Mercury Toxicity Workup

Updated: Dec 12, 2016
  • Author: David A Olson, MD; Chief Editor: Tarakad S Ramachandran, MBBS, MBA, MPH, FAAN, FACP, FAHA, FRCP, FRCPC, FRS, LRCP, MRCP, MRCS  more...
  • Print

Approach Considerations

History and physical examination findings consistent with mercury poisoning are helpful, but blood, urine, and (sometimes) tissue analyses are required to confirm the diagnosis of mercury intoxication (although exact toxicity levels remain undefined).

Correlations have been found between signs, symptoms, and electrophysiologic studies of subjects exposed to mercury with various statistical extrapolations of measures of exposure, such as duration of exposure, peak urinary mercury levels, and estimated cumulative mercury dose.

Whole blood mercury levels are usually less than 2 mcg/dL in unexposed individuals, although individuals with a high dietary fish intake may be an exception.

Obtain a complete blood count (CBC) and serum chemistries to assess possible anemia secondary to GI hemorrhage, to determine if renal failure is present, and to rule out electrolytic abnormalities.

In most laboratories, mercury quantification is not performed on a routine basis; therefore, contact the laboratory to verify the specific collection and precautionary protocols before blood and urinary samples are collected. Reserve neuroimaging and electrophysiologic testing for selected cases. Consider pregnancy tests in women of childbearing age.


Occasional sural nerve biopsies have been performed on patients with mercury toxicity. Two cases of inorganic mercury poisoning revealed a combination of axonal and demyelinating changes. [72] Organic mercury toxicity in Minamata disease resulted in the preferential loss of large myelinated nerve fibers. [73]


Mercury Level Analysis

In the United States, based on the 2003 National Health and Nutrition Examination Survey (NHANES) data, urinary mercury levels of 5 mcg/L and blood mercury levels of 7.1 mcg/L encompassed 95% of the sample. These have been recommended as medically credible comparison levels. [74]


While blood levels are useful for more acute exposures, long-term exposures are best reflected in hair mercury measurements. Hair has high sulfhydryl content. Mercury forms covalent bonds with sulfur and, therefore, can be found in abundance in hair samples.

Because of environmental contamination, hair measurements have been problematic with elemental mercury exposure, but methylmercury hair measurements are considered accurate. [75] A hair value of 1.2 mcg/g encompassed 90% of the NHANES sample. [74]

Interestingly, investigators of Minamata disease identified chronic forms of the disease in which hair mercury levels were not elevated. A delayed neurotoxic effect, with symptoms emerging after age-induced neuronal loss, was hypothesized. [3] Similarly, some researchers have been unable to correlate the fluctuations of mercury blood levels with signs and symptoms of toxicity in mercury vapor exposure. [76]


Methylmercury concentrates in erythrocytes; therefore, mercury levels in blood remain high in acute toxicity. When ingested by humans, methylmercury is easily absorbed and retained by the body; it has a half-life in blood of about 44 days, which makes blood tests useful measures of acute exposure. [77]

The blood level correlation with chronic methylmercury toxicity is more variable. Methylmercury exhibits a blood-to-plasma ratio of up to 20:1, a characteristic of organic mercury. This higher ratio may be useful in determining if the patient was exposed to organic or inorganic mercurials. Aryl mercury compounds accumulate in RBCs but are metabolized to inorganic mercury more rapidly, thus, demonstrating lower blood-to-plasma ratios than those observed with methylmercury exposures. [78]

Following high exposure to inorganic mercury salts, the blood-to-plasma ratio ranges from a high of 2:1 to 1:1. Paraesthesias are expected if blood mercury levels are higher than 20 mcg/dL.

Inorganic mercury redistributes to other body tissue; thus, its levels in the blood are accurate only after an acute ingestion. In general, blood levels of mercury are helpful for recent exposures and for determining if the toxicity is secondary to organic or inorganic mercury, but they are not useful for a guide to therapy.


Urinary mercury levels are typically less than 10-20 mcg/L. Excretion of mercury in urine is a good indicator of inorganic and elemental mercury exposure but is unreliable for organic mercury (methylmercury) because this is eliminated mostly in the feces. In cases of chronic mercury toxicity, the urinary mercury measurement may be falsely low. [79]

No absolute correlation exists between urinary mercury levels and the onset of symptoms; however, levels higher than 300 mcg/L are associated with overt symptoms. Mercury levels in the urine also can be used to gauge the efficacy of chelation therapy, since chelated mercury is excreted primarily through the kidneys. For workers chronically exposed to mercury compounds, urinary excretion with mercury levels higher than 50 mcg/L is associated with an increased frequency of tremor.

Short-chained alkyl mercury compounds are excreted predominantly by the bile, rendering urinary measurements of these invalid. [79]

The position of the American College of Medical Toxicology (ACMT) does not support routine practice of postchallenge urinary metal testing, due to a lack of demonstrable benefits. This practice may be harmful if applied routinely in the assessment and treatment of patients suspected of having metal poisoning. [80]


Toenail mercury has also been used as a measure of long-term mercury exposure, with mean levels of 0.25-0.45 mcg/g among Western samples. Toenail mercury has been correlated with fish and shellfish consumption. [81, 82]

Cerebrospinal fluid

Cerebrospinal fluid (CSF) mercury concentrations have been measured with mass spectroscopy, and normal values vary widely. Nevertheless, increased CSF mercury levels have been found in workers with ongoing exposure to mercury vapors, but these CSF levels, unlike blood levels, normalize several months after such exposures have abated. [83]


Imaging Studies


Obtain a flat plate radiograph of the abdomen to visualize ingested elemental mercury, which appears radiopaque. (See the images below.)

This is a 1-view, abdominal, upright radiograph in This is a 1-view, abdominal, upright radiograph in a male patient who intentionally ingested 8 ounces of elemental mercury. Notice how the mercury outlines the large intestine from ascending to descending. Image courtesy of Fred P. Harchelroad, MD, and Ferdinando L. Mirarchi, DO.
Patient with intentional ingestion of mercury from Patient with intentional ingestion of mercury from blood pressure instrument. Note how mercury beads can be seen deposited in lung fields. Image courtesy of Shuchi Vyas, MD.


Neuroimaging is probably more helpful in excluding other diagnoses than in ruling in mercury toxicity. Nonetheless, magnetic resonance imaging (MRI) in cases of Minamata disease confirms the clinical and pathologic findings. Marked atrophy of the calcarine and parietal cortices, as well as the cerebellar folia, has been visualized. [84]

MRI findings in one patient with inorganic mercury toxicity revealed mild cortical atrophy and T2 hyperintensities in the frontal and subcortical regions. [85] Additionally, a 4-year-old with inorganic mercury poisoning developed transient fluid-attenuated inversion recovery (FLAIR) hyperintensities in cortical white matter during chelation therapy. [86]


Single-photon emission computed tomography (SPECT) demonstrated right cingulate hypermetabolism in a 38-year-old man with emotional lability and inattention following exposure to inorganic mercury. [87]



Electrophysiologic studies have demonstrated sensorimotor neuropathy, typically axonal, in some workers exposed to elemental mercury or mercury vapors. Workers with remote exposures, however, have exhibited only minimal conduction velocity slowing. [88] Abnormalities have also been documented in visual-evoked potential studies among workers exposed to mercury vapors. [76]

In the Faroe Islands, intrauterine methylmercury exposure (as determined by maternal hair and cord blood measures) was positively correlated with prolonged brainstem evoked potentials (III and V latency peaks) 14 years after initial exposure. [89]