Effect of Steroid Pulse Therapy With andWithout Orbital Radiotherapy on Graves’

Ophthalmopathy

KENJI OHTSUKA, MD, AKIHIKO SATO, MD, SATOSHI KAWAGUCHI, MD,MASATO HASHIMOTO, MD, AND YASUO SUZUKI, MD

● PURPOSE: To report the effect of high-dose intrave-nous corticosteroid pulse therapy with and without or-bital radiotherapy on Graves’ ophthalmopathy.● DESIGN: Nonrandomized clinical trial.● METHODS: We selected 39 Japanese patients (agerange, 22–64 years; mean, 48 years; 31 women, 8 men)who had active Graves’ ophthalmopathy among 195consecutive patients. In the first 20 patients, high-doseintravenous methylprednisolone pulse therapy (1 g perday for 3 successive days, repeated 3 times within 3weeks) followed by 24-Gy orbital radiotherapy wasperformed. In the other 19 of the 39 patients, high-doseintravenous methylprednisolone pulse therapy withoutorbital radiotherapy was performed. Coronal computedtomography (CT) of the orbit and exophthalmometrywere performed before the corticosteroid pulse therapy,and 1 and 6 months after the corticosteroid pulse ther-apy. The maximum coronal section area of the mosthypertrophic rectus muscle in each eye was measuredbased on orbital CT imaging.● RESULTS: Clinical findings at study entry were notsignificantly different between the two groups. No sig-nificant difference was found in the maximum coronalsection area of the most hypertrophic rectus muscle andthe results of exophthalmometry measurements in bothgroups before the therapy. Extraocular muscle hypertro-phy was significantly reduced 1 month and 6 monthsafter the therapy (P < .01) in both groups. However, nobeneficial therapeutic effect on proptosis was observed ineither group at 1 month and 6 months after the therapy.No significant difference in the therapeutic effect onextraocular muscle hypertrophy and proptosis was foundbetween the two groups.

● CONCLUSIONS: Orbital irradiation after corticosteroidpulse therapy had no beneficial therapeutic effects onrectus muscle hypertrophy or proptosis of active Graves’ophthalmopathy during the 6-month follow-up period.(Am J Ophthalmol 2003;135:285–290. © 2003 byElsevier Science Inc. All rights reserved.)

G RAVES’ OPHTHALMOPATHY IS A CHRONIC AUTO-

immune process that affects the retrobulbar tissueand has strong etiological links with autoimmune

thyroid disease. Symptoms include blurring of vision,proptosis, extraocular muscle dysfunction, eyelid swelling,chemosis, redness of the conjunctiva, pain, lid lag, andretraction. Corticosteroid (glucocorticoids) and orbitalradiotherapy have been used for treatment of Graves’ophthalmopathy. Many previous studies have shown thetherapeutic benefits of corticosteroid therapy.1–4 Recently,the efficacy of high-dose intravenous corticosteroid pulsetherapy has been reported.5–10 Patients treated with high-dose intravenous corticosteroid had a better outcome thanthose treated with an oral dose of corticosteroid. Effective-ness of orbital radiotherapy for Graves’ ophthalmopathy isalso reported.11–13 Combination of orbital radiotherapywith high-dose corticosteroid pulse therapy was reported toprovide better results than either therapy alone.14 How-ever, the effect of orbital radiotherapy is still controversialfor the treatment of Graves’ ophthalmopathy.15–17 It ispossible that radiotherapy can provide long-term benefitsfor the treatment of Graves’ ophthalmopathy, and thecorticosteroid pulse therapy can provide short-term bene-fits. Therefore, systemic corticosteroid therapy with orbitalradiotherapy has been used in many cases.18 However, thebeneficial effects of orbital radiotherapy after the cortico-steroid pulse therapy have not been clarified.

In this study, we evaluated the effect of high-doseintravenous corticosteroid pulse therapy with or withoutorbital radiotherapy on extraocular muscle hypertrophyand proptosis in patients with active Graves’ ophthalmop-athy. Before the therapy, we evaluated the activity ofophthalmopathy using the classification system proposed

Accepted for publication Oct 1, 2002.From the Department of Ophthalmology, Sapporo Medical University,

School of Medicine, Sapporo, Japan.Inquiries to Kenji Ohtsuka, MD, PhD, Department of Ophthalmology,

Sapporo Medical University School of Medicine, S-1, W-16, Chuo-ku,Sapporo 060-8543, Japan; fax: (�81)-011–613-6575; e-mail: kohtsuka@sapmed.ac.jp

© 2003 BY ELSEVIER SCIENCE INC. ALL RIGHTS RESERVED.0002-9394/03/$30.00 285PII S0002-9394(02)01970-0

by Mourits and associates19 and performed the corticoste-roid pulse therapy with or without orbital irradiation intwo groups of patients whose clinical findings at studyentry were not significantly different between the twogroups. We evaluated differences in extraocular musclehypertrophy and proptosis between the two groups duringthe 6-month follow-up period.

METHODS

● EXPERIMENTAL DESIGN: A nonrandomized clinicaltrial was designed to investigate the effect of high-doseintravenous corticosteroid pulse therapy with and withoutorbital radiotherapy on Graves’ ophthalmopathy.

● PARTICIPANTS AND INCLUSION CRITERIA: The di-agnosis of Graves’ disease was based on history; thepresence of conventional symptoms of thyrotoxicosis asso-ciated with a diffusely enlarged goiter; elevated levels ofserum T3, free T3, T4, and free T4; and increased thyroidal131I uptake. Elevated titers of antithyroid-stimulating an-tibody, antithyroglobulin antibody, and antithyroid perox-idase antibody or the presence of ophthalmopathy, such asproptosis, extraocular muscle hypertrophy, orbital conges-tion, eyelid retraction, and orbital inflammation, providedsupporting evidence for the diagnosis. A complete oph-thalmic examination, including applanation tonometry,exophthalmometry, funduscopy, and measurement of eyemovements, was performed by the same ophthalmologist atthe first visit, before commencement of any systemicmedical treatment, in all patients. The activity of ophthal-mopathy was evaluated using the classification systemproposed by Mourits and associates.19 The classificationsystem (activity score) ranged from 0 to 10 points based onthe classical signs of inflammation (pain, redness, swelling,impairment of visual acuity, and impairment of eye move-ment).19

We selected 39 Japanese patients (age range, 22–64years; mean, 48 years; 31 women, 8 men) who had activeophthalmopathy, defined as a clinical activity score of fouror more points based on the classification system orimpaired function for at least 1 month (decrease in visualacuity or decrease in eye movements) among 195 consec-utive patients examined at Sapporo Medical UniversityHospital between 1998 and 2000. Mourits and associates19

indicated that patients with a clinical activity score ofthree or more points improve after oral corticosteroidtreatment. In our clinic, patients with three points aretreated by oral corticosteroid.

Patients with four or more points on the classificationsystem of Mourits and associates19 participated in theclinical trial of the present study. Twenty patients (agerange, 22 to 64 years; mean, 47 years; 16 women, 4 men)whose illness was diagnosed from January 1998 to June1999 received high-dose intravenous methylprednisolone

pulse therapy followed by orbital radiotherapy (pulse-irradiation group). The other 19 patients (age range, 30 to62 years; mean, 50 years; 15 women, 4 men) whose illnesswas diagnosed from July 1999 to December 2000 receivedhigh-dose intravenous methylprednisolone pulse therapywithout orbital radiotherapy (pulse group). The distribu-tion of the clinical activity scores of the 39 patients isshown in Figure 1. The mean clinical activity score was 5.8in the 39 patients, 6.0 in the pulse-irradiation group and5.7 in the pulse group. The clinical activity score was notsignificantly different between the two groups (Mann-Whitney U test).

● TREATMENT OF OPHTHALMOPATHY: Corticosteroidpulse therapy was performed following the administrationschedule used in the previous study.10 Methylprednisolonewas administered intravenously at a daily dose of 1 g for 3successive days. This treatment was repeated three timeswithin 3 weeks. A daily dose of oral prednisone 30 mg wasadministered for interpulse therapy (4 successive days, twotimes). Oral prednisone therapy following intravenousmethylprednisolone pulse therapy was also performed (ini-tial daily dose of 30 mg, tapered to discontinuationapproximately 3 months after the corticosteroid pulsetherapy).

Orbital radiotherapy for the pulse-irradiation group wasperformed immediately after the corticosteroid pulse ther-apy. The dose of orbital irradiation was 24 Gy in 12fractions over 16 days using a 4-MeV linear accelerator(Siemens, Berlin, Germany). The head was immobilizedwith a thermoplastic mask. The location of the isocenterwas determined with pretreatment planning computedtomography (CT) imaging. The beam arrangement con-sisted of a 60-degree wedged pair, � 45 degrees fromlateral. The isocenter of irradiation was located 2.0 cmposterior to the anterior corneal surface. A daily dose oforal prednisone 20 mg was administered during the radio-therapy. Oral prednisone therapy after the radiotherapywas also performed (initial daily dose of 30 mg, tapered to

FIGURE 1. Distribution of clinical activity score based onfindings at study entry.

AMERICAN JOURNAL OF OPHTHALMOLOGY286 MARCH 2003

discontinuation approximately 3 months after the cortico-steroid pulse therapy).

● ASSESSMENT OF OPHTHALMOPATHY: Serial coronalCT scans of 2-mm sections were acquired of both orbitsusing the General Electric 9800 scanner (GE MedicalSystems, Milwaukee, Wisconsin, USA). We identified themaximum coronal section area of the most hypertrophicrectus muscle from serial scans of orbital CT in each eyeand measured the maximum coronal section area of therectus muscle using the analysis package of GE MedicalSystems (Milwaukee, Wisconsin, USA). We used a thresh-old setting of 0 to � 200 for evaluation of the maximumcoronal section area of the rectus muscle. The measure-ment of exophthalmometry was performed by one ophthal-mologist using a Hertel exophthalmometer (Handaya,Tokyo, Japan). The assessment of ophthalmopathy wasperformed before the corticosteroid pulse therapy, and 1month and 6 months after the therapy in both groups.

● STATISTICAL ANALYSIS: In a prospective study, weinvestigated the differences in the maximum coronalsection area of the rectus muscle and the results ofexophthalmometry measurements of both groups before, 1month after, and 6 months after the therapy. To determineif these values differed among the groups, two-way repeat-ed-measures analysis of variance (ANOVA) was used.

● HUMAN STUDIES CONSIDERATIONS: Informed con-sent was obtained from all the patients after the risks ofcorticosteroid pulse therapy and the risks of orbital radio-therapy had been explained. If optic neuropathy developedand was unresponsive to the corticosteroid pulse therapy,orbital decompression was recommended. All study proto-cols were approved by the Sapporo Medical UniversityClinical Study Committee.

RESULTS

THERE WAS NO SIGNIFICANT DIFFERENCE IN THE MAXIMUM

coronal section area of the rectus muscle and the results ofexophthalmometry between the two groups at study entry(Figures 2 and 3) (two-tailed t test).

Extraocular muscle hypertrophy improved immediatelyafter the therapy in both groups. Figure 4 shows changes inthe maximum coronal section area of the rectus muscle inboth groups before and 1 month and 6 months after thetherapy. In the pulse group, the mean values of themaximum coronal section area of the rectus muscle were54.0 � 23.6 mm2 (mean � SD; n � 38) before thetherapy, 36.1 � 15.1 mm2 (n � 38) 1 month after thetherapy, and 44.4 � 19.9 mm2 (n � 39) 6 months after thetherapy. In the pulse-irradiation group, the mean values ofthe maximum coronal section area of the rectus musclewere 54.2 � 24.4 mm2 (n � 39) before the therapy, 39.5

� 17.1 mm2 (n � 39) 1 month after the therapy, and 45.4� 23.0 mm2 (n � 39) 6 months after the therapy. Themaximum coronal section area of the rectus muscle wassignificantly reduced 1 month and 6 months after thetherapy (repeated-measures ANOVA, P � .01), and wasnot significantly different between 1 month and 6 monthsafter the therapy in both groups (repeated-measuresANOVA). A significant difference in the maximum coro-nal section area of the rectus muscle was not foundbetween the two groups at 1 and 6 months (two-wayrepeated-measures ANOVA).

Conversely, the effect on proptosis was minimal in bothgroups (Figure 5). In the pulse group, the results ofexophthalmometry measurements were 18.6 � 2.8 mm(mean � SD; n � 38) before the therapy, 17.7 � 2.6 mm(n � 38) 1 month after the therapy, and 17.9 � 3.4 mm(n � 38) 6 months after the therapy. In the pulse-irradiation group, the results of exophthalmometry mea-surements were 18.8 � 3.7 mm (n � 40) before thetherapy, 18.0 � 3.1 mm (n � 40) 1 month after thetherapy, and 18.6 � 3.3 mm (n � 40) 6 months after thetherapy. In the pulse group, proptosis was significantlyreduced both 1 month (repeated-measures ANOVA, P �.01) and 6 months after the therapy (repeated-measuresANOVA, P � .05). In the pulse-irradiation group, prop-tosis was significantly reduced 1 month after the therapy

FIGURE 2. Distributions of the maximum coronal section areaof the rectus muscle before the therapy in (top) the pulse groupand (bottom) the pulse-irradiation group. There were no sig-nificant differences in the maximum coronal section area of therectus muscle between the two groups at study entry (two-tailed t test).

STEROID PULSE AND RADIOTHERAPY ON TAOVOL. 135, NO. 3 287

(repeated-measures ANOVA, P � .05) but was not signif-icantly different between before the therapy and 6 monthsafter the therapy (repeated-measures ANOVA). A signif-icant difference in proptosis was not found between thetwo groups at 1 month and 6 months after the therapy(two-way repeated-measures ANOVA).

DISCUSSION

ALL SIGNIFICANT CLINICAL FINDINGS IN THE EARLY STAGE

of active Graves’ ophthalmopathy can be traced back toinflammatory edema of the extraocular muscle and theretrobulbar connective tissue.20–23 Inflammatory edemacauses extraocular muscle hypertrophy and proptosis. Inthe chronic stage, proliferation of fibroblasts and accumu-lation of glycosaminoglycans were noted in addition toinflammatory edema.20–23 These chronic changes causerestriction of extraocular muscle motion and irreversibleenlargement of the retrobulbar tissue. If the therapy isstarted at the early inflammatory stage before proliferationof fibroblasts and accumulation of glycosaminoglycans arenoted, the therapy will be effective for Graves’ ophthal-mopathy. Therefore, patients with active ophthalmopathyshould be selected for subjects in a study to clarify theefficacy of a treatment. Changes in extraocular musclehypertrophy and proptosis should be the objective criteriaof the efficacy of the treatment.

Serum levels of antithyroid-stimulating antibody, anti-thyroglobulin antibody, and antithyroid peroxidase anti-body are correlated with the autoimmune process ofGraves’ thyroid disease. However, no clear correlation hasbeen established between serum levels of antithyroidantibodies and severity of ophthalmopathy, although it hasbeen reported that severe ophthalmopathy tends to de-velop in the presence of high titers in the serum antithy-roid-stimulating antibody.24–26 Therefore, serum levels ofantithyroid antibodies cannot be the criteria of activeGraves’ ophthalmopathy. In this study, we selected pa-tients with the early stage of active Graves’ ophthalmop-athy using the classification system proposed by Mouritsand associates.19

Although the present study was not a randomizedclinical trial, age, the proportion of gender, and clinicalfindings at study entry were not significantly differentbetween the two groups. Therefore, it is probable thatdifferences in therapeutic outcome between the two groupswould confirm the efficacy of orbital irradiation for Graves’ophthalmopathy. However, we were unable to demon-strate any difference in extraocular muscle hypertrophy orproptosis between the pulse group and the pulse-irradiationgroup in this study. In the pulse group, extraocular musclehypertrophy was significantly reduced after the therapy,and relapse was minimal within 6 months. The effect ofthe corticosteroid pulse therapy on proptosis was minimal.These findings are compatible with results in previousstudies.5–10 In the pulse-irradiation group, extraocular mus-

FIGURE 3. Distributions of proptosis before the therapy in(top) the pulse group and (bottom) the pulse-irradiation group.There were no significant differences in the results of exoph-thalmometry between the two groups at study entry (two-tailedt test).

FIGURE 4. Changes in extraocular muscle hypertrophy be-fore, and 1 and 6 months after the therapy in (top) the pulsegroup and (bottom) the pulse-irradiation group. **P < .01(repeated-measures analysis of variance).

AMERICAN JOURNAL OF OPHTHALMOLOGY288 MARCH 2003

cle hypertrophy was also significantly reduced after thetherapy. The effect on proptosis was also minimal, asshown in the pulse group. However, no significant differ-ence in the effect of the treatment was found between thetwo groups. We did not find any beneficial therapeuticeffect of orbital irradiation after corticosteroid pulse ther-apy during the 6-month follow-up period.

Recently, Gorman and associates27 reported that nobeneficial therapeutic effects of orbital radiotherapy werefound in patients with Graves’ ophthalmopathy within 6months after the therapy. They treated patients withactive ophthalmopathy with 20-Gy external beam therapy.No clinically or statistically significant difference betweenthe treated and untreated orbits was observed in thevolume of extraocular muscle, proptosis, range of extraoc-ular muscle motion, and lid fissure width at 6 months.These findings suggest that neither simple orbital irradia-tion nor irradiation after corticosteroid therapy is benefi-cial for patients with Graves’ ophthalmopathy.

Both corticosteroid pulse therapy and orbital irradiationtherapy have side effects. Systemic side effects may resultfrom corticosteroids, and cataract and radiation retinopa-thy may result from orbital radiation therapy. In this study,the side effects of orbital irradiation were not detectedduring the 6-month follow-up period. Obesity and hyper-glycemia as side effects of corticosteroid administration

were observed in several patients. However, these werecontrollable and improved after cessation of corticosteroidadministration. In previous studies, cataract and radiationretinopathy after orbital irradiation for treatment ofGraves’ ophthalmopathy were reported.28,29 The risk ofvisual disturbance is higher in the corticosteroid pulse-irradiation group than in the corticosteroid pulse group.Orbital irradiation after corticosteroid pulse therapy fortreatment of Graves’ ophthalmopathy should be reassessed,because orbital irradiation had no beneficial therapeuticeffects on rectus muscle hypertrophy or proptosis of activeGraves’ ophthalmopathy during the 6-month follow-upperiod.

REFERENCES

1. Werner SC. Prednisone in emergency treatment of malig-nant exophthalmos. Lancet 1966;1:1004–1007.

2. Day RM, Carroll FD. Corticosteroid in the treatment of opticnerve involvement associated with thyroid dysfunction.Arch Ophthalmol 1968;79:279–282.

3. Prummel MF, Mourits MP, Berghout A, et al. Prednisoneand cyclosporine in the treatment of severe Graves’ ophthal-mopathy. N Engl J Med 1989;321:1353–1359.

4. Jacobson DH, Gorman CA. Endocrine ophthalmopathy:current ideas concerning etiology, pathogenesis and treat-ment. Endocrinol Rev 1984;5:200–220.

5. Nagayama Y, Izumi M, Kiriyama T, et al. Treatment ofGraves’ ophthalmopathy with high-dose intravenous meth-ylprednisolone pulse herapy. Acta Endocrinol 1987;116:513–518.

6. Guy JR, Fageien S, Donovan JP, Rubin ML. Methylpred-nisolone pulse therapy in severe dysthyroid optic neuropathy.Ophthalmology 1989;96:1048–1052.

7. Hiromatsu Y, Tanaka K, Sato M, et al. Intravenous methyl-prednisolone pulse therapy for Graves’ ophthalmopathy.Endocr J 1993;40:63–72.

8. Tagami T, Tanaka K, Sugawa H, et al. High-dose intrave-nous steroid pulse therapy in thyroid-associated ophthalmop-athy. Endocr J 1996;43:689–699.

9. Matejka G, Verges B, Vaillant G, et al. Intravenous meth-ylprednisolone pulse therapy in the treatment of Graves’ophthalmopathy. Horm Metab Res 1998;30:93–98.

10. Ohtsuka K, Sato A, Kawaguchi S, Hashimoto M, Suzuki Y.Effect of high-dose intravenous steroid pulse therapy fol-lowed by three-month oral steroid therapy for Graves’ophthalmopathy. Jpn J Ophthalmol 2002;46:563–567.

11. Prummel MF, Mouritis MP, Blank L, Berghout A, KoornneefL, Wiersinga WM. Randomized double-blind trial of pred-nisone versus radiotherapy in Graves’ ophthalmopathy. Lan-cet 1993;342:949–954.

12. Wilson WB, Prochoda M. Radiotherapy for thyroid orbitopa-thy. Effects on extraocular muscle balance. Arch Ophthal-mol 1995;113:1420–1425.

13. Takahashi T, Mitsuhashi N, Nagashima H, et al. Clinicalexperience of radiation therapy for Graves’ ophthalmopathy.Radiat Med 1996;14:343–347.

14. Bartalena L, Marcocci C, Tanda ML, et al. Orbital radio-therapy for Graves’ ophthalmopathy. Thyroid 2002;12:245–250.

15. Donaldson SS, Bagshaw MA, Kriss JP. Supervoltage orbital

FIGURE 5. Changes in proptosis before, and 1 and 6 monthsafter the therapy in (top) the pulse group and (bottom) thepulse-irradiation group. **P < .01 [repeated-measures analysisof variance (ANOVA)], *P < .05 (repeated-measuresANOVA).

STEROID PULSE AND RADIOTHERAPY ON TAOVOL. 135, NO. 3 289

radiotherapy for Graves’ ophthalmopathy. J Clin EndocrinolMetab 1973;37:276–285.

16. Brennan MW, Leone CR Jr, Janaki L. Radiation therapy forGraves’ disease. Am J Ophthalmol 1983;96:195–199.

17. Van Ruyven RL, Van Den Bosch WA, Mulder PG, Eijken-boom WM, Paridaens AD. The effect of retrobulbar irradi-ation on exophthalmos, ductions and soft tissue signs inGraves’ ophthalmopathy: a retrospective analysis of 90 cases.Eye 2000;14:761–764.

18. Bartalena L, Marcocci C, Chiovato L, et al. Orbital cobaltirradiation combined with systemic corticosteroids forGraves’ ophthalmopathy: comparison with systemic cortico-steroid alone. J Clin Endoclin Metab 1983;56:1139–1144.

19. Mourits MP, Koornneef L, Wiersinga WM, Prummel MF,Berghout A, Gaag R. Clinical criteria for the assessment ofdisease activity in Graves’ ophthalmopathy: a novel ap-proach. Br J Ophthalmol 1989;73:639–644.

20. Tallsted ML, Norberg R. Immunohistochemical staining ofnormal and Graves’ extraocular muscle. Invest OphthalmolVis Sci 1988;29:175–184.

21. Campbell JC. Immunology of Graves’ ophthalmopathy: ret-robulbar histology and histochemistry. Acta Endocrinol1989;121:9–16.

22. Imai Y, Odajima R, Inoue Y, Sishiba Y. Effect of growthfactors on hyaluronate and proteoglycan synthesis by ret-

roocular tissue fibroblasts of Graves’ ophthalmopathy inculture. Acta Endocrinol 1992;126:541–552.

23. Bahn RS, Heufelder AE. Retroocular fibroblasts: importanteffector cells in Graves’ ophthalmopathy. Thyroid 1992;2:89–94.

24. Morris J, Hay ID, Nelson RE, Jiang Nai S. Clinical utility ofthyrotropin receptor antibody assays: comparison of radiore-ceptor and bioassay methods. Mayo Clin Proc 1988;63:707–712.

25. Kazuo K, Fujikado T, Ohmi G, Hosoba J, Tano Y. Value ofthyroid stimulating antibody in the diagnosis of thyroidassociated ophthalmopathy of euthyroid patients. Br J Oph-thalmol 1997;81:1080–1083.

26. Ohtsuka K, Hashimoto M. Serum levels of soluble Fas inpatients with Graves’ ophthalmopathy. Br J Ophthalmol2000;84:103–106.

27. Gorman CA, Garrity JA, Farourechi V, et al. A prospective,randomized, double-blind, placebo-controlled study of or-bital radiotherapy for Graves’ ophthalmopathy. Ophthalmol-ogy 2001;108:1523–1534.

28. Miller ML, Goldberg SH, Bullock JD. Radiation retinopathyafter standard radiotherapy for thyroid-related ophthalmop-athy. Am J Ophthalmol 1991;112:600–601.

29. Kinyoun JL, Kalina RE, Brower SA, et al. Radiation retinop-athy after orbital irradiation for Graves’ ophthalmopathy.Arch Ophthalmol 1984;102:1473–1476.

AMERICAN JOURNAL OF OPHTHALMOLOGY290 MARCH 2003