| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Diagnostic Radiology Department, Warren Grant Magnuson Clinical Center,
National Institutes of Health, 10 Center Dr., Bldg. 10, Rm. 1C-660, Bethesda,
MD 20892-1182.
2 HIV & AIDS Malignancy Branch, National Cancer Institute, National
Institutes of Health, Bethesda, MD 20892-1868.
Received February 6, 2002;
accepted after revision April 1, 2002.
Address correspondence to A. D. Patsalides.
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. We conducted a retrospective study of 567 patients (age range, 1 month-29 years; median age, 5.47 years) who acquired HIV as children. Of these, 426 patients (75%) were evaluated with neuroimaging studies. We reviewed these studies to identify the cerebrovascular abnormalities and classify them by type, anatomic location, and shape.
RESULTS. Eleven children (2.6%) were found to have cerebrovascular lesions. Only one had focal neurologic symptoms at the time of diagnosis. Twenty-six aneurysms were found in seven patients, and 27 infarctions were found in eight patients. In four of the patients with infarctions, fusiform aneurysms of the cerebral arteries were also identified. Most patients had advanced HIV disease. Nine of the 11 patients were infected by a vertical transmission route or during blood transfusion early in the neonatal period. In this group of patients, the diagnosis of cerebrovascular disease was made earlier (mean age at diagnosis, 8.2 years) than in the two patients who were infected later in life (mean age at diagnosis, 14.9 years).
CONCLUSION. HIV-infected children have an increased incidence of cerebrovascular disease that is associated with severe immune suppression and with vertically acquired HIV infection or exposure to the virus in the neonatal period. Despite extensive lesions, most children in our study were asymptomatic. Screening with MR imaging should be considered for high-risk children and is advisable when evidence of neurologic symptoms or neurocognitive dysfunction is noted.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Cerebrovascular complications have also been reported in 4-29% of adult patients with AIDS [2,3,4]. The underlying cause of cerebrovascular disease varies widely. Direct involvement of the arterial wall by HIV, cardiogenic emboli, thrombocytopenia, and infectious vasculitis are among the most commonly reported causes [2, 3, 5,6,7,8,9]. Isolated reports of cerebrovascular disease in children with HIV have described aneurysms, infarctions, and hemorrhage [1, 5, 10,11,12,13,14,15].
Our study summarizes the neuroradiologic findings and clinical outcomes in a cohort of pediatric patients evaluated for participation in clinical research trials for HIV disease. The purpose of this article is to report the prevalence of cerebrovascular complications in children infected with HIV, to elucidate the types of vascular involvement, and to examine the possible etiology.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Initial evaluations included physical examination, neuropsychometric testing, and laboratory monitoring including immunologic and virologic surrogate markers (CD4+ T lymphocyte count, HIV-1 viral p24 antigen, and HIV-1 RNA levels). Neuroimaging studies were performed in 426 patients (75%) as part of screening evaluations or protocol-mandated follow-up and for clinical indications. All patients with documented neuroimaging abnormalities were evaluated with repeated imaging at least once annually for several years. CT alone was performed in 243 patients, both CT and MR imaging were performed in 144 patients, and MR imaging alone was performed in 39 patients. One hundred forty-one patients did not undergo imaging because the evaluation protocol did not require it.
CT scans were obtained without contrast material at 5-mm intervals using various models of scanners (General Electric Medical Systems, Milwaukee, WI). T1-weighted MR images were obtained using TR range/TE range, 400-600/8-14, and T2-weighted MR images were obtained using 2000-4000/80-104 with spin-echo pulse sequences in a 1.5-T magnet (Signa; General Electric Medical Systems). In addition, four patients underwent MR angiography with time-of-flight technique; one of these four also underwent conventional angiography. Angiography was performed because of abnormal radiographic findings on CT or MR imaging (n = 3) or because of focal new symptomatology (n = 1).
All CT and MR imaging studies were reviewed by two neuroradiologists who were unaware of the patients' diagnoses, and the detected cerebrovascular abnormalities were classified by consensus. Cerebral lesions were classified as infarctions if they met the established imaging criteria for CT or MR imaging, which include focal abnormalities in the brain parenchyma that evolve over time to a gliotic scar or lacuna. Cerebral aneurysms, when present, were characterized either as fusiform or saccular. The diagnosis of fusiform aneurysms was made if the lumen of an artery was found to be abnormally dilated in comparison with a comparable artery of the opposite hemisphere. Saccular aneurysms were diagnosed by the characteristic focal arterial dilatation connected to the wall of the parent vessel by a neck.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
In seven of the 11 patients with cerebrovascular lesions, HIV disease had been vertically transmitted; four children had acquired disease from blood transfusions. Two of these children were born prematurely and transfused in the neonatal period. Seven patients died during the study's observation period. Only one death was attributed to central nervous system complications; this patient is the only one for whom autopsy findings of the central nervous system are available [11].
Aneurysms were documented in seven of the 11 patients. In six of the seven patients, aneurysms of the cerebral arteries were detected on both CT and MR imaging. The seventh patient had a history of a saccular aneurysm in the supraclinoid segment of the left internal carotid artery; it had been surgically clipped before the patient was evaluated at our institution. No details were available regarding the decision for that surgery. The patient's neurologic examination was nonfocal and remained unchanged after surgery. CT performed at the National Institutes of Health revealed the presence of a surgical clip at the site of the aneurysm.
In the 11 patients with aneurysms, 25 aneurysms were revealed on MR imaging, but only 15 were seen on CT. Twenty aneurysms involved the anterior and six the posterior circulation. Six aneurysms involved the supraclinoid segment of the internal carotid arteries, eight the proximal middle arteries, and six the proximal anterior cerebral arteries. Two aneurysms were found in the proximal posterior cerebral, two in the basilar, and two in the vertebral arteries. The proximal segment of the middle and anterior cerebral arteries and the supraclinoid segment of the internal carotid arteries were most commonly affected. Twenty-four aneurysms (92%) were fusiform, and two, including the surgically treated aneurysm, were saccular. Four of the seven patients had multiple aneurysms (all fusiform); two of these patients had eight aneurysms each. In four patients, abnormal calcifications were identified in the wall of at least one aneurysm. These calcifications were shown only on CT and appeared identical to atherosclerotic plaques. None of the patients experienced clinically significant hemorrhage, subarachnoid or other, in association with the aneurysms. In only one of the patients was there evidence of hydrocephalus; it was caused by partial obstruction of the foramina of Monro by the aneurysms of the middle cerebral arteries during the late stage of the disease.
Ischemic infarctions were identified in eight (73%) of the 11 patients with cerebrovascular lesions. In four of these patients, fusiform aneurysms were also present, whereas the remaining four patients had no other associated vascular abnormalities. A total of 27 infarctions were documented on MR imaging; CT showed 20 of them. Seven infarctions involved the cortical gray matter, and the remaining 20 were in the basal ganglia (Figs. 1A,1B,1C and 2A,2B). Sixteen infarctions (16/27; 59%) were located in vascular territories harboring a fusiform aneurysm. In two patients, the presence of aneurysms was documented at least 6 months before the infarctions were found.
|
|
|
|
|
In these 11 index patients with cerebrovascular complications, systemic infections other than HIV were documented by clinical history and diagnostic laboratory testing during the observation period. The most common infection was varicella-zoster virus, which was often recurrent and was documented in six patients (55%). Three of the patients with varicella-zoster virus had infarctions only, two developed aneurysms, and one patient had both. Two of the eight patients with infarctions had hematologic abnormalities at the time of infarction; one had protein C and S deficiencies and the other had protein S deficiency only.
Autopsy data of the cerebral aneurysms were available in one patient [11]. Sections of the walls of the aneurysms revealed intimal hyperplasia, fibrosis, microcalcifications, loss of muscularis, and fragmentation or thinning of the elastic membrane. Infiltration of the walls of affected vessels by focal collections of mononuclear cells was also documented.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Two types of cerebrovascular lesions were found in this study; aneurysms and ischemic infarctions. The mechanism by which HIV results in arterial damage leading to aneurysmal dilatation is largely unknown. One theory is that HIV is directly implicated in the pathogenesis of arteriopathy. Park et al. [5], using a monoclonal antibody to gp41, identified HIV viral antigen in the intima of the affected arteries and postulated that the HIV infection of parenchymal and leptomeningeal vessels may be involved in vascular complications in children with AIDS. In support of this theory are the findings reported by Husson et al. [11], that cerebral aneurysms in children with AIDS developed during a period of significant HIV replication as indicated by high serum p24 antigen levels. Furthermore, Mazzoni et al. [12] suggested a possible arrest in the progression of arteriopathy after the beginning of antiretroviral therapy that decreased the viral load. These observations are supported by the characteristics of the affected patients in our cohort: most had advanced-stage HIV disease and, despite antiretroviral treatment, had evidence of substantial viral replication at the time their cerebrovascular abnormality was diagnosed.
A number of investigators have suggested that other viral agents, acting alone or in synergy with HIV, may be involved in the pathogenesis of arteriopathy in pediatric HIV infection. More specifically, varicella-zoster virus has been implicated in central nervous system arteriopathy [8, 10, 20,21,22,23]. Dubrovsky et al. [10] reported either a history of varicella-zoster virus infection or elevated varicella-zoster virus antibody titers in HIV-infected children with cerebral aneurysms. In our study, two of the seven patients with aneurysms, three of the eight patients with infarction, and one patient with both aneurysms and infarction had documented evidence of varicella-zoster virus infection.
Regardless of the causative agent responsible for the vascular damage, a number of pathogenic mechanisms of arterial wall damage have been proposed. According to earlier reports [6, 24], the inflammatory reaction begins in the adventitia and involves the vasa vasorum, leading to ischemia of the arterial wall. Other proposed mechanisms include either exposure to toxic cytokines produced in the blood or in the central nervous system [25] or transendothelial migration of HIV-infected monocytes [26]. The end product of the vascular damage shows medial fibrosis, loss of muscularis, fragmentation of the elastic lamina, and intimal hyperplasia as the most common histologic findings on microscopic examination.
Most of the infarctions in our study occurred in vascular territories harboring aneurysms. We therefore surmise that these infarctions are causally connected to the aneurysms. The lenticulostriate and thalamoperforating arteries of the affected areas had their origin in segments of the anterior, middle, and posterior cerebral arteries that were affected by fusiform aneurysms. We presume that the orifice of these vessels was compromised or occluded either by distortion of the wall anatomy of the ectatic parent artery or by mural thrombi. A second mechanism of cortical ischemic infarction in this group of patients is the lodging of blood clots formed on the wall of the abnormally dilated arteries in peripheral branches of these arteries.
The mechanism of the ischemic infarctions that were not associated with cerebral aneurysmal arteriopathy is difficult to define. It is possible that the same agents that affect the larger vessels in the region of the circle of Willis and cause aneurysms also affect the smaller peripheral branches of the cerebral arteries, leading to occlusion and subsequent infarction of the brain parenchyma supplied by these vessels. The deficiency in protein S found in two patients and in protein C in one patient is a well-recognized anticoagulant abnormality that is an important risk factor for thrombotic complications in HIV-infected children [27, 28]. Katsetos et al. [29] described angiocentric perivascular and transmural CD3 T-cell infiltrates in brain specimens of HIV-infected infants and children. This angiocentric T-cell response could be attributed to an autoimmune reaction. It seems also probable that thromboembolic episodes may account for some of these ischemic lesions from embolic material whose origin is in the cardiopulmonary circulation.
Hemorrhagic infarctions due to thrombocytopenia have been reported in pediatric patients with AIDS; they present clinically with focal neurologic deficits consistent with a clinical diagnosis of stroke [5]. This complication was not documented in any of the affected patients we examined. The only patient in our cohort who presented with focal neurologic deficits had thrombotic infarction associated with protein C and S deficiencies.
In our cohort, multiple aneurysms were common (4/7 patients; 57%), whereas the frequency of multiple aneurysms in children who are not HIV-infected is 2% [18]. Another important observation is that cerebrovascular complications are associated with perinatal HIV infection; nine of 11 patients in our study were infected by vertical transmission or transfusion during the neonatal period. In this group of patients, the diagnosis of cerebrovascular complications was made earlier (mean age, 8.2 years) compared with the two patients infected later in life (mean age, 14.9 years). It is possible that immature fetal or neonatal vessels may be more vulnerable to HIV infection or other infectious agents. Consequently, prophylactic antiretoviral treatment of HIV-infected pregnant women may not only offer long-term control of viral replication via highly active antiretroviral treatment but may also provide another line of defense against these cerebrovascular complications in their children.
Most cerebrovascular complications observed in this cohort occurred before protease inhibitors were available for treatment and before the implementation of highly active antiretroviral treatment as the standard of care in the pediatric HIV-infected population. Although two of the 11 patients with cerebrovascular lesions in this cohort had received highly active antiretroviral treatment (mean duration before cerebrovascular diagnosis of 17.5 months), both had persistent viral replication (data not shown). Although researchers in the United States have observed a decline in the incidence of cerebrovascular disease in recent years, it is important to emphasize that this complication may contribute to increased patient morbidity and mortality in developing countries where antiretroviral prophylaxis and treatment are not readily available.
In conclusion, our study has shown that increased incidence of cerebrovascular disease is present in children infected with HIV. Vascular immaturity is suggested as a possible contributory factor, based on the observation that most index patients had acquired HIV vertically or in the neonatal period. Because most of these patients are asymptomatic during the early stages of the disease, screening of high-risk children—preferably by MR imaging—is advisable for the early detection of cerebrovascular abnormalities. Prophylaxis with antiretroviral medication could prevent entirely or diminish the incidence and possibly the severity of cerebral vasculopathy.
Acknowledgments
We acknowledge Freda Jacobsen and Judy Zuckerman for their assistance in
the preparation of this article.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  P. O'Charoen, J.R. Hesselink, and J.F. Healy Cerebral Aneurysmal Arteriopathy in an Adult Patient with Acquired Immunodeficiency Syndrome AJNR Am. J. Neuroradiol., May 1, 2007; 28(5): 938 - 939. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Tipping, L. de Villiers, S. Candy, and H. Wainwright Stroke Caused by Human Immunodeficiency Virus-Associated Intracranial Large-Vessel Aneurysmal Vasculopathy Arch Neurol, November 1, 2006; 63(11): 1640 - 1642. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. D. Corr Imaging of cerebrovascular and cardiovascular disease in AIDS patients. Am. J. Roentgenol., July 1, 2006; 187(1): 236 - 241. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
