Imaging Approaches in Dementia: A Retrospective Cohort Study of Cross-sectional Imaging in the Indian Population

1. Research Intern, Department of Radiology, Nanavati Max Superspeciality Hospital, Mumbai, Maharashtra, India. 2. Consultant Radiologist, Department of Radiology, Nanavati Max Superspeciality Hospital, Mumbai, Maharashtra, India. 3. Fellow in Neuroradiology, Department of Radiology, Nanavati Max Superspeciality Hospital, Mumbai, Maharashtra, India. 4. Research Intern, Department of Radiology, Nanavati Max Superspeciality Hospital, Mumbai, Maharashtra, India. 5. Head, Department of Radiology, Nanavati Max Superspeciality Hospital, Mumbai, Maharashtra, India.

Abstract

Introduction: Dementia is a broad medical term that describes the progressive cognitive decline of brain function due to disease or impairment, resulting in interference with daily activities. Magnetic Resonance Imaging (MRI) is commonly used for the clinical diagnosis of dementia by identifying cerebral atrophy and structural alterations. Furthermore, Magnetic Resonance Spectroscopy (MRS) can detect biochemical abnormalities in dementia patients, which may be beneficial for early diagnosis and treatment.

Aim: To provide a comprehensive protocol and a guidance tool for radiologists to effectively diagnose dementia and its subtypes {Alzheimer’s Disease (AD), frontotemporal lobe dementia, etc.} based on radiological findings coupled with volumetry, spectroscopy, and Arterial Spin Labelling (ASL) findings.

Materials and Methods: A retrospective observational study was conducted in the Department of Radiology, Nanavati Max Superspeciality Hospital, Mumbai, Maharashtra, India, between June 2022 and June 2023. A total of 125 patients were analysed to observe the correlation between whole brain volume, Intracranial Volume (ICV) of different cortical regions, hippocampal atrophy, and MRS findings. Descriptive statistics were used, and results were expressed as means and standard deviations for continuous variables, and as frequencies and percentages for categorical variables.

Results: Out of 125 patients, 71 (56.8%) were males, and 54 (43.2%) were females, with ages ranging from 41 to 96 years. The majority presented with Mild Cognitive Impairment (MCI) 36 patients (28.8%) and Vascular Dementia (VaD) (18 patients - 14.4%). Some clinical features and imaging findings overlapped, resulting in some cases being a combination of different types of dementia. MRI, MRS, and Medial Temporal Atrophy (MTA) played a crucial role in allowing clinicians to perform a differential diagnosis.

Conclusion: The use of MR volumetry and spectroscopy aids in classifying the type of dementia, which, in turn, gives treating clinicians a better perspective for further treatment and its outcomes.

Alzheimer’s disease, Arterial spin labelling, Frontotemporal lobe dementia, Magnetic resonance imaging, Magnetic resonance spectroscopy

Dementia is a clinical syndrome characterised by cognitive impairment, often involving difficulties with patients’ thoughts, behaviour, mood, and memory, which worsen over time (1),(2),(3). It primarily affects the older population (>65 years), but approximately 9% of cases occur in younger individuals. According to the latest World Health Organisation (WHO) report, the number of patients suffering from dementia is expected to increase from 55 million to 139 million by 2050 (4),(5). Distinguishing between the characteristics of typical ageing and dementia is crucial for an accurate diagnosis. Individuals experiencing normal ageing maintain independent functioning and may only occasionally report memory problems. In contrast, patients with dementia depend on others for daily tasks and have significantly impaired memory, often unable to recall events when asked. Moreover, patients with dementia may exhibit socially unacceptable behaviours, whereas those aging normally typically retain social skills (5),(6),(7).

Alzheimer’s disease is the most common form of dementia, accounting for about 60-70% of cases worldwide (8). Other less common forms include VaD, Frontotemporal Dementia (FTD), MCI, and Dementia with Lewy Bodies (DLB). Accurate classification of different dementias remains a challenge for clinicians, as histopathologic examination alone cannot determine the underlying cause, often revealing mixed pathologies in the patient’s brain (1).

Neuroimaging techniques have been widely used for the diagnosis of dementia by ruling out cognitive impairment due to intracranial haemorrhage or space-occupying lesions. These techniques have become reliable for accurately diagnosing distinct types of dementia (9). Fluorodeoxyglucose (FDG) Positron Emission Tomography (PET) combined with MRI is beneficial for confirming findings and identifying abnormalities at an early stage (10).

The aim of the present study was to diagnose the different subtypes of dementia using volume analysis, MRS findings, morphologic factors, and ASL results. A differential diagnosis can enable the development of an effective treatment plan, thereby providing symptomatic relief to patients.

Material and Methods

This retrospective observational study included all adult patients (confirmed cases of dementia) between June 2022 and June 2023 at the Department of Radiology, Nanavati Max Superspeciality Hospital, Mumbai, Maharashtra, India.

Inclusion and Exclusion criteria: Data were collected retrospectively from 125 cognitively impaired participants. These individuals underwent imaging based on complaints such as memory loss, forgetfulness, gait disorders (imbalance, visual disturbances, etc.), and clinical/cognitive function assessment tests such as the finger tapping test, Mini-mental State Examination (MMSE) total score, and Montreal Cognitive Assessment (MoCA). Participants with memory complaints and a diagnosis of schizophrenia were excluded from the study.

Study Procedure

Patients were scanned using a GE 3.0 Tesla MRI Scanner, and the protocols used are illustrated in (Table/Fig 1). To analyse the variation in head size, the authors also calculated the total ICV for each participant using Neuroshield, an artificial intelligence-based clinical tool (11),(12). MRI has been used for decades to measure structural changes in the brain associated with dementia. It is an imaging modality that provides detailed anatomical information due to its excellent tissue contrast, spatial resolution, and levels of image resolution (1),(13). These studies assist in evaluation of neurodegenerative diseases by quantifying the volume of brain structures. Brain volume measurements are a known biomarkers for diagnosing different types of dementia (14). This volumetric analysis serves as a valuable clinical diagnostic support tool, aiding clinicians in calculating volumetric changes, tracking the clinical progression of the disease, and monitoring the percentage of ICV (11).

The MRS is a widely available, non invasive technique that measures biochemical changes in brain tissue (15). N-Acetyl-L-Aspartate (NAA) is a neuronal marker found in neurons, neuroglial precursors, and immature oligodendrocytes, while myo-inositol (mI) is a glial marker present in the frontal, parietal, temporal, and temporoparietal lobes (1). A decrease in the NAA/Creatinine (Cr) ratio correlates with the severity of dementia and cognitive decline (16). Klunk WE et al., demonstrated that a decrease in NAA is linked to neuronal loss (17).

MRI studies aid in identifying the type of dementia based on visual rating scales such as the MTA score, Fazekas scale, and Koedam score. The MTA and Koedam scores assist in identifying AD, while the Fazekas scale helps in determining VaD and normal aging (18). The MTA is a visual score performed on coronal T1-weighted MRI images based on hippocampal height and the width of the choroid fissure and temporal horn (19),(20). The Fazekas scale is 8used to quantify the amount of white matter hyperintensities on Fluid-attenuated Inversion Recovery (FLAIR) or T2-weighted images (21),(22). (Table/Fig 2) shows the Fazekas score scale. The Koedam score provides a scale for analysing parietal atrophy in sagittal, coronal, and axial planes and has proven to have a positive predictive value in the diagnosis of AD (23).

ACR appropriateness criteria: Imaging findings in structural MRI studies are non specific and have limited potential to differentiate between types of dementia. Advanced imaging techniques like functional neuroimaging with MRI and MRS can provide a better understanding of neurodegenerative disorders. The American College of Radiology Appropriateness Criteria (ACR) offers evidence-based guidelines for clinical conditions that are annually reviewed by an interdisciplinary expert panel. The guideline development and revision involve an extensive analysis of existing medical literature from peer-reviewed journals and the application of well-established methodologies (RAND/UCLA Appropriateness Method and Grading of Recommendations Assessment, Development, and Evaluation or GRADE) to rate the appropriateness of imaging and treatment procedures. In instances where the evidence is not unequivocal, expert opinion may be used to recommend imaging or treatment (24),(25).

The PET with FDG is most commonly used to identify neurodegenerative processes (10). FDG PET scans serve as an in-vivo clinical assessment of the autoradiographic technique. These scans are utilised to visualise cerebral glucose metabolism, which is instrumental in the differential diagnosis of dementia and the prediction of cognitive decline by analysing neural degeneration. Glucose is the primary metabolic substrate for energy production in the brain. FDG leverages deoxyglucose as a tracer to detect the exchange of glucose between plasma and the brain (26). Given that glucose is the sole source of energy for the brain, the loss of neurons in neurodegenerative brain diseases may result in decreased glucose consumption in specific brain regions (27). It is important to note that molecular imaging, such as FDG PET, was not utilised in the diagnosis of dementia types in our study.

Statistical Analysis

Descriptive statistics were used, and results were expressed as means and standard deviations for continuous variables, and as frequencies and percentages for categorical variables.

Results

(Table/Fig 3) summarises the statistical analysis of the dataset, detailing whole brain volume and hippocampal volume with respect to gender. The majority of patients were aged between 71-80 years, with 52 (41.6%) falling within this age range (Table/Fig 4).

In the present study, from a cohort of 125 patients, only 10 patients (8%) were diagnosed with AD. Brain volumetry performed on AD patients revealed a mild to moderate reduction in whole brain volume (mean=877.78 cc), and a moderate to severe reduction in hippocampal volumes (mean value of right hippocampal volume: 2.38 cc; mean value of left hippocampal volume: 2.21 cc). MRI scans showed white matter changes in these patients. Magnetic Resonance Spectroscopy (MRS) indicated a decrease in the N-acetylaspartate to creatinine (NAA/Cr) ratio and an increase in the myo-inositol to creatinine (mI/Cr) ratio, further supporting an AD pathology. (Table/Fig 5) presents a moderate reduction in the NAA/Cr ratio, indicative of neuronal loss, and an elevation in the mL/Cr ratio in the grey matter as observed in AD (in a 78-year-old female patient). (Table/Fig 6) shows a similar reduction in the NAA/Cr ratio within the white matter of the same patient.

The MCI cases accounted for 29% of the total sample (n=36), with an average age of approximately 67 years. Compared to AD patients, those with MCI exhibited a mild reduction in whole brain volume (mean=926.43 cc±107.26) and hippocampal volumes (mean value of right hippocampal volume=2.78 cc±0.422; mean value of left hippocampal volume=2.61 cc±0.382) with a MTA score ranging from 1 to 2.

The number of infarcts and the Fazekas score contributed to the diagnosis of VaD. A multi-infarct etiology was most common among the patients. In individuals with VaD, white matter NAA/Cr levels were lower when compared to AD patients. FTD cases (n=10) were diagnosed based on cerebral atrophy and the specific morphological features of the atrophy, particularly in the frontal and temporal lobes.

Additionally, Normal Pressure Hydrocephalus (NPH) was identified in the present cohort and was diagnosed based on cerebral atrophy, aqueductal or fourth ventricular flow void, the disproportion of ventricular dilatation to the degree of cortical atrophy, and the callosal angle.

(Table/Fig 7) illustrates the total number of patients diagnosed with different subtypes of dementia within the cohort.

While the cases of dementia discussed represent a broad classification, clinical features and imaging findings may sometimes overlap, resulting in various combinations of dementia. The most common mixed type found was AD with FTD, each with an overlay of multi-infarct status (VaD). MRS is a valuable tool in the diagnosis of different dementias. (Table/Fig 8) summarises the diagnostic factors used by clinicians for differential diagnosis. The NAA/Cr ratio in the white and grey matter, as shown in (Table/Fig 9), is frequently employed to classify the various subtypes of dementia. AD patients exhibited an NAA/Cr ratio below 1.4 in the grey matter spectrum, which is a strong indicator of neuronal loss. Alongside an increased mI/Cr ratio, this could potentially assist radiologists in confirming a diagnosis of AD, although further exploration is needed. In contrast, the NAA/Cr ratio in the grey matter of FTD patients was approximately 1.5. However, for diagnosing VaD, the NAA/Cr ratio in the white matter spectrum is considered.

Discussion

Alzheimer’s Disease (AD)

MRI is more sensitive than Computed Tomography (CT) scans to patterns of cortical atrophy and can better exclude other causes of dementia. It plays a crucial role in the diagnosis of AD by assessing volume changes in characteristic locations of the medial temporal lobe (including the hippocampus, entorhinal cortex, and perirhinal cortex) and the temporoparietal cortical region. (Table/Fig 10) shows MRI images of two patients diagnosed with AD. Medial temporal lobe atrophy can be determined directly or indirectly. Direct assessment depends on the volume loss of the hippocampus or the parahippocampus, while indirect assessment relies on the enlargement of the parahippocampal fissures. Direct assessment is comparatively more sensitive and specific and has been shown to predict the progression of MCI to dementia (28). The MTA scale is a commonly used visual assessment scale that has been clinically and neuropathologically validated. The MTA score has demonstrated significant ability to distinguish AD patients from those with vascular cognitive impairment or DLB (29). Clinically, AD is characterised by a cognitive decline in the form of episodic memory deficits. As the disease progresses, patients may also experience psychological and behavioural problems such as mood disorders, aphasia, visuospatial difficulties, executive dysfunction, and sleep disorders (28),(30). A decrease in the concentration of NAA in the frontal, parietal, temporal lobes, and the hippocampi is commonly seen in AD patients. Various neurodegenerative conditions feature reduced levels of NAA (15). Studies have shown that elevated glial metabolite, myo-inositol/creatine (mL/Cr) levels, and decreased NAA/Cr levels are associated with AD patients (17). Thus, the combination of high mI and low NAA in MRS assists in the early diagnosis of AD (18).

Vascular Dementia (VaD)

The VaD is the second most common type of dementia, and MR images of VaD patients show periventricular white matter hyperintensities (31). Multi-infarct dementia, a subtype of VaD, results from a series of small strokes that lead to permanent brain damage. These patients develop early symptoms such as mood changes and difficulties in understanding, concentrating, and planning. As the disease progresses, patients may show signs of confusion, difficulty following instructions, and inappropriate laughing or crying, along with loss of bladder or bowel control (32). Some symptoms of VaD are similar to those of AD (33),(34).

Binswanger’s disease, another subtype of VaD, arises due to arteriosclerosis and thromboembolism, affecting blood vessels that impact white matter and other subcortical structures. (Table/Fig 11) highlights an example of this rare type of VaD. Most patients experience progressive memory loss, urinary urgency, and an unsteady walking pattern (35),(36). Hyperintense signals on T2-weighted images and Fluid-attenuated Inversion Recovery (FLAIR) images aid in the distinct visualisation of both small and large vessel diseases and in identifying microhaemorrhages (37). Additionally, there is a variable appearance with multifocal asymmetrical abnormalities in affected brain regions (10).

Frontotemporal Dementia (FTD)

The FTD consists of three subtypes: behavioural variant, primary progressive aphasia, and motor disorder (31). MRI helps to determine the region and extent of atrophy, which enables the diagnosis of FTD. (Table/Fig 12) is an example of a patient who was diagnosed with FTD (38). The various FTD subtypes seen in literature are the behavioural and semantic variants of FTD, corticobasal syndrome, non fluent agrammatic variant of FTD, and FTD-associated motor neuron disease. Moreover, some patients with FTD present progressive supranuclear palsy in combination, primarily affecting movement. However, in the scope of the present study, only the behavioural variant of FTD and semantic variant of FTD will be discussed (39). (Table/Fig 13) describes the clinical and radiological findings of these two subtypes of FTD.

There is an overlap between the cortical regions affected by atrophy in FTD and AD. FTD shows relatively more atrophy in the frontal lobes, whereas AD shows more atrophy in the lateral, parietal, and occipital cortices (29). This type of dementia is diagnosed in patients within the age group of the 40s to early 60s. The behavioural variant of FTD typically shows asymmetrical frontal and temporal cortical atrophy (31). The semantic variant of FTD typically shows anterior temporal and ventral temporal association area atrophy, in turn, affecting the patient’s ability to use and understand language, whereas progressive non fluent aphasia affects the patient’s ability to speak. Some of the common symptoms of FTD are behavioural changes, impaired judgement, decreased self-awareness, and frequent mood changes, while physical symptoms involve tremors, poor coordination, muscle weakness, etc., (40).

Mild Cognitive Impairment (MCI)

The MCI is an early stage of loss in cognitive abilities, such as memory, language, or judgement. The symptoms of people diagnosed with MCI are not severe enough to interfere significantly with daily activities (41),(42). The level of memory deficit may remain stable in MCI patients for years, which is unlikely in the case of AD, where cognitive abilities decline gradually (43). A person diagnosed with MCI may be at risk of developing AD or other related dementias (42). The MTA score gives a strong indication of the progression of MCI to Alzheimer’s dementia (23). MRI is extensively used to diagnose MCI and to differentiate it from AD.

The hippocampal T2 prolongation serves as a unique marker to distinguish MCI from AD. Apart from this, MRI T2 signal decay has been used to measure white matter damage in patients with MCI, owing to its sensitivity to water content. Moreover, patients with MCI have a smaller entorhinal cortex and hippocampus compared to healthy subjects. According to many recent studies, hippocampal volume predicts the transformation of MCI to AD (44).

(Table/Fig 14) is a comparative illustration of the MRI scans of a control, an MCI patient, and an AD patient. In the present study, The MR volumetry showed a mild reduction in the whole brain and hippocampal volume. MRS exhibited a slight to moderate reduction in NAA, which is suggestive of neuronal loss. In most patients, a symmetric and maintained cerebral perfusion was identified. (Table/Fig 15) is an example of a female patient diagnosed with MCI.

Normal Pressure Hydrocephalus (NPH)

The NPH is one of the few dementias that is potentially reversible, and its timely diagnosis can result in the reversal of symptoms (45). This syndrome can occur in people of any age but is most common in the elderly and may result from head trauma, infection, subarachnoid haemorrhage, tumour, etc., (46). MRI is the best modality to image the morphological changes observed in NPH, and this can be further supported by CSF flow studies, MRS, and CT scans (47).

Radiographic features demonstrate periventricular hyperintensities, thinning of the corpus callosum, widening of temporal horns without hippocampal atrophy, enlarged Sylvian fissures, and basal cisterns (48). In NPH, changes in brain water content are observed as high T2-FLAIR signals on MRI scans and as periventricular hypodensity on CT scans (47). (Table/Fig 16) shows examples of two of the present study NPH patients.

Dementia with Lewy Bodies (DLB)

The DLB is a form of progressive dementia that affects an individual’s ability to reason, think, and process information, ultimately leading to a decline in independent function (49). People with DLB have characteristic features, including recurrent visual hallucinations, changes in attention and alertness, and confusion (50). DLB typically presents in older adults (50-70 years of age) and is usually sporadic (51). Structural studies measuring cortical thickness demonstrate that DLB patients show lower levels of volume loss in the amygdala, temporal lobe (including less pronounced grey matter loss in the temporal lobe), and hippocampus when compared to AD patients (52). Generally, DLB patients do not show deterioration in the formation of episodic memory, as observed in AD. However, impairments in visuospatial and attention tasks have been reported. The midbrain, particularly the substantia innominata and putamen, exhibit greater decline in volume and atrophy in DLB compared to AD (53).

Other Dementias

Apart from the types mentioned in this paper, other dementias or conditions that cause dementia-like symptoms include behavioural variant FTD, brain tumours causing dementia, Creutzfeldt-Jakob disease, dementia with Lewy bodies, Huntington’s disease, chronic traumatic encephalopathy, Human Immunodeficiency Virus (HIV)-associated dementia, and Parkinson-like disease. MRI is used to characterise these dementias at the structural level to guide the diagnosis of the patient and direct them towards appropriate management (10).

Limitation(s)

Although the present study of 125 patients underscores the importance of brain volumetry and MRS in differentiating types of dementia, a more detailed evaluation with a larger study group should be conducted to establish these methods as a gold standard for diagnosis. Molecular imaging was not included in the present study due to the low volume of available data. Compared to existing literature, the present study noted some differences, which may be attributed to demographic, environmental, and genetic factors.

Conclusion

Neuroimaging modalities are enabling healthcare practitioners to diagnose dementia using biomarkers. MR volumetry is yet another tool that assists in the differential diagnosis of dementia types.

DOI and Others

DOI: 10.7860/JCDR/2024/67220.19055

Date of Submission: Aug 28, 2023
Date of Peer Review: Nov 15, 2023
Date of Acceptance: Dec 21, 2023
Date of Publishing: Feb 01, 2024

AUTHOR DECLARATION:
• Financial or Other Competing Interests: None
• Was Ethics Committee Approval obtained for this study? No
• Was informed consent obtained from the subjects involved in the study? No
• For any images presented appropriate consent has been obtained from the subjects. No

PLAGIARISM CHECKING METHODS:
• Plagiarism X-checker: Sep 01, 2023
• Manual Googling: Oct 13, 2023
• iThenticate Software: Dec 18, 2023 (12%)

ETYMOLOGY: Author Origin

EMENDATIONS: 6