Nonsurgical Septal Reduction for Hypertrophic Obstructive Cardiomyopathy
Outcome in the First Series of Patients
Background Some patients with hypertrophic obstructive cardiomyopathy may gain symptomatic relief from a reduction in the extent of obstruction to left ventricular outflow. We present the outcome of the first series of patients treated with an alternative method of gradient reduction using catheter techniques.
Methods and Results Eighteen patients were treated with selective intracoronary alcohol injection to induce localized septal infarction. Patients underwent echocardiographic measurement of left ventricular dimensions and Doppler echocardiographic evaluation of left ventricular outflow tract gradients before the procedure, on the first postoperative day, and at a median follow-up of 3 months after the procedure. In addition, patients underwent exercise testing and symptom evaluation before and 3 months after nonsurgical septal reduction. There was a significant reduction in left ventricular outflow tract obstruction after the procedure (preprocedure, 67 mm Hg [95% CI, 48 to 87 mm Hg]; postprocedure, 25 mm Hg [95% CI, 16 to 34 mm Hg]; P=.0006), which persisted at 3-month follow-up (22 mm Hg [95% CI, 12 to 32 mm Hg]; P=.001). This was associated with a significant improvement in symptoms. There was a small but not significant increase in exercise capacity (n=10; preprocedure, 418 seconds [95% CI, 273 to 563 seconds]; postprocedure, 452 seconds [95% CI, 283 to 621 seconds). Left ventricular dimensions were not significantly altered by nonsurgical septal reduction.
Conclusions Nonsurgical septal reduction significantly reduces left ventricular outflow tract obstruction and improves symptoms in some patients with hypertrophic obstructive cardiomyopathy. The technique may provide an alternative to surgical myomectomy in selected patients.
Asignificant number of patients with hypertrophic cardiomyopathy have evidence of left ventricular outflow tract obstruction1 and may suffer symptoms of angina, dyspnea, and syncope as a consequence. In patients who remain symptomatic despite medical treatment, surgical relief of obstruction has been shown to improve quality of life,2 3 even if its prognostic effects remain uncertain in the absence of prospective, randomized, controlled trials. There are several techniques for the surgical removal of the offending portion of the interventricular septum,4 5 6 but all require extracorporeal circulation and are associated with a moderate surgical risk of at least 5%.4 7
A nonsurgical technique to achieve a reduction in septal mass by producing septal infarction using catheter techniques has been proposed by Sigwart.8 The first septal branches of the anterior descending coronary artery supply the myocardium of the proximal interventricular septum, the area of myocardium whose abnormal structure and function are responsible for the production of the left ventricular outflow tract obstruction in hypertrophic obstructive cardiomyopathy. After preliminary observations that temporary occlusion of the first major septal artery decreased the degree of outflow tract obstruction, the creation of permanent septal necrosis rather than temporary septal ischemia by intracoronary alcohol injection was shown to be effective in permanently reducing gradients in three patients with severe hypertrophic obstructive cardiomyopathy.8
We have now treated 18 patients by this novel technique and present the results of the procedure in this first series of patients in terms of the immediate hemodynamic effects and the results of echocardiographic evaluation, exercise testing, and symptomatic follow-up at a median of ≥3 months.
Eighteen patients with hypertrophic obstructive cardiomyopathy have been treated with nonsurgical septal reduction. Twelve patients underwent the procedure at the Royal Brompton Hospital, London, UK, and six at the Herz-und Diabeteszentrum Nordrhein-Westfalen, Bad Oeynhausen, Germany, between June 1994 and February 1996, using identical protocols.
All patients had echocardiographic evidence of significant outflow tract obstruction and symptoms of angina or dyspnea despite medical treatment. Five patients had previously undergone permanent pacemaker implantation in an attempt to modify outflow tract obstruction by altering the sequence of left ventricular activation; one had an unsuccessful trial of temporary dual-chamber pacing. In all cases, consent for the new procedure was obtained after careful explanation of the proposed technique of ablation and discussion of the surgical alternatives.
Nine of the patients were male and nine were female. The mean age of the patients was 49 years (range, 14 to 81 years). Patient characteristics, symptoms, and baseline therapy are shown in Table 1⇓. Investigations were performed before and after nonsurgical septal reduction on identical medical therapy.
Ethics approval from the Royal Brompton Hospital was first given in August 1992 for five pilot patients and subsequently increased after the initial promising results.
Before the procedure, all patients underwent transthoracic echocardiography. Resting left ventricular outflow tract gradients were measured, as were septal thickness and left ventricular end-systolic and end-diastolic dimensions. Echocardiography was repeated on the first postoperative day and at 3-month follow-up. Echocardiographic examination was performed using a Hewlett-Packard Sonos 1500 echocardiograph with a 2.5-MHz transducer interfaced to it. Left ventricular dimensions were measured from the minor axis M-mode of the left ventricle, obtained from the two-dimensional, guided, standard left parasternal view. Septal thickness was measured from the same trace with the use of leading-edge methodology. End diastole was taken at the onset of the Q wave of the superimposed ECG.
Left ventricular outflow tract gradient was measured by use of a nonimaging, continuous-wave Doppler probe positioned at the cardiac apex as identified in the apical four-chamber view by the imaging probe and using the same echogram used at rest, or by use of the Doptek ultrasound system. The pressure gradient was measured in millimeters of mercury (the equivalent of meters per second of outflow tract velocity when the Hewlett-Packard machine was used, or the equivalent of kilohertz when the Doptek system was used).
Ten patients performed symptom-limited treadmill exercise tests before and 3 months after the procedure. A standard Bruce protocol with the addition of a “stage 0” (3 minutes, 1 mph, 5% gradient) was used. Five patients underwent measurement of V̇o2max during the exercise test: during exercise, subjects breathed through a mouthpiece and a one-way valve attached to a mass spectrometer (Amis 2000 system, Innovision). Using a standard inert gas dilution technique, this allowed on-line measurement of metabolic gas exchange and minute ventilation every 10 seconds. Patients were instructed to exercise to their maximum capacity.
Patients underwent 24 hours of ambulatory monitoring before the procedure and at follow-up. Monitoring was performed with the use of pregelled electrodes to record two bipolar leads, an anterior CM5, and an inferior lead. Two-channel recordings were obtained on magnetic tape by an amplitude-modulated dual-channel recorder (Reynolds Tracker). Ventricular tachycardia was defined as six consecutive beats at a rate of ≥120 bpm.
Retrograde and transseptal cardiac catheterization was performed as previously described.8 Resting measurements of the left ventricular gradient were recorded, and then the patients underwent provocation with Valsalva maneuver and dobutamine infusion when measurement of the gradient was repeated. Gradients were also recorded after extrasystoles.
The first major septal branch of the anterior descending coronary artery was then identified and catheterized with a 2- to 2.5-mm coaxial PTCA balloon catheter (Fig 1A⇓ and 1B⇓). The hemodynamic effect of temporary balloon inflation of this vessel was established, and to delineate the extent of myocardial perfusion of the vessel, the guidewire was withdrawn and contrast injected through the lumen. If the cannulated vessel supplied the correct area of the myocardium and its occlusion resulted in a reduction in the left ventricular outflow gradient, we proceeded to permanent ablation. After the administration of diamorphine (5 mg IV), 2 to 5 mL of absolute alcohol was slowly injected into the septal artery through the central lumen of the angioplasty catheter and left in situ for 5 minutes, with the balloon remaining inflated at the origin of the septal artery. After deflation of the balloon, angiography was repeated to confirm blockage of the target artery (Fig 1C⇓). Measurements of the left ventricular outflow gradient were then repeated at rest and on provocation, as described above. If no single, large, first septal branch was found and the gradient reduction was small after alcohol injection, a second or third septal artery was catheterized and the procedure repeated.
Data were analyzed by use of paired Student’s t tests. A level of P<.05 was considered significant. Data are presented as mean (95% CI).
Immediate Hemodynamic Effects
Resting left ventricular gradient. The left ventricular outflow gradient was reduced or abolished immediately, seconds after the alcohol injection, in most patients. Fig 2⇓ shows pressure tracings before and after alcohol injection in a typical case (patient 2). Fig 3⇓ illustrates that the gradient after an extrasystole has also been abolished (patient 1). The mean resting gradient after the procedure was 8 mm Hg (3 to 13 mm Hg) compared with 51 mm Hg (40 to 63 mm Hg) before the procedure.
Maximal left ventricular gradient. After alcohol injection, patients were restressed with dobutamine, pacing, and the production of premature ventricular contractions. In most cases, gradients on stress persisted after the procedure, albeit reduced. The mean maximal gradient after the procedure was 29 mm Hg (14 to 43 mm Hg) compared with 119 mm Hg (85 to 153 mm Hg) before the procedure.
Individual patient Doppler echocardiographic data, obtained before the procedure and on the first postoperative day, are shown in Fig 4⇓. In two patients, there was no reduction in gradient despite a fall in gradient during the procedure. The mean gradient measured in this way was 25 mm Hg (16 to 34 mm Hg) after the procedure compared with 67 mm Hg (48 to 87 mm Hg) before the procedure.
Five patients had permanent pacemakers implanted at the time of the procedure. In the remaining 13 patients, the most common ECG change was the development of right bundle-branch block, which occurred in 11 patients. In 6 patients, this was the only ECG change. In 3 patients, right bundle-branch block was accompanied by anterior ST-segment elevation, and in another 2, by the development of anterior Q waves. The 2 patients who did not develop right bundle-branch block both developed isolated anterior ST-segment elevation after alcohol injection.
Size of Infarction
The mean peak rise in creatinine kinase (CK) was 2295 IU (345 to 14 960). These data are skewed by one patient with a very large enzyme rise (see “Complications”). The median CK rise was 1222 IU.
Chest pain. At the time of alcohol injection, all patients experienced chest discomfort of moderate severity lasting for 1 to 2 minutes.
Heart block. Four patients experienced transient complete heart block after the alcohol was administered. The longest duration of complete heart block was 5 minutes.
Ventricular arrhythmias. Two patients developed ventricular arrhythmias after nonsurgical septal reduction. Patient 9, a 14-year-old girl, developed profound bradycardia and then ventricular fibrillation during femoral sheath removal 2 hours after the procedure. The arrhythmia responded to a single 200-J DC cardioversion, and there were no adverse sequelae.
Patient 12 developed several episodes of ventricular tachycardia, requiring DC cardioversion and administration of intravenous amiodarone, in the first 6 hours after the procedure. This patient also had transient occlusion of the left anterior descending coronary artery (LAD) immediately after a second alcohol injection into the first septal artery (because the first injection had been thought to be too distal), presumably as a result of a small amount of alcohol leaking down the main lumen of the LAD. This was associated with marked anterior ST elevation and a rise in CK to 14 960 IU. Angiography was repeated on the first postoperative day and showed restoration of flow down the LAD. The patient made an otherwise uneventful recovery, and follow-up echocardiography has shown left ventricular function to be well preserved, even in the myocardium in the territory of the LAD.
Length of stay. Mean length of stay was 5 days (range, 2 to 11 days).
Three-Month Follow-up Data
Symptoms. Table 2⇓ shows the functional status of patients and length of follow-up. The majority of patients experienced considerable improvement in symptoms and quality of life.
Holter monitoring. Ambulatory monitoring at 3 months revealed neither ventricular tachycardia nor heart block in any patient.
Exercise testing. Symptom-limited exercise testing was performed before and after the procedure in 10 patients. Mean exercise time was 418 seconds (273 to 563 seconds) before nonsurgical septal reduction, increasing to 452 seconds (283 to 621 seconds) after (P=NS). Heart rate at maximum exercise was unchanged (preprocedure, 131 bpm [122 to 139 bpm]; postprocedure, 135 bpm [117 to 152 bpm]). Measurement of V̇o2max was performed before and after nonsurgical septal reduction in 5 patients. There was a small but not significant rise in V̇o2max (preprocedure, 24.2 mL·kg−1·min−1 [18.4 to 30 mL·kg−1·min−1]; postprocedure, 26.8 mL·kg−1 ·min−1 [19.1 to 34.5 mL·kg−1·min−1]).
Echocardiography. Repeat Doppler echocardiography at 3 months (n=17) showed that the reduction in gradient was maintained, with a mean gradient of 22 mm Hg (12 to 32 mm Hg) at follow-up compared with 68 mm Hg (49 to 87 mm Hg) before the procedure. Individual patient data are shown in Fig 4⇑. The majority of patients showed little change in gradient between evaluation on the first postoperative day and 3-month follow-up. One patient had an increase in gradient of ≥20 mm Hg, and 3 had a fall of ≥20 mm Hg. In 1 of these latter cases, magnetic resonance imaging showed further septal thinning and fibrosis. (See Fig 5⇓.)
Echocardiographic estimation of left ventricular size showed little variation during the follow-up period. Septal thickness was 2.1 cm (1.8 to 2.4 cm) before the procedure compared with 1.8 cm (1.5 to 2.1 cm) 3 months afterward. There was a small rise in end-systolic dimension (preprocedure, 2.6 cm [2.3 to 3.0 cm]; 3 months after the procedure, 3.0 cm [2.3 to 3.7 cm]) and end-diastolic dimension (preprocedure, 4.5 cm [4.1 to 4.9 cm]; 3 months after the procedure, 4.9 cm [4.3 to 5.5 cm]) during the follow-up period. None of these changes were significant.
The results of nonsurgical septal reduction in this first series of patients with hypertrophic obstructive cardiomyopathy suggest that the procedure produces significant hemodynamic benefit. Follow-up data show that the reduction in left ventricular outflow tract obstruction persists in the medium term and is associated with improved symptoms. The procedure appears to be acceptable in terms of safety and does not predispose to left ventricular dilatation or ventricular arrhythmias.
Established treatments for relief of symptoms in hypertrophic cardiomyopathy are negative inotropic drugs, dual-chamber pacing, and surgical resection. β-Blockers and other negatively inotropic drugs, such as verapamil and disopyramide, can reduce left ventricular outflow tract obstruction and diminish symptoms.1 9 10 11 However, many patients remain symptomatic despite taking these drugs. Dual-chamber pacing can also reduce left ventricular outflow tract obstruction in hypertrophic cardiomyopathy,12 13 both by altering left ventricular excitation and by optimizing the timing of atrial contraction,14 and has therefore been recommended to patients unresponsive to medical therapy before consideration of surgery. Some patients do not respond to pacing and require surgical intervention.
There are several techniques for the surgical removal of the offending portion of the interventricular septum.4 5 6 Surgical resection is undoubtedly successful palliation for symptomatic patients with severe outflow tract obstruction and compares favorably with conservative treatment.3 The majority of patients derive long-term symptomatic benefit without significant impairment of left ventricular function.15 However, surgery requires extracorporeal circulation and is associated with a moderate surgical risk of at least 5%.4 7
The concept of developing an alternative, nonsurgical method of septal reduction using catheter techniques, thereby avoiding the risks of open heart surgery, was therefore attractive to us and was first conceived after observations were made on the effect of balloon occlusion on the function of focal areas of the myocardium.16 17 Furthermore, the functional significance of reduced septal function subsequent to balloon-induced ischemia was shown in 1983 to be favorable in patients with hypertrophic cardiomyopathy, with marked reduction in intracavity gradients on occlusion of the first major septal artery. These findings have been confirmed by others.18
The reduction in gradient subsequent to induction of permanent septal damage with intra-arterial alcohol injection produces a marked and immediate reduction in left ventricular obstruction. This immediate effect is greater than the echocardiographic findings on the first postoperative day, and in three patients, echocardiography failed to demonstrate a significant effect despite gradient reduction at the time of catheterization. This suggests that there may be a degree of recovery of myocardial function in the early postoperative period. This should be borne in mind during the procedure; in a minority of patients, the proximal interventricular septum is fed by a number of small septal branches rather than one larger first septal artery. In such patients, alcohol injection may be required in more than one branch, and the operator should not be deterred from this by the observation of some degree of gradient reduction after injection down a single, small, branch artery, because the degree of myocardial necrosis will inevitably be small and the gradient may recur over the next few hours. This was certainly the case in the three early procedures that failed to produce lasting benefit.
In contrast, there does not seem to be significant recovery of localized myocardial function and hence outflow tract obstruction after the immediate postoperative period, the results of 3-month Doppler echocardiographic estimation of outflow tract gradients being very similar to those made on the first postoperative day. Indeed, in some cases, there is further gradient reduction and septal thinning, presumably as a consequence of scarring and fibrosis after the infarct.
Global left ventricular performance does not seem to be adversely affected by nonsurgical septal reduction, at least in the medium term, with follow-up echocardiography failing to show significant left ventricular dilatation. This suggests that the reductions in outflow gradients that we observed are not the consequence of depression of overall left ventricular contraction. Furthermore, the extent of infarction produced by the technique may be overestimated by measurement of cardiac enzyme rise, because the pattern and extent of intracellular enzyme leakage after a chemical insult may differ from that seen after ischemic injury.
Although symptomatic improvement has been reported by the majority of patients, we have not documented any objective evidence of increased exercise performance, such as enhanced exercise time or V̇o2max. However, the number of patients undergoing serial exercise testing was small, and studies in larger numbers of patients are required to allow a proper evaluation of the effect of the procedure on the patient’s ability to exercise.
A number of complications have occurred in these first 18 cases. At the time of alcohol injection, all patients experienced a degree of chest discomfort that was surprisingly short in duration, typically lasting for only 1 to 2 minutes. We routinely administered diamorphine a few minutes before alcohol injection. Four patients experienced transient complete heart block after the alcohol was administered, and although in none of these cases was the heart block permanent, it is clearly mandatory, in patients who do not have a permanent pacemaker, to insert a temporary ventricular pacing wire to cover the perioperative period.
Two patients have suffered from ventricular arrhythmias after nonsurgical septal reduction. In the first case, that of a young girl, this was secondary to bradycardia on sheath removal. In the other, several episodes of ventricular tachycardia were the consequence of alcohol leakage down the main lumen of the LAD. This caused transient impaired flow to that artery and was associated with marked anterior ST elevation and a large cardiac enzyme rise. However, arterial patency was restored by the following day, and the patient suffered no longer-term adverse events, left ventricular function being well preserved. To avoid this complication, it is clearly essential that the balloon should be firmly inflated just distal to the origin of the first septal artery. If the balloon is positioned too proximally or if it is of inadequate size, alcohol may leak down the LAD. The chances of leakage are also increased if alcohol is injected for a second time down the same septal branch, because the first injection will inevitably cause partial occlusion of the branch and increase impedance to forward flow. A new type of a compliant, spherically shaped balloon has recently been tested, with good results.
We have shown that nonsurgical septal reduction is a safe method of reducing left ventricular outflow tract obstruction in patients with hypertrophic obstructive cardiomyopathy. The procedure produces significant hemodynamic improvement during medium-term follow-up and is associated with symptomatic improvement. To evaluate its future role in the treatment of symptomatic patients with hypertrophic obstructive cardiomyopathy, however, prospective randomized trials are now required to compare catheter ablation with other forms of therapy, such as pacing and surgery.
Dr Knight is a British Heart Foundation Junior Research Fellow.
Reprint requests to Ulrich Sigwart, MD, Department of Invasive Cardiology, Royal Brompton Hospital, Sydney Street, London SW3 6NP UK.
- Received August 20, 1996.
- Revision received November 4, 1996.
- Accepted November 23, 1996.
- Copyright © 1997 by American Heart Association
Ciro E, Nichols PF III, Maron BJ. Heterogeneous morphologic expression of genetically transmitted hypertrophic cardiomyopathy: two-dimensional echocardiographic analysis. Circulation. 1983;67:1227-1233.
Maron BJ, Merrill WH, Freier PA, Kent KM, Epstein SE, Morrow AG. Long-term clinical course and symptomatic status of patients after operation for hypertrophic subaortic stenosis. Circulation. 1978;57:1205-1213.
Bonow RO, Maron BJ, Leon MB. Medical and surgical therapy of hypertrophic cardiomyopathy. Cardiovasc Clin. 1988;19:21-239.
McIntosh CL, Maron BJ. Current operative treatment of obstructive hypertrophic cardiomyopathy. Circulation. 1988;78:487-495.
Flamm MD, Harrison DC, Hancock EW. Muscular subaortic stenosis: prevention of outflow obstruction with propranolol. Circulation. 1968;38:846-858.
Grbic M, Sigwart U, Kappenberger L, Goy J-J. Réduction du gradient intraventriculaire en présence de la cardiomyopathie obstructive par le pacing atrio-ventriculaire avec P-R raccourci. Schweiz Med Wochenschr Suppl. 1987;117:29. Abstract.
McDonald K, McWilliams E, O’Keeffe B, Maurer B. Functional assessment of patients treated with permanent dual chamber pacing as a primary treatment of hypertrophic cardiomyopathy. Eur Heart J. 1988;9:893-899.
Borer JS, Bacharach SL, Green MV, Kent KM, Rosing DR, Seides SF, Morrow AG, Epstein SE. Effect of septal myotomy and myectomy on left ventricular systolic function at rest and during exercise in patients with IHSS. Circulation. 1979;60(suppl I):I-82-I-87.
Sigwart U, Grbic M, Essinger A, Rivier JL. L’effet aigu d’une occlusion coronarienne par ballonet de la dilatation transluminale. Schweiz Med Wochenschr. 1982;45:1631. Abstract.
Sigwart U, Grbic M, Payot M, Essinger A, Sadeghi H. Wall motion during balloon occlusion. In: Sigwart U, Heintzen PH, eds. Ventricular Wall Motion. New York, NY: Georg Thieme; 1983:206-210.
Gietzen F, Leuner C, Gerenkamp T, Kuhn H. Relief of obstruction in hypertrophic cardiomyopathy by transient occlusion of the first septal branch of the left coronary artery. Eur Heart J. 1994;15:125. Abstract.