The Heart in Friedreich AtaxiaClinical Perspective
Definition of Cardiomyopathy, Disease Severity, and Correlation With Neurological Symptoms
Background—This cross-sectional study provides a practical approach for the clinical assessment of Friedreich ataxia (FA) cardiomyopathy (FA-CM).
Methods and Results—A comprehensive cardiac assessment, including standard echocardiography, color Doppler myocardial imaging, cardiac magnetic resonance imaging, ECG, and exercise stress testing, was performed in 205 FA patients. To assess myocardial hypertrophy in FA-CM, the end-diastolic interventricular septal wall thickness (IVSTd) was found to be the best echocardiographic parameter compared with cardiac magnetic resonance imaging–determined left ventricular mass. With the use of this parameter, 4 groups of patients with FA-CM could be defined. Patients with normal values for IVSTd (31.7%) were classified as having no FA-CM. Patients with an IVSTd exceeding the predicted normal IVSTd were classified as having mild FA-CM (40%) if IVSTd exceeded the normal value by <18% or as having intermediate FA-CM (16.1%) if IVSTd exceeded the normal value by ≥18%. Patients with ejection fraction <50% were classified as having severe FA-CM (12.2%). In addition to increased myocardial mass, severe FA-CM was further characterized by dilatation of the left ventricle, reduced systolic strain rate of the posterior wall, and ECG abnormalities. Regional myocardial function correlated negatively with FA-CM groups. Younger patients had a tendency for more advanced FA-CM. Importantly, no clear correlation was found between FA-CM groups and neurological function.
Conclusions—We provide and describe a readily applicable clinical grouping of the cardiomyopathy associated with FA based on echocardiographic IVSTd and ejection fraction data. Because no distinct interrelations between FA-CM and neurological status could be determined, regular follow-up of potential cardiac involvement in FA patients is essential in clinical practice.
Friedreich ataxia (FA) is an autosomal recessive neurodegenerative disease caused by a defect in the gene encoding for the mitochondrial protein frataxin. It is a rare disease affecting ≈1 in 50 000 white persons.1–3 Myocardial involvement in FA is well documented, with concentric or (less frequently) asymmetrical left ventricular (LV) hypertrophy as the dominating cardiac finding.4–6 Cardiac dysfunction, predisposing to congestive heart failure and supraventricular arrhythmias, was the most frequent cause of death (59%) in a retrospective study in 99 patients with FA.7 Life expectancy in FA patients with cardiac involvement is reduced considerably to 29 to 38 years.7–9
Editorial see p 1591
Clinical Perspective on p 1634
Echocardiography is the routine imaging technique of choice for the evaluation and follow-up of cardiac involvement in patients diagnosed with FA (FA cardiomyopathy [FA-CM]), although cardiac magnetic resonance imaging (cMRI) has been shown to be more sensitive and accurate in determining the severity of cardiac remodeling in FA patients.10 The challenge with echocardiography is 2-fold: reliably detecting cardiac involvement and assessing the severity of FA-CM. To meet this challenge, precise evaluation of the extent of LV hypertrophy is mandatory, with LV end-diastolic wall thickness usually being the best acknowledged echocardiographic parameter. However, the clinical phenotypes and patterns of disease progression are highly variable in FA. Therefore, a single cutoff value for the evaluation and definition of FA-CM and the disease progression over time may not suffice.
This is the first study analyzing morphological and functional cardiac data from standard echocardiography, color Doppler myocardial imaging (CDMI), cMRI, exercise performance, and ECG from a large cohort of FA patients with wide ranges in age, neurological symptoms, and cardiac involvement.
The aim of this study was therefore to provide a clinical approach for the comprehensive assessment of FA-CM and to determine possible correlations of the degree of FA-CM with neurological disease symptoms and associated genetic factors.
Two hundred five patients with genetically confirmed FA participating in the Mitochondrial Protection With Idebenone in Cardiac or Neurological Outcome (MICONOS) study (https://www.clinicaltrials.gov; ID: NCT00905268) were studied. The MICONOS study was conducted between April 2006 and January 2010 and enrolled patients in 13 centers in 6 European countries. For genetic confirmation, patients had to present with extended GAA repeat length in the first intron of the frataxin gene on both alleles or with an extended GAA repeat length on one allele and a nonsense mutation on the second allele.1 Patients were excluded if (1) they had been treated with idebenone or coenzyme Q10 within the past month (n=1); (2) were pregnant and/or breast-feeding; (3) had clinically significant abnormalities of clinical hematology or biochemistry; or (4) had a past or present history of drug/alcohol abuse. Cardiac data including standard echocardiography, CDMI, ECG, cMRI, and a bicycle exercise test were acquired within 1 day during the baseline visit of the MICONOS trial. All patients gave prior written consent for imaging studies including digital data storage and systematic analysis of the data. The MICONOS study was approved by institutional ethics committees, and all investigations conformed to the principles outlined in the Declaration of Helsinki.
Standard Echocardiographic Measurements
LV end-diastolic and end-systolic dimensions and end-diastolic thickness of the posterior wall and the interventricular septum (echo-IVSTd) were measured with the use of standard M-mode echocardiographic methods from parasternal LV long-axis images. If appropriate, the anatomic M-mode procedure was used. Derived cardiac parameters, including relative wall thickness, LV myocardial mass, and ejection fraction (EF), were calculated as described in Table I in the online-only Data Supplement. Blood pool pulsed Doppler of the mitral valve inflow was used to extract the ratio of early to late diastolic flow velocity (E/A), the deceleration time, and the isovolumic relaxation time.
Color Doppler Myocardial Imaging
Real-time longitudinal 2-dimensional tissue Doppler data were recorded from the interventricular septum and the LV lateral, inferior, anterior, posterior, and anteroseptal walls with 4-, 2-, and 3- chamber apical views with the wall in the center of the sector. Longitudinal strain rates in the basal, mid, and apical segments of each wall were estimated by measuring the spatial velocity gradient. Strain rate profiles were averaged over 3 consecutive cardiac cycles and integrated over time to derive strain profiles with end-diastole used as the reference point (Speqle, KU Leuven, Leuven, Belgium). From the averaged strain rate and strain data, peak systolic strain rate and systolic strain were calculated. The intraobserver variability for peak systolic strain rate of the inferior septal and lateral segments was assessed on 24 CDMI data sets. The correlation coefficient for intrarater variability was high for both segments (septum: Pearson r=0.82; lateral: r=0.91). The weighted κ values for intrarater variability (septum: κ=0.62; lateral: κ=0.69) indicated satisfying agreement and low intrarater variability.11
In addition, the early diastolic velocity of the septal mitral annulus (E′) was extracted to assess the early diastolic transmitral velocity/mitral annular velocity ratio (E/E′). All standard echocardiography and CDMI data were analyzed centrally in a core laboratory by expert readers blinded to other cardiac data or patient characteristics (St. Georges Hospital, London, UK).
Cardiac Magnetic Resonance Imaging
Cine short-axis multislice images of the LV were collected with the use of the steady state free precision technique and analyzed blinded by a central core laboratory (BioClinica, Leiden, Netherlands) with the use of QMassMR software (version 6.1, Medis, Leiden, Netherlands). Epicardial and endocardial convex-shaped contours were drawn manually in the end-diastolic and end-systolic phases of the LV. Papillary muscles and trabeculations were excluded from the calculated LV mass.12 The density of myocardial tissue was set at 1.05 g/mL. Of note, LV masses analyzed by the steady state free precision method tend to be ≈15% lower than masses recorded by the turbo gradient echo method.13 Both the exclusion of papillary muscles and trabeculations as well as the steady state free precision method should be considered when LV masses are compared across studies.
Bicycle Exercise Test
Subjects underwent testing on a recumbent exercise bicycle, with subjects allowed to pedal with either legs or arms (Lode, Groningen, Netherlands) to assess exercise tolerance. After a 3-minute warm-up phase, subjects were instructed to pedal at 50 to 60 rpm for arm ergometry and 60 to 70 rpm for leg ergometry and to keep pedaling at a constant rate during the test. After warm-up, the resistance was increased in steps of 10 W every minute, and the peak exercise workload (in watts) was recorded. Subjects were advised that they were free to stop whenever they wished but were encouraged to continue for as long as possible. Heart rate and ECG were recorded during the exercise test for safety reasons, and the test would have been discontinued if chest pain, signs of ischemia, or arrhythmias developed.
Resting ECGs were performed in 191 patients, and the following parameters were extracted: heart rate, R-R interval, PR interval, QRS interval, QT time, and QT time corrected with Fridericia's formula (QTcF) or Bazett's formula (QTcB). In general, an abnormal ECG was defined as any of the following: ventricular or supraventricular arrhythmia or atrioventricular or interventricular conduction delay. All ECGs were analyzed by expert readers in blinded fashion in a central core laboratory (ganiMed, Villingen-Schwenningen, Germany).
Assessment of Neurological Function
Neurological function was assessed by the International Cooperative Ataxia Rating Scale (ICARS) by trained raters. The ICARS consists of 19 items addressing oculomotor function, speech, kinetic functions, posture, and gait.14,15 Patients who were able to walk at least 10 m with or without walking aids but without the help of an accompanying person were classified as ambulatory. For assessment of the ICARS score, in every center a rater and a backup rater were trained and certified by the study's clinical research physician, using study-specific training materials (including video training).
Data are shown as mean with SD, median with quartiles and/or range, or frequency and proportion, as appropriate. Correlations are presented as Spearman correlation coefficient ρ. Comparisons between groups of no, mild, intermediate, and severe FA-CM were done by Kruskal-Wallis test (χ2 test) and, if significant, by groupwise Mann-Whitney U test or χ2 test. No formal adjustment for multiple testing was implemented; to account for the multiple comparisons between FA-CM categories, only differences significant at P<0.001 were considered. Skewed variables were log-normalized before use in multivariable regression analysis, in which a stepwise model using forward selection followed by backward elimination (P<0.05 used as criterion for entry and grouping in the model) was used. Independent predictors of FA-CM (yes versus no) were sought by logistic regression with a similar approach.
Two hundred five patients with genetically confirmed FA were included in this cross-sectional study. The mean (SD) age was 30.6 (13.3) (range, 8–70) years, and 91 patients (44.4%) were female. The median (quartiles) age at diagnosis was 13.5 (9.1, 19.7) years, and mean time since diagnosis was 14.6 (9.6) years. The mean ICARS score was 48.4 (21.2) points, and 51.8% of patients were ambulatory (see Table 1 for patient characteristics).
Friedreich Ataxia Cardiomyopathy
The primary objective of this study was to identify a readily accessible clinical echocardiography parameter that would allow the sensitive detection and reliable grouping of FA-CM cardiac phenotypes. Because LV hypertrophy is the main cardiac feature in FA, we first correlated a number of standard (ie, nonderived) echocardiography parameters with LV mass assessed by cMRI, which is generally accepted as the reference standard to determine myocardial mass (Table 2). The best correlation with cMRI-derived LV mass was seen for echo-IVSTd (ρ=0.49; P<0.0001), qualifying this easily accessible parameter as a reliable echocardiographic measure to detect LV hypertrophy in FA.
With the use of reference values for echo-IVSTd, with correction for age and body surface area and calculated according to Henry's nomogram16 (Table I in the online-only Data Supplement), 4 groups of FA-CM could be defined, as follows: (1) no FA-CM: patients with normal values for echo-IVSTd (ie, not larger than the predicted value calculated by Henry's nomogram) and MRI EF ≥50%; (2) mild FA-CM: patients with echo-IVSTd exceeding the predicted normal echo-IVSTd by <18% (corresponding to 2×SD in Henry's nomogram) and MRI EF ≥50%; (3) intermediate FA-CM: patients with echo-IVSTd exceeding the predicted normal echo-IVSTd by ≥18% and MRI EF ≥50%; and (4) severe FA-CM: patients with MRI EF <50% independent of echo-IVSTd.
With the use of cMRI data and comparison of actual echo-IVSTd data from this study cohort of FA patients with predicted values according to Henry's nomogram, 65 patients (31.7%) were classified as having no FA-CM. The other patients were thus diagnosed as having cardiac involvement. Of these, 82 (40.0%) had mild FA-CM, 33 (16.1%) had intermediate FA-CM, and 25 (12.2%) had severe FA-CM. The patient characteristics of all FA-CM groups are shown in Table 1. For comparison, we also grouped patients on the basis of echocardiography data, which resulted in 32.2% of patients with no FA-CM, 41.0% of patients with mild FA-CM, 18.5% with intermediate FA-CM, and 8.3% with severe FA-CM. Because cMRI is generally considered more reliable than echocardiography, we used cMRI-based EF measurements for further analyses.
In general, all patients with cardiomyopathy had a concentric LV hypertrophy pattern by visual assessment (no patient had a septal or apical hypertrophic cardiomyopathy pattern). Patients in the group with no FA-CM were on average older and were included in the study with a longer time since diagnosis than patients with any degree of FA-CM. Interestingly, a higher proportion of patients presenting with progressing FA-CM was male (ie, 66% in the intermediate FA-CM and 72% in the severe FA-CM stage).
Echocardiographic Characterization of FA-CM Groups
The morphological and functional echocardiographic characteristics of all subjects studied across all FA-CM groups are shown in Table 3. Because FA-CM categories were defined according to echo-IVSTd, the intermediate FA-CM group showed higher values than the severe FA-CM group. Echocardiographically determined end-diastolic thickness of the posterior wall was thicker in the mild and intermediate FA-CM groups than in the group with no FA-CM but was thinner in severe FA-CM patients. Interestingly, the echo-IVSTd was larger than the echocardiographically determined end-diastolic thickness of the posterior wall in the more advanced groups of FA-CM, particularly in the intermediate and severe FA-CM groups (2 patients with severe FA-CM and 1 patient with intermediate FA-CM had an echocardiographically determined IVSTd/end-diastolic thickness of the posterior wall >1.5). When we focused on LV dimensions, both end-diastolic and end-systolic diameters were larger in severe FA-CM patients, and higher values for LV myocardial mass were evident with more advanced FA-CM. Conventional diastolic echocardiographic function parameters were not significantly different between the FA-CM groups. Although EF on echocardiography was normal and not significantly different between patients with no FA-CM, mild FA-CM, and intermediate FA-CM, a trend for a lower values of peak systolic global longitudinal strain rate, averaged across all cardiac segments (Table 3 and Figure 1A), was detectable. Likewise, significantly lower values for averaged peak systolic global strain in intermediate and severe FA-CM compared with the group with no FA-CM (Figure 1B) were observed. Analysis of peak systolic strain rate for each of the LV cardiac segments showed that the highest absolute values (ie, indicating good regional contractility) were found in the septum, and the lowest absolute values (ie, indicating poor regional contractility) were detected in the posterior wall (Figure 1C and 1D). Of note, the region with the consistently lowest absolute peak systolic strain rate in the LV (ie, functionally worst) was the basal posterior segment in patients with severe FA-CM.
ECG Characterization of FA-CM Groups
Resting ECG data were available from 191 patients and showed that the percentage of patients with an abnormal ECG increased with the severity of FA-CM groups, as follows: 23.4% for no FA-CM and 26.7%/44.8%/65.2% for mild/intermediate/severe FA-CM, respectively. Patients with severe FA-CM exhibited a higher resting heart rate (Table 4). The QRS duration was normal and not different between FA-CM groups, indicating that even patients with severe FA-CM did not tend to develop bundle branch block. Uncorrected and corrected QT intervals were normal and not different between FA-CM groups, implying that these hearts were not very susceptible to malignant ventricular arrhythmias. In parallel with echocardiography findings, intermediate FA-CM patients showed both a higher S wave in V2 and a higher R wave in V5, indicating ECG signs of LV hypertrophy.
An exercise stress test was performed by 191 patients. No stress test had to be stopped early for safety reasons. Peak workload (median, quartiles) was not different between FA-CM subgroups, as follows: 51 W (28, 90) for no FA-CM; 40 W (22, 91) for mild FA-CM; 38 W (26, 64) for intermediate FA-CM; and 40 W (27, 91) for severe FA-CM. Interestingly, exercise performance was more closely related to neurological status than to FA-CM group. Specifically, peak workload negatively correlated with ICARS score for patients with the use of leg (ρ=−0.39; P=0.0002) or arm ergometry (ρ=−0.52; P<0.0001) (Figure 2). When a stepwise regression analysis was applied (with the use of age, age at diagnosis, time since diagnosis, FA-CM group, GAA1 repeat length, arm/leg ergometry, and ICARS score as possible explanatory factors), the best predictor of exercise performance was ICARS score, accounting for 56% of the variability (P<0.0001). Other independent predictors were GAA1 repeat length (additional 5% of variability; P=0.0070) and whether the test was performed by arm or leg ergometry (additional 2% of variability; P=0.0462).
Relationship Between FA-CM Groups and Age, Neurological Function, and GAA Repeat Length
When FA-CM groups were examined by age categories, there was a clear tendency for a higher proportion of patients at young ages to present with intermediate and severe FA-CM (Figure 3A). In fact, 49% of patients aged <20 years presented with intermediate or severe FA-CM. In contrast, more than two thirds of patients aged >40 years had no FA-CM. Patients with an earlier diagnosis (and thus onset) of disease generally also showed more severe cardiac involvement. For example, 37.8% of patients aged <14 years at diagnosis presented with intermediate or severe FA-CM compared with only 17.8% of patients aged >14 years. Interestingly, a total of 46.7% of patients aged >14 years at diagnosis were categorized as having no FA-CM.
In contrast, no relationship could be found between FA-CM groups and ICARS score (Table 1). The distribution of FA-CM groups was comparable across all ICARS groups (Figure 3B), indicating that the ICARS is not a predictor for the severity of cardiac involvement in FA. For 85 patients, data on GAA repeat lengths were available, and the disease-predicting shorter GAA allele was evaluated for a possible association with FA-CM groups. There was also no clear correlation between GAA repeat length and increasing FA-CM severity (Figure 4).
Multivariate statistical analyses were performed with the use of age, age at diagnosis, time since diagnosis, ICARS score, FA-CM category, and GAA1 repeat length as potential explanatory variables. For ICARS score (in linear regression), time since FA diagnosis was the best predictor (describing 46% of variability; P<0.0001), followed by age at diagnosis (10%; P<0.0001). For FA-CM status (logistic regression), age was the best explanatory factor (χ2=24.9; P<0.001), and ICARS score ranked second (χ2=8.6; P<0.0034).
Because cardiac involvement in FA is commonly described on the basis of data limited to either LV hypertrophy, LV dysfunction, or electric abnormalities,10,17–20 there is still a lack of comprehensive cardiac data sets that allow development of strategies for the reliable detection and assessment of FA-CM. The present cross-sectional study in >200 patients is thus far the largest in FA to present data on cardiac morphology and function determined by standard echocardiography, CDMI, cMRI, ECG, and exercise testing. In agreement with emerging diagnostic criteria and standard of care recommendations for FA,4 the assessment of this heterogeneous patient population with pronounced diversity in regard to age, age at diagnosis, duration of disease, genetic status, and neurological symptoms for the first time allowed a more comprehensive description of FA-CM. Reliable and precise diagnostic and follow-up approaches for FA-CM are needed because cardiac involvement is a major contributor to the increased risk of morbidity and mortality observed in this disease.7 Specifically, criteria were lacking that allow assessment of the severity of cardiac involvement in individual patients according to objective and clinically accessible morphological and functional cardiac parameters. Here we provide comprehensive data that support the grouping of FA-CM based on readily available standard echocardiographic measurements, and we describe in detail the clinical presentation of FA-CM phenotypes.
Friedreich Ataxia Cardiomyopathy
As in earlier studies,21 we compared data from our study sample with reference values using Henry's nomogram16 to assess FA-CM. Among various standard echocardiographic parameters of cardiac morphology, echo-IVSTd was shown to be a reliable predictor for increased LV mass as measured by the reference cMRI standard. In addition to the progression toward LV hypertrophy, a decrease in global LV function also indicates cardiomyopathy. Thus, an additionally reduced MRI EF (<50%) was chosen as a principal criterion to define severe FA-CM.
With respect to LV morphology, we observed that in milder groups of FA-CM, the thickness of the LV walls was very homogeneous and the LV was not dilated, whereas in patients with severe FA-CM, the septum was significantly thicker than the posterior wall. However, the visual impression is that of a homogeneous concentric hypertrophy pattern even in severe FA-CM because these small differences cannot be detected by the human eye. In addition, measurement values for both the septum and posterior wall thickness were slightly lower in severe FA-CM patients. Because the highest myocardial mass was seen in patients with severe FA-CM, the decrease in wall thickness in combination with a dilated LV indicates eccentric LV remodeling in these patients. From autopsy and biopsy studies, it is known that FA patients can develop myocardial fibrosis,18,19,22,23 which might be responsible for the observed shrinking of the myocardium in severe FA-CM. Because the posterior wall was thinner than the septum in these patients, it can be speculated that the progression toward fibrosis is more advanced in the posterior cardiac segments. This assumption is supported by the finding that the posterior wall also showed the lowest values for regional longitudinal function. In a recent article by Mottram et al,6 reduced longitudinal function was also discussed as a typical functional finding in FA-CM, which is in accordance with our study. The phenomenon of developing fibrosis, particularly in the posterior wall, is also well known in other genetic cardiomyopathies such as Duchenne muscular dystrophy and Fabry disease.24,25
Our study did not provide evidence for an association between neurological and cardiac involvement in patients with FA, and further work is needed to better understand the role of other factors influencing the development of FA-CM. Importantly, exercise capacity of FA patients was not affected by the severity of FA-CM but rather by neurological involvement. Thus, in patients with advanced neurological disease, a subclinical cardiomyopathy may be present despite the absence of exertional symptoms usually observed in patients with myocardial disease.
The present study provides guidance in regard to the manner in which cardiac involvement in FA can be assessed during routine follow-up as part of the recommended standard of care for FA.4 From other cardiomyopathies, it is known that the severity of the disease affects treatment strategies.26 In this context, it is clinically important that an observed decrease of posterior wall thickness in FA during follow-up could represent progression toward an advanced stage of cardiomyopathy rather than a positive treatment effect, especially in combination with declining EF.
The present data also provide evidence that patients with FA-CM are developing heart involvement at an early age (ie, age <40 years). Young age appears to better predict the degree of FA-CM than GAA repeat length, which is in agreement with previous observations.27 This finding is also consistent with a recent study demonstrating that most FA patients die from cardiac failure at ages <40 years.7 Our data imply that young patients in particular should be evaluated carefully for cardiac involvement and routinely checked thereafter for disease progression; neurological evaluation alone must be considered insufficient to assess cardiac risks.
Cross-sectional studies are not suited to describe the progression of a disease over time. However, 2 longitudinal studies in FA18,28 suggested a natural course of disease progression from LV hypertrophy to dilated cardiomyopathy that was similar to our findings. Owing to the comparatively large number of FA patients across a wide range of clinical and diagnostic phenotypes, we consider the present study representative in that it allows adequate description of the various types of disease severity and organ involvement in FA-CM. Spiroergometry, which is best suited to describe muscular-cardiac-pulmonary coupling, was not performed in the present study. Hence, a more detailed analysis of the reduced exercise capacity in FA-CM is left for further studies. In addition, a validation of Henry's nomogram was not performed, which is difficult because of the rare nature of the disease. Because myocardial biopsies were not sampled for ethical reasons, we were unable to determine the histological pathology underlying the observed morphological and functional changes.
The preferred approach for grouping FA-CM might be to perform both cMRI and echocardiography at the time when the disease is diagnosed; this would allow setting an accurate baseline for the patient. Later, during follow-up, echocardiographic examinations alone may suffice. However, this issue should be left for future discussion within the format of a guideline/consensus article.
This cross-sectional study in a large group of FA patients provides a detailed characterization of the morphological, functional, and ECG abnormalities associated with FA-CM and will add to the standard of care for this rare disease.
Sources of Funding
This work was supported by grants from the Bundesministerium für Bildung und Forschung (BMBF project 01EO1004) and GeNeMOVE (BMBF project 01GM0501).
Drs Rummey and Meier are regular employees of Santhera Pharmaceuticals, the sponsor of the MICONOS study.
The authors thank Drs Coppard and Metz for discussion and support in the interpretation of the data and Mika Leinonen (4Pharma, Sweden) for statistical support. Investigators of the MICONOS Study Group are as follows: Thomas Klopstock (Munich), Thomas Klockgether (Bonn), Bernd Sereda (Göttingen), Ludger Schöls (Tübingen), Angela Schulz (Hamburg), Rudolf Korinthenberg (Freiburg), Hans-Peter Vogel (Berlin), Georg Ertl and Meinrad Beer (Würzburg), Nick Wood (London), Massimo Pandolfo (Brussels), Alexis Brice (Paris), Sylvia Bösch (Innsbruck), Ewout Brunt (Groningen), and Patrick Chinnery (Newcastle).
The online-only Data Supplement is available with this article at http://circ.ahajournals.org/lookup/suppl/doi:10.1161/CIRCULATIONAHA.111.059477/-/DC1.
- Received August 9, 2011.
- Accepted January 24, 2012.
- © 2012 American Heart Association, Inc.
- Harding AE
- Hewer RL
- Trouillas P,
- Takayanagi T,
- Hallett M,
- Currier RD,
- Subramony SH,
- Wessel K,
- Bryer A,
- Diener HC,
- Massaquoi S,
- Gomez CM,
- Coutinho P,
- Ben Hamida M,
- Campanella G,
- Filla A,
- Schut L,
- Timann D,
- Honnorat J,
- Nighoghossian N,
- Manyam B
- Henry WL,
- Gardin JM,
- Ware JH
- Weidemann F,
- Eyskens B,
- Mertens L,
- Di Salvo G,
- Strotmann J,
- Buyse G,
- Claus P,
- D'Hooge J,
- Bijnens B,
- Gewillig M,
- Sutherland GR
- Raman SV,
- Phatak K,
- Hoyle JC,
- Pennell ML,
- McCarthy B,
- Tran T,
- Prior TW,
- Olesik JW,
- Lutton A,
- Rankin C,
- Kissel JT,
- Al-Dahhak R
- Weidemann F,
- Niemann M,
- Breunig F,
- Herrmann S,
- Beer M,
- Stork S,
- Voelker W,
- Ertl G,
- Wanner C,
- Strotmann J
Friedreich ataxia (FA) is an autosomal recessive neurodegenerative disease caused by a defect in the gene encoding for the mitochondrial protein frataxin. Myocardial involvement in FA is well documented, with concentric left ventricular hypertrophy as the dominating cardiac finding. Average life expectancy in FA patients with cardiac involvement is considerably reduced to 29 to 38 years. Echocardiography is the routine imaging technique of choice for the evaluation and follow-up of cardiac involvement in patients diagnosed with FA. The present cross-sectional study in >200 patients describes in detail the clinical presentation with the use of standard echocardiography, color Doppler myocardial imaging, cardiac magnetic resonance imaging, ECG, and exercise testing. According to findings from this study, clinical grouping of the cardiomyopathy associated with FA should be based on echocardiographic wall thickness of the septum and ejection fraction. Furthermore, exercise capacity of FA patients is not affected by the severity of the cardiomyopathy but rather by neurological involvement. The provided definition of cardiomyopathy, classification of disease severity, and description of the interaction with neurology should add to the standard of care for this rare disease.