Abstract
Purpose:
The Optic Neuritis Treatment Trial(ONTT) showed that an MRI of the brain is a powerful predictor of developing multiple sclerosis(MS). However, surveys of practitioners suggest that the recommended use of imaging is not consistently followed in practice. With this study we aim to assess the rate at which newly diagnosed optic neuritis(ON) patients receive brain MRIs.
Methods:
This is a retrospective cohort study using administrative medical claims data from a large, national US insurer. All incident cases from 2000-2016 of ON in patients without MS were assessed. The primary outcome was a comparison of patterns of MRI scanning usage following diagnosis of ON. Secondary outcomes evaluated steroid treatment and progression to MS.
Results:
Of 2,865 qualified ON patients, 1,755(61.3%) received a brain MRI. At 1 year follow up, 629 patients(30.3%) of patients had progressed to MS, a rate that increased slightly to 34.3%(366 patients) within a 3-yr period of their initial ON diagnosis. 520(18.2%) patients received IV steroids, and 383(13.4%) received oral steroids within 30 days of the ON diagnosis.
Conclusion:
Across the U.S., a surprisingly low number of individuals obtain a brain MRI following onset of ON, suggesting physicians may not be fully assessing the risk of MS.
Keywords: Optic Neuritis, Optic Neuritis Treatment Trial, Multiple Sclerosis, Neuro-Ophthalmology Practice Patterns
Introduction:
Demyelinating optic neuritis is an acute inflammation of the optic nerve that primarily affects young individuals and has a robust association with multiple sclerosis (MS).1 Optic neuritis typically manifests as an acute loss of vision, often with associated eye pain, that progresses for several days or up to 3 weeks and then resolves over the course of several weeks.2 While corticosteroids are commonly used to reduce the optic nerve inflammation and hasten visual recovery, they have little effect on final visual outcomes, merely speeding the recovery of visual function with minimal long term improvements in visual field and contrast sensitivity.3 Importantly, the clinical standard of care is for optic neuritis patients to undergo MRI imaging of the brain to assess their risk of MS.4 Much of what we know about the demographics and relative characteristics of optic neuritis, as well as its strong associations with MS, is derived from the Optic Neuritis Treatment Trial (ONTT).5
The ONTT was the first large randomized clinical trial in neuro-ophthalmology, and provided important information about the disease course of optic neuritis, it’s association with MS, and the potential role of steroid therapy. The ONTT’s impact on the field was in part due to its robust clinical trial nature within a field previously limited to case studies and much smaller analyses.1 However, as with all clinical trials, its true legacy/effect on health care depends on whether the results are adhered to in clinical practice.
The ONTT results offer clear recommendations for the evaluation and management of optic neuritis patients, including: 1. MRI brain should be ordered for all new patients to assess the relative risk and possible therapy for MS prevention; 2. Treatment with IV steroids can be considered to hasten visual recovery or reduce risk of developing MS; and 3. If treatment is initiated, intravenous (IV) methylprednisolone should be used and not oral prednisone alone. While significant controversy exists over the dose of the oral steroid used for comparison in the ONTT, few would argue against the benefits of obtaining an MRI of the brain. Despite this, subsequent surveys of ophthalmology and neurology practitioners suggest that only a subset of practitioners follow these recommendations.6,7 While illuminating, physician surveys are subject to a host of limitations including self-report response bias, social desirability bias, and often, limited sample sizes. Recent advances in the availability of large scale databases of healthcare resources, and methods to analyze such data provide an opportunity to examine the impact of clinical trials on treatment practices using less biased information gathering methods.8–10 This type of approach has recently been applied in neuro-ophthalmology to study modifiable risk factors in ischemic optic neuropathy,11 and provides a potential method for more wide-scale analysis of how practice patterns in care of optic neuritis patients may or may not follow the evaluation recommendations of the ONTT.
In the current study, the impact of the ONTT on treatment of acute demyelinating optic neuritis, as reflected by how frequently its clinical recommendations occur, is assessed through evaluation of a large database of insurance claims. The primary outcome of this study was to evaluate the rate at which patients received the care recommended for incident optic neuritis as outlined in the ONTT. This was primarily assessed by examining the rate of patients who received an MRI of the brain. As secondary analyses, we evaluated the demographic factors associated with receiving the primary outcome, the rate patients received intravenous steroids, and the rate patients progressed to multiple sclerosis after their initial diagnosis of optic neuritis.
Methods:
Database:
This is a retrospective cohort study performed on the Clinformatics™ Data Mart Database (OptumInsight, Eden Prairie, MN) using its de-identified administrative medical claims and prescription claims data from a single large, national U.S. insurer. This database has been used previously to assess practice patterns in trends for ophthalmic disease9,12 as well as risk factors for optic nerve disease.11 The data used for this study included the years 2000-2016. Due to the de-identified nature of the data, The University of Pennsylvania’s Institutional Review Board deemed this study exempt from review.
Cohorts
All patients between the ages of 18 and 50 who had an incident diagnosis of optic neuritis on or after January 1st, 2002, and whose diagnosis of optic neuritis was confirmed on a second provider visit within 90 days of the first, were included. The date of the first diagnosis was considered the index date. Patients were required to have at least two uninterrupted years in the dataset prior to and 30 days after the index date, and to have never been diagnosed with multiple sclerosis on or before the index date. To further limit the occurrence of mis-diagnosis, all patients with a diagnosis before or after the index date that could be confused with optic neuritis or a systemic disease that could cause optic nerve infection or inflammation were also excluded. (Please see eTable 1 for full list of ICD9/ICD10 and CPT codes used during this study).
Outcomes of Interest and Statistical Analysis
Multiple analyses were run to evaluate the rate patients received the recommended care for optic neuritis as outlined in the ONTT. First, all patients were assessed to determine if they had an MRI of the brain within 30 days of the index date. We also tested a 120 day period as a sensitivity analysis. Patients were counted if they had both an orbital and brain MRI, but not if they only had an orbital MRI. Next, all patients were assessed to determine if they received a new diagnosis of MS after the index date. To examine secondary outcomes, all patients that were treated with steroids were assessed to determine if treatment was initiated with IV steroids instead of oral steroids within 30 days of the index date. Those who initiated both forms of steroids were considered to have had IV, but not if only oral prednisone was prescribed. Additional secondary analyses focused on demographic factors associated with having received a brain MRI or IV steroids. Multivariate logistic regression was used to generate odds ratios for these factors and assess this. Demographic covariates of interest evaluated in the model included age, gender, race, education level, yearly income, region of the country, and provider type (ophthalmology, neurology, or other). Results of the analyses were considered statistically significant for p<0.05 (two-tailed). Statistical analysis was performed using SAS (version 9.4; SAS Institute Inc., Cary, NC).
Results:
Of the 2,865 patients that qualified for the study (Figure 1), 1,755 (61.3%) of ON patients received a brain MRI within 30 days of their initial optic neuritis diagnosis. Of the 1,110 patients who did not receive a brain MRI, an additional 169 (5.9%) received an orbital MRI only, and the remaining subjects did not undergo any MRI imaging (Table 1). The number of patients receiving a brain MRI was relatively consistent over the study period and fluctuated between 53.8% (2004) to 67.4% (2013) (Table 1). Our sensitivity analysis of the 120 days showed very similar results to the 30 day analysis of numbers of patients receiving a brain MRI (data not shown). Of the ON patients, 520 (18.2%) received IV steroids, and 383 (13.4%) received oral steroids alone (without IV steroids) within 30 days of the ON diagnosis (Table 1).
Figure 1:
Flowchart of patients included in the study
Table 1:
ON patients receiving MRI and receiving steroid treatment by cohort year
| Cohort Year | Number of patients | Brain MRI (%) | Orbit MRI only (%) | IV steroids (%) | Oral steroids (%) | No steroid treatment (%) |
|---|---|---|---|---|---|---|
| 2002 | 101 | 65 (64.4%) | 2 (2.0%) | 10 (9.9%) | 7 (6.9%) | 84 (83.2%) |
| 2003 | 174 | 97 (55.7%) | 12 (6.9%) | 18 (10.3%) | 21 (12.1%) | 135 (77.6%) |
| 2004 | 197 | 106 (53.8%) | 14 (7.1%) | 27 (13.7%) | 17 (8.6%) | 153 (77.7%) |
| 2005 | 225 | 151 (67.1%) | 11 (4.9%) | 39 (17.3%) | 26 (11.6%) | 160 (71.1%) |
| 2006 | 252 | 150 (59.5%) | 14 (5.6%) | 52 (20.6%) | 26 (10.3%) | 174 (69.0%) |
| 2007 | 201 | 115 (57.2%) | 15 (7.5%) | 31 (15.4%) | 22 (10.9%) | 148 (73.6%) |
| 2008 | 235 | 138 (58.7%) | 11 (4.7%) | 46 (19.6%) | 27 (11.5%) | 162 (68.9%) |
| 2009 | 225 | 140 (62.2%) | 16 (7.1%) | 41 (18.2%) | 28 (12.4%) | 156 (69.3%) |
| 2010 | 234 | 148 (63.2%) | 13 (5.6%) | 51 (21.8%) | 29 (12.4%) | 154 (65.8%) |
| 2011 | 224 | 142 (63.4%) | 16 (7.1%) | 50 (22.3%) | 34 (15.2%) | 140 (62.5%) |
| 2012 | 208 | 133 (63.9%) | 5 (2.4%) | 44 (21.2%) | 37 (17.8%) | 127 (61.1%) |
| 2013 | 181 | 122 (67.4%) | 10 (5.5%) | 31 (17.1%) | 43 (23.8%) | 107 (59.1%) |
| 2014 | 164 | 103 (62.8%) | 10 (6.1%) | 31 (18.9%) | 29 (17.7%) | 104 (63.4%) |
| 2015 | 122 | 73 (59.8%) | 7 (5.7%) | 23 (18.9%) | 19 (15.6%) | 80 (65.6%) |
| 2016 | 122 | 72 (59.0%) | 13 (10.7%) | 26 (21.3%) | 18 (14.8%) | 78 (63.9%) |
| Total | 2865 | 1755 (61.3%) | 169 (5.9%) | 520 (18.1%) | 383 (13.4%) | 1962 (68.5%) |
In terms of demographics, the average age of ON patients who received a brain MRI was 36.9(SD ±8.5) years old, and average age of patients who did not receive a brain MRI was 36.6(±8.3) years old (Table 2). 70% of the population receiving a brain MRI and 65% of the population not receiving a brain MRI was female (Table 2). After accounting for numerous factors in the multivariate logistic regression, males were significantly less likely to have a brain MRI (OR = 0.79, 95% CI:0.67-0.93, p=0.005), and those with higher than a bachelor’s degree were more likely to get a brain MRI (OR=1.41, 95% CI: 1.08-1.84) compared to those with less than a high school diploma, but education as whole variable was not significant (p=0.15) (Table 3). Patients in the pacific (OR =0.58, 95% CI: 0.40-0.82, p<0.001) and mountain regions (OR = 0.64, 95% CI: 0.47-0.87, p<0.001) were significantly less likely to receive brain MRIs after an ON diagnosis (Table 3). Age, race and yearly income were not associated with obtaining a brain MRI (p>0.28 for all comparisons) (Table 3). Comparing across specialties, neurologists (OR=1.58, 95% CI: 1.27-1.96), p<0.001) and other providers (OR=2.08, 95% CI: 1.74-2.48), p<0.001) were significantly more likely to obtain an MRI than ophthalmologists (Table 3).
Table 2:
Baseline characteristics of subjects with or without brain MRI performed or steroid treatment received
| No Brain MRI (N=1110) | Brain MRI (N=1755) | p-value | Oral steroid (N = 383) | IV Steroid (N = 520) | p-value | |
|---|---|---|---|---|---|---|
| Age | 0.311 | 0.001 | ||||
| Mean (SD) | 36.9 (8.5) | 36.6 (8.3) | 37.4 (8.5) | 35.6 (8.2) | ||
| Race | 0.512 | 0.37 | ||||
| White | 787 (71%) | 1259 (72%) | 281 (73%) | 375 (72%) | ||
| Asian | 32 (3%) | 48 (3%) | 7 (2%) | 10 (2%) | ||
| Black | 110 (10%) | 179 (10%) | 43 (11%) | 55 (11%) | ||
| Hispanic | 103 (9%) | 132 (8%) | 33 (9%) | 37 (7%) | ||
| Unknown | 78 (7%) | 137 (8%) | 19 (5%) | 43 (8%) | ||
| Gender | <0.012 | 0.05 | ||||
| Female | 717 (65%) | 1224 (70%) | 244 (64%) | 363 (70%) | ||
| Male | 393 (35%) | 531 (30%) | 139(36%) | 157 (30%) | ||
| Education level | 0.062 | 0.56 | ||||
| Less than 12th Grade | 6 (1%) | 10 (1%) | 3 (1%) | 6 (1%) | ||
| High School Diploma | 297 (27%) | 404 (23%) | 99 (26%) | 124 (24%) | ||
| Less than Bachelor Degree | 584 (53%) | 920 (52%) | 200 (52%) | 266 (51%) | ||
| Bachelor Degree Plus | 182 (16%) | 352 (20%) | 74 (19%) | 106 (20%) | ||
| Unknown | 41 (4%) | 69 (4%) | 7 (2%) | 18 (3%) | ||
| Yearly Income | 0.142 | 0.32 | ||||
| <$40K | 107 (10%) | 133 (8%) | 32 (8%) | 163 (31%) | ||
| $40K - $49K | 53 (5%) | 82 (5%) | 20 (5%) | 43 (8%) | ||
| $50K - $59K | 50 (5%) | 80 (5%) | 22 (6%) | 26 (5%) | ||
| $60K - $74K | 84 (8%) | 142 (8%) | 26 (7%) | 17 (3%) | ||
| $75K - $99K | 123 (11%) | 235 (13%) | 56 (15%) | 51 (10%) | ||
| $100K+ | 334 (30%) | 568 (32%) | 119 (31%) | 66 (13%) | ||
| Unknown | 359 (32%) | 515 (29%) | 108 (2%) | 154 (30%) | ||
| Geographic region | 0.012 | 0.13 | ||||
| Upper Midwest | 285 (26%) | 485 (28%) | 113 (30%) | 149 (29%) | ||
| Southern Midwest | 175 (16%) | 322 (18%) | 65 (17%) | 119 (23%) | ||
| Northeast | 92 (8%) | 181 (10%) | 33 (9%) | 54 (10%) | ||
| Mountain | 109 (10%) | 131 (7%) | 35 (9%) | 33 (6%) | ||
| Pacific | 80 (7%) | 87 (5%) | 17 (4%) | 24 (5%) | ||
| South Atlantic | 368 (33%) | 548 (31%) | 120 (31%) | 139 (27%) | ||
| Unknown | 1 (0%) | 1 (0%) | 0 (0%) | 2 (0%) | ||
| Physician | <0.012 | <0.012 | ||||
| Ophthalmology | 418 (38%) | 426 (24%) | 94 (25%) | 77 (15%) | ||
| Neurology | 228 (21%) | 367 (21%) | 74 (19%) | 110 (21%) | ||
| Other | 464 (42%) | 962 (55%) | 215 (56%) | 333 (64%) |
Equal Variance T-Test
Chi-Square
Table 3:
Multivariate logistic regression of characteristics of patients that received a brain MRI and of patients who received initial IV steroids compared to oral steroids.
| Characteristic | Odds Ratio of receiving an MRI (95%CI) | P-value | Odds Ratio of Receiving IV steroid (95% CI) | P-value | |
|---|---|---|---|---|---|
| Age | 0.99 (0.99, 1.00) | 0.28 | 0.97 (0.96, 0.99) | 0.002 | |
| Race | Unknown | 1.20 (0.81, 1.78) | 0.76 | 1.51 (0.76, 3.00) | 0.52 |
| Hispanic | 0.88 (0.66, 1.18) | 0.72 (0.42, 1.24) | |||
| Black | 1.04 (0.80, 1.36) | 0.91 (0.57, 1.44) | |||
| Asian | 1.02 (0.64, 1.65) | 0.92 (0.34, 2.52) | |||
| White | reference | reference | |||
| Gender | Male | 0.79 (0.67, 0.93) | 0.005 | 0.75 (0.56, 1.01) | 0.06 |
| Female | reference | reference | |||
| Education level | Unknown | 1.29 (0.74, 2.24) | 0.15 | 1.68 (0.55, 5.11) | 0.85 |
| Bachelor Degree Plus | 1.41 (1.08, 1.84) | 1.17 (0.74, 1.86) | |||
| Less than Bachelor Degree | 1.20 (0.98, 1.46) | 1.11 (0.77, 1.60) | |||
| Less than 12th Grade | 1.28 (0.44, 3.68) | 1.60 (0.36, 7.17) | |||
| High School Diploma | reference | reference | |||
| Yearly Income | $100K+ | 1.18 (0.86, 1.63) | 0.28 | 0.81 (0.44, 1.48) | 0.39 |
| $75K - $99K | 1.36 (0.96, 1.93) | 0.77 (0.41, 1.46) | |||
| $60K - $74K | 1.26 (0.86, 1.85) | 1.31 (0.65, 2.63) | |||
| $50K - $59K | 1.24 (0.80, 1.94) | 0.51 (0.23, 1.14) | |||
| $40K - $49K | 1.14 (0.73, 1.77) | 0.83 (0.38, 1.79) | |||
| Unknown | 1.00 (0.73, 1.36) | 0.88 (0.50, 1.56) | |||
| <$40K | reference | reference | |||
| Geographic region | Unknown | 0.32 (0.02, 5.28) | <.001 | 0.06 | |
| South Atlantic | 0.87 (0.71, 1.07) | 0.91 (0.63, 1.31) | |||
| Pacific | 0.58 (0.40, 0.82) | 0.96 (0.47, 1.95) | |||
| Mountain | 0.64 (0.47, 0.87) | 0.68 (0.39, 1.19) | |||
| Northeast | 1.18 (0.87, 1.59) | 1.32 (0.79, 2.23) | |||
| Southern Midwest | 1.11 (0.88, 1.42) | 1.52 (1.01, 2.28) | |||
| Upper Midwest | reference | reference | |||
| Physician | Neurology | 1.58 (1.27, 1.96) | <.001 | 1.82 (1.18, 2.80) | .002 |
| Other | 2.08 (1.74, 2.48) | 1.89 (1.32, 2.71) | |||
| Ophthalmology | reference | reference |
At baseline, patients who received IV steroids within 30 days of ON diagnosis were younger (35.6 yo vs. 37.4 yo, p=0.001) and more likely female (70% vs. 64%, p=0.05) compared to those that received oral steroids (Table 2). They did not significantly differ across race, education level, yearly income, or geographic region. After adjusting for all covariates, younger age (OR=0.97, 95% CI: 0.96-0.99, p=0.002) was significantly associated with receiving IV steroids (Table 3). Additionally, patients of neurologists (OR=1.82, 95% CI: 1.18-2.80, p<0.002) and other providers (OR=1.89, 95% CI: 1.32-2.71, p<0.002) were significantly more likely to receive IV steroids compared to those of ophthalmologists. Patients in the southern Midwest (OR=1.52, 95% CI:1.01-2.28) of the US were more likely to receive IV steroids compared to the upper Midwest reference group, however the variable as a whole was not significantly associated with IV steroid treatment (p=0.06) (Table 3).
Of the ON patients, 2,073 patients had at least 1 year of follow up data, and 1,067 patients had at least 3 years (Table 4). At 1 year follow up, 629 (30.3%) of patients had received a new diagnosis of MS (Table 4). The rate of progression to MS diagnosis slightly increased to 34.3% (366 patients) within a 3 yr period of their initial ON diagnosis (Table 4). The rate of progression to MS was significantly greater in patients prescribed IV steroids compared to patients prescribed oral steroids (Table 5). Of the 189 patients prescribed IV steroids with 3 year follow up data available, 124 (65.6%) progressed to MS, whereas of the 129 patients prescribed oral steroids with 3 year follow up data available, only 50 (38.8%) progressed to MS (Table 5).
Table 4:
Rate of optic neuritis progression to MS over 1- and 3-year follow-up
| Cohort Year | Number of patients with 1 year follow-up | Rate of progression to MS over 1 year follow-up (%) | Number of patients with 3 year follow-up | Rate of progression to MS over 3 year follow-up(%) |
|---|---|---|---|---|
| 2002 | 68 | 18 (26.5%) | 34 | 14 (41.2%) |
| 2003 | 127 | 34 (26.8%) | 66 | 26 (39.4%) |
| 2004 | 146 | 51 (34.9%) | 85 | 36 (42.4%) |
| 2005 | 171 | 52 (30.4%) | 102 | 33 (32.4%) |
| 2006 | 192 | 51 (26.6%) | 113 | 34 (30.1%) |
| 2007 | 159 | 43 (27.0%) | 81 | 23 (28.4%) |
| 2008 | 173 | 49 (28.3%) | 105 | 32 (30.5%) |
| 2009 | 171 | 45 (26.3%) | 97 | 29 (29.9%) |
| 2010 | 178 | 60 (33.7%) | 114 | 42 (36.8%) |
| 2011 | 180 | 64 (35.6%) | 105 | 40 (38.1%) |
| 2012 | 162 | 50 (30.9%) | 86 | 24 (27.9%) |
| 2013 | 129 | 48 (37.2%) | 79 | 33 (41.8%) |
| 2014 | 119 | 32 (26.9%) | ||
| 2015 | 98 | 32 (32.7%) | ||
| Total: 2002-2015: | 2073 | 629 (30.3%) | 2002-2013: 1067 | 366 (34.3%) |
Table 5:
Comparison of rate of progression to MS between ON patients receiving IV steroids and oral steroids
| Number of patients | Rate of MS over 3 years follow-up | p-value* | |
|---|---|---|---|
| <0.01 | |||
| IV steroids | 189 | 124 (65.6%) | |
| Oral steroids | 129 | 50 (38.8%) |
Chi-square test
Discussion:
We used a national dataset to find that of incident optic neuritis patients, only 61.3% received a brain MRI. This rate of imaging is less than physician surveys which found that 66-76.5% of ophthalmologists and 71-87.4% of neurologists order brain MRI scans in the majority of their ON patients.6,7 Similar to the trends in these survey-based studies, our results suggest ophthalmologists are less likely than neurologists to order brain MRIs for optic neuritis patients. Both in our study and the survey studies, the rate of MRI attainment is lower than expected given the results of the ONTT that showed that optic neuritis patients with lesions found on brain MRIs had a 3 fold higher likelihood of progressing to develop MS than individuals without brain lesions.13,14 Additionally, for individuals at high risk for MS, IV methylprednisolone demonstrated a protective effect up to 2-3 years.15–17
While it is clear not all practitioners adhere to the ONTT recommendations, our study shows to what degree this does not take place, demonstrating that 39% of new ON patients are not getting the proper imaging work up. Although important, physician survey data likely overestimates the adherence to clinical trials since they rely on the responses of practitioners. Bias results are likely to occur by either receiving responses selectively from practitioners most adherent to clinical trial data, or an overestimation on the part of the respondents as to how often they do indeed order MRIs. The current study objectively measures the rate of MRI imaging using insurance claims data to identify which patients diagnosed with ON did receive a brain MRI. Our results suggest that rates of recommended imaging being performed are indeed even lower than prior survey studies had reported.
Interestingly, our current results also show a higher rate of subsequent diagnosis/development of MS following ON than seen in the ONTT. Our data show that 30.3% and 34.3% of ON patients, all of whom had no prior history of MS, were subsequently diagnosed with MS within one or three years, respectively, of their index date. In the ONTT, a range of only 6.4-10.5% of patients at one year and 17.3-24.7% of patients at three years developed clinically definite MS across the three treatment groups studied in the ONTT.17 The higher rate found here is not surprising given that diagnostic criteria for MS have evolved since the time of the ONTT to include radiographic features in addition to clinical criteria18, thus allowing diagnosis of additional patients at the time of their initial presentation or on subsequent imaging studies even in the absence of new clinical episodes of MS.
Patients who received a brain MRI were compared to those who did not receive one across a range of demographic factors, most of which found no association. However, patients with a high education level (eg bachelors and above) were significantly more likely to get a brain MRI, which may be due to differences in insight about their disease. In addition, patients in pacific and mountain regions of the US were significantly less likely to get MRIs suggesting practice patterns or MRI access may differ across the country. While it is unclear why these regions would differ from the rest of the country, further study is warranted to understand why MRI utilization would be different in these regions.
Secondary analyses showed that only 18.2% of patients received IV steroids, and 13.4% received oral steroids (without preceding IV steroids), within 30 days of their ON diagnosis. The ONTT demonstrated that oral prednisone alone did not improve visual recovery, and in fact, increased the recurrence rate of ON.3 Thus, a main recommendation of the ONTT was that oral steroids not be used alone for ON, and the 13.4% rate of use found here may represent a deviation from the standard of care. However, this aspect of the ONTT has been a topic of debate because of the differing total amounts of corticosteroid given between the oral and IV arms.
A number of later studies have looked at the potential role of using megadose oral steroids to better approximate the IV dose used in the ONTT.19–25 While results support equivalent visual outcomes in most of these studies, including a recent randomized controlled trial of megadose oral vs IV methylprednisolone,24 such visual outcomes are not necessarily surprising since the ONTT already showed that even IV steroids have no long-term visual benefit compared with placebo.3 Importantly, most subsequent studies have not examined the features noted to be differentially affected by IV vs oral steroids in the ONTT, specifically recurrence rates of ON or rates of MS development. Thus, it remains debated whether oral corticosteroids can be substituted for IV when choosing to treat ON, and our current results suggest that almost half of practitioners who are treating with steroids indeed choose to use oral steroids alone.
Interestingly, our results show only a little more than 30% of ON patients receive any steroid treatment (18.2% IV and 13.4% oral), a rate much lower than that reported by previous survey studies (> 50% IV methylprednisolone usage).6 The low rate of use of steroids found here may suggest that practitioners have indeed understood the lack of visual benefit of this treatment. It may also represent patients’ unwillingness to be admitted to a hospital for 3 days of IV steroid administration or the difficulty in finding an in home IV service in some parts of the country.
Across a range of demographic variables, no difference was found for increasing or decreasing the odds of being prescribed oral vs. IV steroids. Interestingly, of the patients given IV steroids, 65.6% of patients progressed to MS over a 3 year follow up period. This is in contrast to patients that received oral steroids, with a progression to MS of 38.8% over 3 years. This difference is probably not unexpected given that IV steroids are recommended mainly for those patients with high risk features for MS development found on MRI, although the rate of progression to MS diagnosis over three years in this subgroup is higher than that reported in the ONTT.17 This possibly reflects the evolving diagnostic criteria for MS.19 Conversely, it is also important to point out the strong likelihood that many of the 38.8% of patients who progressed to MS after receiving oral steroids may have benefited from having IV instead.
Our results are consistent with others suggesting that the main legacy of the ONTT may not be impacting ON treatment practices as much as expected, given the size and validity of the study. 6,7 Trobe et al. (1999) found that following the ONTT, 84% of ophthalmologists and 72% of neurologists reported changing some component of their treatment regimen for acute ON.6 Biousse et al. (2009) carried out an international survey and found that the rate of obtaining brain MRIs for ON patients by ophthalmologists and neurologists varied widely between countries, ranging from 4.9% - 94.2%, with highest rates in the United States and France.7 These results in the US are well above the percent of MRIs found to be obtained by analysis of insurance claims in our current study, suggesting that adherence to recommendations for brain imaging may be lower than previously recognized. In addition, although most patients did receive acute treatment with IV steroids, Biousse et al. found that between 14-65% of neurologists and ophthalmologists still recommended oral prednisone for the treatment of acute isolated ON, more than expected given the ONTT recommendations, and more than found in our present results.7 Overall, results from our current study suggest a lower number of physicians send patients for MRIs following an acute episode of optic neuritis than previously thought.
Limitations inherent to the study design need to be recognized. First, administrative claims data lacks specific clinical and radiographic information and therefore, is unable to confirm that each patient case is truly ON. In an effort to reduce this limitation and only capture data from true ON cases, a series of exclusions in selecting the study cohort were used. The ages of the cohort were restricted to be between 18 and 50 since ON is most common in young adults and its frequency drops at older ages.14,26 Individuals with diagnoses similar to ON at any time point or with prior diagnosis of systemic inflammation/infectious disease were also excluded in order to limit misdiagnosis. An ON diagnosis was also mandated on two separate physician visits in order to eliminate patients that may have been suspected to have ON by their initial provider, but then did not have a confirmatory diagnosis on a subsequent visit. Despite taking these steps, the possibility that misdiagnosis occurred cannot be ruled out. This may be a particularly important limitation for studies of optic neuritis, due to the significant rate of misdiagnosis that has been reported.27 Next, claims data is unable to capture results from imaging studies, limiting our ability to determine who had high risk findings for progression to MS.
Overall, our results suggest that despite use in the majority of cases, a surprisingly low number of individuals are getting a brain MRI, the greatest predictive and prognostic factor noted in the ONTT to determine progression to MS. Furthermore, there are unexplained demographic factors that may increase or decrease the rate of MRI utilization, as well as significant differences in practice patterns across specialties, suggesting further exploration may be warranted to determine whether there are characteristic presentations that justify differences in practice patterns, or whether there may be innate biases in the healthcare system and insurance practices.
Supplementary Material
Acknowledgements:
Financial Support: National Institutes of Health K23 Award (1K23EY025729 - 01) and University of Pennsylvania Core Grant for Vision Research (2P30EY001583). The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH. Additional funding was provided by Research to Prevent Blindness and the Paul and Evanina Mackall Foundation. Funding from each of the above sources was received in the form of block research grants to the Scheie Eye Institute. None of the organizations had any role in the design or conduction of the study
This research was conducted entirely at the University of Pennsylvania. Brian VanderBeek had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis
Footnotes
Conflicts of Interest: No conflicting relationship exists for any author.
References
- 1.Volpe NJ. The optic neuritis treatment trial: a definitive answer and profound impact with unexpected results. Arch Ophthalmol. 2008;126:996–999. [DOI] [PubMed] [Google Scholar]
- 2.Beck RW, Cleary PA, Jye-yu C. The course of visual recovery after optic neuritis: experience of the Optic Neuritis Treatment Trial. Ophthalmology. 1994;101(11):1771–1778. [DOI] [PubMed] [Google Scholar]
- 3.Beck RW, Cleary PA, Anderson MM Jr, et al. A randomized, controlled trial of corticosteroids in the treatment of acute optic neuritis. New Engl J Med. 1992;326(9):581–588. [DOI] [PubMed] [Google Scholar]
- 4.Petzold A, Wattjes MP, Costello F, et al. The investigation of acute optic neuritis: a review and proposed protocol. Nat Rev Neurol. 2014;10(8):447. [DOI] [PubMed] [Google Scholar]
- 5.Beck RW. The optic neuritis treatment trial: Implications for clinical practice. Arch Ophthalmol. 1992;110(3):331–332. [DOI] [PubMed] [Google Scholar]
- 6.Trobe JD, Sieving PC, Guire KE, Fendrick AM. The impact of the optic neuritis treatment trial on the practices of ophthalmologists and neurologists. Ophthalmology. 1992;106(11):2047–2053. [DOI] [PubMed] [Google Scholar]
- 7.Biousse V, Calvetti O, Drews-Botsch CD, et al. Management of optic neuritis and impact of clinical trials: an international survey. J Neurol Sci. 2009;276(1):69–74. [DOI] [PubMed] [Google Scholar]
- 8.VanderBeek BL, Bonaffini SG, Ma L. The association between intravitreal steroids and post-injection endophthalmitis rates. Ophthalmology. 2015;122(11):2311–2315. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.VanderBeek BL, Shah N, Parikh PC, Ma L. Trends in the Care of Diabetic Macular Edema: Analysis of a National Cohort. PLoS One. 2016;11(2):e0149450. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Vanderbeek BL, Zacks DN, Talwar N, et al. Racial differences in age-related macular degeneration rates in the United States: a longitudinal analysis of a managed care network. Am J Ophthalmol. 2011;152(2):273–282. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Cestari DM, Gaier ED, Bouzika P, et al. Demographic, systemic, and ocular factors associated with nonarteritic anterior ischemic optic neuropathy. Ophthalmology. 2016;123(12):2446–2455. [DOI] [PubMed] [Google Scholar]
- 12.Sarezky D, Orlin SE, Pan W, VanderBeek BL. Trends in Corneal Transplantation in Keratoconus. Cornea. 2017;36(2):131–137. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Optic Neuritis Study Group. Multiple sclerosis risk after optic neuritis. Arch Neurol. 2008;65(6):727–732. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Beck RW, Trobe JD, Moke PS, et al. High and Low-risk profiles for the development of Multiple Sclerosis within 10 years after optic neuritis: experience of the optic neuritis treatment trial. Arch Ophthalmol. 2003;1(7):944–9 [DOI] [PubMed] [Google Scholar]
- 15.Beck RW, Cleary PA. Optic neuritis treatment trial: one-year follow-up results. Arch Ophthalmol. 1993;111(6):773–775. [DOI] [PubMed] [Google Scholar]
- 16.Beck RW, Cleary PA, Trobe JD, et al. The effect of corticosteroids for acute optic neuritis on the subsequent development of multiple sclerosis. New Engl J Med. 1993;329(24):1764–1769. [DOI] [PubMed] [Google Scholar]
- 17.Beck RW. The optic neuritis treatment trial: three-year follow-up results. Arch Ophthalmol. 1995;113(2):136–137. [DOI] [PubMed] [Google Scholar]
- 18.Thompson AJ, Banwell BL, Barkhof F, et al. Diagnosis of multiple sclerosis: 2017 revisions of the McDonald criteria. Lancet Neurol. 2018;17(2):162–173 [DOI] [PubMed] [Google Scholar]
- 19.Sellebjerg F, Nielsen HS, Frederiksen JL, Olesen J. A randomized, controlled trial of oral high-dose methylprednisolone in acute optic neuritis. Neurology. 1999;52(7):1479–1479. [DOI] [PubMed] [Google Scholar]
- 20.Kupersmith MJ, et al. Megadose corticosteroids in multiple sclerosis. Neurology 441 (1994): 1–1. [DOI] [PubMed] [Google Scholar]
- 21.Brusaferri Fabio, and Candelise Livia. Steriods for multiple sclerosis and optic neuritis: a meta-analysis of randomized controlled clinical trials. Journal of neurology 2476 (2000): 435–442. [DOI] [PubMed] [Google Scholar]
- 22.Kapoor R, et al. “Effects of intravenous methylprednisolone on outcome in MRI-based prognostic subgroups in acute optic neuritis.” Neurology 501 (1998): 230–237. [DOI] [PubMed] [Google Scholar]
- 23.Naumovska M, Sheikh R, Bengtsson B, Malmsjo M, Hammar B. Visual outcomes is similar in optic neuritis patients treated with oral and i.v. high-dose methylprednisolone: a retrospective study on 56 patients. BMC Neurology. 2018;18(1):160. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Morrow SA, Fraser JA, Day C, Bowman D, Rosehart H, Kremenchutzky M, Nicolle M. Effect of Treating Acute Optic Neuritis with Bioequivalent Oral vs IV Corticosteroids: A Randomized Clinical Trial. JAMA Neurolology. 2018;75(6):690–696. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Gal RL, Vedula SS, Beck R. Corticosteroids for treating optic neuritis. Cochrane Database Syst Review. 2015;14(8). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Pineles SL, Balcer LJ. Visual loss: optic neuropathies In: Liu GT, Volpe NJ, Galetta SL, eds. Neuro-Ophthalmology, Diagnosis and Management. Edinburgh, UK: Elsevier; 2018:101–198. [Google Scholar]
- 27.Stunkel L, Kung NH, Wilson B, McClelland CM, Van Stavern GP. Incidence and Causes of Overdiagnosis of Optic Neuritis. JAMA Ophthalmol. 2018;136(1):76–81. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.