JP7177429B2 - Early diagnosis of dementia by brain MRI examination - Google Patents

Description

The present invention relates to early diagnosis of dementia and its precursors, mild cognitive impairment (MCI) and subjective cognitive impairment (SCI).

Improvements in the performance of magnetic resonance imaging (MRI) equipment have made it possible to image neurons at high resolution. Diffusion tensor imaging (DTI) is an image that emphasizes the directionality of diffusion of water molecules in vivo. Water molecules exhibit isotropic diffusion in the absence of nerve fibers, but anisotropic diffusion along nerve fibers in the presence of nerve fibers. In DTI, by capturing the diffusion anisotropy of water molecules, it is possible to visualize the running of nerves and evaluate nerve damage and the improvement of damaged nerves.

In DTI, diffusion anisotropy can be represented by an index called FA value (fractional anisotropy) value.

It is well known that the FA value accurately reflects changes in the microstructural structure of nerve tissue together with other diffusion tensor indices AD (axial diffusivity) and MD (mean diffusivity). However, the ability to diagnose dementia, particularly mild cognitive impairment (MCI) or subjective cognitive impairment (SCI), which has recently become a problem, is said to be inferior to MD and AD (Non-Patent Document 1). However, these papers neglected the ability to reflect the degree of diffusion of water molecules in unidirectional nerve fibers, which is the most important feature of the FA value, and fixed the above indices with multidirectional ability to be advantageous. The erroneous conclusions appear to be due to measurements being made only within the region of interest (ROI).

Nir et al, Neuroimage. Clinical 3, 180-195, 2013

The question-and-answer format that predominates in conventional dementia or predementia diagnosis is a potentially highly subjective approach. The method had problems with accuracy and stability. In addition, the full implementation of face-to-face testing required patient cooperation and required long testing times. It is not easy to impose such a test on dementia patients, and sometimes the treatment is continued aimlessly without obtaining the necessary information about the treatment.

The present invention is a method for early and scientifically accurate diagnosis and treatment, including cases of dementia and its precursors, such as mild cognitive impairment (MCI) and subjective cognitive impairment (SCI). for the purpose of providing

The present inventors, in order to accurately detect dementia and prodromal state of dementia, a reliable diagnostic method that can be detected even at an early stage, comprehend and evaluate the course of symptoms that progress relatively slowly, and then Intensive studies were conducted on the development of comprehensive and scientific measures, including the prediction of changes.

We detect dementia or prodementia in a subject by tracing nerve fibers from diffusion tensor images in the brain temporal hippocampal region of the subject and analyzing the tract area of the nerve fibers. In the method, five barriers are provided at regular intervals through which the extracted nerve fiber tracts pass when generating a diffusion tensor image extracting the nerve fiber tracts in the temporal lobe hippocampus of the subject, and By setting the FA (fractional anisotropy) value to 0.2 and measuring the tract area of nerve fibers in the obtained image, it is possible to objectively detect dementia or a prodromal state of dementia using the measured value as an index. This led to the completion of the present invention.

That is, the present invention is as follows.
[1] A method for detecting dementia or a prodementia state in a subject by tracing nerve fibers from a diffusion tensor image in the brain temporal lobe hippocampal region of the subject and analyzing the tract area of the nerve fibers. hand,
(i) generating diffusion tensor images extracting tracts of nerve fibers in the temporal hippocampus of the subject;
(ii) providing five barriers at regular intervals for passage of the nerve fiber tracts extracted in (i);
(iii) setting the FA (fractional anisotropy) value to 0.2 to obtain an image of extracted nerve fibers in the hippocampal region;
(iv) measuring the area of the nerve fiber tract from the number of pixels of the image obtained in (iii);
(v) detecting dementia or predementia based on measurements of nerve fiber tract area;
method including.
[2] The method of [1], wherein the prodementia state is Mild Cognitive Impairment (MCI) or Subjective Cognitive Impairment (SCI).
[3] The method of [1] or [2], wherein the hippocampal region is the hippocampus and the parahippocampal gyrus region.
[4] The method according to any one of [1] to [3], wherein the extracted images of nerve fibers in the hippocampal region are lateral view and axial view.
[5] Furthermore, a step of setting the FA (fractional anisotropy) value to 0.1 to obtain an image in which nerve fibers in the hippocampal region are extracted and measuring the area of the nerve fiber tract from the number of pixels in the image, Calculate the FA ratio obtained by dividing the tract area of the nerve fiber of FA0.2 by the tract area of the nerve fiber of FA0.1, and detect dementia or prodementia based on the FA ratio [1]-[ 4] either method.
[6] The method of any of [1] to [5], wherein in (ii), five consecutive regions of interest (ROI) are created on the diffusion tensor image.
[7] Compare the measured value of the nerve fiber tract area with the measured value of the nerve fiber tract area of a normal person. Any method of [1] to [6].
[8] In the subject, the method of [1] is periodically performed, the measured value of the tract area of the nerve fiber is obtained over time, the change in the measured value is examined, and if the measured value decreases, dementia is diagnosed. Or the method according to any one of [1] to [7], wherein the dementia or the prodementia state is evaluated to be improved when the dementia prodrome state worsens and the measured value increases.
[9] The method of any one of [1] to [8], comprising the step of setting an escape barrier to exclude nerve fibers that have erranted from other parts of the brain.

The method of the present invention can detect dementia and predementia in dementia patients and patients with predementia using the size of nerve fibers as an index, and also evaluate deterioration and improvement. be able to. Conventionally, detection of dementia and prodementia state had to rely on subjective question-and-answer method, but the method of the present invention is a scientifically, objectively and non-observationally safe method. can be detected.

FIG. 3 is a photograph showing a method of installing gates in the Nishijima five-gate method of the present invention. FIG. 2 is a diagram showing the results of analysis of hippocampal nerve fiber tracts in non-cognitive cases by the Nishijima five-barrier method of the present invention. FIG. 3 is a diagram showing the results of analysis of parahippocampal gyrus nerve fiber tracts in non-cognitive cases by the Nishijima five-barrier method of the present invention. FIG. 2 is a diagram showing the results of analysis of hippocampal nerve fiber tracts in dementia cases by the Nishijima five-barrier method of the present invention. FIG. 2 is a diagram showing the results of analysis of parahippocampal gyrus nerve fiber tracts in dementia cases by the Nishijima five-barrier method of the present invention. FIG. 10 is a diagram showing the relationship between the number of barriers and the amount of nerve fiber tract during a period of low tract amount before hydrogen treatment. FIG. 10 is a diagram showing the relationship between the barrier number and the amount of tracts that could be collected at the time when ADAS improved after the start of treatment. FIG. 10 is a diagram showing the relationship between the progress of hydrogen treatment and the barrier number; Contribution to ADAS change of each line showing tract amount over time in lateral view and axial view of left hippocampus at FA value = 0.1, 0.15 and 0.2 in dementia patients under treatment It is a result showing degree. FIG. 10 is a diagram showing the degree of contribution of each line segment with different FA values to the change when ADAS starts to deteriorate after the end of hydrogen therapy. FIG. 11 (ae) shows the analysis results of ADAS-cog scores in 11 dementia patients who were administered H ₂ gas. Figures 6a-e show changes in ADAS-cog scores over time for patients a-e. Straight _and dotted lines indicate the duration of _H2 and _Li2CO3 treatment, respectively. Patient b's _H2 treatment was stopped and then resumed, so the effect at the end of the first _H2 treatment was considered. FIG. 11 is a diagram (f to k) showing the analysis results of ADAS-cog scores in 11 dementia patients who were administered H ₂ gas. Figures 6fk show changes in ADAS-cog scores over time for patients fk. Straight _and dotted lines indicate the duration of _H2 and _Li2CO3 treatment, respectively. MRI visualization of hippocampus and parahippocampal gyrus. FIG. 5 shows five barriers set to volumetric sites so that neuronal tracts pass through the hippocampus or parahippocampal gyrus. FIG. 10 shows changes in hippocampal and parahippocampal gyrus volumes during treatment in patient d. FIG. 10 is a representative example of a diffusion tensor image for the inhalation effect of _H2 gas. FIG. 4 shows the results of semi-quantitative analysis of DTI tracts in hippocampal and parahippocampal gyrus regions. FIG. 11 shows normalized diffusion tensor image tract volumes of 11 AD patients.

The present invention will be described in detail below.

The present invention is a method for early detection of dementia and its precursors, mild cognitive impairment (MCI) and subjective cognitive impairment (SCI). Methods for detecting dementia include methods for detecting the risk of developing dementia in a subject and methods for evaluating or determining whether a subject may have dementia. By the method of the present invention, the degree of progression and prognosis of dementia can be evaluated and determined. In the present invention, the detection of dementia also includes evaluating the degree of progression and prognosis of dementia. Methods of the invention also include methods of obtaining ancillary data for assessing or determining whether a subject is likely to have dementia.

Dementia refers to the symptoms and conditions that occur due to the destruction of nerve cells in the brain due to disease. As dementia progresses, understanding and judgment decline, and social and daily life becomes difficult. . Dementia includes Alzheimer's disease (Alzheimer's disease,
AD), Repie body dementia, vascular dementia, etc.

The method of the present invention is carried out by imaging brain nerve fibers with an MRI apparatus, acquiring diffusion tensor imaging (DTI), and tracking nerve fibers from the diffusion tensor image.

In the method of the present invention, first, a brain image of a patient with dementia or a patient suspected of having dementia is imaged with an MRI apparatus to obtain diffusion tensor imaging (DTI). At that time, five software barriers (seed points) were set up by software in the brain temporal lobe hippocampal region, which is said to be specialized for cognitive function in the brain, and all the nerve fiber tracts of the hippocampus ( An image is created so that the bundle of rays passes through these five barriers. 5 barriers are fixed at regular intervals, such as 3 to 7 mm, preferably 5 mm, from a position 5 to 15 mm, such as 9 mm, from the tip of the hippocampus of the brain temporal lobe using brain MRI DTI software. Separate barriers (seed points) may be placed in the hippocampus and parahippocampal gyrus. Imaging is performed to selectively capture nerve fibers that pass through this barrier only. Images can be taken and created in two directions: a lateral view (horizontal direction) and an axial view (direction from the top of the head to the cervical spine). At this time, nerve tracts originating in other parts of the brain other than the hippocampus are excluded by this barrier, and if nerve fibers originating in other parts of the brain such as the brainstem rarely enter from outside the barrier, they are placed at the same time. A possible avoidance barrier excludes this (Fig. 1b).

This method, which uses five barriers, dramatically increases the amount of nerve fibers that can be captured compared to the conventional method using one or two barriers (Fig. 2). In addition, by placing in the same place each time, stable information on the performance of the nerve fibers themselves as well as the sizes of the tracts of these nerve fibers can be obtained. Since the quantity and quality of these tracts decline as cognitive function declines, this information can be used to make more scientific diagnoses of dementia and judgments of therapeutic effects.

It is well known that as dementia progresses, the number of these neural tracts decreases, leading to brain atrophy. In addition, the functions of various nerves within the nerve tract are diminished. Water molecules exhibit isotropic diffusion in the absence of nerve fibers, but anisotropic diffusion along nerve fibers in the presence of nerve fibers. In DTI, diffusion anisotropy can be represented by an index called FA value (fractional anisotropy value). The anisotropic diffusion rate changes depending on the size of the nerve tract, and the FA value is determined. In the present invention, nerve fibers are classified into five types by the FA value based on the diffusion speed of water molecules in such nerve fibers, and the size of each nerve tract can be measured. Studies have shown, for example, that nerve fibers with an FA value of 0.3 or higher are not found in the hippocampus of dementia patients by normal methods, and that changes in the amount of nerve fiber tracts with an FA value of 0.2 are necessary to improve the worsening of dementia symptoms. It was clarified that they occurred in parallel, and that the ratio of the amount of fibers with an FA value of 0.1 to the amount of fibers with an FA value of 0.2 was sensitive to changes in the size of the nerve bundle.

Brain MRI analysis of a patient who is actually undergoing treatment for dementia may be performed by the method exemplified below. In other words, five barriers established in the hippocampal region detect brain atrophy and deterioration of nerve tract function in the hippocampal region using the FA (functional anisotropy) value, which responds sensitively to changes in the brain microstructure structure, using three types of thresholds. (0.1, 0.15, and 0.2) Separately, the hippocampal nerve fibers that pass through only the five barriers are fully visualized to the periphery, and the area of the nerve fibers is calculated, and the tract measurement value is used for diagnosis. A tract measurement may also be referred to as tract area, amount of tract, or number of tracts. Accurate and objective diagnosis can then be made by comparing the tract measurements to the lower limit of normal (Table 1). The calculation results are obtained with the area of the nerve tract expressed in pixels. Areas of nerve fibers in the hippocampus and parahippocampal gyrus obtained for each FA value are compared with the lower limit of the normal value (e.g., the average value of 5 normal volunteers), and if it is below the normal value or lower than the normal value, it is considered abnormal, dementia. Alternatively, it can be evaluated as possibly being in a prodementia state. Here, the normal value refers to the mean value of tract measurements obtained from normal subjects without dementia. At this time, as an index, it can be evaluated using a sharp decrease in the high FA value tract measurement value (area) and a significant decrease in the FA ratio (especially, the tract measurement value with an FA value of 0.2/the tract measurement value with an FA value of 0.1). .

By periodically measuring with the method of the present invention and comparing the above index with subsequent test values, changes in the measured values over time can be investigated, and changes in symptoms can be quickly grasped. That is, when the measured value decreases, it can be evaluated that the dementia or the prodementia state has worsened, and when the measured value increases, it can be evaluated that the dementia or the prodementia state has improved. can. The interval between inspections is preferably 2-4 weeks. Also, the efficacy of therapeutic agents can be determined. By determining the efficacy of the therapeutic agent, a regimen for subsequent therapeutic agent administration can be formulated. Furthermore, it is also possible to predict the subsequent transition of symptoms. It should be noted that the amount of tracts with FA values of 0.25 and 0.3 decreased to the extent that they could not be identified by MRI, so no further imaging was required in routine clinical examinations.

In the method of the present invention, instead of measuring within a predetermined ROI (Region of interest), a fiber tracking method emphasizing excellent unidirectionality of the FA value (Chen et al. Magn Reson Imaging. 26:103-108.2007 Imaging) to create 5 regions of interest (ROI) for tractography, perform correct tractography, extract nerve fiber tracts three-dimensionally from DTI data, To construct. Here, tractography refers to tracking nerve fibers from a diffusion tensor image and generating a diffusion tensor tractography image from which the tracking trajectory is extracted. At this time, a coronal view should be used to create the ROI, and care should be taken not to miss nerve fibers passing over the surface so that the surface brain tissue is sufficiently included. The image rendered in this manner is processed with at least three types of FA values as described above. Here, the nerve tracking method refers to tracing the running of nerve fibers on a DTI image in which continuous running of nerve fibers is three-dimensionally constructed. In the method of the present invention, nerve fiber tracking is performed over the entire length of the nerve fiber using this nerve tracking method, and data at FA values of 0.1, 0.15 and 0.2 are used. The number of barriers to be installed is usually about 1 to 2, and the image error is proportional to the reciprocal of the square root of the number of barriers (seed number). On the other hand, since the image stability is proportional to the number of barriers, it is considered that the number of barriers should be large (Figs. 2 and 3). However, increasing the number of barriers naturally lengthens the measurement time, so it is necessary to consider the situation of each facility according to the purpose of use and the performance of the MRI equipment (Cheng et al. J Neurosci Methods.203, 264-272.2012). The imaging conditions can be determined by simultaneously considering the number of barriers and the FA value.

The method of the present invention is called the Nishijima five-gate method.

More than 400 types of drugs have already been developed for the treatment of dementia, and all of them have been determined to be ineffective in clinical trials. Therefore, the number of patients with dementia is increasing rapidly all over the world, and it is a heavy burden on society and economy. The reason why all the conventional dementia drug developments ended in failure is that they have a common point of taking up and targeting only a single etiology among the various etiologies of dementia. Recently, treatment with minute substances that simultaneously treats multiple targets has been proposed (Sun-Ho Han et al. JKMS 29:893-902, 2014), and the present inventors are the first in the world to use hydrogen as a very minute substance. A very good therapeutic effect has been confirmed in a clinical trial in which it was administered.

The data obtained by the Nishijima five-barrier method of the present invention almost reflect the transition of clinical symptoms, and further analysis of the data shows that the improvement effect of the clinical situation is accompanied by the improvement of the fine structure of the cranial nerve tissue. It can be shown that In the past, there were no cases in which clinical symptoms were reliably improved by drug administration. For the first time, a period of stable improvement of clinical symptoms was obtained with a treatment method using hydrogen, making it possible to use the Nishijima five-barrier method of the present invention.

Diagnosis of dementia is made by analysis of the above diffusion tensor image data, and the effect of the therapeutic drug is determined by comparing with the subsequent test values. If future deterioration is predicted, the current drug should be increased. . Hydrogen is mentioned as a chemical|medical agent.

Furthermore, by using the Nishijima five-barrier method of the present invention to analyze the brain MRI of a patient who is actually undergoing treatment for dementia, the state of the patient's hippocampus can be analyzed very scientifically. In the past, detection and judgment of dementia had to rely on medical interviews and were extremely subjective. However, the method of the present invention enables objective detection and determination, leading to an improvement in the level of dementia treatment.

The present invention will be specifically described by the following examples, but the invention is not limited by these examples.

Example 1 Nishijima-type 5-gate method 1. Implementation method of Nishijima five-barrier The Nishijima five-barrier method enables accurate and scientific understanding of the fine structure of the hippocampus, which is believed to be related to cognitive functions in the patient's brain. Details of the Nishijima five-gate method will be described below.

(1) First, the size of nerve tracts (bundles) passing through the hippocampal region of the brain is examined, and this is divided into high-functioning and low-functioning nerve tracts and measured. For this, images of the brain are taken with an MRI device, diffusion tensor imaging (DTI) is obtained, and brain MRI DTI software is used to extract images from a position 9 mm from the tip of the hippocampus to a distance of 5 mm. Five fixed barriers (seed points) were placed at intervals of the hippocampus and the parahippocampal gyrus separately (Figs. 1, 2, and 3). mm ² ) was measured and multiplied by 5 mm to calculate the volume (mm ³ ) from the barrier to a position 5 mm away. Data from these five volumes were summed to the hippocampal and parahippocampal gyrus volumes. This volume is required for the calculation of atrophy, which indicates complete destruction of the hippocampus. Even if there is a small cyst or ependymal invagination at this barrier, the site where the barrier is installed does not change.

(2) The software was then manipulated to capture and select nerve fibers passing only through this barrier. At this time, if there were clearly errant nerve fibers from the brainstem, etc., these were excluded by establishing an avoidance barrier (exclusion barrier) (Fig. 1b). Exclusion is done by observing the nerve tract passing through the nerve fiber group, without considering the complex site of origin and the wide range of reach of the nerve fiber group, which neurologists are concerned with. Images were taken and created in two directions: a lateral view (horizontal view) and an axial view (direction from the top of the head to the cervical spine).

(3) Next, using the DTI software attached to MRI, the FA (fractional anisotropy) threshold was set to 0.1, 0.15, and 0.2, and the nerve fibers of the hippocampus and parahippocampal gyrus were sufficiently visualized to the periphery by the nerve tracking method. .

(4) Images were taken and created in two directions: a lateral view (horizontal view) and an axial view (direction from the top of the head to the cervical spine).

(5) These images were first carefully observed with T1- and T2-weighted images to confirm that there was no rotation or left-right displacement. On T1-weighted images, adipose tissue appears white, and water, liquid components, and cysts appear black. On T2-weighted images, water, liquid components, and cysts appear white instead of adipose tissue peaks.

(6) After the image was completed, the image was binarized off-line using ImageJ software and the area was measured. At this time, the crop function of ImageJ is used to limit the calculation range to the extent that it touches the nerve fibers so that the measurement value includes only the nerve tract so that unnecessary parts due to noise outside the nerve fibers will not be included in the calculated values. Noted.

(7) Calculation results were obtained with the area of the nerve tract expressed in pixels. This area value can be converted to mm ² if desired (5.52 pixels = 1 mm ² ). Next, the areas of the hippocampal and parahippocampal gyrus nerve fibers obtained for each FA value were compared with the lower limit of the normal value (average value of 5 normal volunteers). .

Table 1 shows the number of pixels (tract area) in the hippocampus (h) on the left (l) and right (r) sides for each site and FA value, using data from non-dementia volunteers.

This is used to determine whether the tract area of a clinical case of dementia is abnormally low and pathological.

The numbers in the left column indicate FA values. 1 and 2 FA value = 0.1, 3 and 4 FA value = 0.15, 5 and 6 FA value = 0.2, 7 and 8 FA value = 0.25, 9 and 10 FA value = 0.3.

Of the numbers in the left column, odd numbers indicate lateral views, and even numbers indicate axial views. For example, lh3 indicates the tract area (number of pixels) with a FA value of 0.15 in the lateral view of the left hippocampus. Note that 5.52 pixels correspond to 1 mm ² .

In Table 1, "SD" indicates standard deviation, and "average-2SD" indicates the value obtained by subtracting 2SD from the average area value (number of pixels). "Healthy person minimum-maximum value" indicates the range of volunteer data instead of the FA value where the mean-2SD became 0 or less.

For the diagnosis of dementia, it is considered important that these tract areas be below the mean -2 SD or the lowest value in this range.

(8) Next, the FA ratio was calculated. That is, the measured value (area) of the nerve tract at FA0.2 at the same site (hippocampus, lateral view, and axial view in the parahippocampal gyrus) was divided by the measured value at FA0.1.

(9) Perform the same test 2 to 4 weeks later, and if there is even a slight change, it can be identified by this 5-barrier method. Therefore, the direction of future symptom changes can be predicted before clinical symptoms change.

FIG. 1 shows a photograph of a gate installation method in the Nishijima five-gate method of the present invention. Figures 1a and 1c show photographs taken at an interval of 6 months, showing 5 barriers and FA0.2 fascicles passing through the barriers. In the figure, "seed_1" to "seed_5" indicate five barriers. FIG. 1b shows the avoidance barrier being used to clear errant fibers that ascend from the brainstem and slip through the posterior barrier to erode. In the figure, labels such as "avoid_1" indicate avoidance barriers, and labels "seed_1" to "seed_5" indicate five barriers. As shown in Fig. 1d, five regions of interest (ROI) for tractography are created every 5 mm from the 9 mm tip of the hippocampal body. Use a coronal view to create the ROI, taking care not to miss any nerve fibers passing over the surface to include enough of the superficial brain tissue. In demented cases, the atrophy around the hippocampal sulcus is usually strong (the case in this figure is a non-demented case), and it is easy to understand if the upper part is the hippocampus and the lower part is the parahippocampal gyrus. In the drawing, the portion surrounded by the upper frame is the region of interest 1 (ROI1), and the portion surrounded by the lower frame is the region of interest 2 (ROI2). The left line segment indicates the scale.

Figures 2-1 and 2-2 show the results of analysis of nerve fiber tracts in cognitively normal cases by the Nishijima five-barrier method of the present invention. Figure 2-1 shows the results of analysis of the hippocampus, and Figure 2-2 shows the results of analysis of the parahippocampal gyrus. It shows the difference in the tract area that can be captured when the number of barriers is set to 1, 2, 3, and 5. Here, when using one barrier 2 (top row) (1 Seed), when using two barriers 2 and 4 (2 seeds) (second row from the top), and 2nd, 4th, and 5th shows the tract area when using 3 barriers (3 seeds) (third row from the top) and when 5 barriers (5 seeds) (bottom row). In the figure, the 1st and 3rd from the left are lateral images, and the 2nd and 4th are axial images. In each panel, the 1st and 2nd left panels show the tracts when FA=0.1, and the 3rd and 4th panels from the left show the tracts when FA=0.2. The area of both the hippocampus and the parahippocampal gyrus is the widest in the case of 5 barriers. A large number of barriers is more advantageous for capturing fibers with high FA values, and the 5-barrier method is necessary to capture fibers in the hippocampus with FA = 0.2, which decreases sharply in dementia.

Figures 3-1 and 3-2 show the results of analysis of nerve fiber tracts in demented cases by the Nishijima five-barrier method of the present invention. Figure 3-1 shows the results of analysis of the hippocampus, and Figure 3-2 shows the results of analysis of the parahippocampal gyrus. 2-1 and 2-2 show findings taken with 1, 2, 3 and 5 barrier imaging techniques placed on the same site in the same manner. The imaging method and arrangement of the images in the figure are the same as in FIGS. 2-1 and 2-2. As shown in Figures 3-1 and 3-2, in dementia, tracts with an FA value of 0.2 are particularly reduced in the hippocampus, and with 1 and 2 function numbers, no data can be obtained and a warning is given that images cannot be displayed. I got out. With 5 barriers, an image was finally obtained that was sufficient to measure the area, confirming the necessity of the 5-gateway method.

2. Reason and justification for setting the number of barriers to 5 Naturally, the amount of nerve fiber bundles (tracts) that can be collected (the number of pixels on the vertical axis) increases when the number of barriers is increased.

FIG. 4-1 shows the relationship between the barrier number and the tract amount during the period of low tract amount before hydrogen treatment, showing data at FA value=0.1 and FA value=0.2. The horizontal axis indicates the number of barriers, and the vertical axis indicates the amount of tracts. In the figure, ♦ and ▪ indicate axial view data, and ▲ and x indicate lateral view data. Up to the third barrier, the message "cannot be measured at FA = 0.2" frequently appears, indicating that measurement is not possible.

Figure 4-2 shows that the relationship between the barrier number and the amount of tracts that could be collected improved when ADAS improved after the start of treatment. Similar to Fig. 4-1, data at FA value = 0.1 and FA value = 0.2 are shown. The horizontal axis indicates the number of barriers, and the vertical axis indicates the amount of tracts. In the figure, ♦ and ▪ indicate axial view data, and ▲ and x indicate lateral view data. At a barrier number of 2 or less, the capture of the FA=0.2 tract is insufficient. It shows that a barrier number of 5 is necessary in consideration of individual differences.

The data for the number of barriers of 4 is omitted, but the number of barriers is less than that of 5. When the number of barriers is 6, if the hippocampus is atrophied, the sixth barrier protrudes out of the hippocampus, and the posterior part of the amygdaloid nucleus may slip into the first barrier. Moreover, it is not suitable because the operation becomes complicated.

Figure 4-3 shows the relationship between the progress of hydrogen treatment and the barrier number (FA=0.2). The vertical axis indicates the amount of tracts, and the horizontal axis indicates the following.
1: Checkpoint 3 only 2: Checkpoints 2 and 4
3: Gates 2, 4, 5
4: Gates 1, 2, 3, 4
5: Gates 2, 3, 4, 5
6: Gates 1, 2, 3, 4, 5

Series 2 shows data at the start of hydrogen treatment, series 4 shows data at 2 months after hydrogen treatment, and series 6 shows data at 4 months after hydrogen treatment.

Overall, the treatment increases nerve tracts across the barrier. Also, increasing the number of functions increases the number of observable neural tracts.

However, nerve tracts crossing barriers 2, 4, and 5 were reduced as observed in series 6, and the number of nerve tracts crossing barrier 4 was significantly reduced in series 4, as evident when comparing 4 and 5. Nerve bundles passing through barriers 1, 2, 3, and 4 are reduced, and certain observations cannot be made. That is, it can be seen that setting the number of functions to 5 is necessary for observing a reliable tract.

3. Reasons and legitimacy for setting FA = 0.2 In patients with dementia under treatment, tract volume over time in lateral and axial views of the left hippocampus at FA values of 0.1, 0.15, and 0.2 We estimated the degree of influence of each line segment showing the ADAS change, and calculated the degree of contribution.

The results are shown in Figure 5-1. In Fig. 5-1, "lh" in lh1 to lh3 indicates left hippocampal data. Numbers indicate FA values, where 1 and 2 indicate FA value = 0.1, 3 and 4 indicate FA value = 0.15, and 5 and 6 indicate FA value = 0.2. Among the numbers, odd numbers indicate a lateral view, and even numbers indicate an axial view. For example, lh3 indicates the data of FA value 0.15 in the left hippocampal lateral view. The vertical axis in FIG. 5-1 indicates the contribution (%) of each line segment involved in ADAS improvement, and the horizontal axis indicates the date of observation. The example in Fig. 5-1 shows the ADAS-cog as an index, and the ADAS-cog values on each date are shown in the figure. A decrease in ADAS-cog values indicates an improvement in cognitive function. At the time of ADAS improvement (Fig. 5-1, 7.19), the left hippocampal FA = 0.2 tracts (lh6 and lh5) contributed 48 to 62%. That is, when FA=0.2, a correlation was observed between the decrease in ADAS-cog (improvement in cognitive function) and the increase in tracts passing through the five barriers. That is, it was shown that it is necessary to set FA=0.2 to observe improvement in cognitive function.

Fig. 5-2 shows the contribution of each line segment with different FA values to the change when ADAS started to deteriorate after hydrogen therapy. The line segment of FA=0.2 has a very high degree of contribution even at the time of deterioration. On the other hand, the FA=0.15 line has much smaller contributions (lh3 and lh4).

From the above results, it was clarified that the data of FA = 0.2 showed a correlation with cognitive function.

Example 2 Evaluation of therapeutic effect of hydrogen gas on dementia by Nishijima five-barrier method Patient selection Patient selection criteria were the following three.
(1) ADAS-cog score using the formula [70-(MMSEx2.33)] (J. Caro et al., BMC Neurol 2002;2:6.) or MMSE (Mini-Mental State Examination) converted ADAS -cog score greater than 10 with worsening score.
(2) A diagnosis of Alzheimer's dementia has been made and treatment with an acetylcholinesterase inhibitor or an N-methyl-D-aspartate (NMDA) receptor antagonist has been tried, but the ADAS-cog score has worsened is doing.
(3) no significant airway disease such as COPD (chronic obstructive pulmonary disease), pneumonia, bronchitis, or asthma that may prevent adequate inhalation of _H2 ;

Prior to the study, patients were given 400 mg/day lithium carbonate (Li ₂ CO ₃ ) for 1 week to ensure renal and liver function were within normal limits.

Finally, 11 patients were selected and treated with _H2 gas.

Treatment The dose of lithium carbonate was reduced because higher blood lithium levels were associated with higher morbidity (WS Waring, Toxicol Rev 2006;25:221-230.). Patients in the treatment group were orally administered lithium carbonate (Li ₂ CO ₃ ) 200 mg tablets (Mitsubishi Tanabe Pharma Corporation) twice daily, and 3% H ₂ inhalation was performed twice daily. The duration of the study was 6 months. Gases containing _H2 were produced using a portable _H2 generator (Nishijima/Enor gas hydrogen generator) to generate _H2 gas (3%) containing 21% oxygen (H. Ono et al., J Stroke Cerebrovasc Dis 2017;26:2587-2594.). Blood levels were confirmed to increase, as previously reported (H. Ono et al., J Stroke Cerebrovasc Dis 2017;26:2587-2594.). Lithium carbonate is used as a therapeutic agent for manic-depressive disorder (bipolar disorder), and it was used in this example because it was reported that few patients taking lithium carbonate developed dementia.

Evaluation Patients were evaluated in a blinded fashion using the ADAS-cog score.
ADAS-cog tests were performed every 1-2 months. Efficacy of treatment on ADAS-cog scores was determined by comparing ADAS-cog scores at baseline and 2-3 trials during treatment.

Measurement of hippocampal and parahippocampal gyrus volumes
Visualization of the hippocampus and parahippocampal gyrus was performed by MRI (Fig. 7-1). FIG. 7-1 is a view similar to FIG. 1d. Five coronal slices were taken 5 mm apart from 9 mm posterior to the anterior end of the hippocampus. The area of each coronal section was calculated using DICOM ROI software. Areas were automatically recorded by the software in ^mm2 , and the volume of the hippocampus and the parahippocampal gyrus region, including the subiculum, entorhinal gyrus and parahippocampal gyrus, was measured by dividing the area in ^mm2 by 5 mm (between control sections). distance) and converted to ^mm3 units (Fig. 7-2). Briefly, volume measurements were taken over most of the hippocampus from 9 mm from the tip to 34 mm into the body. All processing and calculations were performed blindly by a radiologist.

Measurement of hippocampal and parahippocampal tract tract areas by diffusion tensor imaging (DTI). A volumetric site was set across the entire paragyrus (Fig. 7-2). Briefly, five barriers were set at volumetric sites such that neuronal bundles passed through the entire hippocampus or the parahippocampal gyrus. In FIG. 7-2, it can be seen that nerve fibers pass through five barriers (seeds).

Digital tractography imaging was performed by GRAPPA technology using Neuro3D. Blue gates indicate five barriers for selection of passed neuronal bundles. Yellow indicates neuron bundles. To reduce examination time, we performed digital tract imaging (DTI) using Neuro3D with GRAPPA technology. DTI was obtained at functional anisotropy (FA, diffusion rate of water molecules in nerve fibers) values of 0.10, 0.15, and 0.2. The size of the tract was calculated from the number of pixels in the tract image and the number of pixels within the tract was calculated using Image J software.

Treatment Outcomes ADAS-cog scores in most patients improved with _H2 therapy.
Figures 6a-e and 6f-k show profiles of changes in ADAS-cog scores for each patient with combined _H2 therapy and lithium therapy. a through k show results for each of the 11 patients. In f and h, scores deteriorated immediately after combination therapy ended. On the other hand, in e and i, improvement by H ₂ was observed even after stopping lithium addition. In a, b, c, d, g and h, scores worsened (continued to increase) after _H2 treatment was discontinued. Therefore, the main effect of combination therapy was found to be due to _H2 inhalation only.

At the end of the _H2 treatment period, the scores of 7 patients (c, d, f, g, h, i and j) had improved (decreased) from the start time. Scores at the end of _H2 treatment did not recover to initial levels, but in 3 patients (a, b, e) scores worsened once but improved slightly later. One patient (k) was non-responder. Table 2 shows the analysis results of the ADAS-cog score at the start of hydrogen gas administration, after exacerbation, and after the end of hydrogen gas administration in 11 dementia patients who were administered H ₂ gas. As shown in Table 2, including this non-responder, the mean values were -3.4 (from initial score) and -8.5 (from worst score) at the end of _H2 treatment.

Five controls showed no improvement with treatment. In addition, although about 200 patients were seen at the hospital where the study was conducted, none of the patients with Alzheimer's dementia improved in ADAS-cog score in 6 months without _H2 treatment. Thus, _H2 treatment showed significant improvement in ADAS-cog scores.

The volumes of hippocampal and parahippocampal regions did not change during treatment.
Hippocampal and parahippocampal gyrus volumes were measured at least three times during the treatment period as described in Methods. The follow-up period for patient d during treatment lasted 8 months. ADAS-cog scores changed significantly (Fig. 6d). Figure 7-3 shows changes in hippocampal and parahippocampal gyrus volumes during treatment in patient d. Hippocampal size did not increase during the 6-month treatment period.

Hydrogen treatment increased DTI tracts, and increases correlated with ADAS-cog scores.
DTI is useful to assess neuronal integrity as a useful method to assess AD (TC Chua et al. Curr Opin Neurol 2008;21:83-92.: IK Amlien, Neuroscience 2014;276:206- 215.). In this example, we focused on neuronal tracts that passed through the entire hippocampus or the parahippocampal gyrus. For this purpose, we fixed five barriers through which neuronal tracts pass (Fig. 7-2). Images were acquired at different FA values of 0.10, 0.15, and 0.2 for the entire hippocampal or parahippocampal gyrus region to assess the diffusivity of water along neuronal tracts, which represent neuronal integrity (LG Vasconcels et al. al., Dement Neuropsychol 2009;3:268-274.: B. Zhang et al., CNS Neurosci Ther 2014;20:3-9.).

Figures 8-1 and 8-2 show the effect of _H2 gas inhalation on MRI of the hippocampal (Figure 8-1a) and parahippocampal gyrus (Figure 8-1b) regions of patient d at FA = 0.1, 0.15, or 0.2. FIG. 4 is a diagram showing a representative example of a diffusion tensor image for . Neuronal tracts passing through the hippocampus or parahippocampal region at five barriers were selected as described in Methods. Figures 8-2a and 8-2b show the results of semi-quantitative analysis of DTI tracts in hippocampal and parahippocampal gyrus regions, respectively. Square and triangle symbols indicate images viewed from the left and front, respectively. Straight _and dotted lines indicate the duration of _H2 and _Li2CO3 treatment, respectively. Figures 8-1a and _8-1b show lateral 0.1, Representative DTIs at FA values of 0.15 and 0.2 are shown. Bundles with high FA values (0.2) were the most sensitive to change. DTI tracts were unchanged in the parahippocampal gyrus region by _H2 and lithium treatment. FA=0.15 and 0.2 DTI bundles in the hippocampal region appeared to increase after _H2 and lithium treatment. The therapeutic effect was strongest at FA=0.2 (Fig. 8-1a). Furthermore, the DTI flux of tracts with FA=0.15 and 0.2 decreased when treatment with _H2 was discontinued (Fig. 6a). Thus, at least in this case, the increase in DTI flux was not due to lithium treatment.

This profile was in good agreement with the ADAS-cog scores shown in Figure 6d. On the other hand, DTI tracts in the parahippocampal gyrus did not change (Fig. 8-1b). Figures 8-2a,b show the results of semi-quantitative analysis of DTI tracts at FA = 0.1, 0.15 and 0.2. These data indicate that representative hippocampal DTIs are correlated with ADAS-cog scores, especially at high FA values.

FIG. 9 shows normalized diffusion tensor image tracts of 11 AD patients, showing the results of assessing neural tube bundle areas that crossed five fixed barriers in the hippocampus or parahippocampal gyrus area. Figure 9 shows the mean and standard error of DTI for 11 patients. Figure 9a shows the number of pixels in the tract of FA = 0.2 corresponding neurons in the hippocampal and parahippocampal gyrus regions between the worst and the end of _H2 treatment by the number of pixels in the FA = 0.1 fascicle. Shows a comparison of normalized values. ** indicates p < 0.01 for hippocampal regions. No difference was observed between initiation and termination of treatment in the parahippocampal area. Figure 9b shows a comparison of hippocampal DTI tracts normalized by parahippocampal DTI at the start or end of treatment. *** indicates p < 0.001 at FA = 0.2. At FA=0.1, there was no difference between worst and exit from treatment. In FIG. 9, error bars indicate standard errors.

Thus, we suggest that changes in DTI reflect the effects of _H2 treatment on hippocampal neuronal integrity.

consideration
Combined therapy with _H2 and lithium improved ADAS-cog scores in 10 of 11 patients with Alzheimer's disease. The ADAS-cog score (WG Rosen et al., Am J Psychiatry 1984;141:1356-1364.) has become the most widely used measure of general cognitive function in AD clinical trials, and it , and orientation. There were three possibilities in the combination therapy of the present invention. First, the combination of _H2 and lithium may have been essential for synergy. The second is that mainly lithium is effective, and _H2 may have mitigated the side effects of lithium or augmented the effectiveness of lithium. Third, _H2 alone may have been effective in treatment.

The results of this example show that the therapeutic effect was due to _H2 inhalation alone, as scores continued to improve with _H2 even in the absence of lithium (in e and i). In contrast, scores for a, b, c, d, g and h worsened (continued to increase) when _H2 treatment was discontinued. That is, when not combined with H ₂ , lithium alone did not improve cognitive function, and only H ₂ improved cognitive function.

These improvements with _H2 inhalation are significant when compared with the effects of donepezil, which is used to temporarily improve dementia symptoms in Alzheimer's disease. Donepezil reduced the ADAS-cog score by 3 points after 6 weeks (-3). However, after 6 months, scores in the donepezil group had returned to baseline values, although the placebo group had deteriorated (KR Krishnan et al., Am J Psychiatry 2003;160:2003-2011.). In this example, _H2 treatment resulted in a mean score change of -3.4 for subjects including non-responders from early stage. Furthermore, the mean score change from the worst score to the end of _H2 treatment was -8.5. That is, the effect of _H2 gas was observed in a high percentage of 10 out of 11 patients, and the therapeutic effect was more significant than that of donepezil.

Furthermore, the patient of this example lost the therapeutic effect of an acetylcholinesterase inhibitor and/or an N-methyl-D-aspartic acid (NMDA) receptor antagonist, which are pharmaceutical approved as therapeutic agents for Alzheimer's disease. After that, hydrogen gas inhalation treatment was started. Therefore, it is clear that the ameliorating effect of hydrogen gas inhalation was exerted through a mechanism of action different from that of these drugs approved under the Pharmaceutical Affairs Law. INDUSTRIAL APPLICABILITY The present invention can be applied even after the therapeutic effects of conventional drugs approved under the Pharmaceutical Affairs Law have ceased to be recognized, and is extremely useful.

Furthermore, DTI showed significant recovery of neuronal bundle size, especially at high FA values. Cerebral white matter fascicles are highly oriented and arranged in a compact manner. The diffusivity of water along the bundle direction is much higher compared to other directions. The difference between the diffusivity along and across the fascicle increases with axonal density. FA values reflect the density of axons within the fascicle, with lower values corresponding to lower axon densities. The DTI findings were clearly consistent with ADAS-cog score clinical improvement in AD patients.

The method of the invention allows objective detection of dementia or predementia.

Claims (9)

Ancillary data for detecting dementia or predementia in a subject by tracing nerve fibers from diffusion tensor images in the brain temporal lobe hippocampal region of the subject and analyzing the tract area of the nerve fibers. A method for obtaining ,
(i) generating diffusion tensor images extracting tracts of nerve fibers in the temporal hippocampus of the subject;
(ii) providing five barriers at regular intervals for passage of the nerve fiber tracts extracted in (i);
(iii) setting the FA (fractional anisotropy) value to 0.2 to obtain an image of extracted nerve fibers in the hippocampal region;
(iv) measuring the area of the nerve fiber tract from the number of pixels of the image obtained in (iii);
(v) obtaining ancillary data for detecting dementia or predementia based on measurements of nerve fiber tract area;
method including. 2. The method of claim 1, wherein the prodementia state is Mild Cognitive Impairment (MCI) or Subject Cognitive Impairment (SCI). 3. The method of claim 1 or 2, wherein the hippocampal regions are hippocampal and parahippocampal gyrus regions. The method according to any one of claims 1 to 3, wherein the extracted images of nerve fibers in the hippocampal region are a lateral view and an axial view. Furthermore, an FA (fractional anisotropy) value is set to 0.1 to obtain an image in which nerve fibers in the hippocampal region are extracted, and the area of the nerve fiber tract is measured from the number of pixels in the image, FA0.2 Calculate the FA ratio, which is the tract area of the nerve fiber of FA0. The method according to any one of claims 1-4. A method according to any one of claims 1 to 5, wherein in (ii) five successive regions of interest (ROI) are created on the diffusion tensor image. Compare the measured nerve fiber tract area with the measured nerve fiber tract area of a normal person, and assess possible dementia or prodementia if the measured value is less than or equal to that of a normal person. A method according to any one of claims 1 to 6, wherein ancillary data for is obtained . In a subject, the method of claim 1 is periodically performed, the measured value of the tract area of the nerve fiber is obtained over time, the change in the measured value is examined, and if the measured value decreases, dementia or Acquiring auxiliary data for evaluating that dementia or prodementia state has improved when the dementia prodromal state worsens and the measured value increases, claim 1 to 7 according to any one of Method. 9. The method of any one of claims 1-8, comprising establishing an escape barrier to exclude errant nerve fibers from other parts of the brain.

Images