Is surgery the gold standard in the treatment of obstructive hypertrophic cardiomyopathy?

Abstract

OBJECTIVES

Hypertrophic cardiomyopathy is a complex and relatively common genetic cardiac disease and has been the subject of intensive scrutiny and investigation for over 40 years. The aim of this non-randomized cohort study was to compare subjective and objective outcomes in hypertrophic cardiomyopathy patients undergoing drug therapy, surgical myotomy-myectomy, dual-chamber pacing and alcohol septal ablation.

METHODS

We examined 194 patients: 103 with non-obstructive hypertrophic cardiomyopathy and 91 with obstructive hypertrophic cardiomyopathy. All the patients with a non-obstructive form were on drug therapy. Ninety-one consecutive patients with drug-refractory obstructive hypertrophic cardiomyopathy were treated invasively. Dual-chamber pacemaker implantation was performed for 49 patients with previous positive temporary pacing test (Group 1). In 28 patients with massive left ventricle hypertrophy and obliteration of its cavities, extensive myotomy-myectomy was performed (Group 2). In 14 patients with midventricular obstruction and appropriate coronary anatomy, alcohol septal ablation was performed (Group 3).

RESULTS

The peak left ventricle outflow tract gradient was 84.1 ± 15.2 mmHg in Group 1, 113.3 ± 14.9 mmHg in Group 2 and 97.5 ± 8.9 mmHg in Group 3. Dual-chamber pacing in Group 1 with optimal atrio-ventricular delay (85–180 ms for atrium pacing and 45–120 ms for atrial sensing) leads to dramatic decreases in left ventricle outflow tract gradient to 17.6 ± 11.8 mmHg and degree of mitral regurgitation. After extensive myectomy in Group 2, we observed a reduction of left ventricle outflow tract gradient to 17.3 ± 10.2 mmHg. Septal alcohol ablation in Group 3 leads to a left ventricle outflow tract gradient decrease from 97.5 ± 8.9 to 25.3 ± 5.8 mmHg.

CONCLUSIONS

Surgical myectomy, dual-chamber pacing and alcohol septal ablation are equally effective in reducing obstruction in case of correct indications. Dual-chamber pacing is indicated in functional reversible states characterized by excitation delay. Alcohol septal ablation is preferable in cases with midventricular obstruction and appropriate coronary anatomy. Surgical methods are indicated in anatomical irreversible changes and remain the gold standard for obstructive hypertrophic cardiomyopathy treatment.

INTRODUCTION

Today, a lot of centres use different approaches to the treatment of hypertrophic cardiomyopathy, including medical treatment and invasive strategies. The main purpose of this study was to investigate different methods of treatment and evaluate their effectiveness in patients with an obstructive form of hypertrophic cardiomyopathy (HCMP).

MATERIALS AND METHODS

The cohort study comprised 194 patients (115 males) with a mean age of 48. 7 ± 12.6 years, who were consulted and treated at the National Institute of Cardiovascular Surgery Academy of Medical Sciences of Ukraine, Kiev, Ukraine, between 1993 and 2011. All the patients with a non-obstructive form of the disease underwent drug therapy. Ninety-one consecutive patients with a mean age of 34.54 ± 18.5 years (44 males) with a drug-refractory form were treated invasively and constituted the major patient group in our study.

We used different treatment strategies and modalities depending on the stage of HCMP. For selection of an adequate method of treatment, apart from general clinical methods of studies, the following instrumental methods were used: electrocardiography, a complex echocardiography, coronaroventriculography, haemodynamic study (base condition, during isoproterenol provocation, during temporary dual-chamber pacing), Holter electrocardiography monitoring and magnetic resonance imaging. The complex echocardiography included biplane echocardiography, and continuous and impulse Doppler and coloured Doppler imaging. The baseline patient clinical characteristics are summarized in Table 1.

Table 1:

The baseline characteristics of the study subjects

	Myectomy group (n = 28)	Septal ablation group (n = 14)	DDD group (n = 49)
Age (years)	29.1 ± 16.1	34 ± 8.5	37.8 ± 21.2
Male sex	13	8	23
Drugs (%)	100	100	100
β-Blockers (%)	60	71	70
CCA (disopyramide) (%)	25	28.5	16
Amiodarone (%)	35	28.5	35
Permanent AF	4	0	0
Paroxysmal AF	8	1	12
NYHA class I	8	1	15
NYHA class II	14	10	23
NYHA class III	4	3	9
NYHA class IV	2	0	2
Exertional chest pain (%)	64	78	60
Syncope (%)	42	50	40
LVOTG (mmHg)	113.3 ± 14.9	97.5 ± 8.9	84.1 ± 15.2

CCA: calcium channel antagonist; AF: atrial fibrillation; NYHA: New York Heart Association; LVOTG: left ventricular outflow tract gradient.

Temporary dual chamber pacing test

All patients with left ventricular outflow tract obstruction underwent acute haemodynamic study with the real-time direct measurement of systolic pressure gradient (SPG) on LVOT during temporary pacing test in AAI, VDD and DDD modes with different atrioventricular delays (AVD).

Endocardial left ventricle activation sequence mapping

Twenty patients with the non-obstructive form and 91 patients with hypertrophic obstructive cardiomyopathy (HOCM) underwent invasive electrophysiology study with the aim of measuring the time interval between His potential and left ventricle (LV) apex, hypertrophy zone and a lateral wall. We used a steerable electrophysiologic (EP) catheter with a transaortic approach for endocardial LV mapping.

We performed left ventricle endocardial electroanatomical mapping using the CARTO XP system on 3 patients with HOCM and 3 patients with a non-obstructive form of HCMP.

Echocardiography

To determine the anatomical form of HOCM, systolic pressure gradient and the structural characteristics of subvalvular valve structures, 2D, M-mode and Doppler echocardiography was performed in all patients prior to their intervention and at regular follow-up. Standard views for M-mode and 2D studies were obtained, and conventional techniques were used for sizing the left ventricle. Left ventricle outflow tract gradients were calculated using the Bernoulli equation.

Methods of treatment

Drug therapy included calcium channel blockers in 48 patients (26%), β-blockers in 120 patients (65%) and amiodarone in 57 patients (30%).

Dual-chamber pacemaker implantation [1–3] was performed on 49 patients with previous positive temporary DDD pacing test (Group 1). In 28 patients with massive LV hypertrophy and its cavity obliteration, extensive myotomy-myectomy was performed (Group 2). In 14 patients with midventricular obstruction and appropriate coronary anatomy, alcohol septal ablation was performed (Group 3). We used the standard alcohol ablation method [4–7] in 14 patients and the classical Morrow technique [8–10] in 28 patients.

Statistical analysis

Data are shown as mean ± 1 SD. Comparison of systolic pressure gradient before and after operations was performed with Student's t-test. A P value of <0.05 was considered statistically significant. We evaluated the differences in NYHA class before and after the operations and analysed them with the non-parametric (Mann–Whitney) test. A P value of <0.05 was considered statistically significant.

RESULTS

Of 194 patients involved in our study, 91 had an obstructive form of the disease with evident clinic symptoms.

The baseline characteristics of study subjects (patient with HOCM) are shown in Table 1.

We examined this group of patients for the mechanism of obstruction and defined two main components: functional and anatomic.

The results of eject fraction mapping are presented in the Table 2, which proves a direct correlation between the evidence of systolic gradient, contraction delay rate of LV apex and degree of mitral insufficiency. Most patients with high-pressure gradient have a later LV apex than LVOT with respect to activation time. Among the patients with low gradient, the LV apex took place earlier than LVOT in activation time.

Table 2:

Correlation between the systolic pressure gradient and LVA pre-excitation and mitral regurgitation obtained during haemodynamic and EP study

Number of patients	LVA pre-excitation (ms)	Systolic pressure gradient	(mmHg) Mitral regurgitation degree
9	−20 to 0	5 + 0.6	0
11	0–15	7 + 1.2	mild
24	16–30	28.7 + 2.5	Mild–moderate
57	31–50	66.7 + 9.6	Moderate–severe

LVA: left ventricle apex; EP: electrophysiology.

In the group of 6 patients who underwent CARTO XP (electro-anatomical mapping of LV), we observed considerable differences in LV spatial activation between patients with and without pressure gradient.

In Fig. 1a, we can see the electro-anatomical LV mapping of a patient without a significant pressure gradient on LVOT, but with the marked hypertrophy of interventricle septum (IVS). LV activation comes from the apex and advances the LVOT zone with respect to time. In Fig. 1b, we can see the electro-anatomical LV mapping of a patient with a significant pressure gradient on LVOT and with hypertrophy of IVS. In this case, LV activation comes from the septum and LV apex excitation occurs later than the excitation of the LVOT zone. The results of electrophysiology mapping confirmed our hypothesis about the relationship between SPG value and apex contraction delay.

Figure 1:

Carto XP electroanatomical mapping of left ventrical (a) on patient without a significant pressure gradient on LVOT (thickness of IVS = 3.7 сm; thickness of posterior wall LV = 0.8 сm, ΔР_с = 11 mmHg) and (b) on patient with a significant pressure gradient on LVOT (thickness of IVS = 2.0 сm; thickness of posterior wall LV = 1.3 сm, ΔР_с = 90 mmHg).

Thus, abnormalities of consecutive LV activation are considered the leading mechanism of LVOT obstruction development.

Results of surgery

The mean length of follow-up in the surgery group was 45.4 ± 4.2 months, 18.5 ± 3.2 months in the septal ablation group and 68.2 ± 6.6 months in the DDD pacing group. By the end of follow-up, the peak LVOT gradient fell from 113.3 ± 14.9 to 17.3 ± 10.2 mmHg in the surgery group, from 97.5 ± 8.9 to 25.3 ± 5.8 mmHg in the septal ablation group and from 84.1 ± 15.2 to 17.6 ± 11.8 mmHg in the DDD pacing group (Table 3).

Table 3:

Follow-up in surgery, alcohol ablation and DDD pacing group

Methods of treatment	Number of patients	SPG before procedures (mmHg)	SPG after procedure(mmHg)	P-value	Mortality		Follow-up, month
					Hospital	Late
DDD pacing	49	84.1 ± 15.2	17.6 ± 11.8	0.001	0	3	68.2 ± 6.6
Alcohol ablation	14	97.5 ± 8.9	25.3 ± 5.8	0.001	0	0	18.5 ± 3.2
Surgery	28	113.3 ± 14.9	17.3 ± 10.2	0.001	1	2	45.4 ± 4.2

SPG: systolic pressure gradient.

There was one hospital death in the surgical group due to heart failure (myocardial ischaemia as a result of myocardial bridge), and there were two cardiac deaths in late follow-up. There were two late deaths in the DDD pacing group due to sudden cardiac deaths and one late death due to pulmonary artery thrombosis.

Following surgery, the NYHA functional class improved from 2 ± 0.86 to 1.6 ± 0.79 (P = 0.06). Following septal ablation group, the NYHA functional class improved from 2.14 ± 0.53 to 1.71 ± 0.47 (P = 0.04). Following the DDD pacing group, the NYHA functional class improved from 1.9 ± 0.8 to 1.4 ± 0.57 (P = 0.0004) (Fig. 2).

Figure 2:

NYHA class changes following (a) myectomy, (b) septal ablation and (c) DDD pacing.

DISCUSSION

We have used a differentiated approach to treatment in compliance with our assumption that HCMP is a progressive, staged and genetically caused disease.

This study demonstrates that in patients with drug-refractory symptomatic obstructive HCMP, surgical myectomy, trans-coronary alcohol septal ablation and DDD pacing are equally effective at reducing left ventricular outflow tract obstruction and lead to similar subjective improvements in functional capacity. Surgical myectomy, however, was more effective in improving the NYHA functional class.

We consider that HCMP has three major pathogenetic mechanisms: hypertrophy of the myocardium, electro-mechanical disturbance of spatial activation and contraction and pathology of valvular and chordal apparatus of the mitral valve.

Allowing for the regular dependence of LVOT SPG and mitral regurgitation on the LV excitation sequence and LV apex pre-excitation efficacy with DDD pacing, we think we can suggest a logical explanation for LVOT obstruction in HCMP and physiologically substantiate methods of its treatment. The genetically conditioned changes of myocyte and myofibril orientation provoke the primary hypertrophy of the basal part of IVS initially narrowing LVOT and changing normal sequence of LV excitation. There appears to be a delay in LV apex excitation and papillary muscles holding the anterior mitral valve (MV) cusp. One can assume that systolic anterior motion of MV occurs as a result of papillary muscle systolic contraction delay, relaxation of chords and the resulting flotation of the free fringe of the anterior MV cusp. The anterior MV leaflet devoid of chord control, which have not had time to stretch because of papillary muscle delay at the onset of LV systole activation, is carried by the bloodstream into the LVOT area (systolic anterior motion (SAM) of MV) and as a result of the ‘sail effect’, blocks the entry into the aorta causing SPG in LVOT and MV anterior leaflet prolapse and brings about MV opening and regurgitation. On the other hand, the LV apex excitation delay leads to earlier contraction of the IVS basal part and that induces narrowing of the LVOT lumen. The LVOT narrowing and mitral regurgitation increase the LV muscle working load and trigger its secondary hypertrophy. It aggravates its own causes and LVOT narrowing keeps on going down to total obstruction. The effective SPG decrease with LV apex pre-excitation only and lack of better DDD pacing effect with further AVD shortening prove this assumption. LV apex pre-excitation with DDD pacing leads to earlier excitation of papillary muscles preventing the MV anterior cusp from floating and sticking into the LVOT lumen that prevents the occurance of SAM and its consequences, SPG and mitral regurgitation.

In our research, we can attribute the positive effect from implantation of pacemakers to the fact that a pacemaker was implanted only in cases of positive results from temporary DDD stimulation. This method is very important when selecting patients.

Thus, it allows us to say that use of DDD pacing, as a method of treatment, is possible in early stages of the disease.

In spite of common opinion about the ineffectiveness of DDD pacing in treatment of HOCM our long experience of observing patients with implanted pacemakers showed that only 4 patients (8.1%) moved to a group of surgical treatment and required surgery as such. At the same time, other patients from this group had good short-term and long-term results.

In accordance with our assumption, we used alcohol septal ablation in cases of evident SPG without the electromechanical pathogenesis component and pathologic changes in the mitral valve apparatus. Surgical methods of treatment are preferable in cases of evident SPG and high rate of mitral regurgitation.

A large group of patients were treated with dual-chamber stimulation due to the fact that 15 years ago we had the following original hypothesis: the main pathological mechanism in the early stage of HCMP is electro-mechanical disturbance in spatial activation and contraction of LV. At the same time, there are a number of randomized studies showing the low efficiency of dual-chamber pacing. It should be noted that there was no differential approach to patient selection. Over the last 3 years we have developed a differentiated approach according to the method of treatment chosen. We have developed the following indications for dual-chamber stimulation: (i) Reduced pressure gradient between the aorta and the apex of the left ventricle of more than 30% of the original and a residual gradient of less than 50 mmHg. (ii) The presence of coronary anatomies that are not favourable for alcohol ablation. (iii) The presence of contraindications for surgery, using cardiopulmonary bypass.

Thus, on the basis of the results obtained, we can say that every method of treatment has its clearly defined indications. We had good results with patients who received precise indications for a certain method of treatment. In the retrospective analysis, we noticed that we had not received expected results in cases when indications for treatment methods were chosen incorrectly. This phenomenon can account for the absence of positive effects from DDD in the works of several researchers who did not use a differentiated approach to treatment and did not consider staging of the disease.

Limitations of the study

This study was not designed as a prospective investigation, but rather as a retrospective analysis of the three treatment modalities. Patients were assigned to the three treatment modalities on the basis of patient choice and physician guidance and so the two groups may not be comparable. There were a small septal ablation group, a different follow-up period and a small group of patients who underwent electro-anatomical mapping. There was a relatively short follow-up time in the surgery group.

CONCLUSION

HOCM is a multistage pathology with progressive development.

The mechanism of obstruction formation on the LVOT and mitral regurgitation consists of the delay of the apical part excitation of the left ventricle relative to the LVOT. Surgical methods, DDD pacing and alcohol septal ablation are equally effective in reducing obstruction in case of correct indications and have to be used when taking into account a concrete stage of HCMP. Surgical methods are indicated in anatomical irreversible changes and remain the gold standard for the treatment of obstructive hypertrophic cardiomyopathy.

Conflict of interest: none declared.

APPENDIX. CONFERENCE DISCUSSION

Dr E. Delmo-Walter (Berlin, Germany): I read your manuscript with great interest and have several issues and concerns to raise. You seem to have a very clear pattern of decision-making in which patients belong to these four management strategies, but then, still, you mention in your paper that it's the patient's choice to be in these groups. So it doesn't fit together. Would you please explain your criteria for randomizing, or not randomizing, patients to alcohol septal ablation, dual chamber pacing or septal myectomy?

Dr Rudenko: It's a very important issue, because 15–17 years ago we were deeply interested in electrophysiological disorders, and we applied DDD pacing for patients with such disorders. But lately we have been using different methods of treatment. We have a protocol for this cohort of patients and normally we use temporary DDD pacing during angiocardiography. If we observe positive results after DDD pacing, we implant a pacemaker. If the patient with mild hypertrophy and without severe mitral regurgitation has appropriate coronary anatomy, we use alcohol septal ablation. If a patient has hypertrophy and cavity obliteration with severe mitral regurgitation, we use different methods of surgical intervention.

Dr Delmo-Walter: But was there any crossover from your management strategies, such as patients who underwent septal ablation eventually underwent myectomy, or vice versa?

Dr Rudenko: Unfortunately, we don't have a big experience in alcohol septal ablation, but we have received positive results from the group of 10 patients who underwent this procedure. Thus, in the case of mild ventricular obstruction and appropriate coronary anatomy, we can use this method of treatment. Patients with evident hypertrophy and severe mitral regurgitation require surgery, because septal ablation is not an appropriate method.

Dr Delmo-Walter: So there was a crossover?

Dr Rudenko: Yes.

Dr Delmo-Walter: Okay. And the second concern I have is that you spend some time discussing systolic anterior motion, which is very important in hypertrophic obstructive cardiomyopathy, because, as you said, the large mitral leaflet can cause obstruction to the dynamic flow. And because of this, a lot of surgeons perform myectomy with additional mitral valve repair, or plication leaflet plasty, and in this case in our institution we do anterior leaflet retention plasty. However, you never mentioned any procedures, or any mitral valve repair techniques, that you use in your service in the 35 patients who underwent surgery.

Dr Rudenko: Lately we have been using extensive myotomy, but we don't have much experience in mitral repair. There are four patients who underwent extensive myotomy with the Alfieri technique. I would like to emphasise that our study is still ongoing and we will try to present our results concerning this group of patients later.

Dr Delmo-Walter: So perhaps in your intraoperative assessment you don't see systolic anterior motion, if ever you do intraoperative assessment of the systolic anterior motion. Maybe that's why you only had several patients who underwent Alfieri stitch or Alfieri repair and four patients who underwent mitral valve replacement. Anyway, it's very interesting.

Dr Rudenko: I would like to add one more fact. It's very important that some patients who received drug treatment and a pacemaker moved to another treatment group. Thus, we are able to consider hypertrophic cardiomyopathy as a multistage disease.

Dr Delmo-Walter: That's what I was trying to ask you earlier, whether you have had a crossover from patients who underwent DDD pacing and then underwent myotomy eventually, because of maybe a residual gradient or things like this, and that's very important.

Dr Rudenko: We have three patients from the DDD pacing group who moved to surgery and alcohol septal ablation.

Dr Delmo-Walter: I think it's very important that you discuss these issues when you submit your final manuscript.