|
|
||||||||
CLINICAL STUDIES |
1 Donders Institute of Ophthalmology and2 Department of Endocrinology, University Medical Center Utrecht, Heidelberglaan 100, 3584 CX Utrecht, The Netherlands
(Correspondence should be addressed to R J van Geest who is now at Academic Medical Center, Room L3-153, Meibergdreef 15, 1105 AZ Amsterdam, The Netherlands; Email: r.j.vangeest{at}amc.uva.nl)
| Abstract |
|---|
|
|
|---|
Design: Prospective, placebo (PL)-controlled, double-blind, randomized study.
Methods: Fifteen previously untreated patients with active, moderately severe GO participated in the study; 6 patients received MP and 9 patients a PL. Moderately severe disease was defined using the NOSPECS classification of clinical signs of GO . Activity was measured with the clinical activity score (CAS). A dose of 500 mg MP or only solvent was administered intravenously, over three consecutive days, in four cycles at 4 weekly intervals (6 g of MP in total). Qualitatively, a successful treatment outcome was defined as an improvement in one major and/or two minor criteria in the worst eye at week 48. The major criteria were: improvement in diplopia grade; improvement in eye movement; a decrease in CAS of three points. The minor criteria were: decrease of eyelid retraction; decrease of proptosis; improvement in grade of soft tissue swelling; a decrease in CAS of two points.
Results: The qualitative treatment outcome was successful at the end of the trial in five out of six (83%) patients receiving MP and in one out of nine (11%) patients given the PL (relative risk=7.5; (95% confidence interval 1.1–49.3), P=0.005). The treatment was well tolerated.
Conclusions: In spite of the small number of patients, a significant difference in outcome was observed between MP- and PL-treated patients. We conclude that MP pulse therapy appears to be an effective treatment for active, moderately severe GO.
| Introduction |
|---|
|
|
|---|
The first step in the treatment of GO is controlling the thyroidal disease, as achieving euthyroidism improves ocular symptoms (3). The next step is to distinguish active GO from burnt-out GO, for which the clinical activity score (CAS), based on the classical signs of inflammation, is commonly used (4, 5). When the disease has become stable, sequels can be dealt with surgically. Within this framework, treatment of active, moderate to severe GO has no clear one-line strategy. So far, various immunosuppressive therapies have been reported, including retrobulbar steroids (6), oral prednisone (6, 7, 8, 9), systemic steroids (6, 7, 8, 10, 11), radiotherapy (12) and combinations thereof. In spite of its side effects, glucocorticoid therapy, together with retrobulbar irradiation is a widely used method of immunosuppression for GO.
A number of studies comparing intravenously (i.v.) administered glucocorticoids with oral medication showed that i.v. is equally or more effective, with less adverse effects (6, 7, 8). Controlled studies have been done by our group, testing oral prednisone versus cyclosporine and radiotherapy (12), indicating that oral prednisone is effective in two out of every three patients with GO. So far, orbital radiotherapy is the only immunomodulatory treatment that has been tested against a placebo (PL; e.g. sham-irradiation). Its efficacy appeared to be limited to improvement of motility (13).
In patients with dysthyroid optic neuropathy (DON), prednisolone pulse therapy (PPT) seems to be efficacious with few side effects. The term pulse therapy refers to discontinuous i.v. infusion of high doses of glucocorticoids over a short period of time. PPT shows an impressive improvement of visual acuity. It has been noted that the CAS significantly improves and one-third of DON patients are spared from decompressive surgery (14). Reported side effects of PPT are rare. In small numbers of patients, body weight gain, induction of diabetes mellitus, elevation of blood pressure, pyloric ulcer, increased osteoporosis and increase in insulin dosage were noted (15, 16). Cases of sometimes fatal autoimmune hepatitis following i.v. glucocorticoid pulse therapy have been reported (17, 18, 19), but these patients had received excessive cumulative doses of prednisone.
To assess whether PPT is similarly efficacious and is safe in patients with less severe GO, we designed a prospective, controlled, double-blind, randomized study to compare the effect of methylprednisolone (MP) pulse therapy with the effect of PL treatment in previously untreated patients with moderately severe GO.
| Subjects and methods |
|---|
|
|
|---|
The study was approved by the ethics committee of the University Medical Centre Utrecht. Consecutive untreated patients with active, moderately severe GO were included in the trial. The diagnosis was made on typical clinical features of the disease, such as proptosis, eyelid retraction and swelling, impaired motility and enlarged extraocular eye muscles and increased intraorbital fat on orbital computed-tomography scans.
The severity of GO was assessed using an adapted NOSPECS classification (Table 1). Moderately severe GO was defined as marked soft tissue swelling and/or proptosis of 18 mm or more for females and 20 mm or more for male patients (as measured with a modified Zeiss Jena exophthalmometer (Carl Zeiss Meditec AG, Jena, Thüringen, Germany) (18)), and/or diplopia in primary or reading position (NOSPECS classes 2c and/or 3abc and/or 4b). For inclusion, it was sufficient if one eye satisfied these criteria. Patients with mild eye disease or frozen eyes (NOSPECS class 4c), or symptoms of optic nerve compression (NOSPECS class 6: reduced visual functions, oedema of the optic disc; apical crowding of the extraocular muscles on computed tomography; CT scan) were excluded. Disease activity was measured by the CAS; we defined active disease as a CAS of four points or more.
|
Treatment
After the patient had given informed consent, each was assigned to receive either MP or PL, for which they were randomized by an external office (the Centre of Statistics, The Netherlands) by means of a randomization list. All patients were treated on an outpatient basis.
MP 500 mg diluted in 500 ml Ringer solution was administered intravenously during 1 h for three consecutive days, in four cycles at 4 weekly intervals (high-dose pulsed therapy on day 1, 2 and 3 in week 0, 4, 8 and 12).This amounts to a total dose of MP of six grams in 3 months. The PL group received 500 ml of Ringer solution in the same way as the MP group. The ophthalmologist who assessed the treatment results was not informed about the kind of treatment that was given. This was known only to the endocrinologist.
Assessment of efficacy
All patients were examined by the same ophthalmologist at week 0, 4, 8 and 12 before the first infusion on day 1 and after the third infusion on day 3. Also, patients were checked by the same ophthalmologist 24 and 48 weeks after the treatment had commenced.
The soft tissue swelling was graded as no swelling, minimal, moderate or marked (NOSPECS class 2). Proptosis was measured with a single modified Zeiss Jena exophthalmometer (20), for which the NOSPECS class 3 was adapted, with a subdivision for male and female. The eyelid aperture was measured with a ruler as the distance in the midline between the eyelids. A full orthoptic exam, including duction measurement and field of binocular single vision was done by an orthoptist. Elevation, depression, abduction and adduction of the individual eyes were measured with a modified perimeter (21) and expressed in degrees. The field of binocular single vision was measured with the aid of a Maddox cross and a headlamp. Corneal staining was assessed with a biomicroscope. Pinhole visual acuity was tested with a Snellen chart and expressed as a decimal (e.g. 20/20=1.0). Indirect fundoscopy was performed to check the fundus of the eye. The subjective eye score (SES) was obtained by asking patients to rate their eye condition on a scale from one (the worst) to ten (the best possible). Finally, activity of the disease was measured with the CAS (4).
At week 24, the first evaluation of treatment outcome was performed and week 48 was used as the endpoint of the trial. Treatment outcome was defined using changes in major and minor criteria, adapted from others (22). The three major criteria were: an improvement in diplopia grade, according to NOSPECS class 4; (23) an improvement of eye movement in any direction of 8° or more; (21) a decrease in CAS of three points or more.
The four minor criteria were: decrease of eyelid retraction of 2 mm or more, as measured by reduction in aperture; decrease of proptosis of 2 mm or more; an improvement in grade of soft tissue swelling, according to NOSPECS class 2; a decrease in CAS of two points. Our evaluation of the results was based on the eye with the worst features at inclusion. To assess changes in mentioned criteria, the same eye was used at the endpoint. Success in response was defined as an improvement in one or more major criteria or in two or more minor criteria, in absence of deterioration of any criterion in that eye. Deterioration was defined as worsening of one grade or more soft tissue involvement and/or diplopia, and/or an increase of lid aperture of
2 mm, and/or a
2 mm increase of proptosis, and/or decrease of duction of 8° or more. For example, improvement in diplopia makes the treatment a success. However, an increase in proptosis in that same eye turns it into a failure.
We defined a response as not successful if the success criteria were not fulfilled. In addition, pre-treatment and post-treatment quantitative changes of eyelid aperture, proptosis, eye movement, SES and CAS were compared between the MP and PL groups.
Assessment of safety
During the infusion the patients were checked for possible side effects: they were asked about pain in the back, fever, weakness, nausea and itching. Nurses checked skin colour and pupil size. The examination before and after the infusions, and at week 24 and 48 included measurement of blood pressure and body weight, and laboratory testing of plasma concentrations of thyroid hormones, glucose and liver enzymes. Changes in liver function were monitored with the enzymes alkaline phosphatase (ALP) and gamma-glutamyl transpeptidase (GGT). An increase in serum levels two times the upper limit or higher was considered abnormal.
Statistical analysis
Based on a 30% difference in treatment outcome (65% response in the MP group, compared to 35% in the PL group) following several other studies in which prednisolone was tested, a sample size of 30 patients per treatment group was calculated (power 80%,
0.05). The results were analysed using the statistical software package SPSS version 12.0.2 (SPPS Inc., Chicago, IL, USA).
Clinical, biochemical and ophthalmological variables were compared at baseline between the two groups using the Mann–Whitney test. All data were analysed according to the intention to treat principle. When a patient was withdrawn from the study, all outcome measures were assessed, and the patient was analysed with the last value carried forward. The primary outcome was assessed using the relative risk (RR). A RR was calculated with a 95% confidence interval as the percentage of patients with successful treatment outcome to MP divided by the percentage of patients with successful treatment outcome to PL. Statistical significance was calculated with the Pearson
2 test.
| Results |
|---|
|
|
|---|
Recruitment started in February 2003. Two-and-a-half years later we performed an analysis of the patients who had finished the study thus far. Up until then, 355 patients with GO were referred to our department; 309 patients did not meet our inclusion criteria. Of the 46 eligible patients, 30 refused to participate due to various reasons (fear of treatment, a wish to receive other treatment, practical reasons). The remaining 16 patients were randomly assigned to either the MP group (MP, seven patients) or the PL group (PL, nine patients). After randomization, one patient in the MP group had second thoughts about participating and was excluded, leaving six patients in the trial (Fig. 1).
|
Between the two groups, baseline characteristics did not differ significantly (Table 2). Besides the clinical features of eye disease, this included age and smoking habits. The patients in the MP group were all female, the PL group had three male patients. Thyroid function also did not differ between the two groups. All remained euthyroid during the trial.
|
Efficacy
As mentioned, we chose to follow the eye with the worst features at inclusion to determine treatment success (Table 3). Changes in the other eye are also noted in Table 3, but these were not used for the final outcome. Based on improvement in one major and/or two minor criteria, the qualitative treatment outcome was successful at the end of the trial in five out of six (83%) patients receiving MP and in one out of nine (11%) patients given the PL (RR=7.5; (95% CI 1.1–49.3), P=0.005), with the last value of the patients that stopped prematurely carried forward. The one unsuccessful patient in the MP group did improve in CAS, but also had worsening in motility (adduction).
|
8° in duction); RR=4.5; (0.6–33.7), P=0.095). The CAS fell in all MP patients; a decrease of three points or more in four out of six MP patients, the remaining two patients both had a two-point decrease of CAS. In the PL group, one out of nine patients had a decrease of three points or more, two patients had a decrease of two points. (Regarding the RR of a two-point decrease: RR=3; (1.2–7.6), P=0.01). No difference was noted in eyelid retraction. Aperture improved in four MP patients compared to five PL patients (RR=1.2; (0.5–2.7), P=0.67). Aperture increased in one PL patient. Proptosis improved in two out of five MP patients with proptosis at inclusion and in two out of seven PL patients (RR=1.5; (0.3–7.9), P=0.68). Worsening (
2 mm increase) of proptosis was seen in four PL patients, of which one did not show proptosis at inclusion. Changes in SES were as follows (the numbers are given in Table 3). In the MP group: 4 better, 1 equal and 1 worse; In the PL group: 4 better, 1 equal and 4 worse. Side effects
No serious side effects were noted. At week 12 (after the last infusion), three patients in the MP group complained about Cushingoid symptoms, which were objectified by the blinded ophthalmologist in one case (facial oedema). In the PL group three patients had symptoms subjectively. Stomach ache occurred in two patients in the MP group and once in the PL group. In both groups, changes in body weight were observed, but only two patients in each group had weight gain exceeding 1%. In the MP group, an increase of 2.4 and 4.1% in body weight at week 12 as compared to baseline was noted.
An increase in liver enzyme levels two times the upper limit or higher occurred in none of the patients. In the MP group, one patient showed a minor increase in ALP, which soon returned to normal. Four out of six patients had a slight elevation of GGT during the trial, which in all cases normalized before the end of the study. All MP patients had an increase of glucose levels above normal at one time during the trial, with a maximum of 10.5 mmol/l. In all cases, blood glucose levels had normalized at the check-up at week 24 and 48. Blood pressure was elevated temporarily during the trial in three MP patients (diastolic value above 95 mmHg). One of these was known to have hypertension, which was controlled with medication. In the PL group, two patients also showed high blood pressure values on occasion (one previously known with hypertension). All patients had normal blood pressures at the end of the trial.
Follow-up
In the first year following the trial, all patients in the PL group needed additional treatment, two MP patients needed none. Two patients in the PL group received radiotherapy and oral prednisone after the trial. Orbital decompression was done in two out of six MP patients and in eight out of nine PL patients. Extraocular muscle surgery: MP none, PL 3; eyelid retraction repair: MP 2 (plus 1 planned), PL 2; tarsoraphy: MP 1, PL 3; blepharoplasty: MP 2, PL 1.
| Discussion |
|---|
|
|
|---|
The above-mentioned results are the outcome of a preterm analysis on 15 patients, whereas the original study had been designed for a much larger sample size. However, as the differences between the two arms of the study in terms of results were thus striking and already highly significant, we considered it unethical to continue the study. Moreover, comparison of the baseline characteristics of our small groups with some prior studies (12, 13, 25) did not disclose serious differences, indicating that although our sample size was small, it was nevertheless representative.
The beneficial effect of i.v. MP is in line with results presented by others in recent years. In a randomized single-blind study, Marcocci et al. noted that i.v. glucocorticoids are more effective than oral glucocorticoids with a lower rate of side effects, in 82 patients with moderate to severe GO (11). Both treatment modalities were combined with orbital radiotherapy. Kahaly et al. also concluded that in 70 patients with active and severe GO, compared to oral corticoids, i.v. administration had a better outcome and was safe (8).
In this study, no serious adverse effects were noted. We carefully assessed patients for risk factors of liver toxicity prior to inclusion. A slight increase in liver enzymes was noted in a number of MP patients, but this was temporary. There were no cases of hepatitis as was reported by others (11, 17). The four reported cases of fatal liver failure were all associated with high-dose glucocorticoid treatments (18, 19). The cumulative dose of MP in this study amounted to six grams. This agrees with the recommendation by Krassas & Boboridis, that doses should be kept low (24). Even though no serious adverse effects were seen, the small number of patients receiving MP is insufficient to make a clear statement on the safety issue. Nevertheless, in the above-mentioned larger studies of Marcocci & Kahaly, the administration of i.v. MP was shown to be safe, with less side effects in comparison with oral glucocorticoids.
Additional treatment was required less frequently in the MP group. Two-thirds of the patients in the MP group were spared from orbital decompression. Except for one, all PL patients underwent this operation in the year following the trial.
In line with others, our study confirms efficacy of i.v. glucocorticoids. Moreover, ours is the only PL-controlled study to date and did not include severe patients, for whom PPT has previously been shown to be beneficial (14). MP pulse therapy thus appears to be an effective treatment for active, moderately severe GO. In our patients the treatment was well tolerated and appeared to be safe. The dose of i.v. glucocorticoid used varies among studies. As of yet, there is no standard PPT dosage scheme. As mentioned, the cumulative amount must be high enough to distinguish effect and low enough to prevent adversities. One of the next steps in GO research, therefore, should be a study to find evidence for an optimal dose.
| Acknowledgements |
|---|
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
L. Bartalena and M. L. Tanda Graves' Ophthalmopathy N. Engl. J. Med., March 5, 2009; 360(10): 994 - 1001. [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |