The concept of limb-girdle muscular dystrophy (LGMD) is changing rapidly due to the advances in molecular genetics. Recently, seven different gene loci have been described, demonstrating that limb-girdle muscular dystrophy is a heterogeneous syndrome, which includes different diseases with a similar phenotype. In isolated populations which have little genetic exchange with neighbouring populations, an accumulation of cases may be found. We carried out an epidemiological study in Guipúzcoa, a small mountainous Basque province in northern Spain, and found the highest prevalence rate of LGMD described so far: 69 per million. Genetic studies demonstrated that 38 cases corresponded to the LGMD2A type, due to calpain-3 gene mutations. Only one patient with alpha-sarcoglycanopathy was found, and in 12 patients the genetic defect was not identified. Moreover, the particular calpain-3 mutation predominant in Basque chromosomes (exon 22, 2362AG-->TCATCT), has only been rarely found in the rest of the world. This observation strongly suggests a founder effect in the indigenous population of Guipúzcoa. The clinical characteristics of the patients with calpain-3 gene mutations were quite homogeneous and different from the other groups (sarcoglycanopathy and unknown gene defect), allowing for a precise clinical diagnostic. The disease onset was between the ages of 8 and 15 years, in most cases in the pelvic girdle, and the patients became wheelchair-bound between 11 and 28 years after onset. No pseudohypertrophy of calves or contractures were observed. No clear correlations were found between the nature and site of the mutation and the resulting phenotype.

We present here the clinical, molecular and biochemical findings from 238 limb-girdle muscular dystrophy type 2A (LGMD2A) patients, representing ∼50% (238 out of 484) of the suspected calpainopathy cases referred for the molecular study of the calpain 3 (CAPN3) gene. The mean age at onset of LGMD2A patients was ∼14 years, and the first symptoms occurred between 6 and 18 years of age in 71% of patients. The mean age at which the patients became wheelchair bound was 32.2 years, with 84% requiring the use of a wheelchair between the age of 21 and 40 years. There was no correlation between the age at onset and the time at which the patient became wheelchair bound, nor between the sex of the patient and the risk of becoming wheelchair bound. Of the cases where the CAPN3 gene was not affected, ∼20% were diagnosed as LGMD2I muscular dystrophy, while facioscapulohumeral muscular dystrophy (FSHD) was uncommon in this sample. We identified 105 different mutations in the CAPN3 gene of which 50 have not been described previously. These were distributed throughout the coding region of the gene, although some exons remained free of mutations. The most frequent mutation was 2362AG→TCATCT (exon 22), which was present in 30.7% of the chromosomes analysed (146 chromosomes). Other recurrent mutations described were N50S, 550ΔA, G222R, IVS6-1G→A, A483D, IVS17+1G→T, 2069–2070ΔAC, R748Q and R748X, each of which was found in >5 chromosomes. The type of mutation in the CAPN3 gene does not appear to be a risk factor for becoming dependent on a wheelchair at a determined age. However, in the cases with two null mutations, there were significantly fewer patients that were able to walk than in the group of patients with at least one missense mutation. Despite the fact that the results of phenotyping and western blot might be biased due to multiple referral centres, producing a diagnosis on the basis of the classical phenotype is neither sufficiently sensitive (86.7%) nor specific (69.3%), although western blot proved to be even less sensitive (52.5%) yet more specific (87.8%). In this case LGMD2I was a relevant cause of false-positive diagnoses. Considering both the clinical phenotype and the biochemical information together, the probability of correctly diagnosing a calpainopathy is very high (90.8%). However, if one of the analyses is lacking, the probability varies from 78.3 to 73.7% depending on the information available. When both tests are negative, the probability that the sample comes from a patient with LGMD2A was 12.2%.