Clinical manifestations

Presentation and course

Environmental sources. Because mercury is ubiquitous, exposure to mercury does not necessarily lead to illness. Mercury is present in water, air, and soil. The United States Environmental Protection Agency (EPA) has listed safe levels of mercury for humans in air, soil, and water.

The general population may experience exposure from dental amalgams or seafood. ATSDR has reported that very small mercury exposures are not necessarily harmful to one’s health, but active research is ongoing. Sensitive populations are identified as pregnant woman, children under 6 years of age, those with impaired renal function, and those with hypersensitive immune systems. ATSDR has also reported that chelation and removal of amalgams is not indicated in those without health effects from mercury.

Pyatha and colleagues also reported that patients exposed to mercury through dental amalgam fillings are at a six-fold increased risk for the development of Parkinson disease compared to nonexposed patients (60).

Santos-Sacramento and colleagues systematically reviewed mercury’s neurotoxic impact on individuals in the Amazon with mercury exposure within three river basins (Tapajos, Tocantins, and Madeira) (67). Their findings showed mercury body burden was two to 10 times higher than recommended with neurologic deficits noted in cognitive function, vision, motor function, somatosensory, and emotional deficits.

Exposures may also be experienced by religious groups who use mercurial agents in ceremonies. Exposure may occur to those using the agent as well as those who might inhale contaminated air nearby. Metallic mercury is sold in botanicas under the name azogue and is common in the Haitian and Hispanic communities. Inorganic mercury was found to be a component of a previously unrecognized source of exposure: skin care products in New York City (53). These products were found to be produced mainly in the Dominican Republic. A large study assessing urine mercury in 1840 adults in New York City was accomplished, and it uncovered 13 individuals with levels exceeding the New York State reportable level of 20 ug/L. All of these individuals were either Hispanic or African American and between the ages of 21 and 51. Ten of these women were born in the Dominican Republic. Whitening skin creams were confiscated, tested, and found to contain 0.62% mercury by weight. Blood mercury levels were assessed and found to be positively associated with elevated urine mercury. Seventeen different products were seized from stores and tested for mercury. Eight of these were skin-lightening or beauty creams; nine were made in the Dominican Republic, and seven listed mercury as a component of the cream.

A report from Taiwanese literature identified a 51-year-old male who was dispensed traditional medicine to treat a perianal fistula (87). After using hong-dan, he presented with perianal gangrene, fever, rash, and numbness, and later he presented with progressive weakness. EMG/NCV did, in fact, reveal a mixed axonal along with demyelinating features (reduced sensory amplitude, prolonged F waves). Hypokalemia, normocytic anemia, and elevated CRP were noted. AIDP was suspected, and despite plasmapheresis, his numbness and weakness worsened. A chelating agent, succimer (DMSA), was given, and high urine mercury and arsenic resulted. On day 27, the mixed axonal and demyelinating picture remained. Four years later, the patient had partial resolution of his symptoms, but NCV revealed nonresponse for peroneal motor and sural sensory studies. The authors wrote that hong-dan and tung oil are preparations banned due to lead content, but the paste remains compounded with unlicensed practitioners and may also contain significant mercury and arsenic.

A 4-year-old boy admitted to the hospital for rapidly progressing difficulty in walking had been treated for about 2 months with primarily ammoniated mercury ointment for itchy, macular lesions thought to be insect bites over his scalp and body. On examination, he was irritable but with an unremarkable physical examination. He did have pronounced weakness of the flexors and extensors of the feet and less of the legs, and his muscles of the upper extremities were slightly weak, but he had no atrophies or fasciculations. It was noted that he had no signs of acrodynia; however, with his irritability, polyneuropathy, albuminocytologic dissociation in the CSF, 9% eosinophils of the peripheral blood, and history of prolonged use of ammoniated mercury ointment, mercury intoxication was suspected. This was confirmed by a 24-hour urine mercury level of 0.3 mg/L. The patient was treated with dimercaprol (BAL) 3 mg per kg body weight four times a day, then twice daily. The patient showed progressive strength improvement, and when re-examined as an outpatient a year later, he walked normally and demonstrated no motor weakness. Achilles reflexes, however, were still absent. The authors concluded that pathogenesis was evidently a combination of chronic absorption and the individual’s hypersensitivity to inorganic mercury. The authors also noted that their patient did not display signs of acrodynia but did resemble a case of abortive acrodynia with mild polyneuritis, atypical extrapyramidal hyperkinesia, asynergia, eosinophilia, and albumin-cytologic dissociation in the CSF, after being treated with a mercury preparation rectally as an anthelmintic (61).

In 2010, twin 13-year-old sisters were admitted for generalized body pain, especially in the extremities and back, and they complained of headaches, chilling, and sweating (39). Noted was a 2-month history of pain in their legs, gradually causing them not to be able to walk to school. Anorexia, weakness, lethargy, excessive sweating, and salivation were also noted. Itchy rash on the body and extremities presented 2 weeks before hospital admission. On admission, the patients were irritable and depressed and had tachycardia without elevated blood pressure or body temperature. MRI of the brain and thighs and EMG/NCV testing were normal in both patients. Twenty-four-hour urine mercury concentration were measured and revealed toxic levels (50 and 70 ug/L). The mother later recalled using a compound that contained mercury to treat pediculosis 3 months before the onset of symptoms. The compound was purchased from an herbal shop and was applied locally on the scalps for 2 days. The twins were treated with D-penicillamine (DPCN) 20 ug/kg/d orally in four divided doses. They were additionally treated with gabapentin (300 ug) orally in three divided doses for pain. After 2 weeks of therapy, 24-hour urine mercury concentrations had fallen to 9 to 15 ug/L, and the patients were discharged. The authors reported that the sisters’ clinical presentation was compatible with acrodynia, characterized by dark pink discoloration and scaling of the hands and feet, weakness, pain, irritability, hypertension, tachycardia, peripheral neuropathy, recessive salivation, and sweating. The authors concluded that mercury poisoning should always be included in the differential diagnosis when patients present with bone or joint pain, skin rash erythema, and peripheral neuropathy.

In another pediatric case study, a 14-year-old boy was seen at a rheumatology clinic with severe back and extremity pain, weight loss, and weakness (41). Initial labs and MRI were all normal, but EMG showed mild neurogenic involvement on L2-S1. Clinical investigation discovered he had played with a bottle of mercury on a construction site 2 months prior. Follow-up showed blood mercury levels were significantly high, and he was treated with dimercaptosuccinic acid with reduced complaints over the following months (41).

A 54-year-old female with early onset dementia, epilepsy, and peripheral polyneuropathy was diagnosed with mercury intoxication in 2010, and she failed chelation therapy with 2,3-dimercaptopropane-1-sulfonate (DMPS) in 2013. EMG revealed chronic denervation in the tibialis anterior, but no follow-up studies were performed. Following a detailed patient history, it was found that she had been using a mercury-containing skin-lightening cream daily for 6 years. Treatment with 200 mg oral DMSA three times a day was initiated. Blood mercury levels fell from 13.3 to 5.2 ug/L, NSE levels fell from 31.4 to 20.5 ug/L, and urine mercury levels fell from 18.4 to 3.7 ug/L. Follow-up brain MRI showed unchanged brain lesions and one new lesion seen in the left temporo-occipital subcortical region. On discharge, however, memory function was improved qualitatively and quantitatively (90). The Environmental Protection Agency (USEPA) and U.S. Food and Drug Administration (FDA) “Advice About Eating Fish” states that eating fish while pregnant or breastfeeding can provide healthy benefits; however, it is important to avoid fish high in mercury, such as swordfish, king mackerel, marlin, and shark (81). Other fish associated with high mercury levels are tilefish from the Gulf of Mexico, Orange roughy, and Bigeye tuna.

A diet heavy in shellfish, fish, and marine mammals from mercury-contaminated waters may lead to high blood mercury levels. The largest and oldest fish will have the highest levels of mercury. The U.S. Food and Drug Administration (FDA) has estimated the amount of mercury in a normal adult diet and found it not to be harmful. Literature has described that this risk is likely underestimated (16).

Other food sources of mercury exposure have also been identified. NHANES data combined with that from the U.S. Department of Agriculture’s Food Intakes Converted to Retail Commodities Database (FICRCD) examined 49 different food and environmental metal exposure. Using results from blood and urinary biomarkers, researchers estimated that 4.5% of the variation of mercury among children and 10.5% among adults is explained by diet. A previously unrecognized association between rice consumption and mercury exposure was found (24). Latov and colleagues identified that environmental sources may play a role in idiopathic axonal and small fiber neuropathy (see below) (48).

Various forms of mercury poisoning. The various forms of mercury (organic mercury compounds, inorganic mercury salts, and elemental mercury) are all potentially neurotoxic and may produce different clinical pictures. They will be considered separately.

Organic mercury poisoning and neuropathy. Organic mercury is readily absorbed through the skin and gastrointestinal tract. Data concerning organic mercury (methylmercury) poisoning have come from accidental human exposures. Mercuric chloride is an inorganic compound used as an industrial catalyst. In the Japanese fishing village of Minamata, it was discharged into the sea for decades by an industrial plant and converted into methylmercury by microorganisms. It then entered the food chain and was concentrated in the bodies of fish and shellfish. These organisms served as staple items in the diet of the local population. It was observed that cats and seabirds, who also fed on the fish, were developing neurologic disease, leading to the identification of methylmercury as the toxin (45).

Paresthesias and sensory loss in stocking-glove distributions were noted in the distal extremities of Minamata Bay inhabitants. Muscle atrophy was also reported, and these symptoms appeared to have worsened over a 3- to 10-year period (79).

Another major outbreak arose in Iraq after grain treated with an ethyl mercury fungicide was used for baking bread (37; 08). This grain had originally been intended for planting. Paresthesias were also observed in these caes (29).

During chronic exposure to methylmercury, prominent central nervous system disease develops, including visual field constriction, ataxia, dysarthria, hearing loss, tremor, and cognitive impairment. One of the earliest complaints in many patients is distal paresthesia (77), which progresses proximally and may involve the tongue. This and ataxia are probably related to selective toxic effects on the neurons of the dorsal root and trigeminal sensory ganglia. Excretion of methylmercury is first order in humans, and its half-life has been estimated to be roughly 70 days (22). There may be a long latency period between the time of exposure and the onset of neurologic symptoms; the reason for this latency is not clear (56).

Dimethylmercury, a “supertoxin,” is lethal in doses of only a few drops, even if the exposure is brief. Exposure may occur via inhalation or transdermal absorption; typical disposable latex gloves do not protect against permeation. A tragic case of dimethylmercury poisoning occurred in 1996. A 48-year-old chemistry professor accidentally spilled drops of dimethyl mercury from her pipette onto the back of her latex glove. After a latent period of over 5 months, she developed a progressive cerebellar syndrome with gait ataxia, dysmetria, and scanning speech. Subsequently, she suffered progressive visual and hearing loss, paresthesias in the limbs, and encephalopathy. Despite 3 months of aggressive treatment, including chelation therapy with oral DMSA and exchange transfusion, she was left in a condition resembling a persistent vegetative state; life support was withdrawn in accordance with previous wishes (56).

A link between thimerosal, an ethyl mercury-containing preservative used in vaccines, and autistic spectrum disorders has been suggested. There are no compelling studies to date, however, to substantiate these claims (57). In 2002, The World Health Organization advisory committee deemed it safe to continue to use thimerosal in vaccines (74). Nevertheless, thimerosal-containing vaccines are no longer used in the United States, with the exception of influenza vaccines. Ingestion of a large dose of thimerosal in an adult led to multisystemic toxicities, including peripheral neuropathy, with subsequent recovery (59).

Le Quesne and associates evaluated 19 patients with organic mercury poisoning, four of whom were severe, and none demonstrated peripheral electrodiagnostic abnormalities, suggesting a CNS etiology for their complaints (49). Von Burg and Rustam evaluated Iraqis with poisonings and found no abnormalities. However, they did discover that patients responded better to neostigmine than placebo (82; 62).

In 1981, a brief report described increased mercury vapor released into the breath by chewing (75). Patients with silver-mercury amalgam fillings were compared to patients without these fillings. The authors raised the question that biotransformation might occur, and low levels of ingested or inhaled mercury might be converted into more toxic methylmercury. Despite the lack of direct evidence for this biotransformation, the role of dental amalgam in the production of neurologic disease has repeatedly arisen. Cases have included peripheral neuropathy, dementing disorders, motor neuron diseases, and chronic fatigue syndrome. No published studies have ever been able to link amalgam fillings to neurologic disease, and no trial has ever demonstrated a benefit from removing fillings. After an intensive investigation and review of the evidence, the Public Health Service concluded in 1993 that dental amalgams did not pose a serious health risk (44). A study of 1663 middle-aged men found no detectable association between amalgam exposure and peripheral neuropathy (40). Shapiro, however, found 30% of dentists with raised tissue levels of mercury had electrodiagnostic evidence of reduced mean sural and median motor conduction velocities compared to dentists with normal tissue mercury levels, consistent with subclinical neuropathy (70). For these reasons, it must be concluded that, at present, there is no evidence to link amalgam fillings to specific neurologic disease. Furthermore, the removal process generates mercury vapor, raising mercury blood concentrations substantially before eventual decline (22).

A paper by Latov and colleagues, evaluating blood levels of mercury by mass spectrophotometry in patients with idiopathic axonal and small fiber neuropathy, found an increased incidence of elevated blood levels, which were suspected to be from ingestion of methylmercury in seafood (48). Although cause or contribution to these neuropathies was not concluded, certainly more studies are warranted.

Inorganic mercury poisoning and neuropathy. It is difficult to separate the effects of elemental mercury from those of mercuric salts because most cases arise from occupational exposure to both forms (11). It is probable that inorganic mercury, because of its greater water solubility, is best absorbed through the gut; it produces prominent systemic as well as nervous system effects. Elemental mercury, with greater lipid solubility, may produce nervous system disease with almost no systemic disease.

Mercury salts are used in a number of industrial processes. The major route of absorption of these mercury salts is by way of the gastrointestinal tract. Mercuric sulfate has been found in traditional Chinese herbal medications, which have been implicated in cases of mercury neuropathy (21). Gastrointestinal symptoms and a nephrotic syndrome may be prominent systemic features, especially in acute poisoning.

Inorganic mercury is used in the chloralkali industry. Peripheral nerve dysfunction, often motor more than sensory, was noted in the cases summarized. Albers and colleagues reported 138 chloralkali plant workers with long-term exposure to inorganic mercury vapor who were found to have elevated urine mercury levels, reduced sensation on quantitative testing, prolonged distal motor latencies, reduced sensory evoked response amplitudes, and increased likelihood of abnormal needle EMG findings. Subjects exposed to mercury for 20 to 35 years who had urine mercury levels greater than 0.6 mg/L demonstrated significantly less strength, poorer coordination, more severe tremor, more impaired sensation, and higher prevalence of Babinski and snout reflexes than controls. Subjects with polyneuropathy had higher peak levels of mercury than healthy subjects. Factory workers exposed to elemental mercury vapor with elevated urine mercury concentrations had prolonged motor and sensory ulnar distal latencies. Slowing of the median motor nerve conduction velocity was found to correlate with both increased levels of mercury in blood and urine and increased numbers of neurologic symptoms. Sensory deficits were found with short-term exposure to mercury vapor, whereas motor nerve impairment occurred with longer periods of exposure (05; 04). Motor nerves appeared to be more affected than sensory nerves in a study by Singer in 1987 that assessed 16 workers with inorganic mercury exposure. Median motor NCV slowing correlated with mercury blood and urine as well as an increased prevalence of neurologic symptoms (71).

In another study by Andersen and coworkers, chloralkali workers exposed to inorganic mercury vapor for an average of 12.3 years revealed a higher prevalence of reduced distal sensation, postural tremor, and impaired coordination than controls revealed (07).

Barber reported two employees of a chloralkali plant who had findings suggestive of amyotrophic lateral sclerosis (09). Signs, symptoms, and laboratory findings returned to normal 3 months after withdrawal from exposure.

Adams and associates reported a 54-year-old man with a brief but intense exposure to mercury vapor, which led to a syndrome resembling amyotrophic lateral sclerosis that resolved as urinary mercury levels fell (02). Ross reported that prolonged application of an ammoniated ointment to the skin was a cause of motor polyneuropathy with cerebrospinal fluid findings suggestive of Guillain-Barré syndrome.

Warkany and Hubbard reported the association of acrodynia and symmetrical flaccid paralysis with mercury toxicity (84).

Chloralkali workers who were exposed to inorganic mercury for an average of 7.9 years and had ceased working in that environment an average of 12.3 years before the study were found to have both median sensory nerve conduction velocities and amplitude of the sural nerve associated with measures of cumulative mercury exposure (07; 26). A study reviewing the relationship between exposure-related indices and neurologic and neurophysiologic effects in workers previously exposed to mercury vapor revealed that, of 298 dentists with long-term exposure to mercury amalgam vapor evaluated for peripheral neuropathy, 30% had polyneuropathies (70). Another paper reported that a dentist apparently had an unelicitable sensory superficial peroneal nerve action potential that returned to normal following penicillamine treatment (36).

Industrial workers with long-term mercury exposure were found to have performance decrements in neuromuscular functions that were reversible and correlated with blood and urine mercury levels (11).

A 2019 Norwegian literature review of dentists and dental personnel exposed to mercury in the occupational setting was done to evaluate the relevance between mercury exposure in the dental industry and idiopathic disturbances in motor functions, cognitive skills, and affective reactions, as well as dose-response relationships. This study looked at dentists, dental assistants, and dental students from various countries. In multiple studies reviewed, neuropsychological effects in attention, psychomotor speed, cognitive flexibility, visual memory, short-term memory, visual memory, and visuomotor coordination were noted. In another study of neuropsychological performances in low-level mercury-exposed female dental workers, results showed that even chronic subtoxic levels of inorganic mercury can cause mild modifications in short-term nonverbal recall and generally increased distress, psychoticism, and anxiety. These researchers concluded that dental workers exposed to chronic, low levels of metallic mercury resulted in significant neurologic and sensory symptoms. Of these, memory loss was noted as potentially the most important.

Elemental mercury poisoning. Acute inhalation of elemental mercury vapor can cause irritation to the upper respiratory tract causing acute pneumonitis, bronchitis, and bronchiolitis and is commonly mistaken for viral infection if not properly diagnosed (13).

Chronic inhalation of elemental mercury may cause adverse health effects if exposure is above the minimal risk level of 0.3 μg/m3 (03). Clinical presentations are subtle and typically considered insignificant. Complaints include fatigue, general weakness, lack of appetite, diarrhea, lack of sleep, change in mood, and tremor, which has been called “micromercurialism” (35). The tremor is described as fine and rapid, worsens with activity and emotional excitement, and resembles the tremor seen in patients with hyperthyroidism. The tremor frequency, however, is faster and differs from the characteristic resting pill-rolling seen in Parkinson patients (32).

Elemental mercury is absorbed poorly from the gastrointestinal tract; swallowing mercury is not known to be harmful. Metallic mercury is volatile at room temperature. A saturated atmosphere of mercury vapor contains 18 mg of Hg per cubic meter at 24 degrees Celsius. Friberg and Nordberg have demonstrated that chronic Hg exposure to levels above 0.1 mg/m³ in air is associated with the development of toxicity (35). This level is of particular significance because small quantities of metallic mercury may remain hidden for long periods in poorly ventilated areas. Because of its greater lipid solubility, toxicity often manifests first by central nervous system symptoms of fatigue associated with weight loss, anorexia, and gastrointestinal dysfunction. With higher exposure, personality and behavioral changes, insomnia, and hyperexcitability develop, often associated with a coarse facial, head, and limb tremor. A few cases of predominantly peripheral nervous system involvement associated with exposure to organic and elemental mercury have been recorded (61; 76; 86). Elemental mercury exposure from broken school barometers led to symptoms and signs of mercury toxicity in three teenagers; all three had secondary hypertension, and two of the three developed peripheral neuropathy and CNS involvement (43).

A review of several studies that included more than 3000 workers chronically exposed to mercury between the years 1950 and 2015 found that urine mercury levels from less than 50 ug/L to more than 200 ug/L showed increased frequency in tremor, impaired coordination, and abnormal reflexes on physical examination and reduced neurobehavioral tests of tremor, manual dexterity, and motor speed. The data suggest that response thresholds of urine mercury levels of about 275 ug/L for physical exam findings and about 20 ug/L for neurobehavioral outcomes (34).

Seventeen thermometer factory workers with no neurologic complaints had high urine and blood mercury levels but no symptoms; 14 had subclinical neuropathy, mainly distal and axonal neuropathy (89).

Levine and associates, in 1982, studied right ulnar nerve NCVs in factory workers with elemental mercury exposure (50). Prolonged motor and sensory distal latencies were correlative with increasing urine mercury levels.

Albers and colleagues evaluated 386 chloralkali workers with 20 to 35 years of elemental mercury exposure and uncovered prolonged sensory and motor distal latencies in those workers whose biological monitoring revealed urine mercury concentrations above 600 ug/L (06). EMG was also performed and was noted to show an increased prevalence of abnormal electrical activity.

Authors in the United States reported that a patient developed gait ataxia, diplopia, and vomiting 1 year after self-injection of elemental mercury during a cultural practice (52). He was found to have an indurated plaque on his right shoulder along with cranial neuropathies. Urine mercury was elevated. CSF revealed pleocytosis. After excision of the skin lesion, a reservoir of mercury, and chelation, he experienced mild improvement.

In 2013, Turkish authors reported 179 children were exposed to elemental mercury in schools in two different provinces of Turkey (18). Children took the mercury home and invited others to play with it. Some boiled the material, heated it, and even tasted it. Median exposure duration was 5 minutes, but some played for 24 hours. Fifty-one percent were asymptomatic, and 44% had a normal neurologic examination. Thirty percent had headaches, and 5.5% had symptoms of peripheral neuropathy with normal NCV and EMG. MRIs of the brains were normal. Duration of exposure was correlated with elevated urine mercury. Pupillary dilatation was noted as the most common exam finding. After 6 months, peripheral neuropathy symptoms resolved.

In 2015, Korean workers involved in a fluorescent lamp factory demolition project developed acute symptoms of mercury poisoning from mercury vapor (25). Eighteen of 21 workers participating in the demolition project developed symptoms, with 10 of 18 having persistent symptoms 18 months postexposure. Initial symptoms included skin rash, pruritus, myalgia, and sleep disturbance. Late-onset symptoms included fatigue, insomnia, and anxiety disorder. Three workers developed jerky movement, and two had swan neck deformity of the fingers. Mean blood mercury levels were 5.88 ug/L, and the adjusted urinary mercury levels were 13.48 ug/L - creatinine. The lack of improvement by several of the workers was thought to be the result of mercury being captured by the central nervous system. Unfortunately, for workers whose condition did not improve, their symptoms were predicted to become chronic.

Prognosis and complications

The prognosis for recovery from peripheral nervous system involvement is good, although deficits may persist in severe cases (21). CNS injury is usually permanent. Recovery of elemental mercury exposure and peripheral neuropathy was reviewed in detail by Calabrese and colleagues, who reported that there is significant uncertainly due to a lack of animal and human data for this issue (17).

Biological basis

Etiology and pathogenesis

The cause of disease may be an occupational, accidental, or suicidal exposure to mercury compounds. Industrial uses of mercury, including the manufacture of batteries, latex paint, urethane, and polyvinyl chloride; fossil-fuel plants, incinerators, and municipal sewage systems as well as industrial discharge into rivers and streams may all lead to environmental pollution with mercury. Indeed, everyone is exposed to mercury in some form, whether from a source listed above or from thimerosal- or phenylmercuric nitrate-containing pharmaceuticals, fish consumption, or dental amalgams. Due to its accumulation in the food chain, chronic exposure to methylmercury via consumption of fish and sea mammals is still a major concern for human health, especially developmental exposure that may lead to neurologic alterations, including cognitive and motor dysfunctions (19). Thus, it is common to be able to detect mercury in blood, hair, and nails of normal people (44). Specific sources of accidental mercury exposure are quicksilver used in ethno-religious ceremonies common in some Hispanic and Caribbean communities, paint fumes from phenylmercuric acetate-containing interior latex paint, broken mercury-containing barometers or thermometers, Chinese herbal-ball preparations (27; 21), mercurial skin-lightening soaps and creams, spilled elemental mercury from pressure gauges, mercurials in gold recovery methods, ammoniated mercury ointments for the treatment of eczema or psoriasis (38; 58), teething compounds (85), diaper powders, and broken fluorescent light bulbs.

Following the outbreak of methylmercury intoxication in Minamata, extensive pathologic reports were published. These primarily involved the central nervous system, where prominent atrophy and degeneration of the cerebellar and calcarine cortex are seen (77). In the peripheral nerves of humans, axonal degeneration in the sural nerve and lumbar dorsal roots has been demonstrated (77). In experimental animal studies, the major pathologic change appears to be axonal degeneration without segmental demyelination. This has been shown in the L6 dorsal roots of rats but not in the ventral roots. The rate of axonal degeneration in the sural nerve is the same at mid-thigh and ankle level; therefore, the dorsal root ganglion cell is thought to be the primary site of action in the peripheral nervous system (54). Somjen and colleagues have demonstrated in rats that methylmercury administration resulted reduced the compound action potential evoked in dorsal roots, whereas the ventral root action potential was virtually normal (72; 73). This physiologic action was correlated with the finding that the major site of uptake of radioactive 203 Hg methylmercury in the peripheral nervous system was the spinal dorsal root ganglia. In a rat model study, both methylmercury and mercury vapor produced selective degeneration of dorsal root ganglia and dorsal nerve roots (68; 69). The number of autopsy cases for Minamata disease has risen (28). In one case autopsy, the dorsal roots and sural nerve showed endoneurial fibrosis, loss of nerve fibers, and the presence of Büngner bands. The spinal cord showed Wallerian degeneration of the fasciculus gracilis (Goll tract) with relative preservation of neurons in the sensory ganglia. These findings support the contention that peripheral nerve degeneration exists in patients with Minamata due to toxic injury from methylmercury.

Thus, in humans and animals, it appears that the major pathologic effect of methylmercury is on the dorsal root ganglion cells. In experimental models, it was found that the sensory neurons in the spinal ganglia and granule cells in the cerebellum were most vulnerable to mercury poisoning. Ultrastructural studies indicated that vacuolar degeneration of the neurons was mainly associated with inorganic mercury intoxication, whereas the coagulative type of degeneration was found mostly in organic mercury poisoning. Degenerative changes in the nerve fibers were also observed. Based on the biochemical, physiological, and pathological findings on mercury intoxication, a working hypothesis on the pathogenetic mechanism of mercury on the nervous system is proposed (20). No similar data are available for inorganic or metallic mercury poisoning. Because of the different chemical properties and clinical manifestations, these latter compounds likely exert their action at a different site. Three major mechanisms have been identified as critical in methylmercury-induced cell damage, including (i) disruption of calcium homeostasis, (ii) induction of oxidative stress via overproduction of reactive oxygen species or reduction of antioxidative defenses, and (iii) interactions with sulfhydryl groups (19).

Epidemiology

Epidemiological investigation was important in identifying environmental contamination with organic mercury compounds, like the Minamata Bay outbreak (45; 88). Mercury compounds should be considered possible causative agents in any outbreak of unknown neurologic disease if it involves prominent cerebellar findings.

Prevention

Effective treatment involves removing the source of exposure. Exposure to mercury vapor may occur accidentally in the home. This has occurred when liquid mercury is spilled, either by accident or in situations such as ethno-religious ceremonies. The danger is greatest when exposure occurs in poorly ventilated areas. The vapor is heavy, layering close to the floor. Infants and children, who breathe closest to the floor and often play on the ground, are at greatest risk (22). If there is any suspicion that mercury vapor exposure may occur, air levels should be monitored. Disposable monitors are available through public health departments. If toxic levels (greater than 0.1 mg/m³) are present, subjects should be removed to well-ventilated areas, and the mercury vapor source should be identified and removed.

Differential diagnosis

Mercury intoxication usually involves the subacute onset of mood and intellectual changes, followed by tremor and other signs suggestive of spinocerebellar involvement. There may also be significant constriction of visual fields. Differential diagnosis would include subacute cerebellar degeneration as a paraneoplastic syndrome, multiple sclerosis, and the cerebellar form of Creutzfeldt-Jakob disease. In long-term or chronic exposure with a slowly progressive course, mercury intoxication might resemble spinocerebellar degeneration. There is one case report of mercury intoxication initially mimicking amyotrophic lateral sclerosis (but with subsequent recovery) after a brief, intense exposure to elemental mercury (02). Mercury toxicity may resemble renal artery stenosis or rarer causes of secondary hypertension, such as pheochromocytoma or neurofibroma, because mercury binds and inactivates S-adenosylmethionine, an enzyme involved in catecholamine catabolism (10).

When considering whether a patient’s neuropathy arises from occupational or environmental toxicity, one should consider the following algorithm.

Table 1. Algorithm for Clinical Assessment of Neurotoxic Disease

Diagnostic workup

The diagnosis of mercury intoxication depends on the demonstration of an appropriate neurologic syndrome in the context of exposure of mercury or mercury compounds. If there are symptoms or signs of neuropathy, nerve conduction studies and needle electromyography should be performed. Typical findings are that of a sensorimotor axonal neuropathy; myokymia may sometimes be encountered. EMG needle examination may show chronic neurogenic changes, which are most notable in the intrinsic foot muscles (51). In the case of methylmercury intoxication, circumstantial evidence may be the only form available because the compound has such a high affinity for the nervous system that neurologic disease may be present in the absence of elevated urine mercury levels.

In the case of inorganic or metallic mercury intoxication, this can usually be confirmed by measurement of 24-hour urinary mercury excretion using flameless atomic absorption spectrophotometry or mass spectroscopy. Mercury is excreted slowly from the body. The biological half-life for mercury excretion differs depending on its form; ethyl mercury is estimated to have a half-life of 20 days, whereas inorganic divalent mercury, mercury vapor, and methylmercury are estimated to have half-lives of 40, 60, and 70 days, respectively (22). At the Mayo Clinic trace metal laboratory, the mean level for nonexposed individuals is 5 mcg per 24-hour period. Occupationally exposed individuals may excrete 25 to 30 mcg per 24-hour period; levels above 50 per 24 hours raise the question of toxicity related to mercury exposure. It is important to note that blood mercury may be best to assess organic mercury, whereas urine mercury best assesses inorganic or elemental mercury exposures.

Biological testing of serum and urine to assess absorbed dose may be compared to values published for these data. These data suggest that urine or blood levels are associated with the health effects of mercury exposure.

Table 2. ATSDR Biological Exposure Indices (BEIs)

When occupational exposure occurs, industrial hygiene data may be helpful. The Occupational Safety and Health Administration (OSHA), the National Institute of Safety and Health (NIOSH), and the American Congress of Governmental Industrial Hygienists (ACGIH) are useful resources from which to compare the workers’ inhalation exposure. Estimating dose and duration of exposure and level of protection afforded by personal protective equipment is emphasized. Government and professional organizations publish exposure limits for workers using various chemicals. Physicians may use this information to compare with industrial hygiene data (30; 46; 32; 63; 64; 65; 66).

Table 3. Published Limits for Mercury Exposure

Management

Supportive measures are discussed in Peripheral neuropathies: supportive measures and rehabilitation.

Treatment for mercury neurotoxicity includes chelating agents British anti-Lewisite, meso-2,3-dimercaptosuccinic acid (DMSA), D-penicillamine (DPCN), and oral 2,3 dimercapto 1 propanesulfonate (DMPS). These have been used in cases where dramatic neurotoxicity is confirmed from mercury exposure.

Animal experiments support a role for chelation in acute exposure cases and should be instigated promptly in such cases. Efficacy declines or disappears as the time interval between metal exposure and the onset of chelation increases (42). Side effects include hypersensitivity, renal effects, and aplastic anemia. Hemodialysis may also be used in cases where renal function is compromised. L-cysteine may significantly improve the ability of hemodialysis to remove circulating mercury. DMSA taken orally has been approved by the U.S. Federal Drug Administration (FDA) for the treatment of mercury poisoning (55), and it is the only chelator that has been approved by the FDA for treating mercury poisoning in children (39).

DMSA and DMPS are found to be less toxic and more efficient than British anti-Lewisite in the treatment of heavy metal poisoning and are available in capsules for oral use (01). These agents also do not redistribute mercury to the brain or accelerate excretion, but a decrease in morbidity and mortality is not clearly established for cases of chronic intoxication (42).

With inorganic mercury intoxication, for those who do not respond quickly to removal from the source of the exposure (31), chelation with oral DMPS has been dramatically helpful in individual and case series. Bradberry and colleagues reported reversal of neurologic damage with a 5-day course of DMPS chelation in a jeweler who presented with a 1-year progressive neurologic syndrome after 5 years of exposure to mercury vapor (14).

Chelation may be appropriate for patients with inorganic mercury intoxication who do not respond quickly to removal from the source of the exposure (31). Chelation is rarely accomplished unless toxicity is present from acute occupational exposure that is confirmed by an occupational and environmental medicine physician or toxicologist.

There are practitioners in many communities who perform chelation routinely on patients with vague symptoms, but the literature has not supported this treatment. Chelation was successful in treating one patient who attempted suicide with mercuric chloride (15; 83).

Special considerations

Pregnancy

Methylmercury inhibits migration and division of neurons and disrupts the cytoarchitecture of the developing brain, placing the fetal brain at greater risk of suffering from mercury toxicity (22). Prenatal methylmercury exposure, however, has not been linked to subsequent neuropathy. A developmental study of children in the Seychelles Islands whose mothers had methylmercury ingestion via fish consumption found no evidence of neurotoxicity related to this exposure (47). A pregnant woman who consumed pork from a pig fed methylmercury-treated seed grain gave birth to a child with signs of CNS mercury toxicity; peripheral neuropathy was not clearly described (23). Furthermore, elemental, inorganic, and organic mercury are excreted in breast milk; therefore, children who are breastfed by women exposed to mercury may also be exposed to potentially toxic amounts of mercury (32).