An integrated approach to the diagnosis of muscle disorders: what is the role of muscle imaging?

Abstract

Since the discovery of mutations in the dystrophin gene in 1987, there has been a string of exciting genetic discoveries in the field of inherited neuromuscular disorders, with an always increasing number of genetically recognized forms of muscle disorder.1 This has allowed the identification of new clinical entities and a better understanding of the mechanisms underlying the individual disorders. It has also become obvious that for some neuromuscular disorders we can no longer accept such simplistic equations as ‘one gene equals one disease’ and conversely, ‘one disease equals one gene’. It is now well recognized that a gene can be associated with different disorders (e.g. lamin A/C; phenotypic divergence) and that one disease can be caused by several genes (phenotypic convergence). The best example of phenotypic divergence in muscle disorders comes from the forms of congenital muscular dystrophy with muscle, eye, and brain involvement (Muscle-Eye-Brain disease and Walker-Waburg syndrome) that have been found to be associated with mutations in different genes.2 The complexity of all these new clinically and genetically distinct entities has made the diagnostic process more difficult, especially for the muscle disorders sharing clinical and biopsy findings with other genetically distinct entities. This issue is particularly relevant in young children. In older patients there are often clinical signs, such as the presence and the type of cardiac or respiratory impairment or the pattern of weakness and contractures that may provide important clues for targeting the appropriate genetic investigations. In young children however, clinical signs at onset can be mild and non-specific and the differential diagnosis can be difficult. Muscle biopsy is often highly important in identifying structural or immunohystochemical changes that may help to arrive at a specific diagnosis. In a number of cases, however, these markers, even when present, will help to identify the family of disorders but not always the gene involved. A few examples may illustrate the point. Structural changes such as cores or nemaline rods can be found in several forms of nemaline or core myopathies, but this will not predict the gene involved.3 Similarly, the detection of alpha dystroglycan deficiency on muscle biopsy will help to identify the group of disorders but not which gene is responsible for an individual case.2 There is increasing evidence that muscle imaging may help as an additional tool in the differential diagnosis of muscle disorders with clinical overlap. Since the early studies on muscle imaging in the 1970s using muscle ultrasound or computed tomography (CT),4 it has been reported that in patients with inherited neuromuscular disorders not all muscles are equally affected and individual muscles can be selectively affected or spared. Following the identification of the genetic defect in several muscle disorders, a number of papers have suggested that genetically distinct entities have different patterns of muscle involvement and that these patterns appear to be consistent within individual forms of muscle disorders. These changes can be seen on ultrasound or CT but are better appreciated on muscle magnetic resonance imaging (MRI) that has several advantages over CT, such as the absence of ionizing radiations and the possibility to use multiplanar imaging. Although we are still in a learning phase, there are more and more studies reporting the value of muscle MRI in an increasing number of disorders and its sensitivity and specificity in identifying disease-specific patterns in large series of patients with significant clinical overlap. In a recent study assessing scans from 83 patients with muscle disorders characterized by rigidity of the spine secondary to mutations in four different genes, we found that the sensitivity to detect selective patterns in relation to the genetic diagnosis was 0.9.5 These findings do not mean that muscle MRI will replace clinical examination or muscle biopsy but that we need to address the differential diagnosis in this field with an integrated approach, taking advantage of the information provided by each of these tools. Going back to the example of congenital myopathies, muscle biopsy will help to identify the group of disorders, e.g. central core or nemaline myopathies. A detailed family history, the evaluation of the modality of inheritance, clinical signs and, finally, muscle MRI will help to target the most appropriate genetic investigation. Much can be gained by this approach, saving time, money, and anxiety, making a difference for the children and their families both for genetic counselling and for management.