Table of Contents

Literature Commentary

Literature Commentary Stein MS, Liu Y, Gray OM, Baker JE, Kolbe SC, Ditchfield MR, Egan GF, Mitchell PJ, Harrison LC, Butzkueven H, Kilpatrick TJ. A randomized trial of high-dose vitamin D2 in relapsing-remitting multiple sclerosis. Neurology. 2011;77:1611-1618. Objective: Higher latitude, lower ultraviolet exposure, and lower serum 25-hydroxyvitamin D (25OHD) correlate with higher multiple sclerosis (MS) prevalence, relapse rate, and mortality. We therefore evaluated the effects of high-dose vitamin D2 (D2) in MS. Methods: Adults with clinically active relapsing-remitting MS (RRMS) were randomized to 6-month, double-blind, placebo-controlled, high-dose D2, 6,000 IU capsules, dose-adjusted empirically aiming for a serum 25OHD of 130-175 nM. All received daily low-dose (1,000 IU) D2 to prevent deficiency. Brain MRIs were performed at baseline, 4, 5, and 6 months. Primary end points were the cumulative number of new gadolinium-enhancing lesions and change in the total vol-ume of T2 lesions. Secondary end points were Expanded Disability Status Scale (EDSS) score and relapses. Results: Twenty-three people were randomized, of whom 19 were on established interferon or glatiramer acetate (Copaxone) treatment. Median 25OHD dose ranged from 54 to 69 nM (low-dose D2) vs 59 to 120 nM (high-dose D2) (P = 0.002). No significant treatment differences were detected in the primary MRI end points. Exit EDSS, after adjustment for entry EDSS, was higher following high-dose D2 than following low-dose D2 (P = 0.05). There were 4 relapses with high-dose D2 vs none with low-dose D2 (P = 0.04). Conclusions: We did not find a therapeutic advantage in RRMS for high-dose D2 over low-dose D2 supplementation. Classification of Evidence: This study provides class I evidence that high-dose D2 (targeting 25OHD 130-175 nM), compared to low-dose supplementation (1,000 IU/day) was not effective in reducing MRI lesions in patients with RRMS. The authors randomized patients with RRMS to receive either 6,000 IU or 1,000 IU of vitamin D2 (D2) twice daily. They did not find a significant difference in gadolinium-enhancing lesions or T2 lesion volume on brain MRI. Patients on high-dose D2 had worse EDSS and more relapses than those on low-dose D2. This is the first study to randomize patients with RRMS to receive vitamin D2. Although vitamin D deficiency may be associated with RRMS, this study calls into question exactly how much vitamin D supplementation is safe and effective. -Michael S. Lee, MD Although decreased levels of vitamin D have been found in MS patients and associated with severity, we have yet to see a good clinical trial that shows benefit of treatment with vitamin D. The results of this trial are disappointing in that treatment failed to show benefit on MRI or clinical relapses. However, the trial was only 6-month long, and benefit might take longer to occur. -Mark L. Moster, MD One more thing, this study took place in Australia, which may receive more sun than Minnesota. It remains to be seen whether a high-dose vitamin D supplement may be more effective in areas that receive less sun exposure. -Michael S. Lee, MD Tradtrantip L, Zhang H, Saadoun S, Phuan PW, Lam C, Papadopoulos MC, Bennett JL, Verkman AS. Anti-aquaporin-4 monoclonal antibody blocker ther-apy for neuromyelitis optica. Ann Neurol. 2011. doi:10.1002/ana.22657. Objective: Neuromyelitis optica (NMO) is an inflammatory demyelinating disease of the central nervous system. Circulating autoantibodies (NMO-IgG) against astrocyte water channel aquaporin-4 (AQP4) cause complement-mediated and cell-mediated astrocyte damage with conse-quent neuroinflammation and demyelination. Current NMO therapies, which have limited efficacy, include immunosup-pression and plasma exchange. The objective of this study was to develop a potential new NMO therapy based on blocking of pathogenic NMO-IgG to its target, AQP4. Methods: We generated nonpathogenic recombinant mono-clonal anti-AQP4 antibodies that selectively block NMO-IgG binding to AQP4. These antibodies comprise a tight-binding anti-AQP4 Fab and a mutated Fc that lacks functionality for complement-mediated and cell-mediated cytotoxicity. The efficacy of the blocking antibodies was studied using cell culture, spinal cord slice, and in vivo mouse models of NMO. Results: In AQP4-expressing cell cultures, the nonpatho-genic competing antibodies blocked binding of NMO-IgG in human sera, reducing to near zero complement-mediated and cell-mediated cytotoxicity. The antibodies prevented the development of NMO lesions in an ex vivo spinal cord slice model of NMO and in an in vivo mouse model, without causing cytotoxicity. Interpretation: Our results provide proof-of-concept for therapy of NMO with blocking antibodies. The broad efficacy of antibody inhibition is likely due to steric competition because of its large physical size compared to AQP4. Blocker therapy to prevent binding of pathogenic autoanti-bodies to their targets may be useful for the treatment of other autoimmune diseases as well. Prior studies have demonstrated the importance of NMO-IgG in the pathogenesis of NMO. This includes the 86 Moster and Lee: J Neuro-Ophthalmol 2012; 32: 86-89 Literature Commentary Section Editors: Mark L. Moster, MD Michael S. Lee, MD Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. correlation of disease activity with antibody level, the improvement with removal of antibody, and the production of similar pathology in mice or rats administered human NMO-IgG. The authors have created a modified NMO-IgG, which binds to AQP4, is not pathogenic, and does not induce complement activation or cytotoxicity. It blocks the binding of the pathogenic NMO-IgG. These early findings provide hope for a specific targeted therapy with blocking antibodies for NMO and other diseases without the dangerous risks of immunosuppression, normally required in these patients. -Mark L. Moster, MD What a great idea! The targeted therapy concept works well for NMO since the pathophysiology for some patients likely involves a single autoantibody to AQP4. Hopefully, further preclinical (and clinical) studies will continue to show minimal cytotoxic injury with very few, if any, adverse events. -Michael S. Lee, MD Ponto KA, Kanitz M, Olivo PD, Pitz S, Pfeiffer N, Kahaly GJ. Clinical relevance of thyroid-stimulating immunoglobulins in Graves' ophthalmopathy. Ophthalmology. 2011;118:2279-2285. Purpose: Thyroid-stimulating immunoglobulins (TSIs) likely mediate Graves ophthalmopathy (GO). The clinical rele-vance of these functional autoantibodies was assessed in GO. Design: Cross-sectional trial. Participants: A total of 108 untreated patients with GO. Methods: TSIs, assessed with a novel bioassay, bind to the thyrotropin receptor (TSHR) and transmit signals for cyclic adenosine monophosphate (cAMP)-dependent activation of luciferase gene expression. The cAMP/cAMP response ele-ment- binding protein/cAMP-regulatory element complex induces luciferase that is quantified after cell lysis. The TSI levels were correlated with activity and severity of GO and compared with a TSHR-binding inhibitory immunoglobu-lin (TBII) assay. Main Outcome Measures: TSIs, activity and severity of GO, diplopia, and TBII. Results: TSIs were detected in 106 of 108 patients (98%) with GO. All 53 hyperthyroid patients were TSI positive versus 47 patients (89%) who were TBII positive. All 69 patients with active GO were TSI positive, whereas only 58 of 69 patients (84%) were TBII positive. TSIs correlated with the activity (r = 0.83; P = 0.001) and severity (r = 0.81; P = 0.001) of GO. All 59 patients with GO with dip-lopia were TSI positive, and 50 of 59 patients (85%) were TBII positive. Among patients with moderate-to-severe and mild GO, 75 of 75 (100%) and 31 of 33 (94%) were TSI positive compared with TBII positivity in 63 of 75 (84%) and 24 of 33 (73%), respectively. The TSI levels were higher in moderate-to-severe versus mild GO (489% to 137% vs 251% to 100%; P = 0.001). Chemosis and GO activity predicted TSI levels alone (P = 0.001, multivariable analysis). The TSI levels were higher in patients with che-mosis (527% ± 131%) than in patients without chemosis (313% ± 127%; P , 0.001). Conclusions: TSIs show more significant association with clinical features of GO than of TBII and may be regarded as functional biomarkers for GO. In this study, the authors measured thyroid-stimulating immunoglobulin (TSI) among various patients with thyroid eye disease (TED). They compared the TSI level with the TED clinical activity and severity and found a strong correlation. Comparison of serial TSI levels to serial clinical activity and severity was not performed. Since much of the reconstructive procedures in TED occur during the inactive burned out phase of TED, it would be nice to have a biomarker for this. Although TSI looks promising, I think we need to see serial TSI levels to know for sure. -Michael S. Lee, MD The most impressive finding in this study is how sensitive TSI is in patients with GO, even in those with mild disease. This goes along with how I use TSI in my practice, which is to help diagnose GO in patients with subtle findings. This study used a new, novel, cell-based bioassay for TSI. What we do not know is how sensitive most commercially available assays are. -Mark L. Moster, MD Stein JD, Kim DS, Mundy KM, Talwar N, Nan B, Chervin RD, Musch DC. The association between glaucomatous and other causes of optic neuropathy and sleep apnea. Am J Ophthalmol. 2011;152: 989-998.e3. Purpose: To determine whether an association exists between sleep apnea and open-angle glaucoma, normal-tension glaucoma, nonarteritic ischemic optic neuropathy (NAION), papilledema, or idiopathic intracranial hyperten-sion (IIH) and whether treatment with continuous positive airway pressure affects the development of these condi-tions. Design: Retrospective, longitudinal, cohort study. Methods: Billing records for beneficiaries of 40 years and older enrolled in a large United States-managed care net-work from 2001 through 2007 were reviewed. Incidence of open-angle glaucoma, normal-tension glaucoma, NAION, papilledema, and IIH were determined for the beneficiaries and were stratified by sleep apnea status. Cox regression analyses determined the hazard of each of these conditions developing among individuals with and without sleep apnea, with adjustment for sociodemographic, ocular, and medical conditions. Results: Among the 2,259,061 individuals in the study, 156,336 (6.9%) had 1 or more sleep apnea diagnoses. The hazard of open-angle glaucoma was no different among Moster and Lee: J Neuro-Ophthalmol 2012; 32: 86-89 87 Literature Commentary Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. persons with sleep apnea either treated [adjusted hazard ratio (HR), 0.99; 95% confidence interval (CI), 0.82-1.18) or untreated with continuous positive airway pressure (HR, 1.01; 95% CI, 0.98-1.05) and individuals without sleep apnea. Similar findings were observed when assessing the hazard ratio of developing normal-tension glaucoma (P . 0.05 for both comparisons). A significantly increased hazard ratio of developing NAION (HR, 1.16; 95% CI, 1.01- 1.33) and IIH (HR, 2.03; 95% CI, 1.65-2.49) was observed among individuals with sleep apnea who were not receiving continuous positive airway pressure therapy as compared with individuals without sleep apnea, although similar in-creased risks could not be demonstrated among continuous positive airway pressure-treated sleep apnea patients for these conditions (P . 0.05 for both comparisons). Conclusions: Patients with untreated sleep apnea are at an increased risk for IIH and NAION. Clinicians should consider appropriate screening for these conditions in sleep apnea patients. Are patients with sleep apnea more likely to have NAION, glaucoma, or increased intracranial pressure (ICP) than those without sleep apnea? Does it matter if they are treated with continuous positive airway pressure (CPAP)? This study provides further evidence of an association of sleep apnea with NAION and elevated ICP. The nature of the study has some strengths and weakness. The basis of diagnoses using billing records eliminates any observer bias regarding the diagnosis of sleep apnea or type of optic neuropathy. The study also benefits from a large population, with more than 2 million eligible patients after excluding appropriate patients (e.g., those with an optic neuropathy diagnosis prior to the diagnosis of sleep apnea.) Of course, the nature of the study has limitations. Not having any clinical information about the optic neuropathy is a problem. Additionally, the patients who had billing codes for IIH and papilledema (looked at separately in this study) may have much overlap and may inadvertently include those with an alternative cause of elevated ICP, such as a mass lesion. The bottom line findings are helpful clinically. There was no association of sleep apnea with glaucoma, including normal-tension glaucoma. This corroborates some prior studies and conflicts with others. There is a small increased risk of NAION (HR, 1.16; 95% CI, 1.01-1.33) in untreated sleep apnea patients. For IIH, there was an increase in HR (2.03; 95% CI, 1.65- 2.49) for those with untreated sleep apnea. In contrast, for those with a diagnosis code of papilledema, the HR was increased with untreated sleep apnea (adjusted HR, 1.29; 95% CI, 1.10-1.50) as well as those on those on CPAP (adjusted HR, 2.05; 95% CI, 1.19-3.56). This study lends further support to considering sleep studies in our patients with NAION and elevated ICP. -Mark L. Moster, MD I think it is reasonable to ask patients with IIH and NAION about apnea symptoms and to consider sleep studies among suspicious patients. I would not perform sleep studies for everyone with IIH and NAION. The article reports that the risk of NAION and IIH increased with apnea (not using CPAP) over nonapnea patients but that the risk remained the same between apnea (using CPAP) compared to nonap-nea patients. Disappointingly, the authors did not compare NAION/IIH for apnea patients using CPAP vs apnea patients not using CPAP. Therefore, we do not know whether using CPAP clearly reduces the risk of developing IIH or NAION. The authors conclude by suggesting that clinicians screen patients with apnea for IIH. I do not think this is wise or cost effective given that the incidence of IIH in apnea patients in this study was 4 per 10,000. We would need to screen approximately 2,500 patients with apnea to find 1 patient with IIH. Finally, this study showed that incidence of NAION was 5-7 per 10,000 persons, which is consistent with another large Medicare database study. This incidence is much higher than the often quoted incidences of 2-10 per 100,000. -Michael S. Lee, MD Lee MW, Vedanarayanan VV. Cerebrospinal fluid opening pressure in children: experience in a con-trolled setting. Pediatr Neurol. 2011;45:238-240. Abstract: The literature on cerebrospinal fluid opening pressures for pediatric patients is scant. A retrospective study of measured opening pressures during lumbar punc-tures of pediatric patients in a controlled uniform setting was conducted. These procedures were performed in an out-patient surgery setting under anesthesia. Patients' end-tidal carbon dioxide levels were maintained at 40-45 mm Hg. Opening pressures were measured with the patients lying in left lateral decubitus position with their legs extended. Correlations with patients' ages and body mass index per-centiles were investigated. Forty-four patients (median age, 8.9 years; range, 1.1-16.8 years) were included. Patients with chronic headaches, papilledema, or severe abnormali-ties of cerebrospinal fluid were excluded. The mean opening pressure recorded was 20.3 cm H2O (median, 21 cm H2O; range, 6-36 cm H2O). Poor correlation with both age and body mass index percentiles was evident. The correlation coefficients were 0.09 and 0.14, respectively. Our experi-ence suggests that pediatric reference ranges for opening pressures are closer to those of adults than previously ap-preciated, and values above 20 cm H2O should not neces-sarily be considered abnormal. In August of 2010, a research letter in the New England Journal of Medicine (NEJM) described opening pressures (OP) among pediatric patients undergoing lumbar puncture at all levels of sedation. They found a higher range than previously believed. The current study by Lee (no relation) and Vedanarayanan (definitely no relation) performed a similar study, but all patients underwent general anesthe-sia. The advantage is that they could manage carbon dioxide 88 Moster and Lee: J Neuro-Ophthalmol 2012; 32: 86-89 Literature Commentary Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. levels, which may affect OP. They found the mean OP of 20.3 ± 7 cm H2O. The NEJM study data can be down-loaded, and it found a mean OP of 19.8 ± 6.8 cm H2O. This numbers are almost identical. Now, the authors looked at the 90th percentile, but as we all know, the 95th percen-tile is also used as a cutoff for abnormal. So could an OP of 34 be within that normal range too? Previously, I struggled to know what to do with an OP of 26 cm H2O in a patient who I was convinced had pseudopapilledema. This infor-mation may also throw into question the diagnosis of pseu-dotumor without papilledema. How often does it really occur vs are these simply normal OP in patients with headaches? -Michael S. Lee, MD This study provides some new information about opening pressure in children. However, it provides only limited information of use to the clinician. Although the patients did not have conditions know to raise intracranial pressure, they all had neurologic diseases, including myop-athy, neuropathy, ataxia, seizures, stroke, demyelinating disease, and encephalopathy. I do not think we can be certain that we can consider this population as a normal standard for intracranial pressure. Additionally, we do not really know how being under general anesthesia affects OP, which makes this information less useful for patients having lumbar puncture while awake. -Mark L. Moster, MD Omar R, Rohrer JD, Hailstone JC, Warren JD. Struc-tural neuroanatomy of face processing in fronto-temporal lobar degeneration. J Neurol Neurosurg Psychiatry. 2011;82:1341-1343. Abstract: Impairments of face processing occur frequently in frontotemporal lobar degeneration (FTLD) but the neuro-anatomical basis for these deficits has seldom been studied systematically. Here, a prospective voxel-based morphometry study is described addressing the neuroanat-omy of 2 key dimensions of face processing-face identifi-cation and facial emotion recognition-in a single cohort of 32 patients with FTLD (19 with frontal variant and 13 with temporal variant FTLD). For the FTLD group as a whole, face identification was positively associated with gray matter in the right anterior fusiform gyrus while recognition of angry expressions was positively associated with gray matter in the bilateral insular cortex. FTLD provides a perspective on the neuroanatomy of face processing that is complementary to focal lesion and normal functional imaging work. Traditionally, the elucidation of the functions of specific brain regions has been accomplished via the study of patients with focal brain lesions. More recently, functional imaging of normal subjects has proved helpful. It makes sense that patients who have focal neurodegenerative disease may also provide insights into the neuroanatomic basis of individual functions. FTLD (traditionally known as Pick disease) is an excellent group of disorders to study because of the typi-cal focality and asymmetry, which presents with specific deficits related to unilateral or bilateral frontal or temporal degeneration. For instance, those with a predominantly left temporal FTLD have impaired naming, word comprehen-sion, and semantics. Those with right temporal FTLD are socially inept and unempathic, with occasional sociopathic behavior. Some with early left FTLD have unmasked their "right brain" and have become quite artistic. Many patients have difficulty with recognizing faces, and others have dif-ficulty appreciating facial emotion, which underlies this study. The authors looked at brain volumes in focal regions and correlated that with the ability to recognize famous faces and facial emotion. They found that face identification was associated with amount of gray matter in the right anterior fusiform gyrus, while recognition of angry expressions was associated with gray matter in the bilateral insular cortex. These findings explain some of the social abnormalities in FTLD patients and mostly agree with prior findings from lesion and functional imaging work. Neurodegeneration is never completely focal, and other brain areas are involved. Nonetheless, the preferential involvement of certain areas over others allows much to be learned about normal brain function from patients with diseases such as FTLD. -Mark L. Moster, MD Well said, Mark. -Michael S. Lee, MD Moster and Lee: J Neuro-Ophthalmol 2012; 32: 86-89 89 Literature Commentary Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited.