| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Pediatric Imaging |
1 From the Department of Radiology, Kanagawa Childrens’ Medical Center, Yokohama, Japan (S.K., N.A., M.H., Y.O., K.F.); and Department of Radiology, Yokohama City University School of Medicine, 3–9 Fukuura Kanazawa-ku, Yokohama 236-0004, Japan (S.K., N.A., M.H., Y.O., T.I.). Received September 6, 2002; revision requested November 18; final revision received April 14, 2003; accepted June 10. Address correspondence to S.K. (e-mail: s-koike@ga2.so-net.ne.jp).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
MATERIALS AND METHODS: The authors identified 90 children who had undergone cranial irradiation between 1981 and 2001 and undergone magnetic resonance (MR) imaging with follow-up for at least 6 months. Patients were assigned to low-dose (LD) and high-dose (HD) groups. All 24 children in the LD group received a radiation dose of 18.0 or 19.8 Gy. The 66 patients in the HD group received a dose of 32.0 Gy or greater. Telangiectasias were defined as small low-signal-intensity foci on intermediate- or T2-weighted MR images. For the patients who underwent serial MR imaging, the first depicted appearance of each telangiectatic lesion was recorded. Statistical analyses were performed.
RESULTS: Telangiectasias in at least one area were observed in 18 (20%) patients. The frequency of telangiectasia was 13% (three of 24 patients) in the LD group as compared with 23% (15 of 66 patients) in the HD group; this difference was not significant (P = .22, Fisher exact test). In 12 patients (one from LD and 11 from HD group) who underwent serial MR imaging follow-up for up to 10 years (mean, 8.1 years), a total of 31 lesions were detected. Twelve (39%) of these lesions were detected by the 3rd year, and 21 (68%) were evident by the 5th year. Six (50%) of the 12 patients who underwent serial MR imaging had telangiectatic foci after 5 years.
CONCLUSION: Radiation-induced telangiectasia appears to occur in at least 20% of children who undergo cranial irradiation. In this small series, higher radiation dose was not significantly associated with higher frequency of telangiectasia, although there was a trend in this direction.
© RSNA, 2003
Index terms: Brain, MR, 13.121411, 13.121416 • Radiations, injurious effects, complications of therapeutic radiology, 13.47 • Telangiectasia, 13.47, 13.759 • Therapeutic radiology, in infants and children
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
Diffuse white matter injury is secondary to radiation-induced demyelination (2,4,5). Radiation-induced vascular injury also is often seen and can be classified into two types: large arterial injury and cryptic vascular malformation, or telangiectasia (6). Large arterial injury is caused by atherosclerosis and thrombosis of large arteries such as the internalcarotid artery and the middle cerebral artery (3,7). On the other hand, radiation-induced telangiectasia, which has been referred to as cryptic vascular malformation, is one of the microangiopathies reported by Gaensler et al (6). They concluded that radiation-induced telangiectasia in the brain results in varying amounts of hemorrhage and may have an appearance similar to that of cryptic vascular malformation (6,8). Although occult venous malformations also have been termed telangiectasias, we refer to these conditions as radiation-induced telangiectasias (9).
The purpose of the present study was to determine the frequency, dose relation, and latency of radiation-induced telangiectasias in children following cranial irradiation.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
The final study group consisted of 90 children—50 boys and 40 girls (age range at initial irradiation, 0–17 years; mean age, 7.2 years ± 4.5 [SD])—who underwent follow-up brain MR imaging at least 6 months after irradiation. The pathologic or clinical diagnoses included 25 ALL or AML tumors, 30 gliomas, 13 germ cell tumors, eight medulloblastomas, and 14 other tumors (ie, metastasis, primary central nervous system lymphoma, and rare tumors). We assigned the 90 patients to a low-dose (LD) group or a high-dose (HD) group on the basis of the radiation dose administered. The LD group consisted of 24 patients who underwent irradiation of 18.0 or 19.8 Gy; the remaining 66 children received at least 32.0 Gy of radiation and were assigned to the HD group.
All 24 patients in the LD group received a diagnosis of ALL or AML. As part of the treatment protocol, these patients received preventative radiation therapy of 18.0 Gy (for ALL) or 19.8 Gy (for AML). The mean radiation dose administered to the LD group was 18.07 Gy ± 0.36 (SD). The mean age of these patients at initial irradiation was 9.8 years ± 3.6, and the mean follow-up period was 2.3 years ± 1.9.
Sixty-five of the 66 patients in the HD group received a diagnosis of central nervous system neoplasms and underwent greater than 36.0 Gy of irradiation. One patient with ALL underwent 18 Gy of whole-brain irradiation plus 14.0 Gy of whole-body irradiation before bone marrow transplantation, and we included this patient in the HD group. For the HD group, the mean radiation dose administered was 53.54 Gy ± 7.97, the mean age at initial irradiation was 6.3 years ± 4.4, and the mean follow-up period was 6.1 years ± 3.9.
Brain Radiation Therapy
LD radiation therapy for preventative irradiation was administered across the entire brain in 180-cGy fractions for 10 days. In the patient with ALL in the HD group, additional radiation therapy for pretreatment of bone marrow transplantation was administered across the entire body twice in 200-cGy fractions for 3 days. Thirty-nine patients in the HD group underwent whole-brain irradiation with or without a local boost, and 27 patients with astrocytomas (grade, <3) and craniopharyngiomas underwent local radiation therapy.
MR Imaging
All analyzed MR images were obtained by using 1.5-T MR imaging systems (FX III or Visart; Toshiba Medical Systems, Tokyo, Japan). Transverse T2- and intermediate-weighted spin-echo (dual echo, 2,000–3,000/30, 80–100 [repetition time msec/echo time msec], 6.5–7.0-mm-thick sections, 2.0–2.5-mm intersection gap) or fast spin-echo (3,000–3,500/22–30, 80–100; echo train length of four) MR images and T1-weighted spin-echo MR images (350–500/10–15) were obtained in all patients. Some of the T1-weighted MR images were enhanced with gadopentetate dimeglumine (Magnevist; Schering, Berlin, Germany). Gradient-echo MR images were not obtained. Serial MR imaging was performed in 12 patients.
Image Evaluation and Clinical Data
The criteria for detection of telangiectasia were small foci of very-low-signal-intensity areas and/or foci indicating hemosiderin on T2- or intermediate-weighted MR images. We excluded regions around surgical interventions and the basal ganglia from the evaluation. Two independent radiologists (S.K., N.A.) evaluated the MR images for the presence, number, size, and anatomic region of telangiectasias.
We divided the patients with telangiectasias into two groups: those who underwent whole-brain irradiation and those who underwent local irradiation. Furthermore, we described the location of each telangiectatic lesion according to five anatomic regions. The final decision regarding the presence of radiation-induced telangiectasia at MR imaging was made by means of mutual consent between the two radiologists. The lesions were classified by size into three groups: small, meaning smaller than 6 mm in diameter; medium, meaning 6–9 mm in diameter; and large, meaning larger than 9 mm in diameter. One radiologist (S.K.) also reviewed the patients’ clinical records for data, including age at initial irradiation, radiation field and fraction, use of chemotherapy, and clinical course. In each case, we recorded the first MR imaging appearance of each lesion.
Statistical Analyses
For comparisons between the LD and HD groups, between the patients with and those without telangiectasia, and between the patients who did and those who did not undergo chemotherapy, we used the Fisher exact test. The Fisher exact test was also used to assess the association between patient age at initial irradiation and frequency of telangiectasia. The t test was used to determine the difference in mean follow-up periods between the LD and HD groups. P < .05 was considered to indicate statistical significance.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
|
|
|
|
|
|
Six patients underwent MR imaging once, whereas 12 patients underwent MR imaging more than once (range, two to 15 examinations; total, 77 serial MR imaging examinations performed for up to 10 years [mean follow-up period, 8.1 years]). A total of 31 telangiectatic lesions were detected during this period. The times of the MR imaging appearance of each lesion after the initial irradiation are shown in Table 5. The latency period of the lesions—that is, the time from the initial irradiation to the first appearance of a telangiectatic lesion at MR imaging—ranged from 2 to 10 years. Thirty-nine percent (n = 12) of the 31 lesions were detected by the 3rd year, and 68% (n = 21) were evident by the 5th year. Overall, six (50%) of the 12 patients had telangiectatic foci after 5 years.
|
|
|
|
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
In contrast, small-vessel injury is a progressive impaired cerebral microcirculation and a prominent feature of delayed radiation injury. Classic histologic manifestations of these vascular changes include endothelial injury and proliferation, fibrinoid necrosis, and induction of capillary telangiectasia (12).
Radiation-induced telangiectasia was described in the literature before the advent of MR imaging clinical applications (13). Gaensler et al (6) described the appearance of hemorrhagic foci on MR images following central nervous system irradiation; these foci were similar to the cryptic vascular malformations described as radiation-induced telangiectasia (6,8). These foci were found to be depositions of perivascular hemosiderin and calcifications adjacent to capillary-sized telangiectasia. On the other hand, Pozzati et al (9) described the telangiectasias seen following central nervous system irradiation as occult cerebrovascular malformations. They proposed that the term telangiectasia is somewhat restrictive because radiation-induced changes probably represent a more complex pathologic phenomenon and their clinical behavior is often similar to that of cavernous angiomas.
Zabramski et al (14) classified cavernous cerebral malformations into four types: The type 1 cavernous cerebral malformation appears as a high-signal-intensity core on T1- and T2-weighted MR images and as a high- or low-signal-intensity core with a low-signal-intensity rim on T2-weighted MR images. Type 2 appears as a reticular, mixed-signal-intensity core on T1- and T2-weighted MR images and as a low-signal-intensity rim on T2-weighted MR images. Type 3 appears as a homogenous isointense or low-signal-intensity area on T1- and T2-weighted MR images. Type 4 appears as an area of poorly defined signal intensity or absent signal on spin-echo MR images and as an area of focally and decreased signal intensity on gradient-echo MR images. In the present study, almost all of the patients at T2-weighted MR imaging had hypointense foci that were radiologically identical to type 3 cavernous cerebral malformations (ie, isointense or hypointense lesions on T1- and T2-weighted MR images) according to their classification. Among the familial cavernous cerebral malformations, the type 3 malformations have been reported to be asymptomatic (15,16), like those in our series. Therefore, we believe that asymptomatic small, hypointense lesions seen on T2- and intermediate-weighted MR images should be referred to as telangiectasias.
To our knowledge, the frequency of radiation-induced telangiectasia has not been investigated. Other radiation-induced damage, including focal central nervous system necrosis, diffuse white matter injury, central nervous system atrophy, and mineralizing microangiopathy, reportedly occurs in approximately 25%–50% of patients (1–3). In the present study, the frequency of radiation-induced telangiectasia was 20%, which is close to the frequency of previously reported radiation-induced histologic microangiopathies (13). The frequency may have been even higher in this study, but the follow-up period may not have been long enough for some of the patients. Furthermore, in our evaluation, we excluded areas around surgical interventions and the basal ganglia. Therefore, with a sufficient follow-up period, the true frequency of telangiectasia in the whole irradiated area may have exceeded 20%.
To our knowledge, the radiation dose threshold for the development of telangiectasia in the human brain has not been reported; however, a few investigators have reported that single-dose (ie, 20–23-Gy) irradiation to animal brains causes telangiectasia (17,18). We divided the patients in this study into two dose groups to determine the approximate radiation dose threshold. However, our classification of the HD group may warrant reconsideration because it included patients who underwent 32- or 36-Gy cranial irradiation.
According to our study results, telangiectasias developed in 13% of the patients in the LD group (mean radiation dose, 18.07 Gy). These results suggest that the dose threshold for the induction of telangiectasia during standard cranial radiation therapies may be quite low. Moreover, although there was no significant difference in the occurrence of telangiectasias between the patients who received HD and those who received LD therapies, the LD group had a higher mortality rate and shorter follow-up periods. These differences may have biased our analysis of significant differences in the frequency of telangiectasias.
In the present study, most of the telangiectatic lesions were smaller than 10 mm in diameter. The majority of patients had only one small (ie, <6 mm) telangiectatic lesion. None of the telangiectatic foci disappeared or diminished during the course of follow-up. Although the frequency of telangiectasia did not differ significantly between the HD and LD groups, the telangiectasias tended to be larger in the HD group. This finding may reflect the longer mean follow-up period for the HD group, which increased the potential for microbleeding of the lesions.
To our knowledge, the latency period for radiation-induced telangiectasia has not yet been clarified. Gaensler et al (6) reported that the gross true latency period ranged from 5 months to 22 years. In our study, the latency period with serial MR imaging ranged from 2 to 10 years, and 64% of the lesions were evident by year 5. The mean follow-up periods in this study (6.1 years for the HD group and 2.3 years for the LD group) were shorter than those in the Gaensler et al (6) study. This is particularly relevant considering that telangiectasias may appear more than 10 years after irradiation. We speculate that more than 65% of telangiectatic lesions tend to become evident within 5 years after irradiation and continue to appear for several years thereafter. The latency period for large-vessel vasculopathy is reported to vary from 4 months to 20 years following the completion of radiation therapy (3), which may be similar to the latency period for radiation-induced telangiectasia.
The age of the patient at initial radiation exposure may affect the frequency of telangiectasias. Oi et al (19) at autopsy examined the bodies of 34 children who had had glioma and found evidence that the immature brain may be more sensitive to radiation than the adult brain. They further suggested that the manifestations of radiation-induced injury depend on the time that elapses after irradiation. In the 31 patients aged 10 years or older who underwent radiation therapy in the present study, only three (10%) developed telangiectasias, and none of the eight patients who underwent irradiation at age 14 years or older developed telangiectatic foci. Therefore, the immature brain does seem to be more sensitive to irradiation than the mature brain. The difference in frequency was not significant (P = .06), but this may have been because of the small patient sample size and the shorter follow-up period.
The frequency of radiation-induced telangiectasias seems to be unrelated to the use of chemotherapy. This finding is similar to the findings in children with other radiation-induced damage, such as white-matter abnormalities and calcifications (14). The lack of association between radiation therapy and chemotherapy with regard to radiation-induced telangiectasia may be due to the lack of permeation of the chemotherapeutic agents through the brain parenchyma. The frequency of radiation-induced telangiectasia was also unrelated to the field of radiation. Although telangiectatic foci appeared within the radiation field, the frequency of these foci did not differ significantly between patients who underwent whole-brain and those who underwent local irradiation.
Our study had limitations that we must point out. First, we have neither pathologic nor angiographic evidence that the lesions seen in our patient sample were telangiectasias rather than other abnormalities (eg, chronic small hemorrhage or focal calcification). Second, we may have introduced selection bias by excluding 224 patients who did not undergo MR imaging.
In summary, radiation-induced telangiectasia following cranial irradiation is not an uncommon complication in children. However, this complication tends to be asymptomatic and more frequent in younger patients. The frequency of telangiectasias does not appear to be affected by chemotherapy. In this small series, the radiation dose did not significantly affect the frequency of telangiectasia, although there was a trend in this direction. The latency period appears to range from a few months to long after radiation therapy, although the majority of telangiectatic lesions seem to appear within 5 years after the initial radiation therapy.
|
FOOTNOTES |
|---|
Author contributions: Guarantor of integrity of entire study, S.K.; study concepts and design, S.K., N.A.; literature research, S.K.; clinical studies, S.K., N.A.; data acquisition, S.K., N.A., M.H., K.F., Y.O.; data analysis/interpretation, S.K., N.A., T.I.; statistical analysis, S.K., T.I.; manuscript preparation, S.K., N.A.; manuscript definition of intellectual content, S.K., N.A., M.H., Y.O.; manuscript editing, S.K., N.A.; manuscript revision/review, N.A., T.I., K.F., Y.O.; manuscript final version approval, all authors
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  M.S.M. Chan, D.J. Roebuck, M.-P. Yuen, C.-K. Li, and Y.-L. Chan MR Imaging of the Brain in Patients Cured of Acute Lymphoblastic Leukemia--the Value of Gradient Echo Imaging AJNR Am. J. Neuroradiol., March 1, 2006; 27(3): 548 - 552. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Jain, P. L. Robertson, D. Gandhi, S. K. Gujar, K. M. Muraszko, and S. Gebarski Radiation-Induced Cavernomas of the Brain AJNR Am. J. Neuroradiol., May 1, 2005; 26(5): 1158 - 1162. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
