Dear Editor

Optimal age for screening for familial HCM.

McKenna and Behr´s recommendation that offspring of carriers of mutations causing hypertrophic cardiomyopathy (HCM) should have regular evaluation “during puberty and early adulthood” could be read as suggesting that evaluation for carrier status is not necessary before puberty. This would be a pity as there is now good evidence arguing for echocardiographic screening of children of affected parents at a much earlier age. McKenna and Behr quote an annual incidence of sudden death from 2-4% in children diagnosed before 14 years of age, but neglected to point out that some of the children in these studies received drug therapy. The only complete geographically based cohort study of childhood hypertrophic cardiomyopathy published to date in fact demonstrates a higher mortality in untreated patients, with an annual mortality of 6.6% when heart-failure related deaths are included, and that even completely asymptomatic children had a 3.5% annual mortality in sudden death.[1] In relation to timing for family screening the majority (6/10) of sudden deaths occurred in the pre- pubertal and early pubertal age of 9-13 years.[1] An expanded follow-up study including 122 patients from six centres with a geographical referral pattern confirms that a majority of unexpected sudden deaths occur in the 8-11 years age band.[2] Such an age distribution of sudden deaths in childhood suggests that family screening should take place well before the offspring reaches 8 years of age, and probably before the child starts school. Using relative rather than absolute wall-thickness for screening greatly increases the possibility of diagnosing the abnormal phenotype. Echocardiographic wall-to-cavity ratios which show very high specificity and sensitivity for identifying an abnormal phenotype already in young children are septum-to-cavity ratio >0.26 at any age and/or a systolic LV-wall-to-cavity ratio >0.63 at ages > 4years of age.[3] The same ratios, and also diastolic LV wall-to-cavity ratio are more sensitive measures to monitor any progression of disease in a growing child than the absolute wall thickness which should of course normally be increasing with age-related growth.[4] The above considerations therefore would argue for routine first time echocardiographic screening of offspring to take place no later than the four to six year age range.

Can treatment improve survival in childhood?

Screening of asymptomatic patients is clearly only worthwhile if high risk patients can be identified already in childhood, and a treatment that reduces risk is available. The data from the six centre cohort study shows that both echocardiographic (including wall-to-cavity ratios) and electrocardiographic measures of increasing hypertrophy correlates with risk for childhood death.[1,2] McKenna and Behr qualify their quotation of the study showing a 5-10 fold reduction in risk of death achieved by high dose propranolol in childhood HCM[1] by suggesting that HCM associated to Noonan´s syndrome is likely to have a different risk of sudden death than familial HCM related to sarcomeric contractile protein mutations. However, the study in fact clearly shows that the annual sudden death rate in HCM with Noonan´s syndrome was identical to that in patients with familial or sporadic HCM.[1] Thus, since the morphological appearance of the myocardium in HCM associated with Noonan´s syndrome is identical to that of ordinary familial HCM,[5] and it fits the WHO definition just as well as the HCM seen with sarcomeric contractile protein mutations, there is no reason to put HCM associated with Noonan´s syndrome in quotation marks and pretend it is irrelevant just because we still have not defined the causative mutation. This is particularly so as there are many kindreds with familial HCM who also lack the usual sarcomeric protein mutations and some new mutations described that are unrelated to contractile proteins.[6] McKenna and Behr have also missed the pertinent finding in this study that propranolol showed statistically significant protection against risk for fatal outcome both in HCM associated with Noonan´s syndrome and in HCM not associated with Noonan´s syndrome, i.e. the effect of propranolol appeared independent of the type of causative mutation.[1] It is also very clear in Methods that the non-Noonan HCM group did not include any patients with secondary HCM; the majority had familial HCM, and a few were new sporadic mutations not associated with Noonan´s syndrome. The Oxford familial childhood cases included in this study have all been subject to genetic evaluation, which has shown a variety of HCM-inducing mutations of sarcomeric contractile proteins (E. Blair, I. Östman-Smith & H. Watkins, unpublished data). Thus there is no evidence that the 5-10 fold reduction in risk conferred by high-dose propranolol in childhood1 is irrelevant to the mutations encountered in adult familial HCM. The protection by high dose beta-blockers has now also been confirmed in the enlarged six centre cohort study with 122 patients, which shows that high- dose beta-blocker therapy significantly reduces the risk both of sudden death and of heart-failure related death.[2]

What beta-blocker dose should be employed?

Although McKenna and Behr list beta-blockers as first line medical therapy, they give no guidelines in relation to dose. The data from the above studies suggest that conventional doses of beta-blockers (equivalent to 1-4 mg/kg BW of propranolol/day) do not protect against sudden death, in agreement with previous reports, whereas doses >5 mg/kg/day do protect.[1,2] Likewise, adult HCM patients treated with an average propranolol dose of 6 mg/kg day, titrated by finding a dose that blocked the effects of a standardised isoprenaline infusion, have a low annual mortality of 0.5%.[7] Our experience is that the best way to identify the appropriate dose for an individual is to use 24 Holter ECG to assess the efficacy of beta-receptor blockade. Some young individuals with particularly severe HCM will need up to 20 mg/kg BW/day to achieve effective beta-blockade.[1] Why should really effective beta-receptor blockade be so important? There are experimental data in mice that show that over-expression of the cardiac beta-adrenoceptor G-protein adenyl cyclase pathway results in a late developing HCM that can be prevented by propranolol therapy.[8] Interestingly, over-expression of this pathway also potentiates and accelerates cardiomyopathy due to alpha-myosin heavy chain mutations,[9] i.e. significantly worsens the clinical disease expression in our closest animal model to contractile protein HCM in humans. In humans we know that subjects with clinically overt HCM have increased activity of their cardiac sympathetic nerves,[10] so it would seem that this could initiate a vicious cycle of progressive hypertrophy, as occurs in the mice. The fact that so many different mutations in at least nine different genes encoding contractile proteins, and one affecting AMP-activated protein kinase, can cause essentially the same phenotype argues strongly for the fact that the hypertrophy that is the essential definition of disease expression is a secondary compensatory response, either directly from a trophic effect of increased sympathetic nervous activity,[11,12] or secondary to inability to maintain normal ATP levels.[6] Either way, effective beta-receptor blockade could have a crucial role in modifying disease-expression, either by protecting against the trophic effects of noradrenaline, or by improving ATP-conservation. That beta-blocker therapy can reduce compensatory hypertrophy has been shown in several experimental models.[13,14] Thus the efforts at modifying disease progression must proceed in parallel to efforts at protecting high risk patients from sudden death with ICDs, because progressive disease with severe hypertrophy can lead to heart failure, both early due to impaired diastolic filling, and late due to dilated, fibrotic endstage.

References

(1) Östman-Smith I, Wettrell G, Riesenfeld T. A cohort study of childhood hypertrophic cardiomyopathy: improved survival following high- dose beta-adrenoceptor antagonist treatment. J Am Coll Cardiol 1999;34:1813-22.

(2) Östman-Smith I, Wettrell G, Keeton B, Riesenfeld T, Holmgren D, Ergander U. Risk factors for disease-related death in childhood hypertrophic cardiomyopathy. Cardiol Young 2001;11 Suppl. 1:31.

(3) Devlin AM, Östman-Smith I. Diagnosis of hypertrophic cardiomyopathy and screening for the phenotype suggestive of gene carriage in familial disease: a simple echocardiographic procedure. J Med Screen 2000;7:82-90.

(4) Östman-Smith I, Devlin AM. A simple method for assessing the regression or progression of ventricular hypertrophy in the growing child and adult: the value of left ventricular wall-to-cavity ratios. Eur J Echocardiography 2001;2:22-30.

(5) Burch M, Mann JM, Sharland M, Shinebourne EA, Patton MA, McKenna WJ. Myocardial disarray in Noonan syndrome. Br Heart J 1992;68M:586-8.

(6) Blair E, Redwood C, Ashrafian H, Oliveira M, Broxholme J, Kerr B, et al. Mutations in the gamma(2) subunit of AMP-activated protein kinase cause familial hypertrophic cardiomyopathy: evidence for the central role of energy compromise in disease pathogenesis. Hum Mol Genet 2001;10:1215- 20.

(7) Canedo MI, Frank MJ. Therapy of hypertrophic cardiomyopathy: medical or surgical? clinical and pathophysiologic considerations. Am J Cardiol 1981;48:383-8.

(8) Asai K, Yang GP, Geng YJ, Takagi G, Bishop S, Ishikawa Y, et al. Beta- adrenergic receptor blockade arrests myocyte damage and preserves cardiac function in the transgenic G(s)alpha mouse. J Clin Invest 1999;104:551-8.

(9) Hardt SE, Geng YJ, Montagne O, Asai K, Hong C, Yang GP, et al. Accelerated cardiomyopathy in mice with overexpression of cardiac G(s)alpha and a missense mutation in the alpha-myosin heavy chain. Circulation 2002;105:614-20.

(10) Brush JE, Eisenhofer G, Garty M, Stull R, Maron BJ, Cannon RO, et al. Cardiac norepinephrine kinetics in hypertrophic cardiomyopathy. Circulation 1989;79:836-844.

(11) Östman-Smith I. Cardiac sympathetic nerves as the final common pathway in the induction of adaptive cardiac hypertrophy. Clin Sci 1981;61:265- 272.

(12) Östman-Smith I, Wettrell G. Pathophysiology of hypertrophic cardiomyopathy. Lancet 2000;355:928.

(13) Östman-Smith I. Reduction by oral propranolol treatment of left ventricular hypertrophy secondary to pressure-overload in the rat. Br J Pharmacol 1995;116:2703-9.

(14) Östman-Smith I. Reduction by beta-adrenoceptor blockade of hypoxia- induced right heart hypertrophy in the rat. Br J Pharmacol 1995;116:2698- 702.

Dear Editor

In their study, Jenni et al describe the echocardiographic characteristics of a relatively rare cardiomyopathy, isolated ventricular non-compaction. As stated by the authors, this is a disease entity that is difficult to diagnose during life, and the improvement in image quality of modern echocardiography will probably allow to improve its early detection. In the last two years, our group diagnosed ventricular non compaction by magnetic resonance (MR)imaging in several patients in whom no suspicion of this disease existed after routine echocardiography and cardiac catheterisation. Even in retrospect, it was impossible to detect the abnormalities described by Jenni et al in several of these patients. MR however clearly showed localised thickening of the left ventricular wall ("spongy myocardium" with intramyocardial caverns and excessive trabeculation. Dynamic MR demonstrated the high speed expulsion of blood from the caverns during systole which was not visible on color doppler.

In our series of 7 patients, the reason for referral was heart failure in 4 and ventricular arrhythmias in 1. One patient was examined because he was the father of a baby with heart failure and proven non compaction. Surprisingly, in 2 other patients, typical signs of ventricular non compaction were found even though the referral diagnosis was myocardial infarction or atrial septal defect. The MR characteristics were increased thickness of parts of the left ventricle of 1.3 up to 1.7 cm in diastole and an irregular endocardial surface with increased trabeculation. In all cases, the thickening was seen in the apex and part of the lateral wall. Basal portions of the ventricle were always normal. On dynamic imaging (gradient echo sequence) blood was seen to be expulsed from caverns in the thickened myocardium. On T1 weighted spin echo imaging, the thickened myocardium had a peculiar aspect comparable to the echo finding of Jenni : the endocardial part of the myocardium showed an important increase in signal intensity in comparison with the epicardial part of the muscle the normal myocardium and the normal myocardiyum. This is typical for imaging of slowly flowing blood : the spin echo images were made in mid diastole, a part of the heart cycle with less movement of blood in these caverns.

On the basis of these findings, we would suggest to incorporate an MR study in suspected cases of heart failure cardiomyopathy, apical hypertrophy, and important ventricular arrhythmias. Larger cooperative studies are needed to further delineate the role of MR in this difficult diagnosis.

Dear Editor

The question about how we should manage an asymptomatic individual with Brugada syndrome (BS) is not easy to answer. In the case reported, the patient has remained asymptomatic for years (first ECG in 1987) but despite of this, he is carrier of a disease with demonstrated association with arrhythmic death, sometimes the first manifestation in this population. Also, sudden death (SD) related to the syndrome is higher in symptomatic families. In other genetically determined diseases, a family history of SD is a predictive risk factor of SD. This has not been proven in BS.[1]

There is no doubt that symptomatic patients must be protected with an implantable cardioverter defibrillator (ICD). Asymptomatic individuals, either with spontaneously abnormal ECG as the case discussed, or with abnormal ECG after administration of drugs, should undergo an electrophysiologic study (EPS). If inducible, then an ICD should be recommended. HV measurement is of interest as up to 50% of BS have a prolonged HV reflecting conduction disturbances.[2] Pharmacologic testing with an intravenous sodium channel blocker should be performed in patient's relatives for screening of possible carriers of the disease. There is a 100% correlation between the results of the test and the genotype.[3]

New data on asymptomatic individuals with a follow up of 3 years have been recently presented.[4] A total of 16 SDs occurred in 190 asymptomatic patients with BS (111 had spontaneously abnormal ECG and 79 had abnormal ECG after administration of an antiarrhythmic drug). All SDs occurred in the group with spontaneously abnormal ECG. EPS was performed in 136 subjects, of these, 45 were inducible and 91 were not. Six SDs occurred in inducible patients (6/45, 13.33%) and 1 SD in those not inducible (1/91, 1.09%). In 9 patients with SD no EPS was performed. Therefore, inducible patients with spontaneously abnormal ECG have an increased risk of SD (13.33%) compared to those not inducible (1%). But still risk does exist in this latter group. There is also the limitation of the 9 SDs in individuals in whom EPS was not performed.

It is a fact that asymptomatic individuals may become symptomatic during follow up and at present we don't know how to identify them. Actually all symptomatic patients were carriers of the disease and asymptomatic until the moment they developed the first symptoms. Efforts should be made to find out predictive markers of SD in this subgroup. But, no matter the result of the EPS, the threshold for ICD implantation should be low in asymptomatic patients with a malignant family history. No deaths have occurred in patients implanted with an ICD. Some questions still remain. Is previous syncope predictive of SD? Being the peak incidence of symptoms around the fourth decade of life, is there a higher risk in younger patients as compared to the older ones? Is the magnitude of the ST segment elevation related to the occurrence of arrhythmias? Also, taking into account the data presented before, it seems that individuals with spontaneously abnormal ECG are at higher risk than those with abnormal ECG only after drug challenge. Does the fact that up to 40% of symptomatic patients normalise the ECG during follow up have any predictive value?

Finally, we would like to discuss the study carried out by Hermida et al.,[5] pointed in the article as showing 0% cardiovascular mortality of asymptomatic patients with BS over a follow up of 49 months. In this study, diagnosis of BS was made on the basis of the ECG pattern (ST segment elevation of at least 0.1 mm in leads V1-V3). Prevalence was found to be 6.1% (61 of 1000 subjects analysed) much higher than the prevalence reported by other studies. Only 1 of the 61 individuals exhibited a coved- type ST segment elevation (shown in the publication), characteristic of BS and similar to the first cases reported in 1992. This subject had a history of unexplained syncopes and resulted inducible in the EPS. The other 60 individuals had an ECG with a saddle back-type ST segment elevation (one of these is also shown in the article, and looks only "suspicious" of BS). As the authors say, probably not all the subjects with saddle back-type ST segment elevation had BS (early repolarization may show a similar pattern). These individuals should have been exposed to pharmacological testing to confirm or exclude the diagnosis. It seems reasonable to think that an overestimation of BS occurred in the study population, so the lack of mortality can be related to the fact that only a small number of the subjects really had BS.

References
(1) Brugada P, Brugada R, Brugada J. Sudden death in high risk family members: Brugada syndrome. Am J Cardiol 2000;86(suppl):40K-43K.
(2) Brugada P, Brugada J. Right bundle branch block, persistent ST elevation and sudden cardiac death: a distinct clinical and electrocardiographic syndrome. J Am Coll Cardiol 1992;20:1391-6.
(3) Brugada J, Brugada R, Wang Q, et al. Unmasking of genotype and phenotype abnormalities by administration of class I antiarrhythmic drugs in patients with sudden death (abstract). Eur Heart J 1998;19:557.
(4) Brugada J, Brugada R, Antzelevitch C, et al. Arrhythmic events in patients with right bundle branch block and ST segment elevation in right precordial leads (abstract). Eur Heart J 2001;22:17.
(5) Hermida J, Lemoine J, Bou Aoun F, et al. Prevalence of the Brugada syndrome in an apparently healthy population. Am J Cardiol 2000;86:91-4.