Caregiver burden is greater in frontotemporal dementia (FTD) than in Alzheimer disease (AD). However, little is known of the impact of the 3 main clinical variants of FTD— behavioral-variant frontotemporal dementia (bvFTD), semantic dementia (SemDem), and progressive nonfluent aphasia (PNFA)—or the role of disease severity in caregiver burden. The Zarit Burden Inventory was used to measure caregiver burden of bvFTD (n=17), SemDem (n=20), PNFA (n=20), and AD (n=19) patients. Symptom duration, caregiver age, and relationship type were matched across groups. Moreover, a number of caregiver (mood, social network) and patient variables (functional disability, behavioral changes, relationship with caregiver, and dementia stage) were addressed to investigate their impact on caregiver burden. Caregivers of bvFTD patients reported the highest burden, whereas SemDem and PNFA caregivers reported burden similar to AD. A regression analysis revealed that caregiver burden in FTD, regardless of subtype, was explained by a model combining disease staging, relationship changes, and caregiver depression. Burden increased with disease severity in FTD. This study is the first to show that caregivers of SemDem, PNFA, and AD patients show similar burden, while confirming that bvFTD caregivers show higher burden than AD caregivers. More importantly, this study demonstrates that burden worsens with disease progression in FTD.

Frontotemporal dementia (FTD) is a neurodegenerative brain disorder that affects the frontal and temporal lobes predominantly. Impaired emotion recognition has been reported in two FTD subtypes: behavioral-variant FTD (bvFTD) and semantic dementia (SD), but has not been investigated in the third subtype: progressive nonfluent aphasia (PNFA). Methods: Recognition of six basic facial emotions (anger, disgust, fear, sadness, surprise, and happiness) was investigated in 41 FTD patients (bvFTD = 16; SD = 12; PNFA = 13) and 37 age- and education-matched controls, using two tests. In one task, intensity of emotional expression was increased to identify cognitive components contributing to emotion recognition performance. Results: All patient groups demonstrated impaired overall facial emotion recognition compared to controls. Performance, however, improved with increased emotion intensity in bvFTD and PNFA groups, the effect of intensity on emotion recognition being particularly pronounced for negative emotions. In contrast, increased intensity of facial emotion did not change performance in SD. Conclusions: Patients with SD demonstrate a primary emotion processing impairment, whereas PNFA and bvFTD patients' emotional disturbance is in part mediated by attentional deficits. These findings indicate that a subset of FTD patients may benefit from enhanced emotional intensity that will facilitate facial emotion recognition.

Background and purpose Behavioural-variant frontotemporal dementia (bvFTD) and Alzheimer's disease (AD) patients experience behavioural and emotion recognition alterations, yet understanding of how socioemotional processing is affected with disease progression is minimal. Additionally, evidence suggests that bvFTD patients with limited brain atrophy on neuroimaging at presentation (bvFTD-la) have a more benign course than those with marked atrophy (bvFTD-ma). Longitudinal investigation of these patients, however, is lacking. Methods We investigated general cognition, emotion recognition and sarcasm detection in 20 bvFTD (8 with limited brain atrophy) and 17 AD patients longitudinally and used mixed models analyses to determine the level and rates of decline across groups over time. Results At baseline, all patient groups performed worse than controls on general cognition and emotion recognition measures. The bvFTD-ma group showed significant impairment on the sarcasm detection task compared with controls. Longitudinally, an overall effect of time was present for general cognition (p<0.001); however, the rate of decline did not differ across groups. Trends for interactions between time and diagnosis were observed for both emotion recognition tasks (p=0.055; p=0.062), with the bvFTD-ma group declining more rapidly than AD or bvFTD-la groups. On the sarcasm detection task, the bvFTD-ma and AD patients declined, whereas bvFTD-la patients remained stable over time (p=0.002). Conclusions Tasks of sarcasm detection represent a clinically useful tool to differentiate between bvFTD and AD at baseline. Furthermore, tasks of socioemotional functioning can track progression within bvFTD and identify bvFTD patients more likely to show a faster rate of decline.