Background Published genotype/phenotype data on McArdle disease are limited in sample size. A single national (Spanish) registry of patients with McArdle disease was created with the purpose of analysing their genotypic and phenotypic characteristics.
Methods A cross sectional study was conducted, collecting demographic, family history, clinical, genotype and functional capacity data from all patients diagnosed with McArdle disease in the Spanish National Health System up to December 2010.
Results 239 cases were recorded (all of Caucasian descent, 102 women; mean±SD age 44±18 years (range 9, 93)); prevalence of ∼1/167 000 people. Two mutant PYGM alleles were identified in 99.6% of cases. Although there was heterogeneity in the severity of symptoms, there were four common diagnostic features: (1) 99.5% of patients reported a history of acute crises of exercise intolerance (accompanied by recurrent myoglobinuria in 50% of cases); (2) in 58% of patients, symptoms started in the first decade of life; (3) 86% of patients repeatedly experienced the ‘second wind’ phenomenon over life; and (4) 99% of patients had a high basal serum level of total creatine kinase (>200 U/l). Clinical presentation of the disease was similar in men and women and worsened with age. Patients who were physically active had higher levels of cardiorespiratory fitness (by 23%, p=0.003) and were more likely to improve their clinical course over a 4 year period compared with inactive patients (OR 225; 95% CI 20.3 to 2496.7).
Conclusions The main clinical features of McArdle disease are generally homogeneous and frequently appear during childhood; clinical condition deteriorates with ageing. Active patients have a better clinical outcome and functional capacity.
Statistics from Altmetric.com
Glycogenosis type V (glycogen storage disease type V (GSD V), McArdle disease or myophosphorylase deficiency; OMIM database number 232600) is a disorder of carbohydrate metabolism, described by Brian McArdle in 1951,1 inherited in an autosomal recessive manner and produced by mutations in the gene (PYGM) encoding the muscle isoform of glycogen phosphorylase (myophosphorylase).2 This glycogenolytic enzyme catalyses and regulates the breakdown of glycogen to glucose-1-phosphate in skeletal muscle fibres; therefore, as in other muscle glycogenoses, patients are unable to obtain energy from their muscle glycogen stores.
Patients with McArdle disease typically experience exercise intolerance in the form of reversible acute crises of early fatigue and contractures, sometimes with rhabdomyolysis and myoglobinuria, which are triggered by static muscle contractions (eg, lifting weights) or dynamic exercise (eg, climbing stairs, brisk walking).3 Another feature of this disorder, which distinguishes it from other inborn errors of muscle metabolism, is the ‘second wind’ phenomenon.2 This denotes marked improvement in tolerance to aerobic dynamic exercise such as walking or cycling after ∼10 min of exercise, with disappearance of the tachycardia and undue fatigue that were triggered by the start of exertion.4 5 Yet another characteristic of the disease is a high serum level of total creatine kinase (CK) activity even under basal conditions—that is, in the absence of heavy exercise in the previous few hours or days. Due to its low estimated prevalence worldwide (eg, ∼1:100 000 people in the Dallas Forth Worth area, Texas, USA),6 there are scarce data available on the epidemiological characteristics of McArdle disease. Published genotype/phenotype data on McArdle disease are limited in sample size (ie, consistently <100 cases7–14 except for a recent report).14 Similar to other rare disorders, studies with larger cohorts (ideally all diagnosed patients within a given country) are of clinical interest and will help to improve our knowledge of the disease and of the main problems faced by patients, thereby allowing for targeted lifestyle interventions.
In Spain, it is feasible to gather data on all Spanish McArdle patients due to the diagnostic protocol followed by the National Health System. When McArdle disease is suspected (typically based on hyper-CK-aemia and exercise intolerance), patient's blood samples together with medical history data (and biopsy specimens, when available) are sent to three ‘reference’ public hospitals (Hospital 12 de Octubre, Madrid; Hospital Val d'Hebron, Barcelona; and Hospital Meixoeiro, Vigo) for biochemical and molecular genetic analysis. The neuromuscular genetics laboratories of these hospitals have been responsible for confirming the disease from the time molecular genetics tools were available for PYGM genotyping in our country—that is, starting in 1998 for Hospital 12 de Octubre, in 1999 for Hospital Meixoeiro and in 2001 for Hospital Val d'Hebron. Furthermore, since 2005, Spanish McArdle patients have been offered the chance to perform aerobic capacity tests in the Universidad Europea de Madrid (Madrid).
Therefore, the aim of this study was to create a single national (Spanish) registry of patients with McArdle disease and to analyse their genotypic and phenotypic characteristics.
The present cross sectional study collected data on genotype, demographic, family history, clinical, laboratory (biopsy and serum CK activity) and functional capacity data from Spanish individuals diagnosed with McArdle disease up to January 2011 in the three reference hospitals of the National Health System. We also followed-up the progression of the disease in a subcohort of 89 patients from 2006 to 2010. The study was approved by the ethics committees of the three hospitals and of the Universidad Europea de Madrid (Spain), and was conducted following the ethical guidelines of the Declaration of Helsinki 1961 (Edinburgh 2000 revision), the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement15 16 and Spanish legislation on clinical research.
There were 239 (137 men and 102 women, from 167 independent pedigrees) registered cases in Spain (all of Caucasian descent, 30% being family related (brothers/sisters)) who were diagnosed with McArdle disease through genetic analysis (which was accompanied by muscle biopsy diagnosis in164 patients). Patient blood samples were sent to the reference units based on a clinical suspicion of McArdle disease made by 106 neurologists (most with a subspeciality in neuromuscular diseases) working in the neurology department of 62 public (‘general’) hospitals and two private clinics located in 19 different cities in Spain. Patients living in rural areas (29%) were referred to the aforementioned city hospitals by general/family practitioners. We studied all registered cases irrespective of the fact that clinical suspicion of the disease was made before or after 1998—that is, before or after molecular genomic tool analyses for PYGM were available in our country.
Identification of mutant alleles in the PYGM gene was accomplished using a number of molecular genetics approaches on DNA isolated from patient peripheral blood or skeletal muscle—that is, fast screening of the most frequent PYGM mutations by means of PCR and restriction fragment length polymorphism methods,12 followed by sequencing of the entire coding region and intron/exon boundaries of the PYGM gene.17 Additionally, in some patients, muscle mRNA/cDNA based methods were necessary to demonstrate the molecular pathogenicity of a presumed mutant allele, particularly in those cases when alteration of the splicing mechanism was suspected.17
Patient lifetime reports of the existence (or absence) of exercise intolerance, the second wind phenomenon and recurrent exercise myoglobinuria (the latter been described by patients as ‘dark urines’—eg, ‘urine looking like cola or red wine’2) were obtained from patient clinical records. Anthropometric data as well as clinicians' reports of existence (or not) of permanent muscle weakness18 19 were also retrieved from clinical histories.
Identification of these clinical features allowed us to allocate patients to one of the following clinical severity classes according to the most commonly used phenotype severity scale9: 0=asymptomatic or virtually asymptomatic (mild exercise intolerance but no functional limitation in any daily life activity); 1=exercise intolerance, contractures, myalgia and limitation of acute strenuous exercise, and occasionally in daily life activities; no record of myoglobinuria, no muscle wasting or weakness; 2= same as 1 and also recurrent exertional myoglobinuria, moderate restriction in exercise and limitation in daily life activities; 3= same as 2 and also fixed muscle weakness, with or without wasting, and severe limitations on exercise and most daily life activities.
We interviewed 89 patients about the progression of their disease within the previous 4 year period to ascertain: (1) improvement (change to a lower severity class, for example from class 3 to one of the classes below), (2) worsening (change to a higher severity class) or (3) constant (no change in severity class). Sixty-three of these patients were tested for peak oxygen uptake (VO2peak) determination (see below). Finally, available data on renal failure and other concomitant diseases were also obtained from clinical histories.
We used clinical histories to record serum CK activity under basal conditions—that is, in the absence of heavy exercise in the previous few hours or days.2 In those patients who underwent VO2peak evaluation, CK levels were measured before exercise tests in the same laboratory (Universidad Europea de Madrid). Positive histochemical diagnosis was given when there was negative staining for myophosphorylase activity or no myophosphorylase enzyme activity in patient skeletal muscle.2
Functional capacity and physical activity habits
The patient's VO2peak was determined during cycle ergometry testing until exhaustion, as previously described (workload increases of 10 W/min starting at 0 W).10 In children (<15 years), VO2peak was determined during treadmill testing, as described elsewhere.20 All VO2peak evaluations were performed in the same laboratory (Universidad Europea de Madrid) during the 2006–2010 period, using the same equipment (breath by breath VO2peak analyser)10 and under the supervision of the same researchers.
In autumn 2010, a researcher interviewed all patients who underwent the VO2peak tests on their physical activity habits and pre-exercise carbohydrate ingestion patterns (if any) within the 4 year period. Patients were classified, following the recommendations of the American Heart Association and American College of Sports Medicine, as physically active if they reported doing physical activity, such as walking/brisk walking, bicycling or swimming for at least 30 min/day, for 5 or more days per week.21
Descriptive data are expressed as frequencies (%) and mean±SD. We used the χ2 test and unpaired t test for between gender comparisons of demographic, family history, clinical, laboratory and functional capacity data. In order to determine whether the clinical condition of patients deteriorates with ageing, we compared the patient's age across the (1) different severity groups (scores 0, 1, 2 and 3) and (2) different disease progression groups (improvement, worsening, constant) using one way analysis of variance. Pairwise comparisons were conducted with Tukey's test.
We also calculated the OR and 95% CI to determine the association between the PYGM genotype and disease phenotypes and disease progression as well as the association between physical activity levels and disease phenotypes and disease progression.
All statistical analyses were performed using the PASW (V.18.0 for Windows, Chicago, Illinois, USA).
We identified 239 (137 men and 102 women) registered cases in Spain with complete data on age and genotype. Their age at diagnosis was 44±18 years (range 9–93 years). Therefore, the prevalence of the disease in the Spanish (Caucasian) population is ∼1/167 000 persons. Except for one male patient (who died at the age of 80 years of cardiovascular disease, when PYGM diagnosis had been previously made), the rest of the registered patients are currently alive.
The PYGM genotype spectrum in the total Spanish cohort is shown in table 1. We identified two mutant PYGM alleles in all but one case (99.6% of the total population). In one patient (a 12-year-old male), we only identified one mutant PYGM allele yet, biopsy diagnosis was characteristic of McArdle disease and he presented with all of the main clinical features of the disorder, including the ‘second wind’ phenomenon, both self-reported and objectively determined.5 We therefore included this patient in our study cohort. We identified five novel unpublished mutations (see supplementary file 1, available online only).
The main demographic and phenotype data of the Spanish McArdle disease registered patients are shown in table 2. Irrespective of sex and disease severity, we found three main clinical features of McArdle disease:
A total of 99.5% of patients reported a history of acute crises of exercise intolerance which were not necessarily accompanied by myoglobinuria. Half of the patients were indeed free of recurrent exertional myoglobinuria (or could only remember having experienced isolated cases of myoglobinuria).
In the majority of patients (86%), symptoms started within the first two decades of life; indeed, in 58% of patients, symptoms started in the first decade, typically in school sport activities.
The majority of patients (86%) repeatedly experienced the second wind phenomenon over life. It is also noteworthy that we were able to objectively detect the second wind phenomenon in all adult patients (n=20) who performed the diagnostic cycle ergometer test.5 All showed a decrease in early exertional tachycardia (from ∼140–150 to ∼120 beats/min) starting after around 7 min of a constant load (∼40 W for most adults), cycle exercise eliciting a heart rate value of 60–70% of the predicted maximum heart rate (ie, 220 beats/min minus age in years). We were able to detect the second wind phenomenon in three of four children (<15 years) whom we tested on a treadmill although the magnitude of the phenomenon was less marked than in adults—that is, lower decrease in exertional tachycardia after around 7 min.22
Muscle fixed weakness was present in 25% of patients and consistently affected proximal/trunk (paraspinal, neck flexor, periscapular, proximal upper limb, axial or shoulder girdle) muscles (in a symmetric manner in 99% of cases), with less involvement of the hip girdle and lower limbs. For the latter, gluteus maximus, hamstring and quadriceps were the most severely involved muscles. Distal and facial weakness was not observed. A total of 25% of patients belonged to the highest severity class (class 3)—that is, reporting functional limitations during activities of daily life such as household tasks, personal care, and lifting and carrying weights while shopping or carrying children. In contrast, 8% of patients coped adequately with all activities of daily life and, except for mild exercise intolerance, were virtually asymptomatic (class 0).
Regarding the possibility of renal damage associated with exertional myoglobinuria, 4% of patients (all men) suffered acute renal failure, with one case ending in chronic renal failure. One male patient had an episode of cardiac arrest attributed to hyperkalaemia induced by lifting very heavy weights (more than half of his body weight) with no previous warm-up or carbohydrate ingestion, from which he recovered successfully. We found no reports of adverse pregnancy outcomes. A total of 2% (n=5) of the patient cohort were older than 80 years and 0.4% (n=1) were older than 90 years.
Basal hyper-CK-aemia (>200 U/l) was present in 99% of patients, with the majority (79%) surpassing the 1000 U/l threshold. Biopsy diagnosis was performed in 69% of the total patient population. Despite the possibility of false negative histochemistry results due to expression of the brain isoform of glycogen phosphorylase in regenerating muscle fibres,11 all biopsy results in the Spanish cohort were corroborated by PYGM genotyping, except for the aforementioned child in whom only one mutant PYGM allele has been identified to date. On the other hand, the combination of muscle biopsy and molecular genetics (together with the lack of self-reported second wind) allowed a differential diagnosis with other disorders causing a ‘McArdle-like’ phenotype, notably the ‘myalgia and cramps syndrome’ as a form of dystrophinopathy, and limb girdle muscular dystrophy type 2I, which may present with recurrent episodes of myoglobinuria. Other diagnostic tools such as the classic forearm ischaemic or non-ischaemic test (showing no lactate increase in antecubital vein blood after handgrip exercise) or electrophysiological studies, including exercise or ischaemic response, were performed in 15.5% and 28% of patients, respectively.
The main clinical presentation of the disease was similar in men and women (see supplementary file 2, available online only). However, recurrent exertional myoglobinuria was more frequent in men (p=0.006). Likewise, the prevalence of patients with disease severity class 2 was higher in men (p=0.001).
The phenotypic manifestation of the disease in those 89 patients whom we interviewed on disease progression was similar to that of the whole Spanish cohort (n=239) (see supplementary file 3, available online only), discarding a major selection bias. There was individual variability with regards to the clinical course of the disease, with nearly 28% of patients reporting worsening of symptoms and 21% reporting improvement. Those patients who were physically active were more likely to report an improvement after a 4 year period in the clinical course of the disease compared with those who were not (OR 225; 95% CI 20.3 to 2496.7), after controlling for age. Indeed, 17 of 20 physically active patients reported an improvement whereas only one non-active patient reported an improvement in the disease. Further controlling for baseline clinical severity strengthened the association (OR 608.5; 95% CI 15.6 to 23731.4).
Functional capacity and physical activity
The phenotypic manifestation of the disease in those 63 patients whom we interviewed on physical activity habits and tested for VO2peak determination was similar to that of the whole Spanish cohort (n=239) (see supplementary file 3, available online only). Except for one patient, all active patients reported that they usually consume a carbohydrate (sports) drink (∼330 ml) before engaging in exercise and approximately the same amount per hour of exercise, which is equivalent to a total carbohydrate intake of ∼26 g pre-exercise and ∼26 g per hour of exercise. VO2peak was 23% higher in physically active than in inactive patients (21.0±5.6 vs 16.1±5.5 ml O2/kg/min, respectively, p=0.003). The proportion of patients in the lower severity class (class 0) was significantly higher among active patients (24% vs 0% in inactive patients, p=0.02). Levels of serum CK activity were not significantly higher in physically active patients compared with those who were not active (3023±3587 vs 2026±1996 U/l, respectively, p=0.194), nor was the incidence of exertional myoglobinuria (43.0 vs 49%, respectively, p=0.658). Physically active patients did not report major health problems associated with regular exercise.
Age had an effect on several phenotypic features of the disease. The mean age of those patients in the highest severity class (50±18 years) was greater than those in the lower severity classes 1 (42±18 years, p=0.034) and 2 (39±14 years, p=0.011) but not greater than in those few patients belonging to class 0 (46±27 years, p=0.856). The mean age of those showing worsening in clinical severity (50±17 years) was higher than those showing an improvement (35±19 years, p=0.018) but not significantly different from those with a stable clinical severity (45±17 years, p=0.122). Muscle weakness was more prevalent in patients aged 40 years and over (34%) than in younger patients (16%, p=0.005). On the other hand, 41% of patients aged 40 years and over had a cardiorespiratory fitness (VO2peak) below the limits necessary for independent living—that is, 13 ml O2/kg/min22 versus 15% of younger patients (p=0.017). Older patients had lower cardiorespiratory fitness (reduction of 0.20 ml O2/kg/min per 1 year increase, p<0.001).
The PYGM genotype was not significantly associated with any of the clinical, functional capacity or CK data shown in table 2, after controlling for sex and age (data not shown, all p>0.1).
Despite heterogeneity in disease severity and the fact that ageing has a negative impact in the phenotypic manifestation of the disorder, the clinical presentation and laboratory profile of McArdle disease is clear, regardless of gender. The three main clinical features (1) exercise intolerance, whether severe or not, (2) the second wind phenomenon and (3) basal hyper-CK-aemia, provide solid evidence for the presence of McArdle disease before genetic confirmation is provided. In contrast, permanent muscle weakness is not a common feature of this disorder, at least not in young patients. Our data provide some valuable information on current practices for diagnosing and treating McArdle disease.
Regarding recommendations on the best possible diagnostic approach, we believe that, in any patient reporting acute crises of exercise intolerance accompanied by basal hyper-CK-aemia, ideally an exercise confirmatory test should first be performed to objectively assess the ‘pathognomonic’ second wind phenomenon.5 A second necessary step is PYGM genotyping, searching for the most prevalent mutations—that is, mainly p.R50X, p.W798R and p.G205S which account, alone or in combination, for ∼61% of all PYGM genotypes (table 1). Sequencing of the complete gene coding regions should be performed when the search for the most prevalent mutations has rendered a negative result. Muscle biopsies are indicated only if sequencing of the complete coding regions fails to identify pathogenic mutations. If biopsy analysis confirms negative histochemical reaction for myophosphorylase, complete sequencing of PYGM should be performed to identify a rare genetic defect. Finally, we believe that other evaluations, such as the ‘classic’ exercise forearm test or EMG, are not necessary as they do not provide solid additional information compared with the aforementioned tools.
The genotype data were in overall agreement with previous reports on smaller Caucasian cohorts,7 8 11–14 24 25 with p.R50X/p.R50X and p.R50X/p.W798R accounting for nearly 50% of all PYGM genotypes and p.R50X being clearly the most prevalent pathogenic mutation. There is however between studies variability in the reported allelic frequency of the p.R50X allele—that is, the frequency of ∼55% that we found is higher than that originally reported in North American patients (49% in n=40)26 or in the largest Italian cohort (43% in n=68),7 but below that recently reported in British patients (77% in n=45).11 Moreover, we have identified five new pathogenic mutations that must be added to the catalogue of molecular defects in the PYGM gene. Of note is that PYGM genotypes did not influence phenotype heterogeneity among patients, which is in line with previous research.8 11 25 Vissing et al27 reported the case of two patients with a ‘mild’ form of McArdle disease, both of whom carried a splice mutation in one PYGM allele, leading to residual myophosphorylase activity and thus ameliorated the phenotype (VO2peak of 23 and 24 ml O2/kg/min) compared with ‘normal’ McArdle patients (ie, with absent myophosphorylase activity). In our series, however, the biopsy samples that we obtained in 164 patients consistently showed null myophosphorylase activity.
Regarding clinical history, exercise intolerance frequently starts during childhood and adolescence whereas diagnosis is rarely performed at a young age. Indeed in the past years McArdle disease was rarely diagnosed before adulthood, possibly because the relevant medical information and lack of appropriate diagnostic tools (in particular molecular genetics) for this rare disease were not widely available. Thus although the identification of the second wind phenomenon might be more difficult in children,22 paediatricians should consider the possible existence of McArdle disease in those young children showing exercise intolerance (typically in sport activities or in physical education lessons). In the present cohort, acute renal failure was very infrequent (4% of patients, with no female patient being affected), which is below the previously reported data in smaller cohorts—that is, 25%18 or 11%.11 On the other hand, despite previous reports showing decreased insulin action in the skeletal muscles of McArdle patients,28 the prevalence of type II diabetes in this series was similar to that reported in Spanish adults.29
Our study also provides relevant data that are of potential clinical interest based on the prospective analysis we performed in a subcohort of patients (n=63) with data on physical activity habits and VO2peak determination. Levels of physical activity were positively associated with the clinical course of the disease, and such association was not biased by severity class at baseline before engaging in an active lifestyle, or by patient age. It is important to note that 81% of patients who were physically active changed to a lower severity class after a 4 year period. These epidemiological data are in agreement with preliminary research (using non-controlled designs) showing a significant increase in VO2peak of McArdle patients (including very young children) after regular aerobic exercise10 30 31—for example, 44% increase after an 8 month supervised exercise programme, including five weekly sessions (duration ≤60 min) of walking or cycling at moderate intensity (∼60% of maximum heart rate) with no major side effect.10
Levels of physical activity were also positively associated with patient functional capacity (assessed by VO2peak determination). These data on VO2peak of McArdle patients are clinically relevant because this variable is an integrative indicator that reflects the maximum capacity of different systems involved in the chain of events from the delivery of atmospheric oxygen to mitochondria and it is a strong independent predictor of all cause mortality and cardiovascular events.32 Mean VO2peak was very low, especially in women, where it barely reached the limit (∼13 ml O2/kg/min, 3.7 metabolic equivalents (METs) with 1 MET equalling 3.5 ml O2/kg/min) necessary for independent living.23 Age also had a negative impact on VO2peak. However, the VO2peak in patients who were physically active was ∼1.5 METs higher than in those who were not active. Seven patients (six physically active) had a VO2peak ≥8 METs, which is the minimum threshold for optimal health33; their VO2peak was also above the previously reported levels (∼7 METs) in two patients with a ‘mild’ form of McArdle disease owing to residual myophosphorylase activity.27
Concerns are frequently raised by clinicians as to the potential risks of physical activity programmes for these patients. However, we should consider that basal levels of serum CK activity and incidence of myoglobinuria were not significantly higher in physically active patients than in those who were not. It must be kept in mind that muscle damage, as indicated by hyper-CK-aemia, is a necessary physiological stimulus for muscle to be repaired and adaptive hypertrophy to occur.34 Whether this phenomenon also applies to patients with McArdle disease remains however to be elucidated. It should also be noted that previous research in healthy people has shown that exertional hyper-CK-aemia well above the mean values we obtained in our active patients is not associated with renal dysfunction.34
The present report describes the available data from virtually all Caucasian Spanish patients diagnosed with McArdle disease to date. Nevertheless, a number of patients are likely undiagnosed, for several reasons, such as: the rarity of the disease, the fact that some patients might only show a mild phenotype over life with no interference in daily living activities (ie, disease severity class 0) and the relative novelty of clinical advances in the field. As in other cross sectional studies of rare diseases, the fact that relatively novel and clinically relevant features of the disease are not systematically controlled in the national clinical registries calls for better characterisation and monitoring of the disease from childhood and adolescence. Furthermore, as clinical data were collected retrospectively, our results are limited by the ability of the clinicians to adequately identify disease symptoms and signs from their patients' reports. For instance, actual prevalence of myoglobinuria or the second wind phenomenon might have been underestimated in clinical histories. In addition, we did not systematically record pregnancy outcomes in all women.
In summary, the main clinical features of McArdle disease are generally homogeneous and frequently start during childhood; the clinical course of the disease worsens with age unless an active lifestyle is adopted. Our data also support the idea that those patients with an active lifestyle have a better clinical outcome and functional capacity than those who have a more sedentary lifestyle. A randomised controlled trial of physical activity (eg, aerobic training) would help to determine whether regular physical activity is an effective and feasible therapy for patients with McArdle disease and possibly for other muscle glycogenoses. The results of our study can assist clinicians in (1) improving their knowledge of McArdle disease and (2) recommending targeted lifestyle interventions. This is of clinical relevance because a definitive genetic cure for this disorder is not expected in the foreseeable future.2
The authors thank Sara Jimenez and Laura Rufian for technical support.
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Files in this Data Supplement:
AL, JRR, MAM and ALA contributed equally to this paper.
Funding This study was supported in part by grants from the Spanish Ministry of Science and Innovation (FIS PI10/00036, PI 10/02628, PI09-00194, RD09/0076/00011, and RYC-2010-05957) and from the Isabel Gemio Foundation for Neuromuscular and Other Rare Diseases. JCR and IG-C are recipients of contracts from SMSI (CA 05-0039 and CA 08-0203, respectively).
Competing interests None.
Ethics approval Ethics approval was provided by the institutional review boards of Hospital 12 de Octubre (Madrid), Val Hebron, Hospital Moixeiro and Universidad Europea de Madrid.
Provenance and peer review Not commissioned; externally peer reviewed.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.