Journal of Neuro-Ophthalmology, September 2015, Volume 35, Issue 3

Literature Commentary

Literature Commentary Matthews L, Enzinger C, Fazekas F, Rovira A, Ciccarelli O, Dotti MT, Filippi M, Frederiksen JL, Giorgio A, KükerW, Lukas C, Rocca MA, De Stefano N, Toosy A, Yousry T, Palace J. MRI in Leber's hereditary optic neuropathy: the relationship to multiple sclerosis. J Neurol Neurosurg Psychiatry. 2015;86:537-542. Background: Leber's hereditary optic neuropathy (LHON) and a multiple sclerosis (MS)-like illness appear to coexist 50 times more frequently than would be expected by chance. This association of LHON and MS (LMS) raises an important question about whether there could be a com-mon pathophysiological mechanism involving mitochondrial dysfunction. Objective: The primary aim was to define magnetic reso-nance imaging (MRI) features of LMS and LHON and to assess the proportions of individuals displaying features typical of MS. Secondarily, we investigated the effect of gender on the risk of developing white matter lesions in the context of LHON. Methods: A blinded standardized review of conventional brain MRIs of 30 patients with MS, 31 patients with LHON, and 11 patients with LMS was conducted by 3 independent experts in the field. MS-like MRI features were assessed. Results: All patients with LMS and 26% of patients with LHON had white matter lesions. Of these, all patients with LMS and 25% with LHON were found to have an MRI appearance typical of MS. Female patients with LHON had a significantly greater risk of having white matter lesions, consistent with MS compared with male patients (relative risk 8.3). Conclusions: A blinded review of conventional brain MRIs shows that patients with LMS have a scan appearance indistinguishable from MS. Mitochondrial dysfunction could be a common pathophysiological pathway in the formation of white matter lesions. There appears to be a strong female influence on the radiological appearance and clinical development of MS in patients with LHON. In this observational, retrospective, brain magnetic reso-nance imaging (MRI) study totaling 72 patients, 3 neuroradiology experts performed a blinded analysis on 3 set of patients: MS (n = 30), LHON (n = 31), and LMS (i.e., Harding syndrome (1)) (n = 11). There were several intriguing findings in this study. First, in terms of general numbers, 1. 8 of 13 patients (26%) with LHON had T2 hyperintense white matter lesions, 4 of 13 (13%) had T1 hypointense white matter lesions, and 4 of 13 (13%) had brain atrophy; 2. 2 of 8 patients (25%) with LHON and T2 hyperintense white matter lesions had MRI findings characteristic of MS; 3. 11 of 11 patients (100%) with LMS (Harding syn-drome) had T2 hyperintense white matter lesions and MRI findings consistent with MS; and 4. 27 of 30 patients (90%) with MS were felt to have MRI findings typical of MS. What I find interesting about these data is that nearly one quarter of the LHON patients had an abnormal MRI. All patients with LMS had MRI findings that for all intents and purposes were MS, which strongly supports the concept of a unifying pathophysiological mechanism. Additionally, I found it interesting that 10% of the patients with MS had MRI findings that were not consistent with MS. This suggests to me that you can't always rely on the size, shape, and location of lesions to confirm the diagnosis of MS. Second, in terms of gender findings, 1. of the 12 women with either LHON or LMS, 4 had LHON and 8 had LMS; 2. of the 30 men with either LHON or LMS, 27 had LHON and 3 had LMS; 3. of the 4 women with LHON, 3 (75%) had asymptom-atic T2 hyperintense white matter lesions; and 4. of the 27 men with LHON, 5 (18.5%) had asymptom-atic T2 hyperintense white matter lesions. Even though male patients with the LHON genetic mutation had a greater chance of phenotypic expression than female patients with the same LHON genetic mutation, in this study, there were more female patients with LMS (n = 8) than male patients (n = 4). The calculated relative risk of a LHON woman having cerebral white matters lesions was 8.3 (95% confidence interval, 2.8-25.1; P , 0.01). In the discussion, the authors highlighted the fact that there is growing evidence that mitochondrial dysfunction plays an important role in the pathophysiology of MS, and of course, there is no debate that mitochondrial dysfunction is the primary factor in the loss of vision in patients with LHON. This study adds to the body of evidence that the association of MS and LHON is more than coincidental and that there is a plausible scientific explanation for the Moster and Bhatti: J Neuro-Ophthalmol 2015; 35: 323-328 323 Literature Commentary Section Editors: Mark L. Moster, MD M. Tariq Bhatti, MD Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. MS-like illness that is seen is some patients with LHON. One of the questions that remains unknown is the risk of developing MS in a patient with LHON and asymptomatic white matter lesions on MRI. -M. Tariq Bhatti, MD I would agree with most of what you say. However, your love of LHON and MS overlap, notwithstanding I don't think we know for sure that the incidence of this overlap syndrome is more than a chance occurrence of a common disease (MS) in a population with LHON. Other reports (2) have suggested this, and one would expect a more fre-quent diagnosis of MS by MRI or clinical criteria in a pop-ulation who already presented with severe visual loss. -Mark L. Moster, MD REFERENCES 1. Chalmers RM, Harding AE. A case control study of Leber's hereditary optic neuropathy. Brain. 1996;119:1481-1486. 2. Pfeffer G, Burke A, Yu-Wai-Man P, Compston DA, Chinnery PF. Clinical features of MS associated with Leber hereditary optic neuropathy mtDNA mutations. Neurology. 2013;81:2073- 2081. Chen Y, Morgan ML, Barros Palau AE, Yalamanchili S, Lee AG. Evaluation and neuroimaging of the Horner syndrome. Can J Ophthalmol. 2015;50:107- 111. Objective: To define the efficacy, safety, and cost-effectiveness of a single center approach to evaluating Horner syndrome (HS) including a simplified single neuro-imaging protocol. Design: Case series study. Participants: Medical records of 34 patients diagnosed with HS at the Houston Methodist Hospital (HMH) were reviewed after obtaining institutional review board approval. Methods: A retrospective chart review was performed for all patients presenting with the diagnosis of HS at the HMH from January 2010 to November 2013. All patients had diagnostic imaging with contrast-enhanced brain magnetic resonance imaging (MRI) extending to the T2 level in the chest. They either had documented causative diagnosis for HS or were "idiopathic." Efficacy and cost-effectiveness of the proposed neuroimaging technique were analyzed and compared with other recommended protocols. Results: We initially reviewed 34 charts with presumed diagnosis of HS; 27 charts were included in the analysis. The average age of patients was 46.6 years. Eleven patients (41%) had a final diagnosis of HS secondary to a proven cause, and 16 patients (59%) were diagnosed as "idiopathic." Ten patients (63%) in the idiopathic group had follow-up, and none of those with follow-up had an alterna-tive cause. The estimated cost of our recommended MRI protocol was US $667.76 without magnetic resonance angiography (MRA) or US $1501.71 with MRA. Conclusions: A single contrast-enhanced brain MRI ex-tending to the T2 level in the chest is an effective and simple means of ruling out life-threatening and other causative factors of HS. Compared with previous imaging recommendations, this proposed protocol may be simpler for clinicians to use and more cost-effective. The authors describe their approach to evaluating patients with non-localizable, isolated HS, noting that prior literature has proposed various different approaches. They use standard clinical criteria, selective use of apraclonidine, and then imaging with magnetic resonance imaging (MRI) extending from the brain to the T2 spinal cord level and a magnetic resonance angiography (MRA) of the neck. They find the approach to be effective and lower cost than many others. As the authors note, there are limitations to this study, including not knowing the true cost of studies (they used Medicare reimbursement rates), a small number of patients, limited follow-up, and retrospective data collection. So why review this article? As the subspecialty with the most expertise in HS, I think we have to come up with the best way to evaluate these patients and propose some changes to our radiology departments, institutions, and insurance companies. With this in mind, I think the authors approach makes a lot of sense. Currently, what I order on patients with no-localizable, isolated HS depends on which office I see them in (Wills Eye vs suburban Philadelphia vs south New Jersey), at which institution they are studied, and what insurance they have. The orders might include a combination of the following: MRI brain, MRI neck, MRI chest, MRA neck, CT brain, CT neck, CT chest, or computed tomographic angiography (CTA). Sometimes, the decision of what to obtain depends on my conversation with the medical reviewer at the insurance company. Even so, occasionally, I am not confident that I've adequately imaged the entire oculosympathetic pathway. If we can come up with an agreed upon guidelines along the lines recommended in this article, perhaps in partnership with the American Academy of Neurology or the American Academy of Ophthalmology, we can get appropriate imag-ing, easier insurance approvals, and not have to order up to 4 different studies. Additionally, when I've seen patients who have been evaluated by neurologists or ophthalmologists, often a portion of the oculosympathetic pathway has been overlooked. Since many or most HS patients don't make it to a neuro-ophthalmologist, a simplified recommendation, such as proposed here, will lead to a more appropriate evaluation. As the authors note, their protocol is likely faster to perform, simpler for clinicians to order, less costly, and adequately assesses all areas of concern in patients with HS. -Mark L. Moster, MD I agree with you, Mark, that it is very important for specific guidelines to be developed to aid clinicians in evaluating patients with common disorders, such as HS. However, a step further in the process would be to develop "standard-of-care" protocols that can be accepted univer-sally in both the medical and legal communities. 324 Moster and Bhatti: J Neuro-Ophthalmol 2015; 35: 323-328 Literature Commentary Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. I was happy to see this article because it very much parallels the neuroimaging protocol that we developed here at the Duke University Medical Center with the neurora-diology service, which is a contrasted MRI from the head to the upper chest and cervical MRA. My pharmacological strategy for HS is exactly the same as the authors, which is to confirm the diagnosis of HS with apraclonidine (some-times cocaine) but not to perform the hydroxyamphetamine test. As an aside, I found it very interesting that 59% of the patients evaluated were found to have idiopathic HS. What this study does not answer is whether MRI/MRA is better than CT/CTA in the evaluation of a patient with HS. -M. Tariq Bhatti, MD Heussinger N, Kontopantelis E, Gburek-Augustat J, Jenke A, Vollrath G, Korinthenberg R, Hofstetter P, Meyer S, Brecht I, Kornek B, Herkenrath P, Schimmel M, Wenner K, Häusler M, Lutz S, Karenfort M, Blaschek A, Smitka M, Karch S, Piepkorn M, Rostasy K, Lücke T, Weber P, Trollmann R, Klepper J, Häußler M, Hofmann R, Weissert R, Merkenschlager A, Buttmann M; GRACE-MS (German-speaking Research Alliance for ChildrEn with MS). Oligoclonal bands predict multiple sclerosis in children with optic neuritis. Ann Neurol. 2015;77:1076-1082. Abstract: We retrospectively evaluated predictors of con-version to multiple sclerosis (MS) in 357 children with isolated optic neuritis (ON) as a first demyelinating event who had a median follow-up of 4.0 years. Multiple Cox proportional hazard regressions revealed abnormal cranial magnet resonance imaging (cMRI; hazard ratio [HR], 5.94; 95% confidence interval [CI], 3.39-10.39; P , 0.001), presence of cerebrospinal fluid oligoclonal IgG bands (OCB; HR, 3.69; 95% CI, 2.32-5.86; P , 0.001), and age (HR, 1.08 per year of age; 95% CI, 1.02-1.13; P = 0.003) as independent predictors of conversion, while sex or later-ality (uni- vs bilateral) had no influence. Combined cMRI and OCB positivity indicated a 26.84-fold higher HR for develop-ing MS compared to double negativity (95% CI, 12.26- 58.74; P , 0.001). Accordingly, cerebrospinal fluid analysis may supplement cMRI to determine the risk of MS in chil-dren with isolated ON. Despite the urge to consider pediatric optic neuritis as a form of adult optic neuritis, it is evident that these 2 entities are quite different from each other in terms of clinical presentation, underlying etiologies, visual out-come, and neurological implications (i.e., risk of develop-ing multiple sclerosis [MS]). The Optic Neuritis Treatment Trial (ONTT) was a prospective clinical trial that enrolled 457 patients between the ages of 18 and 46 years with optic neuritis. The ONTT found that the presence of oligoclonal bands (OCB) was not helpful in predicting the future development of clinically definite multiple sclerosis in patients with an abnormal magnetic resonance imaging (MRI) (1). Heussinger et al report the results of a retrospective, multicenter (27 hospitals) study that included 357 children (,18 y of age) with optic neuritis to determine the prognos-tic indicators for the conversion to McDonald criteria MS. A sophisticated statistical analysis was performed using both simple and multiple Cox proportional hazard regressions, and 2 multiple regression analysis models (one model that used the MRI and OCB as binary covariates and the other model that used MRI and OCB as categorical variables) were generated. The regression analysis was done in bootstraps of 1000 repetitions (which I must admit I do not know much about) that does not make any statistical assumptions and allows for more precise confidence intervals and P values. Based on a median follow-up of 4 years, 40.6% of the patient cohort developed CDMS and 1.4% developed neuromyelitis optica (NMO). The univariate and multivar-iate analysis found higher age, abnormal MRI, and presence of OCB predictors for the future development of MS. The presence of an abnormal MRI and OCB had a higher predictive value than each factor alone. Gender and optic neuritis laterality were not predictive factors in the future development of MS. Finally, the authors provided the 4- year conversion risk of MS in all the patients (n = 291) who had a minimum of 4 years of follow-up. The table below summarizes those results. I was struck by the high conversion rate of pediatric optic neuritis to MS. In addition, even though OCB status is helpful in predicting the future development of MS, cranial MRI (cMRI) is the major driving force for predicting the conversion to MS (nearly 50% of patients with a positive cMRI and negative OCB developed MS compared to only 35% of patients with a negative cMRI and positive OCB). The ultimate goal of a study like this is to try to determine as accurately and quickly as possible those patients who will go on to develop MS and institute disease modifying drugs (DMD). I think it is important to understand that the treatment of a clinically isolated syndrome (i.e., optic neuritis) with a DMD is based on prospective clinical trials that enrolled patients with an abnormal cMRI. To my knowledge, there is no clinical trial assessing the value of initiating DMD therapy in a patient with clinically isolated syndrome and a normal cMRI. Therefore, I would not recommend DMD therapy in OCB Negative OCB Positive cMRI positive (150 patients) 49% (18/37 patients) 95% (107/113 patients) cMRI negative (141 patients) 8% (9/115 patients) 35% (9/26 patients) Moster and Bhatti: J Neuro-Ophthalmol 2015; 35: 323-328 325 Literature Commentary Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. a patient with optic neuritis, normal cMRI, and positive OCB. As a result, in most cases (unless I am concerned about an alternative diagnosis), I don't perform a lumbar puncture to determine the future risk of MS based on the OCB status in an adult patient with normal cMRI and I would argue to extend this strategy to children as well. A prospective clinical trial similar to the adult ONTT is needed to establish a level-1 evidence-based approach to the management of pediatric optic neuritis. It is my understand-ing that Dr Grant Liu at the University of Philadelphia and Dr Stacy Pineles at the University of California at Los Angeles are organizing a pilot study (pediatric optic neuritis treatment trial). I think that such a study is desperately needed. -M. Tariq Bhatti, MD There are a few issues that I have with this report. First, one of the reasons for the high rate of conversion to MS in this study compared to older studies is that the criteria for developing MS were McDonald criteria, allowing for the diagnosis of MS based on MRI demonstration of dissem-ination in time and space and not requiring a second clinical event. An additional possible source of bias may be that 52% of those who developed MS were treated with immunomodulatory agents when they initially presented with MS. In this nonstandardized, retrospective study, I have little doubt that those treated with MS medications were watched more carefully, had MRI scans more frequently, and likely had their MRI more carefully scrutinized by the pediatric neurologists who interpreted the scans (neuroradiologists were not used in this study). Another source of bias was the exclusion of any child followed for less than 2 years who did not develop MS but the inclusion of those who did go on to develop MS. Another surprising finding is that 69.7% of those develop-ing MS did so within 1 year. One wonders whether the MRI findings seen at follow-up really occurred at the time of the optic neuritis (ON). Despite these limitations, the data suggest that OCBs in the cerebrospinal fluid independently predict conversion to MS. With this in mind, Tariq, I'm surprised you would still not recommend lumbar puncture in children with ON. If these results are verified in future studies, it would make sense to check CSF in those children with ON with a negative MRI. -Mark L. Moster, MD REFERENCE 1. Cole SR, Beck RW, Moke PS, Kaufman DI, Tourtellotte WW. The predictive value of CSF oligoclonal banding for MS 5 years after optic neuritis. Optic Neuritis Study Group. Neurology. 1998;51:885-887. Martinez-Hernandez E, Sepulveda M, Rostásy K, Höftberger R, Graus F, Harvey RJ, Saiz A, Dalmau J. Antibodies to aquaporin 4, myelin-oligodendrocyte glycoprotein, and the glycine receptor a1 subunit in patients with isolated optic neuritis. JAMA Neurol. 2015;72:187-193. Importance: In patients with isolated optic neuritis (ON), the presence of antibodies to aquaporin 4 (AQP4) has diagnostic and prognostic value. In the same clinical setting, the significance of antibodies to myelin-oligodendrocyte glycopro-tein (MOG) or the glycine receptor a1 subunit (GlyR) is unclear. Objectives: To investigate the frequency of antibodies to AQP4, MOG, and GlyR in patients with unilateral or bilateral, severe, or recurrent isolated ON and to determine their clinical and prognostic correlates. Design, Setting, and Participants: Retrospective case-con-trol study from 11/1/2005, through 5/30/zs outside the optic nerves and 142 controls (30 healthy individuals, 48 patients with neuromyelitis optica, and 64 patients with multiple sclerosis). Main Outcomes and Measures: Clinicoimmunologic analy-sis. We determined the presence of antibodies to AQP4, MOG, and GlyR using cell-based assays. Results: The median age of the patients at the onset of ON symptoms was 28 (range, 5-65) years; 36 patients (71%) were female. Antibodies were identified in 23 patients (45%), including MOG in 10 patients, AQP4 in 6 patients, and GlyR in 7 patients (concurrent with MOG in 3 and con-current with AQP4 in 1). Patients with AQP4 antibodies (median visual score, 3.5 [range, 1-9]) had a worse visual outcome than patients with MOG antibodies alone (median visual score, 0 [range, 0-5]; P = 0.007), patients with seronegative findings (n = 28) (median visual score, 1.0 [range, 0-14]; P = 0.08), and patients with GlyR antibodies alone (n = 3) (median visual score, 0 [range, 0-2]; P = 0.10). The median age of the 7 patients with GlyR antibod-ies was 27 (range, 11-38) years; 5 (71%) of these were female. Among the 3 patients with GlyR antibodies alone, 1 patient had monophasic ON, 1 had recurrent isolated ON, and 1 had conversion to multiple sclerosis. The 3 patients with GlyR antibodies concurrent with MOG antibodies had recurrent isolated ON, and the patient with concurrent AQP4 antibodies had conversion to neuromyelitis optica. Of the 48 controls with neuromyelitis optica, 37 (77%) had AQP4 antibodies, 4 (8%) had MOG antibodies, 2 (4%) had AQP4 antibodies concurrent with MOG antibodies, and 5 (10%) were seronegative. Of the 64 controls with multiple sclero-sis, 5 (8%) had GlyR antibodies. Conclusions and Relevance: Forty-five percent of patients with unilateral or bilateral, severe, or recurrent isolated ON had antibodies to MOG, AQP4, or GlyR. Patients with AQP4 antibodies had the poorest visual outcomes, whereas patients with MOG antibodies had a better outcome that was similar to that of patients with seronegative findings. The significance of GlyR antibodies in the setting of ON is unclear and deserves further study. This study found that the presence of antibodies to AQP4, MOG, or GlyR in 45% of patients with isolated optic neuritis (ON), which was severe and/or bilateral and/or recurrent. These patients had no clinical evidence of neurologic disease elsewhere and did not have magnetic resonance imagings that were typical for either MS or NMO. Twenty percent had MOG antibodies, 12% AQP4, and 14% GlyR. As one might expect, those with AQP4 had 326 Moster and Bhatti: J Neuro-Ophthalmol 2015; 35: 323-328 Literature Commentary Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. a worse visual outcome. Those with MOG had a better outcome, which is similar to prior reports of patients with NMO presentations who were negative for AQP4 but positive for MOG. Additionally, those with MOG were younger and more often male than AQP4 patients. GlyR antibodies were considered to be a "bystander effect," since it was not associated with a change in prognosis and was most often seen with other antibodies as well. As clinicians, we have to identify early on who is at risk for poor outcome and early recurrence in order to begin immune suppression and to perform clinical trials with potentially neuroprotective agents. Along with other reports that help predict prognosis, such as poorer visual outcomes with early ganglion cell layer thinning on OCT, studying these antibodies should help determine which patients should receive early aggressive treatment. -Mark L. Moster, MD I think you would agree with me, Mark, that since the identification of the autoantibody associated with NMO, our ability to make the diagnosis has been greatly enhanced and our understanding of the underlying pathophysiology substantially advanced. In addition, more and more autoantibodies are being discovered that may or may not have a direct relationship to certain disease entities. The one comment I would make regarding this study is the fact that all the patients evaluated had severe and/or bilateral and/or recurrent optic neuritis. I personally don't order anti-AQP4 (NMO IgG) on every patient with optic neuritis, especially those with "typical" optic neuritis as defined by the Optic Neuritis Treatment Trial. -M. Tariq Bhatti, MD Siddiqui FM, Dandapat S, Banerjee C, Zuurbier SM, Johnson M, Stam J, Coutinho JM. Mechanical thrombectomy in cerebral venous thrombosis: systematic review of 185 cases. Stroke. 2015;46:1263-1268. Background and Purpose: Cerebral venous thrombosis is generally treated with anticoagulation. However, some patients do not respond to medical therapy, and these might benefit from mechanical thrombectomy. The aim of this study was to gain a better understanding of the efficacy and safety of mechanical thrombectomy in patients with cerebral venous thrombosis, by performing a systematic review of the literature. Methods: We identified studies published between January 1995 and February 2014 from PubMed and Ovid. We included all cases of cerebral venous thrombosis in whom mechanical thrombectomy was performed with or without intrasinus thrombolysis. Good outcome was defined as normal or mild neurological deficits at discharge (modified Rankin Scale, 0-2). Secondary outcome variables included periprocedural complications and recanalization rates. Results: Our study included 42 studies (185 patients). Sixty percent of patient had a pretreatment intracerebral hemor-rhage, and 47% were stuporous or comatose. AngioJet was the most commonly used device (40%). Intrasinus throm-bolysis was used in 131 patients (71%). Overall, 156 patients (84%) had a good outcome, and 22 (12%) died. Nine patients (5%) had no recanalization, 38 (21%) had partial, and 137 (74%) had near to complete recanalization. The major periprocedural complication was new or increased intracerebral hemorrhage (10%). The use of AngioJet was associated with lower rate of complete recanalization (odds ratio, 0.2; 95% confidence interval, 0.09-0.4) and lower chance of good outcome (odds ratio, 0.5; 95% confidence interval, 0.2-1.0). Conclusions: Our systematic review suggests that mechan-ical thrombectomy is reasonably safe, but controlled studies are required to provide a definitive answer on its efficacy and safety in patients with cerebral venous thrombosis. Although not very common, cerebral venous thrombosis (CVT) is an important condition for neuro-ophthalmologists to be aware of because it can mimic idiopathic intracranial hypertension. In addition, a miss or delay in diagnosis can result in a fatal outcome. The management of CVT is based on identifying the condition, determining the underlying etiology, and initiating anticoagulation therapy. Treatment in those cases that are nonresponsive to anticoagulation may require an invasive surgical or endovascular procedure. With the systematic review, it was concluded that mechanical thrombectomy (MT) (with or without intra-sinus thrombolysis) should be considered in a select group of patients with severe CVT not responsive to systemic anticoagulation. However, in those patients with impending transtentorial herniation, the authors recommended a de-compressive hemicraniotomy instead of MT. The authors acknowledge many limitations of their study and also point to the disadvantages of MT, including the high cost of the procedure, technical difficulty of performing the procedure, and inherent complications of the procedure. Just as a footnote, I have limited personal experience with MT based on a 23-year-old woman who initially presented with papilledema due to extensive CVT and then subsequently developed bilateral hemispheric venous infarctions. Thanks to our interventional neurosurgeon/neuroradiologist, the patient underwent MT with intrasinus thrombolysis, resulting in an amazing neurological outcome (i.e., no neurological deficits). -M. Tariq Bhatti, MD As noted by the authors, small case series tend to report their good outcomes and not the disasters. The authors were surprised that they found only 1 case reported of a vessel perforation. What we can conclude at this point is that in patients who are really in dire straits with CVT, in addition to intrasinus thrombolysis, MT can be considered as a last resort and may possibly improve outcome. -Mark L. Moster, MD Moster and Bhatti: J Neuro-Ophthalmol 2015; 35: 323-328 327 Literature Commentary Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Stein JD, Childers D, Gupta S, Talwar N, Nan B, Lee BJ, Smith TJ, Douglas R. Risk factors for developing thyroid-associated ophthalmopathy among individuals with Graves disease. JAMA Ophthalmol. 2015;133:290-296. Importance: Thyroid-associated ophthalmopathy (TAO) is a common and debilitating manifestation of Graves disease (GD). Presently, little is known about the factors that may increase the risk of developing TAO among patients with GD. Objective: To identify the risk factors associated with the development of TAO among individuals with newly diag-nosed GD. Design, Setting, and Participants: In this longitudinal cohort study, all beneficiaries 18 years of age or older with newly diagnosed GD who were continuously enrolled in a large, nationwide, US-managed care network and who visited an eye care professional 1 or more times from 2001 to 2009 were identified. International Classification of Dis-eases, Ninth Revision, Clinical Modification billing codes were used to identify those who developed manifestations of TAO. Multivariable Cox regression was used to determine the hazard of developing TAO among persons with newly diagnosed GD, with adjustment for sociodemographic fac-tors, systemic medical conditions, thyrotropin levels, and medical and surgical interventions for the management of hyperthyroidism. Main Outcomes and Measures: Manifestations of TAO measured by hazard ratios (HRs) with 95% confidence intervals (CIs). Results: Of 8404 patients with GD who met the inclusion criteria, 740 (8.8%) developed TAO (mean follow-up, 374 d since initial GD diagnosis). After adjustment for potential confounders, surgical thyroidectomy, alone or in combina-tion with medical therapy, was associated with a 74% decreased hazard for TAO (adjusted HR, 0.26; 95% CI, 0.12-0.51) compared with radioactive iodine therapy alone. Statin use (for $60 days in the past year vs ,60 d or nonuse) was associated with a 40% decreased hazard (adjusted HR, 0.60; 95% CI, 0.37-0.93). No significant association was found for the use of nonstatin cholesterol-lowering medications or cyclooxygenase 2 inhibitors and the development of TAO. Conclusions and Relevance: If prospective studies can confirm our finding that a thyroidectomy and statin use are associated with substantially reduced hazards for TAO among patients with GD, preventive measures for this burdensome manifestation of GD may become a reality. One of the most frustrating conditions we treat is TAO, related to the complexity of the symptoms and the limited efficacy of our treatments. Optimally, patients with GD should be treated in a way that would decrease the risk of developing TAO. This study found that 8.8% of patients who were diagnosed with GD and were followed for a mean of 374 days went on to develop TAO. Associated with a higher risk of TAO were treatment with radioactive iodine, elevated thyroid stimulating hormone, and an elevated thyroid stimulating immunoglobulin. Associated with a lower risk were treatment with thyroidectomy and use of statins. The study has the limitation of lack of clinical information-merely using coding from a billing database. For instance, is it possible that the risk of TAO may be modified by the lipid abnormalities rather than the treat-ment with a statin, but lipid levels and comorbidities were not assessed. Additionally, there is no information available on smoking status in this study. Interestingly, in addition to verifying our prior under-standing of radioactive iodine increasing the risk of TAO, this study provides evidence that thyroidectomy decreases the risk. Further study is warranted to determine if thyroidectomy, statin use, and the avoidance of posttreat-ment elevations of thyroid stimulating hormone will result in fewer patients developing TAO. -Mark L. Moster, MD This is a very interesting report that emphasizes the need for more research regarding the risk factors that influence the development or worsening of TAO in patients with GD. The authors found that surgical thyroidectomy was associated with a decreased risk of developing thyroid eye disease. The literature contains a mix of results (no change, improvement, or worsening) regarding the influence of surgical thyroidectomy and its influence on TAO. Prior to this study, a meta-analysis of 3 prospective trials comparing total vs subtotal thyroidec-tomy found no difference between these 2 procedures in the development or worsening of TAO (1). This study confirms what has already been well established in the literature, that radioactive iodine is associated with the progression of TAO. However, what I found intriguing was the decreased risk of TAO with statins, which has not been well studied to date. I was surprised to see that COX-2 inhibitor use had no effect on TAO. Mark, you had mentioned the limitation of the study and I think it is worth emphasizing that the status of smoking and over-the-counter medications were not as-sessed, which I think are 2 very important factors that need to be analyzed. Finally, it is important to appreciate the fact that most patients with GD do not develop TAO! -M. Tariq Bhatti, MD REFERENCE 1. Guo Z, Yu P, Liu Z, Si Y, Jin M. Total thyroidectomy vs bilateral subtotal thyroidectomy in patients with Graves' diseases: a meta-analysis of randomized clinical trials. Clin Endocrinol (Oxf). 2013;79:739-746. 328 Moster and Bhatti: J Neuro-Ophthalmol 2015; 35: 323-328 Literature Commentary Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited.