Introduction

Hypertrophic cardiomyopathy (HCM) is the most common genetic cardiac disorder. Its prevalence is given with 0.2% in a cohort study of probands age 23–35 years.1 About 70% of the patients develop left ventricular dynamic obstruction.2, 3 In symptomatic patients despite maximal medical tolerated therapy, surgical myectomy was introduced as successful treatment option for gradient reduction even in younger patients. Since its introduction by Sigwart in 1994, percutaneous alcohol septal ablation4 achieved widespread acceptance, especially if this interventional procedure is performed in experienced centres. Actual American guidelines reserve alcohol septal ablation to adult patients with increased surgical risk due to serious co-morbidities or advanced age.5 The European guidelines underline the controversial standpoints regarding alcohol septal ablation due to the lack of long-term data on the late effects of a myocardial scar in children, adolescents, and young adults.6 Therefore, we report on the long-term follow-up after alcohol septal ablation with respect to patient's age in a large single-centre cohort.

Methods Patients

The study includes 952 consecutive patients with symptomatic hypertrophic obstructive cardiomyopathy who underwent first alcohol septal ablation (PTSMA) at Leopoldina Hospital Schweinfurt, Germany, between May 2000 and June 2017. Clinical indications were dyspnoea (New York Heart Association) and/or angina (Canadian Cardiovascular Society) Functional Class III or IV, and/or recurrent exercise-induced presyncope or syncope. Left ventricular gradients should be at least 30 mmHg at rest or 50 mmHg at provocation (Valsalva manoeuvre or post-extrasystolic beat).

We divided the cohort into three groups: young (<40 years; Group A), middle-aged (40 to <60 years; Group B), and older (≥60 years; Group C) patients. After intensive explanation of both septal reduction treatment options (surgical myectomy and PTSMA), all patients—and in all underage patients their parents—gave written consent. The study was in compliance with the Declaration of Helsinki.

Baseline examinations

All patients underwent medical history taking including family history with respect to HCM and sudden cardiac death (SCD). Baseline echocardiography was performed in each patient including outflow tract gradient measurement at rest and Valsalva manoeuvre. Electrocardiographic studies included ECG at rest and Holter monitoring in all patients. Ergospirometry could be performed in 81.7% of the patients. Invasive studies including coronary angiography, simultaneous LV gradient measurements at rest, Valsalva and post-extrasystolic beat, and LV angiography were mainly performed in one session with PTSMA.

Alcohol septal ablation technique

The technique of PTSMA was described before.7, 8 In brief, PTSMA was performed with local anaesthesia, with continuous simultaneous pressure recording of left ventricular and aortic pressure after exclusion of aortic valve gradient, and protection of a temporary pacemaker in patients without pre-implanted device [pacemaker or implantable cardioverter defibrillator (ICD)]. Intraprocedural echocardiographic monitoring and angiographic exclusion of intraseptal collateralization were mandatory. After alcohol injection, monitoring on the intensive care unit for at least 48 h or until definitive decision on the necessity of device implantation due to permanent total heart block was standard. The type of device was chosen according to the currently valid recommendations. Hospital discharge was earliest 1 week after PTSMA.

Follow-up

The patients underwent a first non-invasive follow-up control after 3 months. Subsequent cardiac examinations had been annually performed either in our institution or by the referring cardiologists. Vital status, cardiac or other clinical events, and symptomatic status compared with the pre-interventional time had been evaluated by a questionnaire. Direct telephone contact to the patients or general practitioner or referring cardiologist was performed in doubtful answers and non-returned questionnaires.

Definitions

Echocardiographic measurements were obtained following the current guidelines of the American Society of Echocardiography. Wall thickness of interventricular septum and posterior wall as well as left atrial dimension in parasternal long axis was indexed by body surface area. Left ventricular outflow tract gradients were assessed by continuous wave Doppler echocardiography at rest and at Valsalva manoeuvre.

Holter monitoring was taken for at least 24 h before ablation. According to risk stratification models, a non-sustained ventricular tachycardia (NSVT) was defined as three consecutive ventricular beats at a rate of 120 b.p.m. and <30 s in duration. Ergospirometry was performed using a ramp protocol with measurement of absolute and indexed maximal workload (watts and watts per BSA), peak O2 consumption (mL/min), and O2 consumption at anaerobic threshold (mL/min).

All-cause mortality was defined as death due to any cause. Cardiovascular death was defined as death related to any cardiovascular disease, including stroke. Cardiac death was defined as death related to any cardiac disease including SCD. SCD was defined as sudden and unexpected death within 1 h after a witnessed collapse in a previously stable patient or death that occurred during sleep.

Statistics

All data were collected in an SQL database. Statistical analysis was performed using Stata 15 (StataCorp, College Station, TX). Continuous variables were expressed as mean ± standard deviation and in addition median and inter-quartile range in case of non-normal distribution (body weight, body mass index, left atrial diameter, and left ventricular end-systolic diameter). Frequencies were given for discrete variables. Comparison of continuous variables was carried out using ANOVA or Kruskal–Wallis test. Continuous variables of two groups were compared with the unpaired Student's t-test. Paired Student's t-test was used for comparison of continuous variables at different times within one group. Categorical variables were compared using χ2 test. A P-value <0.05 was considered statistically significant. Survival estimations were analysed with Kaplan–Meier curves. Differences in survival were assessed using the log-rank test.

Results Baseline characteristics

We subdivided our study cohort of 952 patients (range 14.9–85.1 years) with first PTSMA at our centre in three age groups (Table 1). Figure 1 shows age distribution of the cohort in decades; 133 (14.0%) patients were <40 years of age at the time of PTSMA (Group A, mean age 30.3 ± 7.6 years), 422 (44.3%) patients were between 40 and <60 years (Group B, mean age 50.6 ± 5.8 years), whereas 397 (41.7%) patients were ≥60 years of age (Group C, mean age 69.7 ± 6.1 years). The small number of patients with prior septal reduction treatment either by PTSMA or myectomy was comparable.

Table 1. Baseline characteristics of 952 patients with alcohol septal ablation (PTSMA) with respect to age groups

Group A

<40 years

Group B

40 to <60 years

Group C

≥60 years

P-value Patients 133 (14.0) 422 (44.3) 397 (41.7) Women 35 (26.3) 114 (27.0) 239 (60.2)* *P < 0.00001 (C vs. A and B) Age (years) 30.3 ± 7.6 50.6 ± 5.8 69.7 ± 6.1 Height (cm) 175.4 ± 10.2 173.9 ± 9.6 166.6 ± 10.1* *P < 0.00001 (C vs. A and B) Weight (kg) 81.7 ± 17.4 87.1 ± 16.0 78.2 ± 14.8 P < 0.00001 Median (range) 82.0 (70–92) 86 (77–98) 76 (68–88) BMI (kg/m2) 26.5 ± 4.9* 28.8 ± 4.6 28.2 ± 4.7 P < 0.001 Median (IQR) 26.3 (22.9–29.9) 28.2 (25.6–31.3) 27.3 (24.8–30.6) BSA (m2) 1.96 ± 0.23 2.00 ± 0.21 1.85 ± 0.20 P < 0.001 Symptoms NYHA III/IV 91 (68.4) 286 (67.8) 321 (80.9) P < 0.001 Angina pectoris 59 (44.3) 215 (50.9) 217 (54.8) n.s. Syncope Unexplained 13 (9.8) 37 (8.8) 43 (10.8) n.s. Effort induced 59 (44.3) 155 (36.8) 174 (33.8) n.s. Palpitations 45 (33.8) 135 (32.0) 113 (28.5) n.s. Family history Hypertrophic CM 67 (50.4)* 107 (25.4) 53 (13.4) *P < 0.00001 (A vs. B and C) Sudden cardiac death 23 (17.3)* 47 (11.1) 28 (7.0) *P < 0.01 (A vs. B and C) Smoker 47 (35.3) 142 (33.7) 65 (16.4)* *P < 0.0001 (C vs. A and B) Cardiac diseases Hypertension 19 (14.3)* 213 (50.6) 284 (71.5) *P < 0.00001 (A vs. B and C) Coronary artery disease 1 (0.8) 39 (9.3) 83 (20.9) P < 0.00001 Atrial fibrillation 4 (3.0) 56 (13.3) 73 (18.4) P < 0.001 Paroxysmal 4 (3.0) 49 (11.6) 60 (15.1) Permanent 0 (0) 7 (1.7) 13 (3.3) Medication n.s. Beta-blocker 94 (70.7) 280 (66.4) 264 (66.5) Verapamil 36 (27.1) 94 (22.3) 115 (29.0) Disopyramide 4 (3.0) 6 (1.4) 3 (0.8) Prior septal reduction Alcohol septal ablation 8 (6.3) 17 (4.1) 16 (4.2) n.s. Myectomy 3 (2.4) 7 (1.7) 10 (2.6) n.s. Prior device therapy Pacemaker 6 (4.7) 16 (3.9) 30 (7.8) n.s. ICD 25 (19.7) 42 (10.1) 23 (6.0) P < 0.0001 Holter Sinus rhythm 122 (96.1) 392 (94.5) 352 (90.0) P < 0.01 Atrial fibrillation 0 (0) 11 (2.7) 29 (7.5) P < 0.001 SV tachycardia 6 (4.7) 42 (10.1) 65 (16.6) P < 0.01 Non-sustained VT 10 (7.9) 47 (11.3) 41 (10.5) n.s. Ergospirometry Work capacity (W) 134.7 ± 44.3 124.4 ± 44.1 92.6 ± 36.5 P < 0.00001 Work capacity (W/kg) 1.7 ± 0.5 1.5 ± 0.5 1.2 ± 0.4 P < 0.00001 Peak VO2 (mL/min/kg) 23.0 ± 6.2 20.7 ± 5.5 17.6 ± 4.6 P < 0.00001 VO2 at AT (mL/min/kg) 15.3 ± 4.7 13.7 ± 4.2 13.0 ± 3.9 P < 0.00001 AT, anaerobic threshold; BMI, body mass index; BSA, body surface area; CM, cardiomyopathy; ICD, implantable cardioverter defibrillator; IQR, inter-quartile range; NYHA, New York Heart Association; SD, standard deviation; SV, supraventricular; VO2, oxygen consumption; VT, ventricular tachycardia. Values are given in mean ± SD, median and IQR (in non-normal distribution), and n (%).

Distribution of age (in decades) in 952 patients at the time of alcohol septal ablation (PTSMA).

The old patients suffered more often from dyspnoea New York Heart Association Class III/IV (80.9%; P < 0.001 each) compared with patients of Group A (68.4%) and Group B (67.8%). Differences in other cardiac symptoms like angina and exercise-induced as well as unexplained syncope were not found.

Young patients reported more often family history of HCM (50.4%; P < 0.00001, each) and SCD (17.3%; P < 0.01, each) compared with patients of Group B (HCM 25.4% and SCD 11.1%) and Group C (HCM 13.4% and SCD 7.0%). In contrast to comparable number of pre-interventional pacemaker implantation in all groups, young patients had more often a pre-implanted ICD (19.7%, P < 0.0001 each) compared with patients of Group B (10.1%) and Group C (6.0%). Pre-interventional Holter monitoring showed no differences in incidence of NSVT. A history of atrial fibrillation was less commonly reported by young patients (3.0%; P < 0.001, each) compared with patients in Group B (13.3%) and Group C (18.4%).

Not unexpected, the incidences of systemic hypertension (Group A 14.3% vs. Group B 50.6% vs. Group C 71.5%; P < 0.00001 between all age groups) and coronary artery disease (Group A 0.8% vs. Group B 9.3% vs. Group C 20.9%; P < 0.00001 between all age groups) were significantly higher with increasing age. Comparable number of patients reported on smoking in Group A (35.3%) and Group B (33.7%), whereas smoking was less often seen in Group C (16.4%; P < 0.0001, each).

Acute results and hospital course

According to larger maximal septal thickness (Table 2), we injected a larger amount of alcohol in younger patients of Group A (2.4 ± 0.7 mL; P < 0.00001, each) compared with patients of Group B (2.0 ± 0.4 mL) and Group C (2.0 ± 0.3 mL). Older patients of Group C had less maximal CK rise (820 ± 394 U/L; P < 0.05, each) compared with patients of Group A (916 ± 446 U/L) and Group B (908 ± 574 U/L).

Table 2. Baseline echocardiographic measurements of 952 patients with alcohol septal ablation (PTSMA) with respect to age groups

Group A

<40 years

Group B

40 to <60 years

Group C

≥60 years

P-value Patients 133 (14.0) 422 (44.3) 397 (41.7) Maximal IVS thickness (mm) 23.9 ± 5.6* 20.9 ± 4.0 20.1 ± 3.5 *P < 0.00001 (A vs. B and C) Maximal IVS/BSA (mm/m2) 12.4 ± 3.5* 10.5 ± 2.3 10.9 ± 2.2 *P < 0.00001 (A vs. B and C) Subaortic IVS thickness (mm) 21.2 ± 4.8 19.8 ± 3.8 18.9 ± 3.5 P < 0.00001 Subaortic IVS/BSA (mm/m2) 11.0 ± 2.8 9.9 ± 2.1 10.3 ± 2.1 P < 0.001 LVPW thickness (mm) 13.6 ± 3.7 13.4 ± 2.8 12.9 ± 2.7* *P < 0.05 (C vs. A and B) LVPW/BSA (mm/m2) 7.1 ± 2.1 6.7 ± 1.5 7.0 ± 1.6 P < 0.05 LVEDD (mm) 42.1 ± 5.7 44.3 ± 6.7 43.0 ± 6.4 P < 0.01 LVEDD/BSA (mm/m2) 21.6 ± 3.2 22.3 ± 3.5 23.3 ± 3.8 P < 0.001 LVESD (mm) 22.1 ± 5.4 23.7 ± 5.7 23.3 ± 5.8 n.s. Median (IQR) 22 (19–26) 23 (19–27) 23 (19–27) LVESD/BSA (mm/m2) 11.3 ± 2.8 11.9 ± 2.9 12.6 ± 3.2* P < 0.001 (C vs. A and B) LA diameter (mm) 46.0 ± 7.5 46.6 ± 6.6 45.5 ± 6.7 n.s Median (IQR) 46 (41–51) 47 (42–51) 45 (41–49) LA/BSA (mm/m2) 23.5 ± 4.1 23.5 ± 3.6 24.6 ± 4.0* P < 0.001 (C vs. A and B) LVOT Doppler gradients Rest (mmHg) 69.7 ± 34.7 59.5 ± 35.3 66.5 ± 41.7 P < 0.01 Valsalva (mmHg) 98.8 ± 40.1* 102.8 ± 40.1 108.5 ± 48.6 *P < 0.05 (A vs. C) BSA, body surface area; IQR, inter-quartile range; IVS, intraventricular septum; LA, left atrial; LVEDD, left ventricular end-diastolic diameter; LVESD, left ventricular end-systolic diameter; LVOT, left ventricular outflow tract; LVPW, left ventricular posterior wall; SD, standard deviation. Values are given in mean ± SD, median and IQR (in non-normal distribution), and n (%).

Echocardiographic gradient reduction at hospital discharge (Table 3) was less in young patients of Group A [45.3 ± 30.7 mmHg at rest (P < 0.001 each) and 68.3 ± 38.0 mmHg at Valsalva (P < 0.001 each)] compared with patients of Group B (31.7 ± 28.2 mmHg at rest and 54.5 ± 39.3 mmHg at Valsalva) as well as Group C (31.6 ± 30.3 mmHg at rest and 54.8 ± 43.0 mmHg at Valsalva).

Table 3. Hospital course of 952 patients with PTSMA with respect to age groups

Group A

<40 years

Group B

40 to <60 years

Group C

≥60 years

P-value Patients 133 (14.0) 422 (44.3) 397 (41.7) Angiographic LVEF (%) 71.4 ± 10.5 73.1 ± 8.4 72.4 ± 10.0 n.s. Invasive LV gradients Rest (mmHg) 62.2 ± 34.5* 49.3 ± 38.0 49.7 ± 38.8 *P < 0.001 (A vs. B and C) Valsalva (mmHg) 87.4 ± 37.9* 92.7 ± 36.1 99.9 ± 40.8 *P < 0.001 (C vs. A and B) Post-extrasystole (mmHg) 128.3 ± 42.5* 132.8 ± 47.3 139.1 ± 51.9 *P < 0.05 (A vs. C) More than 1 branch treated 5 (3.8)* 5 (0.9) 2 (0.5) *P < 0.05 (A vs. B and C) Alcohol injected (mL) 2.4 ± 0.7* 2.0 ± 0.4 2.0 ± 0.3 *P < 0.00001 (A vs. B and C) Maximal CK rise (U/L) 916 ± 446 908 ± 574 820 ± 394* *P < 0.05 (C vs. A and B) Third-degree AV block at any time 38 (28.6)* 166 (39.3) 173 (43.6) *P < 0.05 (A vs. B and C) Echo gradients at discharge Rest (mmHg) 45.3 ± 30.7* 31.7 ± 28.2 31.6 ± 30.3 *P < 0.001 (A vs. B and C) Valsalva (mmHg) 68.3 ± 38.0* 54.5 ± 39.3 54.8 ± 43.0 *P < 0.01 (A vs. B and C) Complications Death 1 (0.8) 0 1 (0.3) n.s. Permanent pacemaker 5 (3.8)* 39 (9.2) 56 (14.1) *P < 0.01 (A vs. B and C) Pericardial effusion 1 (0.8) 6 (1.4) 17 (4.3) n.s. AV, atrioventricular; CK, creatine kinase; LV, left ventricular; LVEF, left ventricular ejection fraction; SD, standard deviation. Values are given in mean ± SD and n (%).

One patient of Group A died at Day 3 after PTSMA due to pulmonary embolism, and one patient of Group C died at Day 33 due to pneumonia. A 47-year-old woman got a dissection of the left main coronary artery at the first attempt, which could be fixed with a stent followed by successful PTSMA 6 months later.

Conduction abnormalities were less often observed in younger patients of Group A (Table 2). Furthermore, temporary total heart blocks during PTSMA were more often seen in patients of Group C (43.6%) and Group B (39.3%) compared with the young patients (28.6%; P < 0.05, each). Consequently, young patients required less often implantation of a permanent pacemaker due to ongoing total heart block (3.8%; P < 0.01, each) compared with patients of Group B (9.2%) and Group C (14.1%).

Follow-up results

Follow-up was longer in younger patients (7.4 ± 5.5 years) and patients in Group B (6.5 ± 5.1 years) compared with Group C (5.6 ± 4.8 years; P < 0.001, each) (Table 4). The proportions of living patients who reported clinical improvement were comparable in all groups (94% in Group A vs. 93.8% in Group B vs. 95.0% in Group C).

Table 4. Follow-up of 952 patients with PTSMA with respect to age groups

Group A

<40 years

Group B

40 to <60 years

Group C

≥60 years

P-value Patients 133 (14.0) 422 (44.3) 397 (41.7) Follow-up (years) 7.4 ± 5.5 6.5 ± 5.1 5.6 ± 4.8* *P < 0.001 (C vs. A and B) Clinical symptoms n.s. Improvement 125 (94.0) 396 (93.8) 377 (95.0) No change 6 (4.5) 18 (4.3) 10 (2.5) Worsening 1 (0.8) 5 (1.2) 2 (0.5) Echo gradients at last follow-up Rest (mmHg) 23.7 ± 24.9* 16.4 ± 18.1 17.7 ± 21.4 *P < 0.001 (A vs. B and C) Valsalva (mmHg) 34.3 ± 33.5 30.9 ± 30.6 33.5 ± 36.4 n.s. Maximal IVS thickness (mm) 21.1 ± 5.4* 18.2 ± 4.0 17.5 ± 3.6 *P < 0.001 (A vs. B and C) Holter NSVT 11 (10.9) 30 (9.0) 24 (8.3)