Clare, Gerry BSc MRCOphth; Colley, Stephen FRANZCO; Kennett, Robin BSc MD, FRCP; Elston, John S BSc, MD, FRCS, FRCOphth
Oxford Eye Hospital, Radcliffe Infirmary, Woodstock Road, Oxford, UK.
Address correspondence to Mr John Elston, Oxford Eye Hospital, Radcliffe Infirmary, Woodstock Road, Oxford OX2 6HE, UK; E-mail: firstname.lastname@example.org
Intracellular metabolism of folate, part of the B-vitamin complex, involves several regulatory enzymes for which genetic polymorphisms affecting the efficiency of the metabolic cycle have been described (1,2). Both the anti-inflammatory efficacy of methotrexate and its toxicity are complex and at least partly related to its antifolate properties (3-5). The combination of methotrexate therapy and another insult such as a dietary deficiency or a genetically reduced activity of the folate pathway enzymes leads to a decrease in serum folate levels (5,6). One of the consequences of low folate is an elevation of plasma homocysteine (HCy), which in animal and cell culture studies leads to inhibition of neuronal mitochondrial function by inducing reactive oxygen species (ROS) and ultimately results in neurodegeneration (7,8). The metabolically active papillomacular bundle could be affected preferentially, producing profound bilateral loss of visual acuity and field defects similar to those seen in hereditary optic neuropathies (9).
We present a patient who had a reversible symmetric optic neuropathy with deep central scotomas and dyschromatopsia attributable to methotrexate treatment and abnormal folate metabolism. Although testing failed to disclose the C677→T polymorphism for 5,10-methylenetetrahydrofolate reductase (MTHFR), the most common variant affecting the folate pathway, tests for less common polymorphisms were not performed.
A 66-year-old woman was referred with progressive bilateral visual loss over 5 weeks. She had been taking oral methotrexate 2.5 mg orally three times per week for rheumatoid arthritis for the previous 10 months (total intake 322.5 mg) without folic acid supplementation. She had never smoked or abused alcohol, her diet was healthy, and her medical history was otherwise unremarkable.
At presentation, best-corrected visual acuity was 20/200 OU. No relative afferent pupillary defect was present. Ishihara pseudo-isochromatic plate testing showed absent color vision. The optic discs and macula appeared normal. Goldmann field testing showed extensive central scotomas (Fig. 1). General physical and neurologic examination was normal.
Baseline hematological indices were as follows: hemoglobin 11.7 gm/dL, hematocrit 0.380 L/L, and mean cell volume 107.4 fL. Brain magnetic resonance imaging without contrast showed no abnormality. Electroretinography showed normal scotopic and photopic flash responses with retained oscillatory potentials, but the pattern reversal cortical visual evoked potentials (VEP) showed reduced amplitudes bilaterally (right 4 muV, left 3 muV, with normal being >5 muV) with normal latencies (Fig. 2).
Serum folate was reduced at 1.6 ng/mL (normal >4 ng/mL) with a normal vitamin B12 concentration of 256 pg/mL. Neither red blood cell folate nor plasma HCy levels were measured at this stage.
Methotrexate treatment was stopped and oral folic acid 5 mg/d was commenced. Over the next 4 months, visual acuities improved progressively to 20/15 OD and 20/20 OS with complete recovery of color vision and visual fields (Fig. 1). All hematological indices, including repeat folate levels, returned to normal.
By 6 months, repeat-pattern reversal cortical VEP amplitudes had recovered to 10 muV OD and 9 muV OS (Fig. 2). At this time, folate therapy was stopped. Plasma HCy and serum folate were measured after a 6-month interval free of therapy and found to be abnormal. HCy was 26.3 μmol/L (normal 5.0-15.0 μmol/L) and folate was reduced at 3.70 ng/mL. Testing failed to disclose the C677→T polymorphism for MTHFR, the most common variant affecting the folate pathway. Tests for less common polymorphisms were not performed.
The timing of this patient's visual recovery implicates abnormal folate metabolism as the cause of the optic neuropathy. Untreated dietary B-complex vitamin deficiencies associated with chronic alcoholism or malnutrition typically cause bilateral discrete central or cecocentral scotomas and less commonly precipitate profound visual field defects like those seen here (8). In this case, there was no history of smoking or excessive alcohol consumption. The normal VEP latencies after visual recovery and the normal brain magnetic resonance imaging suggest that the cause of the symmetric reversible optic neuropathy was not demyelination.
This patient's severe field defects are similar to those seen early in the course of inherited optic neuropathies associated with mitochondrial deficiencies, such as the maternally inherited Leber hereditary optic neuropathy. The derangement in folate metabolism may cause a severe optic neuropathy through an unknown mechanism in which the final common pathway is mitochondrial dysfunction. The optic neuropathy appears to be completely reversible.
Folate deprivation is implicated in a diverse range of disorders including fetal abnormalities and neurodegenerative diseases (1,2). Side effects of methotrexate may be related to its antifolate properties and toxicity involves a wide variety of tissues (3,4). Methotrexate inhibits intracellular dihydrofolate reductase and MTHFR. Dihydrofolate reductase is required to produce tetrahydrofolate (THF), an important substrate in purine and pyrimidine biosynthesis. MTHFR partly regulates production of 5-methyl THF, a cofactor required along with vitamin B12 to remethylate HCy to methionine, which acts as a methyl donor for DNA. Serum folate enters the cell via specialized receptors to replenish 5-methyl THF. If folate levels decline, substrates such as HCy increase and DNA methylation and repair is compromised.
Low levels of serum folate seen in patients on low-dose methotrexate treatment probably result from decreased intestinal absorption of folate (10) and consequent raised levels of Hcy (11). Interruption of remethylation of HCy is thought to be the mechanism by which vitamin B12 deficiency leads to a neuropathy in which an accumulation of 5-methyl THF is seen (1). HCy may be neurotoxic by induction of ROS within the cytosol and mitochondria, resulting in disruption of the respiratory chain (8). Electron microscopy studies in rat brains suggest that elevated HCy levels resulting from decreased folate result in mitochondrial degeneration (7). Our patient's elevated plasma HCy and low folate levels, noted after methotrexate and folate therapy were stopped, may indicate either an intrinsically defective folate cycle or inadequate intake of dietary folate. However, elevated HCy is not commonly associated with non-ischemic optic neuropathy, suggesting that other mechanisms are involved.
A genetic polymorphism C677→T resulting in reduced enzymatic activity has been described for MTHFR and is associated with greater toxicity of methotrexate. Functional polymorphisms have been described for other folate enzymes (2,5). Prevalence of homozygous (TT) genotypes with the C677→T MTHFR polymorphism is estimated to be approximately 12% (6). Our patient does not carry this variant, and tests for the other polymorphisms, which are uncommonly performed, were not performed. Therefore, we may never know whether her folate processing is genetically normal.
In most patients, dietary folate is enough to prevent methotrexate-related folate deficiency manifestations (6). Patients who have side effects at low methotrexate doses despite an absence of related risk factors may have abnormal folate metabolism (5,6).
The dyschromatopsia and symmetrical central scotomas seen in this case may represent an acquired mitochondrial dysfunction of the small metabolically active fibers of the papillomacular bundle. The pre-laminar unmyelinated portion of these fibers is particularly rich in mitochondria (9). The profundity of the visual field defects described here suggest a severe defect in the metabolic pathways involved that would not normally be seen in patients treated with low doses of methotrexate. However, given that the TT genotype is common, as are methotrexate therapy and dietary folate deficiency, it is difficult to know why this clinical picture is not seen more frequently. The widespread supplementation of methotrexate treatment with oral folic acid may mask this finding in the general population.
An isolated low level of serum folate has been implicated in reversible nutritional-toxic optic neuropathy in studies of patients who regularly consumed tobacco and alcohol (12,13) and in an isolated case report of a woman with a poor diet and moderate tobacco use (14). Low-dose methotrexate therapy has also been reported as a cause of optic neuropathy in a woman in whom visual field defects corresponded to dose changes (15) but whose serum folate was not recorded, and in a woman in whom serum folate was found to be normal (16). In both of these patients, transient optic disc swelling was noted, and visual field defects lessened but did not disappear when the drug was stopped.
In our patient, a direct neurotoxic effect of methotrexate cannot be excluded. The dramatic resolution of the optic neuropathy with normalization of the VEP and visual fields on administration of folate therapy suggests that methotrexate precipitated the visual loss. A toxic optic neuropathy should be suspected in any patient in whom painless visual loss develops while using methotrexate therapy, particularly if folate supplementation is not being provided. In such a case, discontinuation of methotrexate should be considered and folate supplements commenced. Electrophysiological testing and visual field testing, coupled with measurement of serum folate and plasma HCy levels, are helpful in establishing the diagnosis and monitoring visual recovery.
1. Scott JM. Folate and vitamin B12. Proc Nutr Soc 1999;58:441-8.
2. Zijno A, Andreoli C, Leopardi P, Marcon, F, Rossi S, Caiola S, Verdina A, Galati R, Cafolla A, Crebelli R. Folate status, metabolic genotype, and biomarkers of genotoxicity in healthy subjects. Carcinogenesis 2003;24:1097-103.
3. Chan ES and Cronstein BN. Molecular action of methotrexate in inflammatory diseases. Arthritis Res 2002;4:266-73.
4. Endresen GK, Husby G. Folate supplementation during methotrexate treatment of patients with rheumatoid arthritis: an update and proposals for guidelines. Scand J Rheumatol 2001;30:129-34.
5. van Ede AE, Laan RF, Blom HJ, Boers GH, Haagsma CJ, Thomas CM, de Boo TM, van de Putte LBA. Homocysteine and folate status in methotrexate-treated patients with rheumatoid arthritis. Rheumatology 2002;41:658-65.
6. Shelnutt KP, Kauwell GP, Chapman CM, Gregory JF 3rd, Maneval DR, Browdy AA, Theriaque DW, Bailey LB. Folate status reponse to controlled folate intake is affected by the methylenetetrahydrofolate reductase 677C→T polymorphism in young women. J Nutr 2003;133:4107-11.
7. Kim JM, Lee H, Chang N. Hyperhomocysteinemia due to short-term folate deprivation is related to electron microscopic changes in the rat brain. J Nutr 2002;132:3418-21.
8. Ho PI, Ashline D, Dhitavat S, Ortiz D, Collins SC, Shea TB, Rogers E. Folate deprivation induces neurodegeneration: roles of oxidative stress and increased homocysteine. Neurobiol Dis 2003;14:32-42.
9. Carelli V, Ross-Cisneros FN, Sadun AA. Optic nerve degeneration and mitochondrial dysfunction: genetic and acquired optic neuropathies. Neurochem Int 2002;40:573-84.
10. Leeb BF, Witzmann G, Ogris E, Studnicka-Benke A, Andel I, Schweitzer H, Smolen JS. Folic acid and cyanocobalamin levels in serum and erythrocytes during low-dose methotrexate therapy of rheumatoid arthritis and psoriatic arthritis patients. Clin Exp Rheumatol 1995;13:459-63.
11. Yxfeldt A, Wallberg-Jonsson S, Hultdin J, Rantapaa-Dahlqvist S. Homocysteine in patients with rheumatoid arthritis in relation to inflammation and B-vitamin treatment. Scand J Rheumatol 2003;32:205-10.
12. Golnik KC, Schaible ER. Folate-responsive optic neuropathy. J Neuroophthalmol 1994;14:163-69.
13. Chisholm IA. Serum cobalamin and folate in the optic neuropathy associated with tobacco smoking. Can J Ophthalmol 1978;13:105-9.
14. Hsu CT, Miller NR, Wray ML. Optic neuropathy from folic acid deficiency without alcohol abuse. Ophthalmologica 2002;216:65-7.
15. Johansson BA. Visual field defects during low-dose methotrexate therapy. Doc Ophthalmol 1992;79:91-4.
16. Balachandran C, McCluskey PJ, Champion GD, Halmagyi GM. Methotrexate-induced optic neuropathy. Clin Experiment Ophthalmol 2002;30:440-1.
© 2005 Lippincott Williams & Wilkins, Inc.