Diabetes is a group of metabolic diseases known to cause hyperglycemia (high blood sugar levels).
The two primary types of diabetes, type 1 and type 2, are chronic diseases that impact how the body controls blood sugar or glucose.
Glucose is the fuel that feeds the body's cells, and it needs insulin to enter the cells.
People with type 1 diabetes cannot produce insulin. Meanwhile, people with type 2 diabetes do not respond to insulin as well as they should, and then, later in the disease, they often do not produce enough insulin.
The two types of diabetes can lead to chronically high blood sugar levels, which increases the risk of diabetes complications (1).
Over time, high blood sugar levels result in damage, dysfunction, and even failure of various organs, including the eyes, kidneys, nerves, heart, and brain.
Having diabetes raises one's risk of heart disease, stroke, kidney disease, and other health problems.
Having high blood pressure also increases this risk. If one has both diabetes and high blood pressure, these health conditions raise the risk of issues even more.
The initial studies on CBD for diabetes have been promising that, in 2006, Dr Joseph Alpert of the American Journal of Medicine called on the National Institute of Health (NIH) and the Drug Enforcement Administration (DEA) for more funding and collaboration on further research (2).
Insulin resistance is linked with chronic inflammation, as the study published in the Journal of Clinical Investigation indicated (3).
One risk factor for diabetes is obesity, and research published in the American Journal of Medicine in 2013 has shown that CBD has been associated with low levels of fasting insulin and reduced waist circumference (4).
Results of a study in Diabetes, Obesity and Metabolism suggested that the stimulus of cannabinoid receptors CB-1 in the islet (cluster) cells might be linked to insulin production (5).
Meanwhile, in another study published in Clinical Hemorheology and Microcirculation in 2016, CBD was shown to reduce the incidence of diabetes in non-obese diabetic animal models (6).
In another 2016 study, researchers from the University of Nottingham found that when CBD was used in conjunction with cannabis compound THCV, it helped increase insulin production and lower blood sugar in those with type 2 diabetes (7).
THCV (tetrahydrocannabivarin), as the name suggests, is a cannabinoid similar to THC in molecular structure. Both of these cannabinoids possess psychoactive properties.
The subjects in the said clinical study were required to either not receive any oral hypoglycemic medications or take stable doses of noninsulin glucose-lowering drugs, such as metformin and sulfonylurea, for three months prior to screening.
Results from the study also showed that CBD, when used alone, did not show any detectable metabolic effects.
With several studies that demonstrate CBD’s therapeutic benefits, its potential to help treat diabetes symptoms cannot be easily overlooked.
A growing interest in some people is making them look into this natural, non-psychoactive phytocannabinoid as a potential remedy for symptoms or conditions linked to diabetes.
Diabetic complications linked to the endocannabinoid system (ECS) include blindness, atherosclerosis, kidney failure, heart disease, and neuropathic pain (8).
Meanwhile, CBD’s anti-inflammatory properties might help in managing a diabetes patient’s condition by reducing inflammation.
In a study published in the American Journal of Physiology-Heart and Circulatory Physiology, CBD showed the potential to help reduce the arterial inflammation common in diabetes (9).
CBD might also reverse the harmful effects of high glucose levels, as a study published in Future Medicinal Chemistry demonstrated (10).
In his work with animal subjects, Dr Raphael Mechoulam and his colleagues noted that CBD not only blocked the onset of diabetes but the development of it as well (11).
A study demonstrated that CBD helped protect the retinas of diabetic animal subjects (12).
Diabetes can damage the nerves, resulting in pain from performing normal activities. Thus, people with diabetes often suffer from complications, such as chronic pain in the nerves, which can even be fatal.
However, one study in 2014 showed that CBD and tetrahydrocannabinol might help people suffering from this type of nerve pain (13).
Participants in the study applied a spray containing both cannabinoids to the areas of their bodies that had diabetic neuropathy. These areas are usually the hands or feet.
Study participants who had used the CBD and THC spray reported fewer symptoms before the spray.
More importantly, CBD can help with blood pressure, which increases the risk of having diabetes.
In a 2017 study published in the journal JCI Insight, results showed that a single dose of CBD reduced resting blood pressure and the blood pressure response to stress (14).
Previous studies looking at CBD as a means to alleviate diabetes symptoms have shown promising results. However, much of the research was conducted on animals or healthy human volunteers.
A better understanding of how CBD may be used to treat, manage, or prevent diabetes is needed. More extensive and more longitudinal studies, especially on humans with diabetes, or who are at risk of diabetes, are necessary.
Before using CBD as an adjunct diabetes therapy or using CBD alone as a remedy for any existing condition, consult with a doctor experienced in cannabis use.
|1.4 Identification numbers|
|1.4.1 CAS number|
|1.4.2 Other numbers|
|1.5 Brand names, Trade names|
|1.6 Manufacturers, Importers|
|2.1 Main risks and target organs|
|2.2 Summary of clinical effects|
|2.4 First aid measures and management principles|
|3. PHYSICO-CHEMICAL PROPERTIES|
|3.1 Origin of the substance|
|3.2 Chemical structure|
|3.3 Physical properties|
|3.3.1 Properties of the substance|
|3.3.2 Properties of the locally available formulation|
|3.4 Other characteristics|
|3.4.1 Shelf-life of the substance|
|3.4.2 Shelf-life of the locally available formulation|
|3.4.3 Storage conditions|
|3.4.5 Specific properties and composition|
|4.2 Therapeutic dosage|
|5. ROUTES OF ENTRY|
|6.1 Absorption by route of exposure|
|6.2 Distribution by route of exposure|
|6.3 Biological half-life by route of exposure|
|6.5 Elimination by route of exposure|
|7. PHARMACOLOGY AND TOXICOLOGY|
|7.1 Mode of action|
|7.2.1 Human data|
|7.2.2 Relevant animal data|
|7.2.3 Relevant in vitro data|
|7.7 Main adverse effects|
|8. TOXICOLOGICAL ANALYSES AND BIOMEDICAL INVESTIGATIONS|
|8.1 Material sampling plan|
|8.1.1 Sampling and specimen collection|
|126.96.36.199 Toxicological analyses|
|188.8.131.52 Biomedical analyses|
|184.108.40.206 Arterial blood gas analysis|
|220.127.116.11 Haematological analyses|
|18.104.22.168 Other (unspecified) analyses|
|8.1.2 Storage of laboratory samples and specimens|
|22.214.171.124 Toxicological analyses|
|126.96.36.199 Biomedical analyses|
|188.8.131.52 Arterial blood gas analysis|
|184.108.40.206 Haematological analyses|
|220.127.116.11 Other (unspecified) analyses|
|8.1.3 Transport of laboratory samples and specimens|
|18.104.22.168 Toxicological analyses|
|22.214.171.124 Biomedical analyses|
|126.96.36.199 Arterial blood gas analysis|
|188.8.131.52 Haematological analyses|
|184.108.40.206 Other (unspecified) analyses|
|8.2 Toxicological Analyses and Their Interpretation|
|8.2.1 Tests on toxic ingredient(s) of material|
|220.127.116.11 Simple Qualitative Test(s)|
|18.104.22.168 Advanced Qualitative Confirmation Test(s)|
|22.214.171.124 Simple Quantitative Method(s)|
|126.96.36.199 Advanced Quantitative Method(s)|
|8.2.2 Tests for biological specimens|
|188.8.131.52 Simple Qualitative Test(s)|
|184.108.40.206 Advanced Qualitative Confirmation Test(s)|
|220.127.116.11 Simple Quantitative Method(s)|
|18.104.22.168 Advanced Quantitative Method(s)|
|22.214.171.124 Other Dedicated Method(s)|
|8.2.3 Interpretation of toxicological analyses|
|8.3 Biomedical investigations and their interpretation|
|8.3.1 Biochemical analysis|
|126.96.36.199 Blood, plasma or serum|
|188.8.131.52 Other fluids|
|8.3.2 Arterial blood gas analyses|
|8.3.3 Haematological analyses|
|8.3.4 Interpretation of biomedical investigations|
|8.4 Other biomedical (diagnostic) investigations and their interpretation|
|8.5 Overall Interpretation of all toxicological analyses and toxicological investigations|
|9. CLINICAL EFFECTS|
|9.1 Acute poisoning|
|9.1.3 Skin exposure|
|9.1.4 Eye contact|
|9.1.5 Parenteral exposure|
|9.2 Chronic poisoning|
|9.2.3 Skin exposure|
|9.2.4 Eye contact|
|9.2.5 Parenteral exposure|
|9.3 Course, prognosis, cause of death|
|9.4 Systematic description of clinical effects|
|184.108.40.206 Peripheral nervous system|
|220.127.116.11 Autonomic nervous system|
|18.104.22.168 Skeletal and smooth muscle|
|9.4.7 Endocrine and reproductive systems|
|9.4.9 Eye, ear, nose, throat: local effects|
|22.214.171.124 Acid-base disturbances|
|126.96.36.199 Fluid and electrolyte disturbances|
|9.4.13 Allergic reactions|
|9.4.14 Other clinical effects|
|9.4.15 Special risks|
|10.1 General principles|
|10.2 Relevant laboratory analyses|
|10.2.1 Sample collection|
|10.2.2 Biomedical analysis|
|10.2.3 Toxicological analysis|
|10.2.4 Other investigations|
|10.3 Life supportive procedures and symptomatic/specific treatment|
|10.6 Antidote treatment|
|10.7 Management discussion|
|11. ILLUSTRATIVE CASES|
|11.1 Case reports from literature|
|11.2 Internally extracted data on cases|
|11.3 Internal cases|
|12. Additional information|
|12.1 Availability of antidotes|
|12.2 Specific preventive measures|
|14. AUTHOR(S), REVIEWER(S), DATE(S) (INCLUDING UPDATES), COMPLETE ADDRESS(ES)|
PHARMACEUTICALS 1. NAME 1.1 Substance
Insulin1.2 Group Antidiabetic agent1.3 Synonyms Amorph IZS Amorphous IZS Biphasic Insulin Cryst IZS Crystalline IZS Extended Insulin Zinc Suspension Globin Insulin Globin Insulin with zinc Globin Zinc Injection Insulin Globin Zinc Insulin GZI Injectio Insulini Protaminati cum zinco Insulin cum zinco (crystallisati) suspension Insulin Hydrochloride Insulin Injection Insulin lente Insulin semilente Insulin Ultralente Insulin zinc suspension (mixed) Insulin zinci crystallisati suspension injectabilis Insulini cum zinco (Amorphi) suspension injectabilis Insulini cum zinco suspensio composita Insulini Isophani Protaminati Suspension injectabilis Insulini Solution Injectabilis Insulini Zinci Injectabilis Mixta Insulini zinci protaminati injectio Insulini zinci protaminati suspension injectabilis PZI Isophane Insulin Isophane Insulin (NPH) Isophane Insulin Suspension Isophane Protamine Insulin Injection IZS Neutral Insulin NPH Insulin Ordinary Insulin Prompt Insulin Zinc Suspension Protamine zinc injection Regular Insulin Soluble Insulin Unmodified Insulin1.4 Identification numbers 1.4.1 CAS number 9004-10-81.4.2 Other numbers 53027-39-7 8049-62-5 8063-29-4 9004-17-5 9004-21-11.5 Brand names, Trade names Acid Insulin Injection1.6 Manufacturers, Importers Hypurin Soluble (CP) Regular Iletin (Lily USA) Highly purified Animal Insulins Actrapid MC (Novo UK, Favillon UK) Hypurin Neutral (Weddel UK) Neusulin (Wellcome UK) Nuso Neutral Insulin (Boots UK, Evans Medical UK, Wellcome UK) Velosulin (Nordisk UK, Leo UK) Velosulin Cartridge (Nordisk, Wellcome) Quicksol Boots Human Sequence Insulins Human Actrapid (Novo) Human Actrapid Purified (Novo) Human Velosulin (Nordisk, Wellcome) Humaline (Lily) Highly Purified Animal Insulins Insulin zinc suspension Lente (Evans) Hypurin Lente (CP) Lentard MC (Novo) Tempulin (Boots) Human sequence insulins Human monotard (Novo) Humulin Lente (Lily) Semitard MC (Novo) Human Sequence Insulins Human ultratard (Novo) Humulin zinc (Lily) Rapitard MC (Noro) Manufacturers - Boots, Evans Medical UK, Wellcome UK. Highly Purified Animal Insulins Isophane Insulin Injection (Evans) Hypurin Isophane (CP) Insulatard (Nordisk, Wellcome) Monophane (Boots) Mixed Highly Purified Animal Insulins Initard 50/50 (Nordisk, Wellcome) Mixtard 30/70 (Nordisk, Wellcome) Human Sequence Insulins Human Insulatard (Nordisk, Wellcome) Human Protaphane (Novo) Humulin (Lily) Mixed Human Sequence Insulins Human Actraphane (Noro) Human Initard 50/50 (Nordisk, Wellcome) Human Mixtard 30/70 (Nordisk, Wellcome) Humulin M1 (Lily) Humulin M2 (Lily) Humulin M3 (Lily) Humulin M4 (Lily) Hypurin Protamine Zinc (Weddel UK) Also Marketed in Great Britain by Boots, Wellcome, Evans Medical and Weddel Other Proprietary Names Protamine Zinc and Iletin Local agents2. SUMMARY 2.1 Main risks and target organs Insulin Hayleys (Boots) Morrison Son & Jones (Novo) Robert Hall & Co. (Nordisk) Hypoglycaemia is the main risk of insulin overdose. The brain relies on glucose as its source of energy and hypoglycaemia may lead to coma, convulsions and even death.2.2 Summary of clinical effects Hypoglycaemia:2.3 Diagnosis The early symptoms of hypoglycaemia are weakness, hunger, giddiness, pallor, sweating, sinking feeling in the stomach, palpitations, irritability, nervousness, headache and tremor. Symptoms resemble those of sympathetic stimulation. Later, symptoms such as depression or euphoria, inability to concentrate, blurring of vision, drowsiness, lack of judgement and self control and amnesia may be present due to neuroglycopenia. Other features are hemiplegia, ataxia, tachycardia, diplopia and paraesthesia. If untreated the condition progresses to convulsions, coma and death. In the precoma stage, Babinski reflex is often present. Pupils are often dilated but react to light. Later pupils are constricted and no longer react to light. Hypokalaemia may be present. Speed of onset of hypoglycaemia varies with the preparation of insulin used. Symptoms do not usually appear unless the blood glucose concentration falls below 3.5 mmol/l. Convulsions can occur if the blood glucose concentration falls below 2 mmol/l. In diabetic patients with chronic hyperglycaemia, symptoms of hypoglycaemia may occur at higher blood glucose concentrations. A few patients may develop hypoglycaemic coma without prior warning symptoms. Other effects: Non-specific local reactions such as allergic reactions, atrophy of fat or induration and hypertrophy sometimes occur at the site of injection (usually not a problem with highly purified insulins). Early symptoms include weakness, hunger, giddiness, pallor, sweating, sinking feeling in the stomach, palpitations, irritability, nervousness, headache and tremor.2.4 First aid measures and management principles Later, depression or euphoria, inability to concentrate, blurring of vision, drowsiness, lack of judgement and self control and amnesia occur. Other features are hemiplegia, ataxia, tachycardia, diplopia and paraesthesia. If untreated the condition progresses to convulsions, coma and death. In the precoma stage, Babinski reflex is often present. Pupils are often dilated but react to light. Later pupils are constricted and no longer react to light. Hypokalaemia may be present. Estimation of blood glucose level: (see 2.3.1) Estimation of plasma insulin is usually not relevant. The serum potassium concentration should be measured explicitly. Determination of urinary glucose and ketones for diabetic ketoacidosis. Serum creatinine, blood urea and serum electrolytes to asses renal function. If patient is conscious and cooperative - Oral glucose or 3-4 lumps of sugar should be given with water. This could be repeated in 15 minutes or earlier if symptoms recur. This should be supplemented by one or more carbohydrate meals until the patient improves.3. PHYSICO-CHEMICAL PROPERTIES 3.1 Origin of the substance If unconscious or uncooperative - Intravenous glucose or intramuscular glucagon should be given at once. Hospital admission comes later. 50ml of 50% dextrose should be given IV. NB - It is very important to advise patients and relatives regarding the prevention of hypoglycaemia. They should know the warning symptoms of hypoglycaemia. Correction of hypoglycaemia is the most important aspect of management. If consciousness is impaired even after correction of hypoglycaemia, cerebral oedema should be suspected. Cerebral oedema should be treated with mannitol and corticosteroids. Extracted from beta cells of the islets of Langerhans of pork or beef pancreas and purified by crystallisation.3.2 Chemical structure It is also made biosynthetically by recombinant DNA technology using Escherichia coli or semisynthetically by enzymatic modification of porcine material. Molecular weight: 60003.3 Physical properties 3.3.1 Properties of the substance Consists of two chains, A and B, of amino acids, joined together by two disulphide bonds. Insulin is synthesised from a single chain precursor named proinsulin. On conversion of human proinsulin to insulin, 4 basic amino acids and the remaining connector or C peptide is removed by proteolysis. The resultant insulin molecule has 2 chains. The acidic or A chain with glycine at the amino terminal residue and the basic or B chain consisting of 30 amino acids with phenylalanine at the amino terminus. An even larger molecule prepoinsulin has been identified as a precursor of proinsulin. Insulin can exist as a dimer, monomer or hexamer. Two molecules of Zn2+ are coordinated in the hexamer which is stored as granules in the beta cell. The biologically active form of the hormone is the monomer. The porcine hormone is most similar to man and differs only by the substitution of an alanine residue for threonine at the carboxy terminus of the B chain. Bovine insulin differs by three amino acids and therefore is more antigenic than porcine insulin. White or almost white crystalline powder. Slightly soluble in water. Practically insoluble in alcohol, chloroform and ether. Soluble in dilute solution of mineral acids and with degradation in solutions of alkali hydroxide.3.3.2 Properties of the locally available formulation Insulin Injection3.4 Other characteristics 3.4.1 Shelf-life of the substance This may be prepared by dissolving crystalline insulin containing not less than 23 units/mg in water for injections containing a suitable substance to render the injection iso-osmotic with blood; hydrochloric acid to adjust the pH to 3 to 3.5; and a suitable bactericide. The USP specifies sterile, acidified or neutral solution of insulin USP containing 40, 80, 100 or 500 units per ml as well as 1.4 - 1.8% w/v of glycerol and 0.1 - 0.25% (W/V) of phenol or cresol. pH of acidified injection 2.5 - 3.5 pH of neutral injection 7 - 7.8 Insulin injection is a colourless injection or straw coloured liquid practically free from solid matter which deposits on standing. Contains not more than 40 g zinc/100 units of insulin. Sterilised by filtration and kept in multidose containers. Neutral insulin BP Sterile buffered solution of bovine or porcine insulin of potency not less than 23 units/mg; pH 6.6 - 8 colourless liquid. Contains not more than 20 g zinc/100 units of insulin and a suitable bactericide. Available in multidose containers. Insulin Zinc Suspension BP Sterile buffered suspension of mammalian insulin in the form of a complex obtained by addition of zinc chloride. Insulin is in a form insoluble in water. Prepared by mixing 3 volumes of insulin zinc suspension (amorphous) and 7 volumes of insulin zinc suspension (crystalline). Contains 40, 80, or 100 units/ml. White suspension available in multidose containers. pH 6.9 - 7.5. Complies with a test for prolongation of insulin effect. Insulin Zinc Suspension BP (Amorphous) Sterile buffered suspension of mammalian insulin in the form of a complex obtained by addition of zinc chloride. Prepared fromcrystalline insulin containing not less than 23 u almost colourless suspension in which the particles have no uniform shape and rarely exceed 2 m in dimension; pH 6.9 - 7.5. Iso-osmotic with blood. Containing suitable bactericide, the preparation contains 40 and 80 units/ml. U.S.P. describes a sterile suspension of insulin U.S.P. in buffered water for injection is modified by addition of zinc chloride so that the solid phase of suspension is amorphous. Contains 40, 80 or 100 units/ml. Also contain sodium acetate 0.15 - 0.17%, sodium chloride 0.65 - 0.75%, methyl hydroxy benzoate 0.09 - 0.11% and for each 100 units of insulin, 120 - 250 g of zinc. pH 7.2-7.5 Insulin zinc suspension (crystalline) BP Sterile buffered suspension of bovine insulin to which zinc chloride is added. Crystalline form is insoluble in water. Prepared from crystalline insulin containing not more than 23 units/mg. White or almost colourless suspension. Particles are mainly crystalline. Majority of crystals having a maximum diameter greater than 10 m. pH 6.9 - 7.5 Iso-osmotic with blood. Preparation contains 40 and 80 units/ml. U.S.P. - Sterile suspension of insulin contain 40, 80, 100 units/ml. Contains sodium acetate, sodium chloride and methyl hydroxybenzoate (Concentration same as for amorphous insulin) and zinc 120 - 250 ug. pH 7.2 - 7.5. Biphasic Insulin BP Sterile buffered suspension of crystals of bovine insulin containing not less than 23 units/mg. in a solution of porcine insulin of similar potency. White suspension. pH 6.6 - 7.2 iso-osmotic with blood. Contained 27.5 - 37.5 ug zinc for each 100 units/insulin. Quarter of insulin in soluble form. Multidose glass container. Globin zinc Insulin BP Sterile preparation of mammalian insulin in the form of a complex obtained by addition of suitable globin and zinc chloride. USP specification: Insulin modified by addition of zinc and globin obtained from beef blood, 40, 80, 100 units/ml. Colourless liquid pH 3 - 3.8; iso-osmotic with blood. Each 100 units of insulin also contains 3.6 - 4 mg of globin and 250 - 350 g zinc. USP specification: Also contains as preservatives phenol, glycerol and cresol. Multidose container. Isophane Insulin Sterile buffered suspension of insulin in the form of a complex obtained by addition of suitable protamine. Prepared from crystalline insulin. pH 6.9 - 7.5 iso-osmotic with blood. Contains for each 100 units of insulin, 300 - 600 g protamine sulphate and not more than 40 g zinc, a suitable bactericide and sodium phosphate as buffering agent. USP specification: Sterile suspension of zinc insulin crystalline and protamine sulphate in buffered water for injection. Solid phase contain crystals of insulin protamine and zinc; 40, 80, 100 units/ml. Contains glycerol, metacresol, phenol sodium phosphate and zinc. Protamine Zinc Insulin Sterile buffered suspension of mammalian insulin to which protamine and zinc chloride are added. White suspension. pH 6.9 - 7.5 iso-osmotic with blood. Contains for each 100 units of insulin, 1 - 1.7 mg of protamine sulphate and zinc chloride equivalent to 200 per g of zinc, 10 - 11 mg of sodium phosphate. USP specification: Buffered sterile suspension to which zinc chloride and protamine sulphate are added. 40, 80, 100 units/ml. Also contain glycerol, cresol, phenol, sodium phosphate for each 100 units of insulin in addition to protamine and zinc. Up to 2 years at storage conditions of 2° - 8°C3.4.2 Shelf-life of the locally available formulation Up to 2 years at storage conditions of 2° - 8°C3.4.3 Storage conditions Store at 2° - 8°C. Avoid freezing.3.4.4 Bioavailability To be added by PCC using the monograph3.4.5 Specific properties and composition To be added by PCC using the monograph4. USES 4.1 Indications (a) Diabetes mellitus4.2 Therapeutic dosage 4.2.1 Adults (b) Complications of diabetes mellitus eg. Hyperglycaemic ketoacidotic coma Hyperglycaemic hyperosmolar non-ketotic coma (c) Hyperkalaemia (d) Insulin hypoglycaemia can be used as a test of anterior pituitary function and to test completeness of vagotomy in reducing gastric secretion. Blood sugar < 16.5 mmol/l (300 mg%): 20 units.4.2.2 Children Blood suger 11 - 16.5 mmol/l (200 - 300 mg%): 10 units. Dose is than adjusted according to the usual monitoring of blood and/or urine glucose. Daily dose increments should be 4 units. When stabilised, two-thirds of the daily dose is generally given 30 minutes before breakfast and one-third 30 minutes before the evening meal.4.3 Contraindications If only one injection per day is required, 10 - 14 units of an intermediate-acting insulin can be given. Dose increment is 4 units given on alternate days. Soluble or neutral may be added or special mixed insulins used according to the patient's response. Absolute Hypoglycaemia5. ROUTES OF ENTRY 5.1 Oral Relative Allergic reactions may occur to beef or porcine insulins. Precautions: Differing immunological response to bovine and porcine insulin have been reported and hypoglycaemia has been reported in patients changing from bovine to porcine insulin. Care is recommended to avoid inadvertant change of insulin from one species to another. Care may be necessary in changing over to a highly purified insulin. The hypoglycaemia caused by insulin may be enhanced by alcohol, monoamine oxidase inhibitors and propranolol and other beta blockers. Propranolol may mask the symptoms of hypoglycaemia. NB - Any deterioration in renal function and severe hepatic disease may reduce insulin clearance and may result in hypoglycaemia. When taken orally insulin has no hypoglycaemic effect since it is inactivated in the gastrointestinal tract (Reynolds, 1982)5.2 Inhalation Not relevant5.3 Dermal Not relevant5.4 Eye Not relevant5.5 Parenteral Insulin is administered by subcutaneous, intramuscular or intravenous injections and poisoning can occur only through this route.5.6 Other Not relevant6. KINETICS 6.1 Absorption by route of exposure Insulin must be injected SC, IM or IV. It is absorbed into the blood and peak plasma insulin concentration with subcutaneous insulin occurs at 60 - 90 min. Absorption is slower if there is peripheral vascular disease or smoking, and faster if the patient is vasodilated eg. by a hot bath or ultraviolet exposure or exercise.6.2 Distribution by route of exposure Any changes in mode of administration either accidentally (eg. accidental IM or IV injection) or deliberately (eg. constant subcutaneous insulin infusion) may potentiate the absorption and action of insulin, leading to hypoglycemia. Severe hypoglycaemia may occur during constant infusion and several deaths have been reported (Paterson et al, 1983). Absorption of insulin from injection site affected by insulin lipodystrophy is very unpredictable and rapid absorption may lead to hypoglycaemia. A fraction of endogenous or exogenous insulin in plasma may be associated with certain proteins but the bulk appears to circulate in blood and lymph as the free hormone. The volume of distribution of insulin approximates the volume of extrecellular fluid.6.3 Biological half-life by route of exposure Insulin is inactivated in the liver and kidneys (about 40% in a single passage). About 10% appear in the urine. The plasma half-life is:6.4 Metabolism intravenous injection 10 minutes subcutaneous injection 4 hours intramuscular injection 2 hrs. Metabolism occurs mainly in the liver and kidneys; 10% of the dose appears in the urine. Insulin is normally filtered at the glomeruli and then completely reabsorbed or destroyed at the proximal tubule. In patients with impaired renal tubular function, urinary clearance approaches glomerular filtration rates. 50% of insulin that reaches the liver via the portal vein is destroyed in a single passage, never reaching the general circulation. Proteolytic degradation of insulin occurs both at cell surfaces and in the lysosomes. A proteolytic enzyme that degrades insulin has been purified from muscle. An enzyme, glutathione insulin transhydrogenase, which utilises reduced glutathione to reduce disulfide bridges of insulin and produce separate chains, has been implicated.6.5 Elimination by route of exposure Severe impairment of renal function appears to affect the rate of disappearance of circulating insulin to a greater extent than does hepatic disease. About 10% of the drug appears in urine.7. PHARMACOLOGY AND TOXICOLOGY 7.1 Mode of action 7.1.1 Toxicodynamics Insulin in overdose causes hypoglycaemia. Hypoglycaemia deprives the brain of substrate glucose upon which it is almost exclusively dependent for its oxidative metabolism. During insulin coma, oxygen consumption in the human brain decreases by nearly half.7.1.2 Pharmacodynamics A prolonged period of hypoglycaemia causes irreversible damage to the brain as evidenced in experimental animals by histological changes in the cortex, basal ganglia and rostral parts of the medulla. Convulsions, coma, mental retardation, hemiparesis, ataxia, incontinence, aphasia, choreiform movements, and parkinsonism may occur in man. Insulin binds to a receptor on the surface of the target cell and probably also enters the cell in this state. The receptors vary in number inversely with the insulin concentration to which they are exposed.7.2 Toxicity 7.2.1 Human data 188.8.131.52 Adults The receptor becomes phosphorylated on addition of insulin and ATP. The cellular mechanism of action of insulin after combination with the receptor is uncertain; the complex may activate a "second messenger" which in turn causes the release of third messenger Ca++ ions. Insulin also has a membrane effect in increasing gluose uptake and utilization, especially by muscle and adipose tissue. Its effects include the following: a) Reduction in blood sugar due to increased glucose uptake in the peripheral tissues which convert it to glycogen or fat, and reduction of hepatic output (diminished breakdown of glycogen and diminished gluconeogenesis). When blood glucose falls below renal threshold (180 mg/100 ml or 10 mmol/l) glycosuria ceases as does the osmotic diuresis of water and electrolytes. Polyuria and excessive thirst are thus alleviated. If blood glucose falls much below normal levels, appetite is stimulated. b) Other metabolic effects: Insulin stimulates the transit of amino acids and potassium into the cells. Insulin regulates utilization of carbohydrate and energy products and enhances protein synthesis. c) Insulin increases concentrations of the active form of the enzyme pyruvate dehydrogenase. Hence pyruvate is oxidized or converted to fat and is unavailable for glucose formation. In addition to enhanced synthesis of fat, insulin increases the activity of membrane bound lipoprotein lipase which makes fatty acids derived from circulating lipoproteins available to the cell. In most normal adults, 0.1 - 0.2 Units/kg intravenously is sufficient to cause profound hypoglycaemia. However, in insulin dependent diabetics, it is not possible to indicate the amount of insulin necessary to cause toxicity because the level of hypoglycaemia and its duration are the important factors. For example, patients who have taken 80 to 500 times of the normal dose taken for suicidal purposes have recovered (Martin et al, 1977).184.108.40.206 Children No data available.7.2.2 Relevant animal data No data available.7.2.3 Relevant in vitro data No data available.7.3 Carcinogenicity Development of cancer at the site of long term insulin injection has been reported. (Eisenbad and Walter, 1975).7.4 Teratogenicity Congenital malformations occurred in 17 of 117 babies born to diabetic mothers taking insulin at the time of conception (Pay and Insley, 1976).7.5 Mutagenicity No data available.7.6 Interactions Alcohol, beta blockers, salicylates, oxytetracycline and monoamine oxidase inhibitors potentiate the hypoglycaemic effects of insulin. Bezafibrate and clofibrate may improve glucose tolerance and have an additive effect. Corticosteroids, corticotrophin, diazoxide, diuretics like bumetanide, furosemide and thiazides and oral contraceptives antagonise the effects of insulin. Lithium may occasionally impair glucose tolerance.7.7 Main adverse effects (a) Hypoglycaemia - symptoms are directly related to duration and depth of hypoglycaemia. Initial sympathetic overactivity is followed by signs of neuroglycopenia.8. TOXICOLOGICAL ANALYSES AND BIOMEDICAL INVESTIGATIONS 8.1 Material sampling plan 8.1.1 Sampling and specimen collection 220.127.116.11 Toxicological analyses (b) Non specific local reactions at site of injection eg. pain, oedema. (c) Allergic reactions (d) Lipoatrophy or induration and hypertrophy at the site of injection is asociated with chronic use. (e) Insulin resistance. Type of Plasma Insulin a) Bioassay18.104.22.168 Biomedical analyses 22.214.171.124 Arterial blood gas analysis 126.96.36.199 Haematological analyses 188.8.131.52 Other (unspecified) analyses 8.1.2 Storage of laboratory samples and specimens 184.108.40.206 Toxicological analyses 220.127.116.11 Biomedical analyses 18.104.22.168 Arterial blood gas analysis 22.214.171.124 Haematological analyses 126.96.36.199 Other (unspecified) analyses 8.1.3 Transport of laboratory samples and specimens 188.8.131.52 Toxicological analyses 184.108.40.206 Biomedical analyses 220.127.116.11 Arterial blood gas analysis 18.104.22.168 Haematological analyses 22.214.171.124 Other (unspecified) analyses 8.2 Toxicological Analyses and Their Interpretation 8.2.1 Tests on toxic ingredient(s) of material 126.96.36.199 Simple Qualitative Test(s) 188.8.131.52 Advanced Qualitative Confirmation Test(s) 184.108.40.206 Simple Quantitative Method(s) 220.127.116.11 Advanced Quantitative Method(s) 8.2.2 Tests for biological specimens 18.104.22.168 Simple Qualitative Test(s) 22.214.171.124 Advanced Qualitative Confirmation Test(s) 126.96.36.199 Simple Quantitative Method(s) 188.8.131.52 Advanced Quantitative Method(s) 184.108.40.206 Other Dedicated Method(s) 8.2.3 Interpretation of toxicological analyses 8.3 Biomedical investigations and their interpretation 8.3.1 Biochemical analysis 220.127.116.11 Blood, plasma or serum 18.104.22.168 Urine 22.214.171.124 Other fluids 8.3.2 Arterial blood gas analyses 8.3.3 Haematological analyses 8.3.4 Interpretation of biomedical investigations 8.4 Other biomedical (diagnostic) investigations and their interpretation 8.5 Overall Interpretation of all toxicological analyses and toxicological investigations 8.6 References 9. CLINICAL EFFECTS 9.1 Acute poisoning 9.1.1 Ingestion b) Immunoassay Plasma insulin estimated by bioassay is called insulin-like activity (ILA) and plasma insulin estimated by immunoassay is called immunoreactive insulin (IRI); ILA and IRI differ qualitatively and quantitatively. In vivo bioassays depend on lowering of blood glucose in rabbits or production of convulsions in mice. Several in vitro methods have become popular due to high degrees of sensitivity and relative simplicity of execution. One method is based on the capacity of insulin to increase the glycogen content and glucose uptake of rat diaphragm. Adipose tissue assays are based on the capacity of insulin to stimulate glucose metabolism by the epididymal fat pad of the rat or by a suspension of isolated fat cells. The sensitivity of radio-immunoassay is greater than that of other assays. The following could be determined by radio- immunoassay. a. Total immunoreactive insulin (IRI) b. NEIRI - nonextracted immunoreactive plasma insulin c. Free plasma insulin ('Free' IRI) d. I125 insulin binding in vitro e. Acute insulin sensitivity (KITT) f. Half-time of immunoreactive insulin disappearance (Gilman and Goodman, 1985; Martin et al, 1977). Blood glucose estimation Concentration of plasma IRI of normal persons after an overnight fast is under 20 microunits/ml. Not relevant9.1.2 Inhalation Not relevant9.1.3 Skin exposure Not relevant9.1.4 Eye contact Not relevant9.1.5 Parenteral exposure 0.4 units of monocomponent insulin injected subcutaneously three times a day into each quadrant of a pitted scar did not cause hypoglycaemia (Amroiwalla, 1977). Insulin toxicity causes hypoglycaemic symptoms.9.1.6 Other 9.2 Chronic poisoning 9.2.1 Ingestion Not relevant9.2.2 Inhalation Not relevant9.2.3 Skin exposure Not relevant9.2.4 Eye contact Not relevant9.2.5 Parenteral exposure Not relevant9.2.6 Other Not relevant9.3 Course, prognosis, cause of death Symptoms do not usually appear unless the blood glucose concentration falls below 3.5 mmol/l. Convulsions can occur if the blood glucose concentration falls below 2 mmol/l. Mild hypoglycaemia is usually relieved rapidly by ingestion of carbohydrates.9.4 Systematic description of clinical effects 9.4.1 Cardiovascular Even when the level of consciousness is impaired, parenteral admiistration of glucose or glucagon will lead to full recovery. Sometimes brain damage is irreversible. Severe or prolonged neuroglycopenia may respond only slowly to restoration of the plasma glucose concentration as concomitant cerebral oedema may itself depress the level of consciousness. Prolonged hypoglycaemia may cause irreversible cerebral damage as manifested by chronically impaired cognitive functions, convulsions and hemiparesis, eventually causing death. Acute - Hypoglycaemia promotes the release of adrenalin which causes palpitations and tachycardia. When hypoglycaemia is severe, angina, arrhythmias, premature beats and coronary thrombosis may occur.9.4.2 Respiratory Shallow breathing is present in severe, prolonged hypoglycaemic coma.9.4.3 Neurological 126.96.36.199 CNS Signs of neuroglycopenia due to prolonged hypoglycaemia include intellectual impairment, irrational behaviour, blurring of vision, diplopia, headache, confusion, abnormal and often aggressive behaviour, mental retardation, hemiparesis, ataxia, incontinence, aphasia, choreiform movements, parkinsonism, epilepsy, convulsions, drowsiness and coma.188.8.131.52 Peripheral nervous system Numbness of lips, nose or fingers.184.108.40.206 Autonomic nervous system Initial sympathetic overactivity is observed due to hypoglycaemia as a result of the release of adrenaline.220.127.116.11 Skeletal and smooth muscle 9.4.4 Gastrointestinal Symptoms include anxiety, hunger, sweating, weakness, pallor, palpitations and tremors. No data available.9.4.5 Hepatic No data available.9.4.6 Urinary 18.104.22.168 Renal No data available.22.214.171.124 Other Urinary incontinence.9.4.7 Endocrine and reproductive systems No data available.9.4.8 Dermatological Cutaneous bullae occur due to prolonged coma (similar to those associated with barbiturate poisoning).9.4.9 Eye, ear, nose, throat: local effects Lipodystrophy - Fat atrophy and hypertrophy usually occur with prolonged use of non purified insulins. Eye - Pupils are initially dilated and react to light. Later pupils are constricted and cease to react to light.9.4.10 Haematological Hypercoagulation has been reported.9.4.11 Immunological Immediate hypersensitivity: this occurs most frequently during first few weeks of therapy. Following an injection a 'wheal and flare' response occurs around injection site within 2 hours. Local swelling is maximal at 6 - 12 hrs but the reaction usually settles within 24 - 48 hrs.9.4.12 Metabolic 126.96.36.199 Acid-base disturbances Rarely, a widespread general hypersensitivity reaction with widespread urticaria, gastrointestinal disturbances and angioneurotic oedema will occur. Immediate hypersensitivity is mediated by IgE antibodies. Delayed hypersensitivity may occur shortly after introducing insulin therapy. Erythema occurs around injection site 2 - 24 hrs after injection and settles slowly over 2 - 3 days. Rarely, local residual scarring may occur. Type IV hypersensitivity is mediated by lymphocytes. These reactions subside spontaneously and are rarely generalized but occasionally they may persist and necessitate a change in insulin therapy. Most cases occur with non-highly purified bovine insulin. Acute anaphylaxis to insulin is extremely rare. Hypoglycaemic coma may lead to respiratory depression and anoxia with consequent acidosis.188.8.131.52 Fluid and electrolyte disturbances Hyperinsulinaemia may itself cause hypokalaemia and hypoglycaemia provokes the release of adrenalin, which may also cause hypokalaemia184.108.40.206 Others 9.4.13 Allergic reactions see 220.127.116.11.4.14 Other clinical effects 9.4.15 Special risks Hypoglycaemia, severe in 17 cases, occurred during the first 6 hours of life in 22 of 34 infants born to diabetic mothers who received insulin. Clinical features were present only in two (Martin et al, 1975).9.5 Other Pregnancy: It is unclear whether insulin causes congenital malformations, though diabetes certainly does. Neonatal hypoglycaemia is caused by neonatal pancreatic hyperstimulation when diabetic mothers have been hyperglycaemic. Alcohol is an important risk factor - and most cases of death from insulin have been associated with alcohol. No data available.9.6 Summary 10. MANAGEMENT 10.1 General principles Correction of hypoglycaemia and the maintenance of normal blood glucose concentration are the most important aspects in management. In conscious, cooperative patients, oral therapy with a few lumps of sugar or glucose is adequate.10.2 Relevant laboratory analyses 10.2.1 Sample collection Patients who have large subcutaneous depots of insulin continue to absorb it over several days, and may need glucose infusions. Venous blood glucose concentration is measured on a fresh sample taken into a tube containing fluoride- acetate. A volume of 1-2 ml of whole blood is sufficient. If measurement is likely to be delayed, the sample should be stored at 4°C, though even at this temperature there is some glucose consumption. Where assay within a few hours is impossible, a reliable result can be obtained by separating the plasma and then freezing it at -20°C.10.2.2 Biomedical analysis Diagnosis of hypoglycaemia is suggested by a low glucose oxidase strip reading. Insulin assay is not helpful in patients with antibodies, unless free insulin concentration can be measured. The blood glucose concentration should be measured. Hypokalaemia can be detected by assay of serum potassium.10.2.3 Toxicological analysis Refer to section 18.104.22.168 Other investigations 10.3 Life supportive procedures and symptomatic/specific treatment If the patient is conscious and cooperative, give oral glucose or sucrose (3-4 sugar lumps) and repeat after 15 minutes or earlier if symptoms recur. This should be supplemented by a carbohydrate meal until blood sugar is stable.10.4 Decontamination If the patient is unconscious or uncooperative, give 50 ml of 50% dextrose (25 g of glucose) IV. If there is no response within 15 minutes repeat the dose; glucagon 1 mg IM or IV or SC is a less reliable alternative. This can also be repeated but failure of the first glucagon injection is unlikely to be followed by success with further doses. For children give approximately 1 ml/kg of 50% glucose and repeat if there is no response. A 10% glucose infusion via a peripheral vein may be enough, but commonly 20% glucose has to be given by the central vein. If the patient is still unconscious despite this treatment and blood sugar is normal or above normal then cerebral oedema is likely and should be treated with 20% mannitol. (Unless contraindicated due to cardiovascular disease). Blood glucose concentration should be measured every 15 - 30 minutes and the rate of infusion altered to keep the blood glucose concentration within the range of 5 - 10 mmol/l. Give dexamethasone 10 mg IV followed by 16 mg daily in four divided doses. Give oxygen. Treatment with parenteral dextrose at doses up to 25 g/hr may be necessary for protracted periods, depending on the preparation of insulin injected. Initially, 1% - 20% solution via a large central vein may be necessary to supplement oral carbohydrate ingestion. Occasionally neurological deficits recover after several days if full supportive care is maintained. Frequent blood sugar estimations (venous or capillary) may be necessary, particularly initially, according to the clinical condition. Plasma insulin estimations do not help in the management. Treat hypokalaemia with potassium supplements. Intravenous potassium chloride 20 - 60 mEq per litre of fluid may be given. Not relevant10.5 Elimination Not relevant10.6 Antidote treatment 10.6.1 Adults No specific antidotes are available10.6.2 Children No specific antidotes are available10.7 Management discussion Glucagon 1 mg IM is useful to correct hypoglycaemia. However, there is a risk that glucagon-induced insulin secretion may be a complication (Marri et al, 1968).11. ILLUSTRATIVE CASES 11.1 Case reports from literature Excision of the skin and fat of an insulin injection site under a local anaesthetic has been performed in the management of insulin overdose (McIntyre et al, 1986). Hypoglycaemia may develop later than predicted from the duration of action of various insulin preparations (Haskell and Stapezyriski, 1983). Kaminer and Robbins (1988) reported a case of a 16 year old girl with insulin-dependent diabetes mellitus who gave herself an injection of 600 units of regular insulin in a suicidal attempt. She lost consciousness for 12 hours and was found confused and disoriented. She recovered fully.11.2 Internally extracted data on cases 11.3 Internal cases 12. Additional information 12.1 Availability of antidotes 12.2 Specific preventive measures Martin et al (1977) reported a case of a young man with a prior history of depressive illness who was found unconscious about 12 hours after self-administration of about 1,600 units of NPH insulin along with a large amount of alcohol and barbiturates. Despite the intravenous administration of a large amount of glucose and 20% fructose solution, hypoglycaemia recurred and his conscious state deteriorated after an episode of respiratory obstruction. He did not require further insulin for 6 days by which time the blood glucose level had risen to 15 mmol/l. Improvement in his conscious state was slow and he had evidence of marked mental impairment and emotional lability. Patients should be advised regarding the correct dose and injection technique. They should have an adequate knowledge of early features of hypoglycaemia so that they can take glucose or sugar immediately.12.3 Other 13. REFERENCES Amroliwalla FK (1977). Br Med J 1: 1389 - 9014. AUTHOR(S), REVIEWER(S), DATE(S) (INCLUDING UPDATES), COMPLETE ADDRESS(ES) British National Formulary (1988). British Medical Association and the Pharmaceutical Society of Great Britain. Dukes MNG (1988). Meyler's Side Effects of Drugs. Amsterdam, Elsevier Scientific Publishers. Eisenbad E, Walter RM (1975). J Am Med Ass 233: 985. Gilman AG, Goodman LS, Rall TW, Murad F (1985) ed. In: The Pharmacological Basis of Therapeutics 7th Ed. Pergamon. p 1490 - 1504. Haskell RJ, Stapezynski JS (1983). Intravenous glucose for the treatment of intentional insulin overdoses. Ann Emerg Med 12: 260. Kaminer Y, Robbins DR (1988). Attempted suicide by insulin overdose in insulin-dependent diabetic adolescents. Paediatrics 81: 526 - 528. Laurence DR, Bennett PN. Clinical Pharmacology. Churchill Livingstone. Edinburgh. McIntyre AS, Woolf VJ, Burnhem WR (1986). Local excision of subcutaneous fat in the management of insulin overdose. Br J Surg 73: 538. Martin FIR, Hansen N, Warne GL (1977). Attempted suicide by insulin overdose in insulin-requiring diabetics. Med J Austr 1: 58-60. Martin FIR et al (1975). Arch Dis Child 130: 998 Paterson KR, Paice BJ, Lawson DH, (1983). Undesired effects of insulin therapy. Adv Drug React Ac Pois Rev 2: 219-234 Pay IR, Insley J (1976). Arch Dis Child 51: 935 Reynolds JEF (1982) ed. Martindale, The Extra Pharmacopoeia. 28th ed. The Pharmaceutical Press, London. p. 2025. Trevor M. Avery's Drug Treatment. 3rd ed. ADIS Press. Auckland. p. 530-535 Author(s):Dr Ravindra Fernando Dr (Mrs) Geetha Fernando National Poisons Information Centre General Hospital Colombo 8 Sri Lanka Tel: 94-1-94016 Fax: 94-1-599231 Date: April 1990 Reviewer: Dr R. Ferner West Midlands Poisons Unit Dudley Road Hospital Birmingham B18 7QH United Kingdom Tel: 44-21-5543801 Fax: 44-21-5236526 Date: February 1991 Peer review: Adelaide, Australia, April 1991
INTOX Home Page
INTERNATIONAL PROGRAMME ON CHEMICAL SAFETY
ENVIRONMENTAL HEALTH CRITERIA 61
This report contains the collective views of an international group of
experts and does not necessarily represent the decisions or the stated
policy of the United Nations Environment Programme, the International
Labour Organisation, or the World Health Organization.
Published under the joint sponsorship of
the United Nations Environment Programme,
the International Labour Organisation,
and the World Health Organization
World Health Orgnization
The International Programme on Chemical Safety (IPCS) is a
joint venture of the United Nations Environment Programme, the
International Labour Organisation, and the World Health
Organization. The main objective of the IPCS is to carry out and
disseminate evaluations of the effects of chemicals on human health
and the quality of the environment. Supporting activities include
the development of epidemiological, experimental laboratory, and
risk-assessment methods that could produce internationally
comparable results, and the development of manpower in the field of
toxicology. Other activities carried out by the IPCS include the
development of know-how for coping with chemical accidents,
coordination of laboratory testing and epidemiological studies, and
promotion of research on the mechanisms of the biological action of
ISBN 92 4 154261 6
The World Health Organization welcomes requests for permission
to reproduce or translate its publications, in part or in full.
Applications and enquiries should be addressed to the Office of
Publications, World Health Organization, Geneva, Switzerland, which
will be glad to provide the latest information on any changes made
to the text, plans for new editions, and reprints and translations
(c) World Health Organization 1988
Publications of the World Health Organization enjoy copyright
protection in accordance with the provisions of Protocol 2 of the
Universal Copyright Convention. All rights reserved.
The designations employed and the presentation of the material
in this publication do not imply the expression of any opinion
whatsoever on the part of the Secretariat of the World Health
Organization concerning the legal status of any country, territory,
city or area or of its authorities, or concerning the delimitation
of its frontiers or boundaries.
The mention of specific companies or of certain manufacturers'
products does not imply that they are endorsed or recommended by the
World Health Organization in preference to others of a similar
nature that are not mentioned. Errors and omissions excepted, the
names of proprietary products are distinguished by initial capital
ENVIRONMENTAL HEALTH CRITERIA FOR CHROMIUM
- SUMMARY AND RECOMMENDATIONS
1.1.1. Analytical methods
1.1.2. Sources of chromium, environmental levels and exposure
1.1.4. Effects on experimental animals
1.1.5. Effects on human beings
22.214.171.124 Clinical and epidemiological studies
1.1.6. Evaluation of risks for human health
1.2. Recommendations for further research
1.2.1. Analytical methods
1.2.2. Sources of chromium intake
1.2.3. Studies on health effects
1.2.4. Interaction with other environmental factors
- PHYSICAL AND CHEMICAL PROPERTIES AND ANALYTICAL METHODS
2.1. Physical and chemical properties
2.2. Analytical methods
2.2.2. Analytical methods
- SOURCES IN THE ENVIRONMENT, ENVIRONMENTAL TRANSPORT AND DISTRIBUTION
3.1. Natural occurrence
3.1.5. Plants and wildlife
3.1.6. Environmental contamination from natural sources
3.2. Production, consumption, and uses
3.3. Waste disposal
3.4. Miscellaneous sources of pollution
3.5. Environmental transport and distribution
- ENVIRONMENTAL LEVELS AND HUMAN EXPOSURE
4.1. Environmental levels
4.2. General population exposure
4.2.1. Food and water
4.2.2. Other exposures
4.3. Occupational exposure
4.3.1. Inhalation exposure
4.3.2. Dermal exposure
- KINETICS AND METABOLISM
5.1.1. Absorption through inhalation
126.96.36.199 Animal studies
188.8.131.52 Human data
5.1.2. Absorption from the gastrointestinal tract
184.108.40.206 Animal studies
220.127.116.11 Human data
5.2. Distribution, retention, excretion
5.2.1. Animal studies
5.2.2. Human data
18.104.22.168 Concentration in tissues, blood, urine,
and hair including possible biological
indicators of exposure
22.214.171.124 Dynamic aspects of metabolism
and the influence of pathological states
5.3. Influence of chemical form
- EFFECTS ON ORGANISMS IN THE ENVIRONMENT
6.3. Aquatic organisms
- EFFECTS ON EXPERIMENTAL ANIMALS AND IN VITRO TEST SYSTEMS
7.1. Nutritional effects of chromium
7.1.1. Effects of deficiency on glucose metabolism
7.1.2. Effects of deficiency on lipid metabolism
7.1.3. Effects of deficiency on life span, growth, and reproduction
7.1.4. Other effects of deficiency
7.1.5. Mechanism of action of chromium as an essential nutrient
126.96.36.199 Enzymes, nucleic acids, and thyroid
188.8.131.52 Interaction of chromium with insulin
7.1.6. Chromium nutritional requirements of animals
7.2. Toxicity studies
7.2.1. Effects on experimental animals
184.108.40.206 Developmental toxicity and other
220.127.116.11 Cytotoxicity and micromolecular syntheses
7.2.2. Observations in farm animals
- EFFECTS ON MAN
8.1. Nutritional role of chromium
8.1.1. Biological measurements and their interpretation
8.1.2. Chromium deficiency
18.104.22.168 Malnourished children
22.214.171.124 Patients on total parenteral alimentation
126.96.36.199 Epidemiological studies
8.1.3. Mode of action
8.2. Acute toxic effects
8.3. Chronic toxic effects
8.3.1. Effects on skin and mucous membranes
188.8.131.52 Primary irritation of the skin
and mucous membranes
184.108.40.206 Allergic contact dermatoses
8.3.2. Effects on the lung
220.127.116.11 Bronchial irritation and sensitization
8.3.3. Effects on the kidney
8.3.4. Effects on the liver
8.3.5. Effects on the gastrointestinal tract
8.3.6. Effects on the circulatory system
8.3.8. Mutagenicity and other short-term tests
18.104.22.168 Lung cancer
22.214.171.124 Cancer in organs other than lungs
126.96.36.199 Relative risk between cancer risk
and chromium compound
- EVALUATION OF HEALTH RISKS FOR MAN
9.1. Occupational exposure
9.1.1. Effects other than cancer
188.8.131.52 Respiratory tract
184.108.40.206 Other organs and systems
9.2. General population
WHO TASK GROUP ON CHROMIUM
Professor Chen Bingheng, Department of Environmental Health,
Shanghai Medical University, Shanghai, China
Dr H.N.B. Gopalan, University of Nairobi, Department of Botany,
Professor C.R. Krishna Murti, Integrated Environmental
Programme on Heavy Metals, Department of Environment,
Government of India, New Delhi, India (Vice-Chairman)
Professor Aly Massoud, Department of Community, Environmental
and Occupational Medicine, Faculty of Medicine, Ain Shams
University, Cairo, Egypt
Dr W. Mertz, Human Nutrition Research Center, US Department of
Agriculture, Beltsville, Maryland, USA (Chairman)
Professor I.V. Sanotsky, Department of Toxicology, Institute
of Industrial Hygiene and Occupational Diseases, Academy
of Medical Sciences of the USSR, Moscow, USSR
Professor W. Stöber, Fraunhofer Institute for Toxicology and
Aerosol Research, Hanover, Federal Republic of Germany
Dr J. Parizek, International Programme on Chemical Safety,
World Health Organization, Geneva, Switzerland (Secretary)
Dr R.F. Hertel, Fraunhofer Institute for Toxicology and
Aerosol Research, Hanover, Federal Republic of Germany
(Temporary Adviser) (Rapporteur)
Dr T. Ng, Office of Occupational Health, World Health
Organization, Geneva, Switzerland
Mr J.D. Wilbourn, Unit of Carcinogen Identification and
Evaluation, International Agency for Research on Cancer,
NOTE TO READERS OF THE CRITERIA DOCUMENTS
Every effort has been made to present information in the
criteria documents as accurately as possible without unduly
delaying their publication. In the interest of all users of the
environmental health criteria documents, readers are kindly
requested to communicate any errors that may have occurred to the
Manager of the International Programme on Chemical Safety, World
Health Organization, Geneva, Switzerland, in order that they may be
included in corrigenda, which will appear in subsequent volumes.
* * *
A detailed data profile and a legal file can be obtained from
the International Register of Potentially Toxic Chemicals, Palais
des Nations, 1211 Geneva 10, Switzerland (Telephone no. 988400 -
ENVIRONMENTAL HEALTH CRITERIA FOR CHROMIUM
A WHO Task Group on Environmental Health Criteria for Chromium
met in Geneva from 24 to 27 March 1986. Dr J. Parizek opened the
meeting on behalf of the Director-General. The Task Group reviewed
and revised the draft criteria document and made an evaluation of
the health risks of exposure to chromium.
The initial draft was prepared by the INSTITUTE FOR GENERAL AND
COMMUNITY HYGIENE, MOSCOW. The second draft criteria document was
prepared by DR W. MERTZ, HUMAN NUTRITION RESEARCH CENTER, US
DEPARTMENT OF AGRICULTURE, USA, PROFESSOR ANNA BAETGER, JOHN
HOPKINS UNIVERSITY, BALTIMORE, USA (deceased), and Dr R.F. HERTEL,
FRAUNHOFER INSTITUTE OF TOXICOLOGY AND AEROSOL RESEARCH, Federal
Republic of Germany.
The efforts of all who helped in the preparation and
finalization of the document are gratefully acknowledged.
* * *
Partial financial support for the publication of this criteria
document was kindly provided by the United States Department of
Health and Human Services, through a contract from the National
Institute of Environmental Health Sciences, Research Triangle Park,
North Carolina, USA - a WHO Collaborating Centre for Environmental
Health Effects. The United Kingdom Department of Health and Social
Security generously supported the cost of printing.
- SUMMARY AND RECOMMENDATIONS
1.1.1. Analytical methods
Many analytical methods are available for the determination of
chromium at trace levels, often in the 0.001 mg/kg range. Among
these are flameless atomic absorption spectrometry, atomic emission
spectrometry with various excitation sources (the inductively-
coupled plasma torch is particularly advantageous), gas
chromatography, destructive or non-destructive neutron activation
analysis, and mass spectrometry using double-isotope dilution.
Depending on the particular sample under examination as well as the
analytical technique selected for the determination, wet or dry
ashing procedures may be necessary to destroy the organic/inorganic
matrix and minimize interelemental effects.
Determination of very low chromium concentrations in
"unexposed" biological material (animal and human tissues, blood,
urine, food, as well as water and air) is extremely difficult and
many problems still have to be solved. An accurate assessment of
human exposure and nutritional chromium requirements depends on
reliable analytical results. Chromium concentrations in blood,
urine, and some low-chromium foods are close to or less than 1
µg/kg, which is near the detection limit of even the most sensitive
analytical methods. Thus, agreement as to "normal" levels of
chromium among analytical investigators has been poor, and results
of interlaboratory comparisons have differed widely, usually by one
order of magnitude. Only in recent years has agreement been reached
that "normal" chromium concentrations in unexposed blood and urine
are in the range of 0.1 - 0.5 µg/litre. In this concentration
range, it is not only the sensitivity of the final determination
step that is limiting. The preceding steps of sample collection,
preparation, and digestion are equally important. Contamination,
easily introduced through cutting instruments and dust during
collection, must be carefully controlled. Digestion procedures are
of the greatest importance. Too rigorous treatment by heat or
certain acids can cause a loss of chromium. Few biological
standard reference materials, certified for chromium, are available
and almost all of the older, and most of the recent, published data
were not checked using certified standards. For this reason,
quantitative data concerning chromium concentrations in the range
of < 1 - 100 µg/kg in biological materials must be considered
uncertain, and caution must be used in interpreting their health-
Differential analysis for chromium species is of great
scientific and public health concern, in view of the substantial
differences in the biological availability and in the toxicity of
hexavalent chromium (Cr VI) compared with trivalent (Cr III).
Though methods based on solvent extraction, with or without prior
oxidation, differentiate between these two oxidation states, few
analytical data contain this important information.
The understanding of the chemical and physical principles of
chromium determination is increasing, and existing methods are
being improved and new methods developed. However, at present,
analysis for chromium is a sophisticated procedure requiring the
full attention of a highly trained analytical chemist.
1.1.2. Sources of chromium, environmental levels and exposure
Chromium occurs ubiquitously in nature (< 0.1 µg/m3 in air).
Natural levels in uncontaminated waters range from fractions of 1
µg to a few µg/litre.
The concentration of chromium in rocks varies from an average
of 5 mg/kg (granitic rocks) to 1800 mg/kg (ultramafic/basic and
serpentine rocks). The earth's most important deposits are either
in the elemental or the trivalent oxidation state.
In most soils, chromium occurs in low concentrations (2 - 60
mg/kg), but values of up to 4 g/kg have been reported in some
uncontaminated soils. Only a fraction of this chromium is
available to plants. It is not known whether chromium is an
essential nutrient for plants, but all plants contain the element
(up to 0.19 mg/kg on a wet weight basis).
Almost all the hexavalent chromium in the environment arises
from human activities. It is derived from the industrial oxidation
of mined chromium deposits and possibly from the combustion of
fossil fuels, wood, paper, etc. In this oxidation state, chromium
is relatively stable in air and pure water, but it is reduced to
the trivalent state, when it comes into contact with organic matter
in biota, soil, and water. There is an environmental cycle for
chromium, from rocks and soils to water, biota, air, and back to
the soil. However, a substantial amount (estimated at 6.7 x 106 kg
per year) is diverted from this cycle by discharge into streams,
and by runoff and dumping into the sea. The ultimate repository is
Chromium compounds are used in ferrochrome production,
electroplating, pigment production, and tanning. These industries,
the burning of fossil fuels, and waste incineration are sources of
chromium in air and water. Most of the liquid effluent from the
chromium industries is trapped and disposed of in land fills and
sewage sludges, the chromium being in the form of the insoluble
In chromium ore mines, the concentration of chromium in dust
ranges from 1.3 to 16.9 mg/m3. During the production of refined
ferrochromium, the air in the work-place may contain large amounts
of dust (0.03 - 3.2 mg/m3). In chromium plating factories,
concentrations of 1 µg/m3 up to 1.4 mg/m3 have been measured. In
Portland cement from 9 European countries, the contents of chromium
(VI), extractable with sodium sulfate, varied from 1 to 83 g/kg
Today, it is generally accepted that only the zero-, di-, tri-,
and hexavalent oxidation states have biological importance. The
effects of the last 2 oxidation states are so fundamentally
different that they must always be considered separately. The
trivalent form is an essential nutrient for man, in amounts of 50 -
The kinetics of chromium depend on its oxidation state and the
chemical and physical form within the oxidation state. Most of the
daily chromium intake (50 - 200 µg) is ingested with food and is in
the trivalent form. About 0.5 - 3% of the total intake of
trivalent chromium is absorbed in the body. It is possible, but it
has not yet been proved, that chromium in the form of some
complexes, such as a dinicotinic-acid-complex, glucose tolerance
factor, is better available for absorption. The gastrointestinal
absorption of 3 - 6% of the total intake of hexavalent chromium has
been reported. Once absorbed, chromium is almost entirely excreted
with the urine; the daily urinary-chromium loss of 0.5 - 1.5 µg is
approximately equal to the amount absorbed from the average diet.
However, dermal losses, losses by desquamation of intestinal cells
and by perspiration have not been quantified. Ingested or injected
chromium leaves the blood rapidly. Blood-chromium levels do not
reflect the overall chromium content of tissues, except after a
glucose load, which induces an immediate increase in the plasma-
and urine-chromium levels of chromium-sufficient subjects.
Trivalent chromium inhaled from the air is trapped in the lung
tissues, if in the form of small particles within the respirable
range. The chromium concentrations in lungs increases with age.
Larger particles (greater than 5 µm), regardless of oxidation
state, are moved to the larynx by ciliary action and become part of
the dietary intake.
The intestinal absorption of hexavalent chromium is 3 - 5 times
greater than that of trivalent forms; however, some of it is
reduced by gastric juice. Soluble chromates are rapidly absorbed
through the epithelium of the alveoli and bronchi and cleared into
the circulation, where part is preferentially accumulated by the
red cells and part is excreted by the kidneys. With the exception
of the lungs, tissue levels of chromium decline with age.
1.1.4. Effects on experimental animal
Doses of hexavalent chromium greater than 10 mg/kg diet affect
mainly the gastrointestinal tract, kidneys, and probably the
haematopoetic system. When a similar dose is introduced
parenterally, the principal effect is on the kidney, mainly in the
proximal convoluted tubules, without evidence of glomerular damage.
Toxic effects from trivalent chromium have been reported only
following parenteral administration. Dietary toxicity has not been
reported, even in studies on cats administered amounts of up to 1
g/day for 1 - 3 months. When intravenously injected in mice, the
LD50 of chromium carbonyl was 30 mg/kg body weight; this represents
a 10 000-fold excess over the therapeutic dose required to cure
signs of chromium deficiency.
Many studies on experimental animals have been conducted with
chromium compounds in efforts to reproduce cancer similar to that
found in man, when exposed to chromium.
Most tests have involved subcutaneous, intramuscular, or
intrapleural injection. In addition, several hexavalent chromium
compounds have been administered to rats by intrabronchial
implantation or intratracheal instillation. Relatively insoluble
compounds, calcium chromate, strontium chromate, and certain forms
of zinc chromate produced bronchogenic carcinomas; lead chromate,
and barium chromate produced weak responses. Intratracheal
instillation of soluble sodium dichromate and dissolved calcium
chromate produced bronchogenic tumours. Injection of lead
chromate, lead chromate oxide, and cobalt-chromium alloy resulted
in the production of local sarcomas. Thus, there is sufficient
evidence that certain hexavalent chromium compounds are
carcinogenic for experimental animals. No increased tumour
incidence was observed when trivalent compounds were given orally;
however, the doses administered were low.
Hexavalent chromium has been reported to cause various forms of
genetic damage in short-term mutagenicity tests, including damage
to DNA, and misincoporation of nucleotides in DNA transcription.
It was mutagenic in bacteria in the absence of an exogenous
metabolic activation system, and in fungi. Hexavalent chromium was
also mutagenic in mammalian cells in vitro and in vivo.
Hexavalent chromium caused chromosomal abererations and sister
chromatid exchanges in mamalian cells in vitro. A few positive
results in in vitro assays for mammalian cell chromosomal
aberrations and sister chromatid exchanges were obtained only with
very high doses and could be explained by nonspecific toxic
effects. It induced formation of micronuclei in mice in vivo.
Potassium dichromate induced dominant lethal mutations in mice
treated in vivo.
Trivalent chromium is genetically active only in in vitro
tests, where it can have a direct interaction with DNA, e.g., in
experiments using purified DNA or tests to measure decreased
fidelity of DNA synthesis in vitro. Reduction of chromium (VI)
within the cell nucleus and the formation of chromium (III)
complexes suggests that chromium (III) would be the ultimate
mutagenic form of chromium. Trivalent chromium was present in RNAs
from all sources examined and probably contributes to the stability
of the structure. Injected chromium trichloride (CrCl3)
accumulated in the cell nucleus (up to 20% of cellular chromium
content). It enhanced RNA synthesis in mice and in regenerating
rat liver, suggesting that chromium (III) is involved directly in
RNA synthesis. On the other hand, chromium (VI) inhibited RNA
synthesis and DNA replication in several systems.
1.1.5. Effects on human beings
Studies on man and experimental animals have established the
essential role of trivalent chromium for the maintenance of normal
glucose metabolism. Chromium deficiency has been demonstrated in
malnourished children, in two patients on total parenteral
nutrition, and in middle-aged subjects, the basic disturbance being
an impairment of the action of circulating insulin.
220.127.116.11. Clinical and epidemiological studies
In adult human subjects, the lethal oral dose is 50 - 70 mg
soluble chromates/kg body weight. The most important clinical
features produced following this route of entry are liver and
kidney necrosis and poisoning of blood-forming organs.
Hexavalent chromium causes marked irritation of the respiratory
tract. Ulceration and perforation of the nasal septum have
occurred frequently in workers employed in the chromate producing
and hexavalent chromium-using industries. In addition to
inhalation, direct contact of the nasal septum with contaminated
hands contributes to nasal exposure. Cancer of the septum has not
been reported. Rhinitis, bronchospasm, and pneumonia may result
from exposure to hexavalent compounds together with impairment of
pneumodynamics during respiration.
Chromate compounds, mainly sodium and potassium chromate and
dichromate, cause irritation of the skin and ulcers may develop at
the site of skin damage. Exposure to trivalent chromium does not
produce such effects. Certain persons manifest allergic skin
reactions to hexavalent and possibly trivalent chromium. Skin
reactions through dermal exposure to chromium are often described,
chromate being the most common contact allergen. However, cancer
of the skin due to chromium exposure has not been reported.
Chronic effects of exposure to chromium (excessive industrial
exposure of the skin to hexavalent chromium, when associated with
damaged skin or inhalation of airborne chromium (VI) or mixed dust)
occur in the lung, liver, kidney, gastrointestinal tract, and
ciculatory system. Teratogenic risks from chromium exposure have
not been reported for human subjects; a mutagenic potency is shown
for potassium dichromate and therefore cannot be excluded for
chromates in the chromate-using industries.
The results of epidemiological studies in various countries
have demonstrated that men working in chromate-production plants
before 1950 had a very high rate of bronchogenic carcinoma,
compared with control populations. Because of the long period
between initial exposure and the detection of cancer and the lack
of data on the extent and type of exposure, the dose-response
relationship has not been quantified. However, the few data
available indicate that, before the danger of cancer was
recognized, the exposure levels in such plants were very high.
Recent data show clearly that, though the risk of cancer in workers
in modern plants has been greatly reduced, it still remains a
Some epidemiological data suggest that an excess of lung cancer
has also occurred in the chromate-pigment industry. A few cases of
cancer involving the upper respiratory tract have been reported,
but cancer has not been convincingly demonstrated in other body
tissues. The specific compounds responsible for the cancers have
not been identified. Both hexavalent and trivalent compounds were
present in the old plants. However, on the basis of experimental
animal studies, it is currently assumed that the slowly soluble,
hexavalent chemicals, such as calcium and zinc chromate are
responsible for the cancers. This is based on the theory that
these compounds remain in contact with the tissues for long periods
of time (depot effect).
Trivalent chromium is not considered to be carcinogenic for the
following reasons: (a) there was no evidence of excess cancer in
studies in two industries where only trivalent compounds were
present; (b) results of experimental animal and mutagenicity
studies with trivalent chromium, were negative; and (c) because of
the chemical and biological characteristics of the trivalent state,
i.e., non-oxidizing, non-irritating, and probably unable to
penetrate cell membranes.
1.1.6. Evaluation of risks for human health
Chromium in the form of trivalent compounds is an essential
nutrient. The daily human intake of chromium varies considerably
between regions. Typical values range from 50 to 200 µg/day. Such
intakes do not represent a toxicity problem, and they coincide with
the calculated human requirements. Not enough data are available
for a quantitative assessment of the risk of chromium deficiency in
Evidence from studies on experimental animals shows that
hexavalent chromium compounds, especially those of low solubility
can induce lung cancer. Mutagenicity and related studies have
shown convincingly that hexavalent chromium is genetically active.
On the other hand, trivalent chromium compounds are inactive in
most test systems, except in systems where they can directly
interact with DNA.
Both oxidation states, when injected at high levels
parenterally in animals, are teratogenic, with the hexavalent form
accumulating in the embryos to a much greater extent than the
A number of effects can result from occupational exposure to
airborne chromium, including irritative lesions of the skin and
upper respiratory tract, allergic reactions, and cancers of the
respiratory tract. The data on other effects in, e.g., the
gastrointestinal, cardiovascular, and urogenital systems are
insufficient for evaluation.
Epidemiological studies have shown that workers engaged in the
production of chromate salts and chromate pigments are at increased
risk of developing bronchial carcinoma. No detailed data on dose-
response relationships are available. Although a suspicion of
increased lung cancer risks in chromium-plating workers has been
raised, the available data are inconclusive and so are data for
other industrial processes where chromium compounds are used rather
than produced. There is insufficient evidence to implicate
chromium as a causative agent of cancer in any organ other than the
lung. The frequency of sister chromatid exchanges in the
lymphocytes of workers in chromium-plating factories was higher
than in control groups.
The general population living in the vicinity of ferro-alloy
plants and exposed to ambient air concentrations of up to 1 µg/m3
did not show increased lung cancer mortality.
The results of many studies suggest that exposure to chromium
through inhalation and skin contact can pose health problems for
the general population, but no data on dose-response relationships
are available. Thus, there is no reason, at present, to be
concerned that chromium in the air presents a health problem,
except under conditions of industrial exposure.
1.2. Recommendations for Further Research
1.2.1. Analytical methods
Data from the determination of chromium should not be accepted
unless proper quality assurance procedures have been used,
including the analysis of a certified reference material of similar
composition. There is a great need for the preparation and
certification of additional standards, especially of blood, serum,
or plasma, urine containing only physiological chromium
concentrations, hair, and foods.
All analyses related to the environmental role of chromium
should differentiate between hexavalent and trivalent forms and
these values should be reported separately. While the
differentiation between hexavalent and trivalent chromium can be
accomplished by established methods, the definition of the exact
chemical species of the trivalent and hexavalent forms in air,
water, food, and tissues will require much further research.
Further development of analytical instrumentation and
preanalytical processing techniques to extend the detection limit
by one order of magnitude is recommended. The need for
interlaboratory comparison persists to improve existing methods and
to validate new procedures.
1.2.2. Sources of chromium intake
More data are needed on the chemical and physical properties of
airborne chromium, such as the oxidation state, particle size, and
solubility. These are important determinants of biological and
toxic action. Existing information on the chromium contents of
foodstuffs is unreliable and incomplete and more composition data
are needed for a valid assessment of the human chromium requirement
and the supplies available in different regions of the world to
meet these requirements. Diagnostic procedures to detect marginal
deficiency and marginal overexposure in man must be developed and
the long-term effects of both these imbalances must be defined.
Finally, not enough is known about the fate of chromium in
landfills, sewage sludges, and aquatic environments. Further
studies are needed to investigate environmental factors that
influence the mobilization, migration, and bioavailability of
chromium in the biosphere.
1.2.3. Studies on health effects
Prospective studies on the health of industrial workers,
combined with the determination of the composition and
environmental levels of the chromium compounds to which they were
exposed, are needed to determine the specific chemical or chemicals
responsible for cancer, and the dose-response relationship between
hexavalent chromium and bronchogenic carcinoma. Smoking histories
should be recorded and, when possible, information on exposure to
ionizing radiation and other chemical carcinogens should be
obtained in order to evaluate possible synergistic relationships.
More studies should be carried out on the chrome-using industries.
Preventive measures include searching for more specific biochemical
indicators of chromium exposure and early effects.
Epidemiological studies are needed to assess the incidence and
severity of chromium deficiency. The relation of chromium status
to cardiovascular diseases needs further investigation,
particularly in areas with protein-energy malnutrition.
1.2.4. Interaction with other environmental factors
The interaction of other pollutants in the atmosphere with
chromium, particularly with respect to particle size, adsorption at
the particle surface, etc., require further studies.
Interactions between trivalent chromium in the diet and other
dietary constituents are poorly understood and should be
- PHYSICAL AND CHEMICAL PROPERTIES AND ANALYTICAL METHODS
2.1. Physical and Chemical Properties
Chromium (atomic number 24, relative atomic mass 51.996) occurs
in each of the oxidation states from -2 to +6, but only the 0
(elemental), +2, +3, and +6 states are common. Divalent chromium
is unstable in most compounds, as it is easily oxidized to the
trivalent form by air. Only the trivalent and hexavalent oxidation
states are important for human health. In the context of this
publication, it is of great importance to realize that these two
oxidation states have very different properties and biological
effects on living organisms, including man. Therefore, they must
always be examined separately: a valid generalization of the
biological effects of chromium as an element cannot be made.
This discussion will concentrate only on the aspects of
chromium chemistry that are of concern for health.
The relation between the hexavalent and trivalent states of
chromium is described by the equation:
Cr2O72- + 14H+ + 6e -> 2 Cr(III) + 7H2O + 1.33v.
The difference electric potential between these 2 states reflects
the strong oxidizing properties of hexavalent chromium and the
substantial energy needed to oxidize the trivalent to the
hexavalent form. For practical purposes, it can be stated that
this oxidation never occurs in biological systems. The reduction
of hexavalent chromium occurs spontaneously in the organism, unless
present in an insoluble form. A gradual reduction of hexavalent
chromium to the trivalent state is demonstrated by the colour
change of the conventional chromate cleaning solution in the
laboratory from bright orange to green, in the presence of organic
matter. In blood, chromate is reduced to the trivalent state, once
it has penetrated the red cell membrane and becomes bound to the
haemoglobin and other constituents of the cell and therefore unable
to leave the cell again. The rapid reduction of injected
51chromium-labelled chromate in the rat has been demonstrated by
Feldman (1968). Although a compound CrF6 is well known, the stable
forms of hexavalent chromium are almost always bound to oxygen
(e.g., CrO4-2, Cr2O7-2). The trivalent form exists in coordination
compounds, but never as the free ion. As a rule, its coordination
number is 6, the complexes being generally octahedral.
A large number of complexes and chelates of chromium have been
investigated, ranging from simple hexa- or tetra-aquo complexes to
those with organic acids, vitamins, amino acids and others. The
rate of ligand exchange of chromium complexes is slow in comparison
with other transition elements, with the exception of the even
slower rate of cobalt complexes; most of the Cr(III)-complexes are
kinetically stable in solutions. This property adds to the relative
inertness of trivalent compounds, in addition to the
electrochemical stability of the trivalent state. However, at near
neutral or alkaline pH, the milieu of the animal organism, the
simple chromium compounds to which the organism is exposed in the
environment or through supplementation, rapidly become insoluble,
because hydroxyl ions replace the coordinated water molecules from
the metal and form bridges, linking the chromium atoms into very
large, insoluble complexes. Coordination of trivalent chromium to
biological ligands is the prerequisite for its solubility at
physiological pH and therefore for its biological function and for
its availability for intestinal absorption. The coordination
chemistry and the specific biochemical reactions have been reviewed
by Cotton & Wilkinson (1966) and Mertz (1969), respectively. The
physical and chemical properties of chromium and some chromium-
compounds are summarized in Table 1.
Common chromium compounds
Poorly soluble "sandwich complexes" of metallic chromium
(oxidation state = O) are known, e.g., Cr(C6H6)2; these have little
practical application. Divalent compounds, such as chromium (II)
chloride (CrCl2) are used as strong reducing agents in the
laboratory, but have little industrial use. Of the many hundreds
of trivalent chromium compounds known, chromic oxide (Cr2O3 x
nH2O), is used as a pigment in paints and as a faecal marker in
digestive studies. It dissolves in acids and forms the hexa-aquo
or tetra-aquo complex, e.g.,
Cr2O3 x 9H2O + 6HCl -> 2 [Cr(H2O)6] Cl3
2 [Cr(H2O)4Cl2] Cl + 4H2O
(colour: dark green).
Chromium chloride ([Cr(H2O)6]Cl3 or [Cr(H2O)4Cl2]Cl) is used in
basic solution for leather tanning. The fluoride is used
industrially in printing and dyeing, and chromium sulfates and
nitrates are used as colouring and printing dyes.
One of the numerous organic complexes of chromium, a
dinicotinatoglutathionato-chromium complex has been isolated from
yeast. It is postulated as the physiologically active form in the
animal organism, but its exact structure is not known (Toepfer et
Table 1. Physical and chemical properties of chromium and some selected chromium compounds
Name Chemical Relative Specific Melting Boiling Colour Solubility CAS registry
symbol molecular gravity point point in water number
mass (g/cm3) (°C) (°C) (weight %)
Chromium Cr 51.996 7.19 1857 2672 steel- insoluble 7440-47-3
Chromium (III)- Cr2O3 151.99 5.21 2266 4000 green insoluble 1308-38-9
Chromium (VI)- CrO3 99.99 2.70 196 decompo- red 62.41 1333-82-0
Potassium- K2CrO4 194.20 2.732 968.3 decompo- yellow 39.96 7789-00-6
chromate (VI) sition
Potassium- K2Cr2O7 294.19 2.676 398 decompo- red 11.7 7778-50-9
dichromate (VI) sition
Calcium- CaCrO4 192.09 1025 decompo- yellow 3.5 13765-19-0
chromate (VI) x 2H2O sition
Calcium- CaCr2O4 208.07 4.8 2090 - olive- insoluble
chromium (III)- green
For vapour pressure at 20°C, no data.
The earth's most important deposits of chromium are in either
the elemental or the trivalent oxidation state. Hexavalent
compounds of chromium in the biosphere are predominantly man-made,
and experience with hexavalent chromium is relatively short.
Chromates and dichromates are produced from chromite ore by
roasting in the presence of soda ash. From these, chromium (VI)
oxide, (CrO3), is precipitated out by the addition of sulfuric
acid. Sodium and potassium dichromates are widely used
industrially as sources of other chromium compounds, particularly
of chromium (VI) oxide, and these processes are a major source of
hexavalent chromium pollution (US EPA, 1978).
2.2. Analytical Methods
Methods for the determination of chromium in biological and
environmental samples are developing rapidly, as shown by the fact
that chromium concentrations in the blood and urine of unexposed
subjects, reported as normal, have been revised downwards by 2
orders of magnitude, in only 15 years (Versieck et al., 1978).
This development is not only due to the increasing powers of
detection and specificity of more recent methods, but also to the
better methods of contamination control that have become available.
For these reasons, all data concerning chromium levels in blood and
urine (particularly the early results), should be interpreted with
caution following scrutiny of all experimental details. On the
other hand, analytical results concerning the much higher chromium
levels in foodstuffs and human tissue have not changed as much and
can be accepted with more confidence. However, all interpretations
of chromium data should take into account the need for caution
expressed in section 2.2.2.
As chromium is present in biological materials in very low
concentrations, care must be taken to avoid contamination. The
collection of dust from air samples may introduce contamination
from the chromium in the filters; blood or tissue samples may
become highly contaminated by the chromium in needles, knives,
blenders, and other instruments (Behne & Brätter, 1979). Water
samples may extract chromium from containers. Finally, reagents
used in sample dissolution, separation, chelation, acid digestion,
and other reactions, may contribute significant amounts of
chromium. Thus, it is necessary to control for these influences by
simultaneously performing a blank analysis, i.e., by carrying out
the whole analysis, including sampling, preparation, and digestion,
using all reagents, excluding a sample (Davis & Grossman, 1971).
Conversely, chromium in low concentrations may be adsorbed on
the surface of containers during long periods of storage. This
aspect has not yet been sufficiently investigated (Shendrikar &
West, 1974). Procedures for the sampling of different materials
for chromium determination have been reviewed by Beyermann (1962),
Brown et al. (1970), Murrman et al. (1971), Versieck & Speecke
(1972), Skogerboe (1974), Johnson (1974), and US DHEW (1975). All
suggest strictest contamination control (clean rooms or laminar
2.2.2. Analytical methods
The voluminous literature on analysis for chromium was reviewed
by US EPA (1978). A discussion on analytical methods must
distinguish between two categories: (a) methods for measuring
large, potentially toxic concentrations of chromium as a
contaminant, and (b) methods of analysis for chromium as an
essential nutrient. The first category requires reliable
determinations of chromium at the µg/kg level; the second requires
greater sensitivity, e.g., to determine accurately the chromium
level in urine at several hundred ng/litre.
The sensitivity of instrumental analysis for the determination
of chromium does not present any problems for concentrations in the
mg/kg range, and a number of techniques can furnish satisfactory
precision and accuracy (Table 2). On the other hand, the
sensitivity of instrumentation for the determination of chromium in
the ng or µg/kg range is severely limited, and no one method is
entirely satisfactory, at present (Seeling et al., 1979). The
biologically active concentrations are near the detection limits of
the most sensitive methods, such as neutron activation analysis or
flameless atomic absorption spectrometry. In an inter-laboratory
comparison, there was poor agreement between the analytical results
obtained by well-established, experienced analytical laboratories
in several countries (Parr, 1977). Some of the results are
presented in the Table 3. It is of paramount importance for the
interpretation of all published analytical data on chromium to
realize the great variation in reported results, even for high
concentrations. These results indicate the following conclusions:
- No one analytical method can be expected to produce
"true" results of absolute chromium concentrations, unless
the analyses are controlled by the use of a certified
reference material with a matrix composition similar to
that of the material to be analysed.
- No valid comparisons can be made on the basis of
analytical results obtained by different laboratories,
unless the same reference materials have been used by
both, or samples have been exchanged.
- There is a great need for certified Standard
Reference Materials with many different matrix
compositions. Six such standards of biological materials
have been certified for chromium content (tomato leaves,
pine needles, citrus leaves, oyster tissue, unexposed
bovine serum, and brewer's yeast). In addition, seven
standard reference materials of environmental samples are
available (coal, fly ash, water, sediment, urban
particulate, etc.) and more than 180 industrial samples of
various steels and metal alloys. These are available from
the National Bureau of Standards, Washington DC, USA. New
reference specimens of blood and urine have been produced
for the quality control of heavy metals in industrial
medicine and toxicology (Müller-Wiegand et al., 1983).
The assigned values were determined by reference
laboratories of the "Deutsche Gesellschaft für
Arbeitsmedizin; the control blood and urine preparations
are offered by Behringwerke AG, D-3550 Marburg, Federal
Republic of Germany.
- In inter-laboratory quality assurance studies, it is
preferable to use the methodology developed in the
WHO/UNEP project on biological monitoring for lead and
cadmium (Vahter, 1982).
In 1983, the German DIN-Committee AAS adopted a method for the
determination of the chromium content of water and sewage (by means
of the flame AAS) (Kempf & Sonneborn, 1976); inductively coupled
plasma emission spectrometry is recommended with regard to serial
analyses (Kempf & Sonneborn, 1981).
Two special problems in the analysis for chromium may add to,
or subtract from, the true concentrations, i.e., contamination and
possible loss through volatilization or formation of refractory
compounds during sample preparation. Contamination is a serious
problem when low concentrations in blood or urine are measured.
Dust in laboratories may contain a chromium level of 700 mg/kg
(Mertz, 1969), approximately 6 orders of magnitude higher than the
concentration in urine of 0.2 - 0.7 µg/litre (Guthrie et al.,
1979). In other words, contamination of a one-ml urine sample by
only 1 µg of dust will increase the apparent chromium concentration
two-fold. Special precautions, such as those proposed by Tölg
(1974), must be taken to control this problem. The second problem,
that of potential loss during sample preparation, has been
discussed by Wolf & Greene (1976). There is evidence from several
studies that certain methods of sample preparation, such as heating
or acid digestion in open systems, may lead to the loss of
detectable amounts of chromium (Kotz et al., 1972). A typical
example, in which identical samples were determined by the
identical method, by the same analyst, in the same laboratory is
presented in Table 4.
Table 2. Instrumental methods for the determination of chromiuma
Analytical Relevant Detection Interfering substance Selectivity
method applications limit
Atomic fresh and saline 2 µg/litreb interfering substances all of the extracted
absorption water, industrial present in the original chromium is measured,
spectroscopy waste fluids, sample are usually not but only hexavalent
(flame) dust, and sediments extracted into the chromium is extracted
biological solids organic solvent from the original sample,
and liquids, alloys unless oxidative
pretreatment is used
Atomic biological solids 0.005 µg/ no interfering sub- total chromium is
absorption and fluids; tissue, litreb substances reported determined
(electrothermal) blood, urine; for samples of urinen,
industrial waste and bloodo; less than
waters 10% interference ob-
served for Na+, K+,
Ca2+, Mg3+, Cl-,
F-, SO4-3, and PO4-3
in certain industrial
Emission a wide variety of 4 µg/litrel no interfering total chromium is
spectroscopy biological and substances reported determined
coupled plasma samples
Emission a wide variety of 0.5 ngc total chromium is
spectroscopy environmental determined
Table 2. (contd.)
Analytical Relevant Detection Interfering substance Selectivity
method applications limit
Spectrophotometry natural water and 3 µg/litred iron, vanadium, and after chelation, only
industrial waste mercury may interfere the hexavalent chromium
solutions having in solution is determined
5 - 400 mg hexa-
litre can be
must be reduced
X-ray fluorescence atmospheric 2 - 10 µg/g the particle size of total chromium is
particulates, (liver)e; the sample and the determined
geological 1.5 µg/g matrix may influence
materials (coal)f the observed measure-
Neutron activation air pollution depends on interference may arise total chromium is
analysis particulates, activation from gamma ray activity measured
fresh and saline procedure; from other elements,
waters, biological typical especially Na-24, Cl-38,
liquids and solids, limit: K-42, and Mn-56; P-32
sediments, metals, 10 ngm may also cause inter-
Gas chromatography blood, serum, 0.03 pgg excess chelating agent only chromium that is
(electron capture natural water or other electron- chelated and extracted
detection) samples capturing constituents is measured; other
in the sample may lead electro-negative substances
to erroneous results may elute from the column
and be detected at the
same time as the chromium
Stable isotope all biological not expected high precision and accuracy
dilution mass materialsk,q (1%) complete sample
spectrometry digestion and exchange of
endogenous chromium with
added stable isotope
Table 2. (contd.)
Analytical Relevant Detection Interfering substance Selectivity
method applications limit
Gas chromatography blood, serum, ~1 ngh no interference only chromium that is
(atomic biological material reported chelated and extracted is
spectroscopic detected; atomic
detection) spectroscopic methods
of detection are
inherently more selective
for chromium in complex
Gas chromatography blood, plasma, 0.5 pgi no interference only chromium that is
(mass serum reported chelated and extracted
spectrometric can be detected
Chemiluminescence fresh, natural 30 ng/ Co(II), Fe(II), and only trivalent chromium
waters; dissolved litrek Fe(III) interfere but ion is measured
biological material may be compensated for
by running a blankc
a Modified from: US EPA (1978).
b From: Welz (1983).
c From: Seeley & Skogerboe (1974).
d From: American Public Health Association, American Water Works Association,
and Water Pollution Control Federation (1971).
e From: Kemp et al. (1974).
f From: Kuhn (1973).
g From: Savory et al. (1969).
h From: Wolf (1976).
i From: Wolf et al. (1972).
k From: Seitz et al. (1972).
l From: Welz (1980).
m From: Keller (1980).
n From: Schaller et al. (1973).
o From: Environmental Instrumentation Group (1973).
p From: Morrow & McElhaney (1974).
q From: Veillon et al. (1979).
Table 3. Results for 3 IAEA intercomparison studiesa
Laboratory Method Number of Laboratory SD (%)
code codeb determinations mean
- Simulated air filter
(true chromium concentration: 1.85 µg/filter)
a 7 2 1.3 3
b 3 1 1.6 7
c 2 4 1.78 5
d 2 10 1.85 6
e 2 6 1.86 52
f 2 2 2.00 30
g 2 3 2.07 10
h 2 6 2.07 9
i 2 6 2.07 6
j 7 10 2.16 10
k 5 6 2.83 40
l 7 2 3.00 -c
m 3 5 3.17 4
n 7 1 4.20 8
o 2 5 6.14 6
p 2 1 7.50 22
(true chromium concentration: 11.1 µg/kg)
a 3 4 1.85 18
b 3 5 3.80 12
c 7 6 4.16 15
d 7 1 4.50 11
e 7 6 4.77 6
f 7 2 5.25 -
g 2 5 5.51 3
h 2 5 5.84 4
i 7 1 6.00 -
j 2 3 6.08 4
k 7 2 6.50 -
l 7 2 6.85 17
m 2 2 7.00 -
n 7 2 7.30 -
o 7 3 8.67 3
p 7 2 8.90 20
q 3 1 9.00 20
r 7 5 9.60 12
s 7 6 9.92 10
t 2 5 10.8 11
u 2 4 11.3 11
v 7 3 11.5 45
w 5 2 18.0 30
x 7 1 73.0 14
Table 3. (contd.)
Laboratory Method Number of Laboratory SD (%)
code codeb determinations mean
- Bovine liver (SRM 1577)
(certified chromium concentration: 88 ± 12 µg/kg)
a 2 -c 5 -c
b 1 -c 51 13
c 1 -c 140 -c
d 2 -c 150 13
e 1 -c 150 33
f 1 -c 160 24
g 1 -c 240 53
h 2 -c 490 39
i 1 -c 540 64
j 2 -c 1300 -c
k 2 -c 1600 25
a From: Parr (1977).
b Method code:
- Destructive activation analysis.
- Nondestructive activation analysis.
- Emission spectroscopy.
- Spark source mass spectrometry.
- Atomic absorption, unspecified.
c Information not given.
At present, there is no explanation of the reason why "losses"
of almost 90% were associated with the direct graphite furnace
ashing, compared with oxygen plasma ashing in the case of molasses,
but not of refined sugar. Canfield & Doisy (1976) and Tuman et al.
(1978) suggested that the loss of chromium in biological samples,
such as urine, yeast extracts, or synthetic glucose tolerance
factor (GTF) preparations represented the biologically active GTF
fraction of the chromium. They correlated this "volatile" fraction
in yeast extracts with the antidiabetic activity of the extract in
genetically diabetic mice, and the amount of "volatile" chromium in
the urine of human subjects with the efficiency of the glucose
metabolism of these subjects. This hypothesis of "volatile"
chromium has been confirmed by some investigators (Behne et al.,
1976; Koirtyohann & Hopkins, 1976; Shapcott et al., 1977;
McClendon, 1978), and contradicted by others (Jones et al., 1975;
Rook & Wolf, 1977). While the question of the "volatility" of
chromium, under various conditions, remains unanswered, it is
obvious that chromium determination presents many problems, the
most pressing of which is the selection and control of sample
digestion (Wolf & Greene, 1976).
Table 4. Apparent chromium content depending on the method of
Type of sugar Number Chromium content + SEM (µg/kg)
of Oxygen Muffle Graphite
samples plasma furnace furnace ashing
ashing ashing (1000 °C)
(150 °C) (450 °C) (direct
molasses 3 266 ± 50 129 ± 54 29 ± 5
sugar (unrefined) 8 162 ± 36 88 ± 20 37 ± 13
sugar (brown) 5 64 ± 5 53 ± 8 31 ± 2
sugar (refined) 7 20 ± 3 25 ± 3 < 10
a From: Wolf et al. (1974).
Finally, it is important in any study of the environmental
effects of chromium, to distinguish analytically between the
trivalent and hexavalent forms. This can be accomplished by
dithiocarbamate chelation and solvent extraction (for example, with
methyl isobutylketone) prior to oxidation. Only the hexavalent
chromium remains after this process, and thus it is possible to
differentiate between the oxidation states (Feldman et al., 1967;
Cresser & Hargitt, 1976; Bergmann & Hardt, 1979; Joschi & Neeb,
1980). When determining chromium in biomaterial, the samples are
usually ashed with strong acids to destroy the organic components.
The relationship between the acids used and the behaviour of
chromium were investigated by Hara (1982) who showed that the
oxidation state of chromium was apt to change (hexavalent to
trivalent), because of the reducing action of each acid and the
conditions under which they were used.
- SOURCES IN THE ENVIRONMENT, ENVIRONMENTAL TRANSPORT AND DISTRIBUTION
3.1. Natural Occurrence
Chromium is ubiquitous in nature; it can be detected in all
matter in concentrations ranging from less than 0.1 µg/m3 in air to
4 g/kg in soils. Naturally occurring chromium is almost always
present in the trivalent state: hexavalent chromium in the
environment is almost totally derived from human activities.
Merian (1984) has compiled the global sources of chromium in
the environment. Total input (100%) consists of inputs by:
volcanic emissions (less than 1%); the biological cycle (30%)
including extraction from soil by plants (15%) and weathering of
rocks and soils (15%); and man-made emissions (70%) including those
from general ore and metal production (3%), from metal use (60%),
and from coal burning and other combustion processes (7%).
Almost all the sources of chromium in the earth's crust are in
the trivalent state, the most important mineral deposit being in
the form of chromite (FeOCr2O3) which, however, is rarely pure.
Living matter does not produce the energy necessary to oxidize
trivalent to hexavalent chromium in the organism, therefore, it can
be stated that nearly all hexavalent chromium in the environment is
produced by human activities. The industrial use of chromium and
the oxidation to the hexavalent state on an industrial scale did
not begin until 1816. Thus, man's experience with this form is
The concentration of chromium in rocks varies from an average
of 5 mg/kg (range of 2 - 60 mg/kg) in granitic rocks, to an average
1800 mg/kg (range, 1100 - 3400 mg/kg) in ultrabasic and serpentine
rocks (US NAS, 1974b).
Chromium deposits in the hexavalent oxidation state (crocoite
PbCrO4), were described by Lomonossow, in the year 1763 (Hintze,
1930), who found it in the Ural Mountains. Being a rare mineral,
chromium is found in the oxidized zones of lead deposits in regions
in which lead veins have traversed rocks containing chromite. It
may be associated with pyromorphite, cerussite, and wulfenite.
Notable localities are: Dundas, Tasmania; Beresovsk near
Sverdlovsk, Ural Mountains (Aleksandrov & Kainov, 1975), and
Rezbanya, Rumania. It is found in small quantities in the Vulture
district, Arizona, USA (Dana, 1971) and in the German Democratic
Republic in Callenberg, Saxony (Rohde et al., 1978).
Chromium concentrations in igneous rocks are positively
correlated with concentrations of silica, magnesium, and nickel.
Of agricultural importance, is the high chromium concentration in
sedimentary rocks, where the element is present in phosphorites.
This material is used as a phosphate fertilizer in agriculture and
is a significant source of chromium for agricultural soils.
Chromium-containing rocks and ores are found in all regions of
the world, but the major sources of the world's chromium supplies
are the ultra basic rocks of South Africa, Turkey, the USSR and
Zimbabwe (US NAS, 1974b). While underlying undisturbed rocks
contribute little chromium to the vegetation directly, the chromium
content is strongly correlated with that of the overlying soils.
Chromium can also be found in coal (5 - 10 mg/kg) (Merian,
The weathering of rocks produces chromium complexes that are
almost exclusively in the trivalent state. In most soils, chromium
occurs in low concentrations; an average of 863 soil samples from
the USA contained 53 mg/kg (Shacklette et al., 1970). The highest
concentrations, as high as 3.5 g/kg (Swaine & Mitchell, 1960), are
always found in serpentine soils. In a small area in Maryland,
USA, with soil infertility, the chromium concentration (as Cr2O3)
was as high as 27.4 g/kg (Robinson & Edington, 1935). Conversely,
low chromium concentrations (10 - 40 mg/kg) have been detected in
soils derived from granite or sandstone (Swaine & Mitchell, 1960).
Only a fraction of the chromium in soil is available to the plant;
thus, it is important to determine "available" soil-chromium. A
rough approximation of this available chromium fraction can be made
by extracting soil with acids or chelating agents and by measuring
the chromium in the extract. Though the amount of extractable
chromium is not identical with that truly available to the plant,
it is a much better measure of availability than the total
chromium. In the study of Swaine & Mitchell (1960), the amount of
chromium extracted from the soil with acetic acid varied much less
than the total soil content, and was not correlated with the latter
The comparisons in this Table indicate that the amount of
chromium available to the plant is relatively independent of the
total concentration. The complex principles determining the
availability of chromium for plants are poorly understood.
Table 5. Total versus extractable chromium in different
Soil Total chromium Extractable chromium
derived from: (mg/kg) (mg/kg)
Granite 20, 40, 20 0.15, 0.1, 0.11
Serpentine, 3500, 2000, 3000 0.31, 0.24, 0.63
a From: Swaine & Mitchell (1960).
It is now generally agreed that, except in areas with
substantial chromium deposits, high chromium levels in water arise
from industrial sources (US NAS, 1974b).
With the exception of areas bearing chromium deposits or in
highly industrialized areas, most surface waters contain very low
levels of chromium. The chromium content in surface water in the
Tia-ding county, Shanghai, ranged from 0 to 80 µg/litre (256
samples). According to the Yang-Pu water works, which is the
biggest water works in Shanghai, the chromium levels in well water
are below 50 µg/litre. Between 1980 and 1982, chromium was not
detectable in the Yellow River. There is no information concerning
the analytical methods used (Chen Bingheng, personal communication
to the Task Group, 1986). Kopp & Kroner (1968) detected chromium
in only 25% of surface water samples from sources in the USA, with
a range of 1 - 112 µg/litre, and a mean concentration of 9.7
µg/litre. The remaining 75% contained less than 1 µg/litre, the
detection limit. Another survey of 15 rivers in the USA revealed
levels ranging from 0.7 to 84 µg/litre, the majority of samples
containing between 1 and 10 µg/litre (Durum & Haffty, 1963). On
the other hand, chromium contents in natural water of up to 215
µg/litre were reported by Novakova et al. (1974). Although modern
methods of water treatment remove much of the naturally present
chromium, it should be noted that chlorinated drinking-water
usually contains traces of hexavalent chromium. The mean level in
the drinking-water supplies in 100 cities in the USA was only 0.43
µg/litre, with a range from barely detectable to 35 µg/litre
(Durfor & Becker, 1964).
Sea water contains less than 1 µg chromium/litre (US NAS,
1974b), but the exact chemical forms in which chromium is present
in the ocean, and surface water are not known. Theoretically,
chromium can persist in the hexavalent state in water with a low
organic matter content. In the trivalent form, chromium will form
insoluble compounds at the natural pH of water, unless protected by
complex formation. The exact distribution between the trivalent
and hexavalent state is unknown.
Chromium occurs in the air of non-industrialized areas in
concentrations of less than 0.1 µg/m3. The natural sources of air-
chromium are forest fires and, perhaps, volcanic eruptions (section
3.5). Man-made sources include all types of combustion and
emissions by the chromium industry (section 4.1.1). The chemical
forms of chromium in the air are not known, but it should be
assumed that part of the air-chromium exists in the hexavalent
form, especially that derived from high-temperature combustion.
Chromium trioxide (CrO3) may be the most important compound in the
air (Sullivan, 1969).
3.1.5. Plants and wildlife
It is not known whether chromium is an essential nutrient for
plants, but all plants contain the element in concentrations
detectable by modern methods.
Chromium concentrations in food plants growing on normal soils
range from not detectable to 0.19 mg/kg wet weight (Schoeder et
al., 1962). In addition, chromium of vegetable origin has a
relatively low biological activity (Toepfer et al., 1973).
Much higher concentrations have been reported in plants growing
on chromium deposits. For example, ash analysis showed a chromium
level of 0.34% in New Zealand lichen and 0.3% in Yugoslav Allysium
markgrafi (US NAS, 1974b). Growing on a serpentine soil (chromium
concentration 62 000 mg/kg in old mine tailings), the plant
chromium concentrations (on the basis of ash analysis) ranged from
700 mg/kg in Phormium colensoi and Liliacae to 5400 mg/kg in
Gentiana corymbifera (Lyon et al., 1970). Not all plants tolerate
high concentrations of available soil-chromium; chlorosis of citrus
trees has been observed in high-chromium areas and in laboratory
experiments. Plants grown in the vicinity of chromium-emitting
industries or those fertilized by sewage sludge are exposed to
substantial amounts of chromium. The chromium contents of plants
growing were determined by Taylor et al. (1975) near cooling
towers, where chromates were present as corrosion inhibitors. It
was shown that chromium levels in grasses, trees, and litter,
decreased with increasing distance from the towers. No information
was given as to whether the variations in chromium concentrations
were the result of surface contamination or of true absorption by
the roots of the plant.
The atmospheric deposition of metals and their retention in
ecosystems were studied by Mayer (1983). He measured mean annual
deposition rates in a beech and spruce forest ecosystem in the
Solling (Federal Republic of Germany) in 1974-78 and found that
chromium deposition in the forest canopy was in the range of 13.5 -
15.1 mg/m2 per year; the deposition on the soil below the forest
canopy ranged from 1.6 to 2.3 mg/m2 per year. Thus, up to 80% of
the metals from the atmosphere were retained in the canopy, and 30
- 50% of chromium remained in the noncycling parts of forest
biomass (bark and wood).
Sewage sludge can contain chromium levels as high as 9000
mg/kg. Application of sewage sludge to soils, which increased the
chromium levels from 36.1 to 61 mg/kg on a dry weight basis,
increased the contents of chromium in plants growing in the soil
from, e.g., 2.6 to 4.1 mg/kg in fodder rape (Andersson & Nilsson,
1972). However, most of the increased uptake in plants is retained
in the roots, and only a small fraction appears in the edible part
(Cary et al., 1977). Other elements within the sludge, e.g.,
cadmium or nickel, pose a greater problem for human health (Chaney,
Of particular importance is the chromium concentration in the
forage of meat animals. Kirchgessner et al. (1960) found strong
seasonal variations in the chromium levels in 3 different kinds of
grasses; the highest level found was 590 µg/kg dry weight in hay.
Higher levels of chromium in vegetation not used for human
consumption may account for the generally higher chromium contents
in the organs of wild animals, compared with man (Schroeder, 1966).
Schroeder (1970) determined the chromium concentrations in
different organs and muscles of wild animals and found that they
ranged from 0.04 to 0.48 mg/kg on a wet weight basis. Chromium
concentrations in the hair of several wild-animal species,
collected by Huckabee et al. (1972), ranged from 640 mg/kg in a
pronghorn antelope living in Lemhi Range, Idaho, USA, to about 0.6
mg/kg in a coyote, sampled in Jackson Hole, Wyoming, USA.
3.1.6. Environmental contamination from natural sources
No data have been found that indicate any significant
contamination of the environment from natural sources, though major
catastrophic events, such as large forest fires or volcanic
eruptions, could conceivably contribute to the concentration of
chromium in air. Water supplies originating in areas with chromium
deposits may contain elevated chromium concentrations (section
4.1.2). However, none of these natural sources contributes enough
chromium to pose a hazard for human or animal health.
3.2. Production, Consumption, and Uses
The world's mining production of chromium ore was approximately
9.73 million tonnes (gross weight) in 1980; it fell to 9 million
during 1981 (Thomson, 1982), but rose again to 11 million tonnes in
- Exact data on the yield of elemental chromium are not
available, but may range around half of the gross weight of the
The major uses and amounts of chromium used in the USA in 1968
in thousands of tonnes were: transportation, 77; construction
products, 105; machinery and equipment, 72; home appliances and
equipment, 25; refractory products, 68; plating of metals, 20;
pigment and paints, 15; leather goods, 10; and other uses, 66;
giving a total of 458 thousand tonnes.
The principal uses of chromium are in the metallurgical
processing of ferrochromium and other metallurgical products,
chiefly stainless steel, and, to a much lesser extent, in the
refractory processing of chrome bricks and chemical processing to
make chromic acid and chromates.
Chromates are used for the oxidation of various organic
materials, in the purification of chemicals, in inorganic
oxidation, and in the production of pigments. A large percentage
of chromic acid is used for plating. Dichromate is converted to
chromic sulfate for tanning. Fungicides and wood preservatives
consume an estimated 1.3 million kg of chromium annually.
Chromates are used as rust and corrosion inhibitors, for example,
in diesel engines. Because chromite has a high melting point and
is chemically inert, it is used in the manufacture of bricks for
lining metallurgical furnaces.
In 1981, the demand for chromium was at its lowest level since
- However, on the basis of 1978 figures, the demand for
chromium is expected to increase at an annual rate of about 3.2%,
up to 1990. While the level of stainless steel production will
continue to be the principal influence affecting markets for
chromite and ferrochrome, other factors could have a significant
impact on future trends, e.g., purchases for government stockpiles
(in 1981, the USA had a stockpile of 1.48 million tonnes, and
France and Japan announced the build up of stockpiles) or the
development of new alloys and steels (Thomson, 1982).
3.3. Waste Disposal
Substantial amounts of chromium enter sewage-treatment plants
in major cities. Klein et al. (1974) estimated a total daily
chromium burden for New York city treatment plants of 676 kg, of
which 43% came from electroplating, 9% from other industries, 9%
from runoff, 11% from unknown sources, and 28% from residential
homes. This waste from one city alone (amounting to 2.4 x 105
kg/year), if untreated, would add a significant burden to the
ocean, in comparison with the estimated global natural chromium
mobilization by weathering of 3.6 x 107 kg/year (Bertine &
Goldberg, 1971). The high chromium discharge from homes is
difficult to explain; it has been suggested that this could arise
from the corrosion of stainless steel or the use of waste disposal
units in domestic sinks. The contribution from excreta, estimated
at 100 µg chromium/day per person, should not exceed 1 kg/day for
the 10 million people in the New York area.
The concentration of chromium in the waste-water received at
the New York treatment plants varied between 40 and 500 µg/litre;
this range is probably representative of the chromium discharge in
major cities. The removal of chromium from the waste-water was
studied by Brown et al. (1973). Primary sewage treatment removed
only 27%, secondary treatment using a trickling-filter method
removed 38%; the most effective secondary treatment method
(activated sludge) removed 78%. In another study of a treatment
plant (Chen et al., 1974), the primary effluent contained 300
µg/litre and the secondary effluent, after the activated sewage
sludge and sedimentation process, 60 µg/litre (80% removal). The
final discharge from the plant, a mixture of primary and secondary
effluent and digested sludge had quite a high chromium content of
200 µg/litre. This level is substantially higher than the natural
chromium content of surface water and represents a significant
source of contamination.
Waste-waters from chromium industries contain very high levels
of chromium, ranging from 40 mg/litre (leather industry) to 50 000
mg/litre (chromium plating) (Cheremisinoff & Habib, 1972). These
levels must be reduced by precipitation before the waste-water can
be discharged. The steps include reduction of hexavalent to
trivalent chromium at an acidic pH, followed by precipitation of
the hydroxides at pH 9.5 (Ottinger et al., 1973). The precipitates
containing chromium and other metals are then collected in settling
ponds and disposed of by landfill, incineration, or dumping in the
ocean (US EPA, 1980). If the last procedure is used, the waste-
water treatment itself will contribute to the contamination of the
Landfill and sewage sludge operations are, in turn, potential
sources of contamination of soil and groundwater by chromium.
However, at alkaline pH values, chromium hydroxides are insoluble
and leaching by any but very acidic water should be minimal.
Pohland (1975) did not detect any measurable concentrations of
chromium in the leachate from a simulated landfill.
Similary, the chromium in sewage sludge is very poorly soluble.
Berrow & Webber (1972) found a mean concentration of only 22
mg/litre (range, < 0.9 - 170 mg/litre) in samples of 42 sludges
extracted with 2.5% acetic acid. This represented 0.7 - 8.5% of
the chromium concentration in the original sludge. As acetic acid
is a good complexing and extracting agent for chromium, the
reported levels of extractable chromium are probably much greater
than those resulting from extraction with water at near neutral pH.
However, sludge application to land does increase the chromium
content of the soil (LeRiche, 1968). The application of 66
tonnes/hectare each year, for 19 years, resulted in an increase in
the acetic acid-soluble chromium in the soil from 0.9 to 2.6 mg/kg,
7 years after sludge application was discontinued. This
extractable chromium is presumably available to the plant. The
final link in the cycle of the soil-chromium derived from sewage
sludge is not well known. Undoubtedly, some will be removed by the
growth of vegetation (section 3.1.5). The rate of migration into
ground water depends on the properties of the soil and climatic
conditions. Thus, it is not surprising that, in one study
(LeRiche, 1968), a very slow rate of disappearance was reported
(reduction of extractable chromium from 2.8 to 2.6 mg/kg in 8
years), whereas in another, there was a very rapid rate of
disappearance (reduction of total chromium from 118 to 30 mg/kg, in
3 years) (Page, 1974).
3.4. Miscellaneous Sources of Pollution
As discussed earlier, waste-waters from residential areas in
New York carried approximately 200 kg of chromium daily to the
treatment plants. Of this amount, only 1 kg can be accounted for
by the human excreta of 10 million persons. If a water use of 200
litres per person and a (high) chromium content of 10 µg/litre is
assumed, this concentration would account only for an additional 20
- The origin of the rest is unknown (section 3.3). It should be
pointed out that analytical accuracy is difficult to achieve in
chromium analysis (section 2.2.2) and will affect the results of
all "balance" calculations.
It is evident that the chief source of air pollution with
chromium is ferrochromium refining. Appreciable, but far smaller
emissions, come from refractory operations and inadvertent sources.
The lowest emissions come from the chemical processes in the
production of dichromate and other chrome chemicals. Combustion of
coal and oil, and cement production, large-scale, spray-painting
operations (e.g., ships and planes) and glass plants constitute
other major sources of chromium emissions.
3.5. Environmental Transport and Distribution
Industrial effluents containing chromium, some of which is in
the hexavalent form, are emitted into streams and the air. Whether
the chromium remains hexavalent until it reaches the ocean depends
on the amount of organic matter present in the water. If it is
present in large quantities, the hexavalent chromium may be reduced
by, and the trivalent chromium adsorbed on, the particulate matter.
If it is not adsorbed, the trivalent chromium will form large,
polynucleate complexes that are no longer soluble. These may
remain in colloidal suspension and be transported to the ocean as
such, or they may precipitate and become part of the stream
sediment. Similar processes occur in the oceans: hexavalent
chromium is reduced and settles on the ocean bed. It is replaced
by an estimated 6.7 x 106 kg of chromium from rivers (Bowen, 1966).
In a study of the oxidation state of chromium in ocean water, Fukai
(1967) detected an increased proportion of the trivalent form with
Chromium is emitted into the air, not only by the chromium
industries, but also by every combustion process, including forest
fires. The oxidation state of chromium emissions is not well
defined quantitatively, but it can be assumed that the heat of
combustion may oxidize an unknown proportion of the element to the
hexavalent state. While suspended in the air, this state is
probably stable, until it settles down and comes into contact with
organic matter, which will eventually reduce it to the trivalent
form. Living plants and animals absorb the hexavalent form in
preference to the trivalent, but once absorbed, it is reduced to
the stable, trivalent state.
The transport of chromium in the environment is summarized in
Fig. 1. It should be noted that there is a complete cycle from
rocks or soil to plants, animals, and man, and back to soil. Only
part of the chromium is diverted to a second pathway leading to the
repository, the ocean floor. This part consists of chromium from
rocks and soil carried by water (concentrations, a few µg/litre)
and animal and human excreta, a small part of which may find their
way into water (e.g., runoff from sewage sludge). Another cycle
consists of airborne chromium from natural sources, such as fires,
and from the chromate industry. This cycle also contains some
hexavalent chromium, with byproducts going into the water and air.
Part of the air-chromium completes the cycle by settling on the
land, but a very significant portion goes into the repository, the
ocean, where it ends up as sediment on the ocean floor.
- ENVIRONMENTAL LEVELS AND HUMAN EXPOSURE
4.1. Environmental Levels
Chromium concentrations in air vary with location. Background
levels determined at the South Pole ranged from 2.5 to 10 pg/m3 and
are believed to be due to the weathering of crustal material (US
NAS, 1974a). Data collected by the US National Air Sampling
Network in 1964 gave the national average concentration for
chromium in the ambient air as 0.015 µg/m3, varying from non-
measurable levels to a maximum of 0.35 µg/m3. Chromium
concentrations in most non-urban areas and even in many urban areas
were below detection levels. Yearly average concentrations for
cities in the USA varied from 0.009 to 0.102 µg/m3. Concentrations
ranging from 0.017 to 0.087 µg/m3 have been reported for Osaka,
Japan (US EPA, 1978). The chromium content of the air in the
vicinity of industrial plants may be higher. In 1973, the reported
chromium concentrations ranged from 1 to 100 mg/m3 for coal-fired
power plants, from 100 to 1000 mg/m3 for cement plants, from 10 to
100 mg/m3 for iron and steel industries, and from 100 to 1000 mg/m3
for municipal incinerators (US EPA, 1978). Ferrochromium plants
have the highest emission rates (Radian Corporation, 1983).
However, modern chromium chemical plants contribute very little to
pollution today, because of the installation of collecting
equipment that returns the material for reuse. Drift from cooling-
towers contributes to atmospheric pollution, when chromium is used
as a corrosion inhibitor (section 3.1.5). Little information
exists on the particle size distribution of chromium in the air.
The mass median diameter in a study in the United Kingdom was
1.5 µm (Cawse, 1974).
The chemical form of chromium in air depends on the source.
Chromium from metallurgical production is usually in the trivalent
or zero state. During chromate production, chromate dusts can be
emitted. Aerosols containing chromic acid can be produced during
the chrome-plating process; chromate is also the form found in air
contaminated by cooling-tower drift.
Except for regions with substantial chromium deposits, the
natural content of chromium in surface waters and drinking-water is
very low, most of the samples containing between 1 and 10 µg/litre
(US NAS, 1974a). Substantially higher concentrations are almost
always the result of human activities, reflecting pollution from
industrial activities or sewage waste (Perlmutter & Lieber, 1970).
Thus, the chromium concentration in untreated surface water
supplies reflects the extent of the industrial activity in an area
Table 6. Chromium in water supplies
Country Chromium Range Reference
flat district -a 7 - 8 Novakova et al.
hilly district - 60 - 215 (1974)
Great Lakes 1 0.2 - 19 Weiler & Chawla
Ottawa River 0.01 - Durum & Haffty
Yellow River undetectable - Chen Bingheng
Tia-Ding country - 0 - 80 (personal
surface water communication, 1986)
Rhine River 18 DeGroot & Allersma
Wisla - 31 - 112 Pasternak (1973)
Illinois River 21 5 - 38 Mathis & Cummings
Lake Tahoe < 0.07 - Bond et al. (1973)
Mississipi River 3 - 20 Bond et al. (1973)
New York area, 1250 - Lieber et al. (1964)
a No data available.
Drinking-water from 100 public water supplies in the USA had a
median chromium content of 0.43 µg/litre, ranging from non-
detectable concentrations to 35 µg/litre. In the Federal Republic
of Germany, levels in about 90% of drinking-water samples from 1062
public water supplies were below 0.5 µg/litre; in 1.4% of the water
samples, levels exceeded the prescribed limit value of 50 µg/litre
(Kempf & Sonneborn, 1981).
Both trivalent and hexavalent forms of chromium occur in water.
National and international drinking-water standards reject
drinking-water containing hexavalent chromium concentrations of
more than 50 µg/litre. Such high concentrations occur naturally,
only in areas with substantial chromium deposits (Novakova et al.,
1974); in all other regions they would be caused by industrial
The available food data (Schroeder et al., 1962; Schlettwein-
Gsell & Mommsen-Straub, 1971; Toepfer et al., 1973; Kumpulainen et
al., 1979) indicate a range of the chromium concentrations in
different foodstuffs of 5 -250 mg/kg (Table 7). Highly refined
foods, such as sugar and flour of low extraction, contain the
lowest levels. Very high concentrations have been reported in
pepper (Schroeder et al., 1962) and brewer's yeast.
Table 7. Ranges of chromium concentrations in some
Food Chromium content
(µg/kg of wet weight)
Flour, refined < 20
Meat (beef, pork, chicken) 10 - 60
Fish, fresh < 10 - 10
Vegetables 5 - 30
Whole Milk 10
Cheeses 10 - 130
Sugar, refined < 20
Egg yolk 200
a From: Koivistoinen (1980).
4.2. General Population Exposure
4.2.1. Food and water
The chromium intake from diet and water varies considerably
between regions (Table 8). However, these variations should be
interpreted with caution because not all the analyses have been
controlled by the use of standard reference material or proper
quality assurance procedures, and discrepancies in methods cannot
be completely discounted.
Table 8. Chromium intake from diet and water
Region Chromium Remarks Reference
Canada 189 - Canada, National
(136 - Health and Welfare
Germany, Federal 62 DAa; 4 subjects, Schelenz (1977)
Republic of (11 - 1 week
Japan 723 urban adults Murakami et al.
(202 - (1965)
943 rural adultsb Murakami et al.
(> 180 - (1965)
New Zealand 81 ± 32 DAa; 11 women, Guthrie (1973)
(39 - 190) self-selected
United Kingdom (80 - 100) Facer, J.L.
USA 52 DAa Levine et al.
(5 - 115) (1968)
USA 78 ± 42 DA; 28 diets, Kumpulainen
(25 - 224) complete et al.
Tatar (88 - 126) childrend Goncharov (1968)
a DA = Direct analysis of composite diets as consumed.
b Analysis of composite of cooked servings for one complete day
collected from 10 families in different localities.
c Personal communication to Dr M. Mercier, IPCS (United Kingdom
Ministry of Agriculture, Fisheries and Food, London).
d Analysis of diets in kindergartens.
Most reported chromium intakes range from 50 to 200 µg/day.
However, a comparison of the chromium levels reported by different
investigators reveals substantial differences, some of which may be
due to the influence of the location where the foods were grown.
Only one study (Kumpulainen et al., 1979) was controlled by the use
of standard reference materials. The data should therefore be
treated as preliminary. Furthermore, data concerning total
chromium concentrations do not include information on the species
of chromium in the food and their biological availability. In an
attempt to estimate the biologically available chromium in food,
Toepfer et al. (1973) measured the effects of extracts from foods
on the potentiation of insulin action in epididymal fat tissue in
vitro. No correlation was found between the insulin potentiation
and the total chromium extracted from the foods by acid hydrolysis,
but a significant correlation ( P = 0.01) appeared between the
ethanol-extractable amount of chromium and biological activity.
The highest concentrations of ethanol-extractable chromium were
found in brewer's yeast, black pepper, calf liver, cheese, and
4.2.2. Other exposures
Since chromium compounds are increasingly present in products
used in daily life, chromium eczemas are often observed in the
general population. Polak et al. (1973) surveyed the most
important chromium-containing materials or objects: chromium ore,
baths, colours, lubricating oils, anti-corrosive agents, wood
preservation salts, cement, cleaning materials, textiles, and
leather tanned with chromium. According to Polak et al. (1973),
people who work with material containing mere traces of chromium
salts are more at risk than workers who come into contact with high
concentrations of chromium salts. Some less frequently occurring
cases include sensitization by tattooing (especially green and
light-blue)( Tazelaar, 1970), artificial dentures made of chromium-
containing steel, metal pins used for internal fixation of broken
bones, and bullets retained in the body (Langard & Hensten-
4.3. Occupational Exposure
4.3.1. Inhalation exposure
In chromium ore mines, the concentration of dust in the air in
different work-places ranged from 1.3 to 16.9 mg/m3. In the
crushing and sorting factory, it varied from 6.1 to 148 mg/m3. The
chromium content in settled dust (calculated as Cr2O3) varied from
3.6 to 48%. During the period 1955-69, levels of trivalent
chromium in the dust in different work-places in ferro-alloy
factories ranged from 16 to 42%, while the concentrations of dust
in the air varied from 14 to 38 mg/m3 (Pokrovskaja & Shabynina,
In the past, the production of refined ferrochromium led to
high concentrations of dust in the air of the work-place (10 - 30
mg/m3) (Velichkovsky & Pokrovskaja, 1973). The concentrations of
hexavalent chromium after implementation of a number of sanitation
and hygienic measures were 0.03 - 0.06 mg/m3 (Velichkovsky &
In the manufacture of chromates, the oxidation state,
solubility, and composition of air-borne material varied in
different areas of the plant. Exposure in the ore-crushing area
was to trivalent, insoluble particulates; in the leaching area,
exposure was to tri- and hexavalent, soluble and insoluble
particulates and droplets; at the dry end of the process, the
workers were exposed to the very soluble hexavalent chromates in
particulate form, and to the insoluble residue after leaching
(Velichkovsky & Pokrovskaja, 1973). In plants using calcium in the
roasting process, the residue that is recycled contains, among
other products, calcium chromate, currently believed, on the basis
of animal studies, to be at least partly responsible for the
carcinogenicity of chromium.
Chromium plating of metal surfaces was accompanied by the
release of hexavalent chromium into the air in work premises, in
concentrations ranging from 0.04 to 0.4 mg/m3 (Yunisova &
Pavlovskaja, 1975). In one electroplating factory, the
concentration of chromic acid vapours in the air varied from 0.1 to
1.4 mg/m3 (Gomes, 1972). In the vicinity of 3 different baths in a
Swedish chromium plating factory, chromium concentrations ranged
from 20 to 46 µg chromium (VI)/m3, while, at another factory, the
exposure levels near all baths were below 1 µg/m3 (Lindberg et al.,
Occupational exposure to chromium during welding has been
analysed and the results published by several authors (Stern,
1981). Welding of metals using chromium and nickel electrodes
require high temperatures that melt both the material welded and
the electrode, producing a complex mixture of gases, oxides, and
other compounds, the chemistry of which is determined by the
technology, materials, and welding parameters used in each case
(Lautner et al., 1978). Hexavalent chromium compounds were found in
the respiratory zone of the welder at concentrations ranging from
3.8 to 6.6 µg/m3 (Migai, 1975). For the welding industry as a
whole, the average exposure arising from welding is not homogeneous
but depends on the type and conditions of the welding process
In a cement-producing factory, the concentration of hexavalent
chromium in the air in the work-place varied from 0.0047 to 0.008
mg/m3. The presence of chromium was explained by the fact that the
lining of the kilns was composed of chrome-magnesium bricks
containing 17 - 28% chromium compounds (Retnev, 1960). Forty-two
types of American cement were analysed for total chromium content
and particularly for hexavalent chromium. It was found that
hexavalent chromium was present in 18 out of 42 samples in
concentrations varying from 0.1 to 5.4 g/kg cement, while the total
chromium content ranged from 5 to 124 g/kg (Perone et al., 1974).
Analysis of 59 samples of Portland cement from 9 European countries
showed that the contents of hexavalent chromium extractable with
sodium sulfate varied from 1 to 83 g/kg of cement, while the total
chromium contents ranged from 35 to 173 g/kg (Fregert & Gruvberger,
4.3.2. Dermal exposure
Occupational dermal exposure can result in percutaneous
absorption and in harmful effects on the skin (section 8.3.1),
though the percutaneous absorption of chromium (III) sulfate has
been questioned by Aitio et al. (1984).
Chromium, especially chromate, is the most common contact
allergen and of great importance in occupational contact dermatitis
(Thormann et al., 1979).
Chromium eczema occurred most frequently in building labourers
followed by painters, galvanizers, machine drillers, metal-workers,
graphic artists, and workers in the timber, chemical, leather, and
textile industries (Polak et al., 1973). This is likely to reflect
the exposure to chromium compounds from a large number of every-day
products (section 4.2.2). The skin exposure to cement may be of
particular importance as building labourers belong to the most
- KINETICS AND METABOLISM
5.1.1. Absorption through inhalation
18.104.22.168. Animal studies
Few animal studies have been performed to determine the
absorption of chromium compounds via inhalation. In one early
study, mice and rats were exposed to chromium particulates in an
inhalation chamber for various periods of time. The concentrations
of soluble chromium (CrO3) in air were between 1 and 2 mg/m3 for
the mice and 2 and 3 mg/m3 for the rats. The concentrations of
soluble versus insoluble chromium in the lung tissue of the mice
varied greatly. The soluble chromium concentrations ranged from
4.3 to 10.7 µg/kg dry tissue, after 100 weeks of exposure (Baetjer
et al., 1959a).
The amount of chromium that is absorbed through inhalation
depends on the size of the particles and droplets, on their
solubility in body fluids, and on their reaction with the
respiratory mucosa. Particles greater than 5 µm in diameter
(aerodynamic size) are deposited on the mucosal surface of the
nasal membrane, trachea, and bronchi and are carried by the action
of the cilia to the pharynx, where they are swallowed. Smaller
particles and droplets, especially those below 2 mm in size,
penetrate to the alveoli. Particles and droplets of soluble
compounds, such as hexavalent chromium compounds, are rapidly
absorbed in the blood. Insoluble particles, such as chromite, are
taken up by macrophages and slowly cleared. Soluble materials that
react with the constituents of the lung tissue, such as soluble
trivalent compounds, are also cleared slowly. Baetjer et al.
(1959b) were the first to describe the differences in the clearance
rates of soluble chromates and chromic chloride, when injected
intratracheally into the lungs of animals. The hexavalent chromate
was more rapidly transported from the lungs to other tissues than
the trivalent chromic chloride. Ten minutes after injection, only
15% chromium (IV) remained in the lung compared with 70% chromium
(III). After 60 days, the corresponding figures were 1.7% and 13%
(Baetjer et al., 1959b). Hexavalent chromium is taken up by the
red blood cells in much larger quantities than trivalent chromium.
This finding has been confirmed by Wiegand et al. (1984b)
performing intratracheal instillation (Na251CrO4) studies on
anaesthetized rabbits, as shown in Fig. 2. Confirmation of the
macrophage uptake of insoluble chromate was obtained by exposing
hamsters to 0.5 - 1 mg chromic oxide dust/m3 for 4 h. The median
diameter of the particles was 0.17 µm. Over 90% of the oxide was
found in the macrophages (Sanders et al., 1971).
22.214.171.124. Human data
A mean chromium concentration of 0.22 mg/kg wet weight was
found in the lung tissue of subjects from various locations in the
USA, but there was no correlation between chromium levels in the
lungs and those in the air (Schroeder et al., 1962).
A Committee of the National Research Council (US NAS, 1974a)
concluded: "It is unlikely that the intake from air under ordinary
conditions contributes significantly to the total intake of
available chromium; the intake from the air is calculated to be
less than 1 µg/day; but excessive exposure to airborne chromium
does result in some increased intake".
5.1.2. Absorption from the gastrointestinal tract
The absorption of ingested chromium compounds can be estimated
by measuring the amount of chromium excreted in the urine, as
almost all of intravenously injected chromium is excreted via the
urine and only 2% is found in the faeces. Although a potential loss
of endogenous chromium via the skin and its annexa has not yet been
measured and quantified, it can be stated that this organ, as well
as the gastrointestinal tract are of minor importance in the
excretion of endogenous chromium. The gastrointestinal tract is,
of course, the major organ for the excretion of exogenous chromium.
When considering the gastrointestinal absorption of chromium,
it is essential to recognize the substantial differences in the
efficiency of absorption of trivalent and hexavalent compounds.
These differences exist in both man and animals. Many trivalent
chromium compounds are so poorly absorbed that they have been used
as faecal markers in man and animals. The absorption of hexavalent
chromium, administered orally, was higher in all species examined,
but did not exceed 5% of the dose (Donaldson & Barreras, 1966). No
physiological regulation has yet been established for chromium
126.96.36.199. Animal studies
The gastrointestinal absorption of chromate in rats has been
reported to be between 3 and 6% of a tracer dose (MacKenzie et al.,
1958; Byerrum, 1961). As in man, trivalent chromium compounds are
less well absorbed in the rat, with reported efficiencies ranging
from less than 0.5% (Visek et al., 1953) to 3% (Mertz et al.,
1965a). Within the category of trivalent compounds, there are
moderate differences in absorption, depending on the chemical form.
Binding of the chromium ion to suitable ligands, such as certain
organic acids, stabilizes the metal against precipitation in the
alkaline milieu of the intestines and increases absorption
efficiency by a factor of 3 - 5 times, compared with that for
chromium chloride. This has been shown for certain chelating
agents (Chen et al., 1973), a yet unidentified small peptide
complex isolated from yeast (Votava et al., 1973), and synthetic
glucose tolerance factor (GTF), a dinicotinic-acid-glutathione-
chromium complex (Mertz et al., 1974). Nothing is known about the
interaction of chromium with the flora of the gastrointestinal
tract. Absorption of chromium chloride by ruminant species is
similar to that in rats, with a mean efficiency of 0.76% (Anke et
al., 1971); laying hens have been found to absorb almost 15% of a
tracer dose of the element (Hennig et al., 1971).
188.8.131.52. Human studies
Donaldson & Barreras (1966) studied the gastrointestinal
absorption of hexavalent chromium by administering trace doses of
Na251CrO4 orally to 6 volunteer patients, who were hospitalized,
and by measuring the amount of radioactivity in the faeces and
urine. The mean urinary excretion, expressing the absorption
efficiency, was 2.1 ± 1.5% of the dose given. Administration by
jejunal infusion in 4 volunteers increased these values, suggesting
reduction of the chromate to trivalent compounds by the acid
content of the gastric juice. The same authors reported a mean
absorption efficiency of only 0.5 ± 0.3% for trivalent chromium,
administered as CrCl3 x 6H2O, with a range of 0.1 - 1.2%.
On the basis of the chromium content in diets (60 µg) and
chromium excretion (0.22 µg) in healthy subjects, Anderson et al.
(1983) calculated a minimum chromium absorption of about 0.4%.
Increasing intake by supplementation with chromium (chromic
chloride tablets, furnishing 200 µg chromium/day) led to an
excretion of 0.99 µg, equivalent to 0.4% of the intake.
Aitio et al. (1984) investigated the intake and urinary
excretion of chromium (III) in leather tanning workers. The
environmental concentrations were recorded as low, but chromium was
present in air in the form of large droplets that were not
collected by the standard air measurement technique. It was assumed
that the large droplets were cleared by the upper respiratory tract
and swallowed, and that the chromium in the droplets was absorbed
from the gastrointestinal tract. A calculation showed that this
would explain the urinary excretion levels. No absorption of
chromium through the skin was detected.
In a recent study, the minimum chromium absorption calculated
on the basis of urinary-chromium excretion was about 0.4%.
Increasing intake 5-fold, by chromium supplementation, led to a
nearly 5-fold increase in chromium excretion, suggesting that the
extent of absorption of supplemental inorganic chromium was similar
to that from normal dietary sources (Anderson et al., 1983a).
A similar absorption for trivalent chromium of 0.69% was
reported by Doisy et al. (1968) in healthy human subjects,
regardless of age. However, a group of 14 insulin-requiring
diabetic patients absorbed 4 times as much of the chromium dose as
the non-diabetic or maturity-onset diabetic subjects, as shown by
elevated levels of 51chromium in blood plasma and urine (Doisy et
5.2. Distribution, Retention, Excretion
5.2.1. Animal studies
Most animal studies on chromium metabolism have been performed
on rats. From the site of intestinal absorption, chromium is taken
up by plasma-protein fractions. Small, physiological doses of
51chromium have been shown to bind almost entirely to the iron-
binding protein, transferrin (Hopkins & Schwarz, 1964). On the
other hand, inhaled chromium (Glaser et al., 1984) was bound to
albumin rather than to transferrin. With larger quantities of
trivalent chromium, non-specific binding to other proteins also
occurred, but not to the red blood cells. Visek et al. (1953)
measured the effects of the different chemical forms of chromium on
tissue distribution and found that soluble, chelated forms, such as
acetate or citrate complexes, were cleared quite rapidly, in
contrast with colloidal or protein-binding forms (chromite, chromic
chloride), which have a great affinity for the reticulo-endothelial
system (bone marrow, liver, spleen), and clear more slowly. The
blood clearance of hexavalent chromium, such as chromate, was slow,
because of irreversible binding within the red blood cells. Tissue
distribution of 51chromium, administered in nanogram doses to rats
was studied by Hopkins (1965). As in the preceeding studies, the
element accumulated in bone, spleen, testes, and epididymis; much
less was retained in the lungs, brain, heart, and pancreas. This
obvious difference in chromium distribution between man and rats is
As in man, trivalent 51chromium in the rat was rapidly cleared
from the blood, after absorption, and was retained by the tissues
(Mertz et al., 1965a). These tissue stores, labelled with
51chromium chloride, administered orally or intravenously, were not
immediately available for specific physiological functions. For
example, 51chromium, administered as CrCl3 x 6H2O to pregnant rats,
was not transported into the embryos (Mertz et al., 1969), nor did
any 51chromium appear in the blood in response to glucose or
insulin injections (Mertz & Roginski, 1971). The fact that fetal
chromium concentrations are low, when the pregnant rats are fed a
low-chromium Torula yeast diet, and increase when a high-chromium
natural stock ration is fed, indicates that placental transport and
possibly, the acute chromium response depend on a special form of
chromium, which is different from chromium chloride. It is
possible, but has not yet been proved, that this form is the
dinicotinic acid-glutathione-chromium complex, known as glucose
tolerance factor. Yeast extracts containing this factor labelled
with 51chromium have been shown to cross the placenta (Mertz et
al., 1969) and, in preliminary studies, to furnish chromium for the
acute chromium response (Mertz & Roginski, 1971).
With reference to interactions between chromium and other trace
elements, competition with iron by way of their common carrier
(transferrin) has been suggested in rats (Hopkins & Schwarz, 1964)
and in human beings (Sargent et al., 1979). Goncharov (1968)
reported a close interaction between chromium and dietary iodine.
In iodine-deficient white rats, addition of chromium to the diets
in amounts supplying from 0.6 to 600 µg/animal per day stimulated
thyroid function, as indicated by morphological and functional
changes. Conversely, chromium, in all but the lowest dose,
decreased thyroid function in animals receiving adequate iodine
levels. This relation is in agreement with epidemiological data
from the USSR (Goncharov, 1968).
5.2.2. Human data
184.108.40.206. Concentration in tissues, blood, urine, and hair
including possible biological indicators of exposure
The most comprehensive survey of tissue-chromium concentrations
is that of Schroeder et al. (1962), who carried out a
spectrographic analyses on 20 - 39 samples for each autopsy tissue,
all of which had been carefully collected to avoid extraneous
contamination. The following results were obtained (mean values in
mg/kg ash) for a group of subjects who had died between the ages of
30 and 40 years: lung, 15.6; aorta, 9.1; pancreas, 6.5; heart, 3.8;
testes, 3.1; kidney, 2.1; liver, 1.8; spleen, 1.7. In all tissues,
except for the lungs there was a rapid decline in chromium
concentrations from time of birth to the age of 10 years, followed
by a more gradual decrease to the age of 80 years. It cannot be
stated with certainty whether the decline is an expression of a
physiological mechanism or of a dietary deficiency. The lungs lost
their initially high chromium levels (85.2 mg/kg ash) up to the age
of 20 years (6.8 mg/kg); subsequently, the concentrations increased
to between 20 and 38 mg/kg. This discrepancy demonstrates that the
chromium in lungs is not in equilibrium with the general pool. The
decline of chromium in the aorta, quite pronounced in subjects in
the USA, was much less dramatic in the aortas from subjects of
other countries (Schroeder et al., 1970).
Mancuso & Hueper (1951), as well as Baetjer et al. (1959b),
found concentrations of chromium in the lungs of former chromate
workers that were several orders of magnitude higher than those in
control subjects. In a study on 16 chromate workers, including 11
with lung cancer, Baetjer et al. (1959b), using a colorimetric
method, observed a median concentration of water- and acid-soluble
chromium in the lung of 70 mg/kg dry weight and a median
concentration of acid-insoluble chromium of 17 mg/kg. The chromium
concentration did not differ between cancer cases and non-cancer
cases. The tissue specimens were obtained 0 - 23 years after the
termination of occupational exposure, which had lasted 1.5 - 42
years. With the use of emission spectrometry and atomic absorption
spectrophotometry, Hyodo et al. (1980) found a chromium content of
3.6 mg/kg wet weight in the lung in a male smoker with lung cancer,
who died 10 years after employment for 30 years in a chromate-
producing plant. The concentration in other tissues ranged from
0.05 (bone marrow) to 1.5 mg/kg (suprarenal gland). In five
unexposed controls, lung concentrations ranged from 0.09 to 0.88
mg/kg; concentrations in other tissues ranged from 0.003 (kidney)
to 0.156 mg/kg (suprarenal gland). The ratio of hexavalent to
total chromium in the lungs was 29% in the worker and 22.7 ± 10.6%
(mean ± SD) in the controls.
Brune et al. (1980) gave tissue concentrations for chromium and
other metals in the lung, liver, and kidney of 20 deceased copper
smelter workers, who had retired 0 - 19 years prior to death, and
of a control group (8 subjects). Tissue analysis was carried out
using neutron activation analysis as well as atomic absorption
spectrophotometry, and included a comparison with certified
reference samples (National Bureau Standards, bovine liver). In
the controls, the concentrations ranged from below detection (0.003
mg/kg wet weight) to 0.07 mg/kg in kidneys and 0.11 mg/kg in liver.
There was no marked difference between the chromium contents of
these 2 tissues and those in the exposed workers. On the other
hand, the lung concentrations were between 3 and 4 times higher in
the workers than in the controls (median levels, 0.29 and 0.08
Chromium determinations were carried out on lung and kidney
samples of 45 autopsies from the Northern Bavaria area (Federal
Republic of Germany) (Zober et al., 1984). The analyses were
carried out using electrothermal AAS after wet oxidative digestion.
Median values of 0.097 mg/kg wet weight (range, 0.0006 - 1.230
mg/kg) in lung tissue and 0.0096 mg/kg wet weight (range, 0.0002 -
0.690 mg/kg) in kidney were found.
Very limited data are available on chromium levels in tissues
other than those referred to above. Shmitova (1978) estimated
chromium levels in fetal and placental tissues (abortive material),
derived on the 12th week of pregnancy, and found 92.8 and 30 µg
Cr/kg tissue, respectively.
It can be concluded that chromium may be retained in the lungs,
several years after the termination of occupational exposure.
However, it is not known whether this observation has any
biological relevance to the appearance of lung cancer. In this
context, it is of interest to note that chromium is retained in the
human lung for a relatively long period, even without occupational
The values reported by various investigators for chromium
concentrations in the blood of unexposed human beings range from
0.2 to 70 µg/litre in serum and plasma and 5 to 54 µg/litre in red
blood cells (US EPA, 1978). Insufficient evidence is available to
state whether the blood values in the normal population are
influenced by concentrations in the ambient air. The most
extensive determination of blood-chromium levels in chromate
workers was made by Mancuso (1951). Blood values during exposure
varied from 5 to 170 µg/litre and from 10 to 140 µg/litre, 74 days
after work ended. No decrease in blood-chromium levels was found,
even after 74 days without exposure.
Because of a selective affinity of hexavalent chromium for the
erythrocyte, substantially increased environmental exposure to
chromates is reflected in an increased ratio of the hexavalent
chromium level in red blood cells to that in plasma. Baetjer et
- (1959a) found the following concentrations in 3 exposed
chromate workers: blood cells, 30, 54, and 140 µg/litre; plasma, 0,
20, and 17 µg/litre, respectively. In the absence of exposure to
chromate, the chromium concentrations in erythrocytes and plasma,
as serum, were nearly identical (Paixao & Yoe, 1959). Chromium
(VI) is incorporated into human red blood cells and remains there
over a long period of time (Wiegand et al., 1985), since the
approximate lifetime of human red cells is about 100 days. Exposure
to 2 mg trivalent chromium/day, for 3 months, resulted in an
increased concentration (0.2 mg/kg) in the red cells of 5 human
males, compared with 0.11 mg/kg in 5 controls without supplement
(Schroeder et al., 1962). However, later studies did not show any
increase in chromium concentrations in red cells, following
exposure to trivalent chromium (Beyersmann et al., 1984; Wiegand et
al., 1985). Monitoring of red blood cell-chromium may be a useful
indicator of exposure to hexavalent, but not to trivalent,
Most of the later studies show that the true chromium
concentration in the plasma or serum of healthy subjects is of the
order of 1 µg/litre or less (Guthrie et al., 1978; Versieck et al.,
1978). Seeling et al. (1979) determined chromium in serum and
plasma using flameless AAS. This study was supported by measuring
a standard reference material (National Bureau of Standards, 1569
brewer's yeast, reference data: 2.12 ± 0.05 mg/kg, measured data:
2.3 ± 0.2 mg/kg). The chromium levels in serum ranged from 0.7 to
2.2 µg/litre and in plasma from 1 to 1.5 µg/litre (central
parameters of a long normal distribution).
Pooled serum samples of 6 healthy Finnish volunteers were
studied by Kumpulainen et al. (1983). A mean value of 0.11 ± 0.05
µg chromium/litre (range, 0.06 - 0.20) was found. Nomiyama et al.
(1980b) analysed 20 blood samples of Japanese subjects, using
direct flameless ASS, and reported a value of 2.9 ± 1.7 µg
chromium/litre whole blood.
Zober et al. (1984) analysed the blood of 45 autopsied subjects
from the Northern Bavaria area (Federal Republic of Germany). A
median chromium concentration of 2.8 µg/litre of postmortem blood
(range, 0.20 - 24 µg/litre) was found, the value was not influenced
by age or sex.
A reference material for chromium, bovine serum (RM 8419) is
available from the National Bureau of Standards, Washington DC,
Chromium concentrations in the urine of non-occupationally
exposed subjects have been reported to range from 1.8 to 11
µg/litre (Imbus et al., 1963). Except for exposed persons and
juvenile diabetic patients, the reported values for daily chromium
excretion in urine (Table 9) do not differ as much as those for
blood. In later studies, such as that of Guthrie et al. (1979), a
method of flameless atomic absorption was used in which it was
possible to correct for spectral interference (Zander et al.,
1977). Such interference was difficult to eliminate with earlier
methods (Guthrie et al., 1978). Urine samples from 189 Japanese
volunteers, aged 10 - 80 years, in 4 pollution-free areas, were
analysed for chromium by Nomiyama et al. (1980a), using direct
flameless AAS. The average level was 0.4 ± 0.37 µg/litre (X ± SE)
or 0.47± 0.42 mg/kg creatinine. Urinary chromium tended to be
higher in males than in females and to decrease with age, but the
differences were not significant. Using electrothermal AAS, the
urinary excretion of various metals in non-occupationally exposed
adults was measured by Schaller & Zober (1982). For smokers, a
median value of 1.6 µg chromium/litre was found (non-smoker: 1.4 µg
chromium/litre). Commercially available urine samples were used
for quality control (Angerer et al., 1981). Although a "normal"
level of chromium excretion for healthy, unexposed persons cannot
yet be established with certainty, such a level may be less than 1
Several investigators have measured urinary-chromium excretion
in exposed workers. In a study on chromate workers, the urine
values ranged from 5 to 380 µg chromium/litre during exposure and
from 10 to 54 µg chromium/litre, 74 days after the end of exposure
(Mancuso, 1951). No relationship was evident between the urine
levels and the weighted average number of years of exposure. In
every case where urine values were recorded, both during, and 74
days after the end of, exposure, the chromium concentrations
decreased with time away from exposure.
A study of 12 workers in a galvanizing plant in the Federal
Republic of Germany showed an average concentration of chromium in
urine of 9.5 µg/litre (range, 1.4 - 24.6 µg/litre) compared with a
value of 1.8 ± 1.1 µg/litre in 60 unexposed workers (Schaller et
al., 1972). Gylseth et al. (1977) investigated a group of 14
welders exposed to about 0.05 mg chromium/m3 air, and reported a
urinary chromium concentration of approximately 40 mg/litre. At
the same exposure level, Tola et al. (1977) found a concentration
of 30 mg chromium/kg of creatinine in the urine. In both of these
studies, there was a correlation between recent exposure to
airborne chromium and chromium concentrations in urine. In the
study by Tola et al. (1977), it was shown that the water-soluble
fraction of airborne chromium was better correlated with the
concentration excreted in the urine than with total chromium. It
was also shown that the water-soluble fraction consisted mainly of
Table 9. Daily chromium excretion in urine
Subjects Number Excretion Range Reference
(mean ± SD)
Adult males 2 0.72a 0.58 - 0.86 Schroeder et al.
Adult males, fed 3 31.0a 20.8 - 46.5 Schroeder et al.
2 mg trivalent, (1962)
Normal adults 16 13 ± 6 4 - 24 Voelkl (1971)
Chromate workerb 16 000
Normal adults 60 1.6 ± 1.1 Schaller et al.
Galvano-technical 12 9.7 ± 6.6 1.4 - 24.6 Schaller et al.
Normal young 20 8.4 ± 5.2 Hambidge (1974)
Normal children, 18 5.5 ± 2.9 Hambidge (1974)
8 years old
Insulin-dependent 7 19.2 ± 18.9 Hambidge (1974)
11 years old
Young women 9 7.2 ± 1.2 5.9 - 10.0 Mitman et al.
Adult males 12 0.8 ± 0.4 0.4 - 1.8 Guthrie et al.
Table 9. (contd.)
Subjects Number Excretion Range Reference
(mean ± SD)
Adult males 91 0.48 ± 0.41a 0.42 - 0.53 Nomiyama et al.
Adult females 98 0.34 ± 0.31a 0.22 - 0.43 Nomiyama et al.
Adult males 48 0.20 ± 0.01a 0.05 - 0.58 Anderson et al.
Adult females 28
Adult males 27 0.17 ± 0.10 Anderson et al.
Adult females 15 0.20 ± 0.12 Anderson et al.
Adult males and 299 0.80 ± 0.6a 0.4 - 2.1 Fang (1983)
Normal adults 10 0.11 ± 0.05a 0.06 - 0.20 Kumpulainen et al.
Normal adultsc 10 4.9 0.10 - 14.2 Zober et al.
a Data calculated as µg/litre urine.
b Tanner, suffering from an acute ulceric gasteroenterocolitis.
c Samples taken post-mortem from autopsies. Original values are related
to mass (kg) instead of volume (litre).
Lindberg & Vesterberg (1983a) measured airborne, and urinary-
chromium levels among platers. Concentrations of chromium in urine
of < 5 µg/litre occurred when the time-weighted average values of
exposure were about or below 2 µg/m3 air. Severe damage to the
nasal septum and effects on lung function have not been found at
levels lower than this. It was shown that post-shift urinary-
chromium determinations could be used to monitor exposure in this
The urinary-chromium excretion and chromium clearance in 22
welders, who had been exposed to airborne hexavalent chromium (5 -
150 µg/m3) during 2 - 40 years (mean working time, 18.9 years) were
measured by Mutti et al. (1979). The method used was flameless
atomic absorption spectrometry. A highly significant correlation
was detected between the ratio of urinary-chromium to creatinine
and the airborne chromium concentration in the workplace, with
excretions ranging from 5.3 ± 3.7 to 33.3 ± 6.9 mg chromium/kg
creatinine in slightly exposed and heavily exposed welders,
respectively. The authors also reported an increase in chromium
clearance with increasing body burden of chromium, which indicates
that high urinary-chromium excretion may be caused by previous high
exposure as well as by current exposure.
Baseline data for chromium excretion in unexposed subjects in
the report by Mutti et al. (1979) are approximately 10 times higher
than the values recently proposed and generally accepted as normal.
However, there is reason (section 2.2.2) to accept relative
differences in analytical results in studies by one author using
one method, even if the absolute values reported may be questioned.
To test whether chromium excretion is also associated with
exercise-induced increases in glucose utilization, the urinary
chromium excretion, serum glucose, insulin, and glucagon of 9 male
runners (23 - 46 years old) were evaluated by Anderson et al.
(1982a). The mean urinary-chromium concentration was increased
nearly 5-fold, 2 h after running; excretion of sodium, potassium,
and calcium was unchanged. These data demonstrate an increase in
chromium excretion with exercise-induced increase in glucose
The chromium concentration was determined in 261 samples of
breast milk collected by manual expression from 45 American women.
Chromium levels were measured in whole, liquid milk by graphite-
furnace AAS, using the method of standard additions. The mean
chromium content of the breast milk samples was 0.30 µg/litre. The
range of individual values was 0.06 - 1.56 µg/litre and did not
change significantly with duration of lactation (Casey & Hambidge,
1984). Kumpulainen et al. (1983) analysed frozen samples of pooled
breast milk taken from women in different stages of lactation and
obtained from the Milk Bank of the Children's Hospital of Helsinki.
The mean chromium content ± SD was 0.49 ± 0.067 µg/litre (range,
0.37 - 0.57 µg/litre).
Hair-chromium concentrations in children during the first 6
months of life were significantly higher than at any other age
(Hambidge & Baum, 1972); they declined from an initial value of
1493 µg/kg to an average of 412 µg/kg at 2 - 3 years. Hambidge
(1971) compared the chromium concentration in the hair of 15
newborn babies and that of their mothers: in only one case was the
chromium level in the mother's hair higher than that in the newborn
baby. Chromium levels were significantly lower than those
mentioned above in 50 Turkish women and their newborn babies (203
and 119 µg/kg, respectively) and only 12 newborn babies were found
to have higher concentrations than their mothers, suggesting
suboptimal chromium status (Gürson, 1977). Hair appears to reflect
the nutritional chromium status of groups. The hair-chromium level
is significantly lower in parous women than in nulliparae (Hambidge
& Rodgerson, 1969; Mahalko & Bennion, 1976) and in diabetic
children compared with normal controls (Hambidge et al., 1968). It
is low in adult-onset diabetic adults (Benjanuratra & Bennion,
1975). These findings are in agreement with the expected changes
in chromium balance during pregnancy and in diabetes.
220.127.116.11. Dynamic aspects of metabolism and the influence of
Once chromium is absorbed into the organism, it clears rapidly
from the blood stream and is excreted or taken up by the tissues.
In a clinical study, Sargent et al. (1979) detected a 4-
compartment-type clearance from blood, with mean half-times of 13
min, 6.3 h, 1.9 days, and 8.3 days. However, the disappearance
from 3 tissue compartments was much slower, with half-times of
0.56, 12.7, and 192 days. The half-times for blood 3-compartment
clearance in rats (Hopkins, 1965) were calculated to be 0.56, 5.33,
and 57 h (Withey, 1983).
Whether any organ is specifically responsible for the storage
and release of the "metabolically responsive" chromium is not
known. The "metabolically responsive" chromium in blood is defined
as the fraction that increases acutely in response to an elevation
of blood-glucose or blood-insulin levels. It is believed that this
chromium increment interacts with the increased insulin secreted in
response to a glucose load, to facilitate the action of the hormone
on the insulin receptors of the insulin-sensitive cells.
In young, healthy subjects, but not in elderly subjects and
diabetic patients, an oral glucose load or the injection of insulin
results in a sudden increase in serum- or plasma-chromium
(Glinsmann et al., 1966; Levine et al., 1968; Hambidge, 1971; Behne
& Diel, 1972; Liu & Morris, 1978). This increase may appear 30
min, or as late as 120 min, after the challenge; much of the
chromium responsible for the increase is subsequently lost in the
urine. Lack of this increase, also termed "relative chromium
response", is often associated with impaired glucose tolerance,
indicative of chromium deficiency. It should be noted that, when
the glucose load was given intravenously (Pekarek et al., 1975) to
healthy volunteers, the serum-chromium level decreased rapidly,
while the blood-glucose level increased. Supplementation with
chromium chloride or high-chromium yeast extracts for several weeks
resulted in the reappearance of the relative chromium response and
improvement of glucose tolerance (Glinsmann et al., 1966; Liu &
Morris, 1978). These findings support the conclusion that the
"relative chromium response" measured during a glucose or insulin
tolerance test may serve as an indicator of the adequacy of
metabolically responsive chromium.
Much of the chromium increment secreted into the blood stream
in response to glucose or insulin is subsequently lost in the
urine. Hambidge (1971) observed greatly increased urinary-chromium
excretion in 2 diabetic children after insulin therapy had begun,
compared with the excretion in the same children before insulin
treatment. It is not known whether the increased urinary loss of
chromium is compensated for by an increase in absorption efficiency
from the intestines. Hambidge et al. (1968) reported significantly
lower chromium concentrations in the hair of diabetic children
compared with normal children, and Morgan (1972) found that
the chromium contents in the livers of 31 diabetic adults at
autopsy were lower than those in 24 control livers from non-
diabetic persons (8.57 versus 12.7 mg/kg ash; P = 0.05).
On the other hand, Doisy et al. (1971) demonstrated greatly
increased intestinal absorption, together with elevated chromium
excretion, in 14 insulin-dependent diabetic patients administered
51CrCl3 x 6H2O, orally. It is not known whether the greater
absorption efficiency is adequate to compensate for the increased
urinary losses; the decreased chromium concentrations in hair and
liver, discussed above, suggest that a negative balance may prevail.
Several other pathological conditions affect chromium
metabolism. Sargent et al. (1979) detected significantly less
retention of intravenously administered 51chromium in 11 patients
with haemochromatosis compared with 5 normal controls. This may be
related to the high saturation with iron of transferrin, which is
also the carrier protein for newly absorbed chromium.
Chronic ischaemic heart disease also affects chromium
metabolism. Neiko & Del'va (1978) observed a significantly
increased urinary-chromium loss, greater by a factor of 1.5 - 1.6
than that of normal controls, in 65 heart patients. The urinary-,
and to a lesser extent, the faecal-chromium loss increased
progressively with increasing severity of signs and symptoms and
resulted in a negative chromium balance in the post-infarct state.
The balance became positive again on discharge from the hospital
after medical treatment.
Acute infections also appear to influence chromium metabolism.
Pekarek et al. (1975) measured glucose tolerance, and insulin and
chromium levels in human volunteers, before and after infection
with the benign sandfly fever virus. Impaired glucose tolerance and
a significantly increased insulin response to a glucose load,
observed at the height of the infection, were accompanied by very
significantly depressed serum-chromium levels (0.5 µg/litre,
compared with 1.4 µg/litre before infection; P < 0.05) In
contrast with the sharp decline in serum-chromium following the
intravenous injection of glucose in the healthy state observed by
these authors, the depressed serum-chromium levels declined very
little during the glucose tolerance test at the height of
infection. The mechanism of the changes in chromium metabolism in
heart disease and sandfly fever is not clear. Of great potential
importance is the unanswered question of whether the observations
described here reflect an increased chromium requirement in the
patients or a normal reaction to various forms of stress.
The information on the dynamic aspects of chromium metabolism
in animals is limited and should be considered in connexion with
the more detailed studies on human subjects. Diabetes, induced in
rats by the injection of streptozotocin, affected the tissue
distribution of injected 51CrCl3. Sixteen days after injection of
streptozotocin, 51CrCl3 was injected into 5 diabetic rats and 6
normal controls and the 51Cr content measured 5 days later. The
serum of the diabetic rats contained more than 3 times the 51Cr
activity found in the controls (0.24 versus 0.07% of the injected
dose P < 0.01). Significant differences were also detected in the
distribution of 51Cr in the subcellular fractions of the liver; in
the diabetic tissue, 51Cr activity was higher in the nuclear and
supernatant fractions ( P < 0.01) and lower in mitochondria and
microsomes ( P < 0.05). The mechanism responsible for these
changes is not known (Mathur & Doisy, 1972).
5.3. Influence of Chemical Form
The diverse biological effects of chromium on living organisms
cannot be understood without knowledge of the chemical and physical
forms in which the element is present. As stated earlier, the
metallic state (zero valence) is biologically inert, the trivalent
state represents the essential element chromium, and the hexavalent
state is of concern to the toxicologist. Compounds of trivalent
chromium are poorly absorbed, whereas those of the hexavalent state
easily penetrate physiological barriers, such as cell membranes.
Hexavalent chromium compounds are easily reduced by living matter,
but oxidation of trivalent to hexavalent chromium does not occur in
The physical form of hexavalent compounds (such as particle
size) and chemical properties (such as solubility) determine
metabolic pathways after inhalation and, therefore, health effects.
There is an equally strong influence of the chemical form of
trivalent chromium on metabolism and health effects. When chromium
is bound to water or small anions (e.g., CrCl3 x 6H20), it
precipitates in the neutral or alkaline milieu of the body fluids.
When it is bound to ligands, such as organic acids, the element is
light in solution and is available for intestinal absorption. The
forms in which trivalent chromium occurs in nature are not really
known. Plants probably contain chromium complexes with organic
The biological availability of chromium compounds in foods is
of great nutritional importance, but is poorly defined. One
compound or a group of closely related compounds, glucose tolerance
factor, has been isolated from yeast and shown to be more active
than chromium chloride in genetically diabetic mice and in the in
vitro potentiation of the action of insulin on rat epididymal fat
tissue (section 18.104.22.168). It has been identified as a dinicotinic-
acid glutathione complex, but the exact stereochemical structure is
not yet known (Toepfer et al., 1977). It has been postulated, but
not proved, that this factor is the active form of chromium within
- EFFECTS ON ORGANISMS IN THE ENVIRONMENT
The environmental effects of chromium as a pollutant have been
reviewed by US EPA (1978), Anderson (1982), and EIFAC (1983).
Various effects have been reported, but, because of the presence of
other chemicals, it remains doubtful whether chromium alone is
responsible for the effects observed. Data are available on
microorganisms, plants, and aquatic organisms.
Most microorganisms (protozoa, protophyta, fungi, algae,
bacteria) are able to absorb chromium. The active uptake of
chromate by the sulfate transport system has been shown in
Neurospora crassa (Roberts & Marzluf, 1971). No distinction has
been made between ab- and adsorption in other studies (e.g., algae)
(Calow & Fletcher, 1972), and it has not yet been shown that
chromium is an essential element for microorganisms. In general,
toxicity for most microorganisms occurs in the range of 0.05 - 5 mg
chromium/kg of medium. The internal concentration of chromium
depends on the species. In most groups of microorganisms, it
ranges between the levels of 0.6 mg dry weight present in one litre
of sample of microplankton from Monterey Bay, California, USA, and
21.4 mg/litre phytoplankton collected in the Pacific Ocean (Martin
& Knauer, 1973).
Trivalent chromium is less toxic than hexavalent. The main
features are inhibition of growth (at concentrations greater than
0.5 mg/litre in Chlorella cultures) (Nollendorf et al., 1972) and
inhibition of various metabolic processes, such as photosynthesis
or protein synthesis (US EPA, 1978).
The toxicity of chromium for soil bacterial isolates was
studied by measuring the turbidity of liquid cultures supplemented
with hexavalent chromium and trivalent chromium. Gram-negative
bacteria were more affected by hexavalent chromium (1 - 12 mg/kg)
than gram-positive bacteria. Toxicity due to trivalent chromium
was not observed at similar levels. The toxicity of low levels of
hexavalent chromium (1 mg/kg) indicates that soil microbial
transformations, such as nitrification, may be affected (Ross et
Although chromium is present in all plants, it has not been
proved to be an essential element for plants. Most substances,
including chromium, can be absorbed through either the root or the
leaf surface. Several factors affect the availability of chromium
for the plant (Black, 1968), including the pH of the soil,
interactions with other minerals or organic chelating compounds,
and carbon dioxide and oxygen concentrations.
Little chromium is translocated from the site of absorption;
however, the chelated form is transported throughout the plant
Chromium in high concentrations can be toxic for plants, but
Yopp et al. (1974) stated that there was no specific pattern of
During the smelting of chromite, considerable quantities of
waste are produced, which contain soluble chromates. When combined
with a high pH, Gemmell (1973) showed an inhibition of germination
and growth in white mustard plants (Sinapis alba) growing on waste
heaps. Covering the waste with a 25-to 30-cm layer of granular-
free-draining subsoil followed with layers of soil, peat, or sewage
sludge was shown to be the best revegetation technique (Gemmell,
The main feature of chromium intoxication is chlorosis, which
is similar to iron deficiency (Hewitt, 1953).
Soybeans, treated in nutrient culture containing 0 - 5 mg
hexavalent chromium/litre showed decreased uptake of calcium,
potassium, phosphorus, iron, and manganese (Turner & Rust, 1971).
Death of plants occurred within 3 days of treatment with 30 or 60
A reduction in leaf dry weight occurred after treatment with
0.01 mg hexavalent chromium/litre (Rediske et al., 1955). Chromium
affects the carbohydrate metabolism, and the leaf chlorophyll
concentration decreased with increasing hexavalent chromium
concentration (0.01 - 1 mg hexavalent chromium/litre) (Rediske,
1956). Hexavalent chromium appears to be more toxic than trivalent
chromium (Hewitt, 1953; Stanley, 1974; Verfaillie, 1974). At
present, no data are available concerning the mechanism of action
or the dose-dependent pattern of chromium intoxication.
6.3. Aquatic Organisms
More studies have been performed with aquatic species than with
free-living (non-parasitic) animals. Depending on the species,
chromium can be less toxic for fish in warm water, but marked
decreases in toxicity are found with increasing pH or water
hardness; changes in salinity have little if any effect on its
toxicity. Chromium can make fish more susceptible to infection;
high concentrations can damage and/or accumulate in various fish
tissues and in invertebrates such as snails and worms.
Reproduction of Daphnia was affected by exposure to 0.01 mg
hexavalent chromium/litre (EIFAC, 1983). Numerous other factors
influence the availability of chromium and, therefore, its
toxicity. These include the presence of other minerals and organic
pollutants, and the temperature of the environment; this has been
shown in mice (Nomiyama et al., 1980a).
Hexavalent chromium is accumulated by aquatic species by
passive diffusion (US EPA, 1978). Ecological factors, in the
abiotic and living environment, are involved in this process, which
varies according to the sensitivity of different species. The
physiological state and activity of the fish also affect
accumulation (Reichenbach-Klinke, 1977, 1980). Kittelberger (1973)
analysed the organs and tissues of the roach (Leuciscus rutilus)
from the river Rhine and found that concentrations of chromium in
the spleen, bronchi, and intestine (between 30 and 37.5 mg/kg) were
10 - 30 times higher than those in the heart, skin, muscle, and
LC50s are listed in Table 10 for hexavalent and trivalent
chromium compounds in the aquatic environment.
Table 10. The toxicity of chromium for fresh-water organisms
(expressed as 50% mortality)a
Compound Category Exposure Toxicity range Most
hexavalent invertebrate acute 0.067 - 59.9 scud
chromium long-termb - -
vertebrate acute 17.6 - 249 fathead minnow
long-term 0.265 - 2.0 rainbow trout
trivalent invertebrate acute 2.0 - 64.0 cladoceran
chromium long-term 0.066 cladoceran
vertebrate acute 33.0 - 71.9 guppy
long-term 1.0 fathead minnow
a From: US EPA (1980).
b No data available.
In general, invertebrate species, such as polychaete worms,
insects, and crustaceans are more sensitive to the toxic effects of
chromium than vertebrates, such as some fish (Mathis & Cummings,
1973). The lethal chromium level for several aquatic and
nonaquatic invertebrates has been reported to be 0.05 mg/litre (US
EIFAC (1983) reviewed the literature on the occurrence and
effects of chromium in fresh water and proposed tentative water-
quality criteria that distinguish between salmonid and non-salmonid
waters. To protect salmonid waters, the mean aqueous concentration
of "soluble" chromium should not exceed 0.025 mg chromium/litre,
and the 95 percentile should not exceed 0.1 mg chromium/litre.
However, more stringent values may be necessary in very soft, acid
waters, and less stringent values in alkaline waters.
- EFFECTS ON EXPERIMENTAL ANIMALS AND IN VITRO TEST SYSTEMS
7.1. Nutritional Effects of Chromium
The criteria for an essential nutrient have been defined in
different ways by different authors, but all definitions postulate
that a reduction in the total daily intake of the nutrient below a
certain level must consistently induce signs of deficiency, and
that the supplementation of the daily intake above this level must
prevent and cure the deficiency signs. Chromium deficiency has
been produced experimentally in mice, rats, and squirrel monkeys,
and the full reversal of the deficiency signs by the oral
administration of chromium has been demonstrated in rats (Mertz,
1969). For these reasons, chromium must be considered an essential
micro-nutrient. It is physiologically active in the trivalent
oxidation state at concentrations of approximately 100 µg/kg diet.
7.1.1. Effects of deficiency on glucose metabolism
Semipurified rations, containing Torula yeast as the source of
protein, were fed to groups of 10 male Sprague Dawley rats at
weaning, and intravenous glucose tolerance tests were performed by
injecting 1250 mg glucose/kg body weight and measuring the
subsequent decline in blood-glucose levels. The rate of glucose
disappearance can be calculated from the straight line plot of log
increment glucose (excess of glucose at any time (t) over fasting
glucose), versus time. The disappearance constant k is expressed
as % decline of the increment glucose per min; it is a measure of
the efficiency of glucose utilization. In weaning rats, the rate
constant was found to decline, within 3 weeks or less, from an
average of 4%/min to 2.6%/min in animals fed the Torula yeast diet,
but not in animals administered a diet in which 4 or 8% of Torula
yeast was replaced by an equal amount of brewer's yeast. This
observation suggested that brewer's yeast, but not Torula yeast,
contained an unknown substance necessary for the maintenance of
normal glucose tolerance. Because the only known effect of the
substance was that on glucose tolerance, it was named "Glucose
Tolerance Factor" (GTF) (Mertz & Schwarz, 1959). GTF was extracted
from brewer's yeast and pork kidney powder, concentrated and
purified, and its active ingredient was identified as trivalent
chromium (Schwarz & Mertz, 1959). Chromium in the form of most
common complexes (except for very stable ones) cured the impairment
of glucose tolerance in deficient rats, either as one oral dose of
200 µg/kg body weight, or as an intravenous (iv) injection of 2.5
µg/kg body weight. Chromium in the diet also prevented the
impairment of glucose tolerance.
In order to produce more pronounced deficiencies of chromium
and other trace elements, Schroeder et al. (1963) constructed a
special animal house on a mountain top in Vermont, USA, far removed
from traffic and industry. The interior was specially coated with
organic resins to reduce any metallic exposure. Air was introduced
through special filters, and strict practices were enforced to
avoid the introduction of dust and dirt. This will be referred to
as the "controlled environment". Under these conditions, a more
severe chromium deficiency resulted in very low glucose removal
rates of 1.12%/min in 6 rats and of 0.18%/min in 4 female breeder
rats (482 days old), after 11 days on a chromium-deficient diet.
The removal rate in 4 control rats receiving the same diet with a
chromium supplement improved from 0.51 to 1.39% during the same
period (Mertz et al., 1965b). This strong impairment of glucose
tolerance in rats kept in the "controlled environment" was
reflected in their fasting blood-glucose levels, compared with
those of chromium-supplemented controls: 1370 ± 68 versus 1170 ± 17
mg/litre in males and 1390 ± 68 versus 960 ± 65 mg/litre in
females. More than half of 185 deficient rats excreted more than
0.25% glucose in the urine, whereas glycosuria was found in only 9
of the chromium-supplemented controls (Schroeder, 1966). A
significant reduction in the intravenous glucose removal rate was
also observed in 8 rats in plastic cages, fed an EDTA-washed low-
protein diet (7% casein), compared with 8 controls receiving 50 µg
chromium as the chloride, by stomach tube, daily for 2 weeks (2.1
versus 3.6%/min; P < 0.01). However, chromium supplements did not
improve the near-normal removal rates in rats receiving a 20%
casein diet (Mickail et al., 1976). Significantly lower plasma-
glucose levels, due to supplementation with chromium, were reported
in rats fed the chromium-deficient Torula yeast diet (Whanger &
Weswig, 1975). In another series of studies, only a slight
reduction in blood-glucose (from 1240 to 1180 µg/litre) was found
in 10 rats receiving a diet supplemented with chromium (10 µg/kg)
Impaired glucose tolerance in squirrel monkeys, fed a
commercial laboratory chow, was shown to respond to chromium
supplementation. Of 9 monkeys with an impaired glucose removal
rate (1.38%/min), 8 responded after 22 weeks of supplementation of
their drinking-water (chromium acetate, 10 mg/litre) with a
normalization of their glucose tolerance (average removal rate, N =
9, 2.33 ± 0.3%/min) (Davidson & Blackwell, 1968). There was no
effect on food consumption, growth rate, or serum-insulin
concentrations. The chromium content of the commercial chow was
stated to be 3.3 mg/kg, a very high concentration. In view of the
uncertainties of methods of analysis for chromium (section 2.2), it
is not possible to interpret the results as indicative of a very
high chromium requirement of the squirrel monkey or of an unusually
poor bioavailability of the chromium in that particular ration. A
marginal chromium deficiency may have existed in mice fed a bread
and milk diet, as daily administration of 10 µg chromium for 16
days to 8 months produced a 10 - 30% decline in blood-glucose
levels (Vakhrusheva, 1960). However, this effect was not specific
for chromium, as it was also observed when manganese was
The glucose tolerance of guinea-pigs did not differ
significantly between groups fed diets containing chromium at
0.125, 0.625, or 50 mg/kg, even though the animals fed the 2 higher
levels exhibited a lower mortality rate during pregnancy (Preston
et al., 1976).
Although the administration of synthetic glucose tolerance
factor (a chromium-dinicotinic acid-glutathione complex) to 6 pigs
did not affect glucose tolerance tests, it resulted in a
significant increase in the hypoglycaemic effect of insulin
injected at 0.1 U/kg body weight (Steele et al., 1977a).
Turkey poults, fed a practical ration containing chromium
levels of 5 mg/kg, responded to chromium supplementation (20 mg/kg
diet) with a significant increase in liver glycogen and in glycogen
formation, following a fast, and with a significant increase in
glycogen synthetase (EC 22.214.171.124) activity in the liver (Rosebrough
& Steele, 1981).
It can be concluded that the impairment of glucose tolerance in
rats fed a low-chromium Torula yeast diet is due to chromium
deficiency. The effects of chromium in squirrel monkeys, pigs, and
turkeys, though statistically significant, are somewhat difficult
to interpret, because of the reported high chromium content of the
basal diet. No evidence for chromium deficiency has yet been
obtained through glucose tolerance tests on other animal species.
7.1.2. Effects of deficiency on lipid metabolism
Though trivalent chromium in high doses (2.5 mg/kg body weight)
has been shown to increase the synthesis of fatty acids and
cholesterol in the liver (Curran, 1954), lower, physiological doses
appear to decrease serum-cholesterol concentrations in rats.
Schroeder & Balassa (1965) found an average level of 927 mg
cholesterol/litre serum in 24- to 26-month-old male rats, kept in a
controlled environment and administered chromium in the drinking-
water at a concentration of 5 mg/litre, compared with an average of
1229 mg/litre in controls not receiving chromium ( P < 0.01). The
effects in female rats were ambigous, one study producing the
expected reduction in cholesterol due to chromium, another showing
an elevation in the supplemented female rats. Schroeder's
observations of a cholesterol-reducing effect of chromium in male
rats were confirmed by Staub et al. (1969) and by Whanger & Weswig
(1975) but were contradicted by results of a third study in which
there were not any significant effects of chromium on sucrose-
induced triglyceridaemia and cholesterolaemia (Bruckdorfer et al.,
1971). Perhaps more significant than the effect on circulating
cholesterol is the direct effect of chromium on the occurrence of
aortic plaques. Schroeder & Balassa (1965) observed 6 plaques in 54
male and female chromium-deficient rats, but only one plaque in 48
animals receiving 5 mg chromium/litre in drinking-water. These
results are in agreement with those from a subsequent report of the
protective effect of the natural chromium content of water (60 -
215 µg/litre) against atherosclerosis in cholesterol-fed rabbits
(Novakova et al., 1974). Abraham et al. (1980) extended these
observations by demonstrating that daily chromium injections (20 µg
K2CrO4) reversed the established atherosclerosis in the aorta of 11
cholesterol-fed rabbits, compared with 12 controls. The mean
plaque area was reduced from 95% to 63%, the total aortic
cholesterol from 729 mg to 458 mg, and the atheromatous lesions, as
measured by technetium incorporation from 285 000 cpm to 114 000
cpm, all differences being statistically significant. These
results are reinforced by observations in man discussed in section
7.1.3. Effects of deficiency on life span, growth, and
The mortality of male, but not of female mice, raised in a
"controlled environment" (section 7.1.1) was reduced by trivalent
chromium administered as acetate in the drinking-water at a
concentration of 5 mg/litre (Schroeder et al., 1964). The survival
rates at 12 months were 92.6% and 68.8% ( P < 0.0001) in
supplemented and deficient animals, respectively. Similary, male,
but not female, rats receiving a chromium concentration of 5
mg/litre in the drinking-water had longer life spans than deficient
controls. The mean age of the last surviving 10% of animals was
1249 days, compared with 1141 days in the deficient animals ( P <
0.01). Survival of male rats fed a low-chromium (< 100 µg/kg),
low-protein ration and subjected to a controlled acute haemorrhage
was significantly less than that of chromium-supplemented rats, in
2 studies (67 versus 92%; P < 0.05 and 27 versus 60%; P < 0.01,
respectively) (Mertz & Roginski, 1969).
In Schroeder's study, growth rates in treated mice and rats of
both sexes raised in a "controlled environment", were higher after
6 and 12 months, with highly significant differences in body weight
ranging from 9 to 17% ( P < 0.005) compared with the controls.
Again, the effects of chromium supplementation were greater in
males than in females (Schroeder et al., 1964, 1965).
Similar results were reported by Djahanschiri (1976), who
studied a total of 2750 rats of a special inbred strain (Hk51) fed
a basal diet (0.15 mg chromium/kg diet) and chromium supplements
ranging from 10 to 500 mg/kg diet. At 12 weeks, the average
weights of the chromium-supplemented animals, regardless of dose
level, were significantly higher (by 6% in the males and 3% in the
females) than those of the animals on the basal diet. The same
author reported a progressive diminution in both the milk
production of lactating rats and weight gain in 3 consecutive
generations fed the low-chromium diet, compared with rats receiving
chromium supplementation. Increased mortality was reported in
pregnant guinea-pigs fed a low-chromium diet (125 µg/kg diet)
compared with animals receiving a chromium supplement of either 625
µg/kg or 50 mg/kg (Preston et al., 1976).
When rats raised on a low-chromium Torula yeast diet (< 100
g/kg) mated with those on a normal diet, they were able to
impregnate the females at a 100% conception rate only up to the age
of 4 months. After this age, the conception rate declined to 25%,
25%, and 0%, at the age of 7, 8, and 9 months, respectively. This
decline was accompanied by a significant ( P < 0.01) decrease in
the sperm count in the chromium-deficient males to approximately
half of the count in supplemented controls at the age of 8 months
(Anderson & Polansky, 1981).
7.1.4. Other effects of deficiency
Male weanling rats, fed a 10% soya protein ration with a
chromium content of less than 100 µg/kg, developed a visible
opacity of the cornea in one or both eyes. In several studies, the
incidence of this effect ranged from 10 to 15% in deficient rats.
No opacities developed in control animals receiving 2 mg
chromium/kg diet (Roginski & Mertz, 1967).
Chromium deficiency has been shown to reduce the physical
performance of rats under stress. Ten male rats raised on a
chromium-deficient diet (150 µg/kg diet) swam for an average of 250
min, until exhaustion, in contrast with 10 rats receiving a
supplement of 10 mg chromium/kg diet, which were exhausted only
after 320 min (Djahanschiri, 1976).
7.1.5. Mechanism of action of chromium as an essential nutrient
126.96.36.199. Enzymes, nucleic acids, and thyroid
Chromium is present in nucleic acids in very high
concentrations, but the function of these is not clear at present
(Mertz, 1969). However, recent work suggests a biological function
of chromium in nucleic acid metabolism (Okada et al., 1984).
Ribonucleic acid synthesis in mouse liver was significantly
increased by as little as 1 µmol trivalent chromium, in the
presence of DNA or chromatin (Okada et al., 1981). These effects
were also present when the DNA or chromatin were first complexed
with chromium prior to incubation. However, prior complexation of
RNA polymerase with chromium depressed activity. These effects
were obtained in vitro with a concentration (52 µg/litre) that is
similar to physiological levels. Goncharov (1968) presented data
suggesting that chromium is involved in the function of the thyroid
gland. These findings have been supported by Lifschitz et al.
An oligopeptide with a relative molecular mass of 1480, which
was crystallized from liver tissue, had a specific affinity for
chromium (Wu, 1981).
188.8.131.52. Interaction of chromium with insulin
The interaction of chromium with insulin has been extensively
studied and can therefore be presented in some detail, but this
does not imply that this is the only, or the most important,
function of chromium.
The effects of chromium in vitro, and probably in vivo, depend
on the presence of endogenous or exogenous insulin, no effects
having been demonstrated in in vitro systems that did not either
depend on, or contain, insulin. Chromium deficiency causes an
impaired response to added insulin in rat epididymal fat tissue,
and, when glucose uptake or glucose oxidation or utilization for
lipid synthesis is measured, the dose-effect curve is flat.
Addition of suitable chromium compounds significantly increases the
slope of the curve (Fig. 3). This demonstrates the true
potentiation of the insulin action and indicates that chromium
alone does not act as an insulin-like substance (Mertz et al.,
1961; Mertz & Roginski, 1971; Mertz, 1981). Chromium was also
shown to stimulate the transport of D-galactose into epididymal fat
cells. This suggests cell transport, the first step of sugar
metabolism, as a major site of action for chromium (Mertz &
Roginski, 1963). Insulin-potentiating effects have also been
observed on the swelling of liver mitochondria (Campbell & Mertz,
1963) and on glucose utilization in isolated rat lens (Farkas &
Stimulation of the effects of insulin has been observed in a
glucose-independent, but insulin-responsive, system. Significantly
more alpha-amino isobutyric acid (a non-metabolizable amino acid
analogue) was incorporated into the heart and liver tissue of
chromium-supplemented male rats than in the tissues of chromium-
deficient controls, in response to the in vivo injection of the
labelled analogue and insulin (Roginski & Mertz, 1969).
These observations suggest a peripheral action of chromium to
facilitate the action of insulin; no evidence has been produced
indicating that chromium plays any role in the production, storage,
or release of insulin by the pancreas. Thus, the primary result of
chromium deficiency is a diminution in the effectiveness of
insulin. The resulting metabolic impairment may be compensated for
by increased insulin production in some cases, resulting in
elevated concentrations of the hormone, but not enough data exist
from experimental animal studies to assess the action of the
element on insulin metabolism. More information is available for
human subjects and this is discussed in section 8.1. The
interaction between chromium, insulin, and receptor sites of liver
mitochondrial membranes was studied using polarographic techniques.
The results formed the basis for the hypothesis that chromium may
facilitate bond formation between the intra-chain disulfide of
insulin and sulfur-containing groups of the receptors, by
participating in a ternary complex (Christian et al., 1963).
This hypothesis is consistent with results of studies on rats
fed, either a low-chromium Torula yeast diet or a brewer's yeast
diet known to be adequate in chromium. While the insulin-binding
capacity of hepatocytes was not significantly different, the
insulin affinity of the cells was significantly greater ( P < 0.01)
for the chromium-adequate rats than for the deficient Torula yeast
rats (Steele et al., 1977b).
7.1.6. Chromium nutritional requirements of animals
In the preceding sections, studies were evaluated in which the
effects of chromium supplementation were determined in animals that
were at least marginally chromium deficient. In other studies, the
effects of chromium were investigated in animal systems in which
the existence of chromium deficiency was either not ascertained or
not investigated. Before these studies are described and
interpreted for the determination of chromium nutritional
requirements of animals, it is helpful to consider them against the
background of Venchikov's (1974) model. This model is generally
applicable to trace element effects defining 3 zones of action, the
zone of biological action, that of pharmacodynamic action, and that
of toxicity (Fig. 4). The biological zone, in response to
supplements with low amounts of an element, represents the
correction of a deficiency and the resulting level of biological
activity is that of optimal function. Increasing the amount of
supplement further may lead to a certain depression, followed by a
zone of new, increased activity, in which the element no longer
acts as an essential nutrient, but as a drug. Still greater
supplements, beyond the homeostatic control capability of the
organism produce toxic effects and death. Because all the studies
described subsequently involved amounts of chromium supplements
that were higher than the levels normally needed to correct a
deficiency, it is possible, according to Venchikov's definition,
that the observed effects might be pharmacological.
Tuman & Doisy (1977) studied the effects of yeast concentrates
of high chromium (glucose tolerance factor) content and of
synthetic chromium complexes with GTF activity (Tuman et al., 1978)
in mice, raised on a presumably complete commercial stock diet.
Six animals were used for each test, either genetically diabetic
mice or their control litter mates of the C57Bl-KSI strain.
Injections of either 5 mg of the GTF-containing yeast extracts or
0.1 mg of the synthetic chromium complex, acutely reduced the
elevated plasma-glucose levels in the diabetic and the non-fasting
normal mice by 10 - 38% of the initial values ( P < 0.01) and the
plasma-triglycerides by 26 - 56% ( P < 0.01), compared with control
mice injected with saline. Injection of insulin into diabetic mice
produced only an 11 - 18% decrease in plasma-glucose levels,
whereas injection of the GTF-containing extract together with
insulin reduced plasma-glucose levels by 39 - 51% and plasma-
triglyceride levels by 76% (Table 11).
The results suggest either a much higher increase in the
chromium requirement of the genetically diabetic mice or their
inability to use chromium in the diet.
Table 11. Acute effects of GTF and exogenous insulin on non-
fasting plasma-glucose and plasma-triglyceride (TG) concentrations
in 19-week-old genetically diabetic micea
Treatment Plasma- deltaGlucose Plasma- deltaTG
saline 11 120 ± 3960 ± 160(5)c
GTF 9320 ± -180 (16%) 2790 ± 170(5)d -117 (30%)
insulin 9840 ± -128 (12%) 3020 ± 400(5) - 94 (24%)
insulin 7060 ± -406 (37%) 940 ± 220(5)e -302 (76%)
and GTF 840(5)e
a Modified from: Tuman & Doisy (1977).
b Glucose and triglyceride values represent mean ± SEM for 6 mice in
each treatment group. Dose of GTF was 5 mg (WL-10-AT) administered
intraperitoneally, 12 h prior to collection of blood. Lente insulin
(0.1 U per mouse) was administered subcutaneously, 12 h prior to
collection of blood. Data were treated by analysis of variance to
detect differences between the various treatment groups; independent
orthogonal comparisons were performed in the following groups ( P value
indicates level of significance for each comparison).
c Saline versus all other treatments, P < 0.005.
d GTF versus insulin, P < 0.05.
e GTF and insulin alone versus combined GTF and insulin, P < 0.005.
Thus, GTF and insulin > GTF = insulin > saline.
Steele & Rosebrough (1979) reported a significant stimulation
of the growth rate of one-week-old turkey poults (both sexes) by
supplementation of a practical ration with 20 mg chromium (as
chloride)/kg. The weight gains within the 2-week study were 235 g
and 267 g for the 60 controls and 60 supplemented turkeys,
respectively ( P < 0.001). As the practical ration contained
ground yellow corn and soybean meal, limestone, and dicalcium
phosphate, chromium deficiency would appear unlikely. The amount
of the chromium supplement (20 mg/kg diet) is quite high, and
further studies are needed to decide whether the observed effects
were of a pharmaco-dynamic nature or truly nutritional, i.e.,
correcting a deficiency. A similar interpretation should be
applied to a report of improved egg quality in the laying hen
(Jensen et al., 1978).
The quantitative aspects of the effects of chromium on animals
can be summarized as follows: normal rats fed semi-purified, semi-
synthetic rations, with Torula yeast or individual proteins and
sucrose or starch as the source of carbohydrates, develop mild
signs of deficiency at a dietary level of 100 - 150 µg chromium/kg.
To prevent deficiency, most authors used very high supplements of
several mg/kg diet and did not determine the biological
availability of the chromium complexes used for the
supplementation. Diets containing raw ingredients and supplying
chromium levels between 0.5 and 1 mg/kg do not induce signs of
deficiency and probably meet the requirement of the rat. A very
tentative estimate of the dietary chromium need of the rat and
probably the mouse would be approximately 0.5 mg/kg diet. This
estimate should be interpreted with caution, because of the lack of
knowledge concerning the biological availability of chromium and of
its interaction with dietary constituents.
7.2. Toxicity Studies
The toxicology of chromium compounds has been reviewed by the
US National Academy of Science (US NAS, 1974a), Langard & Norseth
(1979), the International Agency for Research on Cancer (IARC,
1980), Langard (1980a, 1982), and Burrows (1983).
In discussing toxicological problems, it is important to
differentiate between the various oxidation states of chromium and
its compounds. Trivalent chromium, when administered to animals in
food or water, does not appear to induce any harmful effects, even
when given in large doses (US NAS, 1974a) (section 7.2.1). Acute
and chronic toxic effects of chromium are mainly caused by
hexavalent compounds. Since it has been shown that both industrial
trivalent chromium compounds as well as reagent-grade trivalent
chromium compounds can be contaminated by hexavalent chromium
(Petrilli & DeFlora, 1978a; Levis & Majone, 1979), the evaluation
of experimental studies becomes difficult, especially when the
purity of the chemical compounds used is not known.
Discrimination between the biological effects, caused by
hexavalent chromium and trivalent chromium is difficult, because,
after penetration of membranes in tissues, hexavalent chromium is
immediately reduced to trivalent chromium (Gray & Sterling, 1950;
US NAS, 1974a), and it is not evident whether the observed
phenomena are caused by this reduction or even by the trapping of
trivalent chromium by ligands after uptake in the cells. Another
problem in evaluating the data is associated with the route of
administration. Hexavalent chromium, introduced by the oral route,
is partly reduced to trivalent chromium by acidic gastric juice
(Donaldson & Barreras, 1966; DeFlora & Boido, 1980); thus, the
effects or lack of effects observed may be caused mainly by
trivalent chromium and not by the hexavalent chromium, actually
7.2.1. Effects on experimental animals
Many local effects on human beings have been reported (section
8.3), but only a few studies have verified these effects in
experimental animals. A comprehensive survey of hexavalent
chromium-induced effects is given in Table 12 (US NAS, 1974a). For
most studies, details were not given of the length of exposure,
number of treated animals and controls, etc. Diagnoses were stated
without presenting all the original data. Thus, in this section,
some papers will be discussed that refer to the most prominent
local and systemic effects to support and clarify the effects shown
in human beings.
It is evident that the toxicity of hexavalent chromium in
animals varies with the route of entry into the body. Low
concentrations of hexavalent chromium may be tolerated, when
administered in the feed or drinking-water, the extent of
absorption being a factor of importance. For example, rats
tolerated hexavalent chromium in drinking-water at 25 mg/litre, for
1 year, and dogs did not show any effects from chromium
administered as potassium chromate at 0.45 - 11.2 mg/litre over a
4-year period (US NAS, 1974a). However, oral exposure of both male
and female rabbits to sodium dichromate (0.1% solution, 0.2 - 5
mg/kg body weight, for up to 545 days) resulted in significant
morphological changes in the gonads, including atrophy of the
epithelium and dystrophic alterations of the Sertoli and Leydig
cells in the testes, and sclerotic and atrophic changes in ovaries
Larger doses of hexavalent chromium are highly toxic and may
cause death, especially when injected iv, subcutaneously (sc), or
intragastrically. The LD50 of chromium compounds was determined
for several experimental animal species. The LD50 of potassium
dichromate (hexavalent chromium), administered orally (stomach
tube) to rats, was 177 mg/kg body weight in males and 149 mg/kg
body weight in females (Hertel, 1982). When injected iv in mice
(sex not given), the LD50 of chromium carbonyl was 30 mg/kg body
weight (IARC, 1980).
Performing a life-time inhalation study on the rat, Glaser et
- (1984) found an LC50 for Na2Cr2O7 of 28.1 mg/m3 (range, 16.7 -
47.3 mg/m3). Assuming a deposition rate in the lung of 30% of the
dose administered, the LC50 dose was 1 mg/kg body weight in male
and 1.2 mg/kg in female rats.
Table 12. Effects of hexavalent chromium in animalsa
Animal Route Compound(s) Average dose Duration Effect Reference
Rabbit, inhalation chromates 1 - 50 mg/m3 14 h/day for pathological Lukanin (1930)
cat 1 - 8 months changes
in the lungs
Rabbit inhalation dichromates 11 - 23 mg/m3 2 - 3 h/day none Lehmann (1914)
as dichromate for 5 days
Cat inhalation dichromates 11 - 23 mg/m3 2 - 3 h/day bronchitis, Lehmann (1914)
as dichromate for 5 days pneumonia
Mouse inhalation mixed dust 1.5 mg/m3 4 h/day, no tumours Baetjer et al.
containing as CrO3 5 days/week, (1959a);
chromates for 1 year Steffee &
Mouse inhalation mixed dust 16 - 27 mg/m3 1/2 h/day tumours in Baetjer et al.
containing as CrO3 intermit- some strains (1959a);
chromates tently Steffee &
Mouse inhalation mixed dust 7 mg/m3 37 h over increased Baetjer et al.
containing as CrO3 10 days tumour rate (1959a);
chromates Steffee &
Rat inhalation mixed dust 7 mg/m3 37 h over barely Baetjer et al.
containing as CrO3 10 days toleratedb (1959a);
chromates Steffee &
Rabbit, inhalation mixed dust 5 mg/m3 4 h/day, none marked Baetjer et al.
guinea- containing as CrO3 5 days/week, (1959a);
pig chromates for 1 year Steffee &
Table 12. (contd.)
Animal Route Compound(s) Average dose Duration Effect Reference
Rat, inhalation hexacarbonyl 1.6 mg/m3 4 months, anaemia; lipid Roschina (1976)
rabbit 4 h, 5 days and/or protein
a week dystrophia
Rat, inhalation hexacarbonyl 0.16 mg/m3 4 months, anaemia; no Roschina (1976)
rabbit 4 h, 5 days biochemical
a week or morphological
Rat inhalation hexacarbonyl 35 mg/m3 30 min 100% death Roschina (1976)
Rat inhalation dichromates 0.006 - 0.2 28 days/ increase in Glaser et al.
mg/m3 90 days lung-macrophages (1985)
7 days/week lymphocytes
Rat intratra- dichromates 5 per week up to 30 toleratedc Steinhoff
cheal in- 0.01 - 0.25 months et al. (1983)
Rat intratra- dichromates 1 per week up to 30 tolerated; Steinhoff
cheal in- 0.05 - 1.25 months 1.25 mg/kg et al. (1983)
stillation mg/kg harmful
Rat oral potassium 500 mg/litre daily maximal Gross & Heller
chromate in non-toxic (1946)
Rat, oral zinc chromate 10 g/kg daily maximal Gross & Heller
mouse in feed non-toxic (1946)
Table 12. (contd.)
Animal Route Compound(s) Average dose Duration Effect Reference
Rabbit oral sodium di- 0.2 - 5.0 mg/kg 545 days morphological Kucher (1966)
chromate changes in
Rat oral zinc chromate 1.2 g/kg daily maximal Gross & Heller
(young) in feed non-toxic (1946)
Rat oral potassium 1.2 g/kg daily maximal Gross & Heller
(young) chromate non-toxic (1946)
in feed concentration
Dog, oral monochromate 1.9 - 5.5 mg 29 - 685 none harmful Lehmann (1914)
cat, or dichromates chromium/kg days
rabbit body weight per
day 1 mg chrom-
to 2.83 mg
or 3.8 mg
Table 12. (contd.)
Animal Route Compound(s) Average dose Duration Effect Reference
Dog oral potassium 1 - 2 g as daily fatal in 3 Brard (1935)
dichromate chromium months anaemia
Dog stomach potassium 1 - 10 g as - rapidly fatald Brard (1935)
tube dichromate chromium
Monkey subcutan- potassium 0.02 - 0.7 g in - fatald Hunter & Roberts
eous dichromate 2% solution (1933)
Dog subcutan- potassium 210 mg as - rapidly fatal Brard (1935)
eous dichromate chromium
Guinea- subcutan- potassium 10 mg - lethald Ophüls (1911a)
pig eous dichromate Ophüls (1911b)
Rabbit subcutan- potassium 1.5 cc of 1% - 80% fatald Hasegawa (1938)
eous dichromate solution/kg body
Rabbit subcutan- potassium 20 mg - lethald Ohta (1940)
Rabbit subcutan- potassium 0.5 - 1 cc of - nephritisd Ohta (1940)
eous dichromate 0.5% solution/kg
Rabbit, subcutan- sodium 0.1 - 0.3 g as - rapid deathd Priestley
guinea- eous or chromate CrO3 fall in blood (1877)
pig intravenous pressure
Rabbit intra- potassium 0.7 cc of 2% - lethald Mazgon (1932)
venous solution/kg 8-10 days after
body weight injection
Dog intra- potassium 10 grains - instant death Gmelin (1826)
Table 12. (contd.)
Animal Route Compound(s) Average dose Duration Effect Reference
Dog intra- potassium 1 grain - none marked Gmelin (1826)
Dog intra- potassium 210 mg as - rapidly fatal Brard (1935)
venous dichromate chromium
Dog intra- potassium 3 mg/100 cc 2 doses marked renal Hepler & Simonds
venous dichromate blood per dose damage (1946); Simonds
& Hepler (1945)
a Modifed from: US NAS (1974a).
b Pathological changes in experimental and control rats, 101 weeks after exposure.
c The same weekly dose distributed over 5 days was clearly better tolerated than a single weekly
d Renal damage.
A local corrosive action of hexavalent chromium on the skin,
similar to that seen in man, was described by Samitz & Epstein
(1962), who induced chrome ulcers in guinea-pigs at 4 trauma sites,
with daily exposure to 0.34 MK2Cr2O7 solution for 3 days. Mosinger
& Fiorentini (1954) showed the same effects using potassium
Following parenteral administration, the most common systemic
effects of chromium were parenchymatous changes in the liver and
kidney (Mosinger & Fiorentini, 1954). Later studies showed
selective damage in the renal proximal convoluted tubules, without
evidence of glomerular damage, as demonstrated after one single sc
injection of potassium dichromate of 10 mg/kg body weight (Schubert
et al., 1970) or after one single intraperitoneal (ip) injection of
sodium chromate of 10 or 20 mg/kg body weight (Evan & Dail, 1974).
Effects have also been found in fish (Strik et al., 1975). After 32
days of continuous exposure to 0.1, 1, or 10 mg hexavalent chromium
(as potassium dichromate)/litre, the fish Rutilus rutilus developed
lysis of the intestinal epithelium with haemorrhages as well as
hypertrophy and hyperplasia of the gill epithelium.
Franchini et al. (1978) found an increase in urinary protein,
lysozyme, glucose, and beta-glucuronidase in rats after a single sc
injection of potassium dichromate at 15 mg/kg body weight. After
sc injection (3 mg/kg body weight), every other day for 2 - 8
weeks, the authors observed a correlation between the chromium
contents of the renal cortex and chromium clearance.
Five-week-old male Wistar rats of the strain TNO-W-74 were
continuously exposed in inhalation chambers to submicron aerosols
of sodium dichromate at concentrations ranging from 25 (low level)
to 200 µg chromium/m3 (high level) (Glaser et al., 1985). Exposure
for 28 days to 25 or 50 µg chromium/m3 resulted in "activated"
alveolar macrophages with stimulated phagocytic activity, and
significantly elevated antibody responses to injected sheep red
blood cells. After 90 days of low-level exposure, there was a more
pronounced effect on the activation of the alveolar macrophages,
with increased phagocytic activity. However, inhibited phagocytic
function of the alveolar macrophages was seen at the high
hexavalent chromium exposure level (200 µg/m3). In rats exposed to
this chromium aerosol concentration for 42 days, the lung clearance
of inert iron oxide was significantly reduced. The humoral immune
system was still stimulated at a low chromium aerosol concentration
of 100 µg/m3, but significantly depressed at 200 µg chromium/m3.
Exposure of rats, through inhalation, to chromium carbide or
chromium boride dust at very high levels (300 - 350 mg/m3 for each
substance) for 3 months (2 h/day) resulted in effects on the
vascular system of the lungs, e.g., endothelial hyperplasia.
Bronchitis and a decrease in the blood-haemoglobin concentration
were also observed (Roschina, 1964). The effects of chromium boride
were more pronounced than those of chromium carbide.
Steinhoff et al. (1983) performed an intracheal injection study
on rats (930 rats, 30 months, 0.05 - 1.25 mg Na2Cr2O7/kg body
weight per week; 1.25 mg CaCrO4/kg body weight per week). Half of
the rats were intratracheally injected once a week and the other
half received the same weekly dose distributed over 5 injections
per week. In rats receiving doses 5 times a week, there were some
significant changes in levels of total plasma-protein and
-cholesterol, in some haematological variables, in organ weights,
and in survival times in females. Only male rats receiving 1.25 mg
sodium dichromate/kg body weight, once a week, showed a sharp
reduction in body weight, female rats being less affected. Rats
receiving calcium chromate in the same dose showed reduced body
weight but to a lesser extent. The weight of lung and trachea was
increased by both substances in all doses.
Marked histopathological changes (congestion, fairly large
areas of focal necrosis, bile duct proliferation) described after
long-term exposure of rabbits to hexavalent chromium (ip injection
of 2 mg chromium/kg body weight per day for 6 weeks) (Tandon et
al., 1978), as well as the increase in hepatic metallothionein and
decrease in cytochrome P-450 levels after ip injection of 400 µmol
chromium/kg per day (type of chromium compound not mentioned)
(Eaton et al., 1980) need further confirmation. The finding of an
accumulation of hexavalent chromium in the reticuloendothelial
system including bone marrow (Baetjer et al., 1959b; Langard, 1977)
may be of importance for a disturbed blood picture.
Merkurieva et al. (1980a,b) studied the effects of potassium
dichromate in the drinking-water on the activities of different
enzymes in rats. Exposure included daily doses of 0.0005, 0.005,
0.05, or 0.5 mg/kg body weight for up to 6 months. After 20 days
of exposure at the highest dose level, enzyme activities increased
by 15 - 28% in liver microsomes (inosine-5-diphosphatase),
lysosomes (beta-D-galactosidase), and cytosol (lactate dehydrogenase
(EC 184.108.40.206)). For some of these enzymes, as well as for
acetylesterase (EC 220.127.116.11), increases in activity of up to 54%
were found in the gonads, kidneys, seminal fluid, and serum. At a
dose of 0.05 mg/kg, the only statistically significant finding was
a 54% increase in the activity of acetylesterase in the gonads.
After a 6-month exposure at this dose level, there was a 70%
increase in lactate dehydrogenase activity in the seminal fluid as
well as a 50% increase in free beta-galactosidase in the liver. There
were no changes in enzyme activities at the two lowest dose levels.
Painting of rat skin with an aqueous solution of potassium
dichromate (0.5%), daily, for 20 days, resulted in a local
inflammatory reaction, an increased level of hexose glyco-proteins
in the skin and serum, and an elevated concentration of serotonin
in the skin and liver (Merkurieva et al., 1982). Most of these
effects were also seen earlier in the exposure period, though they
were not as pronounced. Ten days after the start of exposure, a
nearly 3-fold increase in the serum-acetylcholine concentration
occurred together with decreased acetylcholinesterase activity.
Thus, the data of Merkurieva et al. show systemic effects following
both oral and dermal exposure to hexavalent chromium.
Cats fed chromic phosphate or oxydicarbonate at 50 - 1000 mg/day
for 80 days did not exhibit signs of illness or tissue damage.
Similarly, toxic reactions were not observed in rats administered
drinking-water containing 25 mg trivalent chromium/litre, for 1
year, or 5 mg trivalent chromium/litre throughout their lifetime
(US NAS, 1974a). The toxicity of trivalent chromium is so low that
even by parenteral administration, a chromic acetate level of 2.29
g/kg body weight or a chromic chloride level of 0.8 g/kg body
weight is required to kill mice. Even very large doses given
intragastrically were not fatal for dogs. Brard (1935) reported
that 10 or 15 g of chromium as chromic chloride proved fatal in one
dog (US NAS, 1974a). Some fatal doses of trivalent chromium
compounds reported in the literature are listed in Table 13.
Rats exposed through inhalation to chromic oxide (trivalent
chromium) at 42 mg/m3 or to chromic phosphate at 43 mg/m3 (5 h/day
for 5 days/week) for 4 months developed chronic irritation of the
bronchus and lung parenchyma, and dystrophic changes in the liver
and kidney (Blokin & Trop, 1977).
Inhalation exposure of rats (number not given) to dusts
containing 36 or 50% chromite for 4 months (2 h/day), at
concentrations of 375 - 400 mg/m3, resulted in thickening of the
walls of pulmonary vessels and bronchi (Roschina, 1959). The high
exposure levels in these studies make it difficult to evaluate the
Inhalation studies have also been performed with chromium
carbonyl, where chromium is in the 0 oxidation state (Roschina,
1976). Twelve rabbits and 48 rats were exposed for 4 months (4
h/day, 6 days/week) at a concentration of 1.6 or 0.16 mg/m3. At
both exposure levels, there was loss of body weight (25 and 12%,
respectively) as well as anaemia and leukocytosis. In the higher
exposure group, the animals showed an elevated gamma-globulin level
in serum and an increased transaminase activity. The contents of
cholesterol and SH-groups were reduced, and there was a decrease in
cholinesterase (EC 18.104.22.168) activity. Lipid and/or protein
dystrophy were noted in several organs, e.g., in the liver and
kidneys. No such effects were detected in the animals in the low-
Table 13. Fatal doses of trivalent chromium in animalsa
Animal Number of Routeb Compound Chromium Effect Reference
animals dose (g/kg)
Dog 2 sc chromic chloride 0.8 fatal Brard (1935)
Rabbit 1 sc chromic chloride 0.52 fatal Brard (1935)
Rat 38 iv chrome alum 0.01 - 0.018 LD50 Mertz et al.
Mouse -c iv chromic chloride 0.8 MLDd Windholz et al. (1960)
Mouse -c iv chromic acetate 2.29 MLD Windholz et al. (1960)
Mouse -c iv chromic chloride 0.4 MLD Schroeder (1970)
Mouse -c iv trivalent chromium ? 0.25 - 2.3 MLD Windholz et al. (1960)
Mouse -c iv chromic sulfate 0.247 MLD Windholz et al. (1976)
Mouse -c iv chromic sulfate 0.085 MLD Schroeder (1970)
Mouse -c iv chromium carbonyl 0.03 LD50 Schroeder (1970)
a Modified from: US NAS (1974a).
b sc = subcutaneous; iv = intravenous.
c No figures given.
d Minimum lethal dose.
Various types of chromium chemicals, methods of administration,
and species of animals have been studied (IARC, 1980a).
Ideally, carcinogenicity should be tested with the methods
recommended by IARC (1980b), but, in many of the early studies,
this was not done. Carcinomas of the lung have been reported in
animals as a result of the administration of chromium chemicals.
Hueper (1958) found 2 squamous cell carcinomas and one
carcinosarcoma in 25 rats following intra-pleural injection of
chromite ore roast. After intrapleural implantation of strontium
chromate lasting 27 months, Hueper (1961) found tumours (type
unspecified) in 17/28 rats. Laskin et al. (1970) and Levy & Venitt
(1975) produced a number of bronchogenic carcinomas by implanting
pellets of cholesterol mixed with various chromium compounds
encased in a wire mesh cage in the bronchi of rats. Calcium and
zinc potassium chromate produced a number of bronchogenic
carcinomas, but soluble chromates and trivalent chromium chemicals
failed to produce cancer.
Using the same technique, Levy & Martin (1983) tested 21
different chromium-containing materials (pigments, intermediates,
and residues from the bichromate-producing industry, relatively
pure crystalline compounds) in 2250 random-bred rats and found that
chromates, described as sparingly soluble, were carcinogenic in the
rat lung. These materials included strontium and calcium chromate
and, to a far lesser extent, certain forms of zinc chromate.
Barium and lead chromate evoked only a very weak carcinogenic
response compared with strontium and calcium chromate. In the
study of Laskin et al. (1970), it was shown that chromium trioxide
produced hepatocellular carcinomas in 2/100 rats (controls, 0/24).
After inhalation of 13 mg calcium chromate/m3 (5 h/day, 5
days/week, for lifetime), Nettesheim et al. (1971) found 14 lung
adenomas in 136 treated mice and 5 in 136 untreated controls, but
no carcinomas. Steffee & Baetjer (1965), performing inhalation
studies on rats, mice, guinea-pigs, and rabbits (inhalation of
mixed chromate dust, corresponding to 3 - 4 mg CrO3/m3, 4 - 5
h/day, 4 days/week, for lifetime, or 50 months, respectively) could
only find 3 alveologenic adenomas in 50 treated guinea-pigs.
Laskin (1972) and Laskin et al. (1970) found 1 squamous cell
carcinoma of the lung, 1 of the larynx and 1 "peritruncal tumor" in
rats (inhalation of calcium chromate, 2 mg/m3, 589 exposures of 5 h
over 891 days) and 1 squamous cell carcinoma and 1 papilloma of the
larynx in hamsters. The number of treated animals was not
specified in either paper. Steinhoff et al. (1983) performed
intratracheal instillations of chromates in rats for 30 months with
one treatment/week and the same weekly dose distributed over 5
treatments/week (Table 14). In 880 exposed rats, 28 adenomas of
alveolar-bronchiolar origin (benign) and 12 malignant tumours (3
adenocarcinomas and 9 squamous cell carcinomas) were found. All
lung tumours developed very late and were only detected at the end
of the lifetime study, often in lungs with callosities. The
tumours were tiny and none of them caused the animal to die.
Sodium dichromate was not carcinogenic after exposure on 5
days/week. With calcium chromate, the carcinogenic effect was more
pronounced after treatment once per week, than after treatment 5
times per week.
Table 14. Incidence of benign and malignant lung tumours among
880 rats intracheally injected with Na dichromate and Ca chromatea
INTOX Home Page
Hyperglycaemia is defined as a blood glucose concentration greater
than 115 mg/dL (6.3 mmol/L), although a level of 150 mg/dL (8.3
mmol/L) is more commonly recognized as abnormal.
Beta-1-adrenergic blocking drugs
Calcium channel blockers
Cocaine and amphetamines
Somatotrophin (Human Growth Hormone)
Other endocrine disorders
Stress with sympathetic system activation
Moderate hyperglycaemia causes no symptoms. At higher blood glucose
concentrations, glucosuria leads to osmotic diuresis and dehydration.
Very high concentrations (greater than 600 to 800 mg/dL [33 to 44
mmol/L]) can cause obtundation or coma as a result of serum
Patients with drug-induced hyperglycaemia usually have other
manifestations of the intoxication which help suggest the diagnosis.
For example, overdose of salbutamol (albuterol) or other
beta-adrenergic agents causes tachycardia, widened pulse pressure,
agitation, and hypokalaemia. Similar findings may be seen with
intoxication by caffeine or theophylline, both of which are also
associated with seizures at high levels. Calcium antagonists such as
verapamil cause hyperglycaemia accompanied by hypotension and cardiac
conduction defects. Iron poisoning causes vomiting and diarrhea, and
radiopaque iron tablets are often visible on abdominal radiographs.
Other causes of coma and dehydration including:
Hypernatraemia (eg, diabetes insipidus)
Hypovolaemia from vomiting, dehydration, etc.
Ingestion of alcohols
Rapid blood glucose measurement. This may be performed by the
hospital laboratory or at the bedside using fingerstick capillary
blood and a portable battery-operated analyzer or a test strip. The
presence of glucose on dipstick testing of the urine suggests an
elevated blood glucose concentration.
Renal function tests (urea, creatinine)
In general, drug-induced hyperglycaemia does not require treatment,
and efforts can be focused on other manifestations of the specific
overdose, such as treatment of shock or seizures. For patients with
evidence of dehydration, administer intravenous fluids (preferably
normal saline). For significantly elevated blood sugar concentrations,
consider intravenous insulin.
CLINICAL COURSE AND MONITORING
Serum glucose levels should be monitored only if they are very high
(greater than 19 to 22 mmol/L [350 to 400 mg/dL]). Decisions about
hospital admission and length of emergency monitoring will depend
largely on the specific overdose.
Not common. Permanent insulin-dependent diabetes mellitus may occur
after poisoning by Vacor, pentamidine, alloxan or streptozocin.
AUTHORS AND REVIEWERS
Author: Dr K R Olson, University of California, San Francisco.
Peer review: Cardiff 9/96: V. Afanasiev, M. Burger, T. Della Puppa,
- Fruchtengarten, K. Olsen, J. Szajewski.
INTOX Home Page
A serum sodium concentration greater than 145 mmol/L (mEq/L).
Secondary to insufficient water intake
CNS depression from toxic cause
Secondary to excessive water loss
Cholinergic syndrome with severe sweating
Drug-induced nephrogenic diabetes insipidus
Severe gastroenteritis from toxic cause
Secondary to excessive sodium ingestion
Environmental exposure with dehydration
Hyperglycaemia leading to osmotic diuresis
Iatrogenic (hypertonic fluid administration, inadequate free water)
Polyuric phase of renal failure (eg, after relief of prolonged urinary
Delirium and decreased level of consciousness may occur with severe
hypernatraemia. Associated illness or circumstances of exposure may
result in hypovolaemia and hypotension. Patients with heat exposure
may also be hyperpyrexic.
Serum potassium, chloride, and bicarbonate
Renal function tests (urea, creatinine)
Blood glucose (to exclude hyperglycaemia as a cause of free water
Serum calcium and magnesium
Urine osmolality (This is the most useful test to determine the cause
of hypernatraemia. Patients who are dehydrated but with normal renal
function usually have an elevated urine osmolality [greater than 400
mOsm/kg]. Patients with impaired ADH secretion or reduced
responsiveness of the kidney to ADH, will usually have a urine
osmolality of less than 250 mOsm/kg.)
Use caution, as overly rapid correction of serum sodium can lead to
cerebral oedema. The goal of treatment should be to correct the serum
sodium at a rate no faster than 1 mmol/L/hour or 25 mEq/L/day. In
asymptomatic patients, a rate of 0.5 mmol/L/hour is acceptable.
Obtain frequent measurements of the serum sodium and adjust treatment
Patients with hypovolaemia should be initially treated with
intravenous isotonic saline. Once volume is restored, this should be
changed to half-normal saline with dextrose.
Patients with normovolaemic hypernatraemia may be treated with oral
water administration or intravenous 5% dextrose solution. Patients
who are hypervolaemic may need a loop diuretic such as furosemide to
remove excess volume.
Patients with lithium-induced nephrogenic diabetic insipidus who do
not improve after discontinuation of lithium may be treated with
indomethacin, 50 mg every 8 hours, and hydrochlorothiazide 50 to
CLINICAL COURSE AND MONITORING
Follow volume status and electrolytes carefully, and avoid overly
rapid correction of the serum sodium. Patients should be monitored
for altered mental status and coma.
Overly rapid correction of hypernatraemia may cause cerebral oedema
and permanent brain damage.
Author: Dr Kent R. Olson, University of California,
San Francisco, USA (February 1999).
Reviewers: Birmingham 3/99: B Groszek, H Kupferschmidt,
N Langford, K Olson, J Pronczuk.