Cerebral hemorrhage produced by ruptured dissecting aneurysm in miliary aneurysm.код для вставкиСкачать
BRIEF COMMUNICATIONS Cerebral Hemorrhage Produced by Ruptured Dissecting Aneurysm in Miliary Aneurysm William I. Rosenblum, MD This report describes, for what may be only the second time, a ruptured miliary aneurysm within a cerebral hemorrhage. The report is unique in that the aneurysm has arisen at the site of a dissection within the wall of an arteriole at a site of fibrinoid necrosis. The case not only is a unique illustration of this pathogenetic pathway to miliary aneurysm formation, but also reemphasizes the relationship between fibrinoid and miliary aneurysm formation. Ann Neurol 2003;54:376 –378 Recently, Fisher1 reported what is apparently the first demonstration of the remains of a ruptured miliary aneurysm within a cerebral hemorrhage. This demonstration is especially important because the role of miliary aneurysms and indeed their very existence has been disputed2–5 since their initial description by Charcot and Bouchard.6 Serial sections may be required to demonstrate miliary aneurysms, and even then their remains may be “lost” within the hemorrhage.1 Thus, prior to Fisher’s report1 histological evidence for their importance was indirect, for example, in demonstrations of unruptured miliary aneurysms4,7 or of fibrosed miliary aneurysmal sacs in sections taken at random from sites distant from the hemorrhage itself.4 Such cases4,7 often also demonstrated fibrinoid necrosis in arteriolar segments, offering support to the hypothesis that such areas might either rupture or might be sites for formation of miliary aneurysms that might rupture.5 In view of the debate concerning the importance of these aneurysms, it is important to provide additional examples of ruptured aneurysms within a hemorrhage. This might best be accomplished by serially sectioning smaller hemorrhages, thereby reducing the required la- From the Department of Pathology, Division of Neuropathology, Virginia Commonwealth University’s Medical College of Virginia, Richmond, VA. Received Mar 10, 2003, and in revised form Apr 29. Accepted for publication Apr 29, 2003. Address correspondence to Dr Rosenblum, 305 Tarrytown Drive, Richmond, VA 23229. E-mail firstname.lastname@example.org 376 bor and perhaps reducing the chance that large amounts of rapidly accumulated blood may have displaced the remnants of the aneurysm, making it more difficult to demonstrate its presence. The following report demonstrates such an aneurysm but does so in a context never before reported to my knowledge. The aneurysm has undergone dissection at its point of origin at a site of fibrinoid necrosis, and the outer wall of the dissection has ruptured, causing the parenchymal hemorrhage. Case Report A lethargic, 83-year-old African American man was brought to the hospital in a cachectic (84lb) bedridden state. He was hypoglycemic without a history of diabetes or insulin therapy. Hypoglycemia responded to glucose. He developed a right-sided seizure and coma and died 2 days later. Significant general autopsy findings were serous atrophy of fat, hypocellular bone marrow, mild to moderate generalized atherosclerosis, emphysema, mild right ventricular hypertrophy, and arteriolonephrosclerosis. The heart weighed 275gm with serous atrophy of adherent epicardial fat.. Although the patient was not hypertensive, the brain displayed lesions commonly associated with hypertension. These are described below. Because of the presence of lesions, it was thought that the patient had been hypertensive and that his cachectic state reflected by the serous atrophy was responsible for the absence of a high blood pressure and for the heart falling within normal weight limits. Results The brain showed old and fresh lesions. Old lesions included thin, bilateral subdural membranes, small bilateral contusions, several lacunar infarcts in the basal ganglia, and two old hemorrhages in basal ganglia. A completely fibrosed miliary aneurysm was found adjacent to the old hemorrhages. Recent lesions were bilateral infarctions in frontoparietal lobes and two fresh hemorrhages. One was 5mm wide, in the left putamen. The other was 3mm wide in the right basis pontis. A portion of the former and all of the latter were serially sectioned. Alternate sections were stained with hematoxylin and eosin, Verhoeff Van Giesen stain for elastic tissue, and an azocarmine stain. Fibrinoid degeneration was defined by the presence of amorphous or granular eosinophilic material within the vessel wall, which stained red with the azocarmine stain.8 This characteristic red staining parallels the identification of fibrinlike material with other stains such as phosphotungstic acid–hematoxylin, or the Putz stain, while giving a clear tinctorial distinction between blue collagen–hyalinized or normal- and red fibrinoid. A miliary aneurysm was found in the fresh basal ganglia hemorrhage. A point of rupture was not found, but a portion of the aneurysm was missing because only part of the hemorrhage was sampled, and the remaining tissue was discarded after cutting the brain. © 2003 American Neurological Association Published by Wiley-Liss, Inc., through Wiley Subscription Services Fig. (A) Typical double-barreled lumen of a dissecting aneurysm. The outer barrel is 1mm wide. The inner barrel is marked by an “X.” Panel B shows a different level through the dissection. The inner barrel is marked by an “X.” An arrow in the upper left portion of the outer barrel marks a point of rupture. The area in the white box has been enlarged in panel C to better show the neck between the inner barrel and the remnant of the parent arteriole. Panel D shows another level stained with hematoxylin and eosin. The dissecting space between the right side of the inner barrel and the parent vessel is labeled “Y.” The wall of the neck between inner barrel and parent arteriole is filled with amorphous eosinophilic material as is the outer wall of the parent arteriole (labeled “F”). This material stained red, like fibrin, with azocarmine stain as shown by the dark amorphous material in the outer wall (small arrowheads) of the parent arteriole at another level (E) at high magnification. In this panel, the entire image has been rotated 90 degrees so that what was the right edge of the bridge connecting inner barrel to parent arteriole is now its lower edge (large arrowheads). Fibrinoid was present at several points at the junction of the aneurysm with the parent arteriole. The smaller hemorrhage in the basis pontis was completely sectioned and revealed the unique findings shown in the Figure. A blood filled aneurysm approximately 1mm wide was found. Within its lumen, there was another lumen (see Fig, A). Serial sections showed a point of rupture (see arrow in Fig, B) in the wall of the outer lumen. The wall of the inner lumen also showed evidence of disintegration and a possible point of rupture (see Fig, D, upper left portion of lumen). The “double-barreled” configuration is typical of dis- Rosenblum: Miliary Aneurysm and Hemorrhage 377 secting aneurysms seen in arteries but never before, to my knowledge, demonstrated in arterioles. The lumen of the inner “barrel” was contiguous with a very thin neck only approximately 40m in diameter with its own narrow lumen connecting the aneurysm to the parent arteriole (see Fig, B, C). Serial sections showed that the parent vessel was approximately 100m in diameter. The neck that connected the inner “barrel” to the parent arteriole contained masses of fibrinoid, as did portions of the wall of the parent arteriole (see F in Fig, D) also shown at higher magnification (small arrowheads) in panel E. Because miliary aneurysms have been reported not only in hypertensives but also in cases of congophilic (ie, amyloid) angiopathy, note that there was an absence of vascular amyloid as defined by Congo red staining and green birefringence of vessel walls after Congo red staining. This absence is consistent with the central and pontine locations of the hemorrhages, because congophilic angiopathy would be expected to produce so-called lobar hemorrhages that are in the periphery of the lobes of the hemispheres. Discussion This case strongly supports the assertion that cerebral hemorrhage may be produced by a ruptured miliary aneurysm. An unruptured portion of an aneurysm was found within the partially sectioned fresh hemorrhage. A ruptured aneurysm was found in the completely sectioned fresh hemorrhage, thus supplying important confirmation of Fisher’s1 demonstration. However, the ruptured miliary aneurysm in this case presents a unique pathogenetic feature. The aneurysm is, in effect, a dissecting aneurysm whose 1mm-wide outer “barrel” is of microaneurysmal dimensions and whose inner “barrel” is also “aneurysmal,” being several times wider than the parent arteriole. The presence of fibrinoid necrosis presumably weakened the wall of the parent arteriole thereby leading to both aneurysmal dilation and dissection. The dissection, in turn, may have played a key role in producing the parenchymal hemorrhage because the thin outer barrel represents a locus of reduced resistance to rupture. To my knowledge, this is the first demonstration of either microdissection at a point of fibrinoid degeneration or of miliary aneurysm participating in dissecting aneurysm after arising at the site of the fibrinoid degeneration. Fisher9 has reported the dissection of a branch of the basilar artery, producing occlusion of the lumen and an infarct. There was neither fibrinoid nor aneurysm formation at the site. Note that the term fibrinoid degeneration or fibrinoid necrosis has been used here rather than the term lipohyalinosis. The latter was introduced as a replacement for the former in the cerebral arterioles and emphasizes other elements in the degenerated wall of the arteriole.1 378 Annals of Neurology Vol 54 No 3 September 2003 However, recent studies2,3 suggest that the fibrinoid may precede the other changes found in lipohyalinosis and that the hyalinized collagen represents healed fibrinoid. Whatever the merits of these hypotheses, it appears probable that it is the fibrinoid that leads to hemorrhage either directly or via formation of miliary aneurysms. References 1. Fisher CM. Hypertensive cerebral hemorrhage. Demonstration of the source of bleeding. J Neuropath Exp Neurol 2003;62: 104 –107. 2. Kalimo H, Kaste M, Haltia M. Vascular diseases. In: Graham DI, Lantos PL, eds. Greenfield’s neuropathology. 7th ed. Vol 1. London: Arnold, 2002:295. 3. Lammie GA. Hypertensive cerebral small vessel disease and stroke. Brain Pathol 2001;12:358 –370. 4. Rosenblum WI. Miliary aneurysms and “fibrinoid” degeneration of cerebral blood vessels. Hum Pathol 1977;8:133–139. 5. Rosenblum WI. The importance of fibrinoid necrosis as the cause of cerebral hemorrhage in hypertension. Commentary. J Neuropath Exp Neurol 1993;52:11–13. 6. Charcot JM, Bouchard C. Nouvelle recherches sur la pathogenie de l’hemorrhagie cerebrale. Arch Physiol Norm Pathol 1868;1: 643– 645. 7. Fisher CM. Pathological observations in hypertensive cerebral hemorrhage. J Neuropath Exp Neurol 1971;30:536 –550. 8. Feigin I, Prose P. Hypertensive fibrinoid arteritis of the brain and gross cerebral hemorrhage. A form of “hyalinosis.” Arch Neurol 1959;1:98 –110. 9. Fisher CM. Bilateral occlusion of basilar artery branches. J Neurol Neurosurg Psychiat 1977;40:1182–1189.