Author + information
- Received January 14, 2003
- Revision received April 11, 2003
- Accepted April 17, 2003
- Published online July 16, 2003.
- ↵*Reprint requests and correspondence:
Dr. Nasser M. Lakkis, Section of Cardiology, Department of Medicine, Baylor College of Medicine, 6550 Fannin, SM 677, Houston, Texas 77030, USA.
Objectives The purpose of this paper is to examine the incidence and determinants of permanent complete heart block (CHB) after nonsurgical septal reduction therapy (NSRT), and to evaluate the clinical impact of permanent pacemaker (PPM) placement.
Background Nonsurgical septal reduction therapy with ethanol improves the clinical and hemodynamic parameters in patients with symptomatic hypertrophic obstructive cardiomyopathy. Complete heart block is a common complication after NSRT.
Methods The database of 261 consecutive patients who underwent NSRT at Baylor College of Medicine was reviewed. Clinical variables that were considered as possible determinants for CHB after NSRT were: age, gender, New York Heart Association (NYHA) functional class, left ventricular outflow tract (LVOT) gradient at rest or with provocation, septal thickness, and baseline exercise duration. For electrocardiographic (ECG) variables, the presence of first-degree atrioventricular (AV) block, bifascicular block, left bundle branch block, atrial fibrillation, and left ventricular hypertrophy were analyzed. In addition, the volume of ethanol injected, the method of administration of ethanol (i.e., bolus vs. slow injection [over 30 to 60 s]), number of septal arteries occluded, use of myocardial echocardiography, and infarct size as determined by peak creatine kinase level.
Results Of 261 consecutive patients, 37 had PPM or automatic implantable cardiac defibrillator placed before NSRT. Of the remaining 224 patients, 31 (14%) developed CHB after the procedure. Multivariate logistic regression analysis showed that female gender (odds ratio [OR] 4.3; p = 0.02), bolus injection of ethanol (OR 51; p = 0.004), injecting more than one septal artery (OR 4.6; p = 0.016), the presence of left bundle branch block (OR 39; p = 0.002), and first-degree AV block (OR 14; p = 0.001) on the baseline ECG are independent predictors of CHB after NSRT. Patients requiring PPM placement had a similar improvement in their NYHA functional class, septal thickness reduction, LVOT gradient reduction, and improvement of exercise capacity when compared with patients who did not require pacing.
Conclusions Multiple demographic, electrocardiographic, and technical factors seem to increase the risk of CHB after NSRT. Patients with CHB after NSRT derive similar clinical and hemodynamic benefit to patients who did not require permanent pacing.
Hypertrophic obstructive cardiomyopathy (HOCM) is characterized by left ventricular hypertrophy and dynamic left ventricular outflow tract (LVOT) obstruction (1,2). Nonsurgical septal reduction therapy with ethanol (NSRT) has been shown to improve symptoms and LVOT obstruction (3–9). We and others have reported on the clinical outcome and the associated complications with this procedure (4–10). Specifically, complete heart block (CHB) has been reported to occur transiently or permanently in up to 50% of the treated patients (5,6,10,15). The incidence of permanent pacemaker (PPM) requirement varies among different studies, and the determinants of this complication as well as its clinical impact at follow-up are not yet completely known.
Therefore, the purpose of this study is to identify the determinants of CHB, and its clinical impact after NSRT.
The technique of NSRT has been previously published (6). The database of 261 consecutive patients who underwent the procedure at Baylor College of Medicine was reviewed. In brief, only patients with drug-resistant symptoms of dyspnea (New York Heart Association [NYHA] functional class III or IV), or syncope underwent NSRT. All patients were evaluated by echocardiography before and after the procedure. All patients had a septal-to-posterior wall thickness ratio of at least 1.3 and a resting LVOT gradient >30 mm Hg. Patients with classic symptoms, systolic anterior motion of the anterior mitral leaflet, and a resting LVOT gradient <30 mm Hg were provoked with up to 15 μg/kg/min of dobutamine (mean 10 ± 3 μg/kg/min), and they were considered candidates for this treatment if their dobutamine-provoked gradient exceeded 60 mm Hg (n = 41).
The study was approved by the Institutional Review Board of Baylor College of Medicine, and written informed consent was obtained from all patients. On the day of the procedure, an electrocardiogram (ECG) was performed. The ECG data were collected, reviewed, and analyzed, and blood was collected for baseline creatine kinase (CK) measured in international units. A temporary pacemaker was placed in all patients except for those who already had a PPM in place. Two-dimensional echocardiography and Doppler was done at baseline and during the procedure. A 6F, 7F, or 8F guiding catheter was engaged in the left main coronary artery, and a 2.5 × 9-mm Ranger or Maverick balloon catheter (Scimed, Minneapolis, Minnesota) was advanced over a 0.014-inch wire into the target septal perforator artery or arteries. In all except 10 patients, myocardial contrast was injected through the balloon lumen to delineate the area to be infarcted. One to 3 cc (mean 2.4 ± 0.3 cc) of ethanol were injected into the septal artery or arteries, which supplied the culprit septal segments, and left in place for 5 min. Patients who remained in CHB 72 h after the procedure underwent permanent dual chamber pacemaker placement. Patients stayed in the hospital for a mean of 3.2 ± 0.5 days. Follow-up was completed as previously described for up to four years.
Predictors of PPM requirement
The following clinical variables were considered in the analysis: age, gender, NHYA functional class, LVOT gradient at rest or with provocation, septal thickness, and baseline exercise duration. For ECG variables, the presence of first-degree atrioventricular (AV) block, bifascicular block, left bundle branch block, atrial fibrillation, and left ventricular hypertrophy were analyzed. In addition, the volume of ethanol injected, the method of administration of ethanol (i.e., bolus vs. slow injection [over 30 to 60 s]), number of septal arteries occluded, use of myocardial echocardiography contrast (MCE), and infarct size as determined by peak CK level.
To determine the clinical impact of permanent pacing after NSRT, the group of patients who needed permanent pacing after NSRT was compared with all the remaining HOCM patients who were not pacemaker-dependent after the procedure. In addition, we identified a control group that matched the pacemaker group with respect to age, baseline LVOT gradient, and follow-up period but did not need permanent pacing.
Data are presented as mean ± SD. The two-sample unpaired Student ttest was used to compare the outcome variables between the pacemaker-dependent group and the other group. Logistic regression was applied for the prediction of CHB. A p value of <0.05 was considered significant.
A total of 261 patients were included in this analysis. The mean age was 51 ± 6 years, and 127 patients were women. The baseline NYHA score was 2.7 ± 0.5. The resting LVOT gradient was 62 ± 28 mm Hg, and the average septal thickness was 2.0 ± 0.4 cm. Left ventricular ejection fraction was 71 ± 7%. The peak CK was 1,335 ± 653 at 9 h after alcohol injection.
The baseline demographic, procedural, and clinical characteristics of different subgroups of patients are shown in Table 1. The ECG data before the procedure are shown in Table 2. The prevalence of left bundle branch block was more common in the group of patients who developed CHB and subsequent permanent pacing.
Incidence of pacemaker requirement
A total of 37 patients had a PPM or automated implantable cardiac defibrillator in place before the procedure. Only six of them were not paced on baseline ECG before the procedure. Of the 224 patients who did not have a PPM before NSRT, 31(14%) required permanent pacing. All patients had their PPMs implanted before discharge, except for one patient who developed CHB one week after the procedure. A higher incidence of permanent pacing was present in our first group of 33 patients (33%) compared with the subsequent 191 patients in whom only 10% required permanent pacing. Of note, MCE was not employed in the first 10 patients. Furthermore, ethanol was administrated as a bolus injection (rather than slow delivery) in the first 17 patients. There was no significant association between peak CK level and the need for PPM (1,333 ± 873 vs. 1,331 ± 681; p = 0.99). Furthermore, when the CK level was adjusted to the interventricular septal thickness, there was still no difference between the two groups (629 ± 345 vs. 669 ± 567; p = 0.9).
Predictors of CHB after ablation
Multiple baseline and procedure-related characteristics were studied to determine their relationship to CHB. Table 3shows the outcome of the univariate analysis for predictors of occurrence of this complication after NSRT. As shown, female gender (p = 0.045), lack of use of MCE (p = 0.001), injecting ethanol as a bolus instead of gradual injection (p < 0.001), injecting ethanol into more than one septal artery (p = 0.05), and the presence of left bundle branch block on the baseline ECG (p < 0.001) were all found to be significant determinants for the development of CHB. The volume of ethanol injected and the peak CK obtained after the procedure did not predict AV block in this model. On multivariate analysis (Table 4), female gender (odds ratio [OR] 4.3; p = 0.02), bolus injection of ethanol (OR 51; p = 0.004), injecting more than one septal artery (OR 4.6; p = 0.016), first-degree AV block on the baseline ECG (OR 14; p = 0.001), and the presence of left bundle branch block (OR 39; p = 0.002) were all significant predictors of CHB after NSRT.
Clinical impact of PPM requirement
At follow-up (mean 24 ± 19 months), 28 of the 31 patients who required permanent pacing after the procedure remained in paced rhythm and, of these patients, 25 were pacemaker-dependent during treadmill exercise testing. As shown in Table 5, when compared with patients (n = 193) who did not develop CHB, the 31 patients who were paced after the procedure derived similar benefit from NSRT. Specifically, 77% of patients who required a PPM became ≤ NYHA functional class II for dyspnea compared with 67% of patients without PPM (p = 0.26). Sixty-one percent of patients with PPM improved by ≥2 NYHA classes compared with 43% of patients without PPM (p = 0.06).
We then selected 31 nonpaced patients who were matched for age, baseline LVOT gradient, and follow-up duration to the 31 patients who were pacemaker-dependent after NSRT. As seen in Table 6, baseline characteristics were similar between the two matched groups. The group with CHB had more septal arteries injected (1.45 ± 0.5 vs. 1.17 ± 0.38; p = 0.023), and a larger number of patients who (7 compared with 0) received bolus injection of ethanol. As shown in Table 7, both groups had a similar improvement in clinical and hemodynamic variables after NSRT. Notably, 61% (19/31) of patients who required PPM have improved by ≥2 NYHA dyspnea classes on follow-up compared with 49% (15/31) of patients who did not require a PPM (p = 0.3). In addition, 77% (24/31) of patients with PPM have become ≤ NYHA functional class II for dyspnea compared with 68% (21/31) of patients who did not require a PPM (p = 0.4).
Nonsurgical septal reduction therapy is currently performed at multiple centers in the U.S. and Europe with a high procedural success rate and an acceptable complication rate. Many published reports have confirmed the persistent improvement in symptoms, along with reduction in LVOT gradient and left ventricular remodeling in the intermediate-term follow-up (8,10–12). The success rate of NSRT was also comparable to that of surgical myectomy in nonrandomized studies (13). Complete heart block is the most commonly reported complication after NSRT; therefore, the overall results after NSRT can be further improved by decreasing the incidence of CHB.
Nonsurgical septal reduction therapy results in several changes in the AV and intraventricular conduction. In general, first-degree AV block, right bundle branch block, left anterior fascicular block, left bundle branch block, and nonspecific intraventricular conduction delay have been reported after NSRT. First-degree AV block is often transient, and only 2% of patients develop permanent first-degree AV block after one year (6,14). On the other hand, right bundle branch block occurs in 60% to 100% of patients and is more likely to be permanent (4,6,14). However, the most serious conduction abnormality is CHB, which develops transiently in up to two-thirds of patients (4,10)or permanently in 0% to 25% according to different series (4–6,9,10,15–18). The overall incidence in our series is 14%. This number is much lower than what we have initially reported with our first 33 patients (6). With experience and modification of technique such as the use of MCE and slow injection of alcohol, the incidence of CHB requiring PPM decreased to 10% in the last 191 patients. Thirty of the 31 patients had CHB immediately after the ethanol injection. Only one patient returned with this complication one week after the procedure. Most of the patients remained paced at follow-up, both at rest and on exercise.
Lack of conduction abnormality during transient occlusion of septal arteries without ethanol has been proposed as a potentially useful method in screening patients for developing CHB (17); however, subsequent studies showed suboptimal predictability of this method in regard to reduction of LVOT gradient after the procedure (4).
Baseline conduction abnormalities have been suggested as a risk factor for development of CHB (14). Using multivariate analysis, we have shown that the presence of left bundle branch block and first-degree AV block on the baseline ECG are associated with a higher incidence of CHB. This finding is expected, knowing that more than two-thirds of patients undergoing NSRT develop complete right bundle branch block (6). Additional insult to a previously diseased conduction system is more likely to result in this complication.
Our analysis also revealed a gender difference in the susceptibility of CHB. Female patients were at higher risk. They tended to be older (55 ± 17 years vs. 48 ± 15 years) had worse NYHA functional class, higher rest LVOT, and thinner interventricular septum as compared with their male counterparts. However, they also had a smaller infarct size and CK level/interventricular septum ratio. Nevertheless, after adjusting for all the baseline demographic, clinical, ECG, and procedural characteristics, female gender still predicted a higher risk for CHB.
In addition to baseline conduction disorders, certain procedural techniques are associated with a higher propensity for CHB such as the number of injected septal arteries, (probably due to involvement of multiple sites of the conduction pathway) and bolus injection of ethanol, which deprives the operator of the ability to slow down or abort the injection of ethanol in the face of a developing conduction disorder. We believe that the concomitant technical changes that occurred after our first 17 patients (use of MCE and injecting ethanol slowly) accounted for the significant reduction in the incidence of CHB after NSRT. A repeat analysis was done after we excluded the first 17 patients in whom no MCE was used and ethanol was injected as a bolus. By univariate analysis, older patients (>55 years), female gender, left bundle branch block, and first-degree AV block continued to be significant determinants of CHB. On univariate analysis, all except age >55 years significantly predicted CHB after NSRT.
The clinical impact of CHB has not been well studied. Our data suggest that the development of CHB does not compromise patients’ outcome after NSRT. However, more prospective data is needed to verify that right ventricular pacing does not reduce the potential benefit of NSRT, as some recent studies have suggested that right ventricular pacing may impair ventricular performance in the presence of hypertrophy and failure, as compared with normal conduction or biventricular pacing.
In conclusion, our findings suggest that multiple factors may contribute to the development of CHB after NSRT. Female gender and baseline conduction abnormalities seem to increase the risk for this complication. Patients with CHB after NSRT seem to derive similar clinical and hemodynamic benefits.
Our study has several limitations. It is retrospective in nature and, therefore, it has all the problems associated with such analysis. We cannot exclude the presence of other important predictors of CHB. Certain “tricks” learned by the operators with experience were difficult, if not impossible, to standardize. One potentially important factor that was not analyzed is the detailed evaluation of the septal artery anatomy by angiography. Other important data of interest are the vascular territory distribution by MCE and transient conduction changes on ECG during the procedure. Although our series is the largest reported so far, it included a relatively modest number of patients who required pacemaker placement. Future prospective studies are needed to validate our observations.
- complete heart block
- creatine kinase
- hypertrophic obstructive cardiomyopathy
- left ventricular outflow tract
- myocardial contrast echocardiography
- nonsurgical septal reduction therapy
- New York Heart Association
- odds ratio
- permanent pacemaker
- Received January 14, 2003.
- Revision received April 11, 2003.
- Accepted April 17, 2003.
- American College of Cardiology Foundation
- Knight C.,
- Kurbaan A.,
- Seggewiss H.,
- et al.
- Seggewiss H.,
- Gleichmann U.,
- Faber L.,
- et al.
- Lakkis N.,
- Nagueh S.,
- Kleiman N.,
- et al.
- Lakkis N.,
- Nagueh S.,
- Killip D.,
- et al.
- Lakkis N.,
- Nagueh S.,
- Dunn K.,
- et al.
- Boekstegers P.,
- Steinbigler P.,
- Molnar A.,
- et al.
- Faber L.,
- Meissner A.,
- Ziemssen P.,
- et al.
- Mazur W.,
- Nagueh S.F.,
- Lakkis N.M.,
- et al.
- Knight C.
- Nagueh S.F.,
- Ommen S.R.,
- Lakkis N.M.,
- et al.
- Kazmierczak J.,
- Kornacewicz-Jach Z.,
- Kilsy M.,
- et al.
- Kuhn H.,
- Gietzen F.,
- Leuner C.
- ↵Bhargava B, Agarwal R. Can we predict complete heart block after alcohol ablation for hypetrophic cardiomyopathy? Circulation 1999;100:e144
- Qin JX, Shiota T, Asher CR, et al. The incidence of conduction system block after septal reduction in patients with hypetrophic cardiomyopathy. Circulation 106;19:II655