One of the major challenges in the clinical management of patients with hypertrophic cardiomyopathy (HCM) is the identification and treatment of the small number of individuals prone to serious events. The impact of new technologies for answering these problems is unclear. O’Hanlon et al. report on the significance of fibrosis detected by late gadolinium enhancement cardiovascular magnetic resonance for the prediction of major clinical events in HCM (7).
In this prospective study the authors analyzed the development of morbidity and mortality with respect to the presence and amount of myocardial fibrosis in 217 consecutive HCM patients with a mean follow-up of 3.1 ± 1.7 years. The pre-specified primary end point was the composite of cardiovascular death, unplanned cardiovascular hospital stay, sustained ventricular tachycardia or ventricular fibrillation, or appropriate implantable cardioverter-defibrillator (ICD) discharge.
Two separate secondary end points were predefined. A composite heart failure (HF) end point included unplanned HF hospital stay, progression to New York Heart Association (NYHA) functional class III or IV status, or HF-related death.
A composite arrhythmic end point included sustained ventricular tachycardia or ventricular fibrillation, appropriate ICD discharge, or SCD.
136 out of 217 patients (63%) showed fibrosis. In this group the mean percentage of fibrosis was 15.5% (range 1.4% to 54.9%). The analyzes of the conventional risk factors for SCD showed in the fibrosis group a greater percentage of patients with maximal wall thickness >30 mm (8.8% vs. 0%; p=0.0006) and non-sustained VT (11.8% vs. 3.7%; p=0.04) whereas no significant difference was seen in patients with family history of SCD (14.0% vs. 7.4%; p=0.14) and syncope (16.9% vs. 14.8%). The authors did not give results of abnormal blood pressure response during exercise.
Overall, the proportion patients with 2 or more risk factors for SCD was higher in the fibrosis group (19.1% vs. 8.6%). No difference was seen in the proportion of patients with rest LVOTO > 30 mmHg (28.7% of patients in the fibrosis group vs. 22.2%; p=0.30) – the numbers of patients with provocable gradients are not given.
Finally, the fibrosis group patients were more often symptomatic according to the NYHA classification (43.3% class II and 17.9% class III/IV vs. 34.6% class II and 7.4% class III/IV; p=0.01) and under medical treatment with betablocker (44.1% vs. 28.4%; p=0.02) and antiarrhythmics (16.9% vs. 7.4%; p=0.05).
Thirty-four of the 136 patients (25%) in the fibrosis group but only 6 of 81 (7.4%) patients without fibrosis reached the combined primary end point (hazard ratio [HR]: 3.4, p=0.006). Analyzing the single endpoints a significant difference was seen in the percentage of unplanned cardiovascular hospital admission (17.7% in the fibrosis group vs. 6.2%; p=0.036) whereas no significant differences were seen in CV mortality (5.9% in the fibrosis group vs. 1.2%; p=0.163), VT/VF (5.9% vs. 1.2%; p=0.131), and ICD discharge (1.5% in the fibrosis group vs. 0%).In the fibrosis group, the overall risk of the primary end point increased with the percentage of fibrosis present (HR: 1.18/5% fibrosis increase, 95% CI: 1.05 to 1.33, p=0.008). Every 5% increase in fibrosis increased the risk of reaching the combined primary endpoint by 15%.
A composite secondary HF endpoint was more often seen in the fibrosis group (24.5% vs. 9.9%; HR: 2.5, p= 0.021), and this risk increased as the extent of fibrosis increased (HR: 1.16/5% increase, p=0.017). All relationships remained significant after multivariate analysis. LAVi (HR: 1.021, 95% CI: 1.01 to 1.03, p<0.001) and LVOTO >30 mmHg (HR: 2.45, 95% CI: 1.2 to 4.9, p<0.013) are independently associated with HF endpoints, too.
A composite arrhythmogenic endpoint showed no significant increase in the fibrosis group (7.3% vs. 2.5%; HR: 3.15, 95% CI: 0.69-14.4, p=0.138). The extent of fibrosis (HR: 1.30, 95% CI: 1.05 to 1.61, p=0.014) and non-sustained ventricular tachycardia were univariate predictors for arrhythmic end points (sustained ventricular tachycardia or ventricular fibrillation, appropriate implantable cardioverter-defibrillator discharge, sudden cardiac death). Non-sustained ventricular tachycardia remained an independent predictor of arrhythmic end points after multivariate analysis, but the extent of fibrosis did not.
Therapeutic goals in the treatment of patients with HCM are prevention of cardiovascular (CV) and especially sudden cardiac death (SCD) on the one hand and improvement of clinical symptoms on the other hand. It has been shown by Maron et al. (4) that sudden cardiac death is the predominant mode of HCM related death in younger patients (mean age 45 ± 20 years) whereas heart failure causes mortality in elderly HCM patients (mean age 56 ± 19 years) and stroke is the main mode of HCM death in old patients (mean age 73 ± 14 years), mainly due to development of atrial fibrillation.
Furthermore, Varnava et al. could show in a post-mortem study of 75 HCM deaths that fibrosis was related to an increase of heart failure and non-sustained VT, whereas myocardial disarray was linked to premature death, SCD, and abnormal blood pressure response during exercise (9).
Since the introduction of late gadolinium enhancement (LGE) in MRI for the diagnosis of scar/fibrosis in patients with HCM there had been conflicting data about the use of LGE as a new risk factor of SCD in addition to the standard clinical risk stratification model of Elliott et al. (2). In the last months 3 papers discussing this conflict had been published (1,7,8).
O´Hanlon et al. used MRI guided diagnosis of fibrosis by late gadolinium enhancement in order to estimate the prognostic significance (7). The authors excluded patients with gradient-reduction therapies in contrast to Rubinshtein et al. (8). No study gave an exact description of the clinical risk factors according to the mentioned model (2).
There was low risk of SCD during follow-up without significant difference with respect to the detection of fibrosis, whereas there was a trend towards a higher incidence of total cardiovascular mortality due to heart failure death (7). These results confirm the study of Varnava et al. (9). Taking into account the mean age of O’Hanlon’s study (51.1 years) the findings correspond to the study of Maron et al. (4) who reported that heart failure death was more often found in elderly HCM patients. Rubinshtein at al. (8) and Bruder at al. (1) reported on higher incidence of SCD in patients with LGE. But, Mayo clinic data included gradient-reduction therapy in half of the patients (7). The given data of Bruder et al. are inconsistent with 1 observed SCD in the non-LGE group which is ignored in the published figures (1).
Fibrosis has been shown as the extract for promoting re-entrant ventricular arrhythmias and increased ventricular stiffness. O’Hanlon found more arrhythmogenic events in the fibrosis group but the data are not significant probably caused by the small cohort (7). Due to that result the authors point out that at present the presence or amount of fibrosis do not support the routine deployment of an ICD.
O’Hanlon et al. describe that myocardial fibrosis is an independent predictor of adverse outcome including heart failure (7). The main difference between patients with and without fibrosis is the higher incidence of unplanned cardiovascular hospital stay, heart failure with NYHA class III and IV, and heart failure death. Both the presence and the amount of fibrosis are of independent prognostic significance. Serial studies could show whether the amount of fibrosis would increase over time with increasing risk for HF.
In addition to the clinical risk stratification model it has been shown that LVOT-obstruction has negative impact on disease progression and survival (3). Including provocation maneuvers up to 70% of the HCM patients suffer from significant obstruction (5,6). Therefore, the non-use of provoking maneuvers in the baseline examinations and the low rate of resting obstruction (26.3%) is a disadvantage of the study (7). The 2 further papers (1,8) gave also no exact information about obstruction. The Mayo group did not describe the proportion of patient with and without obstruction (8), whereas Bruder et al. did not include the information of the type of obstruction (resting/provocable) (1).
O’Hanlon, R., A. Grasso, M. Roughton, J.C. Moon, S. Clark, R. Wage, J. Webb, M. Kulkarni, D. Dawson, L. Sulaibeekh, B. Chandrasekaran, C. Bucciarelli-Ducci, F. Pasquale, M.R. Cowie, W.J. McKenna, M.N. Sheppard, P.M. Elliott, D.J. Pennell, S.K. Prasad. J Am Coll Cardiol 2010;56 (E paper ahead)
Presented by Dr. Barbara Pfeiffer and Prof. Dr. Hubert Seggewiss
Medizinische Klinik 1, Leopoldina Krankenhaus Schweinfurt, Germany
European Society of Cardiology
European Heart HouseLes Templiers2035 Route des CollesCS 80179 Biot
06903, Sophia Antipolis, FR
Our mission: To reduce the burden of cardiovascular disease
© 2018 European Society of Cardiology. All rights reserved