Frontotemporal dementia is clearly a very important condition, not only because alongside Alzheimer's disease, vascular dementia, and Lewy body dementia it is one of the four most common causes of dementia, but because its broad symptom profile and younger age of onset makes it a particularly challenging condition to diagnose and manage. In recent years, great progress has been made in understanding genetic and pathological underpinnings of frontotemporal dementias, and in characterizing specific pathological causes. In particular, three major subtypes can be delineated: those associated pathologically with tau; transactive responsive DNA binding protein (TDP-43); or fused in sarcoma (FUS). Of these, tau and TDP-43 are by far the two most common subtypes of frontotemporal dementia. In addition, there have been important genetic advances and several autosomal dominant mutations have been described, for example associated with MAPT, progranulin, and C9ORF72. However, despite these important advances regarding pathophysiological heterogeneity, relatively little is known about the frequency, nature, and correlates of psychotic symptoms in the disorder. It is therefore very timely that in this month's Paper of the Month, Waldo and colleagues (Waldo et al., Reference Waldo, Gustafson, Passant and Englund2015) describe a very detailed description of a cohort of 97 consecutive subjects seen in Lund who received a neuropathological diagnosis of frontotemporal lobar degeneration.
Although this was a retrospective study, and so relied on the characterization of symptoms and features from case notes, several extremely interesting findings emerge. One of the most striking findings is that only 14% of patients with frontotemporal dementia received that as an initial clinical diagnosis. So early misdiagnosis is common, indeed the norm. Interestingly, psychiatric disorders, such as depression and schizophrenia, accounted for the majority (42%) of initial misdiagnoses, with other dementias being diagnosed in 34% and around 9% having a label of other cognitive or neurological disorders. Misdiagnosis was particularly common amongst younger patients. Another interesting finding was the broad spread and heterogeneity of the condition. Thus, although the mean age at presentation was 58 years, the range was from subjects as young as 30 years to as old as 84 years, and duration of illness ranged from 1 to 28 years. Given this heterogeneity, it's not surprising that this is a very challenging condition to diagnose at presentation.
However, the main aim of the study was an investigation of psychotic symptoms, which have been well described in previous studies as occurring in frontotemporal dementia, but with little consistent data regarding their frequency or clinical or pathological correlates. In this paper, psychotic features were found to be present in around a third of subjects at some point during the illness. Most frequently these consisted of hallucinations, especially visual hallucinations, and delusions. The authors found an association between the presence of psychotic features and right-sided predominance of pathology, an intriguing finding and something that has also been noted in other organic disorders. Psychotic symptoms occurred both in tau-positive and tau-negative cases, and within the tau-negative groups were seen in those with both TDP-43 and FUS pathology. Not surprisingly, family history of dementia was common in this cohort, in around 42% of cases where this had been documented, but there was no association between the presence of psychotic symptoms and a positive family history.
This study, from the widely regarded Lund group, who have provided so much important data that has assisted in the current characterization of frontotemporal dementia, emphasizes on several things. The first is the challenge in making an accurate diagnosis of frontotemporal dementia at presentation. The second is the high prevalence of psychotic symptoms associated with the disorder. Third is the fact that psychotic features can occur in any subtype of frontotemporal dementia and do not particularly point towards a tau-positive or tau-negative pathology. Finally, there is the intriguing observation that the presence of such symptoms may be more common in those with greater pathological change on the right side. The study has major strengths, being very large well characterized series of pathologically proven cases of frontotemporal dementia. The importance of such pathological series, in addition to clinical series, is indicated by the fact that a third of the subjects with frontotemporal dementia and psychotic symptoms in this study did not receive a diagnosis of frontotemporal dementia during life, and so would have been missed from clinical samples. Like any study, there are limitations which the authors discuss, perhaps the most important being that any study relying on documentation retrospectively from case notes suffers from the inevitable variability that non-standardized ascertainment of symptoms bring. There is also the potential issue of ascertainment bias, in that those reaching autopsy may not necessarily be representative of the wider population with frontotemporal dementia.
Whilst the previous literature has varied in defining the frequency of symptoms in frontotemporal dementia, the current study certainly suggests that these are a frequent occurrence, and certainly therefore should not put clinicians off making the diagnosis. The frequency of symptoms also indicates a need for better understanding of their pathophysiology, which very likely will differ from pathophysiology in other disorders such as Lewy body dementia (where psychotic symptoms are particularly associated with cholinergic loss, something not seen in frontotemporal dementia) and Alzheimer's disease, where the genesis of psychotic symptoms may be multifactorial. A better understanding of the pathophysiological underpinnings of such symptoms in different dementias is key to generating better treatments, which are badly needed, especially for frontotemporal dementia which is one of the dementias that does not benefit from the current generation of anti-dementia drugs. This study also shows the scientific importance of the continuing study of clinical and pathological series, and in future larger multisite pathological collections with high quality, prospectively collected, clinical information will undoubtedly further inform our knowledge and characterization of the different subtypes of dementia.
Conflict of interest
None.
You have
Access
Frontotemporal dementia is clearly a very important condition, not only because alongside Alzheimer's disease, vascular dementia, and Lewy body dementia it is one of the four most common causes of dementia, but because its broad symptom profile and younger age of onset makes it a particularly challenging condition to diagnose and manage. In recent years, great progress has been made in understanding genetic and pathological underpinnings of frontotemporal dementias, and in characterizing specific pathological causes. In particular, three major subtypes can be delineated: those associated pathologically with tau; transactive responsive DNA binding protein (TDP-43); or fused in sarcoma (FUS). Of these, tau and TDP-43 are by far the two most common subtypes of frontotemporal dementia. In addition, there have been important genetic advances and several autosomal dominant mutations have been described, for example associated with MAPT, progranulin, and C9ORF72. However, despite these important advances regarding pathophysiological heterogeneity, relatively little is known about the frequency, nature, and correlates of psychotic symptoms in the disorder. It is therefore very timely that in this month's Paper of the Month, Waldo and colleagues (Waldo et al., Reference Waldo, Gustafson, Passant and Englund2015) describe a very detailed description of a cohort of 97 consecutive subjects seen in Lund who received a neuropathological diagnosis of frontotemporal lobar degeneration.
Although this was a retrospective study, and so relied on the characterization of symptoms and features from case notes, several extremely interesting findings emerge. One of the most striking findings is that only 14% of patients with frontotemporal dementia received that as an initial clinical diagnosis. So early misdiagnosis is common, indeed the norm. Interestingly, psychiatric disorders, such as depression and schizophrenia, accounted for the majority (42%) of initial misdiagnoses, with other dementias being diagnosed in 34% and around 9% having a label of other cognitive or neurological disorders. Misdiagnosis was particularly common amongst younger patients. Another interesting finding was the broad spread and heterogeneity of the condition. Thus, although the mean age at presentation was 58 years, the range was from subjects as young as 30 years to as old as 84 years, and duration of illness ranged from 1 to 28 years. Given this heterogeneity, it's not surprising that this is a very challenging condition to diagnose at presentation.
However, the main aim of the study was an investigation of psychotic symptoms, which have been well described in previous studies as occurring in frontotemporal dementia, but with little consistent data regarding their frequency or clinical or pathological correlates. In this paper, psychotic features were found to be present in around a third of subjects at some point during the illness. Most frequently these consisted of hallucinations, especially visual hallucinations, and delusions. The authors found an association between the presence of psychotic features and right-sided predominance of pathology, an intriguing finding and something that has also been noted in other organic disorders. Psychotic symptoms occurred both in tau-positive and tau-negative cases, and within the tau-negative groups were seen in those with both TDP-43 and FUS pathology. Not surprisingly, family history of dementia was common in this cohort, in around 42% of cases where this had been documented, but there was no association between the presence of psychotic symptoms and a positive family history.
This study, from the widely regarded Lund group, who have provided so much important data that has assisted in the current characterization of frontotemporal dementia, emphasizes on several things. The first is the challenge in making an accurate diagnosis of frontotemporal dementia at presentation. The second is the high prevalence of psychotic symptoms associated with the disorder. Third is the fact that psychotic features can occur in any subtype of frontotemporal dementia and do not particularly point towards a tau-positive or tau-negative pathology. Finally, there is the intriguing observation that the presence of such symptoms may be more common in those with greater pathological change on the right side. The study has major strengths, being very large well characterized series of pathologically proven cases of frontotemporal dementia. The importance of such pathological series, in addition to clinical series, is indicated by the fact that a third of the subjects with frontotemporal dementia and psychotic symptoms in this study did not receive a diagnosis of frontotemporal dementia during life, and so would have been missed from clinical samples. Like any study, there are limitations which the authors discuss, perhaps the most important being that any study relying on documentation retrospectively from case notes suffers from the inevitable variability that non-standardized ascertainment of symptoms bring. There is also the potential issue of ascertainment bias, in that those reaching autopsy may not necessarily be representative of the wider population with frontotemporal dementia.
Whilst the previous literature has varied in defining the frequency of symptoms in frontotemporal dementia, the current study certainly suggests that these are a frequent occurrence, and certainly therefore should not put clinicians off making the diagnosis. The frequency of symptoms also indicates a need for better understanding of their pathophysiology, which very likely will differ from pathophysiology in other disorders such as Lewy body dementia (where psychotic symptoms are particularly associated with cholinergic loss, something not seen in frontotemporal dementia) and Alzheimer's disease, where the genesis of psychotic symptoms may be multifactorial. A better understanding of the pathophysiological underpinnings of such symptoms in different dementias is key to generating better treatments, which are badly needed, especially for frontotemporal dementia which is one of the dementias that does not benefit from the current generation of anti-dementia drugs. This study also shows the scientific importance of the continuing study of clinical and pathological series, and in future larger multisite pathological collections with high quality, prospectively collected, clinical information will undoubtedly further inform our knowledge and characterization of the different subtypes of dementia.
Conflict of interest
None.