Corticosteroid Usage in Giant Cell Arteritis

ABSTRACT

Giant cell arteritis (GCA) is a condition that can cause irreversible visual loss if untreated. While corticosteroids remain the mainstay of treatment to prevent visual loss, the type, dose, route, and duration of corticosteroid treatment of GCA remain controversial. Our study surveyed neuro-ophthalmologists to determine commonly prescribed dosages of corticosteroids for the treatment of GCA with or without visual loss. For patients with acute visual loss, 52% would use intravenous (IV), 46% would use IV or oral and 2% would use oral corticosteroids. Seventy-three per cent would use 500 to 1000 mg IV methylprednisolone in this group. For patients with GCA without acute visual loss, 67% would use the oral route, 30% would use IV or oral, and 3% indicated they would use IV route of treatment. Seventy-five per cent would use 1.0 to 1.5 mg/kg oral prednisone in this group. Our results suggest a majority but not a complete consensus for route and dose of corticosteroid treatment in GCA and confirm conventional recommendations for high dose IV corticosteroids for GCA with visual loss and lower dose oral regimens for GCA without visual loss.

Introduction

Giant cell arteritis (GCA) is a vasculitis involving medium- to large-sized vessels that preferentially affects the elderly. The diagnosis of GCA is initially made clinically and usually associated with a high erythrocyte sedimentation rate (ESR) and C-reactive protein (although 22% can have a normal ESR) and can be confirmed with a temporal artery biopsy.^1,2 Treatment for GCA should begin immediately because of the risk of irreversible blindness. The main goal of therapy is to preserve vision in the fellow eye, as the disease can spread to the fellow eye if left untreated.³ One large retrospective review of 245 patients with biopsy-proven GCA treated with corticosteroids reported permanent visual loss in 32 patients who had not received appropriate steroid therapy.⁴

Although many prior studies have shown a reduction of risk for developing or worsening visual loss in GCA with prompt treatment with high-dose corticosteroids, the ideal type, route, dose, and duration of therapy remain variable and somewhat controversial.^5–13 Corticosteroids, like all medications, have different risks and costs based upon type, route, and dosage. The current recommendations for corticosteroid treatment in GCA have been largely consensus based rather than evidence based because of the lack of a large sample size, adequately powered, multicentre, “head to head” (intravenous [IV] versus oral), randomised, controlled, clinical trial.⁵ Many references recommend an individual, “case by case” clinical decision for corticosteroid dose and route that is largely based upon the presence or severity of visual loss and cardiovascular risk factors or other corticosteroid related or corticosteroid impacted comorbidities.^5,7,9,14

Prior reviews of the existing literature have recommended variable corticosteroid regimens that interestingly differ by subspeciality.¹⁵ One study reported that rheumatologists tended to use lower doses of corticosteroids, while ophthalmologists used higher doses.¹⁶ The main reason for this difference is because rheumatologists see patients with rheumatological manifestations of GCA, while ophthalmologists see patients with visual loss as a result of GCA. This view of ophthalmologists of GCA as a blinding disease may cause them to prescribe higher doses of corticosteroids for GCA. While that study examined how rheumatologists and ophthalmologists treat GCA, it did not examine how neuro-ophthalmologists treat this disease.

The primary objective of this study was to determine the most commonly prescribed dosages and routes of administration for corticosteroid therapy for the treatment of GCA with and without acute visual loss among practising neuro-ophthalmologists. To our knowledge, this is the first such report in the English language, ophthalmological literature.

Methods

All study procedures adhered to the principles outlined in the Declaration of Helsinki for research involving human subjects, and the research and methods were HIPAA-compliant. All responses were anonymous and confidential. Institutional Review Board (IRB) and Ethics committee approval was obtained from the Houston Methodist Hospital (HMH) Research Institute in accordance with HMH policies and procedures. The study survey design, method, and dissemination were performed in accordance with all applicable North American Neuro-ophthalmology Society (NANOS) Research Committee protocols.

A multiple-choice survey of four questions was composed and administered using a standard, commercially available, on-line, cloud-based resource (SurveyMonkey®).

The survey questions were as follows:

1. Which of the following is the route of steroid administration that you would give to a patient with giant cell arteritis with acute visual loss?

   1. Intravenous (IV) only
   2. Oral only
   3. Either IV or oral
2. Which of the following is the route of steroid administration that you would give to a patient with giant cell arteritis without acute visual loss?

   1. IV only
   2. Oral only
   3. Either IV or oral
3. Which of the following is the daily dose of steroid that you would give to a patient with giant cell arteritis with acute visual loss?

   1. Less than 1 mg/kg oral prednisone
   2. 1 to 1.5 mg/kg oral prednisone
   3. More than 1.5 mg/kg oral prednisone
   4. 500–1000 mg IV methylprednisolone
4. Which of the following is the daily dose of steroid that you would you give to a patient with giant cell arteritis without acute visual loss?

   1. Less than 1 mg/kg oral prednisone
   2. 1 to 1.5 mg/kg oral prednisone
   3. More than 1.5 mg/kg oral prednisone
   4. 500–1000 mg IV methylprednisolone

The survey link with a corresponding explanation of the study was sent via email to all 700 members of NANOS. The study period was open for 2 months (November 2019 to January 2020). A follow-up email was sent at the beginning of January 2020 as an additional reminder to complete the survey. Survey participation was anonymous, confidential, and entirely voluntary. No financial or other incentives or disincentives were provided for responding to the survey. Survey responses were collected and analysed using the standard, commercially available (SurveyMonkey®) software.

Results

One hundred and eighty (180) neuro-ophthalmologists out of the total 700 members belonging to NANOS completed the study within the two month period. This was a response rate of 26%. The average amount of time respondents spent on the survey was 52 seconds. A summary of the results of the survey is shown in Figure 1–4.

Figure 1.

Survey results from question 1

Figure 2.

Survey results from question 2

Figure 3.

Survey results from question 3

Figure 4.

Survey results from question 4

Discussion

Although it is widely agreed upon that corticosteroids are the preferred initial treatment for GCA, the precise optimal type, dosage, and route of administration of corticosteroid therapy remain unproven.¹⁶ Both rheumatologists and ophthalmologists diagnose and treat GCA, but they differ dramatically in their preferred prescribed dosage and route of corticosteroid therapy for GCA.¹⁶ Even among neuro-ophthalmologists, as demonstrated in our survey, there is variability in prescribing preferences.

Our results suggest that a large proportion (50%) of respondents prefer the IV route of corticosteroid therapy for patients with acute visual loss in GCA, but a significant percentage (40%) prefer either an IV or oral route. In question 3, when asked to choose the daily dosage of therapy for patients with acute visual loss, 75% chose the IV route of administration. These results could be a result of the 40% of respondents in question 1 who chose either IV or oral being split among answers in question 3. When forced to choose a dosage in question 3, about half of the 40% of respondents from question 1 plus the 50% of respondents choosing IV in question 1 could have resulted in 75% choosing the IV route in question 3. Only 27% of respondents chose an oral route of treatment in question 3, approximately half of the 40% of respondents in question 1 who answered either IV or oral. It is important to note that our survey was de-identified, so these observations are speculative, and it is not possible to track individual respondent’s answers to questions.

Furthermore, our study question was not designed to determine the type, dose, or duration of IV therapy (loading dose, three versus five days of initial therapy, oral tapering regimen) etc. In addition, our study was not designed to determine the rationale for choosing IV versus oral therapy or the corticosteroid dosing schedule after initial therapy. Although greater than 60% of respondents preferred the oral route for patients with GCA without visual loss, there was a significant proportion of respondents who still prefer the IV route.

Although high-dose IV corticosteroids have been hypothesised to be superior to oral corticosteroids in terms of effects on lipid peroxidation, there is no proof that this is sufficient rationale to prefer IV over oral corticosteroids in GCA.¹¹ Hayreh et al. reported that 7% of the patients treated initially with IV corticosteroids experienced visual improvement as opposed to 5% who were treated with oral corticosteroids only. However, this difference was not statistically significant.¹⁷ A later study performed by the same authors over a 27-year period studied 145 patients with biopsy-proven GCA (96 patients with visual loss and 29 patients without visual loss). Thirty-three per cent of patients were treated initially with IV corticosteroids and then transitioned to oral corticosteroids. Of the patients who had only oral corticosteroid therapy, the median starting dose was approximately 1 mg/kg/day with 40% of the patients on greater than 1.5 mg/kg/day. Of the patients studied, only 4% with visual loss showed any improvement with higher dose corticosteroid therapy. The study did not find a difference in effectiveness between oral and IV therapy in terms of preventing visual loss.¹⁶

In regard to the dosage of corticosteroid therapy in our study, greater than 70% would use 500–1000 mg IV methylprednisolone daily for GCA with acute visual loss. Various studies in the literature have cited this same dosage range for IV treatment of GCA with visual loss. One study performed by Salvarani et al. recommended giving patients an initial IV dose of 1000 mg methylprednisolone for three days with recent or impeding visual loss.¹⁸ A literature review performed by Lee et al. made the same recommendation.⁶

In contrast, greater than 70% would prescribe 1.0 to 1.5 mg/kg daily of oral prednisone daily to patients with GCA without acute visual loss. Earlier studies have recommended 0.7 mg/kg oral prednisone for patients with GCA.¹⁹ The same study mentioned earlier performed by Hayreh et al. found that 0.7 mg/kg daily was not adequate to control GCA and protect patients from visual loss.¹⁶ Based on the literature, Lee et al. recommended that GCA patients without visual loss receive 1–1.5 mg/kg per day oral prednisone.⁶

We recognise the limitations of our study. First, the survey response rate was 26%. All of the typical limitations inherent in both the survey study design and the typical response rates apply to our study including sample size, sampling error, recall bias, survey fatigue and participation bias, and selection and ascertainment bias. This response rate, however, is consistent with multiple prior medical survey rate averages. Physician response rates to web-based surveys on average were 35% across all subspecialties in one study. The response rate for general surgeons was one of the lowest at 29.6%.²⁰ Interestingly, a meta-analysis of survey-based studies has shown that non-respondent bias is less of a concern with physician surveys and that the demographic variables of early and late respondents do not differ appreciably.²¹ Second, our study did not provide incentives to participate. Prior studies have found that monetary incentives can increase response rates among physicians. One study analysing physician response rates to a mailed survey provided either a 20 USD incentive or a 50 USD incentive. The response rate rose from 52.1% to 67.8% with the increase in dollar amount of the incentive.²²

Another reason for our lower response rate could be that only one email reminder to complete the survey was issued. More frequent automated reminders may have increased response rates. Finally, the timing of the survey could have been a limiting factor. The survey was only open for a total of two months, from November 2019 to January 2020, an often very busy time of year for physicians.

Third, our survey questions were purposefully limited to route and dose for GCA with and without visual loss. We wished to limit the participant time commitment for the survey and to avoid survey fatigue both for our own survey and for future researchers wishing to survey the NANOS membership. We recognise that our study was not designed to answer questions about the rationale for individual clinicians choosing a specific corticosteroid route or dose.

In summary, our study results suggest that there is a potential consensus but not universal agreement among responding neuro-ophthalmologists on type, route, and dosage of corticosteroid therapy for the treatment of GCA with and without visual loss. Future longitudinal, prospective, “head to head”, multi-centre, adequately powered clinical trials are needed to make evidence-based recommendations regarding treatment of GCA.

Declaration of interest statement

The authors report no conflict of interest.