| OCR Text |
Show ORIGINAL CONTRIBUTION Retinol- Binding Protein and Retinol Analysis in Cerebrospinal Fluid and Serum of Patients With and Without Idiopathic Intracranial Hypertension Judith E. A. Warner, MD, Alexander J. Larson, BA, Prakash Bhosale, PhD, Kathleen B. Digre, MD, Courtney Henley, MD, Stephen C. Alder, PhD, Bradley J. Katz, MD, PhD, and Paul S. Bernstein, MD, PhD Background: Several studies have implicated vitamin A- related compounds in the pathogenesis of idiopathic intracranial hypertension ( IIH). The goal of this study was to compare cerebrospinal fluid ( CSF) and serum concentrations of retinol and retinol- binding protein ( RBP) in subjects with and without IIH. Methods: CSF and serum samples were collected from 87 subjects. The study population was composed of subjects with IIH ( IIH group, n = 28), subjects with non- IIH neurologic conditions ( neurology controls, n = 42), and subjects undergoing preoperative lumbar puncture but with no known neurologic conditions ( anesthesia controls, n = 17). RBP levels ( nM) were determined using radial immunodiffusion, and retinol levels ( nM) were determined using high- performance liquid chromatography Results: The retinol/ RBP ratio was greater in CSF than in serum, especially in subjects with IIH. Conclusions: The finding of increased levels of unbound retinol in the CSF of subjects with IIH provides further evidence that vitamin A may be involved in the pathogenesis of IIH. Comparative statistical analyses revealed multivariate relationships that demonstrate the need to further investigate correlations between vitamin A and RBP levels in CSF and serum. (/ Neuro- Ophthalmol 2007; 27: 258- 262) I diopathic intracranial hypertension ( IIH), also known as pseudotumor cerebri, is a condition of increased intracranial pressure ( ICP) with normal cerebrospinal fluid ( CSF) composition. IIH is generally accompanied by headache and papilledema, which can cause progressive vision loss. Initially described in the 19th century ( 1), the underlying cause of IIH remains unknown. Careful clinical evaluation, including appropriate imaging studies, is used to eliminate other causes for the increased pressure and associated signs and symptoms. IIH is usually observed in otherwise healthy obese women in their childbearing years ( 2,3). Intracranial hypertension associated with elevated levels of the essential nutrient vitamin A was first reported in 1954 and evaluated in detail in 1970 ( 4,5). Reports from dermatologic, oncologic, and other clinical settings describe intracranial hypertension associated with vitamin A intoxication ( 6- 9). In American diets, approximately 75% of vitamin A is derived from retinol and 25% from provitamin A car-otenoids, particularly ( 3- carotene ( 10). Dietary vitamin A is absorbed primarily as long- chain fatty acid retinyl esters that are stored in the liver and hydrolyzed in the lumen of the small intestine to produce retinol. Normally, retinol is bound to the plasma carrier retinol- binding protein ( RBP) in a 1: 1 ratio for transport through the circulation. Human RBP has a molecular mass of 21 kDa and a single binding site for retinol; it appears to mediate retinol entrance into cells ( 6). When not bound to RBP, vitamin A is toxic to cell membranes ( 11). A retinol/ RBP ratio of > 1 suggests the presence of free, unbound retinol, a potential toxin. Departments of Ophthalmology and Visual Sciences ( JEAW, PB, BJK, KBD, PSB, AJL), Neurology ( JEAW, BJK, KBD), Anesthesiology ( CH), and Family and Preventive Medicine ( SCA), University of Utah Health Sciences Center, Salt Lake City, Utah. This study was supported in part by an unrestricted grant from Research to Prevent Blindness, Inc., New York, NY, to the Department of Ophthalmology and Visual Sciences, University of Utah. P. S. B. is a Sybil B. Harrington RPB Research Scholar. Mr. Larson is currently at The Ohio State University College of Medicine, Columbus, Ohio. Dr. Henley is currently at LDS Hospital, Salt Lake City, Utah. Address correspondence to Judith E. A. Warner, MD, John A. Moran Eye Center, University of Utah Health Sciences Center, 65 Medical Drive, Salt Lake City, UT 84132; E- mail: judith. warner@ hsc. utah. edu. 258 J Neuro- Ophthalmol, Vol. 27, No. 4, 2007 Retinol and RBP in IIH J Neuro- Ophthalmol, Vol. 27, No. 4, 2007 RBP and transthyretin have been localized to the choroid plexus, a key site for delivery of vitamin A and other nutrients to the CSF ( 12- 15). Little is known, however, about the normal relationships between retinol and RBP in CSF. One study showed elevated serum RBP in some individuals with IIH ( 16). Previous studies documented elevated serum ( 17) and CSF ( 17- 19) retinol levels in some subjects with IIH. Excess CSF retinol may be toxic to arachnoid villi and inhibit CSF resorption, leading to increased ICR However, the absolute values of retinol in any setting are not useful without correlation with RBP. One needs to know the extent to which RBP is saturated by retinol to conclude that free retinol is present. No study has yet compared CSF and serum concentrations of both retinol and RBP in individuals with and without IIH. We investigated possible correlations between retinol and RBP levels in subjects with IIH, in subjects with other neurologic conditions, and in subjects without known neurologic pathologic conditions. We postulated that the relationship between serum and CSF retinol and RBP values could provide further insight into their possible role in the etiology of IIH. METHODS Subjects The University of Utah Institutional Review Board ( IRB) approved this project. Written informed consent was obtained from all study subjects before participation. Investigators obtained 1 mL of CSF and 2 mL of whole blood from subjects with IIH, as well as from subjects who either had other neurologic diseases ( with or without increased ICP) or who had undergone lumbar puncture ( LP) to exclude a diagnosis of IIH (" neurology controls"). Neurology controls did not have IIH based on normal ICP and fundus examination and, frequently, absence of signs of IIH. All LPs on these subjects were performed by the same person ( J. W), with pressure carefully measured when subjects were in the lateral decubitus position with legs extended and head in a neutral position immediately after CSF flow was established. Subjects with IIH and neurology controls, all of whom were consecutively recruited during ophthalmology or neurology clinic visits at the University of Utah Health Sciences Center, had complete documentation of their medical history and current medications. All subjects with IIH met modified Dandy diagnostic criteria for IIH ( 20). Anesthesiology department personnel obtained study- specific written informed consent from the control subjects with no known neurologic diseases (" anesthesia controls") before scheduled surgery in the Same- Day Surgery Unit of the University of Utah Health Sciences Center. Anesthesia controls had already consented to receive preoperative spinal anesthesia. Most of the LPs in this group were performed by the same person ( C. H.), with pressure measured in the lateral decubitus position. All analyses were performed on new subject specimens. Specimens used in prior studies at the Moran Eye Center ( 18) were not included in this study. Sample Preparation Serum for all samples was extracted from whole blood by centrifugation at 2,500 rpm for 5 minutes. CSF and serum samples were protected from light exposure, which could have altered the retinol concentration. All samples were stored at - 76° C until analysis. RBP Measurement RBP levels were measured using a NANORID radial immunodiffusion ( RID) kit ( The Binding Site Limited, Birmingham, England), which was calibrated for serum with a 1: 20 [ serum- 7% bovine serum albumin ( BSA)] dilution. CSF was concentrated 40X and applied in 5 | JLL quantities to immunodiffusion plate wells using a micropi-pette. After sample application, the lid of the RID plate was tightly closed, and the plate was incubated at room temperature ( 20- 24° C) in an airtight container for 96 hours. Incubation involved RBP ( antigen) migrating from a cylindrical well through an agarose gel containing antibody to RBP. Antigen- antibody complexes formed a precipitin ring; concentration was determined by ring diameter measurement. Retinol Measurement For protein precipitation 250 | JLL CSF was treated with 250 ( JLL methanol. Retinol was then extracted using 250 | JLL hexane containing 0.05% butylated hydroxyto-luene ( BHT) ( w/ v). Single step extraction was sufficient for 90% extraction of retinol; the extraction was then repeated to obtain a 500 ( JLL sample. The retinol level in each CSF sample was measured by high- performance liquid chromatography ( HPLC), using a DYNAMAX- 60A silica column ( Ranin Instrument, Emeryville, CA). Samples were eluted isocratically at a flow rate of 1.0 mL/ min with a mixture of 10% 1,4- dioxane ( v/ v) in hexane on a Waters HPLC system ( Waters Corporation, Milford, MA) with single wavelength detection at 325 nm. Additional details concerning this methodology have been published elsewhere ( 10,18). For serum retinol analysis, the HPLC equipment ( Thermo Electron, San Jose, CA) had an autosampler, two-channel solvent degasser, binary gradient pump, and UV-visible photodiode array detector. The vitamin A peak on each system's chromatogram was quantified by external standardization against synthetic all- trans- retinol. Peak identities were confirmed by photodiode array ( PDA) spectra 259 J Neuro- Ophthalmol, Vol. 27, No. 4, 2007 Warner et al and by coelution with authentic standards. Retinyl esters in CSF were never present in greater than trace amounts. Statistical Methods Data were entered into Microsoft Excel 2003 ( Microsoft, Seattle, WA) and imported into SPSS 14.0 for Windows ( SPSS Inc., Chicago, IL) for analysis. Comparisons were made among the IIH, neurology control, and anesthesia control groups using independent samples t tests, x2 tests, and one- way analysis of variance. Serum and CSF RBP and retinol levels and the ratio of retinol ( nM) to RBP ( nM) were compared among groups using one- way analysis of variance. Where parametric assumptions were violated, Wilcoxon rank- sum and Kruskal- Wallis tests were used. Analysis of covariance, using the general linear model, was used to adjust, respectively, for age, sex, body mass index ( BMI), opening pressure ( OP), and use of vitamins, thyroid medication, and estrogens. Multiple linear regression was used to estimate the association between age and total protein, controlling for IIH status. For all analyses, a was set at 0.05. Previous studies from the authors and others provided normal reference ranges for serum and CSF retinol and RBP ( 16- 18,21- 24). Statistical analyses were performed by a biostatistician ( S. C. A.). RESULTS Table 1 summarizes the demographics of our three subject groups. There were more women in the IIH group, and the mean age of the IIH group was less than that of the other two groups. These demographics are consistent with the fact that the majority of patients with IIH are premenopausal women. The mean BMI of the IIH group was higher than mean BMI in the other two groups. The mean OP of the IIH group, 328.6, exceeded the 250 minimum value required for diagnosis of IIH according to Modified Dandy Criteria ( 3). Four subjects with IIH had normal ICP recorded in this study but had had a prior diagnostic LP showing elevated ICP. Some neurology control patients had elevated ICP due to conditions such as venous thrombosis, renal failure, and parathyroid disorders. Data comparing RBP, retinol, and the retinol/ RBP ratio in serum and CSF of subjects in the three groups are presented in Table 2. IIH subjects had higher serum and CSF retinol levels than did neurology controls or anesthesia controls ( P = 0.001 and P = 0.05, respectively). Conversely, CSF RBP levels were lowest in the IIH group ( P < 0.001). Serum RBP levels were highest in the IIH subjects, but this difference was not statistically significant ( P = 0.07). The serum retinol/ RBP ratio ranged from approximately 0.40 to 0.55, whereas the CSF retinol/ RBP ratio was > 1.0 in all three groups. In both serum ( P = 0.05) and CSF ( P = 0.001), IIH subjects had the highest retinol/ RBP ratio. We analyzed our data for differences in retinol and RBP in subjects with newly diagnosed IIH vs previously treated subjects with IIH and found no significant differences. One study subject in the IIH group was removed from the CSF retinol and CSF retinol/ RBP ratio analyses because of extreme values, but results remained statistically significant. Because of limitations imposed by the IRB, CSF protein and cell counts were not measured in the anesthesia controls. Consequently, comparisons using these data were not possible. However, in comparing IIH subjects with neurology controls, CSF total protein increased with age [( 3 = 0.49 ( slope of curve), P = 0.001] and showed a trend toward being lower in the IIH group [( 3 = 8.69 ( difference between the two groups), P = 0.07]. This is consistent with other research findings. We also evaluated the potential confounding factor of CSF contamination with red blood cells ( RBC). We found no significant association between CSF RBC and retinol levels ( data not shown). Our results maintained statistical significance even if the IIH subject with markedly elevated TABLE 1. Demographics of subjects with idiopathic intracranial pressure ( IIH), non- IIH Controls), and undergoing lumbar puncture for surgical procedures IIH ( n = 28) Age, mean ( SD) 32.1 ( 10.2) Female, f (%) 26 ( 92.9) Body mass index, mean ( SD)* 36.5 ( 8.7) Opening pressure ( mm CSF), mean ( SD) 328.6 ( 86.5) Vitamins, f (%) 3( 10.7) Thyroid medications, f (%) 2 ( 7.1) Estrogens, f (%) 6( 21.4) " Calculated by the formula [ weight in pounds X ( 39.37 )] / [ 2.2 X ( hei tn = 5. ' ( Anesthesia Controls) Neurology controls ( n = 42) 44.8 ( 16.4) 22 ( 52.4) 29.0 ( 5.7) 189.4 ( 62.7) 14 ( 33.3) 1 ( 2.4) 3 ( 7.1) ght in inches )]. neurologic conditions Anesthesia controls ( n = 17) 59.5 ( 19.7) 9 ( 52.9) 28.3 ( 4.2) 192.8 ( 77.9) t 6 ( 35.3) 1 ( 5- 9) 1 ( 5- 9) ( Neurology P < 0.001 0.001 < 0.001 < 0.001 0.07 0.62 0.13 260 © 2007 Lippincott Williams & Wilkins Retinol and RBP in IIH J Neuro- Ophthalmol, Vol. 27, No. 4, 2007 TABLE 2. Comparison of levels of retinol- binding protein fluid ( CSF) in three subject groups Serum RBP ( nM) Serum retinol ( nM) Serum retinol/ RBP ratio CSF RBP ( nM) CSF retinol ( nM) CSF retinol/ RBP ratio Values are mean ( SD). * n = 40. tn = 16. in = 39. § n = 26. IKruskal- Wallis test. » n = 27. IIH ( n = 28) 3668.3 ( 1599.9) 1834.9 ( 768.0) 0.54 ( 0.19) 6.9 ( 4.4) 22.0 ( 60.3) § 2.89 ( 6.92) 11 ( RBP), retinol, and retinol/ RBP ratios in serum Neurology controls ( n = 42) 3465.7 ( 1710.5)* 1662.8 ( 846.0)* 0.49 ( 0.19)$ 12.4 ( 6.4) 12.4 ( 7.0)$ 1.14 ( 1.27)* Anesthesia controls ( n = 17) 2545.6 ( 817.1) t 972.5 ( 380.1) 0.39 ( 0.14) t 8.9 ( 3.3) t 15.3 ( 5.7) t 1.84 ( 0.76) t and cerebrospinal P 0.07 0.001 0.05 < 0.001 0.05^ 1 o. ooiu retinol was removed from analysis. This subject's spinal fluid was initially bloody and then cleared during the LP, with the aliquot used for study being the last tube collected. When comparisons across study subject groups were controlled for BMI, serum RBP ( P - 0.050), serum retinol ( P = 0.001), CSF RBP ( P = 0.001), and the CSF retinol/ RBP ratio ( P - 0.032) were significantly different, whereas the serum retinol/ RBP ratio ( P = 0.090) and CSF retinol ( P - 0.195) were not. DISCUSSION We have confirmed previously reported ranges of serum and CSF retinol and RBP concentrations in subjects with and without IIH ( 16- 19). Furthermore, our results are consistent with previous studies that documented higher serum retinol and RBP values in individuals with IIH than in others ( 16,17). Previously reported IIH- associated elevations of serum and CSF retinol are made more compelling by our findings, which for the first time compare and relate all four levels to each other in the same subject population. In addition, our study shows not only that CSF retinol is elevated, but also that the compensatory elevation of RBP is less than expected. In fact, CSF RBP was lowest in the subjects with IIH. The resulting retinol/ RBP ratio of > 1.0 in the CSF of our subjects with IIH suggests the presence of unbound toxic retinol that might interfere with pressure regulation. Furthermore, the presence of unbound retinol in the CSF supports the hypothesis that vitamin A may be involved in the pathogenesis of IIH. Previous studies document the fact that CSF RBP and CSF protein increase with age ( 23,25). This increase was confirmed in our subject population. One could speculate that high CSF protein, concomitant with high CSF RBP, may protect older individuals from development of IIH. It is also possible that there is another carrier protein in the CSF, in addition to RBP, that would nullify the toxic effects of the measured retinol excess. Thus far, no other carrier protein has been identified; but research in this area could prove to be fruitful. Our subject groups were understandably different demographically because of the strong association between IIH, obesity, youth, and female sex. However, when we controlled for BMI, the observed relative elevation of retinol compared with RBP in CSF persisted, although the absolute elevation of retinol lost its statistical significance. In serum, the elevations of retinol and RBP persisted, whereas the ratio was no longer significantly different. These findings suggest again that the unbound retinol in the CSF may be toxic because of insufficient RPB transfer into CSF, excessively dilute CSF with relatively low RBP, or other potential mechanisms. The regulatory system in serum may be more tightly controlled even in subjects with IIH. We recommend that future research be conducted with a larger sample size, which would permit investigation of possible influences of vitamin supplements, thyroid medication, and estrogen use on retinol and RBP levels. In our study, use of these medications did not differ significantly in the three groups. Vitamin use was lower in our subjects with IIH. Although we adjusted for these variables, our sample size was not sufficient to optimally address their impact on results. A larger sample size would also permit further investigation of the wide range of CSF retinol levels we observed in the IIH subject population of this study and our previous study ( 18). This broad range of CSF retinol levels raises the question as to whether there are some subgroups in which retinol plays a role in IIH and 261 J Neuro- Ophthalmol, Vol. 27, No. 4, 2007 Warner et al other subgroups in which retinol does not assume such a role. Further analysis of these hypothetical subgroups may provide useful results. The problem of finding appropriate age- matched control subjects willing to undergo LP solely for research purposes limits this and other studies. To procure optimal data for comparisons, it will be necessary to obtain reliable OP measurements and to determine CSF protein levels and cell counts from sex-, BMI- and age- matched controls. Comparative analyses of our IIH, neurology control, and anesthesia control groups demonstrated complex multivariate relationships. Further study is required to investigate the correlations we have identified between IIH and both CSF and serum levels of retinol and RBR Additional studies to determine a possible relationship between retinol saturation of RBP in the CSF and the development of IIH may be critical to further understanding of the pathophysiology and treatment of this disease. Acknowledgments The authors thank Clinton Thompson for assistance with statistical analyses and Susan Schulman for assistance with manuscript preparation. REFERENCES 1. Bandyopadhyay S. Pseudotumor cerebri. Arch Neurol 2001; 58: 1699- 701. 2. Binder DK, Horton JC, Lawton MT, et al. Idiopathic intracranial hypertension. Neurosurgery 2004; 54: 538- 51. 3. Digre KB, Corbett JJ. Idiopathic intracranial hypertension ( pseudotumor cerebri): a reappraisal. Neurologist 2001; 7: 2- 68. 4. Gerber A, Raab AP, Sobel AE. Vitamin A poisoning in adults; with description of a case. Am J Med 1954; 16: 729^ 5. 5. Feldman MH, Schlezinger NS. Benign intracranial hypertension associated with hypervitaminosis A.. Arch Neurol 1970; 22: 1- 7. 6. Smith FR, Goodman DS. Vitamin A transport in human vitamin A toxicity. N Engl J Med 1976; 294: 805- 8. 7. Vollbracht R, Gilroy J. Vitamin A induced benign intracranial hypertension. Can J Neurol Set 1976; 3: 59- 61. 8. Spector RH, Carlisle J. Pseudotumor cerebri caused by a synthetic vitamin A preparation. Neurology 1984; 34: 1509- 11. 9. Visani G, Manfroi S, Tosi P, et al. All- trans- retinoic acid and pseudotumor cerebri. LeukLymphoma 1996; 23: 437^ 2. 10. Eitenmiller RR, Landen WO Jr. Vitamin Analysis for the Health and Food Sciences. Boca Raton, FL: CRC Press; 1999. 11. Dingle JT, Fell HB, Goodman DS. The effect of retinol and of retinol-binding protein on embryonic skeletal tissue in organ culture. J Cell Sci 1972; 11: 393^ 02. 12. Aldred AR, Brack CM, Schreiber G. The cerebral expression of plasma protein genes in different species. Comp Biochem Physiol B Biochem Mol Biol 1995; 111: 1- 15. 13. Smeland S, Bjerknes T, Malaba L, et al. Tissue distribution of the receptor for plasma retinol- binding protein. Biochem J 1995; 305( Pt 2): 419- 24. 14. Herbert J, Wilcox JN, Pham KT, et al. Transthyretin: a choroid plexus- specific transport protein in human brain: the 1986 S. Weir Mitchell Award. Neurology 1986; 36: 900- 11. 15. MacDonald PN, Bok D, Ong DE. Localization of cellular retinol-binding protein and retinol- binding protein in cells comprising the blood- brain barrier of rat and human. Proc Natl Acad Sci USA 1990; 87: 4265- 9. 16. Selhorst JB, Kulkantrakorn K, Corbett JJ, et al. Retinol- binding protein in idiopathic intracranial hypertension ( IIH). J Neuro-ophthalmol 2000; 20: 250- 2. 17. Jacobson DM, Berg R, Wall M, et al. Serum vitamin A concentration is elevated in idiopathic intracranial hypertension. Neurology 1999; 53: 1114- 8. 18. Warner JE, Bernstein PS, Yemelyanov A, et al. Vitamin A in the cerebrospinal fluid of patients with and without idiopathic intracranial hypertension. Ann Neurol 2002; 52: 647- 50. 19. Tabassi A, Salmasi AH, Jalali M. Serum and CSF vitamin A concentrations in idiopathic intracranial hypertension. Neurology 2005; 64: 1893- 6. 20. Friedman DI, Jacobson DM. Diagnostic criteria for idiopathic intracranial hypertension. Neurology 2002; 59: 1492- 5. 21. Combs GF Jr. Vitamin Safety. In: Combs GF Jr. The Vitamins: Fundamental Aspects in Nutrition and Health. 2nd ed. San Diego, CA: Academic Press; 1998: 537^ 9. 22. Bendich A, Langseth L. Safety of vitamin A. Am J Clin Nutr 1989; 49: 358- 71. 23. Zheng W, Lu YM, Lu GY, et al. Transthyretin, thyroxine, and retinol-binding protein in human cerebrospinal fluid: effect of lead exposure. Toxicol Sci 2001; 61: 107- 14. 24. Krasinski SD, Russell RM, Otradovec CL, et al. Relationship of vitamin A and vitamin E intake to fasting plasma retinol, retinol-binding protein, retinyl esters, carotene, - tocopherol, and cholesterol among elderly people and young adults: increased plasma retinyl esters among vitamin A- supplement users. Am J Clin Nutr 1989; 49: 112- 20. 25. Johnston I, Hawke S, Halmagyi M, et al. The pseudotumor syndrome. Disorders of cerebrospinal fluid circulation causing intracranial hypertension without ventriculomegaly. Arch Neurol 1991; 48: 740- 7. 262 © 2007 Lippincott Williams & Wilkins |
