| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Department of Endocrinology (C.M., R.R., M.M., F.M., E.M., B.M., A.P.) and Oncology (S.M., L.C.) and Neuroscience (M.S.S., M.N.), University of Pisa, 56124 Pisa, Italy; University of Insubria (L.B.), Varese, Italy; and Radiation Oncology Unit, Hospital of Ferrara (F.C.), Ferrara, Italy
Address all correspondence and requests for reprints to: Claudio Marcocci, M.D., Department of Endocrinology, University of Pisa, Via Paradisa 2, 56100, Pisa, Italy. E-mail: c.marcocci{at}endoc.med.unipi.it.
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
At the University of Pisa, patients with severe GO have been treated with OR (almost always combined with high-dose glucocorticoids) for the last 30 yr. Based on the above-mentioned considerations, the aim of the present study was to evaluate the long-term outcome of OR in terms of untoward effects in a large cohort of GO patients.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Enrollment and recall of patients were restricted to GO patients who underwent OR in our University between 1972 and 1996, to have a follow-up period of at least 5 yr. During the 25-yr period considered, 511 patients with moderate to severe GO were irradiated. One hundred and twenty-three patients were irradiated from 1972-l985 with a cobalt unit (CU), whereas the remaining 388 patients were irradiated from 1985–1996 with a high voltage linear accelerator (LA). Most of the patients were managed by a team of endocrinologists, ophthalmologists, and radiation oncologists.
Three hundred and ninety-four patients were reached by telephone or mail; 310 agreed to undergo complete ophthalmological and medical evaluations and a computed tomography (CT) scan of the orbit and the adjacent regions, whereas 84 refused to be evaluated. Of the remaining 117 patients, 98 could not be traced, mostly because of address change, and 19 patients were dead. Causes of death in the latter subgroup were myocardial infarction (n = 5), breast cancer (n = 3), uterine cancer (n = 3), kidney cancer (n = 1), age-related disease (n = 4), and undefined reasons (n = 3). Between January 2001 and the end of July 2002, 204 of the 310 patients (44 treated by CU, 160 treated by LA) were evaluated. The results of these 204 patients are reported in the present study. The ethical committee approved the study, and patients signed an informed consent according to institutional guidelines.
Radiotherapy procedure
The cumulative radiation dose and the radiation field were identical in patients treated by CU or LA. A total of 20 Gy were delivered to each eye in 10 fractionated doses over a period of 2 wk. The head of the patient was stabilized with a head holder. All patients were treated bilaterally. The radiation field was 4 x 4 cm, with the anterior border located just behind the lateral canthus of the eye, and the posterior border located just anterior to the sella turcica; the inferior border of the field was the floor of the orbit, and the superior border was its roof. The field was angled 10° posteriorly to avoid irradiation of the contralateral lens.
Up to 1985 irradiation was delivered using a CU with a 50-cm source to skin distance. The isodose distribution, evaluated by CT scan, showed that the dose delivered to the lens, cornea, pituitary, and hypothalamus was less than 5%, with very low penumbra and side scatter (8). From 1985 a 4-MeV LA was used, and after 1990, a CT scan of the orbit was performed in each patient before treatment to precisely individualize the radiation field (2).
Follow-up evaluation
All patients were seen at the University of Pisa, and complete medical history and physical examination were performed. Ophthalmologists examined the patients among other patients with GO seen at our clinic and were blind to which patients were included in the study. The ophthalmological evaluation included assessment of visual acuity with correction, fundoscopy, computerized perimetry, and assessment of irradiated tissues. When indicated, fluorescein angiography was performed. For those patients who had been diagnosed with cataract or retinopathy before our examination, the time of diagnosis was considered for statistical analysis. One hundred and fifty-seven patients underwent a CT scan of the orbit and adjacent regions (paranasal sinuses and frontal lobe). Eighty-six consecutive patients with untreated GO seen in our department in the last 2 yr who underwent complete ophthalmological evaluation and orbital CT scan before OR were used as controls.
Statistical analysis
The data were reported as the mean ± SD, median, range, and frequency rate, as appropriate. Mann-Whitney and 
2 tests were used as appropriate. P < 0.05 was considered statistically significant for each analysis.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
As reported above, the study group consisted of 204 GO patients, 154 females (75%) and 50 males (25%; Table 1
). Patients had been treated with OR alone (n = 3) or with OR combined with glucocorticoids, either systemically (n = 173) or locally (n = 28) administered. The age at irradiation ranged from 17–68 yr (median, 47 yr). Forty-four patients (12 females and 32 males) were treated with CU, and 160 (38 females and 122 males) were treated with LA. No patient had received treatments for GO before those reported above. Eighteen patients had type 2 diabetes mellitus (9%), and 7 of these had associated hypertension (3%). Thirty-one patients (15%) had hypertension alone. The median age at the time of irradiation did not differ between the 2 groups (CU, 45 yr; LA, 48 yr; P = 0.31); CU patients were significantly older at follow-up than LA patients (P = 0.005), and they had a significantly longer (P = 0.0001) follow-up (median, 17 yr) than patients treated with LA (median, 9 yr). The overall duration of the follow-up ranged from 5–25 yr for both groups.
|
Twenty-one GO patients developed cataract after they had been submitted to OR, resulting in an overall prevalence of 10.3%. Ten of these 21 patients (4.9% of the total cohort) had a mature cataract, 8 of whom had been treated by phacoemulsification (bilaterally in 6 cases, monolaterally in 2 cases) 3–18 yr (mean, 9.6 yr) after OR and before our reevaluation. Two of these 10 patients with mature cataract had been advised to undergo surgery. Eleven of the 21 patients with cataract (5.4% of the total cohort) had a polar posterior cataract.
The mean age of the 21 patients with cataract when this was diagnosed (61 yr) was greater than that of patients who did not develop cataract (56 yr). The estimated time of cataract development was 12 ± 4 yr (range, 3–21 yr) after OR. The prevalence of cataract was greater in patients treated with CU (8 of 44, 18%) than in patients treated with LA (13 of 160, 8%), but the difference did not reach the statistical significance (2 = 2.8; P = 0.1). Because the mean age of patients at reevaluation was different between patients treated with CU (60 ± 10 yr) and patients treated with LA (56 ± 10 yr), the prevalence of cataract in the 2 groups was also analyzed according to age at diagnosis (
60 or <60 yr) and duration of follow-up. As indicated in Table 2 there was no difference in the development of cataract in the CU group according to age, whereas among patients treated with LA, cataract was observed more frequently in patients 60 yr or older (2 = 5.8; P = 0.02). There was no significant difference between patients treated with CU or LA according to age.
|
, there was an increased prevalence of cataract in patients treated with CU less than 60 yr of age compared with the control population, although the difference did not reach statistical significance. On the contrary, no increase in the prevalence of cataract was observed in patients treated by LA regardless of their age.
|
Retinal changes were detected in 15 patients and consisted of minimal signs of hypertensive retinopathy in 13 of 15 cases. Signs of retinopathy, possibly related to orbital irradiation, were observed in 2 patients (1%) treated by LA. Both of these patients had hypertension, and 1 had also diabetes mellitus. The time elapsed between OR and the diagnosis of retinopathy was 4 and 2 yr, respectively. In the patient with diabetes mellitus and hypertension, a venous retinal thrombosis was observed at diagnosis (Table 3). A clear relationship between OR and retinopathy could not be established in this patient, although several considerations suggest that radiation might indeed be responsible. This patient had a recent-onset, mild, type 2 diabetes, with no evidence of retinopathy before orbital irradiation and no significant deterioration in metabolic control during the follow-up. Furthermore, signs of retinopathy were detected 2 yr after completion of radiotherapy, a period rather short for retinopathy to be interpreted as diabetic and more typical of radiation-induced retinopathy.
|
CT scan
To investigate the possible occurrence of secondary tumors in the irradiated areas, 157 patients (80% of the present cohort; 117 females and 40 males; age, 37–79 yr, median, 53 yr) underwent CT scan of the orbit and adjacent areas. In the remaining 47 patients the CT scan was not performed because of the presence of dental metallic devices or because the patient refused to undergo the scan. CT scans were read in a blind fashion. No case of secondary tumor in the head and neck region was detected. A 2-cm arachnoid cyst in the frontal area and an ethmoid polyp were observed in a patient treated with LA 12 yr earlier. Mucosal thickening and rarely polyps of the paranasal sinuses were observed in 53 of 157 cases (34%). There was no difference in the prevalence of paranasal changes according to the radiotherapy procedure (CU, 12 of 38, 32%; LA, 41 of 119, 35%; 2 = 0.2; P = 0.9). The prevalence of these changes was significantly higher (2 = 5.5; P = 0.02) than that observed in 86 consecutive control GO patients of comparable age (30–75 yr; median, 51 yr) and sex (63 females and 23 males) who underwent CT scan before radiotherapy (16 of 86, 19%; Fig. 2).
|
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Concerns regarding the possible long-term side-effects and risks of OR include the possible development of cataract, retinopathy, and radiation-induced tumors. A few studies have addressed this issue. It is well known that the risk of cataract is dose related and can be reduced by dose fractionation, appropriate well collimated irradiation techniques, and well designed radiation fields (3). With the current treatment procedure, which delivers a cumulative dose of 20 Gy to each eye by high voltage LA in fractionated doses, the lens is approximately exposed to 5–10% of the total dose or even less (16, 17). The latency period between irradiation and development of cataract averages 2 yr, ranging from 6 months to a few years (7). In the present study OR was mostly used in association with glucocorticoids. Because the latter may also favor the development of cataract, this should be considered in evaluating our results. We found an overall prevalence of cataract of 10.3% in our large series of GO patients, with a median estimated time for cataract development of 12 yr after OR. It must be pointed out that only half of the patients with cataract had the typical lens alterations induced by irradiation (polar posterior cataract). In contrast, the remaining half with cataract had the mature form, which, although it can be the outcome of a polar posterior cataract, may also be unrelated to irradiation. Patients with cataract were older than those without cataract, in agreement with the results reported by Petersen et al. (17) and more recently by Marquez et al. (18). Interestingly, in our series there was no significant difference in the frequency of cataract development between patients treated with CU or LA, even though the former radiotherapy procedure is known to be characterized by lower collimation and greater penumbra and side scatter compared with the latter. This observation may be used as an argument against a causative role for OR in the development of cataract in our GO patients. In an attempt to address this issue further, we compared the frequency of cataract in our series with that reported in Caucasian Italian subjects (11). We found that patients less that 60 yr of age and treated with CU had an increased prevalence of cataract than the control population, although the difference was not statistically significant. The finding suggests that the use of CU in GO may increase the risk of cataract when administered to relatively young patients. In contrast, LA treatment was not associated with an increased frequency of cataract regardless of the patient’s age at the time of irradiation. Similar results were obtained by comparing our results with those reported in white noninstituzionalized subjects from the United States of America (19) using 65 yr as a cut-off, but in the latter case the increased prevalence of cataract in subjects treated with CU and less than 65 yr of age was statistically significant (not shown). The increased prevalence of cataract related to CU in young patients is also indicated by the fact that, unlike in LA-treated patients, there was no difference in the prevalence of cataract according to age in the CU group.
It may be argued that it would have been more appropriate to use as controls for cataract other patients with GO not treated with OR or patients treated with glucocorticoids for diseases other than GO. However, this was not feasible in our case, as in our clinic the common treatment modality for moderate to severe GO is OR combined with systemic glucocorticoids. Therefore, very few patients not treated with radiotherapy were available. In addition, we could not access patients treated with glucocorticoids for other diseases in sufficient numbers to draw any conclusions.
Radiation retinopathy following OR has been reported when doses higher than those commonly employed for GO were used (20), and retinopathy has been reported to be overt between 6 months and 3 yr after irradiation (21). Preexisting diabetic retinopathy and microvascular damage caused by either hypertension or previous chemotherapy increase the risk of radiation retinopathy (22). With the exception of 2 cases (23, 24), errors in dose calculation and radiation technique accounted for the few cases of retinopathy reported in nondiabetic GO patients (20). The results of the present study support the conclusion that the risk of radiation retinopathy is very low in GO patients if appropriate irradiation techniques and doses are applied. Thus, we detected only 2 cases of asymptomatic mild retinopathy, and both patients had hypertension, which was associated with type 2 diabetes mellitus in 1 case. These mild changes had no clinical relevance, as all patients with retinopathy were asymptomatic and had normal visual acuity. In addition, as the retinal changes of radiation-induced retinopathy are similar to those of diabetic retinopathy, we could not establish whether the retinopathy observed in the patient with diabetes was radiation induced. Whether there is a certain causal relation between retinopathy and previous OR could not be established. Minimal retinal changes have been reported preliminarily by Wakelkamp et al. (25) in 17 of 104 GO patients treated with OR. In addition, they found severe retinopathy in 5 patients, 3 of whom had diabetes mellitus and 2 who had hypertension (25). In another recent study, 3 of 40 (7.5%) patients treated with OR were found to have asymptomatic, newly developed retinal changes (dilated capillaries or microaneurysms), which were detected by fluorescein angiograms performed before and 3 yr after OR (26). In the latter study, patients with hypertension and/or diabetes mellitus had been excluded from the treatment (26). The significance of these minimal retinal changes and their potential progression toward a clinically relevant retinopathy remain to be established. Based on these considerations, we propose that hypertension and diabetes mellitus, especially if associated, should be considered as relative contraindications to OR, in particular when signs of retinopathy are present before irradiation.
The major long-term OR concern is the potential development of cancer in the head and neck region. Beckendorf et al. (16), using dose-volume histograms and irradiation technique and dose comparable to those currently used, showed that the dose given to the brain and skull bones is minimal. Snijders-Keilholz et al. (9), based on tissue-weighting factors, calculated that the theoretical risk of secondary cancer for a standard patient treated with 20 Gy is 1.2%. The use of higher photon energy, reduced field sizes, and smaller target dose (16 Gy) may decrease the risk of radiation-induced cancer to 0.7% (27). On the contrary, Blank et al. (10) suggested that the above calculations lead to a 2- to 4-fold overestimation, and that a theoretical risk of 0.3% is a more accurate estimation. The risk of secondary radiation-induced tumors should be considerably reduced if older patients are exposed to OR. In the present large cohort of irradiated GO patients, no case of radiation-induced cancer in the head and neck region was detected after a 5- to 25-yr follow-up. Similarly, no evidence of cancer was reported in a study of 311 patients from Stanford University seen 21 yr after OR (18) and in a series of 250 patients reported by Schaefer et al. (28) with a median follow-up of 31 yr. In the latter study, the death rate and the death rate due to cancer were similar to those observed in the general population.
An unexpected finding in the present study was the high prevalence of abnormalities of the paranasal sinuses (mucosal thickening and rarely polyps) that occurred in about one third of patients regardless of the radiotherapy procedure used. The prevalence of these changes was significantly higher than that in control GO patients before radiotherapy. Whether these paranasal sinus lesions are related to OR remains to be established. Indeed, a similar prevalence of abnormalities of one or more of the paranasal sinuses has been reported in patients treated by high-dose external radiotherapy for nasopharyngeal carcinoma (29) and also in adults referred for cranial CT scan for conditions such as head injuries and seizures (30).
It may be argued that examination of the whole cohort of 310 patients who agreed to be enrolled in the study might have further strengthened our data and conclusions. Unfortunately, the remaining 106 patients were not readily available for reexamination. Nevertheless, a preliminary analysis of the medical records of the remaining 106 patients showed a similar frequency of cataract and retinopathy as in the 204 patients reported here (not shown).
In conclusion, our results together with data in the available literature indicate that OR is a well tolerated and safe procedure, with very limited side-effects and complications, provided that appropriate irradiation techniques and apparatus are used and patients are properly selected. Although the follow-up period in this and other studies (18, 28) was relatively long, the potential carcinogenic effect of irradiation cannot be ruled out with certainty due to the long latency of radiation-induced tumors. Therefore, it seems prudent to avoid irradiation in patients less than 60 yr of age.
|
Footnotes |
|---|
Abbreviations: CT, Computed tomography; CU, cobalt unit; GO, Graves’ ophthalmopathy; LA, linear accelerator; OR, orbital radiotherapy.
Received February 18, 2003.
Accepted May 6, 2003.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  L. Bartalena, L. Baldeschi, A. Dickinson, A. Eckstein, P. Kendall-Taylor, C. Marcocci, M. Mourits, P. Perros, K. Boboridis, A. Boschi, et al. Consensus statement of the European Group on Graves' orbitopathy (EUGOGO) on management of GO Eur. J. Endocrinol., March 1, 2008; 158(3): 273 - 285. [Full Text] [PDF]
|
|
|
|
|
|
The European Group of Graves' Orbitopathy, P Perros, L Baldeschi, K Boboridis, A J Dickinson, A Hullo, G J Kahaly, P Kendall-Taylor, G E Krassas, C M Lane, et al. A questionnaire survey on the management of Graves' orbitopathy in Europe. Eur. J. Endocrinol., August 1, 2006; 155(2): 207 - 211. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Bartalena Glucocorticoids for Graves' Ophthalmopathy: How and When J. Clin. Endocrinol. Metab., September 1, 2005; 90(9): 5497 - 5499. [Full Text] [PDF]
|
|
|
|
|
|
L. Bartalena, C. Marcocci, and A. Pinchera Orbital Radiotherapy for Graves' Ophthalmopathy J. Clin. Endocrinol. Metab., January 1, 2004; 89(1): 13 - 14. [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
