Journal of Neuro-Ophthalmology, September 2007, Volume 27, Issue 3

Efficacy of Corticosteroids and External Beam Radiation in the Management of Thyroid Eye Disease

Thyroid Eye Disease (TED, Graves ophthalmopathy, thyroid ophthalmopathy) is the most common cause of orbital inflammation and proptosis in adults. There is no agreement on its management although corticosteroids and external beam orbital radiation (XRT) have traditionally been believed to provide benefit in active inflammation. Our review of the published literature in English disclosed an overall corticosteroid-mediated treatment response of 66.9% in a total of 834 treated patients who had moderate or severe TED. Intravenous corticosteroids used in repeated weekly pulses were more effective (overall favorable response = 74.6%, n = 177) and had fewer side effects than daily oral corticosteroids (overall favorable response = 55.5%, n = 265). A combination of corticosteroid and radiation therapy seemed to be more effective than corticosteroids alone. Our conclusions are tempered by a notable lack of standardization within and between study designs, treatment protocols, and outcome measures. Accordingly, the North American Neuro-Ophthalmology Society (NANOS), American Society of Ophthalmic Plastic and Reconstructive Surgery (ASOPRS) and the Orbital Society, in conjunction with Neuro-Ophthalmology Research and Development Consortium (NORDIC), will investigate the design and funding of a multi-center controlled trial.

STATE OF THE ART Efficacy of Corticosteroids and External Beam Radiation in the Management of Moderate to Severe Thyroid Eye Disease Christopher I. Zoumalan, MD, Kimberly P. Cockerham, MD, Roger E. Turbin, MD, Nicholas J. Volpe, MD, Michael Kazim, MD, Raymond S. Douglas, MD, and Steven E. Feldon, MD, for the Neuro- opthalmology Research and Development Consortium ( NORDIC) Thyroid Eye Disease ( TED) Study Commitee Abstract: Thyroid Eye Disease ( TED, Graves ophthalmopathy, thyroid ophthalmopathy) is the most common cause of orbital inflammation and proptosis in adults. There is no agreement on its management although corticosteroids and external beam orbital radiation ( XRT) have traditionally been believed to provide benefit in active inflammation. Our review of the published literature in English disclosed an overall corticosteroid- mediated treat­ment response of 66.9% in a total of 834 treated patients who had moderate or severe TED. In­travenous corticosteroids used in repeated weekly pulses were more effective ( overall favorable re­sponse = 74.6%, n = 177) and had fewer side effects than daily oral corticosteroids ( overall favorable response = 55.5%, n = 265). A combination of corticosteroid and radiation therapy seemed to be more effective than corticosteroids alone. Our con­clusions are tempered by a notable lack of stan­dardization within and between study designs, treatment protocols, and outcome measures. Accord­ingly, the North American Neuro- Ophthalmology Society ( NANOS), American Society of Ophthalmic Plastic and Reconstructive Surgery ( ASOPRS) and the Orbital Society, in conjunction with Neuro- Ophthalmology Research and Development Consor­tium ( NORDIC), will investigate the design and funding of a multi- center controlled trial. (/ Neuro- Ophthalmol 2007; 27: 205- 214) Thyroid eye disease ( TED, Graves ophthalmopathy, thyroid orbitopathy), which is associated with Graves disease ( GD) in over 80% of cases, is an autoimmune disorder characterized by inflammation and expansion of the orbital fat and extraocular muscles. Although it has been identified in all age groups, it primarily affects adults in the fourth and fifth decades. TED can profoundly impair a patient's ability to work and perform activities of daily living. Multiple scoring systems exist to grade the activity and severity of TED. There is, however, no consensus on the most accurate system, nor is there correlation between the currently available scoring systems ( 1- 3). The pathophysiology of TED is not completely understood but there is evidence for both humoral and cell-mediated immune processes ( 4- 7). Active phase TED results from lymphocytic infiltration of the orbital and periorbital fat and muscles. The active phase generally per­sists for six months to three years, and is typically longer in smokers and those with prolonged hypothyroidism ( 8- 10). The duration and severity of disease in an individual case, however, is unpredictable. After the inflammatory process ends, fibrosis and the associated disabling symptoms persist in the chronic, inactive phase. Immunomodulatory agents are believed to affect the activity of orbital lymphocytes and fibroblasts ( 9,11,12). Department of Ophthalmology ( CIZ, KPC), Stanford University School of Medicine, Stanford, California; Institute of Ophthalmology and Visual Science ( RET), University of Medicine and Dentistry- New Jersey Medical School, Newark, New Jersey; Department of Ophthalmology ( NJV), Scheie Eye Institute, University of Pennsylvania, Philadelphia, Pennsylvania; Columbia University Medical Center ( MK), Edward S. Harkness Eye Institute, New York, New York; Department of Ophthalmology ( RSD), Jules Stein Eye Institute, David Geffen School of Medicine at University of California, Los Angeles, Los Angeles, California; and the Department of Ophthalmology ( SEF), University of Rochester Eye Institute, University of Rochester, Rochester, New York. Address correspondence to Christopher Zoumalan, MD, Stanford University Medical Center Department of Ophthalmology, 900 Blake Wilbur Drive, Room W3002 Stanford, CA 94305; E- mail: zoumalan@ stanford. edu J Neuro- Ophthalmol, Vol. 27, No. 3, 2007 205 J Neuro- Ophthalmol, Vol. 27, No. 3, 2007 Zoumalan et al TABLE 1. Studies evaluating corticosteroid treatment of TED Authors Patients Bartalena et al ( 1983) 21 48 ( 36 PO steroids andXRT, 12 only PO steroids) Bartalena et al ( 1998) 31 75 Baschieri et al34 55 ( 30 PO steroids, 25 IVIg) Chang et al42 22 Dandona et al41 37 Hiromatsu et al 23 Kahaly et al33 70 ( 35 IV steroids and 35 PO steroids) Kazim et al43 30 Kendall- Taylor et al38 11 Koshiyama et al 8 Study inclusion criteria Mild to moderate active TED Needed to meet major categories ( variations in 2 mm of proptosis and lid width, diplopia, change in vision) and minor categories ( CAS, self assessment) Grades II- V NOSPECS ( not VI) Grades II- IV NOSPECS ( not V and VI) Not stated Grades II- V NOSPECS ( not VI) Defined as untreated, active, moderate TED Not clearly stated Not clearly stated Mod to severe TED Pretreatment patient profile Not clearly stated Pts with pre- existing TED given iodine and started on PO steroids 2 days after RAI Not clearly stated Not clearly stated Not clearly stated Not clearly stated Not clearly stated Not clearly stated 2 had prior immunosuppressive treatments for TED 2 of 8 already had prior IV pulse steroids Treatment study type Prospective case series in 36 pts that received combined PO steroids ( 70- 80 mg tapered over 6 months) and XRT. 12 pts that received only PO steroids were part of a prospective randomized study Prospective, randomized study. PO steroids ( 0.5 mg/ kg/ day prednisone X 1 month then with 8 wk PO taper Prospective, randomized, blinded study. 80 mg/ day PO steroids X 2 wks then tapered over 5 months Prospective case series. IV steroids 0.5 g/ day X 3 then 5 month PO steroid taper ( starting at 40 mg PO/ day then tapered) Case series, ( IV steroids 1 gm or 0.5 gm/ day X 3 with three week PO oral taper Prospective case series. 1 gm/ day IV steroids X 3, repeat 3- 5 times over 5 weeks ( total 9- 12 gm) followed by 30 mg/ day PO steroids X 1 month, then taper Randomized, single blind study. 35 received IV steroids 0.5 gm/ day X 6 wks ( once weekly), then down to 0.25 gm/ pulse IV steroids X 6 wks ( once weekly); 35 received PO steroids 0.1 gm/ day then taper for 12 wks by 0.01 g/ wk ( cumulative dose of 4.5 gm and 4.0 gm, respectively) Retrospective case series. 80- 120 mg/ day PO steroids tapered " over many months" Prospective case series. IV steroid ( 500 mg / day X 2) then with 40 mg PO steroid taper X 4 wks Prospective case series. IV steroids 1 gm/ day X 3 then tapered to 30- 40 mg/ day PO steroids with variable tapered length ( 6- 14 wks) 206 © 2007 Lippincott Williams & Wilkins Thyroid Eye Disease J Neuro- Ophthalmol, Vol. 27, No. 3, 2007 Adjunctive treatment Outcome measures Duration of follow up ( wks) Comment/ Conclusions 36 combined PO steroids and XRT ( 20 Gy X 2 wks) Clinical activity index, clinical assessment CAS and patient's impression 12 26/ 36 ( 72%) excellent or good response in combined group, 12/ 12 that received only PO steroids had regression or improvement in soft tissue changes, but only 33% had overall good results ( no excellent results reported). Proptosis improved in 19/ 36 ( 56%) of combined group while 5/ 11 ( 45%) improved with PO alone. 5/ 12 ( 41.7%) had improvement in EOM thickness in PO group. Recurrence occurred in 4/ 36 ( 11.1%) patients in combined group 52 50/ 75 ( 67%) improved/ regressed by CAS. Study showed PO steroids reduced the worsening of TED seen with RAI treatment Soft tissue changes, CT, proptosis, NOSPECS NOSPECS, CAS, CT Self assessment, eye exam, muscle size on CT 24 17 Not clearly stated 24/ 30 ( 80%) improved diplopia with PO steroids, 18/ 25 ( 75%) with IVIG; 76% response to NOSPECS with IVIG, 66% response with PO steroids, used CT to evaluate EOM size and found an average improvement in EOM thickness in both PO and IVIG groups 12/ 22 ( 55%) good response esp lacrimation, soreness, soft tissue swelling, proptosis; 4/ 10 poor responders got worse when steroids tapered to 20 mg or 10 mg/ day, found that improvement in CAS correlated well with improvement seen in EOM size on CT 32/ 37 ( 86.5%) improved with reduction in proptosis, 6/ 6 that were imaged had reduction in EOM NOSPECS class, MRI muscle size 24 12/ 23 ( 52.2%) improved diplopia and soft tissue swelling, decrease in mean proptosis values, 13/ 23 ( 56.5%) decrease muscle size on MRI XRT ( 20 Gy X 2 wks) given after completion of pulse IV steroids Proptosis and lid width, visual acuity, IOP in upgaze, diplopia, B- U/ S, CAS, self assessment survey Subjective improvement, improved fusion, proptosis, clinical exam Eye exam with IOP, CT EOM size, photos Diplopia, muscle size on MRI, NOSPECS 12 Favorable response in 27/ 35 ( 77%) of patients receiving IV steroids vs 18/ 35 ( 51%) with PO steroids on CAS; rapid improvement seen in IV group, diplopia improved or resolved in 44% ( 16/ 35) with IV steroids, only in 4/ 35 ( 11.4%) in PO, improved motility in 16/ 35 ( 46%) in IV and 9/ 35 ( 26%) in PO. Proptosis improved in 21/ 35 ( 61%) in IV group and 14/ 35 ( 30%) in PO group. Survey: 80% satisfied with IV vs 54% in PO. No PO steroid taper used for IV pulse steroids. Found a more significant improvement in EOM thickness via B- U/ S in patients that received IV than PO steroids 10/ 30 ( 33%) improved in some way. Proptosis and diplopia improved in 10/ 30 ( 30%); 6/ 16 with optic neuropathy improved, though 9 underwent additional treatment ( XRT of surgical decompression). One patient noted to have recurrence 6/ 7 ( 85.7%) with optic neuropathy improved, 9/ 11 ( 81.8%) improved soft tissue swelling, proptosis persisted in all patients, 8/ 9 ( 88.9%) improved EOM size on CT; of note, the 3 poor responders had TED > 1 year, 9 had CT's performed and 8 had a reduction of EOM size, though variable 3 yr ( 156 wks) 5/ 8 ( 62.5%) with eliminated diplopia, NOSPECS index improved on average, 6/ 8 ( 75%) decreased EOM size on MRI ( EOM size compared with optic nerve thickness), 7/ 8 ( 88%) excellent result, no recurrence seen 24 24 Continued on next page 207 J Neuro- Ophthalmol, Vol. 27, No. 3, 2007 Zoumalan et al TABLE 1. ( Continued) Authors Patients Study inclusion criteria Pretrearment patient profile Treatment study type Macchia et al J 51 ( 26 PO steroids and 25 IV steroids) Not clearly stated Marcocci et al ( 1987) 14 60 ( 30 XRT and PO steroids, 30 combined XRT and retrobulbar steroids Not clearly stated Marcocci et al ( 2001) 3' 82 ( 41 PO steroids, 41 Not clearly stated IV steroids) Matejka et al 8 Ophthalmology Index > 8 on NOSPECS Noth et aT Prummel et al ( 1989) 2, Not clearly stated 18 Severe NOSPECS ( Grade II- VI) Prummel et al ( 1993) 12 56 ( 28 PO steroids, 28 XRT) Staar et alJ Tagami et al 225 27 ( 11 XRT and IV steroids, 16 only IV steroids) Severe NOSPECS ( Grade II- VI) NOSPECS 2- 6, and orbitopathy index Not clearly stated None received prior treatment for TED Not clearly stated 12 received prior immunosuppressive treatment, 1 with prior orbital decompression Not clearly stated No clearly stated Not clearly stated None received prior treatment for TED 187 received prior immunosuppressive ( steroid) treatment Not clearly stated Randomized, prospective study 60- 80 mg PO steroids with 4- 6 month taper. IV group received 1000 mg/ day for two consecutive days each week for total of 6 weeks Prospective, randomized, controlled study. PO steroids group: 70- 80 mg/ day PO steroids X 3 wks then tapered over 5- 6 months, Retrobulbar steroids group: 14 injections ( methylprednisolone, 40 mg/ 1.5 mL) q20- 30 days X 9 months Prospective, single blind, r andomized study. 100 mg/ day PO steroid with taper over 22 wks, ( total dose 6 g) vs IV steroids 15 mg/ kg/ day X 2 days every 2 weeks for 4 cycles, repeat with 7.5 mg/ kg IV steroids X 2 every 2 wks for 4 cycles ( total dose 9- 12 gm) Prospective case series. IV 12.5 mg/ kg q 1 mo), then repeated 3- 6 times monthly, given PO steroids interpulse ( 0.5 mg/ kg/ day) then 4 wk PO steroid taper after last pulse IV dose Prospective case series. 20- 60 mg/ day PO steroids X 3 months Prospective, single blind, randomized study. 60 mg PO steroids/ day vs cyclosporine X 12 wks Prospective, double blind randomized trial. 60 mg PO steorids/ day for 4 wks then taper down over 20 wks Partly retrospective and prospective case series. 60 mg PO steroids simultaneously started with onset of XRT, PO steroids tapered over 6 wks Prospective case series. 1 gm/ pulse IV steroids X 3, repeat ql week X 4 ( if clinically indicated), then followed by 40- 50 mg/ day PO steroids, tapered over next 3- 12 months CAS, clinical activity score; RAI, radioactive iodine treatment; XRT, external beam orbital radiation; TED, thyroid eye disease; PO, oral; ( grade 1), soft tissue involvement with symptoms and signs ( grade 2), proptosis ( grade 3), extraocular muscle involvement ( grade 4), corneal 208 © 2007 Lippincott Williams & Wilkins Thyroid Eye Disease J Neuro- Ophthalmol, Vol. 27, No. 3, 2007 Adjunctive treatment None XRT ( 20 Gy X 2 wks), started at same time as steroids XRT ( 20 Gy X 2 wks) started one week after onset of steroids + / - XRT( 20Gy X 2 wks) Vs Cyclosporine Vs XRT ( 20 Gy X 2 wks) XRT ( 16- 19 Gy over 6 wks) 12 pts received XRT ( followed 2 wks after pulse IV steroids were completed) IV, intravenous; IVIg, intravenous Outcome measures NOSPECS, self assessment survey, proptosis Clinical exam, NOSPECS Proptosis, lid fissure width, diplopia, CAS Self assessment, NOSPECS, EOM and proptosis on CT Clinical exam, NOSPECS NOSPECS, proptosis, EOM size on CT Highest NOSPECS class, CT Subjective impression, NOSPECS Clinical score, CT or MRI of EOM, NOSPECS immunoglobin therapy; EOM. involvement ( grade 5), and sight involvement ( grade 6). Duration of follow up ( wks) 2 yrs 78 52 24 - 3 yrs ( 156 wks) 52 24 52 2 yrs ( 104 wks) , extraocular muscles; Comment/ Conclusions 21/ 25 ( 84%) had improvement with IV steroids, and 15/ 26 ( 57%) had improvement with PO steroids. Proptosis improved similarly in both groups. Four patients that received PO steroids had to withdraw therapy due to severe side effects. No recurrences were seen in a two- year follow up 19/ 30 ( 63%) improvement of EOM with PO steroids and XRT, overall excellent response in 18/ 30 ( 60%). Retrobulbar steroid and XRT group had overall improvement in 80%, 39% improvement in proptosis, and 17% improvement of EOM Overall improvement of 36/ 41 ( 88%) with IV steroids, ( 26/ 41) 63% with PO steroids; less side effects with IV ( 23/ 41 vs 35/ 41 in PO), CAS more improved with IV ( 36/ 41 vs 26/ 41 in PO); diplopia improved in 14/ 40 in IV, 12/ 40 in PO, optic neuropathy better result with IV ( 11/ 14 vs 3/ 9); no PO steroid taper used in IV steroids group 7/ 8 ( 87.5%) had improvement in all parameters, All eight had an improvement in proptosis on CT 5/ 11 ( 45.5%) had good result with PO steroids only, 6 went on to have XRT and 1 went on to have XRT and decompressive surgery 11/ 18 ( 61.1%) responded by EOM size on CT, clinical scores and proptosis at 12 wks; in non responders, combination with cyclosporine was helpful, followed for 52 wks but eventually 17/ 36 had surgery or XRT 14/ 28 ( 50%) responded to steroids, 13/ 28 ( 46%) responded to XRT, both had similar improvement in EOM size on CT, XRT seemed to improve motility more than PO steroids. Soft tissue swelling improved better with PO steroids than XRT Overall, 153/ 225 ( 68%) improved. Proptosis improved in 131/ 207 ( 64%), and diplopia impoved in 133/ 169 ( 78.7%). 72/ 225 ( 32%) eventually had orbital decompressive surgery over the course of the year follow up due to progression/ recurrence 21/ 27 ( 77.8%) diplopia improved ( all pts had diplopia) or disappeared; overall 9/ 11 ( 81.8%) improved in XRT and IV steroids group, and 12/ 16 ( 75%) improved in IV steroids group, 15/ 27 ( 55.6%) improved proptosis; NOSPECS improved as average across the group; used CT to compare EOM size ( by comparing the thickness to optic nerve) and found a correlation in EOM and NOSPECS improvement in responders NOSPECS, no signs or symptoms ( grade 0), only signs 209 J Neuro- Ophthalmol, Vol. 27, No. 3, 2007 Zoumalan et al The most commonly employed immunomodulators include corticosteroids with or without adjunctive external beam orbital radiation ( XRT) in moderate to severe cases of TED. However, disease management varies widely ( 13). The goals of medical therapy are to shorten the duration and minimize the severity of the active phase, thereby reducing the chronic phase disfigurement and disability produced by irreversible fibrosis. Corticosteroids are typically administered orally or intravenously ( IV). Local injections of corticosteroids into the orbit have failed to provide an effect in improving orbitopathy ( 14,15). Some studies have shown that local injections can improve motility and extraocular muscle ( EOM) size and can be a suitable alternative to patients with contraindications to systemic corticosteroids ( 16,17). XRT was first used empirically to treat TED. While the mechanism for its action is not fully understood, radiation ( XRT, typical total dose = 20 Gy) is biologically active against infiltrating lymphocytes, tissue- bound mono­cytes, and fibroblasts so as to alter the local cellular matrix and interrupt the inflammatory process in a more permanent fashion than can be achieved with corticosteroids ( 14,18). One recent prospective, double- masked, sham- controlled clinical trial produced more debate than consensus regarding the efficacy of XRT therapy for TED ( 19,20). There is no agreement on the management of TED ( 18,21,22). As an alternative to corticosteroids and XRT, other immunomodulatory agents such as azathioprine, cyclosporine, intravenous immunoglobulin ( IVIg), and plasmapheresis have been used, but they play a more limited role ( 23- 28). Efficacy studies have not been well modeled, and are mainly small, retrospective, or uncontrolled ( 18,21,22,29,30). Interpretation of the data from the existing studies is limited by the lack of good natural history data and the highly variable nature of the disease. We have evaluated the relevant publications in the English language to compare the outcomes of TED patients treated with corticosteroids and/ or XRT. METHODS We performed a review of published studies iden­tified in a PubMed on- line review from January 1966 to July 2006 using the following key words: Graves ophthal­mopathy, thyroid eye disease, Graves disease, and thyroid orbitopathy. Inclusion criteria required at least eight enrolled subjects within a retrospective or prospective study published in the English language. Our review compared the outcomes of using corticosteroids with or without XRT. We included only those studies that compared the outcomes through definable measurements, orbital imaging studies ( CT or MRI), self- assessment surveys, or clinical examinations. RESULTS Study Profiles We identified nineteen studies for our review. Seven had enrolled patients in randomized prospective studies ( 12,14,28,30- 33); the remainder were either prospective or retrospective case series. General Patient Treatment Profile A total of 834 patients from nineteen studies were reviewed. Study patient populations were dissimilar, and inclusion and exclusion criteria varied. In particular, some studies excluded patients who had had prior treatment for TED ( 12,14,31- 34). Others included patients that had already been treated with XRT, immunomodulatory agents, or decompressive surgery ( 21,28,30,35- 40). Still others failed to detail prior treatment ( 41- 44). The study of Marcocci et al ( 30) was the only one that detailed the thyroid metabolic status of the patients such that 81 ( 99.8%) of 82 patients presented with hyperthyroidism and TED; one patient had euthyroid TED. Of the 834 patients, 597 ( 71.6%) from 13 studies were treated with oral corticosteroids with or without XRT. Of the 597 patients, 265 had received only oral prednisone ( average of 76 +/- 26 mg/ day). The length of oral corticosteroid treatment ( including taper) averaged 17.5 +/- 5.4 weeks. In 10 studies, 237 patients ( 28.4%) were treated predominantly with IV corticosteroids ( methylpred-nisolone) with or without XRT. Seven of these studies mainly used 1000 mg/ pulse and the remaining three studies used a 500 mg/ pulse. The average number of pulses received was 5.8 +/- 3.8 over an average length of 13 +/- 1.5 weeks of treatment. Most of the patients who had received IV corticosteroids ( n = 177) without XRT had also received oral corticosteroids between pulses of IV cortico­steroids and were given a tapered course of oral cortico­steroids that averaged 37.9 +/- 4.9 mg/ day over 14 +/- 17 weeks. The follow- up interval for all studies averaged 59 weeks ( range 12 weeks to 3 years) ( Table 1). Clinical Measurement of Disease Severity and Treatment Response TED severity and activity were assessed using dif­ferent scoring systems, including the NO SPECS classifi­cation system ( No signs or symptoms [ grade 0]; Only signs [ grade 1]; Soft tissue involvement with symptoms and signs [ grade 2]; Proptosis [ grade 3]; Extraocular muscle involvement [ grade 4]; Corneal involvement [ grade 5]; Sight involvement [ grade 6]), American Thyroid Associ­ation classification, Stanford Score, International Index, clinical activity score ( CAS), and self- assessment surveys. Other data came from the clinical examination findings, including proptosis, gaze- evoked changes in intraocular pressure ( IOP), and from neuroimaging abnormalities. 210 © 2007 Lippincott Williams & Wilkins Thyroid Eye Disease J Neuro- Ophthalmol, Vol. 27, No. 3, 2007 The NOSPECS classification system was most commonly used ( 12 of 19 studies). It documents the presence of specific symptoms and signs of which only some are characteristic of active disease. The CAS was used in 6 of 19 studies. It takes into account seven clinical measurements and assigns a point to each symptom or sign ( retrobulbar pain, pain on eye movements, eyelid erythema, conjunctival injection, chemosis, swelling of the caruncle, eyelid edema or fullness). Many studies included self-assessment patient surveys ( Table 1). Neuroimaging Ten studies employed neuroimaging to measure the outcome of TED treatment ( 12,28,33,34,36- 38,40,42,44). CT, MRI, and B- ultrasound were used to evaluate EOM size. There was no standardized grading protocol ( Table 1). All studies showed an overall improvement in EOM thick­ness or proptosis after oral or IV corticosteroid treatment with and without the use of external beam radiation ( x- irradiation, XRT). The degree of improvement corre­lated well with the overall favorable response reported by clinical measures. Matejka et al ( 37) measured the amount of proptosis on CT and found an overall improvement in all eight patients who had received predominantly IV corticosteroids. The study of Hiromatsu et al ( 44) was unique in using MRI to judge response in patients treated predominantly with IV corticosteroids. The activity of TED was measured by MRI signal intensity of the EOMs as well as their thickness. Baschieri et al ( 34) found reduction in EOM thickness on CT after oral prednisone 80 mg/ day for a total two weeks with a five month taper. One of the studies employed B- ultrasound to note significant greater reduction in EOM thickness in patients who had received IV corticosteroids than in those who had received oral corticosteroids ( 33). Intravenous vs. Oral Corticosteroids Because of dissimilar assessment measures within and among studies, treatment outcomes are reported in relation to each study's outcome measures. We judged the results as favorable if the author reported the results as either good or excellent. Corticosteroids were the primary medical therapy used to treat active TED. There was an overall 66.9% favorable response to corticosteroid treatment among all 834 patients, which included those that may also have been treated with XRT. A total of 442 patients were treated with oral or predominantly IV corticosteroids alone. A total of 151 ( 55.5%) of 265 patients treated with oral cortico­steroids alone had a favorable response, and 132 ( 74.6%) of 177 patients treated with IV corticosteroids alone had a favorable response. Patients who had received predominantly IV cortico­steroids seemed to have greater improvement in diplopia, ocular motility, and proptosis than those who had received only oral corticosteroids. But patients who received only oral corticosteroids showed a greater improvement in EOM thickness than patients who received predominantly IV corticosteroids ( Table 2). Two prospective randomized studies compared IV to oral corticosteroid treatment ( 30,32,33). Neither of these studies used a tapered regimen of oral corticosteroids followed a regimen of IV corticosteroids. Kahaly et al ( 33) showed an improvement in the CAS in 27 ( 77%) of 35 patients treated with IV corticosteroids as compared to 18 ( 51%) of 35 patients treated with oral corticosteroids. Based on a self- assessment survey, 80% of patients receiving IV corticosteroids as compared to 54% receiving oral cortico­steroids were satisfied with the treatment results and had an improved quality of life. Macchia et al ( 32) reported similar results, such that 21 ( 84%) of 25 patients treated with IV corticosteroids had a favorable response as compared to 15 ( 57%) of 26 treated with oral corticosteroids. However, proptosis improved equally among both groups. Combined Corticosteroid and XRT Treatment The cumulative radiation dose and the radiation field were similar in nearly all studies ( Table 1). A total of 20 Gy TABLE 2. Outcomes following treatment Total patients Overall favorable response Improvement in diplopia Improvement in motility Improvement in proptosis Improvement in extraocular muscle thickness Cort, corticosteroid; XRT, Cort PO only 265 151/ 265 ( 57.0%) 50/ 135 ( 37.0%) 20/ 53 ( 37.7%) 40/ 94 ( 42.6%) 35/ 42 ( 83.3%) external beam x of patients with active thyroid eye Cort IV only 177 132/ 177 ( 74.6%) 61/ 122 ( 50.0%) 32/ 54( 59.3%) 100/ 163 ( 61.3%) 27/ 38( 71.1%) CortPO and XRT 332 223/ 332 ( 67.2%) 133/ 169 ( 78.7%) 170/ 273 ( 62.3%) 20/ 30 ( 66.6%) disease Cort IV and XRT 60 52/ 60 ( 86.7%) 14/ 19 ( 73.7%) 6/ 8 ( 75%) CortPO ( + / - XRT) 597 374/ 597 ( 62.6%) 40/ 105 ( 38.1%) 20/ 53 ( 37.7%) 31/ 48 ( 64.6%) 50/ 60 ( 83.3%) - irradiation; TV, intravenous administration; PO, oral administration. Cort IV ( + / - XRT) 237 184/ 237 ( 77.6%) 75/ 141 ( 53.2%) 32/ 64 ( 50.0%) 79/ 117 ( 67.5%) 33/ 46 ( 71.2%) 211 J Neuro- Ophthalmol, Vol. 27, No. 3, 2007 Zoumalan et al was delivered to each orbit over a period of two weeks. There was an overall 70.2% favorable response in the 392 patients who underwent both XRT and IV or oral corticosteroid treatment. Compare this to an overall 64.0% favorable response in the 442 patients who received IV or oral corticosteroids without XRT. More specifically, 223 ( 67.2%) of 332 patients who received oral cortico­steroids and XRT ( 14,21,30,35,39) responded favorably as compared to 52 ( 86.7%) of 60 patients who received predominantly IV corticosteroids and XRT ( 30,36,40) ( Table 2). One prospective randomized study compared the outcome following IV corticosteroids and XRT to the outcome following oral corticosteroids and XRT. Marcocci et al ( 30) found a significantly greater short- term improve­ment of periocular edema, erythema, orbital ache, and ocular motility in patients who had received IV cortico­steroids and XRT ( 88%) when compared to oral cortico­steroids and XRT ( 63%). Proptosis improved in 19 ( 47.5%) of 40 patients receiving IV corticosteroids and XRT and in 16 ( 40%) of 40 patients receiving oral corticosteroids and XRT. Optic neuropathy improved in 13 ( 92%) of 14 patients after treatment with IV corticosteroids and XRT, but only in 3 ( 33%) of 9 patients who received oral corticosteroids and XRT. This difference, however, was not statistically significant ( 30). Intravenous vs. Oral Corticosteroid Treatment With or Without Adjunctive XRT Overall, treatment with pulsed IV corticosteroids was more effective and better tolerated than chronic treatment with oral corticosteroids alone. Oral corticosteroids pro­duced a favorable response in 62.6% of patients with or without the use of XRT ( n = 597). This compares to a favorable response in 77.6% of patients treated with IV corticosteroids with or without XRT ( n - 237). In the studies reporting results, the patients who received predominantly IV corticosteroids showed greater improve­ment in diplopia, ocular motility, and proptosis in comparison to those who received oral corticosteroids with or without adjunctive XRT. In contrast, the patients who received oral corticosteroids showed a greater improvement in EOM thickness in comparison to the patients who received predominantly IV corticosteroids with or without adjunctive XRT ( Table 2). Corticosteroid Side Effects There was a higher rate of side effects to cortico­steroids administered via the oral route than to cortico­steroids administered via the IV route. Chronic oral corticosteroid treatment was associated with Cushingoid facies, weight gain, osteoporosis, gastric irritation, labile hypertension, elevated intraocular pressure, elevation in blood sugar, and mood alteration. Marcocci et al ( 30) documented that 35 ( 85.4%) of 41 patients treated with an approximately six- month oral corticosteroid taper ( starting at 100 mg prednisolone by mouth daily for a cumulative dose of 6 grams) demonstrated side effects including weight gain, urinary tract infections, transient hyperglyce­mia, and decreased bone mineral density. Three patients who had received oral steroids in the randomized study of Macchia et al ( 32) had to withdraw from their treatment due to " severe signs or symptoms of hypercortisolism." Prummel et al ( 12) found that 25 of 28 patients who received a 20- week oral corticosteroid taper ( starting at 60 mg prednisolone by mouth daily for four weeks followed by a taper) experienced only minor side effects. One patient developed depression and a second patient man­ifested a recurrent herpetic zoster eruption. Baschieri et al ( 34) reported two cases of hemorrhagic gastritis in patients receiving 80 mg prednisone by mouth daily with a 5- month taper. One patient developed bipolar disorder. More fre­quent side effects included Cushingoid facies ( 5 out of 30 patients) and abnormal glucose tolerance ( 5 out of 30 patients). Intravenous corticosteroid treatment was associated with a lower rate of adverse side effects. Kahaly et al ( 33) reported adverse events in only 6 ( 17%) of 35 patients, including weight gain, insomnia, palpitations, and gastro­intestinal discomfort. However, Marcocci et al ( 30) reported adverse effects in 23 ( 56.1%) of 41 patients, including urinary tract infections and impaired glucose tolerance. Nine patients inexplicably had a mean percent­age increase in bone mineral density after IV corticosteroid treatment. One patient had transient elevation of serum aminotransferase levels ( 34). Radiation Side Effects Radiation was well tolerated and produced few short-term side effects. Koshiyama et al ( 36), Marcocci et al ( 30), Staar et al ( 39), and Prummel et al ( 12) reported no side effects from radiation. In the study of Bartalena et al ( 21), with a follow- up over 26 months, there were no new cataracts. However, Prummel et al ( 12) found that 15 ( 54%) of 28 patients surveyed had side effects, usually minor, including transient hair loss at temples, tiredness, myalgias, headaches, insomnia, and nausea. Reactivation of TED After Treatment Seven of the reviewed studies addressed the inci­dence and timing of disease recurrence after successful initial treatment ( 21,32,35,36,39,42,43). The study of Koshimaya et al ( 36) found no recurrence of TED among all eight patients who had received predominantly IV corticosteroids and combined XRT after a three- year follow- up. The study of Macchia et al ( 32) found no 212 © 2007 Lippincott Williams & Wilkins Thyroid Eye Disease J Neuro- Ophthalmol, Vol. 27, No. 3, 2007 recurrence of TED among 51 patients who had received either oral ( n = 26) or IV corticosteroids ( n = 25). However, four other studies reported substantial recurrence in treated patients. Chang et al ( 42) reported that 4 of 10 patients worsened when the oral corticosteroids were discontinued or quickly tapered to 20 mg/ day over a course of five months. Noth et al ( 35) documented that 12 ( 63.2%) of 19 patients followed for three years required further immunosuppressive therapy and XRT due to disease recurrence or progression. Staar et al ( 39) also reported that because of disease progression or recurrence, 72 ( 32%) of 225 patients eventually proceeded to orbital decom­pressive surgery after a year's treatment with a combination of oral corticosteroids and XRT. Summary of Outcomes A review of the published literature in English has suggested that corticosteroid- mediated treatment produces benefit in 66.9% in TED patients who have moderate to severe disease. IV corticosteroids used in repeated daily or weekly pulses ( average of 5.8 pulses/ treatment epoch) were more effective than daily oral corticosteroids. Intravenous administration of corticosteroids appeared to be more effective than oral administration alone. A combination of corticosteroids and XRT was more effective than cortico­steroids alone. In combination with XRT, IV administration of corticosteroids was more effective than oral administration. Intravenous corticosteroids were associated with fewer side effects than oral corticosteroids. However, single case reports not included in this series have reported fatal cardiac and fatal hepatic necrosis with the use of IV corticosteroids for TED ( 45,46). XRT appeared to be well tolerated with few if any side effects. However, a report by Gorman et al ( 20), not included in this review, discovered newly dilated capillaries or microaneurysms on fluorescein angiograms or fundus photographs in five eyes among 3 of 37 treated patients three years after receiving XRT. Among the three prospective randomized studies, IV corticosteroids had a clear benefit in treatment outcome over oral corticosteroids ( 30,32,33). Patients treated with IVor oral corticosteroids showed improvement in proptosis, diplopia, ocular motility, and in self assessment of benefit, but the treatment outcomes were more substantial in those treated with IV corticosteroids ( over 77% favorable out­come) than with oral corticosteroids ( up to 62% favorable outcome). The only study ( 12) that compared oral cor­ticosteroid administration to XRT in a prospective, double-blind randomized trial showed similar treatment outcomes ( self assessment, EOM size on CT), but XRT seemed to improve ocular motility more than did oral corticosteroids, while oral corticosteroids seemed to improve soft tissue swelling more effectively. Cautions Caution is warranted regarding the interpretation of outcomes in the studies we have reviewed. The studies differed in design, treatment protocol, and outcome measures. The potential clinical impact of corticosteroids and XRT in the treatment of TED was difficult to assess reliably. These studies provide little insight regarding pathophysiology or effect of treatment on quality of life. It was particularly difficult to interpret the results of TED severity and activity among different scoring systems. NOSPECS was the most commonly used scoring system yet it documents manifestations not always characteristic of active disease. Clinical worsening may not represent increased inflammatory activity but rather progressive fibrosis associated with resolving inflammation. CAS, another scoring system used in several of the studies reviewed, does not provide information regarding overall progression or severity of TED. Self- assessment surveys used to document improvements in quality of life are non-standardized and difficult to interpret across studies. The identification of active TED remains an imperfect com­bination of the patient's impression and the clinician's interpretation of the physical signs. We conclude that there is inadequate case- based evidence to ascertain reliably whether medical therapy with corticosteroids or XRT shortens the active phase of disease or improves long- term disfigurement and disability in patients with moderate to severe TED. To answer this question more rigorously, the North American Neuro- Ophthalmology Society ( NANOS), American Society of Ophthalmic Plastic and Reconstructive Surgery ( ASOPRS), and the Orbital Society, working in conjunction with Neuro- Ophthalmology Research and Development Consortium ( NORDIC), have established a committee to pursue the design and funding of a large, multi- center, double- masked, placebo controlled study. REFERENCES 1. Gorman CA. The measurement of change in Graves' ophthalmop­athy. Thyroid 1998; 8: 539^ 3. 2. Dickinson AJ, Perros P. Controversies in the clinical evaluation of active thyroid- associated orbitopathy: use of a detailed protocol with comparative photographs for objective assessment. Clin Endocrinol ( Oxf) 2001; 55: 283- 30. 3. Mourits MP, Koornneef L, Wiersinga WM, et al. Clinical criteria for the assessment of disease activity in Graves' ophthalmopathy: a novel approach. Br J Ophthalmol 1989; 73: 639^ 4. 4. Bahn RS, Heufelder AE. Pathogenesis of Graves' ophthalmopathy. N Engl J Med 1993; 329: 1468- 75. 5. Jacobson DM. Dysthyroid orbitopathy. Semin Neurol 2000; 20: 43- 54. 6. Kazim M, Goldberg RA, Smith TJ. Insights into the pathogenesis of thyroid- associated orbitopathy: evolving rationale for therapy. Arch Ophthalmol 2002; 120: 380- 6. 7. Werner SC. Classification of the eye changes of Graves'disease. Am J Ophthalmol 1969; 68: 646- 8. 213 J Neuro- Ophthalmol, Vol. 27, No. 3, 2007 Zoumalan et al 8. Clauser L, Galie M, Sarti E, et al. Rationale of treatment in Graves ophthalmopathy. Plast Reconstr Surg 2001; 108: 1880- 94. 9. Kennerdell JS, Cockerham KP, Maroon JC, et al. Dysthyroid Orbitopathy. In: Kennerdell JS, Cockerham KP, Maroon JC, et al. Practical Diagnosis and Management of Orbital Disease. Boston: Butterworth- Heinemann Press; 2001: 53- 75. 10. Eckstein A, Quadbeck B, Mueller G, et al. Impact of smoking on the response to treatment of thyroid associated ophthalmopathy. Br J Ophthalmol 2003; 87: 773- 6. 11. Sergott RC, Glaser JS. Graves' ophthalmopathy. A clinical and immunologic review. Surv Ophthalmol 1981; 26: 1- 21. 12. Prummel MF, Mourits MP, Blank L, et al. Randomized double- blind trial of prednisone versus radiotherapy in Graves' ophthalmopathy. Lancet 1993; 342: 949- 54. 13. Weetman AP, Wiersinga WM. Current management of thyroid-associated ophthalmopathy in Europe. Results of an international survey. Clin Endocrinol ( Oxf) 1998; 49: 21- 8. 14. Marcocci C, Bartalena L, Panicucci M, et al. Orbital cobalt irradiation combined with retrobulbar or systemic corticosteroids for Graves' ophthalmopathy: a comparative study. Clin Endocrinol ( Oxf) 1987; 27: 33^ 2. 15. Trobe JD, Glaser JS, Laflamme P. Dysthyroid optic neuropathy. Clinical profile and rationale for management. Arch Ophthalmol 1978; 96: 1199- 209. 16. Ebner R, Devoto MH, Weil D, et al. Treatment of thyroid associated ophthalmopathy with periocular injections of triamcinolone. Br J Ophthalmol 2004; 88: 1380- 6. 17. Garber MI. Methylprednisolone in the treatment of exophthalmos. Lancet 1966; 1: 958- 60. 18. Hatton MP, Rubin PA. The pathophysiology of thyroid-associated ophthalmopathy. Ophthalmol Clin North Am 2002; 15: 113- 9. 19. Gorman CA, Garrity JA, Fatourechi Y et al. A prospective, randomized, double- blind, placebo- controlled study of orbital radio­therapy for Graves' ophthalmopathy. Ophthalmology 2001; 108: 1523- 34. 20. Gorman CA, Garrity JA, Fatourechi Y et al. The aftermath of orbital radiotherapy for graves' ophthalmopathy. Ophthalmology 2002; 109: 2100- 7. 21. Bartalena L, Marcocci C, Chiovato L, et al. Orbital cobalt irradiation combined with systemic corticosteroids for Graves' ophthalmopathy: comparison with systemic corticosteroids alone. J Clin Endocrinol Metab 1983; 56: 1139^ 4. 22. Prummel MF, Bakker A, Wiersinga WM, et al. Multi- center study on the characteristics and treatment strategies of patients with Graves' orbitopathy: the first European Group on Graves' Orbitopathy experience. Eur J Endocrinol 2003; 148: 491- 5. 23. Atabay C, Schrooyen M, Zhang ZG, et al. Use of eye muscle antibody measurements to monitor response to plasmapheresis in patients with thyroid- associated ophthalmopathy. J Endocrinol Lnvest 1993; 16: 669- 74. 24. Burrow GN, Mitchell MS, Howard RO, et al. Immunosuppressive therapy for the eye changes of Graves' disease. J Clin Endocrinol Metab 1970; 31: 307- 11. 25. Claridge KG, Ghabrial R, Davis G, et al. Combined radiotherapy and medical immunosuppression in the management of thyroid eye disease. Eye 1997; 11: 717- 22. 26. Yamamoto K, Saito K, Takai T, et al. Treatment of Graves' ophthalmopathy by steroid therapy, orbital radiation therapy, plasmapheresis and thyroxine replacement. Endocrinol Jpn 1982; 29: 495- 501. 27. Antonelli A, Saracino A, Alberti B, et al. High- dose intravenous immunoglobulin treatment in Graves' ophthalmopathy. Acta Endo­crinol ( Copenh) 1992; 126: 13- 2. 28. Prummel MF, Mourits MP, Berghout A, et al. Prednisone and cyclosporine in the treatment of severe Graves' ophthalmopathy. N Engl J Med 1989; 321: 1353- 9. 29. Bartalena L, Marocci C, Bogazzi F, et al. Glucocorticoid therapy of Graves' ophthalmopathy. Exp Clin Endocrinol 1991; 97: 320- 7. 30. Marcocci C, Bartalena L, Tanda ML, et al. Comparison of the effectiveness and tolerability of intravenous or oral glucocorti­coids associated with orbital radiotherapy in the management of severe Graves' ophthalmopathy: results of a prospective, single-blind, randomized study. J Clin Endocrinol Metab 2001; 86: 3562- 7. 31. Bartalena L, Marcocci C, Bogazzi F, et al. Relation between therapy for hyperthyroidism and the course of Graves' ophthalmopathy. N Engl J Med 1998; 338: 73- 8. 32. Macchia PE, Bagattini M, Lupoli G, et al. High- dose intravenous corticosteroid therapy for Graves' ophthalmopathy. J Endocrinol Lnvest 2001; 24: 152- 8. 33. Kahaly GJ, Pitz S, Hommel G, et al. Randomized, single blind trial of intravenous versus oral steroid monotherapy in Graves' orbitopathy. J Clin Endocrinol Metab 2005; 90: 5234^ 0. 34. Baschieri L, Antonelli A, Nardi S, et al. Intravenous immunoglobulin versus corticosteroid in treatment of Graves' ophthalmopathy. Thyroid 1997; 7: 579- 85. 35. Noth D, Gebauer M, Muller B, et al. Graves' ophthalmopathy: natural history and treatment outcomes. Swiss Med Wkly 2001; 131: 603- 9. 36. Koshiyama H, Koh T, Fujiwara K, et al. Therapy of Graves' ophthalmopathy with intravenous high- dose steroid followed by orbital irradiation. Thyroid 1994; 4: 409- 13. 37. Matejka G, Verges B, Vaillant G, et al. Intravenous methylprednis­olone pulse therapy in the treatment of Graves' ophthalmopathy. Norm Metab Res 1998; 30: 93- 8. 38. Kendall- Taylor P, Crombie AL, Stephenson AM, et al. Intravenous methylprednisolone in the treatment of Graves' ophthalmopathy. Bmj 1988; 297: 1574- 8. 39. Staar S, Muller RP, Hammer M, et al. Results and prognostic factors in retrobulbar radiotherapy combined with systemic corticosteroids for endocrine orbitopathy ( Graves'disease). Front Radiat Ther Oncol 1997; 30: 206- 17. 40. Tagami T, Tanaka K, Sugawa H, et al. High- dose intravenous steroid pulse therapy in thyroid- associated ophthalmopathy. Endocr J 1996; 43: 689- 99. 41. Dandona P, Havard CW, Mier A. Methylprednisolone and Graves' ophthalmopathy. Bmj 1989; 298: 830. 42. Chang TC, Huang KM, Hsiao YL, et al. Relationships of orbital computed tomographic findings and activity scores to the prognosis of corticosteroid therapy in patients with Graves' ophthalmopathy. Acta Ophthalmol Scand 1997; 75: 301^ k 43. Kazim M, Trokel S, Moore S. Treatment of acute Graves orbitopathy. Ophthalmology 1991; 98: 1443- 8. 44. Hiromatsu Y, Kojima K, Ishisaka N, et al. Role of magnetic resonance imaging in thyroid- associated ophthalmopathy: its pre­dictive value for therapeutic outcome of immunosuppressive therapy. Thyroid 1992; 2: 299- 305. 45. Gursoy A, Cesur M, Erdogan MF, et al. New- onset acute heart failure after intravenous glucocorticoid pulse therapy in a patient with Graves' ophthalmopathy. Endocrine 2006; 29: 513- 6. 46. Weissel M, Hauff W Fatal liver failure after high- dose glucocorticoid pulse therapy in a patient with severe thyroid eye disease. Thyroid 2000; 10: 521. 214 © 2007 Lippincott Williams & Wilkins