Micro-MRI improves the accuracy of clinical diagnosis in cerebral small vessel disease

micro-MRI, small vessel disease, cerebral microbleeds, cerebral superficial siderosis, cerebral amyloid angiopathy

Introduction

In many neurological diseases, an autopsy is required for definite diagnosis. However, only limited sections of autopsied tissue can be analysed through histopathological examinations. Cerebral small vessel disease is a representative example of the difficulty in making a definitive diagnosis of the exact lesions detected by in vivo radiological examination. Clinicopathological and clinicoradiological studies have indicated that treatment and prevention for small vessel disease would be beneficial to decelerate cognitive decline.^1–5 For early intervention in small vessel disease, improvement of the disease’s clinical diagnosis is crucial.

Lobar cerebral microbleeds (CMBs), cortical microinfarcts (CMIs), cortical superficial siderosis (cSS) and white matter hyperintensity are key MRI markers for cerebral amyloid angiopathy, which is a form of small vessel disease.¹^,⁶ These in vivo MRI findings are not always caused by cerebral amyloid angiopathy, and the exact underlying pathology remains unclear. To solve this problem, it is important to investigate small vessel disease using in vivo MRI.⁷^,⁸

In most radiological–pathological comparative studies, MRI is used to examine in vivo human tissues.⁹ In the clinical setting, since the MRI machines are dedicated to patient use during the day, the time in which ex vivo brains can be scanned is restricted to after hours (night time).¹⁰^,¹¹ In addition, it is often difficult to obtain thin brain slices (thinner than 10 μm) that include in vivo MRI-detected small vessel disease for pathological examination. Therefore, the exact pathologies of in vivo MRI-detected small vessel disease lesions have been seldom reported.¹² Micro-MRI has been designed for experimental purposes, the performance time is short and because of the microcoils, its spatial resolution is much superior to that of regular 3-T MRI for the whole brain. Few micro-MRI studies have examined human tissue.^13–15In vivo MRI-detected small vessel disease can be efficiently assessed by neuropathological examinations using micro-MRI as a ‘bridge’. In this study, we describe an efficient method of in vivo radiological−ex vivo pathological comparison and novel comparative findings of small vessel disease in patients with in vivo MRI-defined small vessel disease.

Materials and methods

Subjects

We selected patients who had available in vivo MRI scans and findings of small vessel disease detected using in vivo MRI. Only patients whose nearest family provided informed consent allowing an autopsy for diagnostic and scientific purposes were included in this study. Four patients with in vivo MRI-defined small vessel disease were subjected to a radiological−pathological study (see details in Supplementary material and Supplementary Figs 1–3). This study was approved by the Ethics Committee of Mie University Hospital (approval number 2498) and conformed to the tenets of the Declaration of Helsinki.

In vivo 3-T MRI protocol

In vivo 3-T MRI was performed using a 3-T MR scanner (Ingenia, Philips Health Care, Best, the Netherlands), using a 32-channel phased-array head coil. Diagnostic imaging included 3D fluid attenuated inversion recovery (FLAIR), 3D T₁-weighted imaging (WI), diffusion WI and susceptibility WI (SWI). 3D FLAIR parameters included a 250 mm field of view (FOV) with a 256 × 184 matrix, a section thickness of 1 mm with a 0.57 mm overlap, no parallel imaging, a repetition time (TR)/echo time (TE) ratio of 6000/378 (shortest), and an inversion time (TI) of 2000 ms; the number of signals acquired was two and the acquisition time was 4 min 42 s. SWI parameters included a FOV of 230 mm with a 320 × 251 matrix, a section thickness of 0.8 mm over contiguous slices, a minIP of 5 mm, a TR/TE ratio of 22/11.5 (in-phase), 37 (shifted); number of signals acquired, 1; flip angle 20°; and an acquisition time of 6 min 42 s. 3D T1WI parameters included a FOV of 260 mm; matrix, 288 × 288; section thickness, 0.9 mm; TR (ms)/TE (ms) ratio, 7.6 (shortest); TE (shortest), 3.6 ms; flip angle, 10°; and an acquisition time of 4 min 42 s.

Ex vivo MRI protocol

Micro-MRI scanning was performed using a 3-T MR scanner (LabPET/MRI 3016, TriFoil Imaging, Inc., Northridge, CA, USA), using a rat body coil. The formalin-fixed brain was cut to fit the plastic box for setting in the coil in advance. The brain block was placed in the plastic box filled with salt-free water at 35°C. We employed an MR protocol with T1WI, T2WI, T2*WI and FLAIR. Details of the T2WI protocol are as follows: FOV, 80 × 80 mm; slice thickness/gap, 1.0/0.5 mm; acquisition matrix, 256 × 256; pixel size, 0.3125 × 0.3125 mm; TR, 4800 ms; TE, 68 ms; flip angle, 90°; number of averages, 2 and acquisition duration, 310 s. Details of the T2*WI protocol are as follows: FOV, 80 × 80 mm; slice thickness/gap, 1.0/0.5 mm; acquisition matrix, 256 × 256; pixel size, 0.3125 × 0.3125 mm; TR, 400 ms; TE, 16 ms; flip angle, 50°; number of averages, 4 and acquisition duration, 482 s. Details of the FLAIR protocol are as follows: FOV, 80 × 80 mm; slice thickness/gap, 1.0/0.5 mm; acquisition matrix, 264 × 128; pixel size, 0.625 × 0.15625 mm; TR, 7000 ms; TE, 68 ms; TI, 2000 ms; flip angle, 90°; number of averages, 2 and acquisition duration, 423 s.

Neuropathology

This study used formalin-fixed brain tissues from four patients with small vessel disease detected using in vivo MRI. After postmortem ex vivo micro-MRI scanning, brain blocks were dehydrated and embedded in paraffin. Brain blocks were cut in 8 μm thick serial sections using a microtome. Standard histologic staining was performed on these serial sections, which included haematoxylin/eosin staining and Kluver–Barrera staining. Further immunohistochemistry was performed to clarify the aetiology of small vessel disease (Supplementary Table 1).

Radiological−pathological comparative study procedure

The radiological−pathological comparative method for small vessel disease is summarized in Fig. 1. Neuroimaging markers of SVD on in vivo MRI were defined according to the STRIVE consensus criteria.¹⁶ Formalin fixed brains were cut in blocks that included the target lesion and were of a size equivalent to the FOV used in the ex vivo micro-MRI examination.

Figure 1

Radiological−pathological comparative method for small vessel disease. First, the target lesion should be retrospectively detected on in vivo MRI (A). Second, a brain block made to fit the field of view (80 × 80 mm) is set in the micro-MRI (B). Third, the brain block is sliced for neuropathological analysis while carefully inspecting the depth of the lesion from the surface of the brain block using ex vivo MRI (C). Finally, the brain sample is dehydrated, embedded in paraffin and cut in 8 μm-thick serial sections on a microtome at the estimated depth determined by ex vivo MRI. The lesion can be observed on glass slides for standard haematoxylin and eosin staining and additional immunohistochemistry (D).

Data availability statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Results

In vivo MRI-detected small vessel disease findings from four patients were examined (Table 1). Case 1 was neuropathologically diagnosed with cerebral amyloid angiopathy, while the others were diagnosed with non-cerebral amyloid angiopathy. Microinfarcts were detected on micro-MRI in Cases 1, 2 and 4. Three Tesla ex vivo micro-MRI was able to detect CMIs, including those <1 mm (Fig. 2A). Inflammatory cells propagated from the superficial to the deep layers along the perivascular area in the patient with cerebral amyloid angiopathy (Fig. 2B). The comparative study revealed cells positive for iron and cluster of differentiation (CD) 68 around the CMIs (Fig. 2C). An acute CMI in Case 4 was revealed to be due to microembolism showing positivity for peroxiredoxin, CD204, and glial fibrillary acidic protein (GFAP) (Fig. 3). Cortical hyperintensity on FLAIR and T2WI in Case 1 corresponded to positivity for serum protein exudate, astrocytic markers and the complement component C3d (Fig. 4). cSS was detected using in vivo and ex vivo MRI in Case 1 (Fig. 5A and B). A bilinear track-like appearance in the ex vivo MRI of Case 1 was compatible with iron deposition on the cortical surface and within the cortical layer (Fig. 5B and C). These were mostly cells of monocyte/macrophage lineage, but some were iron-positive foamy structures (Fig. 5C) negative for Aβ, GFAP and CD68 but slightly positive for ubiquitin (Supplementary Fig. 4). Although T2* low intensity, suggestive of cSS, was clearly revealed on the ex vivo MRI of Case 2, it was not detectable on in vivo MRI, and there were no cerebral amyloid angiopathy findings on the neuropathological examination (Supplementary Fig. 5A–C). In Case 2, iron-positive foamy structures were not observed (Supplementary Fig. 5D), and neither were cerebral amyloid angiopathy findings (Supplementary Fig. 5E).

Figure 2

Cortical microinfarcts revealed by the radiological−pathological comparative method with micro-MRI. CMIs 0.7−5 mm were detectable as low intensity on the T₁-weighted image of 3-T micro-MRI (A). The sizes and shapes on micro-MRI were almost the same using haematoxylin and eosin staining of the neuropathology (B). CMIs in Case 1 showed iron positivity in the cortical area (C). Some CD68-positive cells showed positivity for iron on double staining of iron and CD68, compatible with iron containing macrophages (inset of C). CD, cluster of differentiation; CMI, cortical microinfarct.

Figure 3

An acute cortical microembolism in Case 4. An acute cortical microembolism was observed on in vivo MRI (A). A hypointense lesion was observed on the T₁-weighted image of ex vivo micro-MRI (B). The radiologically suspected ischaemic lesion was neuropathologically shown to be a microinfarction that extended to the subcortical area (C). The microinfarction was positive for peroxiredoxin (D), CD 204 (E) and GFAP (F), and suggestive of inflammation and a neuroprotective reaction. CD, cluster of differentiation; GFAP, glial fibrillary acidic protein.

Figure 4

Polioaraiosis and a white matter lesion revealed by the radiological−pathological comparative method with micro-MRI. Cortical hyperintensity on FLAIR (A) and T2WI (B). Cortical hyperintensity is shown by the pale area on HE stained sections (C). Some enlarged perivascular spaces were observed (inset of each panel). The hyperintense area included the cortex, as observed by (KB) staining (D). Fib (E) and TF (F) were positive, including in vessel walls, suggesting a linkage with the damaged blood−brain barrier. This area had mild positivity for GFAP, suggestive of the presence of astrocytes (G). Although MSR was negative in the hyperintense area (H), some MSR-positive cells were observed around the vessels with an enlarged perivascular space (inset of H). C3d (I) and ApoE (J) were positive in the hyperintense area. Aβ40 was negative in the vessels with enlarged perivascular space, suggesting that Aβ clearance was maintained in such vessels (inset of K). Aβ42-positive plaques were also positive for MSR-positive cells, suggesting that the macrophages captured Aβ42 (inset of H and L). Aβ, amyloid beta; ApoE, apolipoprotein E; Fib, fibrinogen; FLAIR, fluid attenuated inversion recovery; GFAP, glial fibrillary acidic protein; HE, haematoxylin and eosin; KB, Klüver−Barrel; MSR, macrophage scavenger receptor; T2WI, T₂-weighted imaging, TF; transferrin.

Figure 5

Comparative findings of cortical superficial siderosis. cSS in Case 1 was detected as a hypointense cortical rim on T2*WI of in vivo MRI (A). cSS was detected as curvilinear hypointensity following the cortical surface on ex vivo micro-MRI that was consistent with the in vivo images (B). The low-intensity area was consistent with the iron-positive area on the iron staining of neuropathology (C). Iron-positive anuclear foamy structure was observed in cortical layers II–III (inset of C).

Table 1

Open in new tab

Summary of four cases with in vivo MRI-defined small vessel disease

Cases	1	2	3	4
Neuropathological diagnosis	Alzheimer’s disease, cerebral amyloid angiopathy	Amyotrophic lateral sclerosis	Corticobasal degeneration	Stroke
Age (years) at death	81	77	62	76
Sex (M: Male/F: Female)	M	M	M	M
Hypertension	+	+	−	−
Hyperlipidemia	−	−	−	−
Diabetes mellitus	−	+	−	−
Smoking	−	+	+	+
Atrial fibrillation	−	+	–	+
In vivo MRI	1.5-T	3-T	3-T	1.5-T
Cortical microinfarcts	−	−	−	+
Cerebral microbleeds	+	+	+	−
Cortical superficial siderosis	+	−	−	−
Ex vivo MRI	3-T	3-T	3-T	3-T
Cortical microinfarcts	+	+	−	+
Cerebral microbleeds	–	+	+	−
Cortical superficial siderosis	+	+	−	−
Neuropathology
Cortical microinfarcts	+	+	−	+
Cerebral microbleeds	+	+	−	−
Cortical superficial siderosis	+	−	−	−
Duration between in vivo MRI and ex vivo MRI/neuropathology	9 months	46 days	1.5 years	17 days

Cases	1	2	3	4
Neuropathological diagnosis	Alzheimer’s disease, cerebral amyloid angiopathy	Amyotrophic lateral sclerosis	Corticobasal degeneration	Stroke
Age (years) at death	81	77	62	76
Sex (M: Male/F: Female)	M	M	M	M
Hypertension	+	+	−	−
Hyperlipidemia	−	−	−	−
Diabetes mellitus	−	+	−	−
Smoking	−	+	+	+
Atrial fibrillation	−	+	–	+
In vivo MRI	1.5-T	3-T	3-T	1.5-T
Cortical microinfarcts	−	−	−	+
Cerebral microbleeds	+	+	+	−
Cortical superficial siderosis	+	−	−	−
Ex vivo MRI	3-T	3-T	3-T	3-T
Cortical microinfarcts	+	+	−	+
Cerebral microbleeds	–	+	+	−
Cortical superficial siderosis	+	+	−	−
Neuropathology
Cortical microinfarcts	+	+	−	+
Cerebral microbleeds	+	+	−	−
Cortical superficial siderosis	+	−	−	−
Duration between in vivo MRI and ex vivo MRI/neuropathology	9 months	46 days	1.5 years	17 days

Table 1

Open in new tab

Summary of four cases with in vivo MRI-defined small vessel disease

Cases	1	2	3	4
Neuropathological diagnosis	Alzheimer’s disease, cerebral amyloid angiopathy	Amyotrophic lateral sclerosis	Corticobasal degeneration	Stroke
Age (years) at death	81	77	62	76
Sex (M: Male/F: Female)	M	M	M	M
Hypertension	+	+	−	−
Hyperlipidemia	−	−	−	−
Diabetes mellitus	−	+	−	−
Smoking	−	+	+	+
Atrial fibrillation	−	+	–	+
In vivo MRI	1.5-T	3-T	3-T	1.5-T
Cortical microinfarcts	−	−	−	+
Cerebral microbleeds	+	+	+	−
Cortical superficial siderosis	+	−	−	−
Ex vivo MRI	3-T	3-T	3-T	3-T
Cortical microinfarcts	+	+	−	+
Cerebral microbleeds	–	+	+	−
Cortical superficial siderosis	+	+	−	−
Neuropathology
Cortical microinfarcts	+	+	−	+
Cerebral microbleeds	+	+	−	−
Cortical superficial siderosis	+	−	−	−
Duration between in vivo MRI and ex vivo MRI/neuropathology	9 months	46 days	1.5 years	17 days

Cases	1	2	3	4
Neuropathological diagnosis	Alzheimer’s disease, cerebral amyloid angiopathy	Amyotrophic lateral sclerosis	Corticobasal degeneration	Stroke
Age (years) at death	81	77	62	76
Sex (M: Male/F: Female)	M	M	M	M
Hypertension	+	+	−	−
Hyperlipidemia	−	−	−	−
Diabetes mellitus	−	+	−	−
Smoking	−	+	+	+
Atrial fibrillation	−	+	–	+
In vivo MRI	1.5-T	3-T	3-T	1.5-T
Cortical microinfarcts	−	−	−	+
Cerebral microbleeds	+	+	+	−
Cortical superficial siderosis	+	−	−	−
Ex vivo MRI	3-T	3-T	3-T	3-T
Cortical microinfarcts	+	+	−	+
Cerebral microbleeds	–	+	+	−
Cortical superficial siderosis	+	+	−	−
Neuropathology
Cortical microinfarcts	+	+	−	+
Cerebral microbleeds	+	+	−	−
Cortical superficial siderosis	+	−	−	−
Duration between in vivo MRI and ex vivo MRI/neuropathology	9 months	46 days	1.5 years	17 days

A comparative study of CMBs in the in vivo and ex vivo MRIs revealed hemosiderin deposits in the neuropathological specimens of Case 2 (Fig. 6). A lesion that had been radiologically diagnosed as CMB in Case 3 was revealed to be venous angioma on the neuropathological examination (Fig. 7A–D). An iron positive area was detected around the angioma (Supplementary Fig. 6). One lesion in Case 2 was an ambiguous CMI finding on in vivo MRI (Fig. 7E); this was recognized as a CMI using FLAIR in the ex vivo MRI (Fig. 7F) and continued to a patent microvessel that was occluded by microembolism on the neuropathological examination (Fig. 7G). The lesion was of low intensity, as revealed on T2*WI of the ex vivo MRI (Fig. 7H). Iron staining in the same location revealed iron deposits (Fig. 7I), indicating that the lesion was an embolic microbleed. The low intensity was retrospectively confirmed on in vivo MRI (Fig. 7J).

Figure 6

Comparative findings of a cerebral microbleed in Case 2. A cerebral microbleed (CMB) was observed as hypointensity on the susceptibility weighted image (SWI) of in vivo MRI (A). The hypointensity on SWI of the in vivo MRI was also detected as hypointensity on the T₂*-weighted image of ex vivo micro-MRI (B and C). Neuropathological study was performed to match the exact location of the lesion (D). Hemosiderin deposits were observed in the lesion by haematoxylin and eosin staining (E and F). These hemosiderin deposits were positive for iron staining (G and H). There were no Aβ-positive vessels detected by Aβ staining (I).

Figure 7

Venous angioma clinically diagnosed as a cerebral microbleed on in vivo MRI, and embolic microbleed revealed by the radiological−pathological comparative method with micro-MRI. A lesion in Case 3 diagnosed as CMB on SWI of in vivo MRI (A) was detected as a hypointensity lesion on T2*WI of ex vivo MRI (B), which was compatible with CMB. The lesion was revealed to be a non-CMB by the Elastica van Gieson staining of neuropathology (C). The lesion had vessel walls without internal elastic lamina, which indicated venous angioma (D). Case 2 had a low-intensity lesion on T1WI of in vivo MRI that constituted an ambiguous finding of CMI (E). The lesion showed high intensity on FLAIR of ex vivo MRI (F) and was confirmed as a CMI due to microembolism using neuropathological analysis, which showed fibrinoid degeneration in the adjacent CD31-positive pial arteriole meandering in the CMI (G). The CMI was identified as hypointense on ex vivo micro-MRI (H). Iron staining was positive on the slide showing meandering vessels in CMI (I). The site showed hypointensity on retrospective assessment on SWI of in vivo MRI (J). CD, cluster of differentiation; CMB, cerebral microbleed; CMI, cortical microinfarct; cSS, cortical superficial siderosis; FLAIR, fluid attenuated inversion recovery; SWI, susceptibility weighted image; T1WI, T₁-weighted imaging; T2*WI; T₂*-weighted imaging.

Discussion

To date, in vivo MRI−ex vivo neuropathological studies have rarely been reported because of associated methodological complications and difficulties. In this study, we improved upon an existing method to shed new light on the field of small vessel disease.¹⁰^,¹¹ The major findings of the present study are as follows. First, we showed that the in vivo MRI−ex vivo pathological comparative method using micro-MRI was efficient. Second, findings of in vivo MRI, ex vivo MRI and neuropathology of cSS and CMIs in a patient with cerebral amyloid angiopathy indicated that there is a close spatial relationship between these pathologies. Third, this study revealed the heterogeneity of the neuropathology of CMBs, which includes venous angioma and embolic CMB.

Several radiological−pathological comparative studies using MRI for human brain tissue have been reported.¹¹^,¹⁷ It is often difficult to ensure correspondence between thin brain−specimen slices and small lesions. Our method using ex vivo micro-MRI can form a ‘bridge’ between in vivo MRI and histopathological findings. Postmortem changes can be excluded and additional micro-MRI scans can be performed repeatedly within 30 min per sequence at any angle if necessary to adjust the slice to correspond precisely to the in vivo imaging. The strongest advantage of this method is the size of the scanned images, which precisely corresponds to the neuropathological specimens and therefore allows for accurate comparisons. A study with similar protocol using a microcoil for exvivo MRI-pathological comparison has been reported.¹⁵ The previous study and our study support the advantage of using microcoils for accurate comparison between exvivo MRI and neuropathology. Furthermore, our study has improved the accuracy of comparison between in vivo MRI and ex vivo MRI images because of the shorter time protocol of ex vivo MRI than in the previous studies of overnight protocol.¹⁰^,¹²^,¹⁵

A very limited number of facilities with 7-T MRI have reported pathological comparisons with autopsy brain MRIs.¹⁵^,¹⁷^,¹⁸ Although 7-T MRI can provide clear images, it is not widely used in clinical practice. We confirmed that 3-T micro-MRI was able to detect lesions that were even smaller than 1 mm in size and have so far remained invisible in in vivo and ex vivo 3-T MRI.³^,¹¹ This high resolution was attributed to the microcoils. The smaller the FOV, the higher the spatial resolution. In addition to the time and resolution, another advantage of using micro-MRI is the easy management of temperature. Our previous study using ex vivo 3-T MRI revealed the importance of retaining a proper temperature to enhance the signal-to-noise ratio. Micro-MRI is equipped with a warming device to maintain a constant temperature.

In vivo MRI-detected small vessel disease markers of cerebral amyloid angiopathy are not always caused by cerebral amyloid angiopathy.⁶^,¹⁹ We previously reported the differentiation of CMIs between cerebral amyloid angiopathy and micro-embolisms using in vivo 3-T MRI.²⁰ In this study, CMIs in Case 1 were associated with cerebral amyloid angiopathy, while an acute (subacute at autopsy) CMI in Case 4 was consistent with the features of microembolism. The ischaemic lesion after 17 days from onset showed positivity for peroxiredoxin, CD204 and GFAP, indicating sterile inflammation and a neuroprotective reaction.²¹ The neuroprotective reaction may be related to being obscure of small ischaemic lesions on MRI images over time.

Neuroimaging markers for SVD on in vivo MRI in Case 1 have included white matter hyperintensity, above which there were CMIs in the cerebral cortex. Myelin loss, astrogliosis, and enlarged perivascular spaces were observed, and these findings were in line with neuropathological changes, which have been described in white matter hyperintensities.²² Furthermore, the pallor area in the white matter extended to cerebral cortex, consistent with the finding previously described as polioaraiosis.²³ The neuropathological finding in Case 1 can be regarded precisely as tissue changes of hyperintensity detected on in vivo MRI.

Our data revealed that cSS has a close spatial relation to superficial CMIs. These findings indicate that the pathogenesis of CMIs may be related to inflammatory cells propagating along the perivascular area rather than to vascular occlusion. This result was in line with those of a recent cSS study.²⁴ However, some unique findings were revealed in our patient. A bilinear track-like appearance of the cSS on the ex vivo MRI was compatible with iron deposition on the pia matter and within cortical layers II–III. While iron-positive foamy structures were observed in the iron-deposited cortical area in the patient with cerebral amyloid angiopathy, these structures were not observed in the iron-deposited cortical area in a patient with non-cerebral amyloid angiopathy. More data are needed to clarify whether the iron-positive structures are associated with cerebral amyloid angiopathy as well as the elapsed time course of iron deposition.

One lesion clinically diagnosed as CMB on in vivo MRI was a venous angioma. Iron deposits around the angioma appeared to have low intensity on the SWI of in vivo MRI. CMBs identified using in vivo imaging should be carefully interpreted because lobar CMBs may not be necessarily related to cerebral amyloid angiopathy. The fact that some of the in vivo MRI-defined lobar CMBs might not be real CMBs is an important consideration during the diagnosis of cerebral amyloid angiopathy.²⁵ We previously reported that microembolisms can transform into microbleeds on MRI.²⁶^,²⁷ A CMI due to microembolism in Case 2 showed iron deposits, which were captured as a CMB on both ex vivo and in vivo MRI. The microembolism was probably associated with atrial fibrillation. To the best of our knowledge, this is the first report of embolic CMB confirmed neuropathologically. There is evidence to suggest that lobar CMBs may be more common than deep CMBs in patients with atrial fibrillation.²⁸^,²⁹ Atrial fibrillation is a common cause of cardioembolism. Considering these reports and our results, we should investigate the rate of embolic microbleeds in lobar areas, which potentially mimic cerebral amyloid angiopathy markers in both patients with cardiac risk factors and the healthy population.

Limitations

Limitations of this study include the small sample size and the variation in the time between the MRI imaging and pathological examination. Our data were not enough to examine the validity of recent findings of clinico-radiological SVD study.²⁰^,³⁰^,³¹ One problem with studies on autopsy tissues is that it is often difficult to obtain family consent owing to cultural and religious reasons. In the future, it will be necessary to determine, using a large cohort, how the time from in vivo MRI to postmortem MRI and pathological examination influences these results.

Conclusion

This method improves the diagnosis of cerebral small vessel disease because it allows for bidirectional assessment leading from in vivo images to pathology or from pathological examination to interpretation of in vivo images. Further studies are required to establish the efficiency of our radiological−pathological comparative method and to clarify the exact pathology of in vivo MRI-detected small vessel disease. The next step using our methods would be clarification of the proportion of apparent CMBs due to other aetiologies diagnosed on in vivo MRI, and clinico-patholocical correlation of white matter hyperintensities. To clarify the uncertainty of such SVD lesions could improve the diagnosis of SVDs and enable to make accurate expectation for severity, progression and impairment.

Acknowledgements

We are grateful to Dr Ko Matsuo and Dr Ken-ichiro Uno who provided care to the patients. We thank Mr Takahiro Natsume for setting initial parameters of micro-MRI, and Ms Kaori Kawamura for technical help during histopathological examinations.

Funding

This work was supported by Grant-in Aid from the Research Committee of adult-onset leukoencephalopathy, Research on Policy Planning and Evaluation for Rare and Intractable Diseases, Health, Labour and Welfare Science Research Grants, the Ministry of Health, Labour and Welfare, Japan (21FC1007). H.I. is supported by funding from Japan Society of Promotion of Science KAKENHI (Grant no 18K15446) and Kobayashi Magobe Memorial Medical Foundation. H.T. is supported by funding from Japan Society of Promotion of Science KAKENHI (Grant no 19H03572).

Competing interests

The authors report no competing interests.

Supplementary material

Supplementary material is available at Brain Communications online.

References

Pantoni

Cerebral small vessel disease: From pathogenesis and clinical characteristics to therapeutic challenges

Lancet Neurol

2010

;

(

689

–

701

Ueda

Satoh

Tabei

, et al.

Neuropsychological features of microbleeds and cortical microinfarct detected by high resolution magnetic resonance imaging

J Alzheimers Dis

2016

;

(

315

–

325

van Veluw

Shih

Smith

, et al.

Detection, risk factors, and functional consequences of cerebral microinfarcts

Lancet Neurol

2017

;

(

730

–

740

Takasugi

Miwa

Watanabe

, et al.

Cortical cerebral microinfarcts on 3T magnetic resonance imaging in patients with carotid artery stenosis

Stroke

2019

;

(

639

–

644

Hase

Polvikoski

Firbank

, et al.

Small vessel disease pathological changes in neurodegenerative and vascular dementias concomitant with autonomic dysfunction

Brain Pathol

2020

;

(

191

–

202

Charidimou

Boulouis

Gurol

, et al.

Emerging concepts in sporadic cerebral amyloid angiopathy

Brain

2017

;

140

(

1829

–

1850

Fernando

O'Brien

Perry

, et al.

Comparison of the pathology of cerebral white matter with post-mortem magnetic resonance imaging (MRI) in the elderly brain

Neuropathol Appl Neurobiol

2004

;

(

385

–

395

Gouw

Seewann

van der Flier

, et al.

Heterogeneity of small vessel disease: A systematic review of MRI and histopathology correlations

J Neurol Neurosurg Psychiatry

2011

;

(

126

–

135

Ishikawa

Niwa

, et al.

Comparison of premortem magnetic resonance imaging and postmortem autopsy findings of a cortical microinfarct

J Stroke Cerebrovasc Dis

2018

;

(

2623

–

2626

van Veluw

Biessels

Luijten

Zwanenburg

JJ.

Assessing cortical cerebral microinfarcts on high resolution MR images

J Visualized Exper

2015

;

(

105

53125

OpenURL Placeholder Text

Niwa

Shindo

, et al.

Comparative analysis of cortical microinfarcts and microbleeds using 3.0-Tesla postmortem magnetic resonance images and histopathology

J Alzheimer's Dis

2017

;

(

951

–

959

Crossref

van Veluw

Scherlek

Freeze

, et al.

Different microvascular alterations underlie microbleeds and microinfarcts

Ann Neurol

2019

;

(

279

–

292

Chen

Chodara

Sprecher

, et al.

A new method of reconstructing the human laryngeal architecture using micro-MRI

J Voice

2012

;

(

555

–

562

Lhuaire

Martinez

Kaplan

, et al.

Human developmental anatomy: Microscopic magnetic resonance imaging (muMRI) of four human embryos (from Carnegie Stage 10 to 20)

Ann Anat

2014

;

196

(

402

–

409

Humphreys

Jansen

Muñoz Maniega

, et al.

A protocol for precise comparisons of small vessel disease lesions between ex vivo magnetic resonance imaging and histopathology

Int J Stroke

2019

;

(

310

–

320

Wardlaw

Smith

Biessels

, et al.

Neuroimaging standards for research into small vessel disease and its contribution to ageing and neurodegeneration

Lancet Neurol

2013

;

(

822

–

838

van Veluw

Charidimou

van der Kouwe

, et al.

Microbleed and microinfarct detection in amyloid angiopathy: A high-resolution MRI-histopathology study

Brain

2016

;

139

(

Pt 12

3151

–

3162

PubMed

OpenURL Placeholder Text

van Veluw

Zwanenburg

Engelen-Lee

, et al.

In vivo detection of cerebral cortical microinfarcts with high-resolution 7T MRI

J Cereb Blood Flow Metab

2013

;

(

322

–

329

Charidimou

Martinez-Ramirez

Reijmer

, et al.

Total magnetic resonance imaging burden of small vessel disease in cerebral amyloid angiopathy: An imaging-pathologic study of concept validation

JAMA Neurol

2016

;

(

994

–

1001

Ishikawa

Shindo

, et al.

Cortical microinfarcts detected by 3-Tesla magnetic resonance imaging: Differentiation between cerebral amyloid angiopathy and embolism

Stroke

2020

;

(

1010

–

1013

Shichita

Hasegawa

Kimura

, et al.

Peroxiredoxin family proteins are key initiators of post-ischemic inflammation in the brain

Nature Medicine

2012

;

(

911

–

917

Alber

Alladi

Bae

, et al.

White matter hyperintensities in vascular contributions to cognitive impairment and dementia (VCID): Knowledge gaps and opportunities

Alzheimers Dement

2019

;

107

–

117

Crossref