Simpson’s biplane methods (LVEF, apical four-chamber and two-chamber views), the diastolic function parameters, the LV Global Longitudinal Strain (GLS, apical four/two and three-chamber view) and the mitral regurgitation grade [25,26].

2.3. Genetic Testing

Genetic testing was performed on blood samples through Next-Generation Sequencing (NGS, TruSight Cardio Sequencing kit, Illumina, including 197 genes). The process included the patient’s DNA extraction, purification, amplification and fragmentation, followed by isolation and attachment to labelled beads for short-read sequencing [22]. The resulting alignment against a “reference” human genome sequence allowed the identifications of genetic variants in the patient’s sample: all significant variants identified were confirmed with Sanger sequencing. Genetic variants were then evaluated according to their frequency in the general population (Genome Aggregation Database, Exome Variant Server, 1000 Genomes Project), presence or absence in human genetic variants databases, literature description, localisation and conservation, and they were finally classified according to ACMG guidelines; only pathogenic (P) and likely pathogenic (LP) variants were considered to classify a patient as gene (+) [22].

2.4. Cardiopulmonary Test

Exercise was performed on cycle ergometer, beginning with two minutes of rest, followed by two minutes of freewheeling warm-up, and then by a ramp-incremental load increase by 10 W per minute until volitional exhaustion. During the test, the patient breathed through a non-rebreathing mask connected to a metabolic cart Sensor Medics 2900 (Sensor Medics, 22705 Savi Ranch Pkwy, Yorba Linda, CA, USA) for breath by breath measurements of ventilation (VE, L/min), oxygen consumption (VO₂, L/min) and carbon dioxide production (VCO₂, L/min) [17]. A 12-lead ECG was monitored, and blood pressure and heart rate were measured every two minutes. Respiratory quotient (RER), VO₂ and VCO₂ were averaged during the last 30 s of exercise. Anaerobic threshold (AT) was calculated by the V-slope method. The VE/VCO₂ slope, relating the rate of increase in ventilation per unit increase in CO₂ production, was calculated by linear regression until the anaerobic compensation [26]. Other variables considered were the rate of increase in VO₂ relative to workload (VO₂/Work) that has been interpreted as an indicator of cardiovascular efficiency [27], and O₂ pulse (pO₂, ml/beat), which is computed as VO₂/HR and potentially interpreted as a surrogate measure of stroke volume [28].

2.5. Statistics

Data are expressed as mean ± 1 standard deviation. We used Microsoft Excel statistic package for the analysis. Differences in continuous variables between groups were evaluated by unpaired t-test. Differences in prevalence between groups were analysed by χ² test with Yates’ correction. A p value < 0.05 was considered significant.

3. Results

Out of 120 HCM patients fulfilling all inclusion criteria, 13 were excluded because they carried variants of uncertain significance (VUS).

Of the remaining 107 patients, 53 had a pathogenic/likely pathogenic variant (gene +) and 54 were genotype negative (gene −). The main mutated genes identified were MYBPC3 (33 pts, 62%) and MYH7 (12 pts, 23%); other genes (TNNT2, TPM1 and MYL3) were found in seven patients (15%) (see Figure 1).

Patients’ NYHA class was between I and II; none was involved in competitive sport activity.

The primary genetic analysis was focused on the eight sarcomere genes, according to guidelines [6]; however, all other genes present in the Trusight panel were evaluated and no other P/PL variants were identified.

Table 1. Pathogenic/likely pathogenic variants identified, with the corresponding ACMG classification.

Pt Number	Gene	Type Gene	Nucleotide Change	Amino Acid Change	Class	ACMG Criteria
1	MYBPC3	THICK	c.G1624C	p.E542Q	P	PM3, PP1, PS3, PM2, PP3
2	MYBPC3	THICK	c.T3713C	p.L1238P	P	PS4, PP1, PP3, PM2
3	MYH7	THICK	c.G1231A	p.V411I	P	PS4, PM1, PP2, PM2, PP3
4	MYBPC3	THICK	c.1458-1G>A		P	PS4, PVS1, PM2
5	MYH7	THICK	c.G1063A	p.A355T	P	PP5, PP3, PM1, PM2
6	MYH7	THICK	c.C2080T	p.R694C	LP	PS4, PP1, PM1, PP2, PM2, PM5, PP3
7	TNNT2	thin	c.C862T	p.R288C	P	PS4, PP1, PS3, PM2, PM5, PP2
8	MYBPC3	THICK	c.G532A	p.V178M	LP	PM1, PM2, PP3 (strong)
9	MYL3	thin	c.G447A	p.M149I	P	PS4, PS1, PM5, PM1, PP2, PM2, PP3
10	TNNT2	thin	c.C418T	p.R140C	P	PS4, PP1, PS3, PM1, PP2, PM2, PP3
11	MYH7	THICK	c.G3346A	p.E1116K	LP	PM2, PP3 (strong), PP2, PP5
12	MYBPC3	THICK	c.1351+2T>C		P	PS4, PVS1, PM2
13	MYBPC3	THICK	c.3192dupC	p.K1065Qfs * 12	P	PS4, PP1, PVS1, PM2
	MYBPC3	THICK	c.C1112G	p.P371R	LP	PM2, PM1, PP3 (strong)
14	TNNT2	thin	c.A803T	p.K268I	LP	PS4, PM2, PP3, PP2
15	MYH7	THICK	c.G2770A	p.E924K	P	PP1, PS3, PS2, PM1, PP2, PM2, PM5, PP3
16	TPM1	thin	c.G172C	p.D58H	LP	PP3, PM2, PP2
17	MYBPC3	THICK	c.C1504T	p.R502W	P	PM3, PP1, PM2, PM5, PM1, PP3
18	MYBPC3	THICK	c.C1789T	p.R597W	P	PS4, PM2, PM5, PP3
19	MYBPC3	THICK	c.3331-1G>A		P	PS4, PVS1, PM2
20	MYBPC3	THICK	c.G2198T	p.R733L	LP	PM1, PM2, PM5 (strong), BP4
21	MYH7	THICK	c.G428A	p.R143Q	P	PS4, PP1, PM1, PP2, PM2, PM5, PP3
22	MYBPC3	THICK	c.C1960T	p.R654C	LP	PM1, PM2, PM5 (strong), BP5
23	MYH7	THICK	c.A1615C	p.M539L	P	PS4, PM1, PP2, PM2, PM5, PP3
24	MYBPC3	THICK	c.1458-1G>A		P	PS4, PVS1, PM2
25	MYBPC3	THICK	c.506-2A>C		P	PS4, PP1, PS3, PVS1, PM2
26	TNNT2	thin	c.C862T	p.R288C	P	PS4, PP1, PS3, PM2, PM5, PP2
27	MYBPC3	THICK	c.G772A	p.E258K	P	PP3, PP5, PM2, PP2
28	MYBPC3	THICK	c.2943_2947del	p.Q981Hfs * 67	P	PS4, PVS1, PM2
29	MYBPC3	THICK	c.1227-13G>A		LP	PM2, PS4, PS3, PP1, PP5
30	MYBPC3	THICK	c.2864_2865delCT	p.P955Rfs * 95	P	PS4, PVS1, PM2
31	MYH7	THICK	c.G4402C	p.E1468Q	LP	PM2, PM5, PP3, PP2
32	MYBPC3	THICK	c.G2459A	p.R820Q	P	PS4, PP1, PS3, PM2, PM5, PM1, PP3
33	MYBPC3	THICK	c.2309-2A>G		P	PS4, PP1, PVS1, PM2
34	MYBPC3	THICK	c.2157_2158delTG	p.C719X	P	PS4, PVS1, PM2
35	MYBPC3	THICK	c.2157_2158delTG	p.C719X	P	PS4, PVS1, PM2
36	MYH7	THICK	c.G2680A	p.E894K	P	PS4, PM1, PP2, PM2, PM5, PP3
37	MYBPC3	THICK	c.913_914del	p.F305Pfs * 27	P	PVS1, PP5, PM2, PS4, PP1
38	MYBPC3	THICK	c.G772A	p.E258K	LP	PS4, PP1, PS3, PM2, PP3
39	MYBPC3	THICK	c.G1828C	p.D610H	LP	PM1, PM2, PP3, PM5
40	MYBPC3	THICK	c.C1789T	p.R597W	P	PS4, PM2, PM5, PP3
41	MYH7	THICK	c.T1228C	p.Y410H	LP	PM1, PP2, PM2, PP3
42	MYH7	THICK	c.G428A	p.R143Q	P	PS4, PP1, PM1, PP2, PM2, PM5, PP3
43	MYBPC3	THICK	c.G1624C	p.E542Q	P	PM3, PP1, PS3, PM2, PP3
44	MYH7	THICK	c.C3133T	p.R1045C	P	PS4, PP3, PM2, PM5, PP2
	TNNT2	thin	c.A659T	p.K220M	LP	PM1, PP2, PM2, PP3
45	MYBPC3	THICK	c.913_914del	p.F305Pfs * 27	P	PVS1, PP5, PM2, PS4, PP1
	MYH7	THICK	c.G2012A	p.R671H	P	PS4, PM1, PM2, PP2, PM5, PP3
46	MYBPC3	THICK	c.2258dupT	p.K754Efs * 79	P	PS4, PVS1, PM2, PM5
47	MYBPC3	THICK	c.C3811T	p.R1271X	P	PM3, PS3, PVS1, PM2
48	MYBPC3	THICK	c.2258dupT	p.K754Efs * 79	P	PS4, PVS1, PM2, PM5
49	MYBPC3	THICK	c.2157_2158delTG	p.C719X	P	PS4, PVS1, PM2
50	MYBPC3	THICK	c.G1505A	p.R502Q	P	PS4, PM2, PM5, PM1, PP3
51	MYBPC3	THICK	c.339delC	p.T114Lfs * 45	LP	PVS1, PM2
52	TNNT2	thin	c.C418T	p.R140C	P	PS4, PP1, PS3, PM1, PP2, PM2, PP3
53	TNNT2	thin	c.C341T	p.A114V	LP	PM1, PP2, PM2, PM5, PP3

The symbol used (*) indicates a nonsense genetic variant.

shows all the pathogenic/likely pathogenic variants identified, with the corresponding ACMG classification.

3.1. Relevance of the Presence of P/LP Mutations

Table 2. Patients’ clinical characteristics.

	All Pts (n = 107)	Gene (+) Pts (n = 53)	Gene (−) Pts (n = 54)	p Value
Age (years)	54 ± 16	50 ± 16	59 ± 16	0.01
Female (%)	42 (40%)	26 (42%)	20 (37%)	0.29
BMI (kg/m2)	26.4 ± 4.4	26.2 ± 5.3	26.5 ± 3.3	0.51
Diabetes Mellitus	7 (7%)	3 (6%)	4 (7%)	0.73
Hypertension	52 (49%)	16 (30%)	32 (59%)	0.05
Active smoke	18 (17%)	11 (21%)	7 (13%)	0.56
Dyslipidemia	60 (57%)	28 (54%)	32 (59%)	0.57
ICD, primary prevention	10 (16%)	5 (9%)	6 (11%)	0.31
THERAPY (% of pts)
β-Blockers	81 (74%)	244 (83%)	37 (69%)	0.08
Dysopiramide	5 (5%)	4 (8%)	1 (4%)	0.37
ACEi/ARB	34 (36%)	16 (37%)	18 (35%)	0.72
Ca++Blockers	11 (11%)	5 (10%)	7 (13%)	0.59
MRA	7 (8%)	5 (10%)	3 (6%)	0.63
Amiodarone	4 (4%)	2 (4%)	2 (4%)	0.90
Diuretics	5 (5%)	2 (6%)	2 (4%)	0.25

shows patients’ clinical characteristics: the two groups [gene (+) and gene (−)] were similar in terms of gender distribution, NYHA class, treatment; the only significant differences were that patients with P/LP variants were younger and less frequently affected by hypertension.

Echocardiographic characteristics are shown in

Table 3. Echocardiographic characteristics.

	All Pts (n = 107)	Gene (+) Pts (n = 53)	Gene (−) Pts (n = 54)	p Value
Maximum wall thickness (mm)	19 ± 5	19 ± 5	18 ± 4	0.21
Indexed LA Volume (mL/m2)	46 ± 16	49 ± 19	43 ± 12	0.04
LVEF (%)	64 ± 8	63 ± 9	64 ± 8	0.73
E/e’	11.1 ± 5.0	11.0 ± 4.5	11.3 ± 5.2	0.55
GLS (%)	−15.7 ± 4.1	−16.6 ± 4.4	−15.0 ± 3.7	0.09
Mitral regurgitation, mild to moderate	81 (76%)	43 (83%)	38 (70%)	0.18
LVTO (rest or Valsalva manoeuvre)	25 (23%)	8 (15%)	17 (31%)	0.08

LA = left atrium; LVEF = left ventricular ejection fraction; GLS = global longitudinal strain; LVTO = left ventricular outflow trait obstruction.

: no significant between-group differences were present in LVEF, GLS, MWT, LAVI and E/e’. A mild LVOTO was found more frequently in gene (−) patients.

Table 4. Cardiopulmonary test data.

	All Pts (n = 107)	Gene (+) Pts (n = 53)	Gene (−) Pts (n = 54)	p Value
Peak RER	1.11 ± 0.15	1.12 ± 0.18	1.10 ± 0.10	0.39
Peak VO2 (mL/kg/min)	20.5 ± 7.5	20.1 ± 7.5	20.3 ± 7.0	0.94
Peak VO2 (% predicted)	77.0 ± 19.5	73.0 ± 16.8	80.0 ± 21.5	0.05
AT VO2 (% VO2 max) predicted)	52 ± 19	50 ± 21	52 ± 18	0.59
O2 pulse peak (mL/beat)	13.1 ± 3.7	13.0 ± 3.2	13.3 ± 4.1	0.25
O2 pulse peak (% predicted)	99.0 ± 21.4	96.5 ± 17.9	103.0 ± 23.7	0.03
VO2/Work [(mL/kg/min)]/Watts	9.7 ± 2.3	9.4 ± 1.8	10.0 ± 2.5	0.16
VE/VCO2 slope	31.2 ± 6.3	32.1 ± 7.0	30.7 ± 5.5	0.37
P(ET)CO2 (mmHg)	32.0 ± 5.2	31.2 ± 6.6	33.5 ± 4.6	0.06

shows CPET data.

All patients performed a maximal volitional effort without between-group differences in RER. The overall exercise performance expressed as age-adjusted peakVO₂ and O₂ pulse [16,17,18,19,20,21,22] was slightly below the lower range of normality; in particular, significantly lower age-adjusted peak VO₂ and O₂ pulse were observed in gene (+) patients, i.e., those carrying a P/LP variant, than in gene (−) patients (Figure 2).

Moreover, a moderate to severe reduction in performance (≤65% of the age-adjusted peak VO₂) [16,17] was observed in 47% (25/53 pts) of gene (+) and only in 20% (11/54 pts) of gene (−) patients (p < 0.05, χ² test).

3.2. Relevance of Different Pathogenetic P/LP Mutations

This analysis was specifically carried out in the gene (+) group: we compared patients with P/LP variants in the thick-filament genes [myosin heavy chain MYH7 and myosin-binding protein C MYBPC3 (gene + THICK), n = 45] and those in the thin-filament genes [TNNT2, TPM1 and MYL3 (gene + Thin), n = 7]. One patient, who carried P/LP variants on both MYBPC3 and TNNT2 genes, was excluded from this analysis; thus, we present data on 52 patients. No differences in the baseline characteristics were found according to the genes involved, as shown in

Table 5. Baseline characteristics of the gene (+) populations.

	All Gene (+) Pts (n = 52)	Gene (+THICK) Pts (n = 45)	Gene (+Thin) Pts (n = 7)	p Value
Age (years)	50.1 ± 14.6	49.4 ± 15.5	51.2 ± 11.7	0.71
Female (%)	22 (42%)	20 (44%)	2 (28%)	0.21
BMI (kg/m2)	26.3 ± 5.5	26.0 ± 5.2	27.1 ± 2.9	0.81
Diabetes Mellitus	3 (6%)	2 (4%)	1 (14%)	0.38
Hypertension	20 (38%)	17 (39%)	3 (43%)	0.95
Active smoke	11 (21%)	8 (18%)	3 (43%)	0.05
Dyslipidemia	28 (54%)	23 (52%)	5 (63%)	0.59
ICD, primary prevention	7 (13%)	6 (14%)	1 (13%)	0.93
THERAPY (% of pts)
β-Blockers	42 (81%)	35 (78%)	7 (100%)	0.22
Dysopiramide	4 (8%)	4 (9%)	none	-
ACEi/ARB	20 (38%)	15 (33%)	5 (71%)	0.11
Ca++Blockers	5 (10%)	3 (7%)	2 (25%)	0.11
MRA	5 (10%)	3 (7%)	2 (9%)	0.11
Amiodarone	2 (4%)	2 (5%)	none	-
Diuretics	3 (6%)	2 (5%)	1 (13%)	0.37

.

At echocardiography (

Table 6. Echocardiographic characteristics of the gene (+) populations.

	Gene (+) Pts (n = 52)	Gene (+Thick) Pts (n = 45)	Gene (+Thin) Pts (n = 7)	p Value
Maximum wall thickness (mm)	19.7 ± 5.1	20.3 ± 5.1	16.5 ± 3.4	0.04
Indexed LA Volume (mL/m2)	48.1 ± 15.4	47.5 ± 15.7	51.5 ± 14.3	0.50
LVEF (%)	63.3 ± 8.6	63.9 ± 7.8	60.0 ± 12.2	0.24
E/e’	10.9 ± 4.7	10.8 ± 4.8	11.38 ± 4.0	0.74
GLS (%)	−15.92 ± 4.5	−16.32 ± 4.6	−12.8 ± 3.3	0.15
Mitral regurgitation, mild to mild-moderate	41 (79%)	35 (78%)	6 (86%)	0.31
Mild LVTO at rest or after Valsalva manoeuvre	4 (8%)	4 (9%)	none	0.81

LA = left atrium; LVEF = left ventricular ejection fraction; GLS = global longitudinal strain; LVTO = left ventricular outflow trait obstruction.

), the maximum wall thickness was lower in gene (+Thin) patients compared to those gene (+THICK).

Finally, when gene (+Thin) and gene (+THICK) were considered separately, patients with thin-filament gene variants showed a significantly worse exercise performance: not only age-adjusted peak VO₂ and O₂ pulse were lower, but also VO₂/Work was reduced compared to that observed in the group of gene (+THICK) patients (

Table 7. Cardiopulmonary test data in the gene (+) populations.

	All Gene (+) Pts (n = 52)	Gene (+THICK) Pts (n = 45)	Gene (+Thin) Pts (n = 7)	p Value
Peak RER	1.11 ± 0.12	1.12 ± 0.11	1.10 ± 0.16	0.21
Peak VO2 (mL/kg/min)	20.1 ± 8	20.9 ± 7.6	16.3 ± 2.7	0.12
Peak VO2 (% predicted)	70.5 ± 18.3	74.2 ± 15.6	58.6 ± 10.8	0.01
AT VO2 (% predicted)	49.7 ± 21.0	49.8 ± 15.3	40.9 ± 10.9	0.15
O2 pulse peak (mL/beat)	12.8 ± 3.56	13.4 ± 9.9	11.2 ± 2.3	0.60
O2 pulse peak (% predicted)	94.5 ± 11.5	97.8 ± 17.0	82.2± 9.9	0.02
VO2/Work [(mL/kg/min)]/Watts	9.3 ± 1.8	9.5 ± 1.6	8.3 ± 1.3	0.05
VE/VCO2 slope	31.3 ± 6.9	31.7 ± 6.5	30.6 ± 8.2	0.96
P(ET)CO2 (mmHg)	31.6 ± 7.7	32.0 ± 7.9	30.9 ± 9.2	0.61

). Moreover, the moderate to severe reduction in CPET performance (less than 65% of the age-adjusted peak VO₂) mentioned before was present in the majority of gene (+Thin) patients (6 out of 7 patients, i.e., 86%) and only in 12 out of 45 (+THICK) patients (27%) (p < 0.05, χ² test).

4. Discussion

This single-centre, retrospective study presents data from consecutive HCM patients on optimal medical treatment followed up in our referral outpatients’ clinic, who were asymptomatic or only slightly symptomatic for exertional dyspnoea. As a clinical strategy of our centre, they all performed CPET in addition to the recommended echocardiogram and genetic screening [2,12]. Indeed, in the last 10 years, substantial information has been collected suggesting that the cardiopulmonary test not only clarifies the pathophysiology of HCM, but also offers a prognostic insight on the progression to heart failure [19,20,21] or to the occurrence of malignant arrhythmias [18]. Surprisingly, despite its availability in many centres, its potential usefulness and its small cost, EU and US guidelines do not recommend CPET in the routine assessment of HCM patients, limiting its use for the evaluation of patients with severe symptoms and heart failure [2,12]. Our results suggest that CPET might add potentially useful information on the clinical status of HCM patients at an early stage of the disease.

Current guidelines recommend transthoracic 2D and Doppler echocardiography as the first line exam for the diagnosis and the early evaluation of the hypertrophic phenotype [2,12]. Standardised protocols for cross-sectional imaging from several projections are used to detect the presence, distribution and severity of hypertrophy, and to characterise the presence and severity of LVOT obstruction [25,26]. Based on echocardiographic data, for this study we preliminary excluded patients with reduced LV function and outflow tract obstruction to avoid the potential confounding evidence of these variables on exercise performance.

A positive genetic test in probands with HCM confirms the diagnosis of sarcomeric HCM, excluding the presence of phenocopies that could require a specific treatment. This finding is also important, along with thorough cascade screening, for the early identification of family-members at risk of developing the disease [2,6,7,8,9,10,11,12]. There is an ongoing debate on the role of genotype in risk stratification. Indeed, data from the SHaRe Registry showed that patients carrying a P/LP mutation in sarcomeric genes had a greater risk of developing arrhythmias and/or heart failure [5], but some disagreement is still present on the topic [16]. The 2022 ESC guidelines for the prevention of SCD added the genetic data as an additional risk factor that could help in the decision to implant an ICD in patients with an SCD risk score showing an intermediate risk [29]; however, the 2023 ESC guidelines on cardiomyopathies did not confirm the use of genotype in risk stratification, as the role of sarcomeric variants as a predictor of SCD, independent of SCD risk-prediction models (e.g., HCM Risk-SCD and HCM Risk-Kids), remains to be demonstrated [2]. As a matter of fact, different P/LP sarcomeric variants may define different risk profiles: variants in thin-filament genes, compared with those in thick-filament genes, seem to be associated with an increased likelihood of advanced LV dysfunction and heart failure [7,13,16] and possibly with the development of severe ventricular arrhythmias [16].

When we analysed the results of the cardiopulmonary test keeping in mind the information obtained with echocardiography and genetic analysis, we observed some peculiar patterns of response.

On the whole, the exercise capacity observed in our patients was slightly below normal; this finding agrees with most published data [17,18,19,20,21]. Of note, the ventilatory response to exercise was maintained (both VEVCO₂ slope and P_(ET)CO₂ were normal). This observation is in contrast with previous reports on greater populations: in these studies, however, subjects with heart failure were included, and this could justify the finding of an abnormal ventilatory response [17,20]. Indeed, CPET data obtained on HCM patients without heart failure are consistent with our results [18,19,21].

In the current selected population of asymptomatic or slightly symptomatic HCM patients without signs or symptoms of heart failure and without significant LVOTO, showing only a modest functional limitation, the presence of P/LP sarcomeric variants identified subjects with a significantly poorer performance at CPET: indeed, one third of these patients showed a moderate to severe reduction in CPET performance (less than 65% of the age-adjusted peak VO₂). The worst CPET results were observed when P/LP variants were located in thin-filament genes (TNNT2, TPM1 and MYL3): in this small group of patients, 86% showed a very poor exercise capability; in addition, the associated reduction in the VO₂/Work, compared to the group of gene (+THICK) patients, points to a more advanced impairment of cardiovascular efficiency in these patients.

These observations are in line with previous reports of a less favourable clinical outcome in patients with P/LP variants located in thin-filament genes, who more frequently show an evolution toward heart failure [13,14,15]. The presence of P/LP mutations (and of some of them specifically) and a slightly reduced CPET performance might thus suggest an unfavourable clinical evolution in HCM patients, regardless of similar clinical and echocardiographic characteristics.

As a pathophysiological explanation of our results, we can offer the following considerations. In HCM patients with a known genetic abnormality, mutations in β-cardiac myosin lead to a primary disease of the myocyte, causing abnormal actin–myosin interaction, increased myofilament Ca²⁺ sensitivity with an early phase of hypercontractility, altered transmembrane ion transport and adverse remodelling of the sarcoplasmic reticulum [30]. Furthermore, compared to gene-negative patients, patients with sarcomere myofilament variants have a more severe impairment of microvascular function and an increased prevalence of myocardial fibrosis [31]. These elements determine a progressive failure of energy handling and sarcolemma function, which may explain the worse exercise capability of gene (+) patients. Finally, even if hypercontractility is a shared hallmark of HCM [32], the underlying mechanisms differ between thick- and thin-filament mutations. Thick-filament HCM is primarily associated with increased ATPase activity and an elevated disordered relaxed state of myosin [33]. Conversely, thin-filament mutations initially disrupt calcium regulation: increased Ca²⁺ buffering and altered handling contribute to pathogenesis via Ca²⁺-dependent signalling pathways [34].

So, despite genetic is not yet entered in conventionally used score for risk stratification, our study together with others already published [13,14,15,16] is providing evidence that genotype-positive patients represent a subgroup of HCM patients at higher risk. Multi-centre studies will be needed to evaluate the independent predictive value of genotype, in order to support or exclude its use in risk stratification tools.

5. Conclusions

Cardiopulmonary test results in asymptomatic or slightly symptomatic patients with HCM show a reduced O₂ consumption and O₂ pulse, with an overall CPET performance slightly below normality. Noticeably, patients with P/LP mutations showed a worse exercise tolerance than gene-negative patients. Furthermore, mutations in the thin-filament genes were associated with the poorest test results.

Bearing in mind the limitations of a single-centre, retrospective study, the current results suggest that CPET should be performed in all HCM patients at their enrolment in a dedicated outpatients’ clinic, as this exam could support risk stratification and clinical management with a small additional cost, that in Italy is similar to the cost of an echocardiogram.

Moreover, these data might prompt a detailed analysis of the role of pathogenetic variants on exercise performance in large multi-centre registries.