Accuracy of European diagnostic criteria for familial hypertrophic cardiomyopathy in a genotyped population

International Journal of Cardiology 90 (2003) 33–40 www.elsevier.com / locate / ijcard

Accuracy of European diagnostic criteria for familial hypertrophic cardiomyopathy in a genotyped population a,g ,

b

a

b

c

d

P. Charron *, J.F. Forissier , M.E. Amara , O. Dubourg , M. Desnos , J-B. Bouhour , ´ ¨ a,g , P. Richard e,g , K. Schwartz f,g , M. Komajda a,g R. Isnard a,g , A. Hagege c , A. Benaıche a ˆ ´ ˆ ` , Paris, France Pitie-Salpetriere Service de Cardiologie, Hopital ˆ Ambroise Pare´ , Boulogne, France Service de Cardiologie, Hopital c ˆ ´ Georges Pompidou, Paris, France Europeen Service de Cardiologie, Hopital d ˆ Laennec, Nantes, France Service de Cardiologie, Hopital e ˆ ´ ˆ ` , Paris, France Pitie-Salpetriere Service de Biochimie B, Hopital f ˆ ´ ˆ ` , Paris, France Pitie-Salpetriere INSERM U523, Hopital g ´ ´ Institut Federatif de Recherche no 14., France b

Received 14 March 2002; received in revised form 19 August 2002; accepted 29 September 2002

Abstract Background: Since the sensitivity of conventional diagnostic criteria for familial hypertrophic cardiomyopathy (HCM) is low, new diagnostic criteria were proposed by a European collaboration. However, their diagnostic value remains unknown. The aim of the study was to evaluate the accuracy of these new criteria, using the genetic status as the criterion of reference. Methods: We studied 109 genotyped adults (54 genetically affected, 55 unaffected) from 7 families (mutations in 3 genes). Major European echographic criteria were a maximal wall thickness $13 mm or $15 mm according to the segment involved, or the presence of SAM. Major European ECG criteria were abnormal Q waves, left ventricular hypertrophy, or marked ST-T changes. Combined major / minor European criteria were also evaluated. Results: Sensitivity and specificity of major European criteria (72 and 92%, respectively) were similar to those of major conventional criteria (70 and 94%) and were not improved by combined major / minor European criteria (72 and 90%). When all the minor European criteria were considered, sensitivity increased to 87% but specificity dramatically decreased to 51%. However, one of these minor ECG criteria, deep S V2, was of interest and when added to major European criteria, sensitivity increased to 76% and specificity remained good (90%). Conclusions: The diagnostic value of new European criteria for HCM was evaluated for the first time. We found that it was not different from that of conventional criteria, with a good specificity but a low sensitivity. Additional criteria should be studied to improve the early identification of HCM.  2002 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Hypertrophic cardiomyopathy; Genetics; Diagnosis; Echography

1. Introduction Hypertrophic cardiomyopathy (HCM) is a heart muscle disorder characterized by unexplained ven* Corresponding author. Service de Cardiologie, Hopital ˆ ´ Pitieˆ ` ˆ Salpetriere, 47 Blvd de l’Hopital, 75856 Cedex 13, Paris, France. Tel.: 133-1-4216-2911; fax: 133-1-4217-6800. E-mail address: [email protected] (P. Charron).

tricular hypertrophy, which remains a leading cause of sudden death in young adults, especially in athletes [1–3]. The disease is usually familial, with an autosomal dominant inheritance, and recent developments in the molecular genetics of HCM have identified several genes responsible for the disease [4–6]. Pedigree analyses has resulted in the recognition of highly variable morphologic manifestations of

0167-5273 / 02 / $ – see front matter  2002 Elsevier Science Ireland Ltd. All rights reserved. doi:10.1016/S0167-5273(02)00534-X

34

P. Charron et al. / International Journal of Cardiology 90 (2003) 33–40

the disease and have also revealed that some adults who carry the causal mutation may not fulfil the conventional diagnostic criteria for HCM [7–9]. Indeed, a systematic evaluation of genotyped pedigrees found that up to 30% of adults who carry a mutation were ‘healthy’ with a normal echocardiography and ECG as regards conventional diagnostic criteria [9]. These ‘healthy carriers’ may however present mild abnormalities on echocardiography or ECG which may represent early expression of the disease [10]. In the context of a familial and autosomal dominant disease, the probability for a first-degree relative to carry the mutation is 50%. The presence of mild and unexplained abnormalities in this context may therefore indicate a high probability of carrying the mutation. That is why a European collaboration recently proposed new diagnostic criteria for familial HCM in adults [11]. These criteria might improve the identification of carriers of a mutation, and therefore might improve the identification of subjects at high risk of developing the disease later. Since the diagnostic value of these new criteria remains unknown, the aim of the present work was to evaluate for the first time the accuracy (sensitivity, specificity, predictive values) of these European echographic and electrocardiographic diagnostic criteria, in a genotyped population and using the genetic status as the criterion of reference.

2. Methods

2.1. Population In seven families with HCM, the causative mutation had been previously identified in three genes: MYH7 (R723C, N232S, R403L, E483 K), MYBPC3 (SASint.20, SDSint.23, E1096Stop) and MYL2 (F18L) [12,13]. Clinically affected subjects with HCM and all their first degree-relatives in these seven families were screened for inclusion in the study. One-hundred and eleven adults were included. Written informed consent was obtained from each patient, in accordance with a study protocol approved by the Comite´ d’Ethique du centre Hospitalier Universitaire ˆ ` (Paris), and which conforms to ´ de la Pitie-Salpetriere the ethical guidelines of the 1975 Declaration of

Helsinki. Clinical evaluation, ECG and echocardiography were obtained at the time of the genetic inquest. Examinations were analyzed independently by two observers who had no knowledge of the genetic status. Two subjects were excluded because of systemic hypertension which could interfere with HCM, leaving 109 adults for analysis. Among them, 54 were genetically affected and 55 genetically unaffected (used as the control population).

2.2. Echographic procedure and criteria Subjects underwent M-mode, two-dimensional echocardiography and Doppler ultrasonography according to a previously described standardized procedure [9,14,15]. Images were stored on VHS videotape for subsequent analysis. Hypertrophy was assessed on 2D echocardiography from the parasternal short axis view, at both the mitral and papillary muscle levels and measurements performed for four different segments: anterior septum, posterior septum, lateral free wall, infero-posterior wall [15]. For each location, three measurements were made and averaged. All the measurements were performed in a core laboratory. The maximal wall thickness (MWT) from any location was thus determined. The reproducibility of the measurements was previously reported to be satisfactory [15]. Other views were also used to integrate observations from short axis views. Conventional and European criteria are described in Table 1 [9,11]. These European criteria take into account a different cut-off for MWT according to the segment involved, because the resolution is better in the Table 1 Echographic diagnostic criteria for familial HCM Conventional echo criteria

European echo criteria a

Major MWT.13 mm

Major MWT$13 mm in anterior septum or posterior wall MWT$15 mm in posterior septum or free wall or severe SAM (septal-leaflet contact)

Minor MWT513 mm

Minor MWT$12 mm in anterior septum or posterior wall MWT$14 mm in posterior septum or free wall or moderate SAM (no septal-leaflet contact) or redundant MV leaflets

a Combined European criteria are defined by the presence of two minor Echo criteria; or one minor Echo with two minor ECG criteria.

P. Charron et al. / International Journal of Cardiology 90 (2003) 33–40

35

Table 2 ECG diagnostic criteria for familial HCM Conventional ECG criteria

European ECG criteria a

Major Abnormal Q waves (.40 ms or .1 / 3 R wave) in at least two leads T wave inversion ($3 mm) in at least two leads LV Hypertrophy (Romhilt-Estes score$4)

Major Idem as conventional criteria

Minor Left atrial enlargement (P wave in V1) PR interval ,120 ms Microvoltage (,5 mV) Minor Q waves in at least two leads Bundle branch block or hemiblock

Minor Deep S V2 (.25 mm) Minor repolarization changes Complete BBB or interventricular conduction defect (QRS$120 ms)

a

Combined European criteria are defined by the presence of two minor Echo criteria; or one minor Echo with two minor ECG criteria.

segments perpendicular to the ultrasound beam and lower in the segments which are parallel.

ent, although the difference was low for the posteroinferior wall (10.162.6 vs. 9.0 mm, P50.005).

2.3. ECG procedure and criteria

3.2. Major diagnostic criteria

Standard 12-leads ECGs were performed with subjects in the supine position during quiet respiration. Conventional and European ECG criteria are described in Table 2 [9,11].

In the G1 group, the major conventional echographic criterion was present in 34 subjects and the major ECG criteria found in 28 subjects. Because of dissociation between ECG and echography, the combination of echography and ECG identified 38 of the 54 genetically affected subjects (70%). Indeed, 10 of the 34 subjects with echographic criterion had no major ECG abnormalities and conversely, 4 of the 28 subjects with ECG criteria had no major echographic abnormality. The major European criteria in the G1 group were present in 34 subjects (only one was different from the subjects with Conventional criteria). The combination of European echographic criteria and ECG criteria identified 39 of the genetically affected subjects (72%). The diagnostic values of conventional and European major criteria

2.4. Statistical analyses Sensitivity, specificity, positive and negative predictive values were determined for the different criteria. Continuous data were expressed as mean6S.D. and were analyzed with unpaired twotailed t tests. Discrete variables were compared by x 2 tests. Differences were considered to be statistically significant when P,0.05.

3. Results

3.1. Population Clinical features of the 109 adults are described in Table 3. Mean age, sex ratio and body surface area were similar in genetically affected (G1) subjects (n554) and in genetically unaffected (Go) subjects (n555). The maximal wall thickness (MWT) on echocardiography was 16.4 mm66.9 in the G1 group versus 9.9 mm61.2 in the Go group (P, 0.0001). For each of the four segments of left ventricle, the thickness was also significantly differ-

Table 3 Clinical features of the population n

Go 55

G1 54

P value

Age (years) Sex (M / F) BSA (m 2 ) MWT (mm) Anterior septum Posterior septum Lateral wall Posterior wall

37.7614.7 30 / 25 1.7860.20 9.961.2 9.561.4 8.861.3 8.961.2 9.061.3

37.7617.9 27 / 27 1.7260.16 16.466.9 16.067.1 13.064.6 12.663.8 10.162.6

NS NS NS ,0.0001 ,0.0001 ,0.0001 ,0.0001 0.005

Go, genetically unaffected; G1, genetically affected; BSA, body surface area; MWT, maximal wall thickness.

P. Charron et al. / International Journal of Cardiology 90 (2003) 33–40

36

Table 4 Diagnostic value of major criteria for HCM in adults (n5109)

Table 6 Diagnostic value of minor criteria for HCM in adults (n5109)

Criteria

Sensitivity

Specificity

PPV

NPV

Conventional Echo ECG Echo or ECG

63% 53% 70%

100% 94% 94%

100% 90% 93%

73% 66% 75%

European Echo ECG Echo or ECG

63% 53% 72%

98% 94% 92%

97% 90% 91%

73% 66% 76%

PPV and NPV, positive and negative predictive values.

are indicated in Table 4. Sensitivity and specificity were similar. No false positive was observed with the conventional echographic criterion but one false positive subject with European echographic criteria (MWT of 13 mm in anterior septum in a 60-year-old women). Three false positive were observed with ECG in 3 women because of negative T waves (n52) or left ventricle hypertrophy (n51). The diagnostic value was also analyzed according to the gene involved (Table 5). The results were similar for the three genes.

3.3. Analysis of minor criteria Apart from major conventional criteria, the minor conventional echographic criterion was present in two additional subjects (one G1 and one Go) and a minor conventional ECG criterion in six additional subjects (three G1 and three Go). Four genetically affected subjects were therefore identified by minor criteria (ECG or echography), but there was also four false positive subjects. Sensitivity increased to 78% but specificity decreased to 87% (Table 6). Considering European criteria, and apart from major criteria, a

Criteria

Sensitivity

Specificity

PPV

NPV

Conventional M. or m. M. or SV2

78% 74%

87% 92%

86% 91%

79% 77%

European Combined M. or m. M. or SV2

72% 87% 76%

90% 51% 90%

89% 64% 89%

76% 79% 78%

PPV and NPV, positive and negative predictive values; M. and m., presence of at least one major or minor criteria; combined, two minor echo criteria or one minor echo1two minor ECG criteria; M or SV2, presence of one major criterion or deep SV2 on ECG.

minor echographic criterion was present in four additional subjects (one G1 and three Go, all with a MWT512 mm in anterior septum). One of these four subjects had another minor echographic criterion (presence of SAM) but he was genetically unaffected. None of the three other subjects had two additional minor ECG criteria, so no additional genetically affected subject was identified by the combination of minor / major European criteria (but there was one additional false positive subject). Diagnostic values of European criteria were therefore not improved (Table 6). When the presence of only one of all the minor European criteria was considered, sensitivity increased to 87% but specificity decreased dramatically to 54% (Table 6). However, one of the minor European ECG criterion, deep SV2, provided interesting information. It was present in four additional subjects, including three G1 subjects and one Go subject. Sensitivity and specificity of major European criteria or deep SV2 were, respectively, 76 and 90%. Similarly, sensitivity and specificity of major Conventional criteria or deep SV2 were 74 and 92%.

Table 5 Diagnostic value of Conventional and European criteria according to the gene involved Gene

Total (n5)

G1

Go

Criteria

Se

Sp

PPV

NPV

MYBPC3

47

19

28

Convent. European

68% 68%

92% 92%

87% 87%

80% 80%

MYH7

44

28

16

Convent. European

71% 75%

100% 93%

100% 95%

65% 67%

MYL2

18

7

11

Convent. European

71% 71%

90% 90%

83% 83%

82% 82%

G1, genetically affected; Go, genetically not affected; MYBPC3, cardiac myosin binding protein C gene; MHY7, cardiac myosin heavy chain gene; MYL2, regulatory myosin light chain. Se, sensitivity, Sp, specificity, PPV and NPV are positive and negative predictive values. Criteria are major Conventional or European criteria (echography or ECG).

P. Charron et al. / International Journal of Cardiology 90 (2003) 33–40

4. Discussion In the present work, we evaluated for the first time the diagnostic value of new criteria for Hypertrophic Cardiomyopathy, proposed by a European collaboration made up of experts in this disease [11]. The issue is important because of the low sensitivity of conventional diagnostic criteria. Up to 30% of genetically affected adults are not identified by the conventional criteria [9] and are therefore called ‘healthy carriers’ of a mutation. No prospective follow-up of a large population of these ‘healthy carriers’ of a mutation are available but transversal data and case reports suggest that the large majority of this healthy carriers will develop an obvious form of HCM before the age of 50 years [2,9,11]. The improvement of the diagnosis of these ‘healthy carriers’ is important for many reasons. Firstly, it will lead to the identification of subjects at high risk of developing the disease later. A careful medical follow-up will be proposed to these subjects in order to diagnose early the cardiac expression of the disease (occurrence of left ventricular hypertrophy) and therefore to improve the therapeutic management of these patients. Restriction of physical activity might be discussed at this stage (especially for competitive athletes) because sudden deaths in the absence of hypertrophy were observed at least in one family with HCM [16]. Secondly, the healthy carriers may transmit the disease gene to offsprings (probability: 50% at each pregnancy in this autosomal dominant disease), and this information may be useful for them. The identification of healthy carriers of a mutation might be obviously performed by molecular genetics. However, the first step is to identify the responsible mutation in the proband within the family. This molecular process is very long, costly and uncertain because of the great genetic heterogeneity of the disease. Eleven different genes and more than 150 different mutations are described as responsible for the disease [4–6]. Several months, at least, are therefore required for molecular characterization. That is the reason why the improvement of clinical diagnostic criteria remains a major goal. We found that the diagnostic value of European criteria was very similar to that of the major conventional criteria, even when the combination of minor

37

and major European criteria were taken into account, with a low sensitivity (70–72%) and a good specificity (90–94%). The accuracy of major echographic European criteria was not improved although the rational for them was good (a different cut-of value for maximal wall thickness according to the segments because of a different resolution of echography in these different segments). The sensitivity of the combination of minor / major European criteria was not improved as compared to major conventional criteria, since minor echographic or ECG criteria were not present in the absence of major criteria, except for deep S wave in V2 which led to the identification of three additional genetically affected subjects (sensitivity: 14%). In contrast, many false positives were observed with minor European criteria (especially for minor repolarisation changes on ECG). Additional useful clinical criteria remain therefore to be identified to improve the diagnosis of FHC. This might be achieved either with a prospective and systematic analysis of multiple criteria, as was found in children with HCM [17], or by the use of new diagnostic tools, such as Doppler tissue imaging. Encouraging data were recently reported in a preliminary study with this latter technology [18].

Acknowledgements ´ This work was supported by INSERM (Reseau de Recherche Clinique N8 4R009B), the Association ´ ´ Franc¸aise contre les Myopathies, the Federation ´´ Franc¸aise de Cardiologie, and the Delegation a` la Recherche Clinique de l’Assistance Publique´ Hopitaux de Paris (Credits EMUL et IFR Coeur, Muscle et Vaisseaux). Ph. Charron is the recipient of ˆ a grant from Assistance Publique-Hopitaux de Paris.

Appendix A Coordinators of the French Network on Hypertrophic Cardiomyopathy are M. Komajda and K. Schwartz. Members are: L. Carrier, K. Schwartz ˆ ˆ ` Paris); P. Rich(INSERM U523, Hopital Salpetriere, ard, C. Ledeuil, B. Hainque (Service de Biochimie,

E. Arbustini, L. Tavazzi / International Journal of Cardiology 90 (2003) 38 – 40

38

ˆ ˆ ` Paris); P. Charron, A. Benaıche, ¨ Hopital Salpetriere, R. Isnard, M. Komajda (Service de Cardiologie, ˆ ˆ ` ´ Hopital Salpetriere, Paris); D. Heron, M. Gargiulo ˆ ˆ ` ´ ´ ´ (Departement de Genetique, Hopital Salpetriere, Paris); J.F. Forissier, O. Dubourg (Service de Carˆ ´ A. Hagege, ` diologie, Hopital A. Pare); M. Desnos ˆ ´ (Service de Cardiologie, Hopital europeen Georges ˆ Pompidou, Paris); JM Langlard, JB Bouhour (Hopital ˆ Laennec, Nantes); A. Millaire (Hopital Cardiologique, Lille); J. Feingold (INSERM U155, Lonchamps, Paris); L. Duboscq-Bidot; E. Villard (Ass. ˆ Claude Bernard et Universite´ Paris VI, Hopital ˆ ` Paris). Salpetriere,

References [1] Report of the 1995 World Health Organisation / International Society and Federation of Cardiology Task force on the definition and classification of cardiomyopathies. Circulation 1996;93:841–2. [2] Maron BJ. Hypertrophic cardiomyopathy, a systematic review. J Am Med Assoc 2002;287:1308–20. [3] Wigle ED, Rakowski H, Kimball BP et al. Hypertrophic cardiomyopathy: clinical spectrum and treatment. Circulation 1995;92:1680–92. [4] Bonne G, Carrier L, Richard P et al. Familial hypertrophic cardiomyopathy: from mutations to functional defects. Circ Res 1998;83:579–93. [5] Marian AJ, Roberts R. The molecular genetic basis for Hypertrophic cardiomyopathy. J Mol Cell Cardiol 2001;33:655–70. [6] Seidman JG, Seidman C. The genetic basis for hypertrophic cardiomyopathy: from mutation identification to mechanistic paradigms. Cell 2001;104:557–67. [7] Solomon ST, Wolff S, Watkins H et al. Left ventricular hypertrophy and morphology in familial hypertrophic cardiomyopathy associated with mutations of the b-myosin heavy chain gene. J Am Coll Cardiol 1993;22:498–505.

[8] Watkins H, McKenna W, Thierfelder L et al. Mutations in the genes for cardiac troponin T and a-tropomyosin in hypertrophic cardiomyopathy. New Engl J Med 1995;332:1058–64. [9] Charron P, Dubourg O, Desnos M et al. Diagnostic value of electrocardiography and echocardiography for familial hypertrophic cardiomyopathy in a genotyped adult population. Circulation 1997;96:214–9. [10] Hagege AA, Dubourg O, Desnos M et al. Familial hypertrophic cardiomyopathy: cardiac ultrasonic abnormalities in genetically affected subjects without echocardiographic evidence of left ventricular hypertrophy. Eur Heart J 1998;19:490–9. [11] McKenna WJ, Spirito P, Desnos M, Dubourg O, Komajda M. Experience from clinical genetics in hypertrophic cardiomyopathy: proposal for new diagnostic criteria in adult members of affected families. Heart 1997;77:130–2. [12] Charron Ph, Dubourg O, Desnos M et al. Clinical features and prognostic implications of familial hypertrophic cardiomyopathy related to the cardiac myosin binding protein C gene. Circulation 1998;97:2230–6. [13] Flavigny J, Richard P, Isnard R et al. Identification of two novel mutations in the ventricular regulatory myosin light chain gene (Myl 2) associated with familial and classical forms of hypertrophic cardiomyopathy. J Mol Med 1998;76:208–14. ` A et al. Doppler echocardiography in [14] Dubourg O, Isnard R, Hagege familial hypertrophic cardiomyopathy: the French cooperative study. Echocardiography 1995;12:235–40. [15] Spirito P, Maron BJ. Relation between extent of left ventricular hypertrophy and age in Hypertrophic Cardiomyopathy. J Am Coll Cardiol 1989;13:820–3. [16] Varnava A, Baboonian C, de Cruz L et al. A new mutation of the cardiac troponin T gene causing familial hypertrophic cardiomyopathy without left ventricular hypertrophy. Heart 1999;82:621–4. [17] Charron P, Dubourg O, Desnos M et al. Diagnostic value of electrocardiography and echocardiography for Familial hypertrophic cardiomyopathy in genotyped children. Eur Heart J 1998;19:1377– 82. [18] Nagueh SF, Bachinski LL, Meyer D et al. Tissue Doppler imaging consistently detects myocardial abnormalities in patients with hypertrophic cardiomyopathy and provides a novel means for an early diagnosis before and independently of hypertrophy. Circulation 2001;104(2):128–30.

Accuracy of the diagnostic criteria for hypertrophic cardiomyopathy: a major clinical need unmet by current tools a b, Eloisa Arbustini , Luigi Tavazzi * a

Molecular Diagnostic Division, IRCCS Policlinico San Matteo, P. le Golgi 2, 27100 Pavia, Italy b Cardiology Division, IRCCS Policlinico San Matteo, P. le Golgi 2, 27100 Pavia, Italy Received 22 August 2002; accepted 29 September 2002

*Corresponding author. Tel.: 139-382-503-157; fax: 139-382-503159. E-mail address: [email protected] (L. Tavazzi).

Typical hypertrophic cardiomyopathy (HCM) is easy to identify: the diagnosis coincides with the descriptive definition of the disease [1]. However

0167-5273 / 02 / $ – see front matter  2002 Elsevier Science Ireland Ltd. All rights reserved. PII: S0167-5273( 01 )00533-8