Anderson-Fabry disease (AFD) is a rare X-linked disorder secondary to mutations in the GLA gene, resulting in a reduced activity of the lysosomal enzyme α-galactosidase A (α-Gal A) leading to the abnormal accumulation of globotriaosylceramide. Classic manifestations include angiokeratoma, cornea verticillata, and acroparesthesias, and main affected organs are the kidney, the heart, and the central nervous system [1]. At cardiac level, patients with AFD can present with cardiac hypertrophy, conduction disorders, arrhythmias, and microvascular angina. Cardiac problems currently are the leading cause of morbidity and mortality in patients with AFD [1].

AFD is associated with an important diagnostic delay [2]. The rarity of the disease (affects 1 in 40,000 to 170,000 individuals from the general population) and the lack of awareness among physicians are the principal factors for this diagnostic delay. Early diagnosis of AFD is important since appropriate therapies seem to be more effective when initiated promptly [1]. Screening of AFD in cardiac patients has been classically performed in patients with hypertrophic cardiomyopathy, where AFD has been reported in 1–2% of individuals. Additional efforts to diagnose AFD among other cardiac patients as those with conduction disorders requiring pacemaker implantation have proven not to be effective [3].

Coronary microvascular dysfunction has been described in AFD with and without cardiac hypertrophy and may represent the only manifestation of cardiac involvement in AFD patients [4]. Furthermore, an unexpectedly high prevalence of coronary spastic angina has been recently reported in AFD [5]. With these data, we hypothesized that AFD could be an underdiagnosed cause of angina and that patients with chest pain and non-obstructive or normal coronary arteries could be an appropriate group to perform screening.

For this reason, we designed a multicenter, cross-sectional study to investigate the prevalence of AFD in patients with chest pain and without obstructive coronary arteries (defined as any stenosis ≥ 70%).

A total of 663 patients (52.9% females, mean age 63 ± 11 years) who underwent coronary angiography (CA) or computer tomography angiography (CTA) were identified from databases at ten Spanish and one Portuguese hospitals. Consecutive patients fulfilling inclusion criteria who had undergone CA or CTA evaluation for the study of chest pain, between January 2016 and December 2017, were invited to participate. Males were screened by measuring α-Gal A activity and those with reduced levels were genetically tested for GLA mutations. All females were tested for mutations in GLA.

Baseline clinical characteristics and diagnostic tests are summarized in Table 1. Almost 16% of patients exhibited troponin levels above local reference upper limit and had the CA/CTA performed after seeking attention at an emergency department, whereas a cardiac stress test had been performed in 38% of patients before CA/CTA. In this regard, the treadmill stress test was the most frequently stress test (N = 210, 31.7%) and was reported positive in 51% of cases.

Table 1 Baseline clinical characteristics and diagnostic tests
Full size table

Left ventricular hypertrophy on echocardiogram (≥ 12 mm) was found in 31% of patients, with concentric hypertrophy in 19% of individuals. ECG repolarization abnormalities were present in 21% of individuals and 7% exhibited atrial fibrillation.

Genetic variants in GLA were identified in seven patients. Two polymorphisms (Arg118Cys and Asp313Tyr) were observed in six individuals with normal lyso-Gb3 levels while a known pathogenic variant (His125Tyr) was identified in a 52-year-old male with suppressed α-Gal A enzyme activity and high lyso-Gb3 levels. This patient had also end-stage kidney disease and severe LVH, which are both hallmarks of AFD.

According to our results, the overall prevalence of AFD in unselected individuals with chest pain and non-obstructive coronary artery disease is only 0.15% (1/663). Moreover, the patient diagnosed with AFD in our study could have been diagnosed previously if classical manifestations had been taken into consideration.

Chest pain is a subjective and frequent symptom and it is not clear which tests should be performed once coronary problems have been discarded. Our results suggest that AFD screening should not be routinely performed in this population, despite that it could be of interest in subjects who exhibit other AFD features like left ventricle hypertrophy or kidney disease.

In conclusion, our data suggest that AFD is not a frequent cause of chest pain without obstructive coronary artery disease and that screening efforts should not be conducted in this patient population unless AFD classical manifestations are present.