A Case Series on Longitudinal Hypothalamic Morphology in Mild Cognitive Impairment

Article information

Ann Geriatr Med Res. 2026;30(2):285-289

Publication date (electronic) : 2026 March 31

doi : https://doi.org/10.4235/agmr.25.0122

Jeffrey Cayaban Pagaduan^,1,2, Melvin Barrientos Marzan³, Xiaoling Wang⁴, Benjamin Billot⁵

¹Melbourne School of Psychological Sciences, The University of Melbourne, Melbourne, Australia

²Faculty of Physical Culture, Palacký University Olomouc, Olomouc, Czech Republic

³Department of Obstetrics, Gynaecology & Newborn Health, Melbourne Medical School, Faculty of Medicine, Dentistry and Health Sciences, The University of Melbourne, Melbourne, Australia

⁴Georgia Prevention Institute, Augusta University, Augusta, GA, USA

⁵Computer Science and Artificial Intelligence Laboratory, Massachusetts Institute of Technology, Cambridge, MA, USA

Corresponding Author: Jeffrey Cayaban Pagaduan, PhD Melbourne School of Psychological Sciences, The University of Melbourne, Grattan Street, Parkville, Victoria 3010, Australia E-mail: jcpagaduan@gmail.com

Received 2025 August 5; Revised 2025 December 7; Accepted 2026 March 29.

Abstract

INTRODUCTION

Mild cognitive impairment (MCI) is considered the transitional stage between healthy aging and the onset of Alzheimer’s disease (AD). About one in three individuals with MCI will most likely develop AD within 5 years.1) One phenomenon affected by MCI is an alteration in the volumetric structure of the hypothalamus.2,3) The hypothalamus plays various functions to maintain homeostasis.4,5)

To date, only two studies have examined the volumetric hypothalamus in MCI.2,3) Pecher et al.2) recorded decreased hypothalamic volume in MCI. Quattrini et al.3) detected right anterior-posterior and superior tubular degradation in MCI. The aforementioned studies were conducted using a crossover design, which limited their ability to provide key insights into the progression of hypothalamic atrophy in MCI. There is also no information on hypothalamic lateralization with MCI.6)

In this case series, we investigated the longitudinal hypothalamic structural alteration occurring in early-stage MCI among females, who are more likely to experience MCI than males.7) We employed a hypothalamic segmentation technique that allowed the parcellation of 10 regions, delivering valuable information on hypothalamic morphological changes associated with the progression of MCI.8,9)

CASE REPORT

We examined sex-specific (female) brain MRI data from the Harvard Aging Brain Study, which included imaging at both the first and fifth years.10) Out of 173 females from the first year, 61 qualified as healthy controls (age 71.9±5.86 years; education 16.1±2.68 years; verbal intelligence quotient [VIQ] 123.6±7.62), while nine females (age 73.8±4.80 years; education 16.3±2.74 years; VIQ 120.3±11.8) demonstrated MCI after neuroimaging at fifth year.9) Briefly, we employed the recon-all processing stream in Freesurfer version 7.3.2 on the T1-weighted images, deriving estimated total intracranial volume (eTIV), which was normalized and used for analysis.11) Then, we conducted an automated hypothalamic segmentation to generate the following subunits:

1) Anterior-inferior (AI): suprachiasmatic nucleus, supraoptic nucleus (SON);

2) Anterior-superior (AS): preoptic area, paraventricular nucleus (PVN);

3) Posterior (P): mammillary body (including medial and lateral mammillary nuclei, lateral hypothalamus), tuberomammillary nucleus (TMN);

4) Inferior tubular (IT): infundibular or arcuate nucleus; ventromedial nucleus, SON, lateral tubular nucleus, TMN; and,

5) Superior tubular (ST): dorsomedial nucleus, PVN, lateral hypothalamus.8,9)

Fig. 1 displays an example of longitudinal hypothalamic alteration from an individual with MCI. Additionally, we assessed hypothalamic nuclei asymmetry through the lateralization index (LI), where LI = (left hypothalamus subregion volume – right hypothalamus subregion volume) / (0.5 * (left hypothalamus subregion volume + right hypothalamus subregion volume)) * 100.6) A positive LI depicts larger left than right nuclei, while a negative LI indicates greater right than left nuclei.

Fig. 1.

Hypothalamic alteration from an mild cognitive impairment participant in year 1 and year 5 (AI, anterior inferior; AS, anterior superior; IT, inferior tubular; ST, superior tubular; P, posterior).

To examine the differences in hypothalamic subunits, we used the lme4 package in R v.4.3.3 to implement multivariable linear models.12) The models included hypothalamic regions (right and left) as dependent variables, with time point (Year 1 and Year 5), age, eTIV, years in education, and VIQ as covariates.6,13) Time was modelled as a categorical factor with two levels representing the repeated measurement occasions. We included an interaction term between group and time to test whether longitudinal changes differed between groups. We utilized the Wald Test to assess the significance of the interaction term.14) Significance level was set at 0.05 level alpha.

Table 1 presents brain and hypothalamic volumes in MCI and healthy controls. We found a significant inverse relationship between age and all alterations in the left hypothalamic nuclei, as well as most changes in the right hypothalamic nuclei. A Wald test comparing models with and without group and time interaction resulted in a significantly better model fit (p<0.05) when the interaction was retained. With this method, we confirmed differential longitudinal trajectories in the left anterior-inferior (LAI, p<0.01), left anterior-superior (LAS, p<0.05), and left posterior (LP, p<0.01) between groups. Table 2 depicts the generated linear model for the left hypothalamic nuclei.

Table 1.

Brain and hypothalamic nuclei volumes in MCI individuals and healthy controls

Table 2.

Longitudinal left hypothalamic nuclei alteration in MCI individuals and healthy controls

No significant group differences were found in right anterior-inferior (RAI, p=0.08), right anterior-superior (RAS, p=0.91), right posterior (RP, p=0.35), right inferior tubular (RIT, p=0.98), and right superior tubular (RST, p=0.85). Table 3 displays the right hypothalamic nuclei alteration in MCI and healthy controls.

Table 3.

Longitudinal right hypothalamic nuclei alteration in MCI individuals and healthy controls

There was a significant difference in LI at AS (p<0.05). No LI differences were found in AI (p=0.37), P (p=0.08), IT (p=0.91), and ST (p=0.67).

DISCUSSION

We examined the longitudinal alteration of the hypothalamic nuclei in the progression of MCI. Results revealed volumetric reduction in the LAI, LAS, and LP hypothalamic nuclei with MCI. Further, the MCI demonstrated a greater negative lateralization index at the AS hypothalamus than the healthy group.

Indeed, we detected atrophy in the LAI, LAS, and LP hypothalamic subunits. While baseline age effects reflect cross-sectional, age-related differences in brain structure, the significant interaction of group and time demonstrates that MCI participants experienced accelerated hypothalamic atrophy beyond normal aging trajectories of the regions mentioned above, suggesting MCI-specific neurodegenerative processes. The volume loss in these areas may be due to proximity to the ventricles, increasing susceptibility to degradation.15) Another possibility is the depletion of norepinephrine within these hypothalamic subregions.7,16) Thus, structural degeneration in the LAI, LAS, and LP hypothalamic nuclei implies disruption of metabolic processes critical to cognitive functions.7,17)

In this study, we also identified LAS lateralization in MCI. The AS nuclei include the preoptic area and paraventricular nucleus. The preoptic area is essential in sleep and thermoregulation.18) On the other hand, the paraventricular nucleus is one of the autonomic control centers regulating cardiovascular, gastrointestinal, and renal functions.19) Researchers documented autonomic dysfunction20,21) and sleep disorders22,23) in MCI. Therefore, the left lateralization in the AS under MCI suggests interruption in homeostasis, potentially leading to autonomic dysfunction and poor sleep hygiene.

We acknowledge the limitations of this case series. First, the small sample size may not be sufficient for the generalizability of results to broader populations. Second, only females were included in the study, as only two males progressed to MCI, and were excluded. Such an imbalance reflects evidence suggesting that MCI is more prevalent among older females than males.7) Incorporating male populations in future studies may elucidate sex-specific alterations in MCI. Lastly, including other physiological variables may provide additional insights into the influence of hypothalamic degradation on physiological functioning.

Article information Continued