Therapeutic Outcomes of High-Dose Intravenous Steroids in the Treatment of Dysthyroid Optic Neuropathy

Abstract

Background:

While pulsed intravenous methylprednisolone (iv-MP) has been shown to be effective and well tolerated in moderate to severe Graves' orbitopathy (GO), limited data are available on dysthyroid optic neuropathy (DON). The objective of this retrospective study was to investigate the efficacy of iv-MP in the treatment of DON and to seek parameters predictive of response.

Methods:

Twenty-four DON patients (40 eyes) treated with iv-MP from 2007 to 2012 were included in the study. Concurrent neurological or ophthalmologic diseases or signs of corneal exposure were considered as exclusion criteria. Iv-MP was administered daily for three consecutive days and repeated the following week. At six months, eyes not requiring surgery to preserve visual function were considered as responsive to treatment. Visual acuity, color sensitivity, visual field, and optic discs were analyzed at two and four weeks, and at 3, 6, and 12 months after treatment. Activity of GO was graded using a clinical activity score (CAS). Visual and clinical characteristics of the eyes responsive to iv-MP were studied by comparison to those of nonresponsive eyes.

Results:

At six months, 17 of 40 (42.5%) eyes had complete visual recovery and were spared from surgical decompression. At two weeks, visual acuity, color sensitivity, and visual field improved significantly in almost all eyes, but GO inactivated (CAS<4) only in the eyes that permanently responded to iv-MP (p<0.01). The CAS at two weeks was a good predictor of response (cutoff ≥4; 66.7% sensitivity, 76.9% specificity). Optic disc swelling at diagnosis was highly predictive for unresponsiveness to iv-MP (34% sensitivity, 100% specificity). At baseline, high CAS (cutoff >5; 40.2% sensitivity, 94.1% specificity) and severely altered visual field mean defect (cutoff ≤6.31 dB; 73.9% sensitivity, 58.8% specificity) were associated with unresponsiveness to steroids. No major side effects were observed.

Conclusions:

High-dose iv-MP was effective in permanently restoring visual function in about 40% of the eyes treated. When successful, it generally induced inactivation of the orbital disease within two weeks and normalization of visual function within one month. The presence of optic disc swelling at diagnosis and persistent active disease at two weeks were good predictors of unresponsiveness to steroids.

Introduction

Graves' orbitopathy (GO) is a complex autoimmune disease characterized by inflammation and edema of the extraocular muscles and of the orbital connective/adipose tissue with infiltration of mononuclear inflammatory cells, followed by deposition of glycosaminoglycans and fibrosis (1). Acute dysthyroid optic neuropathy (DON) is a sight-threatening complication that affects up to 5% of GO patients (2). It is caused by the compression of the optic nerve at the orbital apex because of the increase of the orbital content, primarily due to enlarged and/or inflamed extraocular muscles (3).

The diagnosis, assessment, and management of DON are still a matter of controversy. By analyzing a large prospective case series of patients with features of DON, the European Group on Graves' Orbitopathy (EUGOGO) has recently reported that visual acuity, color perception, and evidence of optic nerve compression at imaging are the most sensitive tools in DON diagnosis (4).

Once DON is diagnosed, urgent treatment is required to prevent potential blindness (5). To date, orbital surgical decompression has been considered the more reliable treatment for DON, particularly in patients not responding to other treatment options (6), although evidence of the optimal time for performing surgery is lacking (7). Steroids have also been used, orally or intravenously, to treat severe GO complicated by DON (6), before or concurrently with orbital surgery. The use of large doses of intravenous methylprednisolone (iv-MP), usually 1 g daily for three consecutive days, has been employed since the 1980s for the management of optic neuritis in patients affected by multiple sclerosis (8,9). In 1989, Guy et al. reported successful treatment of five GO patients affected with DON with 1 g of iv-MP divided into four daily doses of 250 mg administered every six hours for three consecutive days (10). More recently, Mourits et al. administered four single pulses of 500 mg of iv-MP every second day, subsequently tapered down with oral prednisone, and reported that 39% of their DON patients avoided surgical decompression because of a good visual recovery. They also showed that the final visual outcome of patients who did not respond to iv-MP and who were then surgically decompressed was comparable to that obtained in patients submitted to immediate surgery alone. These authors therefore recommended iv-MP as an emergency treatment for DON with the purpose of replacing or postponing orbital decompression (11). In 2005, Wakelkamp et al. reported a favorable visual outcome in nine DON patients after administering 1 g of iv-MP daily, for three consecutive days, repeated the following week and followed by oral prednisone tapered down in four months. That randomized prospective study has also shown that immediate surgery alone did not result in a better visual outcome than immediate steroid treatment alone, and therefore the authors suggested adopting this scheme of iv-MP pulsed therapy, defined as “medical decompression,” as the first-choice treatment for DON (12).

Although iv-MP pulses have been used since to treat DON as an alternative to immediate decompressive surgery, no studies have analyzed in detail the effect of steroid administration on optic nerve function. In this retrospective study of patients affected with DON, we analyzed the efficacy of iv-MP in restoring visual function. Moreover, we sought parameters most useful to assess effective treatment response and studied which baseline clinical signs were significantly associated with successful therapy.

Patients and Methods

Patients

We retrospectively studied 24 patients affected with DON—a total of 40 eyes—treated with iv-MP pulses in our thyroid eye clinic from 2007 to 2012. Patients concurrently affected with other neurological or ophthalmologic diseases that could significantly affect visual function (multiple sclerosis, congenital, toxic, ischemic or inflammatory neuritis, cataract, glaucoma, macular diseases, high astigmatism, high myopia, congenital dyschromatopsia, etc.) or presenting with signs of severe corneal exposure (diffuse punctate keratopathy, ulcers, abscesses, leucomas, etc.) were excluded from this study. The study was approved by the ethics committee of our institution, and all patients gave written informed consent to therapy and to the analysis of the results.

Of the 24 patients included, 14 were men and 10 women (mean age 66.7±1.3 years; range 52–83 years), and 11 were smokers. Five patients (21%) had type 2 diabetes, well controlled with insulin (three patients) or oral medication (two patients). Twenty-three patients had Graves' disease (GD) and one primary idiopathic myxedema. Twelve patients were hyperthyroid on thionamides, and 12 were euthyroid (three without treatment and nine on L-T4) after being treated with radioiodine (seven patients) or total thyroidectomy (five patients). Median thyroid and orbital disease duration (determined as the time elapsed from the initial diagnosis of thyroid and orbital disease to the diagnosis of DON) were 12 months ([confidence interval (CI) 10–23]; range 3–241 months) and 4.25 months ([CI 4–6]; range 0.5–182 months) respectively.

Ophthalmologic examination

All patients underwent complete ophthalmologic examination according to the EUGOGO criteria (13). Best-corrected visual acuity (BCVA) was assessed in all patients with Snellen charts and expressed as decimal fraction. All patients were tested by Hardy–Rand–Rittler pseudoisochromatic tables for acquired dyschromatopsia (HRR second edition, Richmond products, Albuquerque, NM) to assess color perception: the number of errors in the six screening tables (5 –10) was registered for each eye, and a test with more than two errors was considered abnormal. Automated perimetry was performed in all patients with the Humphrey visual field analyzer (HFA; Humphrey Instruments, Inc., San Leandro, CA) using the Swedish Interactive Threshold Algorithm Fast Test (30–2 SITA Fast) with a Goldmann size III stimulus on a dim background (31.5 apostilb) and foveal threshold turned on. During visual field testing, every patient was provided with adequate near correction and rest breaks when requested. Visual fields with reliability indices exceeding normal limits (fixation losses >20%, false positive and false negative >33%) were repeated and excluded from the analysis if again unreliable. At baseline, each field was judged as abnormal if at least one of the visual field global indices (mean deviation [MD] or pattern standard deviation [PSD]) was beyond normal limits (p<0.05). Tests with abnormal fields were repeated for confirmation before starting treatment. Moreover MD, PSD, and foveal threshold were registered during follow-up to monitor visual field changes.

Orbital imaging

Orbital imaging was acquired with the same computed tomography (CT) scanner (Ultra Z Marconi Scanner, Marconi Medical System, Cleveland, OH) and reviewed by the same two radiologists in all but three patients. Signs of apical crowding or optic nerve stretching were noted.

Activity of GO

Inflammatory activity of GO was assessed for each orbit with the clinical activity score (CAS), which is the sum of 10 items: two symptoms (orbital pain at rest or associated with ocular movements), five signs of orbital inflammation and/or congestion (soft tissue swelling, redness of the eyelids, redness of the conjunctivae, chemosis, and swelling of the caruncle), and three signs of deterioration (worsening of proptosis, monocular duction, and visual acuity). GO is considered active when the CAS is ≥4/10 (14).

DON diagnosis

The diagnosis of DON was made based on the presence of CT scan signs of apical crowding or optic nerve stretching and at least two clinical signs suggestive of optic neuropathy such as reduction of BCVA, loss of color vision, altered visual field, relative afferent pupillary defect, or optic disc swelling in the absence of another explanation.

Treatment schedule and response criteria

Once DON was diagnosed, all patients were treated with either 500 or 1000 mg pulsed iv-MP administered daily for three consecutive days. The same cycle of treatment was repeated after one week, and then steroids were tapered off either orally or intravenously. This treatment schedule is referred to as “medical decompression” (12). Generally, patients with more severe visual dysfunction were treated with the higher-dose regimen, while patients with relative contraindications to steroids were treated with the lower-dose regimen. No other adjunctive therapies (orbital radiotherapy or immunosuppresive drugs) were administered before, concurrently, or after medical decompression. Patients who showed (in at least one orbit) persistent active inflammatory GO after medical decompression (CAS≥4) were kept on weekly iv-MP pulses (250 or 500 mg) until inactivation of orbital disease, but never exceeding the cumulative dose of 10 g, including steroids administered in the six months preceding medical decompression. Liver function tests, glycemia, blood pressure, heart rate, and adverse effects were monitored at baseline and throughout follow-up.

Eye function was not considered to have recovered adequately when at least two parameters—namely BCVA <0.8, more than three errors at HRR test, or altered visual field (MD <−3 dB or PSD >3 dB or both)—were recorded. After two weeks of medical decompression, eyes showing inadequate visual response to steroids were surgically decompressed independently of the degree of GO activity. Subsequent relapses of DON in eyes initially responding to steroids were also treated with surgical decompression. In this series, surgery was employed exclusively to preserve visual function when iv-MP failed to achieve a complete visual recovery, never to treat persistent high CAS scores.

Outcomes analysis

Eyes that did not require surgery at six months after medical decompression because of visual function recovery in the absence of DON relapses were considered treatment responsive (primary outcome).

Visual function parameters and the CAS were studied as secondary outcomes and analyzed at baseline, at the end of the second and fourth week, and at three and six months after the first iv-MP pulse. A late follow-up examination was performed one year after treatment. Adverse events were also recorded throughout follow-up. One patient was surgically decompressed before the fourth week of follow-up because of sudden deterioration of both visual function and optic disc swelling. This patient was excluded from the analysis of visual parameters at one month.

Statistical analysis

All data are reported as mean±standard error or with confidence intervals for the mean, or as median with range or confidence intervals for the median, as appropriate. Differences in categorical data were tested using the chi-square test or Fisher's exact test; differences in parametric or nonparametric variables were tested using the one-way analysis of variance test (ANOVA) or the Kruskal–Wallis test respectively. All variables were tested for time variation with repeated measures ANOVA for parametric variables or Friedman test for nonparametric variables; Bonferroni correction for repeated multiple comparisons was applied to p-values. Correlations among nonparametric variables were tested with Spearman's analysis of rank correlation. Kaplan–Meier survival analysis was performed to study the timing of decompressive surgery. Receiver operating characteristic (ROC) curve analysis was used to test the performance as response predictors of all the visual and clinical parameters assessed. The area under the curve (AUC), cutoff values, sensitivity, and specificity were calculated and tested at baseline and during follow-up. The minimum criterion for tests of significance was p<0.05. Study results were analyzed with MedCalc statistical software for Windows, v9.5.0.0 (MedCalc Software, Mariakerke, Belgium).

Results

DON characteristics at presentation and treatment

Of the 24 patients included in the study, seven had unilateral DON, one was single-eyed (after traumatic injury), and the remaining 16 had bilateral DON. At diagnosis, 38 of the 40 affected eyes (95%) showed an altered automated visual field, 36 (90%) apical crowding on CT scan, 31 (77.5%) dyschromatopsia, 30 (75%) reduced BCVA, and 4 (10%) optic nerve stretching on CT scan. Only seven eyes (17.5%) had optic disc swelling or a definite relative afferent pupillary defect. In the six months preceding the development of DON, 10 patients (16 eyes) had already been treated with various doses of oral or intravenous steroids for active GO (mean cumulative dose 1.2 g; [CI 0.5–1.9]; range 0.5–6.5 g). In this subgroup of patients, steroids were discontinued at a median of five weeks ([CI 4–9.5]; range 0–12 weeks) before subsequent iv-MP treatment for DON.

Of the 24 patients subjected to medical decompression, 11 (18 eyes) were treated with six pulses of 500 mg of iv-MP, and 13 (22 eyes) with six pulses of 1000 mg of iv-MP. Among baseline clinical and visual characteristics of patients treated with the two dose regimens of medical decompression, only the MD of the visual field was significantly worse in patients treated with the higher-dose regimen, according to the criteria for treatment. Six patients (10 eyes) treated with the lower-dose regimen and three (six eyes) treated with the higher-dose regimen had persistent disease inflammatory activity after medical decompression (p=0.14) and were kept on therapy with weekly iv-MP pulses for 6–10 weeks. The mean cumulative dose of steroids administered (including the six months preceding medical decompression and the six months following) was 6 g ([CI 5–7.1]; range 3–9 g) in the lower-dose regimen group and 7.7 g ([CI 6.9–8.6]; range 6–10.5 g) in the higher-dose regimen group (p<0.01).

Primary outcome analysis

At six months, 17 of 40 eyes (42.5%) showed complete visual function recovery and had no subsequent DON relapses. For these eyes, surgical decompression of the orbit was eventually unnecessary to preserve optic nerve function. The response rate to iv-MP was not different whether DON was treated with the lower-dose regimen (500 mg daily) or with the higher-dose regimen (1000 mg daily; 44% vs. 41% respectively, p=0.92), nor was it different in patients previously treated with steroids or not (38% vs. 44% respectively, p=0.84). Previous treatment with steroids before medical decompression did not show good performance as predictor of response (AUC=0.57; p=0.36).

Orbital surgical decompression was performed 3–24 weeks after DON diagnosis (median=8 weeks; [CI 6–16]) when the visual response to steroids was inadequate or DON relapsed. Visual response to iv-MP was generally symmetrical, as only three patients (12.5 %) had one orbit responding to steroids, while the opposite needed surgery. As shown in Figure 1, the majority of orbits not responding to iv-MP were surgically decompressed between one and three months of follow-up (16 of 23 orbits, 70%). Only one patient was surgically decompressed before the fourth week due to rapid worsening of optic disc swelling and deterioration of visual function. After surgical orbital decompression, no relapses of DON were observed; three patients (six orbits of the 23 submitted to surgery) underwent further weekly intravenous steroid treatment, only for therapy of inflammatory signs. Interestingly, no relapses of DON were observed after the six-month assessment in steroid-responsive eyes: among them, three (12%) underwent surgical orbital decompression for other indications (disfigurement), only after one year of follow-up.

FIG. 1.

Percentage of the 40 orbits treated with high-dose intravenous methylprednisolone that required surgery at different times of follow-up. On the x-axis, the time (weeks) after medical decompression is shown. The dotted vertical reference lines mark the post-treatment follow-up times.

Visual function, disease inflammatory activity, and side effects

Two weeks after medical decompression, all parameters of visual function improved significantly, independently of the final outcome (Fig. 2). At one month, BCVA (Fig. 2A) and color sensitivity (Fig. 2B) improved, compared to baseline values, only in the eyes permanently responding to iv-MP (p<0.005 and p<0.001 respectively). The MD (Fig. 2C) and PSD (Fig. 2D) of the visual field improved at one month in both responding (p<0.001 and p<0.005 respectively) and nonresponding eyes (p<0.0001 and p<0.0001 respectively) but normalized only in responding eyes. Although the foveal threshold improved on average in all eyes, this change was not significant (not shown). At two weeks, optic disc swelling did not improve in any of the affected eyes that eventually underwent surgical decompression of the orbit in the subsequent four weeks. After the first month of follow-up, all visual parameters remained steadily normal in responding eyes. At six months, visual function had also improved significantly in nonresponding eyes, only after surgical decompression. At one year, BCVA and color sensitivity of nonresponding eyes normalized, although their values were significantly worse than those of responding eyes (p<0.05; Fig. 2A and B). On the other hand, the MD and PSD of the visual field, although significantly improved, did not normalize and remained significantly worse than those of responding eyes (p<0.05; Fig. 2C and D).

FIG. 2.

Changes of visual function parameters after medical decompression up to 48 weeks of follow-up. Visual outcomes of eyes responding to intravenous methyprednisolone are shown with empty circles and a continuous line (), while those of eyes not responding are shown with solid triangles and a dashed line (). Values are shown as mean±standard error (SE) for continuous variables and as median (error bars: CI) for discontinuous variables. (A) Best-corrected visual acuity (BCVA; decimal fraction). (B) Color vision (median number of errors at the screening tables of the Hardy–Rand–Rittler [HRR] color vision test). (C) Mean deviation (MD) of the automated visual field. (D) Pattern standard deviation (PSD) of the automated visual field. *p<0.05 versus eyes not responding.

At two weeks, the CAS score improved significantly in both responsive and nonresponsive orbits (p<0.05), but GO became inactive (CAS<4) only in the orbits that permanently responded to medical decompression (p<0.01). Overall, an inadequate visual response that required subsequent surgical decompression was observed in about 70% of the eyes of patients with GO still active at two weeks compared to 20% of those whose disease became inactive (p<0.05). ROC curve analysis of the CAS registered at two weeks showed an AUC of 0.788 ([CI 0.613–0.964]; p<0.001) resulting in the highest Youden index at a cutoff of ≥4 (66.7% sensitivity and 76.9% specificity). At six months, GO had also become inactive in the orbits unresponsive to steroids, only after surgical decompression (Fig. 3).

FIG. 3.

Changes in the clinical activity score (CAS) after medical decompression up to 48 weeks of follow-up. Dotted boxes (): CAS of responding orbits; grey boxes (): CAS of not responding orbits. Values are expressed as median (error bars: [CI]). *p<0.05 versus orbits not responding. Percentage values in black boxes () indicate the percentage of orbits not responding to medical decompression that were surgically decompressed at each time point of the follow-up.

When we looked at the baseline clinical parameters of the eyes in which DON was successfully treated with medical decompression and in those where it was not (Table 1), we observed that optic disc swelling was present exclusively in nonresponsive eyes (p<0.01; AUC=0.652; [CI 0.485–0.796]; p<0.002; 30.4% sensitivity and 100% specificity when optic disc swelling is present), that the MD of the visual field was significantly lower in nonresponsive eyes (p<0.04; AUC=0.659; [CI 0.492–0.801]; p<0.05; 73.9% sensitivity and 58.8% specificity at a cutoff of ≤6.31 dB), and that the CAS was significantly higher in nonresponsive orbits (p<0.05; AUC=0.679; [CI 0.513–0.818]; p<0.03; 40.2% sensitivity and 94.1% specificity at a cutoff of >5).

Table 1.

Baseline Clinical Parameters of the Eyes Affected with Dysthyroid Optic Neuropathy Treated with High-Dose Intravenous Methylprednisolone

	All eyes	Eyes successfully treated	Eyes unsuccessfully treated	p-Value
Patient's age (years)	66.7±1.3	67.6±2.4	67.4±1.5	0.95^b
Eyes belonging to male/female	24/16	10/7	14/9	0.84^a
Eyes belonging to smokers (yes/no)	19/21	6/11	13/10	0.31^a
Eyes belonging to diabetic patients (yes/no)	8/32	3/14	5/18	0.93^a
MP dosage for medical decompression (low/high)	18/22	8/9	10/13	0.92^a
Previous steroids for active GO (yes/no)	16/24	6/11	10/13	0.84^a
Subsequent steroids for persistent active GO (yes/no)	16/24	5/12	11/12	0.39^a
GD duration (months)	12 (3–241)	19 (4–192)	10 (3–241)	0.17^c
GO duration (months)	4.25 (0.5–182)	4 (0.75–182)	4.5 (0.5–33.5)	0.59^c
CAS	5 (2–7)	4 (2–6)	5 (3–7)	0.049 ^c
Proptosis (Hertel; mm)	25.8±0.5	26.0±1.1	25.6±0.4	0.92^b
BCVA (decimal fraction)	0.72±0.03	0.76±0.04	0.70±0.04	0.35^b
Color sensitivity (errors at HRR)	2 (0–6)	2 (1–4)	3 (0–6)	0.06^c
MD (decibel)	−10.17±1.08	−7.59±1.21	−12.08±1.5	0.04 ^b
PSD (decibel)	5.78±0.46	4.88±0.75	6.45±0.55	0.09^b
Foveal threshold (decibel)	29.25±0.82	30.41±0.89	28.39±1.25	0.22^b
Optic disc swelling (yes/no)	7/33	0/17	7/16	0.01 ^d
Apical crowding (yes/no)	36/4	14/3	22/1	0.39^a

Values are expressed as number of events observed or as median with range or as mean±standard error of mean, as appropriate. Boldface indicates statistically significant parameters.

p-Values: ^aChi-square test; ^bone-way analysis of variance; ^cKruskal–Wallis test; ^dFisher's exact test.

MP, methylprednisolone; GO, Graves' orbitopathy; GD, Graves' disease; CAS, clinical activity score; BCVA, best-corrected visual acuity; HRR, Hardy–Rand–Rittler test for acquired color vision defects (the table show the number of errors registered in the six screening tables of the test); MD, mean deviation of the automated visual field; PSD, pattern standard deviation of the automated visual field; foveal threshold, foveal sensitivity threshold detected with the automated visual field.

After medical decompression, we identified only mild and transient side effects independently of the dose regimen employed (Table 2). Moreover, we did not observe a significant relationship between the overall cumulative dose of steroids administered and the frequency of side effects observed, nor between the treatment schedule with 500 or 1000 mg (p=N.S.). A transient increase in blood pressure or glycemia was observed in patients affected with systemic hypertension or diabetes mellitus and only required adjustment of the currently administered therapy. A mild depressive crisis was observed in a patient affected with depression and was well controlled by adjusting the dose of the currently administered antidepressants. At baseline or during follow-up, no patient showed elevated levels of serum aminotransferases (as assessed by values four times upper normal limits); one patient had transient hyperbilirubinemia. No major side effects were observed.

Table 2.

Side Effects Recorded in Patients with Dysthyroid Optic Neuropathy After Treatment with High-Dose Intravenous Methylprednisolone

	All 24 patients	500 mg regimen (11 patients)	1000 mg regimen (13 patients)
Hypertensive crisis^a	2	1	1
Tachycardia	1	1	0
Hyperglycaemia^a	6	3	3
Minor depression^b	1	0	1
Anxiety	1	1	0
Insomnia	2	1	1
Dyspepsia	1	0	1
Asthenia	2	2	0
Myalgia/arthralgia	2	2	0
Hyperbilirubinemia	1	0	1
Total number of side effects observed	19	11	8

Transient and mild increase in blood pressure or glycemia was observed in patients with a history of diabetes mellitus or systemic hypertension.

A mild depressive crisis was observed in a patient with a previous history of minor depression.

Discussion

While various treatment regimens and doses of intravenous glucocorticoids have been shown to be effective and well tolerated in moderate to severe GO (15 –21), very limited data are available on the efficacy and potential risks of high-dose intravenous steroids for the treatment of DON (6,10 –12,22 –24). Moreover, the existing literature does not provide evidence on which specific clinical features are associated with a significant therapeutic response of acute DON to steroids or on the optimal time for performing surgery in eyes unresponsive to medical decompression.

In 2001, Mourits et al. retrospectively studied the efficacy of high-dose iv-MP in DON patients, and observed a long-term response rate of 39%, reporting few minor side effects. Post-treatment visual acuity was assessed as the most important criterion to define successful therapeutic response (BCVA>0.5) while, although tested, visual field and color sensitivity were not studied in detail (11). More recently, Wakelkamp et al. prospectively treated a small group of DON patients with medical decompression, and showed that 55% of them did not need surgery to preserve visual function. Visual acuity was considered as the most important parameter of response (responders were defined as patients with a post-treatment pinhole BCVA≥0.63), but again, a specific description of the changes of visual field and color sensitivity was not provided (12). Except for one patient who had thrombosis of the central retinal vein in one eye, no other major side effects were reported (12). However, the authors recommended not to exceed a total cumulative dose of 8 g of iv-MP in order to reduce the risk of major side effects, with special concern for liver dysfunction, as described in previous reports (25 –27). In 2005, Hart et al. studied the outcome of high-dose iv-MP in severe GO patients, nine treated for DON, seven for progressive deterioration of diplopia, and two for severe proptosis or soft-tissue involvement. They administered three consecutive daily doses of 500 mg of iv-MP tapered off with oral prednisone, and defined the improvement in BCVA by two lines in the Snellen chart in the worst-affected eye as a good response to treatment in DON patients. The authors reported an overall long-term response rate of about 60%, and concluded that a good response to treatment, observed within the first week, was highly related to the final outcome. Unfortunately, they did not provide separate details about the visual outcomes of the subgroup of the nine DON patients treated. Of note, no major side effects were reported (24).

In our retrospective series of patients with DON treated with high-dose iv-MP, we defined the response to treatment by the eyes that did not need surgical decompression of the orbit to preserve visual function. Unlike previous reports, BCVA, color sensitivity, and visual fields were attributed the same weight in deciding for an adequate or inadequate visual recovery, as the persistence of at least two of these parameters significantly altered after medical decompression generally brought about prompt surgical decompression. We observed an overall long-term response rate of 42.5% to medical decompression, consistent with previous reports (11,12) and, interestingly, not related to the dose of iv-MP administered. As previously reported by Guy et al. (10) and Hart et al. (24), almost all patients showed a significant improvement of all clinical and visual parameters of DON within the first two weeks of follow-up. At one month, visual function normalized in the eyes successfully treated by steroids, and subsequently DON did not relapse. On the other hand, visual parameters deteriorated in steroid nonresponsive eyes after two weeks of follow-up, eventually requiring surgical decompression of the orbit to treat DON definitively. Interestingly, the CAS was shown to be more useful than visual parameters in the early assessment of the therapeutic response of DON to medical decompression: at two weeks, GO inactivated only in responding eyes, while their visual parameters normalized later, at one month. Persistent activity of the orbital disease (CAS ≥4/10) at two weeks was significantly associated with long-term unresponsiveness of DON to steroids with the best sensitivity/specificity ratio compared to the other visual and clinical parameters assessed. This would suggest that in patients in whom medical decompression is not able to inactivate GO within a few weeks, orbital congestion, venous stasis, and anatomic factors may play a more important role than inflammation in the pathogenesis of DON, and consequently steroids alone may not treat DON permanently. This may explain why the presence of optic disc swelling, the highest expression of orbital congestion in GO, was, in this study, invariably associated with unresponsiveness to medical treatment (100% specificity). This observation is in good agreement with that of Soares-Welch et al. who in 2003 retrospectively studied 215 DON patients treated with surgical decompression and suggested that the presence of disc edema alone should be considered an indication for surgery (7). By studying our patients at baseline, we have also found that high CAS (>5) and low MD (≤6.31 dB) values were significantly associated with unresponsiveness to steroids. All these observations suggest that we can identify patients at high risk to be nonresponders just before starting medical treatment or a couple of weeks later. In this study, steroid nonresponsive eyes showed good visual recovery after surgical decompression, but significantly lower compared to that of responding eyes. In retrospect, this observation may suggest that surgery should not have been delayed in patients at high risk to be nonresponders that still showed active GO, optic disc swelling, or deteriorating visual function at two weeks.

High-dose steroids could potentially have life-threatening adverse effects. The incidence of major side effects has been related to the cumulative dose administered or to the presence of comorbidity (19,25 –27). Our data suggest that if we exclude patients with concurrent severe systemic diseases from steroid administration, we may avoid potential severe side effects. Indeed, the frequency of adverse effects observed in our patients was mainly related to the presence of comorbidity, while, interestingly, no significant relationship was found with the total dose of iv-MP administered. Although six of our patients received a cumulative dose of steroids >8 g, we identified only minor side effects, mainly characterized by transient worsening of pre-existing diseases (systemic hypertension, diabetes mellitus, dyspepsia, anxiety, or minor depression) that only required therapy adjustments. Therefore we recommend careful examination of the patients' personal history before administering high-dose steroids in order to identify patients potentially at risk for severe side effects.

This retrospective study has some limitations. The patients included were either hyperthyroid or euthyroid, and about one third of them (9/24; 37.5%) was treated with various doses of oral or intravenous steroids before the development of DON. Another potential confounding factor is that the two different dose regimens of medical decompression employed to treat acute DON were not randomly assigned to patients. In addition, a significant number of patients required additional doses of iv-MP and variable duration of pulse therapy to inactivate the inflammatory phase of the disease. Nevertheless, in our study, we observed an overall response rate in good agreement with previously published studies, and therefore we encourage the use of iv-MP therapy as first line treatment for DON. Although a dose-finding study of medical decompression was beyond our objectives, we did not observe a definite superiority of the higher-dose regimen compared to the lower-dose regimen. Considering that the optimal dose and schedule of steroids are not yet established, our findings would suggest that higher doses may not always be required to treat DON. In addition, we also observed response to medical decompression in patients treated with steroids before the development of DON (response rate: 38%), suggesting that previous medical treatment does not necessary prevent successful response to higher-dose steroids. Our data are of interest primarily because they provide useful evidence on the visual outcomes following high-dose steroids in DON and may in fact predict treatment failure, which helps the clinician to choose surgical orbital decompression in patients at higher risk to be unresponsive to steroids.

In conclusion, medical decompression is successful in restoring visual function in about 40% of the eyes treated, particularly in those in which therapy induces inactivation of the orbital inflammation within two weeks. Deterioration of visual parameters after two weeks of treatment seems to be indicative of poor long-term visual recovery if surgical decompression is delayed. In addition, our findings suggest that surgical decompression, particularly when delayed, may not invariably allow complete restoration of normal visual function, consistent with previously published studies (7,28,29). The presence of optic disc swelling at diagnosis and persistent active disease at two weeks are good predictors of unresponsiveness to steroids. Excluding patients with significant systemic comorbidity from high-dose steroid administration may avoid potentially severe side effects. Our findings need to be confirmed in a prospective randomized study.

Footnotes

Author Disclosure Statement

No funding or financial support was required for this study. The authors have no financial interests in any of the topics discussed.

References

Bahn

. 2010. Graves' ophthalmopathy. N Engl J Med, 362:726–738.

Prummel

, Bakker

, Wiersinga

, Baldeschi

, Mourits

, Kendall-Taylor

, Perros

, Neoh

, Dickinson

, Lazarus

, Lane

, Heufelder

, Kahaly

, Pitz

, Orgiazzi

, Hullo

, Pinchera

, Marcocci

, Sartini

, Rocchi

, Nardi

, Krassas

, Halkias

. 2003. 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, 148:491–495.

Lane

, Boschi

. 2010. Management of very severe Graves' orbitopathy (dysthyroid optic neuropathy and corneal breakdown). In: Wiersinga

, Kahaly

(eds) Graves' Orbitopathy: A Multidisciplinary Approach. Second edition. Karger, Basel, pp. 153–159.

McKeag

, Lane

, Lazarus

, Baldeschi

, Boboridis

, Dickinson

, Hullo

, Kahaly

, Krassas

, Marcocci

, Marinò

, Mourits

, Nardi

, Neoh

, Orgiazzi

, Perros

, Pinchera

, Pitz

, Prummel

, Sartini

, Wiersinga

. 2007. Clinical features of dysthyroid optic neuropathy: a European Group on Graves' Orbitopathy (EUGOGO) survey. Br J Ophthalmology, 91:455–458.

Bartalena

, Baldeschi

, Dickinson

, Eckstein

, Kendall-Taylor

, Marcocci

, Mourits

, Perros

, Boboridis

, Boschi

, Currò

, Daumerie

, Kahaly

, Krassas

, Lane

, Lazarus

, Marinò

, Nardi

, Neoh

, Orgiazzi

, Pearce

, Pinchera

, Pitz

, Salvi

, Sivelli

, Stahl

, von Arx

, Wiersinga

. 2008. Consensus statement of the European Group on Graves' Orbitopathy (EUGOGO) on management of Graves' orbitopathy. Thyroid, 18:333–346.

Lucarelli

, Shore

. 1996. Management of thyroid optic neuropathy. Int Ophthalmol Clin, 36:179–193.

Soares-Welch

, Fatourechi

, Bartley

, Beatty

, Gorman

, Bahn

, Bergstralh

, Schleck

, Garrity

. 2003. Optic neuropathy of Graves' disease: results of transantral orbital decompression and long-term follow-up in 215 patients. Am J Ophthalmol, 136:433–441.

Milligan

, Newcombe

, Compston

DAS

. 1987. A double blind controlled trial of high dose methylprednisolone in patients with multiple sclerosis: clinical effects. J Neurol Neurosurg Psychiatry, 50:511–516.

Cleary

, Beck

, Anderson

Jr , Kenny

, Backlund

, Gilbert

. 1993. Design, methods, and conduct of the Optic Neuritis Treatment Trial. Control Clin Trials, 14:123–142.

10.

Guy

, Fagien

, Donovan

, Rubin

. 1989. Methylprednisolone pulse therapy in severe disthyroid optic neuropathy. Ophthalmology, 96:1048–1053.

11.

Mourits

, Kalman

, Sasim

. 2001. Methylprednisolone pulse therapy for patients with dysthyroid optic neuropathy. Orbit, 20:275–280.

12.

Wakelkamp

, Baldeschi

, Saeed

, Mourits

, Prummel

, Wiersinga

. 2005. Surgical or medical decompression as first line treatment of optic neuropathy in Graves' ophthalmopathy? A randomized controlled trial. Clin Endocrinol, 63:323–328.

13.

The European Group on Graves' Orbitopathy (EUGOGO): Wiersinga

, Perros

, Kahaly

, Mourits

, Baldeschi

, Boboridis

, Boschi

, Dickinson

, Kendall-Taylor

, Krassas

, Lane

, Lazarus

, Marcocci

, Marino

, Nardi

, Neoh

, Orgiazzi

, Pinchera

, Pitz

, Prummel

, Sartini

, Stahl

, von Arx

. 2006. Clinical assessment of patients with Graves' orbitopathy: the European Group on Graves' Orbitopathy recommendations to generalists, specialists and clinical researchers. European Journal of Endocrinology, 155:387–389.

14.

Mourits

, Koornneef

, Wiersinga

, Prummel

, Berghout

, van der Gaag . 1989. Clinical criteria for the assessment of disease activity in Graves' ophthalmopathy: a novel approach. Br J Ophthalmol, 73:639–644.

15.

Nagayama

, Izumi

, Kiriyama

, Yokoyama

, Morita

, Kakezono

, Ohtakara

, Morimoto

, Okamoto

, Nagataki

. 1987. Treatment of Graves' ophthalmopathy with high-dose intravenous methylprednisolone pulse therapy. Acta Endocrinol, 116:513–518.

16.

Kendall-Taylor

, Crombie

, Stephenson

, Hardwick

, Hall

. 1988. Intravenous methylprednisolone in the treatment of Graves' ophthalmopathy. Br Med J, 297:1574–1578.

17.

Tagami

, Tanaka

, Sugawa

, Nakamura

, Miyoshi

, Mori

, Nakao

. 1996. High dose intravenous steroid pulse therapy in thyroid associated ophthalmopathy. Endocr J, 43:689–699.

18.

Matejka

, Verges

, Vaillant

, Petit

, Brun-Pacaud

, Rudoni

, Brun

. 1996. Intravenous methylprednisolone pulse therapy in the treatment of Graves' ophthalmopathy. Horm Metab Res, 30:93–98.

19.

Zang

, Ponto

, Kahaly

. 2011. Clinical review: intravenous glucocorticoids for Graves' orbitopathy: efficacy and morbidity. J Clin Endocrinol Metab, 96:320–332.

20.

Kahaly

, Pitz

, Hommel

, Dittmar

. 2005. Randomized, single blind trial of intravenous versus oral steroid monotherapy in Graves' orbitopathy. J Clin Endocrinol Metab, 90:5234–5240.

21.

Bartalena

, Krassas

, Wiersinga

, Marcocci

, Salvi

, Daumerie

, Bournaud

, Stahl

, Sassi

, Veronesi

, Azzolini

, Boboridis

, Mourits

, Soeters

, Baldeschi

, Nardi

, Currò

, Boschi

, Bernard

, von Arx

; European Group on Graves' Orbitopathy. 2012. Efficacy and safety of three different cumulative doses of intravenous methylprednisolone for moderate to severe and active Graves' orbitopathy. J Clin Endocrinol Metab, 97:4454–4463.

22.

Panzo

, Tomsak

. 1983. A retrospective review of 26 cases of dysthyroid optic neuropathy. Am J Ophthalmol, 96:190–194.

23.

Ben Simon

, Syed

, Douglas

, Schwartz

, Goldberg

, McCann

. 2006. Clinical manifestation and treatment outcome of optic neuropathy in thyroid related orbitopathy. Ophthalmic Surg Lasers Imaging, 37:284–290.

24.

Hart

, Kendall-Taylor

Pat

, Crombie

, Perros

. 2005. Early response to intravenous glucocorticoids for severe Thyroid associated ophthalmopathy predicts treatment outcome. J Ocul Pharmacol Ther, 21:328–336.

25.

Marinó

, Morabito

, Brunetto

, Bartalena

, Pinchera

, Marocci

. 2004. Acute and severe liver damage associated with intravenous glucocorticoid pulse therapy in patients with Graves' ophthalmopathy. Thyroid, 14:403–406.

26.

Le Moli

, Baldeschi

, Saeed

, Regensburg

, Mourits

, Wiersinga

. 2007. Determinants of liver damage associated with intravenous methylprednisolone pulse therapy in Graves' ophthalmopathy. Thyroid, 17:357–362.

27.

Salvi

, Vannucchi

, Sbrozzi

, Del Castello

, Carnevali

, Fargion

, Beck-Peccoz

. 2004. Onset of autoimmune hepatitis during intravenous steroid therapy for thyroid-associated ophthalmopathy in a patient with Hashimoto's thyroiditis: case report. Thyroid, 14:631–634.

28.

Liao

, Chang

, Lin

. 2006. Transcaruncular orbital decompression: an alternate procedure for Graves ophthalmopathy with compressive optic neuropathy. Am J Ophthalmol, 141:810–818.

29.

Choe

, Cho

, Elner

. 2011. Comparison of lateral and medial orbital decompression for the treatment of compressive optic neuropathy in thyroid eye disease. Ophthal Plast Reconstr Surg, 27:4–11.