Left ventricular hypertrophy (LVH) detected by imaging techniques in the absence of common physiological or pathological causes is due in about 40-60% of cases to sarcomeric hypertrophic cardiomyopathy (HCM),1 an autosomal dominant trait caused by mutations in cardiac sarcomeric protein genes. The disease has an estimated worldwide prevalence of 1 in 500 individuals on the basis of the echocardiographic phenotype.1 It only affects the heart and is often benign throughout life, although it may also be associated with adverse outcomes, including sudden cardiac death (SCD).1,2 However, in about 5-10% of cases, unexplained LVH can be caused by other non-genetic or rarer genetic disorders that may mimic sarcomeric HCM, for some of which specific treatment is available. This is the case with Fabry disease (FD),3–5 a multisystem lysosomal storage disease caused by a deficiency of the enzyme alpha-galactosidase A (α-Gal A), encoded by the GLA gene on the X chromosome. The disease leads to accumulation of globotriaosylceramide (Gb3) and other glycosphingolipids in lysosomes, resulting in progressive multiorgan damage, with severe complications due to cardiac, renal or cerebrovascular lesions, and ultimately in decreased life expectancy.6–8 Although the classic form of FD affects multiple organs and occurs early in men (hemizygotes), some forms are found particularly (though not exclusively) in women (heterozygotes), manifesting as milder disease, with a late-onset phenotype that is often confined to a single organ such as the heart, kidney, or brain.9–11 The frequent observation of exclusively cardiac involvement in FD suggests that the heart is the most susceptible organ to α-Gal A deficiency.12–14 Cardiac involvement is one of the main determinants of prognosis,6,9,15,16 and includes LVH (the most common manifestation),17 arrhythmias, small-vessel coronary disease and heart failure. Electrocardiographic (ECG) abnormalities are frequent, and brady- or tachyarrhythmias are a significant cause of morbidity and mortality, including SCD.18,19

The diagnosis of FD is difficult and requires a high level of clinical suspicion. Many of the typical extracardiac features may not be evident in non-classical forms of the disease, and in some patients cardiac involvement may predominate, commonly mimicking sarcomeric HCM.20–22 Unlike the latter, LVH is typically concentric in FD,23,24 but it may also manifest as asymmetric septal hypertrophy or even predominantly with involvement of other wall segments.11,13,21,25–27 Left ventricular outflow obstruction (a common feature in sarcomeric HCM)1 may also exist in FD,21 hindering the differential diagnosis between the two conditions.

The accepted prevalence of FD in patients presenting with unexplained LVH is 0.5%-1%, although higher in some series,20,26,28,29 and its exclusion should be considered in this context.11,28,30 When initiated early, enzyme replacement therapy with agalsidase alpha or agalsidase beta can halt the progression of FD and modify its prognosis.3,4,31,32

In men, the first diagnostic test should be measurement of α-Gal A (in plasma, white blood cells or dried blood spots),33 which is generally decreased or absent, followed by GLA gene sequencing. In women, interpretation of enzyme activity results is difficult – only about 40% of female patients have a low enzyme level,34 and sequencing of the GLA gene is necessary for diagnosis.35

In the Portuguese Registry of Hypertrophic Cardiomyopathy (PRo-HCM),36 the possibility of FD was considered in the differential diagnosis of LVH. The aim of the present study was to analyze the registry data in order to assess the awareness of cardiologists regarding the need to exclude this phenocopy in real-world scenarios.

PRo-HCM was a national multicenter registry designed to collect information on the current approach to sarcomeric HCM in Portugal and to facilitate future improvements regarding diagnosis and therapeutic management of the condition.36 The registry was designed and implemented by the Working Group on Myocardial and Pericardial Diseases of the Portuguese Society of Cardiology (SPC), and centralized and managed at the SPC's National Center for Data Collection in Cardiology (CNCDC). It was an observational, multicenter, voluntary, non-mandatory study, with a two-year enrollment period (from 25 April 2013 to 25 April 2015), largely retrospective but also including a prospective update. Participating centers (n = 29) were asked to include all patients with a diagnosis of HCM followed at the center at that time or in the past (with no retrospective time limit), including those already deceased at the time of enrollment. Included individuals were aged over 18 years at inclusion and had a diagnosis of HCM with LVH phenotype by imaging methods (unexplained LVH with maximum left ventricular wall thickness of ≥15 mm in one or more myocardial segments in index patients).1,37 Criteria for non-inclusion in the registry were the presence of secondary LVH (at least stage 2 hypertension,38 moderate or severe aortic valve stenosis,39 or a previously diagnosed cardiac or systemic disease associated with LVH). However, the electronic case report form included some information about the possible exclusion of FD. This was a non-mandatory open-ended yes/no question, at the discretion of the investigator and not requiring any specific clinical or imaging characterization. However, if the answer was yes, a reference to the specific tests carried out for that purpose (measurement of α-Gal A activity and/or GLA gene sequencing) was requested.

The registry complied with the principles of the Declaration of Helsinki (October 2000) and written informed consent was obtained from all living patients before inclusion.

For the purpose of the present study, only index patients included in the registry were considered. This population was first identified and characterized according to whether sarcomeric genetic testing had been performed and the results thereof. They were then divided into three groups: group A – patients with positive sarcomeric genetic testing, in whom pathogenic or likely pathogenic mutation(s) were identified in genes encoding beta-myosin heavy chain (MYH7), myosin-binding protein C (MYBPC3), cardiac troponin I and T (TNNI3 and TNNT2), tropomyosin alpha-1 chain (TPM1), myosin light chain 3 (MYL3), myosin regulatory light chain 2 (MYL2), alpha cardiac actin (ACTC1) and muscle LIM protein (CSRP3); group B – those with negative sarcomeric genetic testing, in whom pathogenic or likely pathogenic mutation(s) were identified in none of the above nine genes; and group C – those in whom genetic testing in sarcomeric genes had not been performed. Each group was then characterized according to whether exclusion of FD had been considered (and methods used in its exclusion), associated extracardiac conditions, ECG abnormalities and echocardiographic (echo) characteristics. Finally, the groups were analyzed for the presence of specific features that could potentially constitute red flags for FD. These included clinical data, specific ECG abnormalities (short or prolonged PR interval, intraventricular conduction disturbances, or the need for pacemaker [PM] implantation for treatment of bradyarrhythmias), or concentric LVH on echocardiographic imaging. Data from group B (patients with no identified sarcomeric mutations) – the particular population of interest in this study – were compared with data from groups A and C.

In this descriptive, observational analysis of a subpopulation of 811 index patients included in the PRo-HCM registry with a diagnosis of HCM on the basis of classic criteria,37,40,41 genetic testing in sarcomeric genes was performed in ≈45% of patients, and a pathogenic or likely pathogenic mutation associated with HCM was identified in 35.5%.

A definitive diagnosis of sarcomeric HCM can only be made when a pathogenic or likely pathogenic mutation associated with the disease is identified. However, when genetic testing targeting the most common mutated genes is performed, findings are negative in up to 50% of cases.1 It may be hypothesized that older patients with a late HCM phenotype may have mutations in other (uncommon) sarcomeric genes that are not usually screened, but it is also possible that FD is the cause of that phenotype.

According to data from the registry, exclusion of FD was infrequent and relied on specific diagnostic tests in less than 40% of patients. When sarcomeric genetic testing was negative or was not performed, consideration of FD predominantly involved older women. Clinical presentation of FD may occur later and be more variable in women than in men, and women are more likely to develop a variant characterized by isolated cardiac involvement only.6–8,10 Interestingly, the GLA gene was most often tested in patients with identified sarcomeric mutations, probably because a large panel of genes was used; but it was the only gene tested in 13% of patients in group C, possibly due to economic restraints.

However, genetic testing for FD was not performed in more than 50% of female patients in group B, or in more than 70% of females in group C. Furthermore, α-Gal A activity was only assessed in a minority of patients, although, appropriately, mainly in males; but it was not assessed in the majority of men in groups B and C. Overall, specific tests for FD exclusion were performed in less than 50% of patients in whom the disease was reported as having been excluded. GLA testing results, when available, were negative.

These data support the conclusion that many cardiologists rely only on clinical or imaging data for exclusion of FD, and consider that in most patients with an HCM phenotype specific tests are not indicated or necessary. However, this may be a misconception. In patients with no male-to-male transmission of unexplained LVH, FD should be considered as a potential diagnosis and specific tests need to be performed, a decision supported by any other additional personal and/or family feature suggesting the diagnosis.35 Needless to say, in the absence of a family history of FD, such features in family members (kidney disease, acroparesthesias, cornea verticillata, etc.) may be absent, difficult to ascertain, or simply overlooked in patients with a predominantly or exclusively cardiac phenotype.

The ECG abnormalities observed in this analysis are commonly described in patients with LVH, whatever the cause, patients with sarcomeric HCM or FD included. A short PR interval is classically described in 21%-40% of patients with FD,42 particularly in its early stages but, although less common, it is also described in HCM (3.3% in the study population). Likewise, a tendency for PR prolongation and AV block is described in the later stages of FD,43 and these abnormalities may also occur during the natural history of sarcomeric HCM. They were found with similar prevalence in all groups of the study population. Conduction abnormalities may always be considered an additional warning for the cardiologist when excluding FD, but they are in no way a decisive diagnostic marker of FD.

All echo patterns of LVH can be observed in sarcomeric HCM.44–46 Concentric LVH (the most common form in FD) was described in less than 10% of patients in the study population. Conversely, obstruction (infrequent in FD) was diagnosed in about 40% of patients. Both echo features were observed similarly, whether or not sarcomeric genetic testing had been carried out, and regardless of its result, but none of them is key to the diagnosis.

Finally, in this population there was no clear evidence of non-cardiac features suggesting FD, but symptoms or signs may be subtle and non-specific in non-classical forms. The subpopulations studied were relatively elderly for sarcomeric HCM and potentially prone to what is termed Fabry cardiomyopathy – despite the rarity of this disease – particularly considering the female gender and the forms that can be expressed with a late and predominantly cardiac phenotype.13,20

The findings herein highlight the need to be alert for both conditions: sarcomeric HCM (the purpose of the PRo-HCM registry), which is often diagnosed late as it frequently evolves without symptoms; and FD, which needs to be systematically considered in the differential diagnosis of most cases of LVH, especially when the proband is negative for sarcomeric mutations.

In the genetic era, it seems reasonable to include GLA in the panel of genes to be tested when HCM is investigated. When genetic testing is not possible, α-Gal A activity should be assessed, because it may be diagnostic in male patients, with GLA testing being reserved as a second step if appropriate. In females, particularly if middle-aged or older and bearing in mind the possibility of a predominantly cardiac phenotype, GLA gene testing is advisable even if it is the only gene that can be tested due to economic constraints or other reasons.

LVH occurs late in FD, but enzyme replacement treatment can favorably modify the long-term prognosis of the disease, reducing the incidence of and/or progression to serious clinical events,3,4,31,32 and when a diagnosis is made in the index patient, early diagnosis in affected relatives is facilitated47 and may be associated with better outcomes.

Analysis of data from subpopulations of patients included in the PRo-HCM registry revealed a low awareness among cardiologists of the need to exclude FD in patients without a genetic diagnosis of sarcomeric HCM. When exclusion of FD was undertaken, it relied mainly on clinical or imaging features, with specific diagnostic tests performed in fewer than 50% of cases in which they could hypothetically be justified. These data, in a cardiology setting, should be taken as an indication for a broader approach to patients with unexplained HCM, aimed at the exclusion of a disease that, although rare, may benefit from specific therapy that has a favorable impact on prognosis.

The Portuguese Registry of Hypertrophic Cardiomyopathy was supported by the following companies (in alphabetical order): Jaba Recordati, Medinfar, Merck Serono, Sanofi Genzyme, Servier, Shire Human Genetic Therapies.

This article was sponsored by a grant from Sanofi Genzyme.

DB, NC, and LR-L conceived and designed the research, and drafted the manuscript; AB performed the statistical analysis; JM, LG, and HM performed critical revisions of the manuscript for drafting and key intellectual content. All authors contribute to revising the paper and gave final approval of the version to be published.

DB reports consultancy and lecture fees from Sanofi Genzyme, and attended meetings sponsored by Shire Human Genetic Therapies; LRL attended meetings sponsored by Shire Human Genetic Therapies and Sanofi Genzyme; the other authors have no conflicts of interest to declare.