|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
© 2000 American Society for Clinical Oncology Silent Lacunar Lesions Detected by Magnetic Resonance Imaging of Children With Brain Tumors: A Late Sequela of TherapyFrom the St Jude Children’s Research Hospital/Le Bonheur Children’s Medical Center Brain Tumor Team; Departments of Hematology-Oncology, Diagnostic Imaging, Behavioral Medicine, Biostatistics/Epidemiology, and Radiation Oncology, St Jude Children’s Research Hospital; and Departments of Pediatrics and Radiology, University of Tennessee, Memphis, TN. Address reprint requests to Amar Gajjar, MD, Department of Hematology-Oncology, St Jude Children’s Research Hospital, 332 North Lauderdale, Memphis, TN 38105-2794; email amar.gajjar@ stjude.org.
BACKGROUND: Cerebral lacunes, which generally appear on magnetic resonance imaging as foci of white matter loss, usually occur in adults after ischemic infarcts. We report the development of lacunes in children after therapy for brain tumors. PATIENTS AND METHODS: We reviewed the clinical characteristics and radiologic studies of 524 consecutive children with brain tumors treated over a 10-year period. We documented the neuropsychologic findings associated with lacunes and the factors predictive of lacunar development. RESULTS: Lacunes developed in none of the 103 patients observed or treated with surgery alone. Twenty-five of the 421 patients treated with chemotherapy or radiation therapy or both had lacunes. Patients were a median of 4.5 years old at the time of both diagnosis (range, 0.3 to 19.8 years) and radiotherapy (range, 1.5 to 20 years). Fourteen patients were treated with craniospinal irradiation, and 11 were treated with local radiotherapy. The median time from radiotherapy to the appearance of lacunes was 2.01 years (range, 0.26 to 5.7 years). For all patients, lacunes were an incidental finding with no corresponding clinical deficits. The factor most predictive of lacunar development was age less than 5 years at the time of radiotherapy (P = .010). There was no significant difference in estimated decline in intelligence quotient scores between patients with lacunes and age and diagnosis-matched controls. CONCLUSION: Lacunes may be caused by therapy-induced vasculopathy in children with brain tumors, with the most significant predictor being age less than 5 years at the time of radiotherapy.
RADIOTHERAPY IS integral to the treatment of children with brain tumors but can damage normal CNS structures. Potential side effects are of particular significance in the treatment of children, for whom radiotherapy to the developing CNS may have devastating consequences, including a significant decline in neuropsychologic status.1,2 As a result of more effective therapeutic regimens and improved survival rates for children with primary CNS tumors, the spectrum of clinically and radiologically apparent treatment-related sequelae are emerging. Late sequelae, such as cerebral atrophy, white matter changes, and vasculopathy, have been documented in addition to the classical subacute sequelae, such as radiation necrosis.3 CNS lacunes are a well-known pathologic entity that were first described by Durand-Fardel in 1842.4 They are believed to represent ischemic infarcts of restricted size in the deeper parts of the brain; these infarcts form irregular cavities usually between 0.2 to 15 mm in diameter.5 Such lesions are relatively common in patients with chronic hypertension and diabetes mellitus; in these patients, lacunes are typically seen in the basal ganglia, pons, and deep white matter of the brain.6 They may be incidental findings or may be associated with a broad array of symptoms, including pure sensory stroke, pure motor hemiparesis, hemiballismus, hemifacial spasm, dementia, and other deficits.4-7 Cerebral lacunes have not been previously described in children with CNS tumors. Stimulated by the incidental detection of such lesions in a small number of patients during routine neuroimaging studies after treatment for brain tumors, we undertook a broader investigation of this phenomenon. Here, we present a review of the clinical, radiologic, and neuropsychologic findings and the clinical significance and neuropsychologic sequelae of lacunar lesions in children with CNS tumors.
Study Design and Population Between January 1986 and November 1996, 534 patients with brain tumors were referred to St Jude Children’s Research Hospital. Of these, 103 patients were either observed (n = 21) or underwent surgery only (n = 82); the remaining 421 patients were given radiotherapy as part of their treatment. Patients were generally enrolled on protocols approved by the institutional review board. Informed consent was obtained from patients, parents, or guardians, as appropriate, at the time of protocol enrollment. The serial neuroimaging of these 534 patients were retrospectively reviewed for the presence of lacunes. The clinical characteristics, treatment, and outcomes of the 25 patients with foci of lacunar white matter loss were then reviewed. All patients underwent either computerized tomography (CT) or magnetic resonance imaging (MRI) at the time of diagnosis and during routine follow-up examinations. A neuroradiologist (J.L.) and a neuro-oncologist (A.G.) concurrently reviewed the neuroimaging studies of all patients to describe the presence of lacunes and the evolution from the time of their first onset. Lacunes (Figs 1 and 2) were defined as small cavities within the white matter of the brain with characteristics similar to those of cerebrospinal fluid on T1- and T2-weighted images. The number and distribution of the lacunes and the size of each were recorded. Size was defined as the product of the largest perpendicular diameter of the lacunes in two dimensions. The cumulative size at the time of each serial imaging evaluation was calculated by adding the size of all observed lacunes.
Our institutional standard of care includes annual neuropsychologic testing for all English-speaking patients capable of participating until at least 5 years after the completion of therapy. The current analysis was restricted to intelligence quotient (IQ) testing, with standardized, individually administered, age-appropriate instruments with a normative IQ of 100 and a SD of 15 or 16.8-12
Statistical Analysis A mixed-effect model for longitudinal data with random coefficients was used to compare the rate of decrease in IQ scores between patients with and without lacunes in age at radiotherapy–matched and diagnosis-matched controls. The controls were selected from all patients with CNS tumors irradiated between January 1986 and November 1996. To increase the sample size, we considered patients’ ages at the time of radiotherapy to be matched if they differed by less than 6 months, except in two patients whose ages differed by less that 12 months. Six patients with lacunes were excluded from comparison for IQ with the control group because either no IQ data were available (n = 2), no matched control could be found (n = 2), or only one IQ measurement was available (n = 2). If more than one matched control was available for a patient, we calculated the time interval between the last IQ examination for the patient and that of each control and selected the control for which that time interval was the shortest. Two mixed-effect models for longitudinal data with random coefficients were used to estimate the increase in the cumulative size of the lacunes and the decrease in IQ score per year for each patient studied. We used Pearson’s correlation coefficient to evaluate the relationship between the estimated increase in cumulative size and the estimated decrease in IQ score. The random coefficient model analyses were implemented by using PROC MIXED (SAS Institute Inc, Cary, NC).15
No lacunes developed in any of the 103 patients who were either observed by serial scans or treated with surgery alone. The remaining 421 patients who were treated with radiation as part of their therapy form the subset that includes 25 patients with lacunes. The clinical characteristics of the study groups are listed in Table 1.
Patient Characteristics The histologic diagnoses of the 25 patients included medulloblastoma (nine patients), supratentorial primitive neuroectodermal tumor (four patients), ependymoma (two patients), grade 1 or 2 astrocytoma (six patients), and anaplastic astrocytoma (four patients). The median age at the time of diagnosis was 4.5 years (range, 0.3 to 19.8 years). All patients were treated with radiotherapy at a median age of 4.5 years (range, 1.5 to 20 years). Fourteen patients were treated with craniospinal irradiation at a median dose of 55 Gy (range, 48 to 66 Gy) to the posterior fossa and 35 Gy (range, 24 to 55 Gy) to the craniospinal axis. The eleven other patients were treated with local radiotherapy to the tumor (median dose, 54 Gy; range, 50 to 70.2 Gy). Sixteen patients were given chemotherapy (details not presented) before irradiation; two were given additional chemotherapy after irradiation. One patient was treated with chemotherapy at the time of progressive disease after radiation therapy. Most chemotherapy protocols contained an alkylating agent (cyclophosphamide) and a platinating agent (cisplatin/carboplatin). The median time to the first appearance of lacunes after radiotherapy was 2.01 years (range, 0.26 to 5.7 years). Except for one lesion that developed at the junction of the pons and the middle cerebellar peduncle of a patient with medulloblastoma, all lacunes occurred in the white matter of the cerebral hemispheres. At the time of the first appearance of lacunes, the mean number of lacunes per patients was 1.75, and the mean cumulative size was 17.7 mm2. Five years after the observation of the first lacune, the mean number of lacunes had increased to eight per patient, and the mean cumulative size had increased to 161.5 mm2. These findings demonstrate the progressive nature of the lesions (Fig 3). The median number of lacunes per patient was eight (range, one to 84; Fig 4). In 14 of the 21 patients (67%) with multiple lacunes, at least 50% of the lesions first appeared simultaneously. Lacunes ranged in size from 1 mm2 x 1 mm2 to 10 mm2 x 24 mm2. Forty percent of the lacunes did not change in size; 9% of the lacunes had increased in individual size by more than 30 mm2 (Fig 5).
The factor most predictive of the development of lacunes was age at the initiation of radiotherapy. The 5-year cumulative incidence (mean ± SE) of developing lacunes was 11.9% ± 3% for children 5 years of age or younger at the time of radiotherapy and 4.2% ± 1% for children older than 5 years of age (P = .010; Fig 6). Other factors such as sex (P = .38), histologic diagnosis (P = .30), and type (ie, local or craniospinal) of radiotherapy (P = .60) were not predictive of lacunar development. Because most patients received both radiation therapy and chemotherapy, the potential impact of the later modality in the development of lacunes could not be ascertained. Likewise, the similarity in radiation therapy doses between index cases and controls precluded a dose effect analysis.
Neuropsychologic Testing IQ scores were observed and predicted for patients and controls. The median duration of the interval from initiation of radiotherapy to last IQ examination was 5.1 years for the patients (range, 3.3 to 8.4 years) and 5.0 years for the controls (range, 1.6 to 11.8 years). The median number of IQ examinations was four for the patients (range, two to eight) and five for the controls (range, two to seven). The projected IQ score at the beginning of therapy was 83.7 for the patients with lacunes and 89.5 for the controls (P = .28; Fig 7). Both scores were below normal expectations for children of the same ages. A longitudinal analysis detected no significant difference in the mean rate of decline in IQ scores between the patients with lacunes and the matched controls (P = .20). The estimated rates of decline (1.8 units per year for the patients with lacunes and 2.7 units per year for the controls) are significantly different from zero (P < .001) and suggest a substantial and similar decrease in IQ score for both groups of patients.
Although for many patients the size of the individual lesions did not change, the increase in the number of lesions over time led to an increase in the cumulative size (Fig 3). Although the mean cumulative size increased over time, Pearson’s correlation coefficient suggests no significant relationship between the estimated change in cumulative size and the estimated decline in IQ (r = -0.22, P = .38).
CNS lacunes were originally believed to be the end result of focal hemorrhages.16 Three types of lacunar lesions have been described: small infarcts (type 1), small hemorrhages (type 2), and dilated perivascular spaces (type 3).17 Among these, type 1 lacunes are the most common. Detailed histologic studies of brains of stroke patients by Fisher et al5 established occlusive microvascular disease as the cause of the lacunes in these patients. Other factors implicated in the pathogenesis of lacunes include: vascular spasm,18-20 focal brain edema,21,22 emboli,20,23,24 arteriopathies such as microatheromas,17,25 and lipohyalinosis5 (which is believed to represent hypertensive cerebral vasculopathy in which the lumina of smaller arteries are occluded). Most lacunes occur in the territories of the lenticulostriate branches of the anterior and middle cerebral arteries, the thalamoperforating branches of the posterior cerebral arteries, and the paramedian branches of the basilar artery.5 Although late therapy-induced changes leading to vasculopathy is the inferred pathogenic mechanism for the development of these lesions in our patient population, no correlative histology has been obtained. In other studies reporting similar lesions in adults, cerebral lacunes are described as small cavities with ragged margins that remain after macrophage removal of the necrotic tissue caused by a local infarct. They are principally found (in descending order of frequency) in the putamen, the caudate nucleus, the thalamus, the pons, and the internal capsule.6 Occasionally they occur in the deep white matter of the cerebral gyri.3 Although all cerebral infarcts resolve by the process of cavitation, most are superficial and result in poorly visible depressions at the surface of the brain. Only those infarcts confined to the interior of the brain result in a visible lacune. Radiologically, lacunes appear on CT images as small, low-density, sharply marginated areas. On T2-weighted MRI images, lacunes are hyperintense white matter lesions that correspond to low-intensity foci seen on T1-weighted images. Comparison of CT and MRI images of patients with a clinical diagnosis of a lacunar stroke found that MRI images are more sensitive than CT images in demonstrating infarction.26 A study of MRI and cerebral blood flow in neurologically normal adults with silent lacunar lesions demonstrated the sensitivity of MRI and the clear relationship between silent lacunar lesions and a decrease in cerebral circulation.27 In our series, 14 of the 21 patients had multiple lacunes, and more than 50% of the lacunes appeared simultaneously. For all patients, most of the new lacunes occurred within a short interval after the occurrence of the first lacune (median, 11 months; range, 5 to 72 months). Interestingly, in some patients some lacunes increased in size and others did not, suggesting that a pathophysiology of unknown origin may affect the further loss of brain tissue. The clinical definition of a lacunar syndrome has varied widely. As first described by Pierre Marie in relation to capsular infarcts detected at the time of postmortem examination of 50 chronically ill patients, the syndrome consisted of sudden incomplete hemiparesis followed by good recovery.4 Since then, Fisher6 has described more than 20 lacunar syndromes, including pure motor hemiplegia (most common), pure sensory stroke, homolateral hemiplegia, dysarthria-clumsy hand syndrome, sensory-motor stroke, dementia, and many others, most of which occur in elderly patients with hypertension and diabetes. Interestingly, none of our patients had these or other new neurologic signs or symptoms between the completion of therapy and the time of detection of these lesions by MRI. These lacunar lesions were clinically silent. In adults, the appearance of white matter lesions on CT images has correlated with cognitive impairment and development of dementia in patients who have suffered a clinical stroke.28-31 However, similar changes seen on MRI have been below the level of clinical significance.32-34 In our series, IQ scores were determined for patients both before and after the first lacune was observed on MRI. Although we initially hypothesized that, for these patients, IQ scores might decrease slightly before the first lacune was detected and decline sharply thereafter, this hypothesis was not confirmed. The IQ profiles did not change in relation to the appearance of lacunes. However, these patients may still be at risk for the early onset of neurodegenerative disease. We considered factors that might predict the development of lacunes in our patients. In a series of adult patients with symptomatic lacunar infarcts, confirmed by CT or MRI, the predictors of death were age, diabetes mellitus, and smoking.35 The 4-year survival estimate was 90% ± 4%. No predictors of progression of lacunar defects were found and the authors concluded that the prognosis associated with lacunar infarcts was relatively favorable. In our population, neither sex nor tumor histology predicted the development of lacunes. Age less than 5 years at the time of radiotherapy, however, was a significant predictor (P = .010). Although in all patients the sites of the lacunes correlated with the sites of radiotherapy, the volume of brain irradiated did not predict the development of lacunes. Even though chemotherapy was used as a part of the combined-modality therapy, the role of chemotherapy or its interaction with irradiation in the pathogenesis of lacunes cannot be clearly delineated in our series. Young children with malignant brain tumors represent the population at greatest risk for lacunar development. Current treatment approaches use combined therapies, limiting the ability to assess any potentiating impact of chemotherapy in conjunction with irradiation. Radiation-induced vasculopathy may involve large or small vessels. Delayed capillary endothelial damage leads to breakdown of the blood-brain barrier, vasogenic edema, endothelial hyperplasia, and fibrinoid necrosis of penetrating arterioles in the white matter of the CNS.36 The changes in vessels are not commonly seen during intervals less than 6 months after radiotherapy. The time frame is consistent with the hypothesis that radiation-induced vasculopathy may be responsible for the development of lacunes in our patients, in whom the median time between the end of radiotherapy and the development of lacunes was 2 years. In conclusion, this is the first report of the development of lacunes as detected by MRI in pediatric brain tumor patients after radiotherapy. We hypothesize that late radiation-induced vasculopathy may be the pathogenic mechanism for the development of lacunes in these patients. Although we have not yet detected any clinical manifestations associated with these lesions, careful long-term follow-up is warranted because these children may be predisposed to the onset of cerebrovascular and neurodegenerative disease.
Supported in part by grants no. P30 CA21765 and P01 CA23099 from the National Cancer Institute and by the American Lebanese Syrian Associated Charities (ALSAC). We thank Jennifer Havens and Annemarie Fraga for assistance with the data collection, Dr James Boyett for assistance with the statistical analysis, Flo Witte for editing the manuscript, and Patsy Burnside for typing the manuscript.
1. Mulhern RK, Kepner JL, Thomas PR, et al: Neuropsychologic functioning of survivors of childhood medulloblastoma randomized to receive conventional or reduced dose craniospinal irradiation: A Pediatric Oncology Group Study. J Clin Oncol 16:1723-1728, 1998[Abstract]
2.
Silber JH, Radcliffe J, Peckham V, et al: Whole-brain irradiation and decline in intelligence: The influence of dose and age on IQ score. J Clin Oncol 10:1390-1396, 1992 3. Phillips TL: Early and late effects of radiation on normal tissues, in Gutin PH, Leibel SA, Sheline GE (eds): Radiation Injury to the Nervous System. New York, NY,Raven Press Ltd, 1991, pp 37-55 4. Fisher CM: Lacunes: Small, deep cerebral infarcts. Neurology 15:774-784, 1965 5. Fisher CM: The arterial lesions underlying lacunes. Acta Neuropath (Berl) 12:1-15, 1969
6.
Fisher CM: Lacunar strokes and infarcts: A review. Neurology 32:871-876, 1982
7.
Kawakami M, Sato T, Tochigi S, et al: Lacunar pontine infarct with hemifacial spasm as initial symptom. Stroke 21:1236, 1990 (letter) 8. Bayley N: Bayley Scales of Infant Development. San Antonio, TX,The Psychological Corporation, 1969 9. Wechsler D: Wechsler Preschool and Primary Scale of Intelligence-Revised. San Antonio, TX,The Psychological Corporation, 1989 10. Wechsler D: Wechsler Intelligence Scale for Children-Revised. San Antonio, TX,The Psychological Corporation, 1974 11. Wechsler D: Wechsler Intelligence Scale for Children (ed 3). San Antonio, TX,The Psychological Corporation, 1991 12. Wechsler D: Wechsler Adult Intelligence Scale (revised). San Antonio, TX,The Psychological Corporation, 1981 13. Kalbfleisch JD, Prentice RL: The Statistical Analysis of Failure Time Data. New York, NY,Wiley, 1980, p 169 14. Gray RJ: A class of K-sample tests for comparing the cumulative incidence of a competing risk. Ann Stat 16:1141-1154, 1988 15. SAS Institute Inc:SAS/STAT Software: Changes and Enhancements Through Release. Cary, NC,SAS Institute Inc, 1996, pp 531-657
16.
Cole FM, Yates PO: Comparative incidence of cerebrovascular lesions in normotensive and hypertensive patients. Neurology 18:255-259, 1968 17. Poirier J, Gray F, Gheardi R, et al: Cerebral lacunae: A new neuropathological classification. J Neuropathol Exp Neurol 44:3, 1985 (abstr 312) 18. Ogata J, Fujishima M, Takami K, et al: Stroke-prone spontaneously hypertensive rats as an experimental model of malignant hypertension. Acta Neuropathol (Berl) 51:179-184, 1980[Medline] 19. Ogata J, Fujishima M, Takami K, et al: Vascular changes underlying cerebral lesions in stroke-prone spontaneously hypertensive rats: A serial study. Acta Neuropathol (Berl) 54:183-188, 1981[Medline]
20.
Millikan C, Futrell N: The fallacy of the lacune hypothesis. Stroke 21:1251-1257, 1990 21. Johansson BB: Hypertension and the blood brain barrier, in Neuwelt EA (ed): Implications of the Blood Brain Barrier and its Manipulations. New York, NY,Plenum Medical Book Company, 1989, pp 389-409 22. Johansson BB: Vascular mechanism in hypertensive cerebrovascular disease. Cardiovascular Pharmacol 19:S11–S15, 1992 (suppl 3) 23. Fisher CM: Bilateral occlusion of basilar artery branches. Psychiatry 40:1182-1189, 1977 24. Fisher CM: Capsular infarcts: The underlying vascular lesions. Arch Neurol 36:65-73, 1979[Abstract] 25. Poirier J: Giant cerebral lacuna due to dilatation of the perivascular space: A case report. Clin Neuropathol 2:138-140, 1983[Medline]
26.
Rothrock JF, Lyden PD, Hesselink JR, et al: Brain magnetic resonance imaging in the evaluation of lacunar stroke. Strok e18:781-786, 1987
27.
Kobayashi S, Kazunori O, Yamashita K: Incidence of silent lacunar lesion in normal adults and its relation to cerebral blood flow and risk factors. Stroke 22:1379-1383, 1991 28. Kertesz A, Polk M, Carr T: Cognition and white matter changes on magnetic resonance imaging in dementia. Arch Neurol 47:387-391, 1990[Abstract] 29. Lee A, Yu YL, Tsoi M, et al: Subcortical arteriosclerotic encephalopathy: A controlled psychometric study. Clin Neurol Neurosurg 91:235-241, 1989[Medline] 30. Steingart A, Hachinski VC, Lau C, et al: Cognitive and neurologic findings in demented patients with white matter lucencies on computed tomographic scans (leuko-araiosis). Arch Neurol 44:36-39, 1987[Abstract] 31. Steingart A, Hachinski VC, Lau C, et al: Cognitive and neurologic findings in subjects with diffuse white matter lucencies on computed tomographic scans (leuko-araiosis). Arch Neurol 44:32-35, 1987[Abstract] 32. Hendrie HC, Farlow MR, Austrom MG, et al: Foci of increased T2 intensity on brain MR scans of healthy elderly subjects. Neuroradiol 10:703-707, 1989
33.
Hunt AL, Orrison WW, Yeo RA, et al: Clinical significance of MRI white matter lesions in the elderly. Neurology 39:1470-1474, 1989 34. Van Sweiten JC, Staal S, Kappelle LJ, et al: Are white matter lesions directly associated with cognitive impairment in patients with lacunar infarcts? J Neurol 243:196-200, 1996[Medline] 35. Clavier I, Hommel M, Besson G, et al: Long-term prognosis of symptomatic lacunar infarcts: A hospital-based study. Stroke 25:2005-2009, 1994[Abstract] 36. Baker DG, Krochak RJ: The response of the microvascular system to radiation: A review. Cancer Invest 7:287-294, 1989[Medline] Submitted May 19, 1999; accepted September 23, 1999. This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2000 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
TOP
ABSTRACT
INTRODUCTION
