Dementia is defined as a syndrome characterized by progressive deterioration of cognitive function, including the ability to process thought, beyond what is expected during normal biological ageing. It affects memory, thinking, orientation, comprehension, calculation, learning capacity, language, and judgement, while consciousness remains unaffected. Impairment in cognitive function is often accompanied by, or may sometimes precede, changes in mood, emotional control, behaviour, or motivation.1 Dementia is one of the major causes of disability and dependency among the elderly population worldwide. In India, its prevalence among individuals aged 40 years and above is 0.43%, while among those aged 65 years and above, it is 2.44%. There is often a lack of awareness and understanding of dementia, resulting in stigmatization and barriers to diagnosis and care. Therefore, it is essential to estimate the disease burden in the community, identify risk factors, and implement appropriate measures.1 The relationship between the hippocampus and memory is well established, and a reduction in hippocampal volume is frequently observed in patients with dementia. However, there is a paucity of statistically significant data from India regarding this association.2 Magnetic resonance spectroscopy (MRS) can be used to assess various metabolites within the hippocampus, which may serve as indicators of dementia-related neuronal and biochemical changes.6 The objective of this study was to assess the hippocampus in clinically diagnosed patients with dementia using MR volumetry and spectroscopy to evaluate structural and biochemical changes, and to compare hippocampal measurements and biochemical alterations with clinical cognitive assessment.

This cross-sectional study was conducted over a period of 18 months, from December 2022 to June 2024, among clinically suspected or diagnosed patients with dementia who were referred for MRI of the brain to the Department of Radiodiagnosis, J.J.M. Medical College, Davangere. Inclusion criteria Patients aged above 40 years with clinically suspected or diagnosed dementia or cognitive impairment who were referred for MRI of the brain were included in the study. Exclusion criteria Patients with claustrophobia, orthopaedic or metallic implants, or cardiac pacemakers incompatible with MRI were excluded. Patients with acute delirium, a history of trauma, alcohol abuse, psychiatric illness, or acute vascular insults such as cerebral infarction or haemorrhage were also excluded. Data Collection and Imaging Technique Informed consent was obtained from all participants after obtaining ethical committee clearance and explaining the procedure. MMSE scores of clinically diagnosed cases of dementia according to the ICD-10 criteria were assessed and recorded. Hippocampal Volumetry MR images were acquired using a 1.5-T MRI system (Philips) with a standard head coil. The imaging sequences included 3D T1-weighted, axial T2-weighted, axial FLAIR, DWI/ADC, GRE, oblique coronal T1-weighted inversion recovery, and TOF MRA sequences. The regions of interest were manually delineated on successive MR images, and dedicated software was used to calculate the volumes of both hippocampi. The average hippocampal volume (HV) was calculated by summing the right and left hippocampal volumes and dividing the value by two. Normal hippocampal volume in the general population was standardized using data from a study published in NCBI. The mean average hippocampal volume of both sides was reported as 2.411 cc (SD = 0.299). In the present study, the reference range for HV was calculated using 90% confidence intervals (1.64 SD), with an upper limit of 2.901 cc and a lower limit of 1.920 cc. Magnetic Resonance Spectroscopy Multi-voxel MRS studies were performed in the region of the hippocampus containing the densest grey matter. The MRS parameters analysed were N-acetylaspartate/creatine (NAA/Cr), myo-inositol/creatine (mI/Cr), and choline/creatine (Cho/Cr) ratios. Statistical Analysis The accuracy of MRI in detecting dementia was evaluated by analysing hippocampal volume and its correlation with MRS changes within the hippocampus, using clinical assessment based on MMSE as the reference standard. The distribution of MRI imaging parameters was examined in relation to MMSE categories. Pearson or Spearman correlation coefficients and Fisher's exact test/chi-square test were used to assess the associations between MMSE scores and MRI findings

Descriptive and inferential statistical analyses were performed to evaluate the demographic, clinical, hippocampal volumetric, and magnetic resonance spectroscopy (MRS) findings in the study population. Demographic and clinical characteristics The study included 45 participants. The highest proportion of participants belonged to the 45–60-year age group (51%), followed by the 60–75-year age group (26%). A slightly higher proportion of cases was observed among men. No significant association was observed between the study findings and the presence of comorbidities such as hypertension, diabetes mellitus, or obesity. Clinical assessment according to MMSE Based on the MMSE categorization used in the study, 20 (44%) participants were classified in the mild category, followed by 12 (27%) in the moderate category and 9 (20%) in the severe category. Four (9%) participants were classified in the probable category. Hippocampal volume according to MMSE category The distribution of hippocampal volume across the different MMSE categories is shown in Table 1. Among the 4 participants in the probable category, all 4 (100%) had hippocampal volumes >1.920 cc, with no significant decrease. Among the 20 participants in the mild category, 18 (90%) had hippocampal volumes >1.920 cc and 2 (10%) had reduced hippocampal volume. Among the 12 participants in the moderate category, 5 (42%) had hippocampal volumes >1.920 cc, 6 (50%) had reduced hippocampal volume, and 1 (8%) had hippocampal atrophy. Among the 9 participants in the severe category, 2 (22%) had hippocampal volumes >1.920 cc, 4 (44%) had reduced hippocampal volume, and 3 (33%) had hippocampal atrophy. Table1: Distribution of hippocampal volume according to MMSE category Hippocampal volume category Probable, n (%) Mild, n (%) Moderate, n (%) Severe, n (%) Total >1.920 cc, no significant decrease 4 (100%) 18 (90%) 5 (42%) 2 (22%) 29 <1.920 cc, reduced 0 (0%) 2 (10%) 6 (50%) 4 (44%) 12 <1.813 cc, atrophy 0 (0%) 0 (0%) 1 (8%) 3 (33%) 4 Total 4 (100%) 20 (100%) 12 (100%) 9 (100%) 45 Association between hippocampal volume and MMSE classification For diagnostic analysis, MMSE categories 2 and 3 were grouped as MMSE-positive, while categories 0 and 1 were grouped as MMSE-negative (Table 2). Among the 16 participants with reduced hippocampal volume (<1.920 cc), 14 were MMSE-positive and 2 were MMSE-negative. Among the 29 participants with normal hippocampal volume, 7 were MMSE-positive and 22 were MMSE-negative. Table 2: Association between hippocampal volume and grouped MMSE classification Hippocampal volume MMSE-positive (2,3) MMSE-negative (0,1) Total Reduced (<1.920 cc) 14 2 16 Normal 7 22 29 Total 21 24 45 The Pearson correlation coefficient of approximately 0.62 indicated a moderate positive correlation between hippocampal volume and MMSE category. The reported chi-square analysis demonstrated a statistically significant association between hippocampal volume and MMSE classification (χ² ≈ 21.12, p ≈ 0.0017). Hippocampal volumetry demonstrated a sensitivity of 87.5% and specificity of 75.9% for identifying dementia based on the reference standard used. MRS findings according to MMSE category MRS findings for NAA/Cr, Cho/Cr, and mI/Cr across the different MMSE categories are presented in Table 3. For NAA/Cr, 4 (100%) participants in the probable category, 16 (80%) in the mild category, 8 (67%) in the moderate category, and 1 (11%) in the severe category had values without reduction. Reduced NAA/Cr (<1.23) was observed in 4 (20%) participants in the mild category, 4 (33%) in the moderate category, and 8 (89%) in the severe category, giving an overall total of 16 (36%) participants with reduced NAA/Cr. For Cho/Cr, 4 (100%) participants in the probable category, 14 (70%) in the mild category, 7 (58%) in the moderate category, and 5 (56%) in the severe category had values without reduction. Reduced Cho/Cr (<0.10) was observed in 6 (30%) participants in the mild category, 5 (42%) in the moderate category, and 4 (44%) in the severe category, giving an overall total of 15 (33%) participants with reduced Cho/Cr. The association between Cho/Cr and MMSE severity was not statistically significant (χ² = 2.975, p = 0.395). For mI/Cr, 2 (50%) participants in the probable category, 15 (75%) in the mild category, 3 (25%) in the moderate category, and 3 (33%) in the severe category had values without elevation. Elevated mI/Cr (>0.23) was observed in 2 (50%), 5 (25%), 9 (75%), and 6 (67%) participants in the probable, mild, moderate, and severe categories, respectively, giving an overall total of 22 (49%) participants. Table 3: Distribution of MRS metabolite ratios according to MMSE category MRS parameter Category Probable, n (%) Mild, n (%) Moderate, n (%) Severe, n (%) Total, n (%) NAA/Cr Normal 4 (100%) 16 (80%) 8 (67%) 1 (11%) 29 (64%) Reduced <1.23 0 (0%) 4 (20%) 4 (33%) 8 (89%) 16 (36%) Cho/Cr Normal 4 (100%) 14 (70%) 7 (58%) 5 (56%) 30 (67%) Reduced <0.10 0 (0%) 6 (30%) 5 (42%) 4 (44%) 15 (33%) mI/Cr Normal 2 (50%) 15 (75%) 3 (25%) 3 (33%) 23 (51%) Elevated >0.23 2 (50%) 5 (25%) 9 (75%) 6 (67%) 22 (49%) Total 4 20 12 9 45 Hippocampal volume and MRS findings Among the 16 participants with reduced hippocampal volume, corresponding to hippocampal volume categories 1 and 2, 9 (56%) showed a reduced NAA/Cr ratio, 7 (43%) showed a reduced Cho/Cr ratio, and 10 (62%) showed an increased mI/Cr ratio. Overall, the findings demonstrated a progressive increase in the proportion of reduced NAA/Cr with increasing severity of cognitive impairment. The distribution of mI/Cr also showed greater elevation in the moderate and severe MMSE categories, whereas the association between Cho/Cr and MMSE severity was not statistically significant.

The diagnosis of dementia is complex and involves multiple clinical and imaging parameters, with the hippocampus being an important region affected by neurodegenerative changes. MRI enables measurement of hippocampal volume, while MRS provides information regarding biochemical alterations within the hippocampus.6 In the present study, MMSE scores were correlated with imaging findings. To ensure a focused assessment, cases with acute vascular insults, including large cortical infarcts or haemorrhage, identified incidentally on imaging were excluded. This approach helped minimize potential confounding by acute structural brain lesions when assessing cognitive impairment using MMSE.1 Hippocampal Volumetry Findings Our study demonstrated a significant association between hippocampal volume and MMSE categories. Both the Pearson correlation coefficient (0.61859) and chi-square test (χ² ≈ 21.12, p ≈ 0.0017) supported this association, suggesting that reductions in hippocampal volume are associated with cognitive decline as assessed by MMSE. Avinash Vijaykumar et al. conducted a study demonstrating that hippocampal volume was reduced by 5% in individuals with mild dementia, 12% in those with moderate dementia, and 34% in those with severe dementia compared with individuals with normal cognitive function. The reduction in hippocampal volume was closely associated with increasing dementia severity, as reflected by MMSE scores.3 MR Hippocampal Spectroscopy Findings A statistically significant reduction in the NAA/Cr ratio was observed in the hippocampus. The high proportion of reduced NAA/Cr among patients with severe cognitive impairment (88.9%) may indicate significant neuronal loss or dysfunction in advanced stages of cognitive impairment. The odds ratio of 28.00 for reduced NAA/Cr in the severe MMSE category compared with the other categories indicates that patients with severe cognitive impairment were 28 times more likely to have a reduced NAA/Cr ratio. This finding may have potential implications for risk stratification and further clinical assessment. In our study, a reduced Cho/Cr ratio was observed among patients with more severe cognitive impairment. However, this association was not statistically significant. These findings suggest that the Cho/Cr ratio, which is often associated with cell membrane turnover and neuronal integrity, may be related to cognitive decline as assessed by MMSE categories. However, correlation does not imply causation, and larger studies with additional clinical parameters are required to establish this relationship more definitively. Normal mI/Cr ratios were more prevalent among patients with milder forms of cognitive impairment. A higher proportion of increased mI/Cr ratios was observed in the severe MMSE category, suggesting that an increased mI/Cr ratio may be associated with more advanced cognitive decline. Rose SE et al., in their study, reported notable metabolic differences in patients with dementia, including decreased N-acetylaspartate (NAA) levels and a reduced NAA/creatine (Cr) ratio. They also observed elevated myo-inositol levels, resulting in an increased myo-inositol/creatine (mI/Cr) ratio. However, no significant differences were observed in the choline/creatine (Cho/Cr) ratio.5

Dementia is a neurodegenerative disorder that significantly affects the hippocampal formation, an important region involved in cognitive and neuropsychiatric functions. Clinical diagnosis can be challenging, making early identification of cognitive impairment important for timely management and preservation of cognitive function. MRI enables quantitative assessment of hippocampal volume and provides information regarding structural changes associated with dementia, while MRS provides information about biochemical alterations within hippocampal tissue. Together, these imaging modalities may complement clinical assessment and provide objective markers of structural and metabolic changes associated with cognitive impairment. Our findings demonstrated a significant association between declining MMSE scores and reduced bilateral hippocampal volume. Hippocampal spectroscopy demonstrated a reduction in the NAA/Cr ratio and an increase in the mI/Cr ratio among patients with reduced hippocampal volume. The NAA/Cr ratio also decreased with declining MMSE scores, suggesting an association with increasing cognitive impairment. The higher proportion of increased mI/Cr ratios in the severe MMSE category further suggests its potential association with advanced cognitive decline.

1. Chui HC, Ramirez-Gomez L. Clinical and imaging features of mixed Alzheimer and vascular pathologies. Alzheimers Res Ther [Internet]. 2015;7(1):21. 2. N Dolek, S Saylisoy, D Ozbabalik and B Adapinar, 2012, Comparison of Hippocampal Volume Measured Using Magnetic Resonance Imaging in Alzheimer's disease, Vascular Dementia, Mild Cognitive Impairment and Pseudo-dementia, The Journal of International Medical Research, Volume 40: page 718. 3. Avinash Vijayakumar and Abhishek Vijayakumar, 2013, Comparison of Hippocampal Volume in Dementia Subtypes, ISRN Radiology, Volume 2013, Article ID 174524, page 3. 4. Wang H, Shu L, Xie J, Zhang H, Zhang D. Diagnostic utility of neuropsychological performance and quantitative MRI-based measurement in Alzheimer disease. Alzheimer Dis Assoc Disord. 2004;18(3):163–70. 5. Rose SE, De Zubicaray GI, Wang D, et al. A 1H MRS study of probable Alzheimer's disease and normal aging: implications for longitudinal monitoring of dementia progression. Magn Reson Imaging. 1999;17:291–299. 6. K Kantarci, MD, 2007,1H Magnetic resonance spectroscopy in dementia, The British Journal of Radiology, 80 (2007), S146 7. Pedro J. Modrego. The contribution of magnetic resonance spectroscopy as biomarker in Alzheimer’s disease, Journal of Biomedical Graphics & Computing, 2013,Vol. 3,No. 1,66-74.1.