Is alcohol ablation of the septum associated with recurrent tachyarrhythmias ?

Questions under study: Alcohol ablation (AA) of the septum has been introduced as new therapy in hypertrophic cardiomyopathy (HCM). It was feared that iatrogenic myocardial infarction due to AA may induce re-entry tachyarrhythmias and increase sudden cardiac death. Methods and results: Twenty-four patients (mean age 52 years) underwent successful AA. Clinical follow-up (FU) ranged from .3 to .7 years (mean 2.8). One patient died (suicide) 4 years after AA. Left ventricular (LV) outflow gradient (peak-to-peak) decreased (median) after AA from 43 (IQR 25 to 4) mmHg to 1 (IQR to 12) mmHg (rest) (p < . 1) and from 13 (IQR 75 to 1 5) mmHg to 13 (IQR to 31) mmHg (postextrasystolic) (p < . 1). Transient AV block occurred in 22% (5/24) necessitating temporary pacing. A permanent pacemaker was implanted in 4% (1/24). NYHA-class was 2.5 (IQR 2. to 3. ) before and 1.5 (IQR 1.3 to 2. ) (p < . 1) after AA. During FU, 2 pacemakers were implanted due to bradycardia (no AV block). A right bundle branch block was found in 13% (2/24) before and 4 % (11/24) after AA (p = . 3). Non-sustained ventricular tachycardia (NSVT) was observed in 13% (2/1 ) before and 22% (5/23) (p = .4 ) after AA. Two patients required ICD implantation. Conclusions: Long-term FU is excellent in HCM after AA. The pressure gradient drops below 25 mm Hg in 95% (23/24) of all patients. Transient AV block occurs in 22% (5/24), but permanent pacemaker implantation is rarely needed (13%, 2/24). Severe NSVT occurs in 13% (2/1 ) before and 22% (5/23) after AA but ICD implantation is only occasionally required.


Introduction
No financial support declared.with symptomatic improvement and reduction in cardiovascular mortality.More recently, alcohol ablation of the septum with injection of small amounts of alcohol into the first, second or third septal branch was introduced in the late nineties [3].Alcohol ablation can be done percutaneously without the need of open heart surgery [3-].A word of caution has been raised that iatrogenic septal myocardial infarction might cause re-entry tachyarrhythmia [7].Previous clinical studies have not indicated an increase in arrhythmogenic substrate but the purpose of the present study is to assess cardiac arrhythmia before and after alcohol ablation in a small group of carefully examined patients with hypertrophic obstructive cardiomyopathy.

Patient population
Tw enty-four patients (mean age 52±1 years, 8 women, 1 men) with hypertrophic obstructive cardiomyopathy were included in the present analysis.Patients were selected from a total group of 54 patients with hypertrophic cardiomyopathies seen between 1999 and 25 at our institution (screening).Genetic screening for determination of the mutations was not done except in one case, which was a family form with severe HOCM.All underwent successful alcohol ablation.Patients were clinically followed for a mean of 2.8 years (4 months to .7 years).Body mass index (BMI) was 2.1±5.Clinical symptoms were: angina pectoris in 71% (17/24), dyspnoea in 92% (22/24), presyncopes and syncopes in 58% (14/24) respectively 4% (11/24) as well as palpitations in 5% (12/24) (table 1).Due to supraventricular arrhythmias or coronary artery disease every 2nd patient was treated with beta-blockers or calcium channel-blockers and every 3rd with acetylsalicylic acid (table 2).Inclusion criteria: patients with symptomatic HOCM (NYHA-class ≥2), refractory to medical therapy and a septal/posterior wall ratio ≥1.5 as well as a pressure gradient at rest ≥3 mm Hg and after provocation ≥ mm Hg.Amiodarone therapy had to be stopped >2 months and beta blockers >24 hours prior to the procedure.Yo ung age <1 years, pregnancy, AV-block 2 or 3 unless a pacemaker was implanted, and trifascicular block were excluded from the study.Symptoms.
Echocardiography: Tr ansthoracic echocardiography was performed in the parasternal long and short axis as well as apical two and four chamber views.The 2D images provided septal and posterior wall thickness, left ventricular (LV) mass, left atrial diameter, ejection fraction (EF), LV systolic and end-diastolic diameter (LVEDD).The magnitude of LVOT gradient at rest and during provocation with amylnitrate was determined from Doppler echocardiography.Systolic anterior motion (SAM) of the anterior mitral leaflet was detected by 2D echocardiography.Mitral regurgitation was estimated from pulsed Doppler and/or continuous wave Doppler echocardiography.
Exercise testing: 14 of 24 patients underwent treadmill exercise testing.Heart rate at rest and during maximal exercise, working capacity, exercise duration and ECG changes were recorded.
Holter ECG: In all patients Holter ECG was recommended but not carried out in 8 (refused by the patient or the treating physician). 1 of 24 patients were monitored over 24 hours to assess heart rate and arrhythmias using a CardioDay ® Ve rsion 1.9 (getemed AG, Germany).Specifically, heart rate variability, atrial and ventricular events and tachy-/bradyarrhythmias were determined.

Alcohol ablation
A temporary pacemaker was placed in the right ventricle.After diagnostic coronary angiography and left ventricular (LV) ventriculography, a F guiding catheter was placed in the ostium of the left main coronary artery as well as a 4F pigtail catheter (2nd puncture) in the apex of the LV for simultaneous pressure gradient measurements.Next a standard guide wire (Guidant Corp., Santa Clara, CA) with a floppy tip was introduced into the first or second septal branch and an over-the-wire balloon (1.5-2.mm diameter) was advanced into a septal branch and inflated with 4 to bar to block the septal artery.Then, contrast material was injected to localise the ablation area under fluoroscopy.Following this procedure echo contrast (Levovist, Schering, Germany) 1 to 2 ml was injected under 2D echocardiography for proper localisation of the ablation site.When the septal bulge or thickened septum was not properly stained, the 2nd or 3rd septal branch was selected and echocardiography repeated until the optimal ablation site was identified.Then, 1 to 3 ml pure (9%) alcohol was slowly injected over 3 to 5 min, while the pressure gradient and the ECG were continuously recorded.The dose of alcohol was chosen according to the response of the pressure gradient, the occurrence of arrhythmias (AV block) and the pain reaction of the patient.If an AV block occurred, alcohol injection was stopped immediately.If pressure gradient did not respond to alcohol ablation, a second branch was ablated (n = 3).At the end of injection, it was waited 4 to 5 min until the balloon was deflated, to prevent alcohol backflow into the LAD.After the interven-tion, patients were monitored for 18 to 3 hours.The temporary pacemaker was removed after 8 to 12 hours (next morning).Cardiac enzymes (creatine kinase, troponin I) were determined after hours and the next morning.

Follow-up
All patients were asked to undergo a follow-up examination after 3, , and 12 months either at the referring cardiologist or at our institution.Mean follow-up was 2.8 years with a range between 4 months and .7 years.Echocardiography was carried out at 3, , and 12 months (mean 2.2 years).Holter monitoring was done in 23 patients at follow-up examination (mean 2.2 years).Exer-cise testing was performed in 11 of 23 patients after the mean FU of 2.9 years.One patient committed suicide 4 years after alcohol ablation, therefore late clinical follow-up consisted of 23 patients.

Statistical analysis
Results of continuous variables are given as median with interquartile range.Wilcoxon test for nonparametric paired variables was used for group comparisons.Frequency distributions were assessed with chi 2 test.Differences were considered significant if the two-tailed p value was <.5.Analysis was performed by using SPSS ® Ve rsion 11 (SPSS Inc., Chicago, IL).

Clinical symptoms
During the intervention 83% (2/24) of all patients developed acute chest pain of moderate to severe intensity.All patients received 5 to 1 mg morphine.Angina was found in 71% (17/24) before and in 52% after ablation (12/23) Pressure recording in a patient with severe hypertrophic cardiomyopathy before and after alcohol ablation.Shown are eCG recordings of 3 standard leads as well as simultaneous measurements of lV and aortic pressures.There is a large gradient of approximately 100 mm Hg (post-extrasystolic 170 mm Hg) before but none after ablation.

Figure 2
Original recording in a patient who developed AV block iii during intervention.Arrows indicate the occurrence of the P-wave in the eCG, respectively A-wave in the lV pressure recording, which is not follewed by a QRS complex or lV contraction.The wide QRS complex after AV block iii is induced by temporary pacing.

Figure 6
Holter recording of a 43-year old patient 1 year after alcohol ablation.He was treated for 4 years with amiodarone and remained asymptomatic.Five years after alcohol ablation he developed NSVT.He was treated by iCd implantation and received beta-blocker therapy.Amiodarone was stopped.

Discussion
Alcohol ablation of the septum has become the new treatment modality in hypertrophic obstructive cardiomyopathy with a high success and a relative low complication rate [8-1].The most dreaded complication is the development of AV block requiring pacemaker-implantation (PM) and prolonged rhythm control after the intervention.In a literature overview (table 4a) 1-14% PM implantations were necessary.However, up to 5% of all studies showed transient AV block although newer studies without bolus injection and lower amounts of alcohol [11,12] showed transient AV block in 1-21% of all cases (table 4a).
Most of the studies showed a dramatic reduction in systolic outflow tract obstruction after al-cohol ablation.Resting pressure gradients in our study were 1 mm Hg during late FU of 2.2 years.A favourable result with regard to elimination of outflow tract obstruction was achieved in 95% of all patients.As a consequence, LV wall thickness decreased by 8% of both septum and posterior wall, whereas LV mass was reduced by 11%.These data indicate that not only the septum but also the rest of the LV show LV remodelling due to the unloading with a decrease in LV pressure gradient of 74% at rest and 8% during provocation.This reduction in LV hypertrophy is probably accompanied by an improvement in diastolic dysfunction with a reduction in LV filling pressure and an increase in filling [13].The occurrence of a RBBB in up to 5% and LBBB in up to 2% may counterbalance these beneficial effects by introduction of LV asynchrony.However, a similar number of patients had left bundle branch block before the intervention.Comparable data with regard to outcome have been reported [8, [14][15][16][17][18][19][20][21][22][23][24][25][26][27][28].LBBB is a relative contraindication [29] because when RBBB (up to 45%) occurs during alcohol ablation (figure 5) total AV block may appear.In this situation the risk of AV block is in the order of 5% which was not the case in our series.

Risk stratification
Risk stratification has been a major issue in patients with HOCM because there is a small risk of unexpected sudden cardiac death (SCD).
Strategies to identify patients at risk for SCD have been manifold and the 5 most important risk factors are [3]: 1. Non-sustained ventricular tachycardia, 2. abnormal exercise blood pressure response, 3. family history of premature sudden death, 4. unexplained syncope, 5. severe left ventricular hypertrophy (septal thickness >3 mm).
In the present study, patients were mildly to moderately symptomatic with moderate to severe outflow tract obstruction.Severe septal hypertrophy (>3 mm) and NSVT was found in 8% (2/24) and in 13% (2/1), respectively.Sudden cardiac death survivors were not present and a family history of SCD was observed in one.Overall risk for SCD was low to minimal in the present analysis.

Arrhythmogenic potential of alcohol ablation
Previous data have indicated that septal ablation may lead to scarring at the ablation site favouring recurrent tachyarrhythmias and nonsustained ventricular tachyarrhythmias (NSVT) [7].This hypothesis has, however, not been supported by scientific evidence but is pure speculation [31].In contrast, previous studies have suggested that arrhythmias may even decrease due to the reduction in LV hypertrophy with a diminution of subendocardial ischaemia.The reduction in outflow tract obstruction not only improves ex-ercise capacity [32] and clinical symptomatology but also haemodynamics and diastolic function.Almost all previous studies have shown not only an improvement in symptoms (NYHA classification, dyspnoea and angina pectoris) but also a reduction in palpitations and syncopes [14].
NSVT were observed in the present study in 13% (2/1) before and 22% (5/23) after alcohol ablation.The careful search for arrhythmias after the intervention during FU was responsible for the identification of 2 patients with NSVT who Ta ble 3 Arrhythmias.
needed ICD implantation although one of the 2 patients already had NSVT prior to the intervention, the other not.Overall, arrhythmias decreased rather than increased after the intervention (table 3).The only significant change in ECG abnormalities was an augmentation of RBBB from 13% (3/24) to 4% (11/24) (p <.3).The induction of re-entry tachyarrhythmias could not be documented as it was suggested by some authors [33].

Conclusions
Tr eatment of hypertrophic cardiomyopathy has dramatically changed over the last 5-1 years starting with surgical myectomy leading to percutaneous cardiac intervention.Alcohol ablation is a safe and effective treatment with similar results as surgical myectomy.However, PM-implantation for complete AV block may be four times more frequent for alcohol ablation but still remains a rare complication occurring in approximately 1 of 8 patients.The fear of re-entry tachyarrhythmias after alcohol ablation could not be confirmed by the present study but NSVT occurs in 1 to 2% of all patients before and after intervention, and requires an ICD implantation in 1 out of 12 pa-tients (9%).However, the present study is rather small and may not be representative for treatment recommendations because it is not randomised.
Correspondence: Otto M. Hess 24 patients were successfully treated with alcohol ablation.The peak-to-peak (invasively measured) pressure gradient decreased from 43 (IQR 25 to 4) mm Hg to 1 (IQR to 12) mm Hg at rest (p <.1) and from 13 (IQR 75 to 15) mm Hg to 13 (IQR to 31) mm Hg during postextrasystolic potentiation (p <.1).A typical recording is shown in figure 1.In 95% (23/24) of all patients the pressure gradient was reduced below 25 mm Hg at rest and mmHg during provocation.As a complication transient AV block occurred in 22% (5/24) between 2 and 3 minutes.In one patient AV block persisted and a pace-maker was implanted after 3 days.A typical recording is shown in figure 2 in a patient who developed AV block during intervention with a pacemaker induced recovery rhythm.Eight patients developed a right bundle branch block (RBBB) after the intervention.
Figure 3 lV outflow tract gradient in patients before and after alcohol ablation (mean follow-up 2.2 years).data are given at rest (upper panel) and after provocation with amyl nitrite (lower panel).Pressure gradients are measured echocardiographically.Median gradients (red line) decreased from 38 to 13, respectively to 10 mm Hg during late follow-up (p <0.001).The pressure gradient after provocation decreased from 85 to 46, respectively to 27 mm Hg during late follow-up (p <0.001).

before late after (2.8 years)
[15]arisons with previous surgical studies have indicated similar results between surgical myectomy and alcohol ablation[15].However, comparisons of current literature data suggest that surgical myectomy results in less complete AV block and requires PM implantation only in about 1 ⁄ 4 of those undergoing alcohol ablation (5% versus 14%; table 4b).Cardiovascular death was higher in the alco-hol ablation group but many of the surgical studies did not specify mortality, and the two cohorts have similar mortality ranges between % and 5%.Procedural outcomes of surgical or catheter-based therapies are similar but alcohol ablation is less invasive with a shorter hospital stay and lower costs[8].
[33,35]achyarrhythmias by scarring of the septum[33,35].However, we and others do not believe that this is true.The reported data