The treatment of thyroid-associated. High dose intravenous methylprednisolone pulse therapy versus oral prednisone for thyroidassociated

Transcription

1 High dose intravenous methylprednisolone pulse therapy versus oral prednisone for thyroidassociated ophthalmopathy Ritva Kauppinen-Mäkelin 1, Anni Karma 2, Eeva Leinonen 1, Eliisa Löyttyniemi 3, Oili Salonen 4, Timo Sane 1, Kirsi Setälä 2, Jorma Viikari 5, Armin Heufelder 6 and Matti Välimäki 1 1 Department of Medicine, Helsinki University Central Hospital, Finland 2 Department of Ophthalmology, Helsinki University Central Hospital, Finland 3 Department of Statistics, University of Turku, Finland 4 Department of Diagnostic Radiology, Helsinki University Central Hospital, Finland 5 Department of Medicine, Turku University Central Hospital, Finland 6 Medizinische Klinik, Klinikum Innenstadt, Ludwig-Maximilians-University, Munich, Germany ABSTRACT. Purpose: To compare the effectiveness of intravenous (i.v.) methylprednisolone pulse therapy and oral prednisone when used as the initial treatment of patients with mild or moderate thyroid-associated ophthalmopathy. Methods: Thirty-three consecutive patients with thyroid-associated ophthalmopathy in Helsinki and Turku University Hospitals were randomly assigned either i.v. methylprednisolone pulse therapy (group A, n Ω 18) or oral prednisone (group B, n Ω 15). Treatment outcomes were measured by subjective changes in the grade of diplopia and quantitatively in several ophthalmic variables at 3 and 12 months. Any decision to proceed with additional treatment at 3 months was made on clinical grounds. The study was open in respect of both the initial treatment and the need for additional therapy. Results: No significant differences in the grade of diplopia, proptosis or soft tissue activity scores were noted between groups A and B from 0 to 3 months. However, group A required additional forms of therapy at 3 months less frequently than did group B (p Ω0.038). Conclusions: Our data suggest that i.v. methylprednisolone pulse therapy and oral prednisone are equally effective as initial treatments for thyroid-associated ophthalmopathy where diplopia, proptosis and signs of soft tissue inflammation are concerned. When additional treatment is required, i.v. methylprednisolone pulse therapy may be more effective than oral prednisone. However, the study s limitations meant that any decision to give additional treatment after the initial therapy was made on clinical grounds. Key words: exophthalmos bone mineral density cell adhesion molecule corticosteroid treatment cytokines Graves disease ophthalmopathy thyroid disease Acta Ophthalmol. Scand. 2002: 80: Copyright c Acta Ophthalmol Scand ISSN The treatment of thyroid-associated ophthalmopathy has so far been largely empirical and unsatisfactory. Glucocorticoids have been administered orally (Prummel et al. 1989) and intravenously (i.v.) as pulse therapy (Nagayama et al. 1987; Kendall-Taylor et al. 1988). The role of orbital radiotherapy has remained controversial (Kriss et al. 1989; Prummel et al. 1993; DeGroot et al. 1995; Mourits et al. 2000; Prummel 2000). Decompression surgery is frequently used as a last resort to promptly relieve intraorbital pressure in patients with optic nerve compression or exposure keratitis (Burch & Wartofsky 1993). Clinical studies comparing various modes of treatment for thyroid-associated ophthalmopathy are scarce (Wiersinga & Prummel 2000). No comparative studies between i.v. methylprednisolone and oral prednisone as a single therapy have been published, but Marcocci et al. (2001) recently reported higher efficacy of a combination of i.v. methylprednisolone pulse therapy and orbital radiotherapy compared to oral prednisone and radiotherapy in severe thyroid-associated ophthalmopathy. The present study aims to determine whether there is a difference in outcome between treatment with high dose i.v. methylprednisolone pulse therapy and treatment with oral prednisone. Patients and Methods Patients Thirty-three patients with thyroid-associated ophthalmopathy were studied at 316

2 either Helsinki or Turku University Central Hospitals in Finland. The inclusion criteria were: (1) Previous or concurrent thyroid-associated hyperthyroidism as ascertained by diffuse uptake on thyroid scan and by the presence of thyroid stimulating antibodies (TSab) in serum. (2) Soft tissue symptoms of ophthalmopathy with an activity score of at least 3, according to the modified method described by Mourits et al. (1989), or proptosis (Hertel exophthalmometer reading of 20 mm or more), or the presence of diplopia. (3) Duration of ophthalmopathy of less than 1year. Exclusion criteria were corneal involvement (exposure keratitis, corneal ulceration, clouding or necrosis) or optic nerve compression. A decrease in visual acuity (VA) described by the patient and/ or a colour vision defect (four or more wrong plates out of 12) when assessed by Ishihara plates were considered to be clinical signs for suspicion of optic nerve compression. Suspicions were confirmed by the ophthalmologist by the presence of delayed visual evoked responses and a visual field defect in the Goldmann perimetry. Patients with optic nerve compression were immediately referred for decompression surgery and excluded from the study. Soft tissue involvement was scored by the modified method described by Mourits et al. (1989). Six of the original 10 categories were assessed, namely retrobulbar pain, eyelid erythema, eyelid oedema, conjunctival redness, chemosis and caruncle swelling. One point was given for each parameter, with a score of 6 points representing the maximum activity score. The presence or absence of diplopia and proptosis, VA and intraocular pressure (IOP) are reported separately. Diplopia was graded as mild when it occurred only in the extreme gaze, and severe when it also occurred in the primary position. It was considered as moderate if it did not occur in the primary position but did otherwise disturb the patient in everyday life. Study design Patients were randomly assigned either i.v. methylprednisolone pulse therapy or oral prednisone as initial therapy. Treatment information was sealed in envelopes to ensure randomization, with a 1:1 ratio for treatment modalities in blocks of 10. The i.v. methylprednisolone group received 500 mg of methylprednisolone in 250mL of physiologic saline as a 30min i.v. infusion (Kendall-Taylor et al. 1988). Infusion was repeated after 48h. On the day after the second infusion, patients were started on oral prednisone as follows: 40mg/day for 1week, 30mg/day for 1week, 20mg/day for 1week and 10mg/ day for 1week. The administration of 500 mg i.v. methylprednisolone was then repeated twice with a 48-hour interval between doses. Thereafter, oral prednisone was given as described above except that the prednisone at 10 mg/day was continued for up to 4weeks, followed by 5 mg/day for 1week and 5mg every other day for 1 week. The total cumulative methylprednisolone plus prednisone dose was 3660 mg over 14 weeks. If 4 mg of methylprednisolone equals 5 mg of prednisone, the total glucocorticoid dose was 4160 mg. The oral prednisone group was given prednisone at 60mg/day for 2weeks, 40 mg/day for 2 weeks, 30 mg/day for 4 weeks, 20mg/day for 4weeks, 10mg/day for 2weeks, 5mg/day for 1week and finally 5 mg every other day for 1 week (Prummel et al. 1989). The total cumulative prednisone dose was 2990mg over 16 weeks. Follow-up visits were at 1, 3, 6, 9 and 12months. At the 3month visit, all patients were assessed to ascertain any requirements for additional therapy. Therapy was given if the disease process was persistent or progressive, with increased or unchanged proptosis, development of diplopia, or persistent diplopia either in the primary position or otherwise disturbing the patient in everyday life, or marked soft tissue involvement in spite of therapy. If additional therapy was necessary, patients in the methylprednisolone pulse group were referred for retrobulbar radiotherapy, whereas patients in the oral prednisone group were either given i.v. methylprednisolone pulse therapy as described above or referred for radiotherapy according to clinical judgement. If no further therapy was required, prednisone was tapered off during the next 2 4weeks. Patients were referred for decompression surgery at any point in the study if they developed proptosis with established or impending severe corneal defects, or showed signs of optic nerve compression. One of the five internist investigators met the patients at every visit and filled in the clinical eye status form (not shown). One of the two ophthalmologists examined the patients at 0, 3, 6 and 12 month visits and filled in the ophthalmologist s form (not shown). Ophthalmologists and internists used the same methods for assessment of soft tissue activity. The grade of proptosis was measured by Hertel exophthalmometer. Eye movements, pupillary reactions, IOP (with the patient looking straight ahead) and fundus examinations were also documented. A Hess test was performed at nearly every visit and Goldmann perimetry was carried out when indicated. Blood samples were drawn at every visit for measurements of free thyroxine (FT4), total triiodothyronine (T3), thyrotropin (TSH), thyroid stimulating antibodies (TSab), soluble interleukin-1-receptor antagonist (IL-1RA) serum levels using an immunosorbent assay (R&D Systems, Abingdon, UK). Soluble cell adhesion molecules, intercellular adhesion molecule-1 (ICAM-1), endothelial-leucocyte adhesion molecule-1 (ELAM-1), and CD44st were also monitored (Bender MedSystems, Vienna, Austria). Orbital computed tomography (CT) (Siemens Somaton HiQ, Erlangen, Germany) was performed at 0, 6 and 12 months, as were bone mineral density (BMD) measurements of the lumbar spine and the upper femur by dual energy X-ray absorptiometry using a Hologic QDR-1000 densitometer (Hologic Inc., Waltham, Massachusetts, USA). Eye photographs were taken at 0, 3, 6 and 12 months. In orbital CT, axial and coronal planes with 3mm thick slices were taken. The interslice space was 3mm for axial scans and 10mm for coronal images. The thickness of the medial and lateral muscles was always measured in the same plane on a single axial section, and that of the superior and inferior muscles in a single coronal section, respectively, by the same radiologist. The study was open in respect of both the initial treatment and the need for additional treatment. The study protocol was approved by the ethical committees of Helsinki and Turku University Central Hospitals. Informed consent was obtained from patients prior to entry to the study. Statistical analysis The equality of means between patient groups A and B at baseline was tested by one-way ANOVA. Frequencies were tested by frequency tables using 2-tailed Fisher s exact test. The effect of the two therapies on continuous variables was analysed by 317

3 Table 1. Baseline characteristics (mean SEM) of group A (i.v. methylprednisolone) and group B (oral prednisone) patients. Group A Group B p Number of patients Sex F/M 17/1 14/ Age (years) Age range Smoker no/yes/not known 6/8/4 5/10/ Radioiodine * no/yes 13/5 11/ FT4 (pmol/l) T3 (nmol/l) TSH (mu/l) TSab (%) * Before entry to the study. ANOVA with repeated measures. The effect of the treatments on proptosis and IOP was analysed by ANOVA with repeated measures, with time and eye being the two within factors. Timepoints 0 and 3 months were included in the analyses. The chosen covariance structure was compound symmetry. All tests were performed as two-sided. The significance level was set at In case of significant results, linear contrasts were created to determine where differences occurred. Visual acuity measurements were tested with the Kruskal-Wallis test, using both the minimum and mean values for each eye. Tests were performed at each timepoint and for changes between timepoints. Friedman and Kruskal-Wallis tests were also used to evaluate the changes in activity scores within and between the groups. Results are given as mean SEM unless otherwise indicated. Results Characteristics of patients All 33 patients completed the study. Eighteen patients received i.v. methylprednisolone therapy (group A) and 15 received oral prednisone therapy (group B). The groups were comparable with respect to age and sex distribution. There were no statistically significant differences between groups in thyroid function tests, but TSab titres were significantly higher in group B (Table1). There were no differences between the groups in terms of the number of patients using methimazole or in the dose of methimazole, or in the number of patients using thyroxine at baseline or during follow-up. Five patients in group A and four patients in group B had received radioiodine therapy prior to the onset of ocular symptoms. Ocular findings Diplopia. There was no statistically significant difference between the groups in the change in diplopia between baseline and the 3month visit. At baseline, clinically significant diplopia was present in four of 17 patients with a known history of diplopia in group A, and in one of 15 patients in group B (p Ω0.338) (Table2). At 3 months, diplopia was reported by one of 18 patients in group A and by one of 15 patients in group B (pω1.000). Proptosis. There were no significant differences between the two groups in Hertel exophthalmometer readings at baseline (Table2). During the 3months of treatment, exophthalmometer readings of patients in group A decreased slightly, but did not change in group B (Table2). A trend towards a significant difference between the two treatment groups was noted (p Ω0.066). Activity score. The activity score of inflammation decreased in both groups between baseline and 3 months (p Ω0.001) (Table 2). However, activity scores decreased to a similar extent in both groups Table 2. Ocular findings in group A (i.v. methylprednisolone) and group B (oral prednisone) patients at baseline and at 3 and 12 months. Proptosis and IOP are given as least square means, based on a statistical model. Visual acuity is given as median and minimum and maximum values (mean value for both eyes). p indicates difference between groups A and B between baseline and 3 months Group A Group B Baseline 3 months 12 months Baseline 3 months 12 months p Diplopia* No/yes 13/4 17/1 17/0 14/1 14/1 15/0 Baseline months Proptosis (mm) Right eye Left eye Activity score Mean SEM Baseline /3 4/5 6 5/5/2 9/1/1 10/2/0 3/8/2 8/3/1 10/2/1 3 months Visual acuity Median Min Max Pressure (mmhg) Right eye Left eye * Diplopia in primary position or otherwise disturbing the patient. 318

4 Fig. 1. Requirements for additional therapy in patients treated with i.v. methylprednisone pulse therapy (group A) and patients treated with oral prednisone (group B). Intent-to-treat figures are in parenthesis. (p Ω0.729 at baseline, p Ω at 3 months). Visual acuity. At baseline, VA measurements (mean value for both eyes) were marginally lower in group A than in group B (p Ω0.098) (Table 2). Between timepoints 0 and 3 months, VA increased slightly in group A, but did not change significantly in group B (p Ω0.024 for the difference between the groups). Intraocular pressure. Intraocular pressure measurements were similar in both groups at baseline (Table2). There was no significant difference in IOP between 0 and 3months in either group (pω0.099) or between the groups (p Ω0.734). Extraocular muscle thickness. There were no significant changes in any of the extraocular muscle thickness measurements between either treatment group, either prior to or during the course of the study (data not shown). group A patients should have been given additional therapy. Thus the intent-totreat difference between the groups is significant at p Ω0.038 instead of p Ω Table 3 shows the presence of diplopia and proptosis and the activity scores at baseline and at 3months both in patients who required additional therapy and in those who did not. Biochemical measurements Serum concentrations of sicam-1 and scd44st decreased in both groups between baseline and 3 months (data not shown). Concentrations of soluble ELAM-1 decreased slightly more in group A than in group B (p Ω0.043). Soluble IL-1RA serum levels increased from baseline to 3months in both groups. In the total study population, sil-1ra serum levels at baseline correlated negatively with the activity score of inflammation (p Ω0.013). Bone mineral density There were no significant differences between the treatment groups with respect to BMD measurements at baseline and during follow-up (data not shown). Adverse effects No serious adverse effects occurred. The most important side-effect was weight gain in both groups. Discussion According to the present study, i.v. methylprednisolone and oral prednisone are equally effective as initial therapies for thyroid-associated ophthalmopathy involving diplopia, proptosis and signs of soft tissue inflammation. However, i.v. methylprednisolone may be superior to oral prednisone with respect to the subsequent need for additional therapy. The borderline increase in VA in group A compared to the unchanged VA in group B to some extent supports the theory that i.v. methylprednisolone pulse therapy may be more beneficial than therapy with oral prednisone. However, baseline visual acuity was slightly lower in group A. There was, therefore, more likelihood Table 3. Activity score, proptosis and diplopia at baseline and at 3 months in patients with and without a need for additional therapy. P indicates the difference between patients with and without a need for additional therapy between baseline and 3 months. With Without p Need for orbital radiotherapy and surgery The number of patients who needed additional therapy is shown in Fig. 1. Of the 18 patients in group A, 16 did not require any additional treatment, whereas only six of the 15 patients in group B could be managed without additional therapy (p Ω0.008). Two patients in group A and seven patients in group B underwent orbital radiotherapy (p Ω0.047). No patients in group A and four patients in group B required orbital decompression surgery (p Ω0.033). However, later considerations indicated that another two Baseline 3 months Baseline 3 months Activity score Mean SEM Baseline /3 4/5 6 1/5/1 3/3/1** 7/8/3 14/1/1 3 months Proptosis (mm) Right eye Left eye Diplopia* No/yes 10/1 11/0 17/4 20/2 Baseline months * Diplopia in primary position or otherwise disturbing the patient. ** Two patients were treated because of unchanged proptosis. Patient with activity score of 5 6 at 3 months should have been treated. The two patients with persistent clinically significant diplopia at 3 months should have been treated, the other one is already mentioned above with an activity score of 5 6 at 3 months. 319

5 that VA would be improved by any treatment modality in group A than in group B. Other minor limitations of the study concern the differences in baseline thyroid function tests between the groups. Group B recorded higher TSab titres than group A. In addition, group B had a slightly higher mean concentration of thyrotropin (TSH), although this was not statistically significant. As TSH stimulation may worsen thyroid-associated ophthalmopathy, and high TSab titres may promote ophthalmopathy (Utiger 1992), any signs of ophthalmopathy noted at baseline might have been expected to be more severe in group B than in group A. However, the clinical eye status at baseline was very similar in both groups. Nonetheless, higher TSab titres in group B may have influenced response to any treatment. Any hypothyroidism was corrected as soon as detected, and serum TSH levels were similar in both groups at 3 months. The study s principal limitation concerned the fact that the decision to give additional treatment after initial therapy was made on clinical grounds. This process is always biased by subjectivity, as well as being influenced by patient opinion. While the present study cannot provide a definitive answer as to which method of administering glucocorticoid therapy is better, its data supports the findings of Marcocci et al. (2001), who reported higher efficacy of i.v. methylprednisolone in combination with orbital radiotherapy compared to oral prednisone and radiotherapy in severe thyroid-associated ophthalmopathy. However, the doses used by Marcocci et al. (2001) were higher than in the present study. Glucocorticoids are a well-known cause of bone loss (Sambrook 1996). In this study, glucocorticoid therapy administered intravenously or by an oral route had no negative effect on BMD at the lumbar spine and upper femur. Thus, our study shows the bone safety of shortterm glucocorticoid treatment in thyroidassociated ophthalmopathy independently of the way it is administered. Interleukin-1 (IL-1) is a potent stimulator of glycosaminoglycan production by orbital fibroblasts. Interleukin-1 can be inhibited by IL-1 receptor antagonist (IL-1RA), the soluble form of which can be measured in serum (Prummel et al. 2000). In the present study, sil-1ra serum concentrations at baseline were negatively correlated with the degree of inflammatory activity observed. Furthermore, IL-1RA serum levels increased markedly during treatment with glucocorticoids, suggesting that an increase of IL-1RA levels may account for at least some of the anti-inflammatory effects conferred by glucocorticoids (Hofbauer et al. 1997). Both IL-1RA and sil-1 receptor are potent inhibitors of IL-1-induced glycosaminoglycan production by cultured orbital fibroblasts (Tan et al. 1996). Thus, clinical improvement in thyroid-associated ophthalmopathy following immunomodulatory therapy with glucocorticoids or local radiotherapy (Hofbauer et al. 1997) may be related to the rise in IL-1RA serum levels. In thyroid-associated ophthalmopathy, cytokines have been shown to up-regulate various adhesion molecules (e.g. ICAM- 1, CD44, ELAM-1) on orbital fibroblasts. Adhesion molecules play an important role in directing circulating lymphocytes towards their target homing. The present study showed that ICAM-1, CD44st and ELAM-1 appear to reflect changes in orbital inflammatory activity in response to anti-inflammatory therapy with glucocorticoids. The decrease in these molecules during the first 3 months of therapy is consistent with findings by Prummel et al. (2000). In conclusion, high dose i.v. methylprednisolone therapy and oral treatment with pharmacological doses of prednisone are equally effective for treatment of thyroid-associated ophthalmopathy in terms of relieving diplopia, proptosis and signs of soft tissue inflammation. However, i.v. methylprednisolone therapy may be more effective than oral prednisone in reducing the need for further therapy. Acknowledgements The study was funded by the Jalmari and Rauha Ahokas Foundation. References Burch HB & Wartofsky L (1993): Graves ophthalmopathy: current concepts regarding pathogenesis and management. Endocrine Rev 14: DeGroot LJ, Gorman CA, Pinchera A, Bartalena L, Marcocci C, Wiersinga WM, Prummel MF & Wartofsky L (1995): Therapeutic controversies. Radiation and Graves ophthalmopathy. J Clin Endocrinol Metabolism 80: Hofbauer LC, Mühlberg T, König A, Heufelder G, Schworm H-D & Heufelder AE (1997): Soluble interleukinª1 receptor antagonist serum levels in smokers and non-smokers with Graves ophthalmopathy undergoing orbital radiotherapy. J Clin Endocrinol Metabolism 82: Kendall-Taylor P, Crombie AL, Stephenson AM, Hardwick M & Hall K (1988): Intravenous methylprednisolone in the treatment of Graves ophthalmopathy. Br Med J 297: Kriss JP, Petersen IA, Donaldson SS & Dougall IR (1989): Supervoltage radiotherapy for progressive Graves ophthalmopathy: results of a twenty-year experience. Acta Endocrinologica 121: Marcocci C, Bartalena L, Tanda ML et al. (2001): Comparison of the effectiveness and tolerability of intravenous and oral glucocorticoids associated with orbital radiotherapy in the management of severe Graves ophthalmopathy: results of a prospective, single-blind, randomized study. J Clin Endocrinol Metabolism 86: Mourits MP, Koornneef L, Wiersinga WM, Prummel MF, Berghout A& van der Gaag R (1989): Clinical criteria for the assessment of disease activity in Graves ophthalmopathy: a novel approach. Br J Ophthalmol 73: Mourits MP, van Kempen-Harteveld ML, Garcia MBG, Koppeschaar L, Tick L & Terwee CB (2000): Radiotherapy for Graves ophthalmopathy: randomized placebo-controlled study. Lancet 355: Nagayama Y, Izumi M, Kiriyama T et al. (1987): Treatment of Graves ophthalmopathy with high-dose intravenous methylprednisolone pulse therapy. Acta Endocrinologica 116: Prummel MF (2000): Graves ophthalmopathy: diagnosis and management. European J Nuclear Med 27: Prummel MF, Mourits MP, Berghout A, Krenning EP, van der Gaag R, Koornneef L & Wiersinga WM (1989): Prednisone and cyclocporine in the treatment of severe Graves ophthalmopathy. N Engl J Med 321: Prummel MF, Mourits MP, Blank L, Berghout A, Koornneef L & Wiersinga WM (1993): Randomized double-blind trial of prednisone versus radiotherapy in Graves ophthalmopathy. Lancet 342: Prummel MF, Wiersinga WM & Mourits MP (2000): Assessment of disease activity of Graves ophthalmopathy. In: Recent Developments in Graves Ophthalmopathy. Prummel, MF, Wiersinga, WM, Mourits, M Ph & Heufelder, AE (eds). Kluwer Academic Publishers. Boston Dordrect London: Sambrook PN (1996): Corticosteroid induced osteoporosis. J Rheumatol 23: Tan GH, Dutton CH & Bahn RS (1996): Interleukin-1 (IL-1) receptor antagonist and soluble IL-1 receptor inhibit IL-1 induced glycosaminoglycan production in cultured human orbital fibroblasts from patients with 320

6 Graves ophthalmopathy. J Clin Endocrinol Metabolism 81: Utiger RD (1992): Pathogenesis of Graves ophthalmopathy. N Engl J Med 326: Wiersinga WM & Prummel MF (2000): An evidence-based approach to the treatment of Graves ophthalmopathy. Endocrinol Metabolism Clinics North America 29: Received on May 11th, Accepted on February 9th, Correspondence: Matti Välimäki, MD, PhD Division of Endocrinology Department of Medicine Helsinki University Central Hospital FIN Helsinki Finland Tel: π Fax: π matti.valimaki/huch.fi 321