Clinicopathological Correlates: Chronic Arsenic Toxicity Causing Bilateral Symmetric Progressive Optic Neuropathy

A 70 year-old man presented with insidiously progressing central visual acuity loss in both eyes over several years. Objectively the only abnormality identified on the exam was questionable granularity in the fovea in each eye. Extensive work up which included neuro-imaging, screening blood work for toxic and nutritional causes of optic neuropathy as well as electroretinogram and fluorescein angiography to rule out subtle maculopathy was all unrevealing. When vision continued to deteriorate over the next several years investigations were repeated and again did not yield any positive results. Levels of heavy metals were then obtained after further progression of visual loss, revealing very high levels of arsenic. Subsequent investigations revealed that patient has been spending almost every weekend for the past 28 years alone at a remote country cottage where the sole supply of water was from the local well. He also recalled that 1.5 months after purchasing the cottage he developed hemorrhagic colitis requiring partial colectomy. The specimen from colectomy was located and total reflection x-ray fluorescence testing performed in a specialized lab revealed greatly increased level of arsenic particle in the colonic biopsy from 28 years ago. This case is a reminder that heavy metal toxicity should be considered in a differential diagnosis of patients with bilateral symmetric optic neuropathy.

Clinical-Pathological Case Study Section Editors: Daniel R. Gold, DO Marc Levin, MD, PhD Clinicopathological Correlates: Chronic Arsenic Toxicity Causing Bilateral Symmetric Progressive Optic Neuropathy Paul Freund, Laila Al-Shafai, Gabriella Mankovskii, David Howarth, Edward Margolin Abstract: A 70 year-old man presented with insidiously progressing central visual acuity loss in both eyes over several years. Objectively the only abnormality identified on the exam was questionable granularity in the fovea in each eye. Extensive work up which included neuroimaging, screening blood work for toxic and nutritional causes of optic neuropathy as well as electroretinogram and fluorescein angiography to rule out subtle maculopathy was all unrevealing. When vision continued to deteriorate over the next several years investigations were repeated and again did not yield any positive results. Levels of heavy metals were then obtained after further progression of visual loss, revealing very high levels of arsenic. Subsequent investigations revealed that patient has been spending almost every weekend for the past 28 years alone at a remote country cottage where the sole supply of water was from the local well. He also recalled that 1.5 months after purchasing the cottage he developed hemorrhagic colitis requiring partial colectomy. The specimen from colectomy was located and total reflection x-ray fluorescence testing performed in a specialized lab revealed greatly increased level of arsenic particle in the colonic biopsy from 28 years ago. This case is a reminder that heavy metal toxicity should be considered in a differential diagnosis of patients with bilateral symmetric optic neuropathy. Journal of Neuro-Ophthalmology 2020;40:423-427 doi: 10.1097/WNO.0000000000000961 © 2020 by North American Neuro-Ophthalmology Society Departments of Ophthalmology and Vision Sciences (PF, EM), and Medical Imaging (LA-S), Faculty of Medicine, University of Toronto, Toronto, Canada; Department of Biomedical Physics (GM), Ryerson University, Toronto, Canada; Department of Laboratory Medicine (DH), University of Toronto, Toronto, Canada. The authors report no conflicts of interest. Address correspondence to Edward Margolin, MD, FRCSC, Departments of Ophthalmology and Medicine, University of Toronto, 801 Eglinton Ave West Suite 301, Toronto, ON M5N 1E3; E-mail: Edward. [e-mail] Freund et al: J Neuro-Ophthalmol 2020; 40: 423-427 Drs. Freund and Margolin: A 71-year-old man was referred for the assessment of slowly progressive insidious vision loss in both eyes. His medical history was significant for hypertension, atrial fibrillation, and a remote colectomy for an episode of hemorrhagic colitis attributed to Crohn's colitis. He had congenital dyschromatopsia but no other significant ocular history. His medications included digoxin, ramipril, and bisoprolol. He was previously taking amiodarone but discontinued it before the onset of visual symptoms. He had no family history of ophthalmic or neurologic diseases. On examination, the best-corrected visual acuity was 20/80 in the right eye and 20/60 in the left eye. Pupils were briskly reactive, and there was no relative afferent pupillary defect (RAPD). Visual field testing (Humphrey field analyzer 24-2 algorithm) demonstrated asymmetrically decreased foveal sensitivities (27 dB in the right eye and 9 dB in the left eye) but was otherwise nonlocalizing. Optic nerves were not swollen or pale on dilated fundus examination, and optical coherence tomography (OCT) of the peripapillary retinal nerve fiber layers (RNFLs) was normal. There was questionable foveal granularity, but a retinal specialist felt that the appearance of the maculae was within normal limits. A neurologic examination, including ocular motility, was unremarkable. On reassessment 1 year later, the central visual acuity improved slightly to 20/50 in the right eye but declined to 20/300 in the left eye. Again, pupils were briskly reactive, and there was no RAPD. OCT of the peripapillary RNFLs was unchanged. He was investigated for possible bilateral optic neuropathy with MRI of the brain and orbits (without contrast) and screening bloodwork (complete blood count, erythrocyte segmentation rate, C-reactive protein, antinuclear antibody titers, venereal disease research laboratory test, angiotensin-converting enzyme level, vitamin B12, and folate levels). Genetic testing for the 3 most common Leber hereditary optic neuropathy (LHON) mutations was performed. All testing was unremarkable, except for mild 423 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Clinical-Pathological Case Study FIG. 1. Automated 24-2 Humphrey Visual Field Analyzer output demonstrates reduced foveal sensitivities in both eyes with mild nonspecific generalized reduction in sensitivities at extrafoveal testing points. leukopenia (3.19 · 109/L [range 3.5-10]) and thrombocytopenia (96 · 109/L [range 130-400]). A multifocal electroretinogram (mfERG) was performed to rule out an occult maculopathy and was reported to show decreased amplitudes within the central 30°. A second retinal specialist was consulted and did not feel that the mfERG results were reliable given a poor visual acuity and fixation. Visual-evoked potentials were not obtained. Dr. Alshafai: MRI of the brain and orbits, without contrast, demonstrates normal-appearing optic nerves with no evidence of decreased optic nerve caliber to suggest optic neuropathy. Elsewhere, there were only nonspecific T2-hyperintense foci in the white matter consistent with chronic microangiopathic changes. Drs. Freund and Margolin: The visual acuity declined to 20/200 in the right eye and counting fingers in the left eye over the next year with an otherwise unchanged neuro-ophthalmologic examination. Neuromyelitis optica (NMO) antibody testing performed using the enzyme-linked immunosorbent assay technique was negative, whereas contrast-enhanced MRI of the spine and repeat MRI of the brain and orbits were all unremarkable. The patient was lost to follow-up for 4 years before returning with a further decreased visual acuity of 20/400 in the right eye and counting fingers in the left eye. Pupils 424 remained reactive to light without an RAPD. Maculae appeared normal with normal foveal reflexes and without definite retinal pigment epithelial changes. Visual field testing demonstrated bilateral central scotomas with decreased foveal sensitivities (Fig. 1). Although the average RNFL thickness on peripapillary OCT was unchanged, there was subclinical thinning of the temporal RNFL, signifying loss of axons in the papillomacular bundle. OCT of the macular ganglion cell complex demonstrated significant thinning in both eyes with preservation of the normal outer retinal architecture and normal ellipsoid zone, arguing against maculopathy as a cause of visual loss (Fig. 2). Repeat investigations (complete blood count, erythrocyte segmentation rate, C-reactive protein, anti-nuclear antibody titers, venereal disease research laboratory test, angiotensin-converting enzyme level, vitamin B12, and folate levels) were again unremarkable, except for persistent leukopenia (2.30 ·109/L) and thrombocytopenia (91 · 109/L). NMO and myelin oligodendrocyte glycoprotein antibodies tested with cell-based assays were both negative. Given the progressive central visual loss, contrast-enhanced MRI of the brain, orbits, and the spine was repeated looking for occult structural lesions but was once again unremarkable. Further review of the possible causes of bilateral symmetric optic neuropathy led to the serum testing for heavy metals (arsenic, cadmium, lead, mercury, and zinc). Unexpectedly, very elevated levels of arsenic (279 nmol/L [range 0-50.7]) and mildly elevated levels of lead (0.28 mmol/L [range 0-0.10]) were found. Repeat testing 1 month later again confirmed Freund et al: J Neuro-Ophthalmol 2020; 40: 423-427 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Clinical-Pathological Case Study FIG. 2. Zeiss Cirrus optical coherence tomography analysis of the optic nerve head/peripapillary retinal nerve fiber layer (RNFL; left) and macular ganglion cell complex (GCC; right). The RNFL analysis shows generalized borderline thinning in both eyes and a sector of thinning in the right eye. The RNFL was unchanged from OCT imaging 4 years earlier. GCC analysis shows marked diffuse thinning in both eyes. OCT, optical coherence tomography; ONH, optic nerve head. very elevated serum levels of arsenic and mildly elevated lead levels. Random urine testing for arsenic and lead levels was within the normal range. His results were reviewed with public health officials who felt the mildly elevated lead level was not significant and unrelated to his symptoms. The patient's partner had normal serum levels of heavy metals. On further probing, the patient denied any exposures to materials that may be contaminated with arsenic through his occupation or hobbies, including exposure to stained glass, pigments, textiles, papers, metal adhesives, wood preservatives, ammunition, or tobacco. He denied taking any supplements and specifically denied taking any nutritional supplements or vitamins ordered from overseas (a possible source of arsenic intoxication). On further questioning of his hobbies, he said that he had gone alone almost every weekend to a remote country cottage for the past 28 years. He relayed that the water in the cottage was sourced from the local well. It was established that he purchased the cottage 28 years earlier, 2 months before his isolated episode of hemorrhagic colitis. Sections from the resected descending colon specimen at that time were located and rereviewed. Dr. Howarth: The initial clinical diagnosis was toxic megacolon, and the pathological diagnosis of Crohn's disease was made; however, on rereview of the slides along with gastrointestinal Freund et al: J Neuro-Ophthalmol 2020; 40: 423-427 pathologists, the diagnosis of nonspecific colitis was made (Fig. 3). There are foci of ulceration and transmural inflammation with cryptitis and crypt abscesses, as would be seen in Crohn's disease, with a patchy nature of the inflammation with some distal sparing (unlike ulcerative colitis); however, evidence of granulomatous inflammation, as would be seen in Crohn's disease, and more importantly, evidence of chronicity such as gland branching in the FIG. 3. Pathologic photomicrograph (hematoxylin & eosin, ·40) of the descending colon demonstrating a nonspecific colitis with patchy foci of ulceration and transmural inflammation with cryptitis and crypt abscesses with distal sparing. There is a lack of both granulomatous inflammation and evidence of chronicity to suggest the underlying inflammatory bowel disease as a cause. 425 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Clinical-Pathological Case Study epithelium, as would be seen in both inflammatory bowel diseases, is not present, excluding both. A research laboratory performed an analysis of the colon sample with total reflection x-ray fluorescence to determine its heavy metal contents. It found very elevated arsenic levels in the sample at 0.0059 ppm (equivalent to 21.43 ng/g). Final Diagnosis Optic neuropathy secondary to chronic arsenic exposure. Drs. Freund and Margolin: The presentation of an insidiously progressing bilateral symmetric optic neuropathy primarily affecting the papillomacular bundle is most consistent with toxic or metabolic optic neuropathy. Metabolic optic neuropathies can result from a variety of etiologies including toxic exposures (common exposures include methanol, carbon monoxide, ethambutol, or chloramphenicol and, much less commonly, heavy metal toxicity), nutritional deficiencies (commonly vitamin B12, folate, or thiamine), or inherited optic neuropathies (including LHON and dominant optic atrophy) and can be exacerbated by metabolic stressors, including smoking and alcohol consumption (1). The shared pathophysiology of these different etiologies is mitochondrial dysfunction, which manifests as bilateral optic neuropathy. The papillomacular bundle is preferentially affected by mitochondrial dysfunction because of high metabolic demands and its constituent small caliber axons, which are more susceptible to energy depletion. Investigations for nutritional deficiencies (vitamin B12, folate, and thiamine levels) and LHON testing were all negative. Inflammatory, demyelinating, or compressive optic neuropathies are rarely so symmetric and insidious but were ruled out in our case by repeated negative neuroimaging and antibody testing for known antibody-mediated optic neuropathies. Retinal mimics such as an inherited cone dystrophy or toxic maculopathy were excluded because of the absence of any progressive retinal changes, including retinal pigment epithelium pigmentary changes/loss, thinning of the outer retina, or disruption of the ellipsoid zone. The reduced amplitudes on mfERG were felt to reflect poor fixation from his poor acuity rather than outer retinal dysfunction. Repeatedly, elevated levels of arsenic and the presence of arsenic in the resected colon shortly after the exposure to arsenic began supported the diagnosis of chronic arsenic toxicity. A toxicologist felt that serum lead levels slightly higher than the recommended threshold set by the Centers for Disease Control and Prevention were not sufficient to produce clinically significant lead toxicity but might have been related to well water with other sources, including contamination from lead water pipes, being possible as well. Arsenic is a toxic and carcinogenic element that is commonly found in the environment. The source of arsenic 426 exposure is most commonly contaminated groundwater, but it can also be ingested from contaminated food or exposure to arsenic from industrial processes, occupational and hobby exposures, and cigarettes. The toxidrome of arsenic varies with the dose and duration of the exposure, as well as the specific arsenic species. Inorganic arsenic, such as arsenate (AsV) or arsenite (AsIII), is frequently present as a groundwater contaminant and is ingested through contaminated drinking water or by food grown or prepared with arsenic-containing water. Organic species of arsenic (such as arsenobetaine or trimethylarsine) are biologically unavailable or rapidly secreted; so, despite higher dietary consumption of organic arsenic from seafood, the morbidity is felt to be less with organic arsenic than inorganic arsenic. The mechanism of arsenic toxicity is believed to be related to its inactivation of up to 200 enzymes, specifically those responsible for cellular duplication and repair as well as energy production. Symptoms of acute arsenic poisoning include muscular pain, weakness, nausea, vomiting, abdominal pain, and diarrhea, as was the case with this patient (2-4). Signs of chronic arsenic exposure include skin changes (hyperpigmentation, palmar and solar keratosis), and neurological symptoms consist of peripheral neuropathy mimicking Guillain-Barre syndrome and sensory neuropathy with glove and stocking anesthesia. Our patient's chronic leukopenia and thrombocytopenia were also consistent with hematologic abnormalities from arsenic exposure. Longterm exposure can lead to changes in behavior, confusion, and memory loss, which had been noticed by the patient's family during the final year of his course. Chronic arsenic exposure also increases the risk of various malignancies involving the skin, lung, liver, kidney, and bladder. Testing for arsenic usually involves measurements of arsenic levels in the urine in which the half-life of the inorganic arsenic is about 4 days (5). In our patients, urine testing was performed only once, and it was 2 weeks after he had last visited the cottage. Besides urine, levels of arsenic in the hair and nails can be measured because they retain it longer than other tissues; however, the measurements can be prone to errors because there is only an approximate relationship between the levels of arsenic in the hair and its toxicity. We repeatedly attempted to arrange hair testing, but the patient had a difficult time following the instructions. Measurements of organic and nonorganic arsenic in the blood reflect recent exposure because the half-life of arsenic in the blood is short. Our patient had both blood tests for the arsenic level performed shortly after he came back from the cottage where he was presumably exposed to contaminated groundwater. In clinical practice, if clinical suspicion for arsenic toxicity is high, the best screening tool is a 24-hour urine collection with arsenic speciation. Other heavy metals such as lead and mercury can be evaluated adequately with serum levels. Chronic arsenicosis causes irreparable damage in multiple organs and increases the risk of various Freund et al: J Neuro-Ophthalmol 2020; 40: 423-427 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Clinical-Pathological Case Study malignancies. There is no proven treatment for arsenic toxicity, although chelation therapy has been used for the reduction of overall arsenic levels in the body. Various chelating agents have been tried (dimercaptosuccinic acid, dimercaptopropane succinate, and D-penicillamine) with varied results (5). In summary, our patient's presentation with an acute onset of hemorrhagic colitis 2 months after he purchased a cottage in rural Ontario and a delayed onset of slowly progressive symmetric optic neuropathy all point out to the contamination of well water by naturally occurring arsenic as the cause of the visual loss. We have indirect evidence of his chronic arsenic exposure with repeatedly high serum levels of arsenic and increased levels of arsenic in the colonic resection performed shortly after the exposure presumably started. Confirmatory testing of the presence of arsenic in his well water could not be completed by the patient because of his declining mental status and confusion. Shortly after his last visit with us, the patient was admitted to the intensive care unit (ICU) with an acute onset of lower gastrointestinal bleeding, the source for which was not Freund et al: J Neuro-Ophthalmol 2020; 40: 423-427 found. He also developed very severe painful diffuse gout (which has been reported to be associated with chronic arsenic exposure) and spontaneous subarachnoid hemorrhage (6). Shortly after his admission to the ICU, the family decided to withdraw care, and he died. The family declined autopsy. REFERENCES 1. Sadun AA. Metabolic optic neuropathies. Semin Ophthalmol. 2002;17:29-32. 2. Hughes MF, Beck BD, Chen Y, Lewis AS, Thomas DJ. Arsenic exposure and toxicology: a historical perspective. Toxicol Sci. 2011;123:305-332. 3. Susan A, Rajendran K, Sathyasivam K, Krishnan UM. An overview of plant-based interventions to ameliorate arsenic toxicity. Biomed Pharmacother. 2019;109:838-852. 4. Ratnaike RN. Acute and chronic arsenic toxicity. Postgrad Med J. 2003;79:391-396. 5. Arsenic in drinking water. World Health Organization Publication. Available at: https://www.who.int/water_sanitation_health/ water-quality/guidelines/chemicals/arsenic.pdf?ua=1. 6. Kuo CC, Weaver V, Fadrowski JJ, Lin YS, Guallar E, Navas-Acien A. Arsenic exposure, hyperuricemia, and gout in US adults. Environ Int. 2015;76:32-40. 427 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited.