Cerebral amyloid angiopathy

DR FRANK S. CELESTINO (Assistant Professor, Department of Family Medicine): We begin today's Grand Rounds with the relatively common story of an older patient presenting with acute mental status changes. We could not realize at the time that this patient's evolving illness over the next 8 months would introduce us to a "new" threat to the aging cerebrum and would improve our understanding of emerging concepts relating dementia, cerebrovascular insufficiency, and normal brain aging.

CASE PRESENTATION

DR JOSEPH C. KONEN (Assistant Professor, Department of Family Medicine): Ms F.M., a 60-year-old woman, was admitted to the hospital in January 1988 for acute mental status changes, having been found at home lethargic, mumbling, incontinent of urine, and poorly responsive to family attempts at arousal. During the month preceding admission, Ms M.'s job performance as a bookkeeper had deteriorated markedly. Her employer had noticed increasing difficulty with even simple mathematics, while her family had noted unsteadiness of gait, difficulty focusing vision, and complaints of headache. For symptom relief she took alprazolam and a butalbital-aspirin combination.

Past medical history was most significant for a 20-year history of moderately well controlled hypertension with a negative radiologic and metabolic workup for secondary causes. Over the years there had been several episodes of symptomatic accelerated hypertension. On two occasions, brief complaints of facial and arm numbness were labeled as possible transient ischemic attacks. In 1985, type II diabetes mellitus was diagnosed. Ms M. denied smoking, rarely drank alcohol, and had no known allergies. No occupational exposures were evident. Family history was positive only for Ms M.'s father dying of atherosclerotic heart disease, two sisters and a child with hypertension, and one brother with a stroke. Medications included 0.5 mg of alprazolam, 1 to 2 tablets of a butalbital-aspirin combination, and 40 mg of propranalol, all taken three times a day, as well as 20 mg of furosemide and 5 mg of glyburide taken once daily.

At admission, her blood pressure was 170/100 mm Hg, her heart rate was 52 beats a minute and regular, her respirations were 20 per minute and unlabored, and her temperature was 36.9 [degrees] C (98.2 [degrees] F). The initial examination revealed an obtunded, mildly obese woman who was barely arousable by loud noises, name calling, or sternal rub. She was in no apparent distress and moved extremities purposefully to painful stimuli. The complete physical examination was otherwise normal except for mild bilateral papilledema as the only focal neurologic finding.

The following admission laboratory tests were either normal or negative: complete blood count, automated (SMAC 20) chemistry panel (excluding glucose of 17.4 mmol/L [313 mg/dL]), coagulation studies, thyroid function tests, serum iron, creatine kinase, sedimentation rate, arterial blood gases, ammonia level, rapid plasma reagin test, and ethanol level. Urine drug screening was positive

only for barbiturates. Serum alprazolam level was therapeutic. Urine for heavy metals initially indicated mild arsenic elevation, but hair analysis and repeat urine tests were negative. Abdominal films were normal. Chest x-ray examination showed only borderline cardiomegaly. Electrocardiogram (ECG) showed mild sinus bradycardia, left anterior hemiblock, and left atrial dilatation. Cerebrospinal fluid (CSF) studies (including full cultures) were unremarkable. The opening pressure was modestly elevated at 28.5 cm but no xanthochromia was noted. An electroencephalogram showed only nonspecific bifrontal intermittent rhythmic delta activity. An ophthalmology consultant verified mild optic disc edema without signs of neuritis or field cut.

Dr Williams will now review the results of the cerebral imaging procedures used to investigate Ms M.'s problem.

DR DANIEL WILLIAMS (Neuroradiology Fellow, Department of Radiology): An unenhanced cranial computed tomography (CT) scan done at admission demonstrated a possible subarachnoid hemorrhage, diffuse deep white matter radiolucencies, and poor delineation of the basilar cisterns and cerebral sulci (felt to be secondary to either cerebral edema or the presence of subarachnoid blood). A repeat CT scan 48 hours later revealed decreasing edema but no new findings.

Because of the clinical and radiologic suspicion of subarachnoid hemorrhage, a four-vessel cerebral arteriogram was performed. No abnormalities were detected. The patient did have prominent infundibula associated with the posterior communicating arteries, but there was no radiographic evidence that either of these had bled. Because of the difficulty in establishing a definitive diagnosis, a cranial magnetic resonance imaging (MRI) scan was performed. This study revealed extensive areas of increased T[.sub.2] signal involving the periventricular and subcortical white matter and brain stem, but failed to further clarify the clinical and CT findings.

DR KONEN: Because of the possibility of pseudotumor cerebri, the patient had been started on acetazolamide. The consulting neurologist believed, however, that the clinical picture was perhaps more compatible with some unknown post-infectious demyelinating encephalitis or encephalopathy than pseudotumor." He recommended conservative, supportive management. The patient's hospital course was punctuated by marked fluctuations in mental status with days of clear sensorium, alertness, and full orientation interspersed with ones of confusion and agitation.

Having been lucid and oriented for nearly 3 days with normal blood pressure, Ms M. was discharged 13 days after admission with tentative diagnoses of nonspecific encephalitis, subarachnoid hemorrhage, possible pseudotumor cerebri, diabetes, and hypertensive vascular disease. She was sent home on acetazolamide, 250 mg daily, and her usual medications. Formal neuropsychological testing at the time of discharge documented persistent cognitive deficits with acquired intellectual impairment and abnormally low scores on receptive vocabulary, attention, and visual and verbal recall.

Three months after discharge, during an emergency department visit for severe headache and moderately elevated blood pressure, Ms M. had another enhanced cranial CT scan showing resolution of edema but persistent deep white matter lucencies.

In September of 1988, Ms M. presented, alert and talkative, to the emergency department with severe bifrontal headache and blood pressure of 160/110 mm Hg. While in the emergency department, she suddenly became unresponsive except to deep pain. Admission vital signs were normal except for a blood pressure of 160/110 mm Hg. As in January, Ms M. was obtunded yet moved extremities purposefully to painful stimuli. Complete physical and neurologic examinations were otherwise unremarkable except for a nonreactive, mid-position (4 to 5 mm) right pupil, mild fight ptosis, and inability to adduct the right eye from the midline. This third nerve palsy was thought by a consulting neurologist to be secondary to edema or possible aneurysm.

As before, a complete laboratory evaluation was unrevealing. A rheumatology panel was negative. Serum and urine heavy metals were normal. CSF analysis was once again normal with an opening pressure of 16 cm and no xanthochromia. ECG was unchanged. As before, several cranial imaging modalities were used.

DR WILLIAMS: During the second hospitalization, an initial uninfused CT scan demonstrated unequivocal subarachnoid hemorrhage, cerebral edema with mild brain stem compression, obliteration of the basilar cisterns and sulci, and unchanged white matter lesions. A repeat CT scan 6 days later showed resolving edema and subarachnoid hemorrhage but also a new small lacunar infarct in the right caudate nucleus. After a second cerebral arteriogram failed to show any intracranial abnormalities, a gadolineum-enhanced MRI scan once again revealed the extensive deep white matter abnormalities and a hemorrhagic infarct of the head of the right caudate nucleus.

In retrospect it is interesting to reexamine the CT scan done during the emergency department visit between hospitalizations. It may, in fact, show a small subarachnoid hemorrhage.

DR KONEN: By the second hospital day, Ms M. was more alert with clear speech. A mental status examination showed loose associations, tangential thinking, flights of ideas, thought perseveration, poor orientation except to person, and deficits injudgment, calculation, and abstraction. Her memory was intact. Optic fundi and visual fields were normal. She was felt to have possible hypertensive encephalopathy and recurrent subarachnoid hemorrhage. Ms M. received corticosteroids, antihypertensive therapy, and increased acetazolamide.

Because of the patient's fluctuating mental status and confusing array of clinical, laboratory, and radiologic findings, an open biopsy of the right frontal lobe was performed to investigate the underlying pathology.

NEUROPATHOLOGY

DR JEAN N. ANGELO (Professor Emeritus, Department of Pathology): A wedge of cerebral cortex and subcortical white matter was submitted. There were no gross abnormalities. Several special stains (silver, myelin, fungal, and Congo red for amyloid) were used to supplement the standard hematoxylin-eosin technique. Many capillary-sized vessels, arterioles, and intermediate vessels showed strong congophilia, which had green birefringence with polarized light. Smaller vessels were markedly thickened. In addition to several intracortical glial scars, there were two small subacute infarcts associated with hemosiderin macrophages and mineralization or calcification of small vessels centrally. Only two immature neuritic plaques were noted, but no neurofibrillary degeneration was seen. Subcortical white matter showed a slight increase in astrocytes, but myelin was not significantly altered. Stains other than for amyloid were negative. Electron microscopy, primarily of subcortical white matter, demonstrated thickened small vessels with split laminae and fibrils consistent with amyloid. These findings amply support a neuropathologic diagnosis of cerebral congophilic (amyloid) angiopathy.

CLINICAL FEATURES

DR CELESTINO: Although the pathologic entity of cerebral amyloid angiopathy (CAA) has been recognized since the first decade of this century, (1) only during the last 10 to 15 years has it attracted increasingly intense interest from the research community. This "rediscovery" has been fueled by two observations: (1) that CAA is the probable cause of nontraumatic primary cerebral hemorrhage in a significant portion of elderly patients, especially those who are normotensive or who suffer recurrent episodes of bleeding (1-3), and (2) that CAA is closely associated with the microscopic hallmarks of Alzheimer's disease and may in fact be causally related to this dementing disease. (4-7) Several recent reviews (1-13) have summarized our current understanding of CAA as well as its relationship to dementia and brain aging in general.

CAA, also known as cerebrovascular amyloidosis, microvascular amyloid, and cerebral congophilic angiopathy, is seldom, if ever, associated with systemic amyloidosis. It is characterized by the deposition of homogeneous (amotphous) eosinophilic material in the walls of arterioles, small arteries, and capillaries of the cortex and leptomeninges. (1,2,11) These deposits, which stain with Congo red and show characteristic yellow-green birefringence under polarized light, are by definition amyloid. The media and adventitia can become replaced by this amyloid, and the elastic lamina may be fragmented, split, or destroyed. The presence of amyloid deposits weakens the vessel wall, which may undergo spontaneous or traumatic rupture or develop microaneurysmal dilatation with subsequent hemorrhage. (1,2)

The incidence of CAA is clearly age related. (1,2,9) It has been estimated that 8% of individuals at age 60 years, 37% at age 80 years, and nearly 60% by the tenth decade are affected. (9) Men and women are equally susceptible. Likewise, the mean age of 72 years for onset of CAA-related brain hemorrhage shows no sex predilection. (1)

CAA predominantly affects the neocortex, relatively sparing the vessels of the basal ganglia, cerebellum, and brain stem. (14) This distribution explains the cortical and subcortical location of the associated hemorrhages, which frequently rupture through the cortical surface to extend into the subarachnoid space. This pattern differs distinctly from hypertensive hemorrhage, which typically involves the basal ganglia, thalamus, cerebellum, and pons, and rarely reaches the subarachnoid space. As in our case, however, hypertension exists concomitantly in up to one third of patients with CAA and may contribute to the occurrence, size, and extent of CAA-related hemorrhages. (1,4,8, 10)

DR WILLIAMS: Radiographically, hemorrhages secondary to CAA are usually large and may be multiple or recurrent. (1-3,11,15-20) They occur in superficial, subcortical, or lobar locations. Episodes of lobar bleeding are most common in the frontal and parietal regions, with a somewhat lesser incidence in the temporal and occipital areas. CAA uncomplicated by intracerebral hemorrhage appears most severe in the parieto-occipital region. (1, 14) The actual site of bleeding from an amyloid-infiltrated microvessel is rarely identified even after angiography or surgery. Our case well illustrates this point.

DR CELESTINO: It is estimated that CAA now accounts for 5% to 15% of all primary nontraumatic brain hemorrhages, which in turn account for approximately 10% of all strokes. (1) The risk of recurrent hemorrhage is 10% to 30%.8 Thus, CAA must be accepted as an increasingly important cause of nontraumatic intracerebral hemorrhage and can be strongly suspected in the setting of an elderly normotensive patient with recurrent or multiple simultaneous cerebral hemorrhages, particularly lobar ones. The burgeoning numbers of older individuals, especially those aged over 80 years, will bring this pathologic entity further into the clinical limelight in the years ahead.

CAA is unrelated to atherosclerosis of cerebral vessels but may of itself cause multiple small cortical and subcortical infarcts in severely affected regions. (1,9,11) These usually asymptomatic microinfarctions occur in about one fifth of patients and are attributed to occlusion of vessels by amyloid deposition and fibrinoid degeneration. Large infarcts should never be attributed to CAA.

CAA most often occurs in isolation as a single entity but may also be associated with other pathologic states. (1,2) It has been recognized in dementia pugilistica, Creutzfeldt-Jakob disease, cerebellar ataxia, granulomatous angiitis, rheumatoid vasculitis, giant cell arteritis, post-irradiation necrosis, vascular malformations, spongiform encephalopathy, other rare degenerative dementias, and in adults with Down's syndrome. There are also familial forms of autosomal-dominant cerebrovascular amyloid reported from Iceland and the Netherlands causing premature intracerebral hemorrhage in the fourth and fifth decades. (1,2,4)

Interestingly, parenchymal brain hemorrhages from CAA are very rare in middle-aged patients with Down's syndrome who otherwise have brain abnormalities, including extensive CAA, identical to those seen in Alzheimer's disease. (4) Given this observation, some have proposed that the degenerative microvascular changes leading to rupture of brittle, amyloid-laden vessels may need to be abetted by hypertensive or atherosclerotic vascular disease occurring in the elderly before hemorrhage becomes likely. Clearly, the entire pathogenesis of CAA-related bleeding is not understood.

Recovery from a single episode of CAA-associated bleeding can be surprisingly good, (1,4,21) although overall long-term prognosis is often poor. Most authors (1-3,8-12) advocate conservative, nonsurgical supportive management of intracerebral hemorrhage related to CAA including aggressive management of intracranial hypertension. Because amyloid replaces the contractile elements of vascular walls, the first step in hemostasis is impaired. Surgery plays a limited role because of excessive intraoperative bleeding and recurrent hemorrhage in the early or late postoperative phase. Brain and meningeal biopsy, the only definitive means of establishing a diagnosis of CAA, may in itself cause fatal cerebral hemorrhage.

RELATIONSHIP TO DEMENTIA

Besides intracerebral hemorrhage, the other major association of CAA is with dementia. (1,2,4,8,9,13) Forty to fifty percent of patients with proven CAA exhibit dementia, and a similar proportion show Alzheimer's disease changes at autopsy. Interestingly, although CAA is found in a variety of dementing diseases, as mentioned earlier, it is most severe in patients with Alzheimer's disease, occurring in 90% to 100% of affected brains. The amount of vascular amyloid, however, has failed to correlate with the severity of dementia. Also, severe CAA with associated intracerebral hemorrhage clearly may occur without any of the microscopic or clinical features of Alzheimer's disease. Likewise, Alzheimer's disease probably can occur in the absence of CAA.

Nonetheless, because the incidence of both CAA and Alzheimer's disease increase strikingly with age, and because the senile plaques and neurofibrillary tangles of Alzheimer's disease both contain amyloid, researchers have wondered whether there might be more than a casual relationship between amyloid deposition and the development of dementia in Alzheimer's disease. (4-7) Further interest was engendered by the observation that senile plaques with amyloid cores are often in close proximity to CAA-affected microvessels. In an exciting development, recent immunochemical and biochemical evidence suggests that senile plaque amyloid, CAA deposits, and neurofibrillary tangle amyloid (in both Alzheimer's disease and Down's syndrome) are identical. Additionally, the gene markers for Alzheimer's disease and for the amyloid precursor protein are both located on chromosome 21 in very close proximity. The above findings have prompted some to designate Alzheimer's disease as a form of cerebral amyloidosis. (6)

An attractive current hypothesis, (1,4-7) increasingly supported by recent work at the molecular level, proposes that the amyloid protein gene on chromosome 21 encodes a normal precursor that is subsequently (posttranslationally) abnormally processed by an aberrant enzyme (protease) encoded by the familial Alzheimer's disease gene also on chromosome (21). In one theory the amyloid precursor, arising in peripheral sites, is disseminated through the blood stream to the central nervous system, where cerebral endothelial cell proteases cleave it to produce amyloid fibrils. These fibrils may disrupt that blood-brain barrier, allowing influx of more amyloid into the central nervous system to accumulate pathologically and destructively in plaques, tangles, and vessel walls. The search is now on to identify the specific abnormal proteases and their location and regulation. Understanding these molecular mechanisms could lead to the development of specific proteolytic inhibitors (drugs) used to halt Alzheimer's disease.

One last set of observations has recently given rise to a novel and certainly heretical hypothesis linking CAA, Alzheimer's disease, and dementia. (21-25) Thirty to sixty percent of patients with Alzheimer's disease exhibit periventricular white matter lucencies on CT scanning compared with only approximately 10% to 20% of age-matched controls. Hachinski et al (24) have coined the term leukoaraiosis (LA) for these lesions, although leukoencephalopathy has sometimes been used. MRI, with its increased resolution (sensitivity), has demonstrated an even more alarming frequency of LA on T[.sub.2]-weighted images. (21,23,26) As with CAA and Alzheimer's disease, these findings increase dramatically with age, with over 30% of all patients aged over 60 years demonstrating LA by MRI. CAA is only one of several pathologic entities associated with LA. Other disease correlations include multiple sclerosis, leukodystrophies, hydrocephalus, previous head trauma, brain irradiation, vasculitides, sickle cell disease, and Binswanger's disease (subcortical arteriosclerotic encephalopathy). (23-25)

Most authors, (21,23-30) in keeping with the near 100% prevalence of LA in vascular dementias, believe from pathological correlations that these areas represent demy-elination, hyalinosis, fibrosis, and injury secondary to hypoperfusion (vascular insufficiency). Indeed, recently Brun et al (22) have documented that incomplete, evolving microinfarctions were noted in 60% of brains from patients with Alzheimer's disease and correlated with the areas of LA caused by CAA. Awad et al (26) have performed pathologic correlations to areas of LA seen on MRI and once again verified areas of demyelination, arteriolar thickening, and hypoperfusion injury. Several reports (20,28,29) have verified the nearly 100% prevalence of LA in CAA-affected brains. Once again pathologic studies implicate vascular insufficiency as the cause of these radiographic lesions.

Given this evolving database, investigators (21 25,29,30) have begun to wonder about the functional role of the white matter lesions of putative hypoperfusion in the development of dementia. As Scheinberg (21) and Brun and Englund (22) have so eloquently stated, perhaps we have overlooked a vascular component (? CAA) to the dementia of Alzheimer's disease. Do these deep, white matter lesions in some way modulate the expression of clinical dementia?

Clearly, despite enormous progress in neuropathology, neuropharmacology, and cerebral imaging, we are still a long way from understanding dementing illness. The etiology and pathogenesis of CAA still remain speculative at best. Links between cerebral amyloid angiopathy, dementia, and brain aging will continue to shed light on the development of dementia and raise anew the question of the role of cerebral hypoperfusion in Alzheimer's disease.

FOLLOW-UP

DR KONEN: Fortunately, Ms M. has not suffered any further intracranial bleeding during this past year of follow-up. Nevertheless, she remains cognitively disabled and unable to return to gainful employment.

References

1. Vinters HV: Cerebral amyloid angiopathy--A critical review. Stroke 1987; 18:311-324

2. Kase-Carlos S, Vonsattel JP: Case records of the Massachusetts General Hospital: Case 10-1988: A 73-year-old hypertensive man with intracranial hemorrhage 15 months apart. N Engl J Med 1988; 318:623-631

3. Gilbert JJ, Vinters HV: Cerebral amyloid angiopathy: incidence and complications in the aging brain. I. Cerebral hemorrhage. Stroke 1983; 14:915-923

4. Vinters HV, Miller BL, Pardridge WM: Brain amyloid and Alzheimer disease. Ann intern Med 1988; 109:41-A

5. Strittmatter WJ, Appel SH: Alzheimer's disease. in Rogers DE (ed): 1989 Yearbook of Medicine. Chicago, Year Book Medical, 1989, pp xiii-xxvii

6. Glenner GG: The pathobiology of Alzheimer's disease. Ann Rev Med 1989; 40:45-51

7. Selkoe DJ: Biochemistry of altered brain proteins in Alzheimer's disease. Ann Rev Neurosci 1989; 12:463-490

8. Cosgrove GR, Leblanc R, Meagher-Villemure K, Ethier R: Cerebral amyloid angiopathy. Neurology 1985; 35:625-631

9. Kameyama M: Amyloid angiopathy. in Kameyama M, Tomonaga M, Aiba T (eds): Cerebrovascular Disease. Tokyo, Igaku-Shoin, 1988, pp 105-108

10. Okazaki H, Reagan TJ, Campbell RJ: Clinicopathologic studies of primary cerebral amyloid angiopathy. Mayo Clin Proc 1979; 54: 22-31

11. Mandybur TI: Cerebral amyloid angiopathy: The vascular pathology and complications. J Neuropathol Exp Neurol 1986; 45:79-90

12. Kalyan-Raman VP, Kalyan-Raman K: Cerebral amyloid angiopathy causing intracranial hemorrhage. Ann Neurol 1986; 16:321-329

13. Esiri MM, Wilcock GK: Cerebral amyloid angiopathy in dementia and old age. J Neurol Neurosurg Psychiatry 1986; 49:1221-1226

14. Vinters HV, Gilbert JJ: Cerebral amyloid angiopathy: Incidence and complications in the aging brain. II. The distribution of amyloid vascular changes. Stroke 1983; 14:924-928

15. Gilles C, Brucher JM, Khoubessehan P, Vanderhaegen JJ: Cerebral amyloid angiopathy as a cause of multiple intracerebral hemorrhage. Neurology 1984; 34:730-735

16. Finelli PF, Kessimian N, Bernstein PW: Cerebral amyloid angiopathy manifesting as recurrent intracerebral hemorrhage. Arch Neurol 1984; 41:330-333

17. Patel DV, Hier DB, Thomas CM, Hemmati M: Intracerebral hemorrhage secondary to cerebral amyloid angiopathy. Radiology 1984; 151:397-400

18. Tucker WS, Bilbao JM, Klodawsky H: Cerebral amyloid angiopathy and multiple intracerebral hematomas. Neurosurgery 1980; 7:611-614

19. Wagle WA, Smith TW, Weiner M: Intracerebral hemorrhage caused by cerebral amyloid angiopathy: Radiographic-pathologic correlation. Am J Neuroradiol 1984; 5:171-176

20. Brown RT, Coates RK, Gilbert JJ: Radiographic-pathologic correlation in cerebral amyloid angiopathy: A review of 12 patients. J Assoc Can Radiol 1985; 36:308-311

21. Scheinberg P: Dementia due to vascular disease--A multifactorial disorder. Stroke 1988; 19:1291-1299

22. Brun A, Englund E: A white matter disorder in dementia of the Alzheimer type: A patho-anatomical study. Ann Neurol 1986; 19: 253-259

23. Nichols FT: Leuko-araiosis. In Meyer JS, Lechner H, Marshall J, Toole JR (eds): Vascular and Multi-Infarct Dementia. Mount Kisco, NY, Futura, 1988, pp 212-215

24. Hachinski VC, Potter P, Merskey H: Leuko-araiosis. Arch Neurol 1987; 44:21-22

25. Bogucki A, Papierz W, Szymanska R, Staniaszczyk R: Cerebral amyloid angiopathy with attenuation of the white matter on CT scans: Subcortical arteriosclerotic encephalopathy (Binswanger) in a normotensive patient. J Neurol 1988; 235:435-437

26. Awad IA, Johnson PC, Spetzler RF, et al: Incidental subcortical lesions identified on magnetic resonance imaging in the elderly. II. Postmortem pathological correlations. Stroke 1986; 17:1790-1796

27. Artigas J, Risch W, Perez-Canto A, et al: Lightmicroscopical and ultrastructural findings in diffuse subcortical angiopathy. in CervosNavarro J, Ferszt R (eds): Stroke and Microcirculation. New York, Raven Press, 1987, pp 519-522

28. Roullet E, Baudrimont M, Hauw JJ, et al: Leukoencephalopathy in cerebral amyloid angiopathy: Clinical, radiologic and pathologic study of six cases. Neurology 1988; 38(suppl 1):325. Abstract

29. Gray F, Dabas F, Roullet E, Escourolle R: Leukoencephalopathy in diffuse hemorrhagic cerebral amyloid angiopathy. Ann Neurol 1985; 18:54-59

30, Celsis P, Agniel A, Puel M, et al: Focal cerebral hypoperfusion and selective cognitive deficit in dementia of the Alzheimer type. J Neurol Neurosurg Psychiatry 1987; 50:1602-1612

COPYRIGHT 1990 Dowden Health Media, Inc.
COPYRIGHT 2004 Gale Group