Autonomic dysfunction and hypothalamic atrophy in frontotemporal dementia and primary psychiatric disorders

Abstract

INTRODUCTION

Evidence suggests that autonomic symptoms in frontotemporal dementia (FTD) may relate to hypothalamic pathology. We investigated (1) autonomic symptoms in FTD, primary psychiatric disorders (PPD), and controls; (2) hypothalamic volumes; and (3) their associations.

METHODS

Autonomic Symptoms Questionnaire (ASQ) data were collected from FTD (n = 31), PPD (n = 30), and controls (n = 30). MRI data from these and additional participants (FTD n = 259, PPD n = 44, controls n = 62) were analyzed using a deep learning approach.

RESULTS

FTD and PPD groups reported more cardiovascular symptoms than controls (p = 0.020; p = 0.049). FTD patients showed greater thermodysregulation (vs. PPD p = 0.026; vs. controls p = 0.012) and altered pain perception (FTD 13%, PPD 2%, controls 1%; p < 0.001). FTD showed smaller hypothalamic subregions, except the left inferior‐tubular area. In FTD, autonomic symptoms correlated with subregional hypothalamic volumes.

DISCUSSION

FTD patients exhibit increased autonomic symptoms across several domains compared to PPD and controls; moreover, FTD is associated with smaller regional hypothalamic volume.

Highlights

• Autonomic symptoms are more common in FTD than in PPD and healthy controls.

• In certain hypothalamic regions, atrophy is more pronounced in FTD than in PPD and controls.

• Autonomic symptoms may support clinical distinction between FTD and PPD; however, autonomic symptoms also occur in PPD.

• The current study combines clinical symptom ratings with hypothalamic volumetry.

• Findings highlight the value of non‐cognitive symptoms in FTD diagnosis.

1. INTRODUCTION

Frontotemporal dementia (FTD) is typically characterized by early impaired social cognition, behavioral symptoms, and language impairments. Increasing evidence shows that FTD symptomatology extends beyond these characteristic features, with autonomic symptoms in early and end disease stages including thermodysregulation, urinary tract dysfunction, gastrointestinal dysfunction, sleep disturbances, increased heart rate, reduced heart rate variability, and altered pain perception. ¹ , ² , ³ In clinical practice, the significant overlap of behavioral symptoms and frontotemporal hypometabolism of FTD and late‐onset primary psychiatric disorders (PPD) hampers diagnostic distinction, leading to misdiagnoses in 50% of cases and resulting in a diagnostic delay of ≈ 6.4 years. ⁴ The clinical differentiation is further challenged by the co‐occurrence of autonomic dysfunction in patients with PPD, possibly due to pathophysiology in hypothalamic subregions, ⁵ side effects of psychotropic drug use, and frequent cardiovascular comorbidity. ⁶ , ⁷

It is conceivable that autonomic symptomatology might originate from FTD pathology in specific brain regions that mediate autonomic function, including the anterior cingulate cortex, insula, amygdala, brain stem, and most importantly, the hypothalamus. Interestingly, regional hypothalamic atrophy in FTD has been reported. ⁸ Although the hypothalamus roughly accounts for 4 g of the total ≈ 1300 to 1400 g of the adult brain, myriad autonomic, endocrine, and behavioral responses are evoked by hypothalamic interplay. Specialized hypothalamic nuclei and related circuitry fulfill a prominent role in regulating the autonomic nervous system, thereby contributing to thermoregulation, hunger and satiety, sleeping behavior, cardiorespiratory responses, and lower urinary tract functioning. ⁹ , ¹⁰ , ¹¹ For example, the dorsomedial hypothalamic nucleus (DMH) regulates cardiac sympathetic output and influences the heart rate. The DMH and pre‐optic nuclei of the hypothalamus are involved in thermoregulation, and the lateral nucleus in the tuberal area contains hypocretine neurons important for sleep regulation, while pain and miction disturbances have been associated with the paraventricular nucleus (PVN). ¹² In addition, the hypothalamus–pituitary axis enables hypothalamic endocrine responses and last, hypothalamic functioning is associated with certain survival or social characteristics. Decreasing hypothalamic volume during the disease course has been reported in FTD, ⁸ , ¹³ and previous studies highlighted specific hypothalamic vulnerabilities in FTD, potentially leading to altered eating behavior, one of the core symptoms of behavioral variant FTD (bvFTD). ¹⁴

RESEARCH IN CONTEXT

1. Systematic review: We reviewed prior literature on autonomic dysfunction in FTD and PPD. While autonomic symptoms have been noted in FTD, no studies directly compare these to symptoms in PPD. Research on hypothalamic involvement in FTD is limited and often lacks clinical correlation. Our study addresses this gap by examining both symptom profiles and structural findings across groups.

2. Interpretation: We found that autonomic symptoms are more prevalent in FTD than in PPD or controls, and that subregional hypothalamic atrophy is more pronounced in FTD. These findings suggest that while autonomic changes occur in both disorders, their extent and potential neural correlates differ. Thus, offering additional clinical context for the differential diagnosis.

3. Future directions: Future research should clarify how specific autonomic symptoms evolve across FTD and PPD, and how these relate to everyday functioning. Longitudinal studies may help define symptom patterns that support earlier and more accurate clinical recognition.

Despite recent efforts to assess autonomic symptoms in FTD, previous studies did not include PPD as a control group, nor assessed the relationship with hypothalamic volume. Defining the type, frequency, and severity of autonomic symptoms in FTD compared to PPD and the relation to hypothalamic subregional volumes might aid in better diagnostic distinction in clinical practice, which is essential for appropriate treatment of both conditions and for accurate patient selection in future clinical trials of FTD. Apart from diagnostic value, it is relevant to further assess the prevalence of autonomic symptoms in relation to clinical awareness. Clinical expert observations indicate that autonomic dysfunction can lead to hazardous situations, such as exposure to hot water during showering, leading to burn injuries due to impaired thermoregulation and altered pain perception. ¹⁵ Also, autonomic dysfunction can cause high symptom burden and have a negative influence on quality of life. Moreover, it might contribute to understanding the pathophysiology of the disease, which is essential for future development of disease‐modifying therapies to reduce symptom burden and mitigate the likelihood of hazardous circumstances resulting from autonomic dysfunction.

Therefore, in this study, we aimed to (1) assess autonomic symptoms among FTD, PPD, and controls; (2) compare hypothalamic volumes among groups; and (3) assess the association between autonomic symptoms and hypothalamic (subregional) volumes.

2. METHODS

2.1. Subjects

Ninety‐one subjects (31 FTD patients: n = 24 probable bvFTD, n = 5 definite FTD, n = 1 FTD motor neuron disease [FTD MND], n = 1 semantic dementia [SD]; 30 PPD patients: n = 14 major depressive disorder [MDD], n = 6 autism spectrum disorder [ASD], n = 4 bipolar disorder [BD], n = 3 psychosis spectrum disorder [PSD], n = 1 attention deficit hyperactivity disorder [ADHD], n = 2 unspecified; and 30 controls) were recruited from the Amsterdam Dementia Cohort from 2000 until 2023 ¹⁶ and diagnosed using a multidisciplinary approach following current diagnostic criteria. ¹⁷ , ¹⁸ , ¹⁹ All participants underwent an extensive diagnostic assessment, including examination by a cognitive neurologist and psychiatrist; a neuropsychological examination assessing all cognitive domains; cerebrospinal fluid (CSF) assessment of amyloid beta (Aβ) and hyperphosphorylated tau to exclude Alzheimer's disease (AD); blood draw to exclude somatic causes; electroencephalogram; magnetic resonance imaging (MRI); and, in case of diagnostic uncertainty, an [¹⁸F] fluorodeoxyglucose positron emission tomography (PET) scan was performed. As a control group, participants with subjective cognitive decline (SCD) were included from the SCIENCe project. ²⁰ Controls presented at the memory clinic but performed normally on clinical and cognitive examinations (i.e., criteria for mild cognitive impairment, any type of dementia, or a psychiatric disorder not met). All controls were negative for CSF AD biomarkers. In all participants, the Frontal Assessment Battery (FAB) and Mini‐Mental State Examination (MMSE) were administered. Additionally, FTD and PPD subjects underwent the Ekman 60 Faces test (facial emotion recognition). All FTD patients had a maximum interval of 12 months between the MRI scan and the Autonomic Symptoms Questionnaire (ASQ) to ensure optimal and reliable correlation analyses. Exclusion criteria for all groups included the presence of co‐morbidity that can cause autonomic symptoms (hypo‐ or hyperthyroidism, phaeochromocytoma, cerebral vascular events, Wilson's disease, brain tumors, metabolic encephalopathy, current or history of substance abuse [alcohol, drugs], cardiac/gastrointestinal co‐morbidity, pre‐existing micturition disorders, pre‐existing sleep disorders) and advanced neurodegeneration in case of FTD: Clinical Dementia Rating scale (CDR) > 2.

2.2. Ethical approval

The study was approved by the medical ethical committee of Amsterdam Universitair Medisch Centrum (UMC), and all participants provided written and oral informed consent. The study was performed in accordance with the ethical standards as laid down in the 1964 Declaration of Helsinki and its later amendments or comparable ethical standards.

2.3. Clinically reported autonomic symptoms

Autonomic symptoms were evaluated during a structured interview by a physician using the ASQ. ¹ The ASQ consists of 44 questions that assess physical symptoms associated with autonomic functioning, encompassing the areas of cardiovascular function, gastrointestinal functioning, temperature regulation, sweating, urinary symptoms, and sleep. Participants were interviewed to rate the frequency of autonomic symptoms on a 5‐point Likert scale, ranging from 0 (never) to 4 (daily or continuously), and the severity on a 3‐point scale, ranging from 0 (not applicable) to 2 (marked) for each symptom experienced in the preceding 6 months. A composite score was calculated for each domain by multiplying the frequency score by the severity score for each question, and the summed composite scores represented the total ASQ score. Because recent studies have reported a changed pain perception in FTD, an additional domain assessing altered pain perception was added (present yes/no, increased/decreased). ³ In case of FTD, the informal caregiver was present during administration of the ASQ to contribute additional information, if necessary, when, due to impaired disease insight, patients were not able to provide reliable answers.

2.4. Image acquisition and processing

MRI of the brain was acquired on a 1 T, 1.5 T (Siemens Magnetom Avanto, Vision, Impact and Sonata, GE Healthcare Signa HDXT), or 3 T (Toshiba Titan and Philips Ingenuity PET/MR) magnetic resonance system. Acquisition protocols have been published previously. ²¹ , ²² Images were manually inspected for motion, artifacts, and structural anomalies. Using a deep neural network approach, two segmentation tools were used for further processing of the images. ²³ , ²⁴ These were first analyzed with the whole‐brain segmentation suite SynthSeg ²⁵ to obtain total intracranial volume (TIV), as well as frontal and temporal lobe volumes. In parallel, these scans were also processed with a hypothalamic tool distributed with FreeSurfer ²³ that provided volumes of five hypothalamic subregions (anterior inferior and superior, tubular inferior and superior, and posterior subunits of the left and right hypothalamus). Hypothalamic nuclei included in the subunits have been described in earlier publications. ¹³ , ²³ Scans with a resolution > 1.5 mm (in any plane) were excluded from this study, as the accuracy of the used segmentation tool lowers beyond this resolution. To ensure the correctness of these automated segmentations, visual quality control was performed, and subjects with failed segmentations were discarded. Finally, the hypothalamus total and subregional volumes were calculated based on the resulting segmentations, enabling group comparisons and correlation analysis with autonomic symptoms.

2.5. Statistical analyses

The statistical analysis of demographic and clinical data was conducted using IBM SPSS Statistics for Windows version 28.0 (IBM). Independent t tests, chi‐squared, and analysis of variance tests were used to compare baseline characteristics among the three groups. Differences in autonomic symptoms were assessed non‐parametrically using the Kruskal–Wallis test (due to non‐normally distributed data), corrected for multiple comparisons using Bonferroni corrections. Z scores within each group were calculated to identify outliers in hypothalamic volumes, with a cutoff set at −3 or +3 for removal. Hypothalamic volumetric differences were compared using analysis of covariance, adjusted for age, sex, TIV, and multiple comparisons (Bonferroni). Spearman correlation coefficients were calculated to test the correlation between autonomic symptoms and hypothalamic subregional volumes in FTD, PPD, and controls, adjusted for age, sex, and TIV. Volumetric correlations between hypothalamic regions and frontal and temporal lobe volume were assessed in the FTD, PPD, and controls to assess the relationship between frontotemporal atrophy and hypothalamic degeneration (Supplementary Materials, Figure S1 in supporting information). For the latter analysis, the Benjamini and Hochberg ²⁶ method for multiple comparisons was applied to avoid false discovery rates (p < 0.01).

3. RESULTS

3.1. Demographics

In Arm A: Assessment of autonomic symptoms using the ASQ (n = 91), the participants who completed the ASQ included 31 FTD patients (n = 24 probable bvFTD, n = 5 definite FTD, n = 1 FTD MND, n = 1 SD), 30 PPD patients (n = 14 MDD, n = 6 ASD, n = 4 BD, n = 3 PSD, n = 1 ADHD, n = 2 unspecified), and 30 controls. Table 1 displays the full demographics. Briefly, in Arm A (n = 91), there were no differences in age (mean ± standard deviation: FTD 67 ± 7.2; PPD 66 ± 7.1; controls 69 ± 6.6; p = 0.258) and sex (p = 0.48). FTD patients had higher scores on the CDR compared to PPD and controls (mean ± standard deviation: FTD 1.2 ± 0.6; PPD 0.5 ± 0.5; controls 0 ± 0). PPD patients used psychotropic drugs more often compared to FTD patients (p = <0.001) and controls (p < 0.001). There were no differences among FTD, PPD, and controls in the use of laxatives (p = 0.40), anti‐arrhythmic (p = 0.60), or antihypertensive medication (p = 0.22).

TABLE 1.

Demographics.

ASQ (Arm A)	FTD	PPD	Controls	p value
n	31	30	30
Age, mean ± StD, years	67 ± 7.2	66 ± 7.1	69 ± 6.6	0.258 ×
Sex, M/F	17/13	20/10	16/15	0.48 †
Disease duration, mean ± StD, years	7.8 ± 3.0	9.4 ± 6.3		0.40 †
Diagnosis, n	Probable bvFTD = 24	MDD = 14
	Definite FTD = 5	ASD = 6
	FTD‐MND = 1	BD = 4
	SD = 1	PSD = 3
		ADHD = 1 Unclassified = 2
CDR, mean ± StD	1.2 ± 0.6	0.5 ± 0.5	0 ± 0	<0.001 ×
MMSE, mean ± StD	23.7 ± 3.4	25.6 ± 3.3	29.5 ± 1	<0.001 ×
FAB, mean ± StD	12.7 ± 3.2	15.1 ± 2.4	17.2 ± 1.2	<0.001 ×
Ekman 60 Faces test, mean ± StD	31.2 ± 9.6	39.0 ± 7.9		<0.001 ∼
Medication use
Psychotropic drugs, n Antihypertensives, n Laxative, n Anti‐arrhythmics, n	8	24	4	<0.001 †
	6	11	6	<0.22 †
	6	3	3	<0.40 †
	2	1	3	<0.60 †
MRI scans (Arm B)
n	259	44	62
Age, mean ± StD, years	64 ± 8.4	60 ± 7.4	70 ± 4.5	<0.001 ×
Sex, M/F	159/100	26/18	44/18	0.33 †
Diagnosis, n	Probable bvFTD = 132	MDD = 27
	Definite FTD = 69	BD = 6
	FTD‐MND = 11	PSD = 4
	Other = 47	Unclassified = 7
Overlapping participants from Arms A and B (Arm C), with available MRI and ASQ
n	18	21	20
Age, mean ± StD, years	67 ± 7.5	64 ± 5.9	69 ± 5.3	0.106 ×
Sex, M/F	10/8	14/7	9/11	0.388 †
Diagnosis, n	Probable bvFTD = 13	MDD = 9
	Definite FTD = 4 FTD‐	ASD = 5
	MND = 1	BD = 4
		PD = 2, unclassified = 1

Abbreviations: ADHD, attention deficit hyperactive disorder; ASD, autism spectrum disorder; ASQ, Autonomic Symptoms Questionnaire; BD, bipolar disorder; bvFTD, behavioral variant frontotemporal dementia; CDR, Clinical Dementia Rating; definite, definite frontotemporal dementia; FAB, Frontal Assessment Battery; FTD, frontotemporal dementia; FTD MND, frontotemporal dementia motor neuron disease; MDD, major depression disorder; MMSE, Mini‐Mental State Examination; MRI, magnetic resonance imaging; PPD, primary psychiatric disorders; PSD, psychosis spectrum disorder; SD, semantic dementia; StD, standard deviation.

^×Analysis of variance. ^†Chi‐squared. ^∼Unpaired t test.

In Arm B: Structural assessment of hypothalamic volume (n = 365), in the larger sample of participants with available MRI scan, diagnoses included n = 259 FTD (n = 132 probable bvFTD, n = 69 definite FTD, n = 11 FTD MND, n = 47 other FTD), n = 44 PPD (n = 27 MDD, n = 6 BD, n = 4 PSD, n = 7 unclassified), and 62 controls. In Arm B, controls were older compared to PPD but not compared to FTD patients (mean ± standard deviation: FTD 64 ± 8.4; PPD 60 ± 7.4; controls 70 ± 4.5; p < 0.001), and there were no differences in sex among FTD, PPD, and controls (p = 0.33).

In Arm C: Overlapping participants from Arms A and B (n = 59), participants who completed the ASQ and had an available MRI scan included n = 18 FTD (n = 13 probable bvFTD, n = 4 definite FTD, n = 1 FTD MND), n = 21 PPD (n = 9 MDD, n = 5 ASD, n = 4 BD, n = 2 PD, n = 1 unspecified), and 20 controls. In Arm C, there were no differences in age (mean ± standard deviation: FTD 67 ± 7.5; PPD 64 ± 5.9; controls 69 ± 5.3; p = 0.106) or sex (p = 0.388) between groups.

3.2. Autonomic symptoms in FTD, PPD, and controls (Arm A)

FTD patients showed increased overall autonomic symptoms (total score ASQ) compared to controls (FTD vs. controls p = 0.006) but not compared to PPD (p = 0.347). Between PPD and controls, there were no differences in overall autonomic symptoms (p = 0.376). Both FTD and PPD showed increased cardiovascular‐related symptoms compared to controls (FTD vs. PPD p = 1.000; FTD vs. controls p = 0.020; PPD vs. controls p = 0.049). In addition, FTD patients showed increased thermoregulatory‐related symptoms compared to PPD (p = 0.026) and controls (p = 0.012). Furthermore, FTD patients more often experienced altered pain perception compared to PPD and controls (FTD 13%; PPD 2%; controls 1%; p < 0.001). From these participants with an altered pain perception, the majority reported an increased pain perception (FTD 85%; PPD: 67%; controls 50%). There were no differences in gastrointestinal (p = 0.302), urinary (p = 0.092), or sleep‐related symptoms (p = 0.317) among FTD, PPD, and controls (Figure 1).

FIGURE 1.

The y axis represents the ASQ score. Confidence interval is represented by the vertical line inside each bar. ASQ, Autonomic Symptom Questionaire; FTD, frontotemporal dementia; PPD, primary psychiatric disorders

3.3. Hypothalamic (subregional) volumes in FTD, PPD, and controls (Arm B)

FTD patients showed bilaterally lower total hypothalamic volumes compared to PPD and controls (p < 0.001, η ² = 0.194 left, η ² = 0.232 right; Table 2 and Figure 2). Additionally, in FTD, the bilateral hypothalamic anterior inferior (p < 0.001 η ² = 0.086 left, η ² = 0.137 right), anterior superior (p < 0.001 η ² = 0.221 left, η ² = 0.184 right), posterior (P < 0.001, η ² = 0.205 left, η ² = 0.226 right), tubular superior (p < 0.001, η ² = 0.087 left, η ² = 0.174 right), and right tubular inferior (FTD vs. PPD: p = 0.009; FTD vs. controls: p = 0.017, η ² = 0.039) subregional volumes were lower, compared to PPD and controls. The volume of the left tubular inferior region was lower in FTD compared to PPD (p = 0.016, η ² = 0.024), but showed no difference compared to controls (p = 0.466). PPD patients showed no differences in total and subregional hypothalamic volumes compared to controls.

TABLE 2.

Volumetric comparison of hypothalamic (sub)regions among FTD, PPD, and controls.

Hypothalamic subunit	FTD n = 259 mean (± SD)	PPD n = 44 mean (± SD)	Controls n = 62 mean (± SD)	Group comparison	Effect size η 2
Left hypothalamus	0.2332 (0.026)	0.2605 (0.021)	0.2579 (0.021)	FTD versus PPD: p < 0.001 FTD versus controls: p < 0.001 PPD versus controls: p = 1.000	0.194
Right hypothalamus	0.2180 (0.030)	0.2521 (0.021)	0.2515 (0.024)	FTD versus PPD: p < 0.001 FTD versus controls: p < 0.001 PPD versus controls: p = 1.000	0.232
Left anterior inferior	0.0090 (0.003)	0.0106 (0.003)	0.0107 (0.003)	FTD versus PPD: p < 0.001 FTD versus controls: p < 0.001 PPD versus controls: p = 1.000	0.086
Right anterior inferior	0.0079 (0.003)	0.0102 (0.003)	0.0104 (0.003)	FTD versus PPD: p < 0.001 FTD versus controls: p < 0.001 PPD versus controls: p = 1.000	0.137
Left anterior superior	0.0112 (0.003)	0.0136 (0.003)	0.0145 (0.003)	FTD versus PPD: p < 0.001 FTD versus controls: p < 0.001 PPD versus controls: p = 0.168	0.221
Right anterior superior	0.0103 (0.003)	0.0134 (0.003)	0.0134 (0.003)	FTD versus PPD: p < 0.001 FTD versus controls: p < 0.001 PPD versus controls: p = 1.000	0.184
Left posterior	0.0616 (0.013)	0.0743 (0.009)	0.0739 (0.009)	FTD versus PPD: p < 0.001 FTD versus controls: p < 0.001 PPD versus controls: p = 1.000	0.205
Right posterior	0.0570 (0.010)	0.0708 (0.010)	0.0720 (0.009)	FTD versus PPD: p < 0.001 FTD versus controls: p < 0.001 PPD versus controls: p = 0.387	0.226
Left tubular inferior	0.0874 (0.011)	0.0923 (0.009)	0.0893 (0.009)	FTD versus PPD: p = 0.016 FTD versus controls: p = 0.466 PPD versus controls: p = 0.575	0.024
Right tubular inferior	0.0804 (0.011)	0.0864 (0.009)	0.0845 (0.010)	FTD versus PPD: p = 0.009 FTD versus controls: p = 0.017 PPD versus controls: p = 0.1000	0.039
Left tubular superior	0.0641 (0.009)	0.0697 (0.007)	0.0695 (0.008)	FTD versus PPD: p < 0.001 FTD versus controls: p < 0.001 PPD versus controls: p = 1.000	0.087
Right tubular superior	0.0625 (0.009)	0.0712 (0.009)	0.0712 (0.008)	FTD versus PPD: p < 0.001 FTD versus controls: p < 0.001 PPD versus controls: p = 1.000	0.174

Note: Data show hypothalamic volumes (mm³) and differences between groups using analysis of covariance adjusted for age, sex, total intracranial volume, and multiple comparisons (Bonferroni).

Abbreviations: FTD, frontotemporal dementia; PPD, primary psychiatric disorders; SD, standard deviation.

FIGURE 2.

Boxplots represent volumes of hypothalamic subregions between groups. All hypothalamic (subregional) volumes were lower in FTD, compared to PPD and controls, adjusted for age, sex, total intracranial volume, and multiple comparisons, except for the left tubular inferior region, which showed no differences compared to FTD and controls but was lower in FTD, compared to PPD. FTD, frontotemporal dementia; PPD, primary psychiatric disorders

3.4. Correlation between hypothalamic subregional volumes and autonomic symptoms in FTD, PPD, and controls (Arm C)

In FTD, increased overall autonomic symptoms (total ASQ score), positively correlated with the left anterior inferior (r _s = 0.57, 95% confidence interval [CI]: −0.04 to 0.91, p = 0.027), bilateral anterior superior (left: r _s = 0.61, 95% CI: −0.11 to 0.84, p = 0.015; right: r _s = 0.58, 95% CI: −0.03 to 0.92, p = 0.023), bilateral posterior (left r _s = 0.57, 95% CI: −0.14 to 0.81, p = 0.025; right r _s = 0.72, 95% CI: 0.03–0.98, p = 0.003), and right tubular superior (r _s = 0.58, 95% CI: −0.07 to 0.88, p = 0.023) hypothalamic subregional volumes. In addition, increased overall autonomic symptoms positively correlated with bilateral lower total hypothalamic volumes (left: r _s = 0.61, 95% CI: −0.21 to 0.74, p = 0.015; right: r _s = 0.65, 95% CI: −0.02 to 0.93, p = 0.009). Increased cardiovascular symptoms positively correlated with the right hypothalamus (r _s = 0.53, 95% CI: −0.26 to 0.69, p = 0.042). Increased thermoregulatory‐related symptoms showed a positive correlation with the left anterior inferior (r _s = 0.60, 95% CI: 0.03–0.98, p = 0.018), left anterior superior (r _s = 0.68, 95% CI: −0.08 to 0.87, p = 0.005), bilateral posterior hypothalamic subregions (left: r _s = 0.57, 95% CI: −0.18 to 0.77, p = 0.027; right: r _s = 0.63, 95% CI: −0.03 to 0.92, p = 0.012), and bilateral total volumes of the hypothalamic volumes (left: r _s = 0.57, 95% CI: −0.43 to 0.53, p = 0.026; right: r _s = 0.63, 95% CI: −0.29 to 0.66, p = 0.012). Increased urogenital symptoms positively correlated with the right anterior superior (r _s = 0.68, 95% CI: 0.17–1.12, p = 0.005) and right tubular superior (r _s = 0.62, 95% CI: 0.11–1.06, p = 0.014) and bilateral total hypothalamic volumes (left: r _s = 0.58, 95% CI: 0.00–0.95, p = 0.022; right: r _s = 0.64, 95% CI: 0.13–1.08, p = 0.010). Increased sleep‐related symptoms negatively correlated to the right tubular inferior (r _s = −0.55, 95% CI: −0.92 to 0.03, p = 0.032) hypothalamic subregion. Increased altered pain perception showed positive correlations to the left anterior superior (r _s = 0.58, 95% CI: −0.18 to 0.77, p = 0.024) and left posterior (r _s = 0.66, 95% CI: −0.09 to 0.86, p = 0.007) hypothalamic subregions. In FTD, there were no correlations between gastrointestinal symptoms and hypothalamic (subregional) volumes. Figure 3 shows the heatmap of correlations between autonomic symptoms and subregional hypothalamic volumes in FTD, adjusted for age, sex, and TIV.

FIGURE 3.

Heatmap shows the results of the partial Spearman correlations, adjusted for age, sex, and total intracranial volume. Shown in the cells: Spearman correlation coefficient, P value and 95% confidence interval. Significant results are highlighted yellow. ASQ, Autonomic Symptom Questionnaire; FTD, frontotemporal dementia

In PPD, increased overall autonomic symptoms positively correlated with the right anterior inferior subregion of the hypothalamus (r _s = 0.45, 95% CI: 0.01–0.89, p = 0.041). Increased gastrointestinal symptoms positively correlated with the left posterior (r _s = 0.58, 95% CI: −0.03 to 0.85, p = 0.012), left tubular inferior (r _s = 0.51, 95% CI: 0.14–1.02, p = 0.029), right tubular inferior subregions (r _s = 0.54, 95% CI: 0.10–0.97, p = 0.020), and bilateral total hypothalamic volumes (left: r _s = 0.56, 95% CI: 0.17–1.04, p = 0.016; right: r _s = 0.47, 95% CI: 0.07–0.95, p = 0.047). Thermoregulatory‐related symptoms positively correlated with the left anterior inferior (r _s = 0.48, 95% CI: 0.00–0.88, p = 0.044), left posterior (r _s = 0.55, 95% CI: −0.07 to 0.81, p = 0.017), right anterior inferior (r _s = 0.51, 95% CI: 0.00–0.88, p = 0.032), and superior (r _s = 0.47, 95% CI: 0.05–0.93, p = 0.049) hypothalamic subregions. Increased altered pain perception positively correlated with the bilateral posterior subregions (left: r _s = 0.62, 95% CI: −0.07 to 0.81, p = 0.006; right: r _s = 0.48, 95% CI: −0.07 to 0.81, p = 0.045), and bilateral total hypothalamic volumes (left: r _s = 0.49, 95% CI: −0.07 to 0.81, p = 0.037; right: r _s = 0.48, 95% CI: −0.07 to 0.81, p = 0.046).

In controls, both increased cardiovascular (r _s = 0.54, 95% CI: 0.01–0.91, p = 0.026) and gastrointestinal symptoms (r _s = 0.72, 95% CI: 0.09–0.99, p = 0.001) positively correlated with the right anterior inferior hypothalamic subregion. Figure S2 in supporting information shows the heatmap of correlations between autonomic symptoms and subregional hypothalamic volumes in PPD and controls, adjusted for age, sex, and TIV.

4. DISCUSSION

We found that compared to PPD patients and controls, FTD patients displayed overall autonomic dysfunction, reflected by cardiovascular and thermoregulatory symptoms, as well as increased pain perception. Most hypothalamic (subregional) volumes were smaller in FTD patients compared to PPD and controls. Additionally, increased overall autonomic, cardiovascular, thermoregulatory, sleep, and urinary symptoms in FTD were related to specific hypothalamic subregions.

These findings align with previous studies investigating autonomic dysfunction in FTD. ¹ , ³ , ²⁷ Unlike end‐stage FTD patients, who display extensive gastrointestinal symptoms, ²⁷ our early‐stage sample did not, suggesting these symptoms emerge later in disease progression. Interestingly, both FTD and PPD showed increased cardiovascular autonomic symptoms compared to controls, consistent with studies indicating altered heart and respiration rates, reduced heart rate variability, bradycardia, and orthostatic hypotension in FTD. ²⁸ , ²⁹ , ³⁰ In PPD, autonomic dysfunction may relate to psychotropic medication, comorbid cardiovascular disease, and hypothalamic pathology. ⁵ , ⁶ , ⁷ , ³¹ , ³² , ³³ , ³⁴ Most PPD patients in our sample used psychotropic drugs, which are associated with cardiovascular risks, especially atypical antipsychotics, tricyclic antidepressants, and mood stabilizers. These drugs can contribute to metabolic syndrome, QT prolongation, and arrhythmias. ³⁵ Lifestyle factors such as poor diet, inactivity, smoking, and limited health‐care access further compound cardiovascular risks. ³⁶ , ³⁷ , ³⁸ This underlines the need for careful monitoring and targeted interventions to address cardiovascular health in PPD and FTD patients.

Interestingly, the hypothalamus contains several nuclei involved in cardiovascular control, including the PVN, supraoptic nucleus, arcuate nucleus (ARC), and dorsomedial nucleus (DMN). ³⁹ , ⁴⁰ , ⁴¹ , ⁴² These nuclei regulate sympathetic outflow (e.g., blood pressure), fluid balance, and responses to psychological stress. Degenerative changes of these and other hypothalamic nuclei could potentially lead to autonomic symptomatology. Future prospective studies should investigate associations between specific hypothalamic nuclei and cardiovascular symptoms across disorders to support transdiagnostic understanding and guide targeted interventions.

We found smaller total and subregional hypothalamic volumes in FTD compared to PPD and controls, with sparing of the left inferior tubular region. This supports prior findings of hypothalamic degeneration in FTD, ⁸ , ²⁵ made possible through advanced neuroimaging techniques ¹³ , ¹⁴ and is corroborated by post mortem studies. ¹⁴ , ⁴³ Our results further show that hypothalamic volume loss in FTD relates to frontotemporal degeneration, unlike in PPD and controls, suggesting the hypothalamus is a relevant subcortical structure in FTD pathology. ⁴⁴

Despite pathophysiological evidence of hypothalamic involvement in PPD, ⁴⁵ we found no volumetric group differences compared to controls. While structural brain changes are reported in PPD ⁴⁶ , ⁴⁷ , overt hypothalamic atrophy is rare. Grouping diverse psychiatric disorders with potentially distinct hypothalamic alterations may dilute group‐level effects.

Contrary to our hypothesis, we found a positive correlation between autonomic symptoms and hypothalamic volume in FTD and PPD. One explanation could be early stage neuroinflammatory processes and gliosis, which may cause temporary tissue swelling before atrophy ensues. ⁴⁸ , ⁴⁹ Given that our FTD cohort was at an early disease stage, this could account for the observed pattern. Similar neuroinflammatory activity in PPD may also play a role. ⁵⁰ However, the small sample size limits interpretation, and larger studies are needed to investigate the temporal dynamics of hypothalamic volume changes.

This study is the first to explore autonomic symptoms and hypothalamic volume in both FTD and PPD, which is important because PPD constitutes the main differential diagnosis of FTD. ⁴ We assessed multiple autonomic domains and used a large MRI sample, though only a subset completed both ASQ and MRI. This limits statistical power for correlation analyses. Potential recall bias from the ASQ, which asks about a 6‐month symptom history, must also be considered. Moreover, FTD patients may underreport symptoms due to disease insight, ²⁸ which we partially mitigated by involving caregivers during ASQ administration. Future research should incorporate objective biometric data to complement subjective reports, as these were not assessed in the current study. While the segmentation tool was developed for 1 mm isotropic resolution, most scans had lower resolution. Nevertheless, the tool showed robustness due to training data augmentation and generalizability to other “unseen” pathologies like AD. ²³

This study, supported by prior research, ¹ , ³ , ²⁷ underscores the high prevalence of autonomic symptoms in FTD, potentially linked to hypothalamic atrophy. Given the prevalence of cardiovascular symptoms in both FTD and PPD, it is imperative to enhance attention to this aspect in clinical practice and, when necessary and feasible, to address and treat these symptoms. Because autonomic symptomatology in FTD is widespread and also occurs in PPD and other neurodegenerative diseases, likely due to cross‐disorder pathology in the hypothalamus, ⁵ , ⁶ , ⁷ it is unlikely to serve as a specific FTD biomarker. However, the presence of altered pain perception does signal more toward FTD than PPD. In light of prospective diagnostic criteria for FTD, it is essential to incorporate this understanding, alongside other, more specific biomarkers. Further research using objective biometric measurements of autonomic function and including a broader range of control groups, such as Lewy body dementia, AD, and vascular dementia, is necessary to determine whether there are disease‐specific autonomic profiles.

FUNDING INFORMATION

No funding was received toward this work.

ETHICAL APPROVAL

All participants provided informed consent.

CONFLICT OF INTEREST STATEMENT

The authors declare no conflicts of interest. Author disclosures are available in the Supporting Information.

Supporting information

Supporting Information

Supporting Information