Published online before print January 23, 2003, doi: 10.1161/01.STR.0000055022.86811.10
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
(Stroke. 2003;34:376.)
© 2003 American Heart Association, Inc.
Letters to the Editor |
Postmortem MRI as a Useful Tool for Investigation of Cerebral Microbleeds
Department of Neuroradiology, Umberto I Hospital and University of Ancona, Ancona, Italy
To the Editor:
We read with interest the article by Dichgans et al about cerebral microbleeds in CADASIL.1 The authors demonstrated a high frequency of focal areas of signal loss on gradient-echo MRI, suggesting past microbleeds, in CADASIL patients. They also performed postmortem examinations on the brains of 7 additional CADASIL subjects and found old microbleeds, defined as focal accumulations of hemosiderin-containing macrophages, in 6 of the 7 brains. In their article, the authors regret that none of these had previously been investigated by gradient-echo MRI, thus precluding a correlation study between neuroimaging and postmortem findings.
We felt obliged to suggest that the absence of previous in vivo MRI, however, does not absolutely preclude imaging-pathologic correlation. It has been demonstrated that MRI of excised and formalin-fixed brains, ie, postmortem MRI (pMRI), also called in vitro MRI, is a feasible and reliable method for delineating normal brain anatomy and detecting intracranial lesions.2–4 It can make correlative studies possible in those cases in which in vivo MRI is lacking, and it also has the advantage of imaging the brain just at the time of pathology, whereas correlation with previous in vivo MRI may be flawed because of changes that occurred between examination and the death of the patient. Changes caused by the fixation process must be considered before extrapolating brain pMRI findings to in vivo MRI, but they have been well described in previous studies.5–8 According to the literature and to our own experience with this technique, started as far as approximately 10 years ago,9,10 its value and usefulness vary according to lesion nature and location. In January 2000, we started a prospective MRI and pathological study of autopsy specimens to be examined for medico-legal purpose. Soon after removal, the brains were placed in buckets of formol saline and allowed to float in the fixative liquid for a minimum of 3 weeks, during which the formol was repeatedly changed. Between 6 and 12 h before the MRI examination, the brains were placed in tap water. MRI was performed, with the whole brain in a plastic box completely filled with water, by means of a superconducting 1.0-T system (Magnetom Expert Impact, Siemens) with head coil. Sagittal, axial, and coronal 5-mm spin-echo (600/23)(TR/TE), turbo spin-echo (4800/96), turbo-fluid attenuated inversion recovery (7000/105) and gradient-echo (600/26) images were obtained. After MRI, the brains were again kept in formol saline until they were sliced into coronal sections 1 cm thick with the help of 2 brass right angles. When abnormal findings were present on MRI and/or macroscopic examination, the corresponding slice/s was/were subjected to MRI again. For this MRI examination, a surface coil was used; initially (case 1) the slices were put into a plastic box filled with water, but subsequently water was no longer used as a medium because of major disadvantages encountered,4 and the slices were examined in air. Two series of 5-mm coronal images were obtained for each 1-cm-thick slice. Then, sections 15 µm thick were cut from the macroscopically and/or MRI-evident abnormal regions and stained with hematoxylin and eosin; if necessary, specific stains were also carried out. In our preliminary experience, dark signal spots revealed by gradient-echo MRI in formalin-fixed brains and brain slices were similar to those detected by in vivo MRI studies, and they corresponded to hemosiderin at pathology (unpublished data). As stated by Dichgans et al in their article, because microbleeds are small and inconspicuous in routinely stained histological sections, they are likely to be missed in autopsy studies. Owing to the high sensibility of MRI in detecting even minimal lesions, pMRI can be a potent help to neuropathologists for addressing detailed and microscopic examination in the light of the signal abnormalities detected, making it possible to optimize the pathological study; examination with gradient-echo pulse sequence, which has been well demonstrated to increase the sensitivity of MRI for hemorrhages, is optimally suited to assess the presence and pattern of past microbleeds. We also found pMRI very sensitive for detection of recent infarcts (which were revealed by MRI as very high signal lesions and diagnosed by subsequent MRI-guided histology), whereas it was of little value with those lesions (well evident at macroscopic pathological examination, however) located superficially, such as cortical contusions and subarachnoid hemorrhages, owing to susceptibility artifacts. Therefore, pMRI appears to be especially suited for detection of deeper lesions, eg, cortico-subcortical junction and white-matter abnormalities.
We would like to emphasize several advantages of pMRI. Fixed brains can be imaged at various distances of time from death, and on standard MR imagers. MRI information from the whole-brain multiplanar study gives a preliminary overall view of the brain condition while it is still intact, can be obtained repeatedly, and makes it possible to optimize the cuts in a single case in order not to miss minimal and/or deep lesions as well as initial changes. MRI of single slices, focusing the attention of the neuropathologist to abnormal regions, will indicate the site of microscopic investigation, ensuring that the signal abnormality is contained in the tissue slice being examined for histologic studies. Predissection multiplanar images of the whole brain will be left as a document once the brain has been sectioned and specimens have been obtained for microscopic examination.
To conclude, we would suggest that pMRI might provide a means for overcoming the limitation of not having in vivo MRI obtained in correlative studies; microbleeds are lesions for which this technique appears to be well suited.
References
1. Dichgans M, Holtmannspötter M, Herzog J, Peters N, Bergmann M, Yousry TA. Cerebral microbleeds in CADASIL. A gradient-echo magnetic resonance imaging and autopsy study. Stroke. 2002; 33: 67–71.
2. Hirsch WL, Kemp SS, Martinez AJ, Curtin H, Latchaw RE, Wolf G. Anatomy of the brainstem: correlation of in vitro MR images with histologic sections. AJNR Am J Neuroradiol. 1989; 10: 923–928.[Abstract]
3. Burger PC. Postmortem (specimen) MR. AJNR Am J Neuroradiol. 1990; 11: 912–913.
4. Van den Hauwe L, Parizel PM, Martin JJ, Cras P, De Deyn P, De Schepper AMA. Postmortem MRI with pathological correlation. Neuroradiology. 1995; 37: 343–349.[Medline] [Order article via Infotrieve]
5. Thickman D, Kundel H, Wolf G. Nuclear magnetic resonance characteristics of fresh and fixed tissue: the effect of elapsed time. Radiology. 1983; 148: 183–185.
6. Kamman R, Go K, Stomp G, Hulstaert C, Berendsen H. Changes of relaxation times T1 and T2 in rat tissues after biopsy and fixation. Magn Reson Imaging. 1985; 3: 245–250.[CrossRef][Medline] [Order article via Infotrieve]
7. Unger EC, Gado MH, Fulling KF, Littlefield JL. Acute cerebral infarction in monkeys: an experimental study using MR imaging. Radiology. 1987; 162: 789–795.
8. Braffman B, Zimmerman R, Trojanowski J, Gonatas N, Hickey W, Schlaepfer W. Brain MR: pathologic correlation with gross and histopathology. AJNR Am J Neuroradiol. 1988; 9: 621–628.
9. Salvolini U, Scarpelli M. Magnetic resonance in the pathologic evaluation of the isolated human brain [in French]. Bull Acad Natl Med. 1994; 178: 337–345.[Medline] [Order article via Infotrieve]
10. Scarpelli M, Salvolini U, Diamanti L, Montironi R, Chiaromoni L, Maricotti M. MRI and pathological examination of post-mortem brains: the problem of white matter high signal areas. Neuroradiology. 1994; 36: 393–398.[CrossRef][Medline] [Order article via Infotrieve]
Department of Neurology
Department of Neuroradiology, Klinikum Großhadern, München, Germany
Response
We thank Drs Messori and Salvolini for their interest in our study. The authors suggest postmortem MRI as a tool for studying cerebral microbleeds. We fully agree. In fact, we would like to draw their attention to 2 studies that were cited in our article.1,2 Fazekas et al performed postmortem MR imaging on the brains of 11 patients who had died of intracerebral hemorrhage.1 They found small areas of signal loss on gradient-echo T2*-weighted MR images in 7 brains. These changes correlated well with focal hemosiderin deposits on histological sections. Similar results were obtained in the study by Tanaka et al.2
References
1. Fazekas F, Kleinert R, Roob G, Kleinert G, Kapeller P, Schmidt R, et al. Histopathologic analysis of foci of signal loss on gradient-echo T2*-weighted images in patients with spontaneous intracerebral hemorrhage: evidence of microangiopathy-related microbleeds. AJNR Am J Neuroradiol. 1999; 20: 637–642.
2. Tanaka A, Ueno Y, Nakayama Y, Takano K, Takebayashi S. Small chronic hemorrhages and ischemic lesions in association with spontaneous intracerebral hematomas. Stroke. 1999; 30: 1637–1642.
This article has been cited by other articles:
 |
|
 |
|
  A V MacLean, R Woods, L M Alderson, S P Salloway, S Correia, S Cortez, and E G Stopa Spontaneous lobar haemorrhage in CADASIL J. Neurol. Neurosurg. Psychiatry, March 1, 2005; 76(3): 456 - 457. [Full Text] [PDF]
|
|
|
|
 |
|
 S. L. Kraft, D. Dailey, M. Kovach, K. L. Stasiak, J. Bennett, C. T. McFarland, D. N. McMurray, A. A. Izzo, I. M. Orme, and R. J. Basaraba Magnetic Resonance Imaging of Pulmonary Lesions in Guinea Pigs Infected with Mycobacterium tuberculosis Infect. Immun., October 1, 2004; 72(10): 5963 - 5971. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||