Sense and sensitivity of novel criteria for frontotemporal dementia

The story goes that when Marco Polo first saw a rhinoceros on Java, he called it a unicorn. As a meticulous observer, he hastened to tell us that these unicorns appeared rather strange (Eco, 2000). Diagnostic categories in medicine guide our attention to discriminating nosological features and prevent us from misidentifying rare disease entities as more familiar diseases they may resemble. Interest in frontotemporal dementia (FTD) was rekindled in the late 1980s when patterns of hypoperfusion and hypometabolism distinct from those seen in functional images of Alzheimer's disease were noted (Neary et al ., 1987, 1988). Clinicopathological series collected in Lund and Manchester and international conferences on FTD in Lund led to the Lund–Manchester criteria for frontotemporal lobar degeneration (FTLD: Brun et al ., 1994; Neary et al ., 1998) suggesting a common denominator for three different phenotypical presentations: FTD, in which behavioural and personality changes predominate; progressive non-fluent aphasia; and semantic dementia. Other authors have since proposed restricting the term FTLD to neuropathologically confirmed disease (Josephs et al ., 2011). The clinical neurological examination in FTD can remain essentially normal until well into the disease course. The history provided, however, often abounds with concrete examples of changes in personal and social conduct that together may be expressed as a qualitative difference in the patient's personality. The Lund–Manchester papers (Neary et al ., 1987, 1988, 1998; Brun et al ., 1994) provided clinicians with a vocabulary that allowed these complex patterns to be encapsulated in discrete components: five core features, and an 11-item list of supportive features and nine exclusionary clinical features. According to the literal interpretation, each of the five core criteria and none of the exclusionary features must be present. While this strict requirement may have increased the specificity of the criteria, it …