Predictors of outcome after alcohol septal ablation for hypertrophic obstructive cardiomyopathy: an echocardiography and cardiovascular magnetic resonance imaging study

Topics: Hypertrophic cardiomyopathy

Authors: Minjie Lu, Hui Du, Zhan Gao, Lei Song, Huaibing Cheng, Yan Zhang, Gang Yin, Xiuyu Chen, Jian Ling, Yong Jiang, Hao Wang, Jinghui Li, Jinghan Huang, Zuoxiang He, Shihua Zhao

Presented by: Angelos G. Rigopoulos, 1^st Department of Internal Medicine, Leopoldina Hospital, Schweinfurt, Germany

Background

Alcohol septal ablation is a septal reduction treatment alternative for patients with hypertrophic obstructive cardiomyopathy and significant symptoms despite optimal pharmacological therapy. Given the considerable anatomical variability of the existing septal perforators amenable to alcoholisation, as well the complex pathophysiology of left ventricular obstruction involving characteristics of left ventricular contraction, cavity dimensions and pathology of mitral valve and mitral valve apparatus, prediction of optimal clinical and haemodynamic result according to clinical features has always been challenging.

The existing American(1) and European(2) Guidelines for Hypertrophic Cardiomyopathy imply that marked septal hypertrophy (≥ 30 mm) could be a factor leading to less optimal results after septal ablation, thus supporting the choice of myectomy for this patient subset. Nevertheless, there has been up to now no evidence to support this. The studies that have tried to find predictors for the success of alcohol septal ablation have not shown that septal thickness could be one of them.(3) Faber et al.(4) suggested that the haemodynamic result could be less favourable in younger patients with very high baseline gradients and marked septum hypertrophy, however, septal thickness was not a predictor of success in the multivariate analysis. On the other hand, less severe septal hypertrophy at SAM-septal contact was a predictor of optimal outcome in the study of Sorajja et al.(5) A cut-off point in septal hypertrophy above which alcohol septal ablation might be less efficacious has not been possible to establish.

Aim of the study

The study(6) investigated the prognostic value of cardiac magnetic resonance imaging (CMR) and echocardiography to predict the outcome of alcohol septal ablation.

Summary

In this study 102 consecutive patients with hypertrophic obstructive cardiomyopathy were enrolled. All patients underwent CMR and transthoracic echocardiography before alcohol septal ablation (ASA). Selection for ASA was based on persistent symptoms despite optimal pharmacological treatment, Left ventricular outflow (LVOT) gradient ≥ 50mmHg at rest or after provocation with Valsalva manoeuvre, accessible septal branches typically originating from the left anterior descending coronary artery and absence of intrinsic abnormality of the mitral valve or other indication for heart surgery.

It should be noted that Valsalva was performed only in patients who had an LVOT gradient < 60 mmHg at rest before the ASA.

At 6-month follow-up (all patients survived) the patients were divided in 2 groups according to the LVOT gradient reduction: ASA responding group with ≥ 50% reduction of baseline gradient (73 patients, 72%) and ASA nonresponding group with < 50% reduction in LVOT baseline gradient (29 patients, 28%).

There was a significant difference in LVOT gradient after ASA between the 2 groups, whereas there was no difference in LVOT gradient at baseline. Patients of the nonresponding group had higher thickness of the basal septum in echocardiography (28.2±5.0 vs. 24.2±5.0 mm, P<0.001) as well as of the posterior wall (11.4±2.4 vs. 9.7±2.3 mm, P<0.001). Both the residual LVOT gradient and the relative reduction of LVOT gradient correlated significantly with the wall thickness of the basal anterior segment (Segment 1) and of the basal anteroseptal segment (Segment 2) as assessed by CMR. Furthermore, there was an even better correlation of the residual LVOT gradient and of the relative LVOT gradient reduction with the summated thickness of both segments (1+2). The corresponding area under the curve (AUC) and 95% confidence interval (CI) of the ROC curve for the outcome prediction analyses, as well as the sensitivity and specificity for the corresponding cut-off values are depicted in the table:

Conclusion

The thickness of basal septal wall is inversely correlated with the haemodynamic outcome of ASA. Both echocardiography and CMR can be used for patient selection. The summation of thicknesses of the basal anteroseptal and basal anterior segments, as measured with CMR, with a cut-off thickness of 50.9 mm, may have higher predictive merit in distinguishing nonresponders after ASA, compared with the basal septal thickness alone.

Comments

This study provides significant and very important input regarding the significance of basal septal thickness as a predictor of haemodynamic success after ASA. Moreover, it offers for the first time cut-off values for basal septal thickness above which a patient may have a suboptimal haemodynamic outcome after ASA. Indeed, these cut-off values seem to be much lower than the arbitrary 30mm wall thickness stated in the American and European Guidelines. On the other hand, as this is the first study that provides such numbers, the accuracy of these cut-off points should be seen through the prism of the limitations of this study.

The authors acknowledge the retrospective, single-centre and nonrandomised character of the study that could have led to selection bias. Furthermore, the sample was not large and “the follow-up was a little bit short”. The latter is very important, since a remodelling process with constant further reduction of LVOT gradient during follow-up has been reported after ASA.(7) Measurement of the LVOT gradient at baseline during Valsalva manoeuvre was not performed in all patients but only in those with a LVOT gradient < 60 mmHg at rest. As was pointed out by Dr. Fifer in the accompanying Editorial, an important limitation of the study lies in the methodology of the measurement of the LVOT gradient, because of the “lack of standardisation in the performance of the Valsalva manoeuvre” in this study, as well as the well-known spontaneous variation of the magnitude of the LVOT gradient.(3)

A technical consideration regarding the methodology of ASA procedure in this study is the use of X-ray contrast to depict the myocardial target area of interest with contrast echocardiography. There is some concern that the probably poor contrasting capacity of X-ray contrast agents for contrast echocardiography could influence the targeting potential of this method in patients with very thick septum.(8)

A significant point that could be addressed in future studies is the combination of thickness measurements with the tissue characterisation potential of CMR. It is plausible that a very thick septum does not consist entirely of muscle but comprises a significant amount of fibrotic tissue, which might be left largely unaltered (at least in thickness) after ASA.