Diagnostic criteria for inclusion body myositis.

Inclusion body myositis (IBM) was first identified as a specific disorder about 40 years ago and is now recognized to be the most frequently presenting primary myopathy in middle age and beyond. Initial characterization was based on the observation of specific pathological features distinguishing it from polymyositis. It was soon appreciated that there were also distinguishing clinical features. The earliest diagnostic criteria were heavily biased towards pathological features, but over time revised criteria have given increasing importance to certain clinical features. Until the specific cause of IBM is determined, and the basic pathogenetic mechanisms are better understood, there can be no diagnostic gold-standard against which to compare the sensitivity and specificity of any proposed diagnostic criteria, but such criteria are essential to ensure that patients entering clinical, epidemiological, genetic, pathological or therapeutic studies represent a homogeneous population. It is likely that any currently accepted diagnostic criteria will, once a gold-standard is eventually established, be shown to have 'missed' patients with atypical features, but that has to be accepted to make certain that current studies are not contaminated by patients who do not have IBM. In other words, in everyday clinical practice there will be the occasional patient who an experienced myologist strongly suspects has IBM, but does not meet current criteria - the criteria lack sensitivity. But if the criteria are so broad as to include all such atypical cases, they would be likely to include patients who do not in fact have IBM - they would lack specificity. The sensitivity and specificity of existing criteria have been reviewed recently, in so far as it is possible to do so, and found to have high specificity but variable sensitivity.