Heavy metal poisoning and its laboratory investigation · Heavy metal poisoning and its laboratory investigation Dianne R Baldwin and William J Marshall From the Department of Clinical

Heavy metal poisoning and its laboratory investigation

Dianne R Baldwin and William J MarshallFrom the Department of Clinical Biochemistry, King's College Hospital, London SE5 9RS, UK

Additional key phrases: cadmium; lead; mercury;thallium; bismuth; arsenic; tin; antimony

INTRODUCTION AND OVERVIEW

The term `heavy metal' is, in this context,imprecise. It should probably be reserved forthose elements with an atomic mass of 200 orgreater [e.g., mercury (200), thallium (204), lead(207), bismuth (209) and the thorium series]. Inpractice, the term has come to embrace anymetal, exposure to which is clinically undesirableand which constitutes a potential hazard. Ourintention in this review is to provide an overviewof some general concepts of metal toxicologyand to discuss in detail metals of particularimportance, namely, cadmium, lead, mercury,thallium, bismuth, arsenic, antimony and tin.Poisoning from individual metals is rare in the

UK, even when there is a known risk ofexposure. Table 1 shows that during 1991±92only 1´1% of male lead workers in the UK and5´5% of female workers exceeded the legal limitsfor blood lead concentration.1 Collectively,however, poisoning with metals forms animportant aspect of toxicology because of theirwidespread use and availability. Furthermore,hitherto unrecognized hazards and accidentscontinue to be described.The investigation of metal poisoning forms a

distinct specialist area, since most metals areusually measured using atomic absorptiontechniques. Analyses require considerable ex-pertise and meticulous attention to detail toensure valid results.2 Different analyticalperformance standards may be required ofassays used for environmental and occupational

monitoring, or for solely toxicological purposes.Because of the high capital cost of good qualityinstruments, the relatively small numbers of testsrequired and the variety of metals, it is morecost-effective if such testing is carried out inregional, national or other centres having thenecessary experience. Nevertheless, patients arefrequently cared for locally, and clinical bio-chemists play a crucial role in maintaining a highindex of suspicion and liaising with clinicalcolleagues to ensure the provision of correctsamples for analysis and timely advice.

EXPOSURE AND TOXICITY

Potentially, toxic exposure to heavy metalsarises in diverse circumstances, includingenvironmental, therapeutic and industrial, byaccident, and by harmful intent. There is aspectrum of exposure patterns from chronic, lowlevels, such as occur among children exposed tolead paint in older houses, to acute, high-doseexposure, either intentionally or by accident.Accidental exposure may be small-scale, as inthe case of a child who swallows a mercurybattery, or large-scale, as with the contamina-tion of a water supply with aluminium sulphate(alum) at Camelford.3

Appreciation of the toxicokinetics of indi-vidual metals is important for an understandingof the rationale of monitoring and treatment.Relevant concepts fall into two categories.Exposure is the amount of a particular physicalor chemical agent that reaches the target;4

biological monitoring (BM) of exposure con-cerns the level and form of the element in theenvironment and in the body. Effective BMrequires an accurate measure of the internaldose, which is the amount or concentration of agiven chemical at the site of effect, and may referto the overall body burden or to that in aparticular compartment. Internal dose is usuallyassessed by some convenient and robust bio-chemical measurement which shows good corre-lation with the body burden, since in vivomeasurement is seldom practicable. Table 2

Review Article Ann Clin Biochem 1999; 36: 267±300

267

Dianne Baldwin died on 24 November 1998. This article isdedicated to her memory.

This review was commissioned by the Clinical LaboratoryInvestigations Standing Committee of the Association ofClinical Biochemists but does not necessarily re¯ect theirviews.

Correspondence: Dr William Marshall.E-mail: [email protected]

268 Baldwin and Marshall

Ann Clin Biochem 1999: 36

TABLE1.

Leadworkersunder

medicalsurveillance,bysex,bloodleadlevelandindustry

sector,1991±19921

Men

Women

%withbloodlead

concentration(mg/dL)in

range

Totalunder

surveillance

Number

withblood

leadconcentration

Totalunder

surveillance

(mg/dL)in

range

540

40±59

60±69

470

540

440

Smelting,re®ning,alloying,casting

61

32´5

5´5

0´9

7759

115

1116

Leadbatteryindustry

56´9

33´9

6´6

2´5

4297

88

28

116

Badgeandjewellery

enamellingandother

vitreousenamelling

100

00

058

27

027

Glass

making

66´6

28´7

4´4

0´3

1059

168

5173

Manufacture

ofpigments

andcolours

90´7

8´1

1´0

0´2

517

89

291

Potteries,glazes,transfers

87´8

11´0

1´2

0´0

425

166

4170

Manufacture

ofinorganic

andorganic

leadcompounds

97´9

1´9

0´0

0´1

2288

60

6Ship

building,repairingandbreaking

91´8

6´8

1´0

0´3

293

10

1Dem

olitionindustry

72´1

17´9

4´7

5´3

513

Paintingbuildingsandvehicles

94´3

5´0

0´5

0´1

799

Workingwithmetallic

leadandleadalloys

78´2

18´5

2´5

0´8

2063

94

11

105

Other

processes

90´8

7´3

1´0

0´8

3730

158

1159

Scrap

57´6

25´4

9´0

7´9

177

Allsectors

73.1

22.0

3.8

1.1

23978

912

52

964

CrowncopyrightmaterialisreproducedwiththepermissionoftheController

ofHer

Majesty'sStationeryOf®ce.

shows some of the factors which may in¯uenceexposure.The second concept relates to the biological

effects of exposure (BEM). Assessment of thesepresents more problems, since individuals maydiffer markedly in their responses to similarexposure. Both toxic and essential substancesexhibit a continuum of effects over a range ofconcentration. Figure 1 uses lead and its adverseeffects as an example of a toxic agent,5 and zincas an example of an essential substance whichhas adverse effects only at extremes of concen-tration. The critical organ is the ®rst organ in

which adverse effects, reversible or irreversible,occur under the particular circumstances, e.g.,the kidney in chronic cadmium exposure; thecritical effect is the ®rst physiological orbiochemical change which can be detected,6

e.g., altered haem synthesis in lead poisoning.Both may vary according to the nature androute of exposure to the metals and are ofparticular relevance when selecting suitableprocedures for monitoring biological effects.

We have included aspects of occupationalmonitoring in this review since poisoning fromindustrial causes may need hospital treatment.In the UK, employees exposed to lead at workmust be monitored according to the Control ofLead at Work Regulations (Code of Practice1985, updated 1998).7,8 This is normally theresponsibility of Employment Medical Advisers(EMAs) or company Appointed Doctors (ADs),who are empowered to suspend an employeewithout loss of pay, institute remedial action andcheck on other employees who may be affected.General practitioners do not have all thesepowers, but may be consulted by employees,self-employed persons and those who have leftsuch work or have performed it outside the UK.

MECHANISMS OF TOXICITY

Metals have been recognized as powerful toxinsfor many years. This has been re¯ected in their

Investigation of heavy metal poisoning 269


TABLE 2. Factors affecting the exposure of individualsto toxic metals

Element Atomic radius, valency, mobility,hydration

Chemical form Organic, inorganic, oxidation stateCompound Sulphide, oxide, saltsPhysical form Dust (particle size), fume,

solubilityConcentration High, low, seasonalityDuration Acute, chronic, variableInteractions With other metals, with

compounds (bioavailability)Route of entry Oral, inhalation, skinGeography Wind direction, water run-off,

distance from source (especiallyindustrial), transport (by man)

Foodchains Plant uptake, agriculture, ®shing

FIGURE 1. Bertrand's principle: the continuum of biological effects of exposure to a toxic element (lead) comparedwith an essential element (zinc). PBG=porphobilinogen; ATPase=adenosine triphosphatase; ALA=aminolaevulinicacid.

wide clinical applications, for example asantibacterial agents (mercury, bismuth, lead),antihelminthics (thallium) and diuretics (organicmercurials). Man-made drugs are designed tohave speci®c actions with minimal other effects,but in contrast, metals, although in some caseshave speci®c actions, generally exert widespreadeffects throughout the organism.9 Some of theseare indicated in Fig. 2.Metals with small atomic radii often have

important cellular functions which depend onthe formation of their preferred co-ordinationcomplexes with oxygen or nitrogen ligands ofcomparable size; for example, to control thequaternary arrangement in structural or cataly-tic domains of proteins (e.g., zinc and copper insuperoxide dismutase), and in the transport ofamino acids with sodium and of electrons byiron and copper in cytochrome c oxidase. Lighttoxic metals (e.g., lithium and aluminium) maycompete with or replace the metals theyresemble. Heavier metals are less likely to beimportant physiologically. They have largeratomic radii, are more easily polarized and tendto form stable co-ordination complexes withequally large, polarizable, sulphydryl ligands.Toxic metals exert their effects by virtue of theseco-ordination properties and their distributionthrough the body compartments. In addition,they may share certain physical or chemicalproperties with physiological elements near themin the periodic table. In acute poisoning, largeexcesses of metal ions can cause disruption ofmembrane and mitochondrial function with thegeneration of free radicals. Consequently, gen-eralized clinical effects, including weakness andmalaise, feature in most cases. However, despitethe potential for metals to disrupt metabolicprocesses, the body has a considerable armouryof defence.

EXTERNAL DEFENCES

The epidermis usually forms an effective barrieragainst metals in the environment, althoughsome topical creams may reduce its ef®ciency. Afew metals are able to reach deeper layers. Forexample, chromium and nickel can causedermatitis, and the fumes of certain metals,e.g., chromium(VI), can lead to severe ulcerationof the nasal septum. Absorption of ingestedmetals from the gut is generally low; adultsabsorb no more than about 10% of ingested leadand rather less of aluminium. Children, espe-cially the very young, probably absorb consider-

ably more lead.10 The absorption of toxic metalscan often be reduced by ensuring adequatedietary intakes of essential elements, becauseeither there are dietary interactions or commonuptake mechanisms which are stimulated byde®ciency. With mercury as mercuric ion(Hg2+), intestinal absorption is about 10%, butas elemental mercury (Hg0) it is negligible. In thelungs however, more than 80% of elementalmercury inhaled as vapour is absorbed. Whenmetal-bearing dust particles are inhaled, theamount of metal absorbed is generally inverselyproportional to the particle size.

SEQUESTRATION

In the body, toxic metals equilibrate with tissueand metabolic proteins, ®rst in the vascularspace and then in other compartments. Lead,aluminium, strontium and other metals mayaccumulate in bone to form a pool having a slowturnover rate and a long half-life, so reducingthe immediate exposure of other tissues. This is aparticular problem when they are also radio-active, e.g., americium. The accumulation mayalso have implications in situations with in-creased bone turnover or loss of bone, such aspost-menopausally, in pregnancy, in renal fail-ure or during treatment with cis-platinum.11,12

Toxic metals which fail to cross the basolateralmembrane of the intestinal mucosal cells areeliminated when these are shed. For this reason,copper toxicity in Wilson's disease can betreated with high doses of zinc. The zinc inducessynthesis of metallothionein (see next section)which binds copper in the cells and prevents itsabsorption. Other metals may be sequestered inthe skin, e.g., silver in argyria.

Protein complex formationMetallothioneinMetallothionein (MT) is a low molecular weight(6±7000Da) intracellular protein with about30% sulphur-containing amino acids, which isfound in most phyla and has a highly conservedstructure. Some 30% of its amino acids aresulphur-containing and high concentrations arefound in liver, kidney, intestine and pancreas.MT is known to be important in the regulationof copper and zinc metabolism and in thedetoxi®cation of heavy metals, particularlycadmium. However, the fundamental role ofMT is still not understood, despite extensive useof its promoter gene and transcription controlmechanisms as tools in molecular biology. Its





FIG

URE2.

Somemolecularandcellularactionsofmetals.AD=

adenosinedi-;AT=

adenosinetri-.

clinical signi®cance has recently been thoroughlyreviewed by Coyle et al.13

MT has alpha (C-terminal) and beta (N-terminal) domains and can bind at least 18different elements. Up to 7 atoms of most metals(e.g., zinc, cadmium) can usually be bound (4alpha, 3 beta) but up to 12 atoms of a fewmetals, e.g., copper, can be bound. Synthesis ofMT is induced by many substances, includingzinc, cadmium, copper, mercury, gold, bismuthand some non-metallic compounds (althoughoften at non-physiological concentrations), butintracellular zinc must be available for this tooccur. Synthesis is regulated by interactions withglucocorticoids and a multiplicity of in¯amma-tory mediators. An increase in MT is one of theearliest cellular events in the acute phaseresponse and may be a means of increasing zincavailability for protein synthesis.In hepatocytes, MT is metabolized in lyso-

somal or non-lysosomal compartments, accord-ing to the metal bound; its half-life in liver alsodepends on the metal bound, being about 80 hfor cadmium-MT and 20 h for zinc-MT. PlasmaMT concentration rises in response to cadmiumor copper toxicity and in the acute phaseresponse. It falls in zinc de®ciency. Inductionof MT by pretreatment with zinc has beenshown to reduce the toxicity of many elements inanimals and of cadmium in humans.

Nuclear inclusion bodiesLead, bismuth and mercury all give rise tocharacteristic eosinophilic nuclear inclusionbodies in renal tubular cells. These are insolubleprotein complexes and form soon after thecommencement of exposure. They can some-times be detected in urine, but it is not clearwhether this is due to an active excretoryprocess, or passive shedding of tubular cells.14

Haemosiderin formation represents a similarmechanism for sequestering excess iron.

BIOTRANSFORMATION

Several toxic elements are metabolized in thebody. Mercury, arsenic and selenium can bemethylated and demethylated by several path-ways. Reductive methylation of arsenate[As(V)O4

37] via arsenite [As(III)O337] leads to

formation of less toxic monomethylarsonic anddimethylarsinic acids, which can be excreted.Excess selenium is excreted as trimethylseloniumion unless exposure is great, in which casevolatile dimethylselonium is formed. (This gives

the breath a characteristic garlic odour, abiological effect which individuals can monitorfor themselves.) Methylation of mercury, leadand tin produces metabolites which are moretoxic. Tetramethyl derivatives of lead and tin aredealkylated in the hepatic microsomal cyto-chrome P450-mediated monoxygenase path-way.15 Elemental mercury is oxidized tomercuric ions via the catalase-mediated hydro-gen peroxide pathway.

THERAPEUTIC APPROACHES TO METALPOISONING

Dialysis, haemo®ltration and plasma exchangehave all been used in attempts to reducecirculating levels of metal in acute poisoning,sometimes in conjunction with chelating agents.However, once the metal has become distributedthrough extravascular compartments, so much isbound in tissues that removal of the metal fromblood has a negligible effect. Thus, bene®t isonly achieved if these techniques are tried early,although they may become necessary to managerenal failure arising as a consequence of thepoisoning.

Although there are speci®c antidotes tocertain metals, e.g., Prussian blue for thallium,treatment is generally by the use of chelatingagents. These were ®rst developed in World WarTwo, when British Anti-Lewisite (BAL, dimer-captopropanol) was introduced for use againstmustard gas. Later, this compound was used forthe removal of copper in Wilson's disease, untilthe introduction of oral penicillamine. Not allchelating agents are suitable for therapeutic use,and Table 3 indicates the criteria required.

Dimercaptosuccinic acid (DMSA, Succimer)and dimercapto-1-sulphonate (DMPS, Unithiol)are examples of new agents which are effective atchelating lead and mercury and have theadvantage that they can be given orally. Theybelong to a family of compounds derived fromBAL, but are less unpleasant.16 In the USA,DMSA is now approved for chelation therapy inchildren, and in the UK both agents areavailable on a named-patient basis from theMedical Toxicology Unit, Guy's and St Tho-mas's Hospitals NHS Trust (London). Fewadverse effects have been noted so far, but theseagents have not been in use long and prospectivetrials are awaited.17

With most chelation therapy it is essential thatthe agent is not administered while externalsources of exposure (including any in the



intestines) remain, otherwise uptake may bedramatically enhanced.18 Iron is an exception tothis important caveat, as any present in theintestine becomes effectively unavailable becausethe chelating agent desferrioxamine, a bacterialsiderophore, binds so strongly to iron whengiven parenterally. Blood concentrations ofmetals frequently rise further during chelationtherapy, since most analytical techniquesmeasure the total amount, including the chelate.Once the chelate has been excreted, bloodconcentrations fall below pretreatment values,but may subsequently rebound as the vascularspace is replenished from other body compart-ments. Careful monitoring of this rebound isessential to ascertain whether further chelationmay be necessary.

CADMIUM

Cadmium has an atomic mass of 112. It is one ofthe commonest environmental metal poisons,largely because of human activities, particularlyzinc, lead and copper (which occur in mixed oreswith cadmium) smelting, although areas of highnatural cadmium content do exist. In addition,cadmium is used in welding and brazing, inalloys and in the manufacture of batteries, whileits compounds are widely employed as pigmentsand stabilizers in the paint, plastics and printindustries. Cadmium usage has increased duringthe last 50 years and has resulted in contamina-tion of more areas of land. Human health

problems due to cadmium have been identi®edin ®ve of thirteen such areas in Japan andincreased prevalence of tubular proteinuriaassociated with environmental cadmium pollu-tion has been reported in parts of China19 andBelgium20 but not in all such areas.21 Air andwater cadmium concentrations in unpollutedareas are low and normal dietary intake is lessthan 267 nmol (30 mg)/24 h, of which adultsabsorb 2±8%. The amount absorbed variesinversely with the plasma ferritin concentration.Inhaled cadmium is more ef®ciently absorbedand twenty cigarettes may contribute 18±36 nmol cadmium (2±4mg).22 The main concernswith cadmium are therefore chronic exposure inpopulations and prevention of accumulatedrenal burdens from exceeding a critical concen-tration. Much knowledge has been gained fromstudies among populations exposed to cadmiumeither occupationally or through living inpolluted environments. However, acute poison-ing from inhalation of fumes or ingestion ofcadmium salts can occur and at least one deathfrom self-poisoning with cadmium chloride hasbeen reported.23

ToxicokineticsAfter absorption, cadmium is transported in theblood bound to albumin. It is taken up by theliver, where it induces synthesis of MT, to whichit binds. Subsequently, cadmium-MT is releasedinto the circulation, ®ltered by the kidneys andreabsorbed by endocytosis into the cells of theproximal tubules, which are the critical organs.The complex is degraded in the lysosomes andcadmium atoms are freed, either to recombinewith MT synthesized locally or to initiate toxicprocesses. Cadmium thus accumulates in renaltubular cells, with a half life of 17±30 years,6

until their synthetic capacity for MT is exceeded.This occurs at a critical concentration in renalcortex of about 1600 nmol (180mg) cadmium/g.Cadmium then binds to other ligands, irrever-sible renal damage ensues, renal cadmiumdeclines and its urinary excretion rises sharply.There is characteristic, but mild, low-molecular-weight renal tubular proteinuria, progressing tohypercalciuria, glucosuria, phosphaturia andamino aciduria. The hypercalciuria is an early®nding and calcium excretion is correlated withurinary cadmium; whether this effect is primaryor secondary to the tubular damage remainsunresolved.24 Effects of cadmium on the synth-esis of procollagen by human lung ®broblasts invitro have also been reported, consistent with the



TABLE 3. Criteria for the suitability of chelatingagents for therapeutic use

. Available in a suitable form for administration,preferably orally

. Neither the agent nor its chelate with the elementmust be toxic

. Any side effects should be minimal, e.g.,ethylenediaminetetra-acetic acid (EDTA) mustbe given as the calcium salt because it chelatescalcium and other essential elements as well aslead

. The agent and chelate must not be metabolizedand release or redistribute the element, e.g.,citrate is a good chelating agent, but is readilymetabolized

. The thermodynamic association constant for theagent with element must be favourable comparedwith that for the element and body constituents

. The chelate must be excreted readily in urine orbile

idea that slowing of repair processes in lungsleads to connective tissue damage, a possiblerole for cadmium in the lung disease of smokersand occupationally exposed persons.25

There is some evidence that the tubularproteins currently used for monitoring occupa-tional cadmium exposure may re¯ect irreversibledamage to the tubules.26 In an effort to developand validate earlier markers for nephrotoxiceffects, a collaborative European project hasstudied changes in urinary and serum constitu-ents following industrial exposure to lead,mercury and cadmium.27 For cadmium, thethreshold concentration (de®ned as the level atwhich 10% of individuals will show the change)varied considerably for the different markerslisted in Table 4.28 The authors postulated thatthese thresholds re¯ect a progression from (a)tubular cell alterations with minor biochemicalchanges only, followed by (b) a reduction innegative charge at the glomerular basementmembrane leading to increased permeability,with raised excretion of high-molecular-weightproteins in some individuals, followed by (c)continuing tubular cytotoxicity progressing toactual tubular damage with increased excretionof low-molecular-weight markers, e.g., beta-2-microglobulin (beta-2-M) and retinol-binding

protein (RBP). The health signi®cance of theseobservations remains to be established andcon®rmed in other exposed workers, but maybe important in future monitoring programmes.

Similar changes in excretion of markerproteins in environmental exposure cannot beassumed, since the route of exposure is moreexclusively oral. However, increased excretion ofcalcium beta-N-acetylglucosaminidase, RBPand beta-2-M were found at urinary cadmiumthresholds of 2 nmol/mmol creatinine in onelarge study.29

Clinical manifestationsWelders or cutters who accidentally inhalecadmium fumes develop a `fume fever', apneumonitis with cough, dyspnoea and alsomyalgia, which may progress after 24±48 h topotentially fatal pulmonary oedema. Loweracute exposure may produce reversible effects,but sometimes leads to massive progressive lung®brosis. Chronic inhalation leads to obstructiveairways disease and emphysema.30

In Japan, cases of Itai-Itai (meaning Òuch!-Ouch!') disease were identi®ed between 1940 and1975 among populations living in cadmium-polluted areas. In this condition there isproximal tubular damage, a mild anaemia anda severe loss of bone mineral which results inpainful fractures. Multiparous and postmeno-pausal women are subject to particularly severeeffects. However, in other countries with areasof comparable pollution, populations do notexhibit such severe symptoms.31 This may relateto other dietary in¯uences. Disturbances ofcalcium metabolism have been noted amongcadmium workers, variously leading to osteo-malacia, osteoporosis and nephrocalcinosis, andthis is more likely to occur in highly exposedsubjects, and in those who have low dietarycalcium and vitamin D intakes.

In some studies, cadmium exposure has beenlinked with increased mortality from cancers ofthe prostate and lungs.32,33 Although longer-term follow-up of the original groups has castdoubt over the link with prostatic cancer, thelink with lung cancer persists in at least twostudies.34,35 However, this could be due to thedif®culty of excluding effects from associatedelements known to be carcinogenic, e.g., arsenicor nickel.36 The unpredictable nature of any suchrisk would imply that dose±response monitoringwas irrelevant and that only minimum exposurewould be acceptable. In some early studies anassociation between hypertension and cadmium



TABLE 4. Values of urinary cadmium associated with asigni®cantly higher probability of abnormal values forrenal markers in adult male cadmium workers

Marker measured

Urinary cadmiumconcentration (nmol/mmol creatinine) atwhich 10% of sub-jects show elevatedmarker concentration

6-Keto-prostaglandin F1-alpha 2´4Sialic acid 2´4Transferrin 3´6Brush border antigen 3´7N-acetyl-beta-D-glucos-aminidase 4Intestinal alkaline phosphatase 4´1Albumin 4´1Tamm±Horsfall glycoprotein 7Tissue non-speci®c alkalinephosphatase 9´7Brush border antigen HF5 10Retinol-binding protein 10´4Glycosaminoglycans 11´5Beta-2-microglobulin 11´5Beta-2-microglobulin (serum) 6´1

Adapted with permission.28

exposure was suggested, but this has not beencon®rmed in later studies.37,38

Laboratory investigationsAssessment of exposure and internal doseIn the blood 90% of cadmium is bound toerythrocytes and has a half life of 70±120 days.The blood concentration depends on bothcurrent exposure and body burden and isgenerally a better guide to the former than isurine cadmium, unless the body burden is high.Blood calcium increases during the ®rst 4months of exposure.39 Urinary cadmium ex-cretion is low in non-occupationally exposedpersons of both sexes and rises slowly until 55years of age and then declines slightly. Bothblood and urine concentrations are higher insmokers.40 Urinary cadmium excretion is pos-itively correlated with cumulative exposure andwith renal and hepatic concentrations, exceptwhen: (a) exposure is so great that all cadmiumbinding sites are used; (b) the critical concen-tration is exceeded; or (c) cadmium-inducedrenal tubular damage has occurred. Both wholeblood and urine concentrations are thereforenecessary for the assessment and monitoring ofchronic cadmium exposure. The low values andthe small increases that occur with time makecadmium assay particularly demanding. Never-theless, within-laboratory relative SDs (CVs)better than 10% at concentrations in theunexposed ranges and better than 4% nearoccupational limits can be achieved in experi-enced hands.The measurement of urinary MT offers little

advantage over measurement of urine cadmium.In hair samples, cadmium measurement issubject to errors from contamination, whilefaecal cadmium is only an approximate guideto intake. Studies of hepatic and renal cadmiumconcentrations in vivo using direct neutronactivation or X-ray ¯uorescence techniques havecon®rmed conclusions based on epidemiological,occupational and post-mortem work, but are tooexpensive and dif®cult for wide application.41,42

Assessing the effects of exposureUrinary beta-2-M was the ®rst biological effectmarker for cadmium to be widely used, but it isincreased in other conditions and degraded inbladder urine with a pH of below 6´0. UrinaryRBP and alpha-1-microglobulin (alpha-1-M)also increase in response to renal tubulardamage and are now often preferred becauseof their greater stability.43 Increases in beta-2-M

and RBP are considered by some to re¯ectirreversible renal effects, which may progressfollowing cessation of exposure,26 and newearlier markers, as discussed above, may replacethem. Meanwhile, other workers consider acombination of high- and low-molecular-weightmarkers to be more effective than a singlemarker. Schaller et al. recommend albumin ortransferrin as the high-molecular-weight marker,together with RBP or alpha-1-M,44 while Jung etal. and Mueller found alpha-1-M combined withthe lysosomal enzyme beta-NAG exceededreference limits in 30% of an environmentallyexposed group and 39% of workers at acadmium battery plant.45,46 Renal proteinuriafrom causes other than cadmium exposure, suchas pregnancy, is not accompanied by an increasein urinary cadmium.47

Current UK guidelines for occupationalmonitoring of cadmium exposure and its effectsare shown in Table 5.48

TreatmentThere are no speci®c methods which eliminatecadmium effectively and therapy is entirelysupportive. This is why emphasis is on preven-tion of accumulation over long periods. Newchelating agents are being developed which maybe capable of reducing cadmium burden.49

LEAD

Lead has been used in industrial and domesticapplications for hundreds of years. Its toxicityhas been recognized for almost as long:Dioscorides in 200 BC wrote that `lead makesthe mind give way' and Benjamin Franklindescribed the `dry gripes' (colic) and `dangles'(wrist drop).50,51 Lead presents one of thegreatest challenges of modern times to thoseresponsible for public health and industrialhygiene. Analytical epidemiological studies sug-gest that some degree of exposure is almostuniversal.52 Lead, its metabolism and its effectshave been more extensively reviewed than anyother metal.53±5

Since the 19th century Factory Acts, progres-sive legislation has resulted in stringent controlof industrial hygiene and reduced the dangers oflarge-scale use of lead to a few cases each year inthe UK (see Table 1). However, sporadic casesmay be increasing. The sub-contracting ofdemolition work may lead to poor continuityof monitoring, and workers from small compa-nies or those who are self-employed may risk



acute exposure on top of a pre-existing bodyburden. People working in Continental Europefor short periods have presented on their returnwith manifestations of lead poisoning (Supra-regional Assay and Advisory Service TraceElement Directors, personal communication).Episodes of poisoning from occasional causessuch as imperfectly glazed ceramics,56 the use ofmedicines (which may contain as much as 60%lead57) available from Asian `healers', andcosmetic preparations, may affect any agegroup, and cases may present as acute emergen-cies. The average daily intake has fallen withreplacement of lead for soldering tins and fordomestic water pipes. The use of lead in paint,though restricted in the UK, still presents seriousproblems when older buildings are redecorated,and the occasional child with pica may stillpresent as an emergency with acute leadpoisoning from ingested paint.During the last decade the risk of exposure

to low levels of inorganic lead affecting thebehaviour and intelligence of children hasbeen a major issue. In 1979 Needleman andco-workers published a study of dentine lead inchildren from the general population.58 Groupswith high and low dentine lead were comparedin neuropsychological tests and differences ofabout 4 IQ points were found. Some of these

differences persisted in follow-up studies after 5and 11 years.59 However, blood lead measure-ments in small subgroups of children from bothgroups during the ®rst study [mean concentra-tions of 1´71mmol/L (35´5 mg/dL) and 1´15 mmol/L (23´8 mg/dL) in children with high and lowdentine lead, respectively] suggest that they mayall have suffered some exposure to lead. Dentinelead measurement has been shown not to beparticularly reproducible.60 Nevertheless, othergroups in the USA, Australia and Europe havereported similar relationships between bloodlead concentrations and performance in intelli-gence and attainment tests.61±63 Not all of theresults achieved signi®cance and if they did, itoften disappeared when they were adjusted forsocioeconomic variables. In Edinburgh, 501children aged between 6 and 9 years werestudied; the geometric mean blood lead was0´50 mmol/L (10´4 mg/dL). Blood lead concentra-tions correlated with performance in a range ofability and attainment tests, although the effectswere small compared with those of other factors.The association remained after controlling for33 confounding variables and was strongest forthe younger children.64 Pocock et al.65 recentlyconducted a systematic review of 21 cross-sectional and prospective surveys which includedmore than 100 children and concluded that



TABLE 5. Cadmium reference values in blood and urine

(a) Unexposed subjects

Non-smokers Smokers

Blood* 527 nmol/L (53mg/L) 554 nmol/L (56 mg/L)Urine 525 nmol/24 h (53 mg/24 h), or

52 nmol/mmol creatinine

(b) Occupationally exposed subjects: guidelines

Blood concentrations*

Urine concentrations 590 nmol/L (510 mg/L) 490 nmol/L (410mg/L)

510 nmol/mmol creatinine (510 mg/g) Maintain good hygiene andmonitoring

Check hygiene control

410 nmol/mmol creatinine (410 mg/g) Investigate hygiene control,check renal tubular functionand serum creatinine

More rigorous hygiene control,monitor renal tubular functionand serum creatinine

*ethylenediaminetetra-acetic acid (EDTA) whole blood, 2mL. Tubes with red or yellow stoppers are likely to causesigni®cant contamination.

doubling of the body burden [blood lead from0´5 to 1´0mmol/L (10±20mg/dL), dentine leadfrom 24 to 48 nmol/g (5±10 mg/g)] is associatedwith a de®cit of 1±2 points on the IQ scale.However, they acknowledged that a non-causalexplanation for the association remained possi-ble. In the USA, the scienti®c methods employedby Needleman for the original studies werecriticized to the Of®ce of Scienti®c Integrity.The resulting public investigation was incon-

clusive.66 Nevertheless, it is dif®cult to ignore thetrends noted in most studies. The debateculminated in 1991, when the US Centers forDisease Control published proposals to reducethe acceptable limit for blood lead, requiringeach state in the Union to establish programmesof monitoring and abatement to ensure that nochild between 0 and 6 years of age will beexposed to lead suf®cient to raise the concentra-tion in blood above 10mg/dL (0´5 mmol/L).67

Since then several surveys conducted in differentareas of the USA have expressed concern aboutthe number of children who may now beclassi®ed as `lead poisoned'.68

Two European Community-wide surveys ofblood lead conducted between 1979 and 1981recommended that 50% of the populationshould have values below 1´0 mmol/L (20 mg/dL), not more than 10% should be above1´5mmol/L (30 mg/dL) and fewer than 2%should exceed 1´7 mmol/L (35 mg/dL). Duringthe second campaign 3500 blood samples fromadults and children at risk of undue exposurewere surveyed; in only 3 out of 35 populationgroups were these criteria exceeded.69 Evidenceis emerging that concentrations of lead in bloodhave fallen further as use of unleaded petrol hasincreased (Delves HT, personal communica-tion). Figure 3 shows the decline in the 95thcentile for blood lead concentrations in environ-mental surveys in the UK since 1985, when theavailability of lead-free petrol was increased,with a consequent reduction in lead emissionsfrom 7500 to 3000 tonnes per annum.70 Theaction limit recommended by the Department ofHealth for the UK remained at 1´21mmol/L(25 mg/dL) and in France at 0´7mmol/L (15 mg/dL) until Ministers for the Environment met inMiami for the G8 summit in May 1997. At thismeeting, Ministers called for further actions thatwill result in reducing blood lead levels inchildren to below 0´5mmol/L (10 mg/dL) andcon®rmed the Commission's proposal to ban themarketing of leaded petrol in the EC fromJanuary 2000.71 Australia had plans for staged

reduction to 0´7 mmol/L (15 mg/dL) by 1998 andthen to 0´5mmol/L (10 mg/dL). Despite thedecline in blood concentrations noted above, ithas been estimated that an appreciable numberof children under 5 years of age in the UK mayhave blood lead concentrations above the newtarget limit.72 Analysis of the results of bloodlead measurements in children referred to fourUK Supraregional Assay and Advisory Service(SAAS) Trace Element laboratories between1992 and 1997 show values above this limit inabout 1 in 8 children (Fig. 4) (SAAS Directors,personal communication), with a slightly higherincidence of raised values in the younger agegroups.

ToxicokineticsIn the body, about 5% of lead forms anexchangeable pool distributed between the softtissues and blood. The remaining 95% issequestered in the bone, as insoluble phosphateswith a half life of 20±30 years. There is someevidence that with prolonged exposure, less leadenters bone and more binds to proteins, possiblyMTs, that have been demonstrated in theerythrocytes of persons exposed to lead. Suchbinding tends to increase the half-life of lead inthe blood.73 About 93% of lead entering theblood is bound rapidly to erythrocyte mem-branes and haemoglobin with a half-life of about



FIGURE 3. UK environmental lead monitoring: thedecline in the 95th centile of concentrations for surveysconducted since 1983. &=Adults and children (for 1984n=2321; for 1985 n=2470; for 1986 n=2376; for 1987n=2306; for 1995 n=420). *=Children only: 1=urban contaminated land, n=71; 2=raised water lead,n=158; 3=rural contaminated land, n=55; 4=mixedenvironments, n=280.

35 days.74 In relatively unexposed persons, bloodlead concentrations show a log-normal distribu-tion. Concentrations are slightly higher in adultmen, but below 12 years of age there is nodifference between the sexes. Less than 0´1 mmol(20 mg) lead per 24 h is normally excreted in theurine (House I, personal communication); al-most all that present in faeces consists ofunabsorbed lead.Classically, the erythroid bone marrow has

been the `critical organ' and inhibition oferythrocyte porphobilinogen synthase (amino-laevulinic acid dehydratase, ALAD: EC4.2.1.24) the `critical effect'. This enzymecatalyses conversion of aminolaevulinic acid toporphobilinogen, and inhibition results in in-creased urinary ALA. The insertion of ferrous(Fe II) iron into the porphyrin ring to form haemis catalysed by the mitochondrial enzymeferrochelatase (EC4.99.1.2) which shows re-duced activity in the presence of lead. Thisleaves the protoporphyrin free to form zincprotoporphyrin (ZPP), a highly ¯uorescent

complex, which is characteristically increasedin lead poisoning and also in iron de®ciency.Once formed, ZPP remains in the erythrocytefor the life of the cell and it should theoreticallybe an ideal marker for the effect of lead exposureduring the lifetime of the population of ery-throcytes. Other enzymes along the haemsynthetic path are also affected, particularlycoproporphyrinogen oxidase (EC1.3.3.3), whichleads to increased urinary coproporphyrinexcretion. All these metabolites have been widelyemployed as markers in screening and monitor-ing programmes for lead exposure.

Some recent work has challenged the assump-tion that lead inhibits ferrochelatase andcoproporphyrinogen oxidase directly. Studiesin lymphocytes have shown no loss of activityfor either enzyme in the presence of lead,75 andin vitro ferrochelatase can use zinc or cobalt asalternatives to iron. In inherited de®ciency offerrochelatase (protoporphyria) only freeprotoporphyrin, not ZPP, is formed. Therefore,lead may act by inhibiting activity of the



FIGURE 4. Blood lead concentrations in children (511 years) deemed clinically to be at risk of lead poisoning(1992±1997): results from clinical requests for lead analyses at four UK Supraregional Assay and Advisory Service(SAAS) Trace Element Units (Regional Toxicology Laboratory, Birmingham; King's College Hospital, London;Medical Toxicology Unit, Guy's Hospital, London; and Leeds General In®rmary).

ferrireductase which reduces iron to the ferrousion, thus limiting its supply to ferrochelatase sothat ZPP formation is preferred.76 The overalldecrease in haem production feeds back tostimulate the whole pathway with furtherincreases in production of ZPP and intermedi-ates and their urinary metabolites, principallyaminolaevulinic acid and coproporphyrin. Amicrocytic hypochromic anaemia develops dueto the effects of lead on globin synthesis,erythrocyte survival and haem oxygenase. Theblood ®lm appears similar to that seen in ironde®ciency and reticulocytosis may also bepresent. Inhibition by lead of another erythro-cyte enzyme, pyrimidine-5'-nucleotidase(EC3.2.2.10), leaves clumps of RNA in theerythrocytes. These appear as basophilic stip-pling on blood ®lms, which also occurs ininherited de®ciency of the enzyme. This stipplingis often the ®rst sign to suggest the diagnosis. Ina patient who presented with severe abdominalcolic, it was this observation which indicated thereal problem and halted consideration of alaparotomy.77

Figure 1 demonstrates the overall sequence ofeffects of lead in relation to the typical bloodlead concentrations at which they occur. Thelatter are somewhat lower for children than foradults, but individual variation is high.78 Forinorganic lead, there is now suf®cient evidencethat the central nervous system should beregarded as the `critical organ' and the `criticaleffects' are those on intelligence and cognitivefunction, which cannot be so convenientlyidenti®ed in individuals as can the productsfrom the haem pathway. The mechanisms ofneurotoxicity remain obscure, but may include:(a) alteration in the permeability of endothelia,leading to cerebral oedema; (b) impaired sialy-lation of adhesion molecules affecting synapseformation; and (c) indirect effects due to thehaem pathway metabolite aminolaevulinic acidon gamma-aminobutyric acid function.55

As exposure continues, the next organ to beaffected is the kidney. Generalized tubulardysfunction results in amino aciduria (particu-larly glycinuria), phosphaturia and glycosuria.Lead-induced changes in haemopoietic and renaltubular function are not usually permanentunless the insult is continual or repeated. Inthe long term this may result in irreversiblechronic interstitial nephropathy with a raisedurate ± `saturnine nephropathy'. A recent studyof leadworkers examined changes in a range ofurinary constituents.79 The authors reported

reduced excretion of eicosanoids, enhancedthromboxane excretion, and enhanced excretionof sialic acid. Excretion of tubular antigens wasrelated to the duration of exposure. Thesigni®cance of such ®ndings is not yet clear,but may become important in future biologicalmonitoring of the effects of exposure to lead.

In contrast to inorganic lead, organic lead islipophilic and primarily affects the centralnervous system. It does not bind to erythrocytes,so the blood lead concentration is not increasedexcept by the small amount of inorganic leadformed by dealkylation.

Clinical manifestationsThe classic presentation, with cerebral oedemaleading to encephalopathy with high mortalityor residual damage, used to be seen frequently inchildren from some urban environments, parti-cularly during summertime.80 This is now rare indeveloped countries. Peripheral and autonomicneuropathies continue to occur. The former isuncommon in children, but weakness, par-aesthesiae and wrist drop (painter's palsy) occurin severely poisoned adults,81 while reducedmotor nerve conduction velocities are found atlower blood lead concentrations. Autonomicneuropathy results in abdominal colic and pain,sometimes with diarrhoea and vomiting, some-times with constipation. Abdominal symptoms,together with marked general weakness, fatigueand malaise are the most common presentingsymptoms. Upper abdominal pain is said toindicate more acute exposure, while lower ab-dominal symptoms may suggest a more chronicexposure.78 Children frequently present withpeculiar eating habits (pica), often exacerbatedby iron de®ciency; the possibility of leadpoisoning should be investigated in all suchcases and probably considered in all childrenwith iron de®ciency or learning dif®culties.Radiography of the hands and feet may showlead lines at the epiphyses which suggest long-standing exposure. Rarely, in both adults andchildren, a blue line around the gums may beapparent.

Exposure to organic (usually tetramethyl ortetraethyl) lead is rare, thanks to stringentregulation, but can occur following petrolsnif®ng, although symptoms of hydrocarbontoxicity usually predominate. Poisoning resultsin a predominantly neurological picture, withsymptoms such as vomiting, weakness, fatigue,anorexia, confusion and convulsions.



Laboratory investigationAssessment of exposure and internal dosesThe `gold standard' for estimating the internaldose of lead has been measurement of itsexcretion after the administration of 1g of cal-cium ethylenediaminetetra-acetic acid (EDTA)(as an i.v. infusion or i.m. injection). Excretionduring the next 24 h of more than 3´9 mmol(800mg) lead in adults, or 2´9±3´4 mmol (600±700mg) in children, indicates a raised bodyburden. Unfortunately, a number of protocolshave been employed, for which the cut-off pointsvary. In the USA this test has been recom-mended for deciding whether or not to treatchildren with borderline blood lead concentra-tions. However, some experiments in animalshave suggested that even a single dose ofchelating agents may result in redistribution ofsome lead from bone to soft tissues, includingthe brain.82 This has prompted a considerationof alternative tests as well as a plea for moreresources for abatement programmes to reducelead exposure.83 In the UK this test is not widelyused, but may occasionally be of assistance, e.g.,when there is suspicion of multiple episodes, orof past exposure, as in children with pica.84

Blood lead and analytical performancestandards The concentration of lead in bloodshows good correlation with the EDTA excre-tion during periods of exposure to inorganic leadand it is now regarded as the measurement ofchoice for investigating environmental, acute oroccupational exposure in the UK. Followingcessation of exposure, the relationship is lessexact, but if the body burden is high, blood leadmay remain elevated for years, re¯ecting ìnter-nal exposure' from the bone compartment.In the past, the analysis of blood lead was

time-consuming and dif®cult. Blood sampleswere often collected using ®ngerprick techniquesor venesection in dusty workplaces and theincidence of contamination was notorious. To-day, industrial hygiene is better and venesectionsimpler, so contamination is less frequent.Nevertheless, elevated concentrations shouldalways be con®rmed in a further sample(urgently if necessary). Contamination fromother sources can also occur. In 1991, largebatches of rubber cap liners used by severalmanufacturers of blood sample tubes werefound to be contaminated.85,86

Since the 1970s a dramatic improvement inthe performance of blood lead analysis hasbeen brought about by three major develop-

ments: (a) the advent of graphite furnaceatomic absorption techniques; (b) the establish-ment of various national and internationalexternal quality assessment programmes;87,88

and (c) characterization of stable referencematerials suitable for use in internal controlprocedures by the Supraregional Assay andAdvisory Service Trace Element laboratories.2

Reductions in acceptable concentrations forlead, requirements for longitudinal stability inoccupational and environmental monitoring andthe interlaboratory accuracy necessary for largesurveys, contribute to the need for performanceto be speci®ed and for stringent controlprotocols.89 For example, during the secondEuropean Community environmental survey ofblood lead concentrations, 10% of all bloodsamples were exchanged between pairs oflaboratories and a further 10% were re-analysedin a central laboratory. The differences betweenthe paired estimates had to match certaincriteria, i.e., 80% of differences less than0´2 mmol/L (4 mg/dL). The UK laboratories,analysing 716 blood samples containing 0´2±3´0 mmol/L (4±60mg/dL) lead achieved over 95%of differences less than 0´2mmol/L (4 mg/dL).69

Minimum standards in external quality assess-ment have been enforced for some time amongUK laboratories undertaking occupational mon-itoring. A new portable instrument for analysinglead is currently being evaluated in the USA.

Urinary lead Measurement of lead in urine isappropriate when exposure to organic, e.g.,tetraethyl, lead is suspected. For persons occu-pationally exposed to organic lead, the frequencyof monitoring is governed by the expectedconcentration (Table 6) and the EmploymentMedical Adviser.8 Urinary lead concentrationcan also be used for monitoring chelationtherapy.

Faecal lead Measurement of faecal lead isoccasionally useful to con®rm the suspectedroute of exposure, e.g., in the occasional case offactitious lead poisoning.90

Hair lead The analysis of hair for trace andtoxic elements was well reviewed by Taylor in1986.91 He concluded that variation in factorsaffecting concentrations (sex, age, colour, cos-metic, hair site, hair length, sample preparation)varied so much that valid use had beendemonstrated only for certain toxic metals inparticular situations. For lead, hair analysis does



not show good correlation with blood leadconcentrations, but has been applied to casesof gross poisoning and in environmental surveyswhen aerial lead contamination is relevant.

Isotope ratios Naturally occurring lead com-prises four stable isotopes: 204Pb, 206Pb, 207Pb,208Pb. Their proportions depend on the age ofthe lead and its geological origin. Isotopes canbe determined using inductively coupled plasmaatomic absorption with mass spectrometry. Bycomparing the abundance ratios of the leadisotopes in biological samples with those inpotentially toxic environmental or other materi-al, the source of lead exposure can be con-®rmed.92,93

Assessing the effects of exposureZinc protoporphyrin (ZPP) ZPP is not af-fected by lead contamination and is readilymeasured in haemato¯uorimeters, which areconvenient and portable. It has been used toscreen children for lead poisoning, particularlywhere prevalence was great and exposure high94

and is still important in such areas whereresources are limited. However, several studieshave shown that when blood lead concentrationsare below 2 mmol/L (40 mg/dL) many childrenhave a normal ZPP and would not bedetected.95,96 The reason is that ZPP starts torise exponentially only at blood lead concentra-tions above 1´5mmol/L (30 mg/dL) (for adults) or1´2 mmol/L (25 mg/dL) (for children), i.e., at



TABLE 6. Lead reference values in blood and urine

(a) Unexposed subjects

Action

Blood* concentration 50´48mmol/L (10mg/dL) None required40´48mmol/L (10mg/dL) In children eliminate exposure41´21mmol/L (25mg/dL) In children under 6 years, consider

treatment if lead fails to fall afterelimination of exposure

42´9 mmol/L (60mg/dL) In children swift actionIn adults consider treatment

Urine{ excretion 5100 nmol/24 h (20 mg/24 h) None required

(b) Occupationally exposed subjects (inorganic lead)8

Newly appointed employees have an initial medical assessment and annual assessments thereafter. Wheresigni®cant lead exposure is present or suspected, lead is monitored at 3 monthly (for inorganic lead) or 6 weekly (fororganic lead) intervals, together with other relevant biological tests as required by the company Appointed Doctoror Employment Medical Adviser

Action levelprompts investigation ofreasons for elevation;a review of hygiene control;direct medical surveillance,and more frequentmonitoring

Suspension levelprompts immediatecon®rmation with a secondsample and suspension ortransfer from work exposingemployee to lead (for certainexceptions, see Regulations)

Inorganic lead [Blood* lead mmol/L (mg/dL)] [Blood* lead mmol/L (mg/dL)]Women of reproductive capacity 1´2 (25) 1´4 (30)Young persons 518 years 1´9 (40) 2´4 (50)All others 2´4 (50) 2´9 (60)

Organic lead [Urine{ lead nmol/mmol(mg/g) creatinine]

[Urine{ lead nmol/mmol/L(mg/g) creatinine]

Women of reproductive capacity 9´7 (20) 12 (25)All others 45 (95) 53 (110)

*2mls ethylenediaminetetra-acetic acid (EDTA) blood is preferred. {20mL urine in sterile universal container.

blood lead concentrations now considered highin the UK. ZPP does have a role in theevaluation of the effects of lead, to ascertainwhether exposure is acute, ¯uctuating or chronicsince, in the absence of iron de®ciency, ZPPrepresents the average lead exposure during thelifespan of the erythrocytes.

Porphobilinogen synthase Inhibition of thisenzyme in the cytosol of erythrocytes is the ®rstmeasurable response to lead exposure (Fig. 1). Ithas been used in monitoring the effects ofoccupational exposure in some European coun-tries. Theoretically, it should be relevant at thenew limits for blood lead concentration in theUSA. Quantifying this inhibition by measuringreactivation of the enzyme with zinc or dithio-threitol is dif®cult;97 the enzyme is not stable andin children total activity is low.

Other tests Inspection of Fig. 1 makes it clearthat as the acceptable limits for exposure fromenvironmental or occupational sources havebeen reduced, most measurable biological effectsare now insuf®ciently sensitive to be of use.Urinary ALA and coproporphyrin both riseexponentially as blood lead concentration in-creases and, although much used in the past, arenow appropriate only at higher occupationalexposures. Abnormalities of renal tubular func-tion may be demonstrable, with increasedexcretion of amino acids (particularly glycine),phosphate and glucose. Plain radiographs of theabdomen are important in revealing the presenceof ingested or other lead, e.g., from paint, toysor gunshots. Careful neurological investigationis required, although the subtle de®cits at lowexposure cannot be detected in individuals.

Responding to a raised blood leadThe ®nding of a raised blood lead concentrationshould prompt a request for con®rmation, theurgency depending on the level. In asymptom-atic patients chelation therapy is generally notconsidered unless the blood lead is above3´0mmol/L (60 mg/dL) in adults. In children,heightened appreciation of the risks and gen-erally falling blood lead concentrations meansthat chelation therapy may now be considered atconcentrations above 1´2 mmol/L (25 mg/dL) ifthere is no response to removal from the source.Whenever exposure is indicated, vigorousattempts must be made to identify and eliminatethe source. For leadworkers, the frequency ofbiological monitoring depends on the blood lead

concentration and type of exposure according tothe Code of Practice for the Control of Lead atWork and it is the responsibility of the companyAppointed Doctor to ensure suitable monitoringand to take appropriate action if the suspensionlevel is breached. New lower limits have takeneffect from April 1998 (Table 6). In these, themonitoring of women of reproductive capacityand young people are considered separately forthe ®rst time and there are action limits andsuspension limits. These limits are de®ned in mg/dL and are the reason why some blood leadmeasurements in the UK are reported in theseunits as well as in mg/L or mmol/L. Particularcare must therefore be taken to interpret bloodlead results and ranges correctly (see Appendix 1for conversion factors) especially when a resultmay be communicated through several labora-tory systems.

TreatmentIf the blood lead concentration is not greatlyraised and the patient is well, elimination of thesource of lead and ensuring adequate dietarycalcium, iron and zinc, together with a low fatintake, may be all that is required. Should theblood lead concentration remain elevated, treat-ment with oral agents may be necessary; thenewer dimercaptosuccinic acid (DMSA, Succi-mer, McNeil Consumer Products Co, FortWashington, PA, USA) is more effective thanpenicillamine. For acutely ill patients, intrave-nous EDTA (Ledclair, Sinclair PharmaceuticalsLtd, Godalming, Surrey, UK) can be used (asthe calcium salt to prevent the severe conse-quences of excessive calcium loss). Much zinc aswell as lead can be lost during treatment.

Case history 1: lead poisoningPresentationA child aged 7 years presented to her GP in 1991with anaemia. Her haemoglobin was 41 g/L andshe was referred to a paediatrician. She was paleand small, but not unwell. She was fussy withfood, preferring only milk and was said to eatplaster off the walls. Seven other children, olderand younger, were well. Her blood lead was2´0 mmol/L (41 mg/dL); zinc protoporphyrin was197 mg/g haemoglobin (upper reference limit2 mg/g). One week later, the blood lead had risento 3´3mmol/L (68 mg/dL). An abdominal radio-graph showed opaque material in her stomach,presumed to be plaster. There was no evidenceof renal tubular involvement nor of lead lines inher bones.



ManagementThe rising blood lead concentration at week 2prompted admission to hospital, to remove herfrom the source. Further increase in the bloodlead could have incurred the risk of serioussymptoms including encephalopathy. It wasconsidered that the lead was actually causingher symptoms, so after ingested lead had beenallowed to clear from the intestine she wastreated with oral DMSA (had treatment beencommenced before the lead had cleared, itsabsorption would have been enhanced). Aftertreatment, her blood lead fell and subsequentweekly monitoring showed a rebound to2´3mmol/L (47 mg/dL) which was not treatedfurther [current practice (1998) would probablyhave been to continue treatment until her bloodlead was below 0´72mmol/L (15 mg/dL).] Urinaryexcretion of lead was 2´3mmol/L (480mg/L) atthe end of treatment and fell to 0´3mmol/L(60 mg/L) two days after cessation of therapy.She was given iron supplements and attemptswere made to ensure her diet contained plenty ofcalcium and zinc, with a low fat content.

Identi®cation of exposure Blood lead concen-trations in the siblings were normal. Had theybeen raised, some common cause such as paintdust at school or home, toys, cosmetics,medicines or household water would have beenlikely. In such a case, testing the parents wouldprovide further clues. Environmental investiga-tion indicated that the source of lead was theplaster in the house, and a recommendation wasmade that the family should be rehoused.Attempts to persuade the family to reduce theexposure in the interim by sending the child toher grandmother were unsuccessful.

THALLIUM

This element has an atomic mass of 204´59 andso is truly `heavy', binding readily to sulphydrylligands. Its salts are colourless, odourless andtasteless, as well as quite soluble in water, so thatthey easily pass undetected. They are exceed-ingly poisonous. Ingestion of more than 10±15mg/kg can be lethal. Thallium achievednotoriety at the hands of the St Albanspoisoner.98 In past times, thallium had a varietyof medical applications, including the treatmentof scalp ringworm. Such use has been banned inthe UK since 1940, but thallium is still used as arodenticide in some parts of the world. Thalliumisotopes are used in clinical imaging. Industrial

uses, including the manufacture of optical glass,semiconductors, low-temperature thermometersand switches, catalytic processes and green®reworks are closely regulated by licensing.Poisoning is generally the result of ingestion,but cases due to inhalation of dusts, fumes fromsmelting, skin absorption and even fromsnif®ng contaminated cocaine have beenreported.99±101

ToxicokineticsDuring the ®rst few hours after ingestionthallium is distributed throughout the vascularspace. Thereafter, up to about 48 h, it enters thecentral nervous system (CNS) and other tissues.Because thallium closely resembles potassium inionic size and charge, its distribution volume islarge. Elimination commences after about 24 h.Approximately two-thirds of ingested thallium issecreted into the intestine, from which signi®cantreabsorption occurs. The remaining one-third isexcreted in urine. The elimination half-life maybe up to 30 days, but can be reduced to as littleas two days depending on the therapeuticintervention. The pattern is at least biphasic,with a slow ®nal stage.102

The toxicity of thallium depends on its abilityto combine with sulphydryl groups of moleculesin mitochondrial membrane and neuronal ax-ons, and on its similarity to potassium, withwhich it can exchange and compete to causeconsiderable disruption of fundamental cellularmetabolism. For example, it has a ten-timesgreater af®nity for membrane Na+,K+-ATPasethan does potassium and alters the pK ad-versely.103 Thallium has also been reported toform insoluble intracellular complexes withribo¯avin, which may account for symptomsmimicking de®ciency of this vitamin (alopecia,dermatitis, neuropathy).104

Clinical manifestationsThese are summarized in Table 7. The mostcharacteristic feature is the hair loss whichusually occurs 2±3 weeks after the exposure.

Laboratory investigationInitially, plasma electrolytes, renal function testsand full blood count may be quite normal.Measurement of thallium in urine and/or bloodis necessary to con®rm the diagnosis (see Table8). Urinary concentrations are about ten-timesgreater than those in blood and since largervolumes are available, more can be extracted, sothat better analytical performance can beachieved, e.g., between-run relative SDs



55%.105 Measurement of thallium concentra-tions in blood and urine 2±3 times weekly areneeded during the early stages of therapy, withmonitoring of plasma potassium. Patients maybe extremely unwell and complications whicharise during the illness require careful biochem-ical monitoring.For occupational monitoring, random urine

samples are collected at the end of the workingweek to assess the internal dose, or after theweekend to assess the body burden. All samplesare collected before the shift, to avoid contam-ination. Monitoring thallium exposure is im-portant, because its long elimination time makesit, in effect, a cumulative poison.

TherapyIn the ®rst 48 h after ingestion charcoalhaemoperfusion or plasma exchange can reducethe amount of thallium available for the second,distribution, phase. Chelating agents are eithernot effective for elimination of thallium, or (e.g.,in the case of sodium dithiocarbamate) willchelate the element but redistribute it to othertissues, particularly the CNS. However, becauseof the resemblance to potassium, there is a

speci®c antidote. Oral potassium ferrihexacyano-ferrate (Prussian blue, Berlin blue) is notabsorbed from the gut. It sequesters thallousions secreted into the intestine, by exchange withpotassium ions from the molecular lattice, thuspreventing the enteral recirculation of thallium.Concurrent intravenous infusion of potassiummay help `¯ush' thallous ions from tissuecompartments, but given too early in treatmentcan accentuate intracellular thallium uptake.Treatment for long periods (months) may berequired before thallium concentrations in bloodand urine fall to acceptable values and complica-tions such as sepsis, cardiac arrhythmias orrespiratory failure can arise. Recovery from theneurological symptoms takes even longer anddepends on good physiotherapy; full recovery offunction may never be achieved.

Case history 2: thallium poisoningMoore et al. described two cases of deliberatethallium poisoning.104 In the ®rst patient, thesymptoms commenced within 10min of inges-tion and were mistaken for myocardial infarc-tion, gallstones or pancreatitis. In the second,serious symptoms resembling gastroenteritis



TABLE 8. Thallium reference values

Reference interval Toxic range

Blood* 55 nmol/L (51mg/L) 4500nmol/L (4100mg/L)Urine(preferred)

55 nmol/L (51mg/L), or51mmol/mol creatinine(51mg/g)

41000nmol/L (4200mg/L), or4100nmol/mol creatinine(4100mg/g)

*Heparinized sample, 10mL in a plastic tube.

TABLE 7. Clinical features and differential diagnosis of thallium poisoning

Symptoms 12±48 h Nausea, vomiting, diarrhoea, gastritis, duodenitis, pancreatic and parotiddamage

2±5 days Paraesthesiae, hyperaesthesia, headaches, respiratory depression, ptosis,nystagmus, optic neuropathy and atrophy, myalgia, myopathy, severe pain,loss of re¯exes, convulsions, coma, delirium, acute motor neuropathy,dementia and psychosis

2±3 weeks Characteristic change at the bases of hair shafts, followed by loss of hair43 weeks Cardiac arrhythmias may occur for up to 2 months (due to autonomic

neuropathy or direct toxicity, or, in the case of sinus tachycardia, probablysecondary to increased catecholamine production.) Skin rash as in ribo¯avinde®ciency

Differentialdiagnosis

early Gastroenteritis, myocardial infarction, alcoholism

later Guillain±BarreÂ , systemic lupus erythematosis, diabetic polyneuritis, acuteintermittent porphyria, toxicity due to alcohol, arsenic, gold, mercury,carbon monoxide, organophosphates

developed within 1 h. In both patients severedeterioration continued during the next 10 days,with loss of walking ability and development ofhair loss suggesting the correct diagnosis. Thiswas con®rmed by thallium measurements andthe patients were ¯own to a specialist MedicalToxicology Unit for treatment including psycho-therapy and physiotherapy.

MERCURY

The most widely occurring mercury ore iscinnabar, or red mercuric sulphide. Whenground, it yields vermilion, which has beenmuch used in embalming. Five thousand yearsago, it was used to preserve bones found inDolmen tombs at La Velilla in Spain and this isbelieved to be the earliest recorded use ofmercury as an antibacterial agent.105 Mercurycompounds have since been used to treat manymedical conditions. The death of Paganini mayhave been hastened by chronic mercury poison-ing from the treatment he received for supposedsyphilis and excessive use of laxatives containingcalomel (mercurous chloride).106 Until 1953, theuse of calomel in ointments for nappy rash andin teething powders led to the admission ofmany children to hospital for mercury poisoning(Pink disease).107 In 1978, when Yeh et al.described a fatal poisoning due to mercuro-chrome antiseptic used for a baby with a largeomphalocoele, they reviewed 13 similar cases inthe literature.108 In 1984, Laundy et al. describeda fatal case of mercury poisoning resulting fromirrigation of the peritoneal cavity with mercuricchloride after surgery for an ovarian tumour andreviewed ten similar cases from the literature.109

Mercury is one of the three major environ-mental metal poisons; as with lead, attention iscurrently focused on the risk of subtle damagecaused by exposure to concentrations belowthose associated with detectable clinical effects.In particular, this concern relates to the degreeand effects of exposure from dental amalgam.This matter has generated widespread concernand requests for mercury measurements byanxious patients.Of the three classes of organic mercury, all

have been used in agriculture and medicine.Alkyl compounds are by far the most toxic.Inorganic mercury compounds are classi®edaccording to their oxidation state (see Table 9).The different mercury species are linked throughthe environment in a complex and fascinatingcycle. Elemental mercury is the principal

atmospheric form. About half is generated fromfossil fuels, mines, metal re®neries, wasteincinerators and crematoria, and the rest arisesnaturally through degassing of the earth's crust.Atmospheric mercury is oxidized to solubledivalent mercury and returns slowly to the earthin rainwater, having become distributed world-wide. In surface waters and sediments, divalentmercury can be reduced again, or methylated bymicro-organisms to the highly toxic methylmer-cury. The half-life of this compound in ®sh islong and the bioaccumulation factor from waterto edible ®sh can be more than 107. Environ-mental methlymercury is therefore of concern inpopulations which depend on ®sh consump-tion,110,111 in which, although overt clinicalsymptoms have not been detected, concentra-tions of methylmercury in urine and bloodoverlap those found in people affected byaccidental environmental contamination. Onesuch disaster was the release into MinnamataBay of waste methylmercury from an acetalde-hyde plant during the 1950s.112 Other outbreaksof methylmercury poisoning occurred during thelate 1960s, the most serious being in Iraq, wherefungicide-treated wheat was used for food. Over6000 people were affected and at least 600 died.Currently, there is evidence that mercury isaccumulating in river sediments, farm animalsand people living alongside rivers which draingold mining sites in Brazil.113

Isolated poisonings are more likely to becaused by elemental or inorganic mercury. Themetal is physically attractive and there are manyrecorded examples of children playing with it tothe detriment of the whole family. Childrenoccasionally swallow small photographic orhearing aid batteries which may contain as muchas a gram of mercuric oxide. These often passthrough the body unbroken, but must bemonitored, because if the battery lodges in thestomach, acid disintegration of the casing canlead to serious poisoning. Mant et al. describe a2-year-old child who needed two laparotomies toremove the battery and treat subsequent adhe-sions.114 Other important but occasional causesof mercury poisoning include ethnic remediesdispensed by Indian Hakim115,116 and cosmeticssuch as skin-lightening soaps containing mercu-ric iodide, which are banned in the EuropeanCommunity but available in other parts of theworld.117,118 Poisoning has also been reportedfrom mercury vapour released when ¯uorescentlight bulbs are crushed119 and from the use ofmercury in magical±religious ceremonies.120



ToxicokineticsInorganic mercury121

At least 80% of elemental mercury (Hg0) inhaledas vapour is absorbed. Curiously, prior ingestionof alcohol can reduce this to 50%, because itinhibits oxidation of Hg0 to Hg(II) by catalaseand hydrogen peroxide in erythrocytes. Absorp-tion of the metal from the gastrointestinal tractis poor. About 10% of the Hg0 absorbed isdistributed to the brain, and about 10% to theblood: most accumulates in the kidney. Excre-tion occurs equally in urine and faeces, with asmall amount being exhaled. The half-lives inthe blood (3 days) and brain (20 days) areshorter than in the whole body (58 days).Elemental mercury crosses the placenta freely.Only 10±15% of ingested Hg(II) is absorbed.

The half-life (48 days) in the body is longer thanthat for Hg0. Hg(I) is less well absorbed. In thebody, both Hg0 and Hg(I) are oxidized toHg(II). In this oxidation state mercury is highlyreactive and can disrupt membranes, combinewith sulphydryl groups to inhibit enzymes anddamage DNA. This may prove to be thecommon mechanism for the toxic effects of allmercury species; apparent variations only re-¯ecting differences in absorption and distribu-tion. The kidney is the target organ for Hg(II),which becomes concentrated in the proximalrenal tubules as various solutes are reabsorbedfrom the glomerular ®ltrate. It causes severedamage to the brush border membranes, withmitochondrial and lysosomal damage, particu-larly on the S3 segment of the proximal tubules.Selenium in the kidney can combine withmercury to form crystalloid inclusion bodies,and MT is induced by mercury. These mechan-isms may be protective.122 After recovery fromthe acute effects of poisoning, mercury may bedetectable microscopically within tissues formany years.

Organic mercuryIn contrast, about 90% of most organic mercurycompounds is absorbed from the intestine, andphenylmercury is also absorbed through theskin. Fig. 5 shows the inter-relationship ofmercury species. Much is known about organic(particularly methyl) mercury from studiesconducted in the wake of environmental acci-dents. This has been well reviewed by Clark-son.123 Methylmercury takes about 30 h toequilibrate in the blood compartment and ratherlonger in the brain. Equilibration with tissuesoccurs more rapidly than excretion so that ratiosof methylmercury concentrations between tis-sues remain constant. In primates, probablyincluding humans, the ratio between brain (thetarget organ) and blood mercury concentrationsis 5:1, higher than that in other species. Thehair:blood ratio is about 250:1. Blood and hairmeasurements can therefore be used to estimateinternal dose to the critical organ. Both havebeen used to establish thresholds for the onset ofvarious symptoms.

Although most textbooks state that methyl-mercury enters the brain and cells because it islipid-soluble, only the chloride (CH3HgCl) islipophilic. Other derivatives form hydrophiliccompounds through thiol groups in proteins andamino acids. Recent studies suggest that methyl-mercury, when combined with cysteine, has astructure which resembles methionine and entersbrain and other cells via the neutral amino acidcarrier. Other studies in rats have shown thatintracellular methylmercury, complexed withglutathione, leaves hepatocytes via the gluta-thione carrier and is secreted into the bile.Reabsorption of the complex occurs in the gallbladder and intestine, but any organic mercuryreaching the lower intestine is demethylated bybacteria to Hg(II). The central nervous system isthe target organ for methylmercury toxicity and,



TABLE 9. Mercury species and industrial uses

Main forms Principal uses

Inorganic Elemental Hg0 Industrial, scienti®c and electrical instruments, gold-re®ning, dentistry, chlor-alkali industry

Mercurous Hg(I)(e.g., calomel, HgCl)

Laxative, vermifuge, teething powders

Mercuric Hg(II) Antiseptic(e.g., corrosive sublimate, HgCl2)

Organic Aryl, alkoxyalkyl (e.g., CH3Hg+) Paper pulp, fungicides (now banned), diuretics,preservatives

in adults, the visual cortex and granular layer ofthe cerebellum are particularly vulnerable.Methylmercury crosses the placenta freely andhas devastating effects on the foetal brain. Thecritical effect is inhibition of the polymerizationof microtubules, a process essential for the celldivision and neuronal migration crucial tonormal development of the foetal brain. Effectsin both adults and foetuses are dose-related, butthe foetus is 5±10 times more sensitive.

In recent years it has been suggested that anumber of illnesses with poorly de®ned symp-toms, such as multiple sclerosis and Alzheimer'sdisease, are related to mercury released fromamalgam ®llings.124 Studies measuring mercuryin oral air and eluates from various types ofamalgam in vitro have indeed shown thatmercury is released from these ®llings. Estimatesof the mean amounts released vary but are up to20 mg per day, according to the amalgam type



FIGURE 5. Mercury: some interactions and effects of the different chemical species within the body. GI=gastro-intestinal.

and number of surfaces (World Health Organ-ization recommended maximum intake in adultsis 45mg/day125). The amount released is in-creased after insertion or removal, by chewinggum and grinding the teeth.126,127 Blood andurine mercury concentrations are higher insubjects with amalgam ®llings and indicate thatafter ®sh, these are the main source of mercuryin the general population.128 One study of tissuemercury in autopsy samples from patients withmultiple sclerosis showed less mercury in thebrain than in control samples,129 while Fung etal. showed no differences in blood or brainmercury concentrations, or mercury:seleniumratios in patients with Alzheimer's disease incomparison with controls.130,131

However, it must be emphasized that: (a) theamounts of mercury released and absorbed aresmall; (b) blood and urine concentrationsgenerally found are considerably lower thanthe maximum limits currently accepted as safefor occupationally exposed people; (c) theselimits may not necessarily be relevant if newertoxic effects are under consideration; (d)analytical methods in general use cold vapouratomic absorption and may not be suf®cientlysensitive unless they include a prior concentra-tion step; and (d) good control groups aredif®cult to establish for such studies. Thepossible harmful effects of mercury fromamalgam have recently been comprehensivelyreviewed by Eley.132

The ®nal conclusion remains uncertain, but itseems likely that a few people with corroded®llings may develop blood or urine mercuryconcentrations comparable with occupationallimits and some who have certain oral condi-tions may bene®t from replacement of amalgam®llings.133 An even smaller group of people, inwhom the smallest trace of mercury, (e.g., in®llings, shampoos, contact lens ¯uids) may leadto contact dermatitis, gingivitis or oral lichenplanus, suffer from true hypersensitivity tomercury, which can be con®rmed by patchtesting.134,135

Mercury affects haem synthesis in renal tissueleading to porphyrinuria. One group in theUSA,136 in a study of dentists exposed tosigni®cant amounts of mercury in their sur-geries, has shown a correlation between urinarycoproporphyrin, behavioural de®cits and mer-cury excretion after a test dose of DMPS. Invitro studies with various blood cells reveal ahost of toxic effects of mercury, includingincreased numbers of chromatid breaks, micro-

nuclei formation, reductions in B cell prolifera-tion and immunoglobulin synthesis.137±140 Otherstudies have failed to demonstrate differences incirculating antibodies, or complement compo-nents, at least at low exposure.141

Clinical manifestationsExposure to high concentrations of mercuryvapour affects primarily the lungs, causingpleuritic pain and shortness of breath, and alsofever, lethargy, nausea, vomiting and blurring ofthe vision, within hours of exposure. The lungsshow bronchiolitis with interstitial in®ltrates,and pulmonary function tests indicate restrictivechanges with reduced diffusion capacity. Levinet al. reported four young men who attempted topurify South American gold at home usingmercury.142 Their urinary excretions of mercuryranged from 0´85±2´50mmol/24 h (170±500 mg/24 h) on presentation. One suffered a cough andshortness of breath, two required hospitalizationfor treatment with penicillamine and one neededintensive care. Less severe exposure to mercuryvapour affects the brain and nervous system.Symptoms include excessive shyness, intoler-ance, irritability, gingivitis, paraesthesiae (alloccurring early), salivation, metallic taste,stomatitis, tremor, marked weight loss, memoryloss, myalgia, muscle spasms, arthralgias, nightsweats, insomnia and hallucinations. The termèrethism' was ®rst coined by Pearson in 1805 todescribe all these effects,143 but now is used torefer only to the early group. The non-speci®cityof these clinical features and their insidious,delayed appearance means that the diagnosismay be missed unless a careful history is taken.They may continue to increase in severity evenafter exposure has ceased.

Metallic mercury has been injected directlyinto tissues or veins, by accident from cathetertips or during arterial gas sampling, and some-times intentionally.144±146 Mercury occludessmall capillaries, and causes pleuritic pain,emboli and transient ischaemia as presentingfeatures. Sometimes sterile abscesses form whichexude droplets of mercury and, rarely, theresulting in¯ammation can provoke more wide-spread distribution of mercury and systemicpoisoning. Mercury persists in the body formany years, giving a speckled appearance toradiographs. Ingested metallic mercury (e.g.,from broken clinical thermometers) may passthrough the intestine uneventfully or becomeentrapped leading to complications such as®stulae.147



Mercuric chloride (corrosive sublimate) isexceedingly toxic in all metabolic systems.Ingested in large quantities it causes severenecrosis throughout the gastrointestinal tract,with cardiovascular shock and acute renalfailure. In smaller amounts it may cause patientsto present with unexplained loss of weight,diarrhoea, renal failure and neurological signssuch as paraesthesia and tremor. Àcrodynia' or`Pink disease' is a further syndrome of mercurytoxicity, found in certain individuals, which maybe caused by stimulation of the sympatheticnervous system. It was formerly associated withthe use of medications containing calomel, butmay occur with other types of exposure (seeCase history 3). The hands and feet become red,peeling and painful, there is loss of weight,irritability and marked personality change,hypertension, excessive paroxysmal sweatingwith photophobia in more severe cases. Someof these effects may be due to increasedcirculating catecholamines, and acrodynia hasbeen confused with phaeochromocytoma.148

Methylmercury exhibits an extremely longlatent period. The earliest symptoms are par-aesthesia, malaise and blurring of the vision.More severe exposure leads to constriction ofthe visual ®elds, deafness, dysarthria, ataxiaand, ®nally, mental derangement, coma anddeath. Exposure in utero results in microcephalyand even more severe brain damage. Many non-alkyl organic mercurials are virtually withouttoxic effects, except phenylmercury which isconverted to Hg(II) and shows similar toxicity.The potential overlap of syndromes caused bythe different mercury species is shown in Fig. 5and is well illustrated in the family describedbelow.

Case history 3: mercury poisoning(McNeil et al.)149

An 8-year-old boy found some metallic mercurywhich he took home and spilt. He remained wellthroughout, but 2 months later, his 14-year oldsister presented with painful, discoloured, hotfeet. She was agitated and uncooperative, andwas found to have raised blood pressure andproteinuria. The father also developed acro-dynia with variable proteinuria, but no aminoaciduria or glycosuria. The mother had had atransient rash initially and amino aciduria at thetime of diagnosis; about 3 weeks later shedeveloped oedema with 10 g proteinuria perday. All members of the family had a completerecovery, including the sister whose `dif®cult'

behaviour (characteristic of erethism) improveddramatically as her mercury level declined.

Laboratory investigationsThe diagnosis of mercury poisoning is based ondemonstrating concentrations higher than thelimits shown in Table 10 for either whole bloodor urine (early morning or 24 h collection).Urine is easier to measure although there maybe problems from skin contamination, mercurialpreservatives in disposable equipment and lossesdue to adsorption or diffusion if stored inpolythene containers. Also, because of effectsof mercury on mood and personality, urinecollections may be unreliable for monitoringtherapy and blood samples are preferable. Hairanalysis is seldom of value in individual cases,unless it is important to estimate the pattern ofexposure with time. It is analytically possible tomeasure total and inorganic mercury separatelyin blood or urine and this is occasionally helpful.Mercury excretion after a test dose of thechelating agent DMPS may re¯ect the bodyburden better than basal excretion, but has notfound wide application. Urinary mercury re¯ectsexposure during the previous 2±3 months; bloodmercury with its shorter half-life re¯ects ex-posure in the preceding week. Either may beused for occupational monitoring, depending onthe task. Measurement of urinary copro-porphyrin is not generally of diagnostic value.

In patients presenting with fever followingacute exposure to high levels of Hg0, measure-ment of blood gases will reveal hypoxaemia;following acute ingestion of Hg(II) salts theremay be signs of developing cardiovascularcollapse and acute renal failure. With lesseringestion, renal failure may be the presentingfeature. Otherwise, `routine' biochemical testsare likely to be normal although increased serumbilirubin and alanine aminotransferase havebeen reported in individuals with prolongedhigh exposure to mercury compounds. Aminoaciduria, glycosuria and phosphaturia may ormay not be present and there may be increasedexcretion of tubular proteins. Urine should betested for protein in case a nephrotic picturedevelops, in which case the serum albumin willalso decline.

TherapyIn the past dimercaprol and later oral penicilla-mine have been used as chelating agents to treatpoisoning from all mercury species. The neworal agent DMPS is proving more effective in



eliminating mercury and can reduce the half-lifeto about 11 days.150,151 The need for treatmentcan be dif®cult to assess; the value of earlytreatment, the risk of progression and late onsetof symptoms must be balanced against wideindividual susceptibility. Patients with injectedmercury metal are monitored using bloodmercury concentration and tests of renal func-tion. Recovery from acute poisoning may becomplete.

ARSENIC

After lead and mercury, arsenic is the thirdmajor environmental poison. Like them it occursnaturally in the environment; contamination ofgroundwater with inorganic arsenic (in Taiwan,northern Chile and most recently in WestBengal) gives rise to endemic arseniasis.152±154

Man's activities in smelting gold, copper andzinc, and making pesticides and fertilizers, havecaused pollution with arsenic as well as with leadand mercury, and like these metals arsenic hasfrequently been used in Western medicine;Fowler's solution of 1% arsenic trioxide wasformerly used for treating leukaemia, psoriasisand asthma, and as a tonic. The use of inorganicarsenical preparations is no longer recom-mended. Like lead and mercury, the use ofarsenic in Ayurvedic medicines dispensed byunquali®ed Hakims is continuing to give rise toreports of sporadic poisonings. These areworrying particularly because the number re-ported probably understates the problem.115

ToxicokineticsIngested inorganic salts are well absorbed,depending on their solubility and other dietary

factors. More soluble salts may also be absorbedfollowing inhalation and through the skin.Organic arsenic compounds are almost comple-tely absorbed. In the body, arsenic is distributedrapidly to lungs, liver, kidney and spleen, afterwhich it is redistributed to the skin, nails andhair, where it binds tightly to keratin. The half-life varies with the tissue, being longest in theskin and lungs. The kinetics of excretion,through the kidneys, suggests a three-compart-ment model.155

Arsenic exists as a number of species andoxidation states having marked differences intoxicity:

Arsine: AsH3

4Arsenite: As(II)O337

4Arsenate: As(V)O437, HAs(V)O4

27, H2As(V)O47

4Methylarsonic acid: CH3AsO(OH)2

4Dimethylarsinic acid: (CH3)2AsO(OH)

4Arsenobetaine: CH3As+CH2COO7

Arsine, the gaseous hydride, is rapidlyabsorbed through the lungs and acts bydisrupting erythrocyte sodium±potassiumpumps, causing haemolysis. Arsine is usuallyliberated accidentally; the best known examplebeing the leakage of a canister of the gas into thehold of a ship, which resulted in eight of the crewbeing poisoned.156 The most severely affectedcrew-member presented with a haematocrit of0%!

Arsine poisoning is rare; arsenite is the mostimportant toxic agent. The adverse effects arewidespread, particularly affecting the gastro-intestinal, cardiovascular and renal systems and,in the longer term, the peripheral and centralnervous systems. Arsenic metabolism varies



TABLE 10. Mercury reference values

Blood* Urine{

Unexposed persons 520 nmol/L54mg/L

550 nmol/24 h (510 mg/24 h),or 55 nmol/mmolcreatinine (510mg/g)

Dental workers 510 nmol/mmolcreatinine (520mg/g)

Industrial workers 595 nmol/L 545 nmol/mmolcreatinine (590mg/g)

Organic 515 nmol/mmolcreatinineInorganic 5120 nmol/mmol creatinine

*Lithium heparin or ethylenediaminetetra-acetic acid (EDTA) whole blood, 10mL in a hard plastic or glass tube.{24 h collection in a hard plastic bottle washed with nitric acid, or an early morning sample in a sterile universal.

between species; in humans, reduction andmethylation take place in the liver to generatemuch less toxic organic arsenic compounds:

Arsenate?Arsenite?

Monomethylarsonic acid?Dimethylarsinic acid

Arsenic excreted in urine consists of 10±15%inorganic arsenic, 10±15% monomethylarsonicacid and 60±80% dimethylarsinic acid. In ®shand shell®sh there is further metabolism totrimethylarsine and other organic species, in-cluding arsenobetaine. Arsenobetaine is a stableand non-toxic compound. It is rapidly absorbedfrom the human intestine and excreted un-changed in urine.Inorganic arsenic species are large ligands and

some of their toxicity resides in a tendency tocombine preferentially with the sulphydrylgroups of proteins. In addition, they havecertain speci®c reactions. The arsenite oxyanionhas a pyramidal structure, such that it can forma stable ring with the alpha-lipoic acid moiety ofpyruvate dehydrogenase, causing inhibition andraised pyruvate concentrations. Because thestereochemical structure of the arsenate oxy-anion resembles that of sulphate and phosphate,arsenate can enter mitochondria and competewith phosphate as a substrate to form highenergy compounds, but the arsenic analoguesare not stable (Fig. 2) and so uncouple oxidativephosphorylation.157 Arsenic can cross the pla-centa easily and is excreted in breast milk.

Clinical manifestationsTaken acutely, arsenic can cause vomiting andhaemorrhagic diarrhoea within minutes, orhours if the poison is taken with food;haematuria and acute renal failure may follow.If the poisoning takes place over a longerperiod abdominal pain, facial oedema, upperrespiratory tract dif®culty and/or obstructivejaundice with hepatomegaly can develop. Overabout 3 weeks anorexia and features of neuritisdevelop (weakness, salivation, trembling, loss oftendon re¯exes, impaired cutaneous sensation),sometimes after the arsenic ingestion hasstopped.158 Transverse white `Mees' lines appearin the nails. Milder poisoning causes similar butmilder symptoms. In a 68-year-old man withprolonged exposure to an arsenical insecticidethe main features were of persistent neuropathy,

macrocytosis without anaemia and hyper-keratosis.159

Occupational exposure to arsenic is associatedwith dermatitis and skin and nasal ulcers,caused by the corrosive action of inorganicsalts. There is an increased incidence ofperipheral neuropathy, cardiovascular disease,cirrhosis and renal tubular impairment. Thoseliving in areas with contaminated groundwaterdevelop hyperkeratosis, pigmentation and ulcersafter 2 or 3 year's exposure. The incidence ofcancers of skin, lung and gastrointestinal tract isincreased.160

Laboratory investigationArsenic can be measured in whole blood orurine, but the latter is to be preferred. Theidenti®cation of stable non-toxic arsenic species,such as arsenobetaine, derived from dietaryseafood is more dif®cult than measurement oftotal arsenic, but their presence can make thecon®rmation of arsenic poisoning from totalarsenic measurements dif®cult. A single meal ofseafood can raise the excretion of total urinaryarsenic more than 50 times. If arsenic ismeasured without prior digestion, inorganicarsenic and the two acids are measured, butarsenobetaine arsenic is not. Some increase inmonomethylarsonic and dimethylarsinic acidsmay arise from the metabolism of other organicspecies in seafood, e.g., arsenosugars fromseaweed.161,162 Seafood should be omitted for 5days before samples are collected. If this is notpossible, information concerning intake is help-ful. If the total urinary arsenic is raised anapproximate separation into non-toxic organicand toxic inorganic arsenic can be achieved bymeasuring the arsenic with and without priordigestion. For more rigorous speciation, com-plex methods based on HPLC to separate theindividual forms of arsenic are available in a fewcentres.

Table 11 shows reference ranges for blood andurine arsenic. Hair analysis is valuable forforensic purposes and in surveying exposedpopulations.163 For occupational monitoring,urine should be collected towards the end ofthe working week.

TherapyBecause of the af®nity of inorganic arsenic forsulphydryl groups, chelation with dimercaprolhas been used to treat arsenic poisoning.Penicillamine has also been used, but may beless effective at promoting excretion.161



However, DMPS has been used in several cases,e.g., in two patients who ingested grams ofarsenic trioxide, mistaken for a substance ofabuse: they made a full recovery.164 Thesymptoms of peripheral neuropathy are notalways reversible.

ANTIMONY

Antimony has come to prominence duringrecent years, as a consequence of an hypothesisthat a fungus (Scropularis brevicaulis), identi®edin PVC mattresses used by babies who died ofsudden infant death syndrome (SIDS), acted on®re retardant compounds containing antimony,arsenic or phosphorus to form poisonoushydride gases (`CookReport'. Cot death (poison-ings). Central Television, 17 November 1994).This hypothesis was supported by the ®nding ofhigher blood antimony concentrations in sam-ples from SIDS babies than in those who haddied from other causes.165 Correspondence inThe Lancet and editorials in the British MedicalJournal have followed.166±169 The original workhas been repeated with the inclusion this timeof essential control experiments.170 This hasshown that the fungus S. brevicaulis was notpresent on all the mattresses. Mercuric iodideand silver nitrate test papers (as used in the ®rststudy) gave colour responses whenever anybacterial growth occurred, possibly due to thepresence of sulphur compounds. No antimony,arsenic or phosphorus could be detected in themattresses using a variety of methods.Antimony is used in glass, paint and in the

ceramics, printing, explosives and semiconduc-tor industries. Its compounds are used as ®reretardants and the sulphide is used in medicalscintigraphy. Antimonials are used for treatingschistosomiasis (bilharzia) and the more resis-tant cutaneous and visceral forms of leish-maniasis (kala-azar) and may give rise totoxicity.

ToxicokineticsAntimony is in the same group (Va) of theperiodic table as arsenic, but is heavier. Its rangeof chemical forms is similar, as are theirproperties and biological effects, although theantimonials are slightly less toxic. Following theconcerns about antimony in babies, studies oninfants' urine have shown a variable but lowexcretion which is slightly higher in pre-termbabies. There was no correlation with urinecotinine, suggesting that tobacco and passivesmoking are not an important source ofexposure to antimony.171

Clinical manifestationsThe hydride gas stibine (SbH3) is the most toxicform and causes headache, nausea, vomiting,jaundice and anaemia. Acute exposure to otherantimonials affects the skin and causes hair lossand weight loss. Damage to heart, liver andkidney may be found and myocardial infarctionis possible. Pentavalent compounds are less toxicthan trivalent antimonials and are preferred fortherapeutic use. Chronic exposure leads to skinproblems, with antimony spots, irritation anddermatitis.172

Laboratory investigations and therapyMeasurement of antimony in urine is used foroccupational monitoring and in EDTA bloodsamples for therapeutic monitoring or suspectedtoxicity. Reference values are shown in Table 12.For treatment, the agent of choice would beDMPS but it has proved possible to managepatients who have had blood antimony concen-trations as high as 820 nmol/L (100 mg/L)following acute exposure without chelation.

BISMUTH

This element has a few industrial uses inpigments, ceramics, and low-melting alloys. Asthe chelate tripotassium dicitratobismuthate, ithas recently become important in the triple



TABLE 11. Arsenic reference values

Blood* Urine

Unexposed persons 5130 nmol/L (510mg/L) 5130nmol/24 h (510mg/24 h), or540 nmol/mmol creatinine (53 mg/g)

Occupational monitoring{ 5230nmol/mmol creatinine(517 mg/g)

NB Seafoods cause greatly increased values and are best omitted for 5 days prior to sampling.*Ethylenediaminetetra-acetic acid (EDTA) whole blood, 10mL in a plastic tube. {Random urine samples collectedtowards the end of the working week.

therapy against Helicobacter pylori in patientswith gastric or duodenal ulcers. It is a well-known over-the-counter remedy for indigestionand constipation.

Toxicokinetics and clinical manifestationsBismuth resembles lead and mercury in form-ing nuclear inclusion bodies in renal cells and,in common with gold and mercury, can causedevelopment of a nephrotic syndrome. In themid-1970s a number of patients taking verylarge doses developed an encephalopathy whichproved to be reversible. Thomas et al.173

showed that some patients absorb considerablymore bismuth than others and maintain highblood and urine concentrations, although renalclearance is normally rapid at low bloodconcentrations. The effect does not seem tobe dose-related as some patients with highblood concentrations do not get encephalo-pathy.

Laboratory investigations and therapyBismuth can be measured in blood or urine.Because it is normally rapidly excreted in urinethere is a greater risk of higher blood concentra-tions and consequent problems in patients withrenal impairment (see Case history 4). Referencevalues are shown in Table 12. When treatment isnecessary, a chelating agent can be used.

Case history 4: bismuth poisoning(Playford et al.)174

A patient with poor renal function (creatinineclearance 15mL/min) who was being treatedwith bismuth developed encephalopathy aftertaking twice the recommended dose for 2months. He was given a course of DMPS.Bismuth clearance increased from 0´24 to2´4mL/min; after 50 days, blood bismuthconcentration had fallen from 4´21 to0´22mmol/L (from 880 to 46 mg/L).

TIN

Tin has uses in the tinplate, container and paintindustries and as a stabilizer of plastics. Tin¯uoride is used in toothpaste. Inorganic tin saltsare seldom toxic, but as with lead, organotincompounds are more toxic. They are widely usedas biocides in agriculture and as anti-foulingagents in marine paints. Tributyltin is the bestknown. Environmental pollution of rivers andestuaries from this source is increasingly recog-nized as is contamination from factories makingthe agricultural biocides.175

ToxicokineticsInorganic tin is an inducer of haem oxygenase.However, tin metalloporphyrin inhibits thisenzyme and has been tested, as have severalother metalloporphyrins, as therapeutic agentsin reducing haem conversion to bilirubin inseveral clinical situations, e.g., Crigler±Najjarsyndrome, acute hepatic porphyria and neonataljaundice.

In the body tetraethyl- and tetramethyltin,like the organic lead equivalents, are dealkylatedrapidly to toxic trialkyl derivatives and are thenmetabolized more slowly to the less toxic di- andmono-derivatives which are less toxic.15 Organo-tin compounds are genotoxic: dibutyltin is themost toxic compound in this respect.

Clinical manifestationsTin hydride is a neurotoxic gas but, unlikearsine, it does not cause haemolysis. Severeirritation and burning occurs if skin is exposedto organic tin, which may be absorbed and leadto systemic features including anaemia and renaland hepatocellular damage. Triethyltin un-couples oxidative phosphorylation, causingmitochondrial damage, and causes cerebraloedema leading to neurotoxicity. Organotincompounds may stimulate catecholamines



TABLE 12. Reference values for antimony, bismuth, tin

Blood (or serum) Urine

Antimony* 2´5±28´7 nmol/L (0´3±3´5mg/L) 1´6±18 nmol/L (0´2±2´2mg/L)Bismuth*Therapeutic 50´5 nmol/L (0´1mg/L) 3´8±7´7 nmol/L (0´8±1´6mg/L)Warning range Up to 240nmol/L (50 mg/L)Toxic 240±480nmol/L (50±100mg/L)

4480 nmol/L (100mg/L)Tin{ 50´25mmol/L (30mg/L) 50´22mmol/L (526 mg/L)

*Ethylenediaminetetra-acetic acid (EDTA) whole blood, 2mL in a plastic tube. {Serum, 2mL in a plastic tube.

leading to hyperglycaemia, changes in bloodpressure and immunotoxicity.176

Laboratory investigationsUrinary excretion of tin is suitable for monitor-ing occupational exposure to inorganic tinspecies, together with haematological indices.Techniques for the speciation of organic tinusing gas chromatography with mass spectro-metry are available,177 but measurement of totaltin in blood or urine can provide adequateindication of recent exposure. Reference valuesare shown in Table 12, although these are basedonly on the small number of reports available.

CONCLUSION

The widespread effects of poisoning by metalsmakes this dif®cult to identify if exposure is notreadily apparent. There are a few presentationswhich suggest speci®c elements, for example thecharacteristic hair loss within thallium, thehepatic effects of arsenic and the renal effectsof mercury. Much also depends on the degreeand duration of exposure. The effects of leadpoisoning with low exposure is a matter of majorconcern. The solution should lie in public healthmeasures to manage the lead content of watersupplies and effective abatement procedures todeal with old leaded paint. Nevertheless, child-ren can still present as acute medical emergenciesrequiring urgent treatment to avoid lead en-cephalopathy and its long-term sequelae. Theimportance of taking a careful history, includingquestions to identify any non-prescribedsupplements, cannot be over-emphasized. Thepossibility that more than one metal may bepresent should not be overlooked: for example,lead and arsenic can occur on demolition sites,mercury and arsenic in old waste land®ll sites,and combinations of features of mercury, leadand arsenic poisoning have been found inpatients using Asian remedies.116,178

In addition to poisoning from the metalsdiscussed, new problems are constantly arising:technological developments in medicine maylead to toxicity, in some cases because thebody's normal defences are by-passed e.g.,aluminium in renal failure and parenteralnutrition, manganese in parenteral nutritionand silicone in breast implants (although theeffects of this remain unproven and the siliconecompound rather than silicon itself is impli-cated).179 The newest concern relates to thepossible effect of wear debris from metal joint

prostheses, which have been shown to accumu-late in lymph nodes. Nutritional supplements arealso potential source of problems: in at least onecase, excessive use has been partly responsiblefor copper overload and liver failure (O'Dono-ghue, personal communication); the sale ofgermanium-containing supplements was even-tually banned because of the risks of renaltoxicity.

AcknowledgementsWe would like to acknowledge the help receivedas a result of the generous spirit of scienti®cexchange which exists among DB's colleagueson the SAAS Trace Element Sub-Committee:Drs Braithwaite, Delves, Fell, Halls, Taylor,Walker, White and Widdop and Messrs House,Newton and Sheehan. In particular, Dr AndrewTaylor has helped with many queries andprovided valuable criticism on early drafts ofthe manuscript. We also thank the members ofCLISC for their patience and enthusiasm duringthe preparation of this review, and for theirinvaluable comments.

REFERENCES

1 Health and Safety Executive Annual Report:Statistical Supplement 1992±93. London: HMSO,1993

2 Braithwaite RA, Girling AJ. Bovine referencematerials for accuracy control of blood leadanalysis. Fresenius Z Anal Chem 1988; 332: 704±9

3 Water Pollution at Lowermoor North Cornwall.Report of the Lowermoor Incident Health Ad-visory Group. Truro: Cornwall and Isles of ScillyDistrict Health Authority, 1989

4 Identi®cation and control of pollutants of broadinternational signi®cance. Document A/CONF. 48/8 prepared for the United Nations Conference onthe Human Environment. Stockholm, 1972

5 World Health Organization. Environmental HealthCriteria 3. Lead. Geneva: WHO, 1977

6 Nordberg GF. Application of the `critical effect'and `critical concentration' to human risk assess-ment for cadmium. IARC Sci Publ 1992; 118: 3±14

7 Control of Lead at Work. Approved Code ofPractice. London: HMSO, 1985

8 Control of Lead at Work Regulations 1998.London: The Stationery Of®ce Ltd, 1998

9 Clarkson TW. Molecular targets of metal toxicity.In: Brown SS, Savory J, eds. Chemical Toxicologyand Clinical Chemistry of Metals. London: Aca-demic Press; 1983: 217±26

10 Chisholm JJ, Barltrop D. Recognition and man-agement of children with increased lead absorp-tion. Arch Dis Child 1979; 54: 249±62

11 El-Sharkaw AM, Morgan WD, Cobbold S, JaibMB, Evans CJ, Somervaille LJ, et al. Unexpected



mobilisation of lead during cisplatin therapy.Lancet 1986; ii: 249±50

12 Silbergeld EK, Sauk J, Somerman M, Todd A,McNeill F, Fowler B, et al. Lead in bone; storagesite, exposure source and target organ. Neurotox-icology 1993; 142±3: 225±36

13 Coyle P, Philcox JC, Rofe AM. Clinical signi®-cance of metallothionein: a dealer in heavy metals.Clin Biochem Revs 1993; 14: 118±25

14 Shelton KR, Egle PM, Bigbee JW, Klann E. Anuclear matrix protein stabilised by lead exposure:current knowledge and future prospects. Neurotox-icology 1993; 14: 61±8

15 Mushak P. Mammalian biotransformation pro-cesses involving various toxic metalloids andmetals. In: Brown SS, Savory J, eds. ChemicalToxicology and Clinical Chemistry of Metals.London: Academic Press, 1983: 227±45

16 Graziano JH, Leong JK, Friedheim E. 2,3-Dimercaptosuccinic acid: a new agent for thetreatment of lead poisoning. J Pharmacol ExpTher 1978; 206: 696±700

17 Glotzer DE. The current role of 2,3-dimercapto-succinic acid (DMSA) in the management ofchildhood lead poisoning. Drug Safety 1993; 9:85±92

18 Rey-Alvarez S, Menke-Hargrave T. Deleadingdilemma: pitfalls in the management of childhoodlead-poisoning. Paediatrics 1987; 79: 214±17

19 Shiwen C, Yin Y, Zhineng H, Xiangu Z, Zhadu Y,Huidong L, et al. Cadmium exposure and healtheffects among residents in an irrigation area withore dressing wastewater. Sci Total Environ 1990;90: 67±73

20 Buchet JP, Lauwerys R, Roels H, Bernard A,Bruaux P, Claeys F, et al. Renal effects of thecadmium burden of the general population. Lancet1990; 336: 699±702

21 Herber RFM. The World Health Organisationstudy on health effects of exposure to cadmium:morbidity studies. IARC Sci Publ 1992; 118:347±58

22 Friberg LT, Piscator M, Nordberg GF. Occur-rence, possible routes of exposure and daily intake.In: Weast RC, ed. Cadmium in the Environment.Cleveland: CRC Press, 1971

23 Buckler HM, Smith WDF, Rees WDW. Selfpoisoning with oral cadmium chloride. BMJ1986; 292: 1559±60

24 Kjellstrom T. Mechanism and epidemiology ofbone effects cadmium. IARC Sci Publ 1992; 118:301±10

25 Chambers RC, McNaulty RJ, Shock A, Campa JS,Newman Taylor AJ, Laurent GJ. Cadmiumselectively inhibits ®broblast procollagen produc-tion and proliferation. Am J Physiol 1994; 267:L300±08

26 Toffoletto F, Apostoli P, Ghezzi I, Baj A, CortonaG, Rizzi I, et al. Ten year follow up of biologicalmonitoring of cadmium-exposed workers. IARCSci Publ 1992; 118: 107±11

27 Roels H, Bernard AM, Cardenas A, Buchet JP,Lauwerys R, Hotter G, et al. Markers of earlyrenal changes induced by industrial chemicals, III.

Application to workers exposed to cadmium. Br JInd Med 1993; 50: 37±48

28 Lauwerys RR, Bernard AM, Roels HA, Buchet JP.Cadmium markers as predictors of nephrotoxiceffects. Clin Chem 1994; 40: 1391±4

29 Bernard A, Roels H, Buchet JP, Cardenas A,Lauwerys R. Cadmium and health: the Belgianexperience. IARC Sci Publ 1992; 118: 15±33

30 Davison AG, Newman Taylor AJ, Darbishire J,Chettle DR, Guthrie CJ, O'Malley D, et al.Cadmium fume inhalation and emphysema. Lancet1988; i: 663±7

31 Carruthers MM, Smith B. Evidence of cadmiumtoxicity in a population living in a zinc-miningarea. Pilot survey of Shipham residents. Lancet1979; i: 845±7

32 Lemen RA, Lee JS, Wagoner JK, Blejer HP.Cancer mortality among cadmium productionworkers. Ann NY Acad Sci 1976; 271: 274±9

33 Armstrong BG, Kazantzis G. The mortality ofcadmium workers. Lancet 1983; i: 1425±7

34 Thun MJ, Schnorr TM, Smith AB, Halperin WE,Lemen RA. Mortality among a cohort of UScadmium production workers ± an update. J NatlCancer Inst 1985; 75: 325±33

35 Kazantis G, Lam TH, Sullivan KR. Mortality ofcadmium-exposed workers. A ®ve-year update.Scand J Work Environ Health 1988; 14: 220±3

36 Kazantsis G, Blanks RG, Sullivan KR. Is cadmiuma human carcinogen? IARC Sci Rep 1992; 118:435±46

37 Beevers DG, Campbell BC, Goldberg A, MooreMR, Hawthorne VM. Blood cadmium in hyper-tensives and normotensives. Lancet 1976; ii:1222±4

38 Staessen R, Bernard A, Buchet JP, Claeys F,Dekempeneer L, Ducoffre G, et al. Effects ofcadmium exposure on the cardiovascular systemand on calcium metabolism: results of a cross-sectional populations study. IARC Sci Publ 1992;118: 263±9

39 Braithwaite RA, Armstrong R, Franklin DM,Chettle DR, Scott MC. Cadmium toxicokineticsfollowing long term occupational exposure. In:Aitio A, Aro A, Jarvisalo J, Vaino H, eds.Proceedings of the Joint International Symposiumof the Nordic Trace Element Society and theInternational Union of Pure and Applied Chemistry:Trace Elements in Health and Disease. Espoo,Finland 5±8 June 1990. Cambridge: Royal Societyof Chemistry, 1991: 95±103

40 Sartor F, Rondia D, Claeys F, Buchet JP, DucoffreG, Lauwerys R, et al. Factors affecting thecadmium body burden in a population study.IARC Sci Publ 1992; 118: 101±6

41 Gompertz D, Chettle DR, Fletcher JG, Mason H,Perkins J, Scott MC, et al. Renal dysfunction incadmium smelters: relation to in-vivo liver andkidney cadmium concentrations. Lancet 1983; i:1185±7

42 Roels H, Lauwerys R, Buchet JP, Bernard A,Chettle DR, Al-Haddad IK. In vivo measurementof liver and kidney cadmium in workers exposed tothis metal. Environ Res 1981; 26: 217±40



43 Donaldson MD, Chambers RT, Woolridge MW,Whicher JT. Stability of alpha-1-microglobulin,beta-2-microglobulin and retinol binding protein inurine. Clin Chim Acta 1989; 179: 73±8

44 Schaller KH, Angerer JH. Biological monitoring inthe occupational setting ± relation to cadmiumexposure. IARC Sci Rep 1992; 118: 53±64

45 Jung K, Pergande M, Graubaum HJ, Fels LM,Endl U, Stolte H. Urinary proteins and enzymes asearly indicators of renal dysfunction in chronicexposure to cadmium. Clin Chem 1993; 39:757±65

46 Mueller PW. Detecting the effects of cadmiumtoxicity [editorial; comment]. Clin Chem 1993; 39:743±5

47 Bernard A, Roels H, Thielemans N, Van-Lierde M,Lauwerys R. Assessment of the causality of thecadmium protein relationships in the urine of thegeneral population with reference to the Cadmibelstudy. IARC Sci Publ 1992; 118: 341±6

48 Health and Safety Executive. Cadmium: Health andSafety Precautions: Guidance Note EH1. London:HMSO, 1986

49 Eybl V, Kotyzova D, Koutensky J, Jones MM,Singh PK. Effect of chelators, monoisoamyl meso-2,3-dimercaptosuccinate and N-(4-methylbenzyl)-4-O-(beta-D-galactopyranosyl)-D-glucamine-N-car-bodithioate on cadmium and essential element levelsin mice. Analyst 1995; 120: 855±7

50 Major RH. Some landmarks in the historyof lead poisoning. Ann Med Hist 1931; 3:218±27

51 Franklin B. The pernicious effects of lead. In:Franklin B, ed. The Works of Ben Franklin.Philadelphia: BC Buzby, 1818

52 Grandjean P. International perspectives of leadexposure and lead toxicity. Neurotoxicology 1993;14: 7±14

53 Fell GS. Lead toxicity: problems of de®nition andlaboratory evaluation. Ann Clin Biochem 1984; 21:453±60

54 Goyer RA. Lead toxicity: current concerns. EnvHealth Perspect 1993; 100: 177±87

55 Coyle P, Haywood MR, Rofe AM. Living withlead. An Australian overview. Clin Biochem Revs1994; 15: 179±87

56 Matte TD, Proops D, Palazuelos E, Graef J, AvilaHA. Acute high-dose lead exposure from beveragecontaminated by traditional Mexican pottery.Lancet 1994; 344: 1064±5

57 Bayly GR, Braithwaite RA, Sheehan TMT, DyerNH, Grimley C, Ferner RE. Lead poisoning fromAsian traditional remedies in the West Midlands ±report of a series of ®ve cases. Hum ExperimentToxicol 1995; 14: 24±8

58 Needleman HL, Gunnoe C, Leviton A, Reed R,Peresie H, Maher C, et al. De®cits in psychologicaland classroom performance in children withelevated dentine lead levels. N Engl J Med 1979;300: 689±95

59 Needleman HL, Schell A, Bellinger D. The long-term effects of exposure to low doses of lead inchildren: an 11 year follow up report.N Engl J Med1990; 322: 83±8

60 Delves HT, Clayton BE, Carmichael A. Anappraisal of the analytical signi®cance of toothlead measurements as possible indices of environ-mental exposure of children to lead. Ann ClinBiochem 1982; 19: 329±37

61 Hatzakis A, Kokkevi A, Maravelias C, Katsouyan-ni K, Salaminios F, Kalandidi A, et al. Psycho-motor intelligence de®cits in lead-exposed children.In: Smith MA, Green LD, Sors A, eds. LeadExposure and Child Development. London: Kluwer,1989: 211±23

62 Baghurst PA, McMichael AJ, Wigg NR, VimpaniGV, Robertson EF, Roberts RJ, et al. Environ-mental exposure to lead and children's intelligenceat the age of seven years; the Port Pirie cohortstudy. N Engl J Med 1992; 327: 1279±84

63 Harvey PG, Hamlin MW, Kumar R, Morgan G,Spurgeon A, Delves HT. Relationship betweenblood lead, behaviour, psychometric and neuro-psychological test performance in young children.Brit J Dev Psychol 1988; 6: 145±56

64 Fulton M, Raab G, Thomson G, Laxen D, HunterR, Hepburn W. In¯uence of blood lead on theability and attainment of children in Edinburgh.Lancet 1987; i: 1221±6

65 Pocock SJ, Smith M, Baghurst P. Environmentallead and children's intelligence: a systematic reviewof the epidemiological evidence. BMJ 1994; 309:1189±97

66 Palca J. Lead researcher confronts accusers inpublic hearing. Science 1992; 256: 437±8

67 CDC Strategic Plan for the Elimination of Child-hood Lead Poisoning. Atlanta: Centers for DiseaseControl Departments of Health and HumanServices, 1991

68 Diermeyer M, Hedberg K, Fleming D. Backing offuniversal childhood lead screening in the USA:opportunity or pitfall? [commentary]. Lancet 1994;344: 1587±8

69 Department of the Environment. European Com-munity Screening Programme for Lead: UnitedKingdom Results for 1981. Pollution Report No 18.London: HMSO, 1984

70 Department of the Environment. UK Blood LeadMonitoring Programme 1984±1987: Results for1987. Pollution Report No 28. London: HMSO,1990

71 Environmental Data Services Ltd, London. ENDSReport 1997 (269, June)

72 Millborne E, Russell J. Britain must replace its leadpipes to meet WHO standards for drinking water[letter]. BMJ 1995; 310: 1408±9

73 Church HJ, Day P, Braithwaite RA, Brown SS.Binding of lead to a metallothionein-like protein inhumanerythrocytes.J InorgBiochem1993;49: 55±68

74 Barltrop D, Smith AM. Kinetics of lead interac-tions with human erythrocytes. Postgrad Med J1985; 51: 770±3

75 Rossi E, Costin KA, Garcia-Webb P. Effect ofoccupational lead exposure on lymphocyte en-zymes involved in haem biosynthesis. Clin Chem1990; 36: 1980±3

76 Labbe RF. Lead poisoning mechanisms [editorial].Clin Chem 1990; 36: 1870±1



77 Pagliuca A, Baldwin D, Lestas AN, Wallis RM,Bellingham AJ, Mufti GJ. Lead poisoning: clinicalbiochemical and haematological aspects of a recentoutbreak. J Clin Pathol 1990; 43: 277±81

78 Lee WR. What happens in lead poisoning? J RoyColl Phys 1981; 15: 48±54

79 Cardenas A, Roels H, Bernard AM, Barbon R,Buchet JP, Lauwerys RR, et al. Markers of earlyrenal changes induced by industrial pollutants. IIApplication to workers exposed to lead. Br J IndMed 1993; 50: 28±36

80 Chisholm JJ Jr. Lead poisoning. Sci Am 1971; 224:15±23

81 Graham JA, Maxton DG, Twort CH. Painter'spalsy: a dif®cult case of lead poisoning. Lancet1981; ii: 1159±61

82 Chisholm JJ Jr. Mobilisation of lead by calciumdisodium edetate: a reappraisal. Am J Dis Child1987; 141: 1256±7

83 Berger OG, Gregg DJ, Succop PA. Using un-stimulated urinary lead excretion to assess the needfor chelation in the treatment of lead poisoning. JPaediatr 1990; 116: 46±51

84 Piomelli S, Rosen JF, Chisholm JJ, Graef JW.Management of childhood lead poisoning. Paedia-trics 1984; 105: 523±32

85 Braithwaite RA, Sheehan TM, Davies RC. Bloodlead analysis: beware of contaminated collectiontubes. Association of Clinical Biochemists NewsSheet 1991 (344, December): 26±7

86 Crick J, Flegal AR. Contaminant lead in blood-collection tubes for trace element studies. ClinChem 1991; 38: 600±1

87 Taylor A. The use of internal quality controlmaterials for the preparation and maintenance ofreliable methods for the measurement of lead inblood. Fresenius Z Anal Chem 1988; 332: 732±5

88 Bullock DG, Smith NJ, Whitehead TP. Externalquality assessment of assays of lead in blood. ClinChem 1986; 32: 1884±9

89 Delves HT. Analytical techniques for blood leadmeasurement. J Anal Toxicol 1977; 1: 262±4

90 Taylor A, Majumdar G, Delves HT. Probableself-administration of lead [letter]. Lancet 1985; i:513

91 Taylor A. Usefulness of measurements of traceelements in hair. Ann Clin Biochem 1986; 23:364±78

92 Campbell MJ, Delves HT. Accurate and precisedetermination of lead isotope ratios in clinical andenvironmental samples using inductively coupledplasma source mass spectrometry. J Anal AtomSpectrom 1989; 4: 235±6

93 Keen RW, Deacon AC, Delves HT, Moreton JA,Frost PG. Indian herbal remedies for diabetes as acause of lead poisoning. Postgrad Med J 1994; 70:113±4

94 Piomelli S, Bernard D, Guinee VF, Young P, GayG. The FEP free erythrocyte porphyrin test: asimple sensitive screening test for lead poisoning.Clin Chem 1972; 50: 254±9

95 Zwennis WCM, Franssen AC, Wijnans M. Use ofzinc protoporphyrin in screening individuals forexposure to lead. Clin Chem 1990; 36: 1456±9

96 Parsons PJ, Reilly AA, Hussain A. Observationalstudy of erythrocyte protoporphyrin screeningtest for detecting low lead exposure in children:impact of lowering the blood lead action thresh-old. Clin Chem 1991; 37: 216±25

97 Hurst PL, Hughes PM. Poor correlation betweenerythrocytic 5-aminolaevulinic acid dehydrataseactivity and blood lead concentration [letter]. ClinChem 1985; 31: 1244±5

98 Holden A. St Albans Poisoner: Life and Crimes ofGraham Young. London: Black Swan (Trans-world), 1995

99 Reed MD, Crawley J, Faro SN, Pieper SJ,Kurland LT. Thallotoxicosis: acute manifesta-tions and sequelae. JAMA 1963; 183: 516±22

100 Insley BM, Grufferman S, Ayliffe A. Thalliumpoisoning in cocaine abusers. Am J Emerg Med1986; 4: 545±8

101 Dolgner R, Brockhaus A, Ewers U, Wiegand H,Majewski F, Soddemann H. Repeated surveil-lance of exposure to thallium in a populationliving in the vicinity of a cement plant emittingdust containing thallium. Int Arch Occup EnvironHealth 1983; 52: 79±84

102 Moore D, House I, Dixon A. Thallium poisoning.Diagnosis may be elusive but alopecia is the clue.BMJ 1993; 306: 1527±9

103 Chandler HA, Archbold GPR, Gibson JM,O'Callaghan P, Marks JN, Pethybridge RJ.Excretion of a toxic dose of thallium. Clin Chem1993; 36: 1506±9

104 Cavanagh JB. What have we learned fromGraham Frederick Young? Re¯ections on themechanisms of thallium neurotoxicity. Neuro-pathol Appl Neurobiol 1991; 17: 3±9

105 Martin-Gil J, Delibes de Castro G, Zapatero-Magdaleno P, Sarabia-Herrero FJ. Preserving theancients with vermilion [letter]. Lancet 1994; 344:1776

106 O'Shea JG. The death of Paganini. J Roy CollPhys 1988; 22: 104

107 Stephenson JBP. Pink disease [letter]. BMJ 1966;i: 1110±1

108 Yeh T, Pildes RS, Firor HV, Szanto PB. Mercurypoisoning from mercurochrome therapy ofinfected omphalocoele [letter]. Lancet 1978; i: 210

109 Laundy T, Adam AE, Kershaw JB, Rainford DJ.Deaths after peritoneal lavage with mercuricchloride solutions: case report and review of theliterature. BMJ 1984; 289: 96±8

110 Peterson DE, Kanarek MS, Kuykendall MA,Diedrich JM, Anderson HA, Remington PL, etal. Fish consumption patterns and blood mercurylevels in Wisconsin Chippewa Indians. ArchEnviron Health 1994; 49: 53±8

111 Svensson BG, Nilsson A, Jonsson E, Schutz A,Akesson B, Hagmar L. Fish consumption andexposure to persistent organochlorine com-pounds, mercury, selenium and methylaminesamong Swedish ®sherman. Scand J Work EnvironHealth 1995; 21: 96±105

112 Harada M. Minnamata disease: methyl mercurypoisoning in Japan caused by environmentalpollution. Crit Rev Toxicol 1995; 25: 1±24



113 Palheta D, Taylor A. Mercury in environmentaland biological samples from a gold mine in theAmazon region of Brazil. Sci Total Environ 1995;168: 63±9

114 Mant TG, Lewis JL, Mattoo TK, Rigden SP,Volans GN, House IM, et al. Mercury poisoningafter disc battery ingestion. Hum Toxicol 1987; 6:179±81

115 Kew J, Morris C, Aihie A, Fysh R, Jones S,Brooks D. Arsenic and mercury intoxication dueto Indian ethnic remedies. BMJ 1993; 306:506±7

116 Sheerin NS, Monk PN, Aslam M, Thurston H.Simultaneous exposure to lead, arsenic andmercury from Indian ethnic remedies. Br J ClinPract 1994; 48: 332±3

117 Lauwerys R, Bonnier C, Evrard P, Gennart JP,Bernard A. Prenatal and early postnatal intoxica-tion by inorganic mercury resulting from thematernal use of mercury containing soap. HumToxicol 1987; 6: 253±6

118 Mucklow ES. Health hazard from mercury soaps.Lancet 1989; i: 448

119 Tunnessen WW Jr, McMahon KJ, Baser M.Acrodynia: exposure to mercury from ¯uorescentlight bulbs. Paediatrics 1987; 79: 786±7

120 Wendroff AP. Domestic mercury pollution[letter]. Nature 1991; 347: 623

121 Friberg L. Inorganic Mercury. InternationalProgramme on Chemical Safety, EnvironmentalHealth Criteria 118. Geneva: World HealthOrganisation, 1991

122 Fowler BA. Mechanisms of kidney cell injuryfrom metals. Environ Health Perspect 1992; 100:57±63

123 Clarkson TW. Mercury: major issues in environ-mental health. Environ Health Perspect 1992; 100:31±8

124 Pendergrass JC, Haley BE, Vimy MJ, Win®eldSA, Lorscheider FL. Mercury vapor inhalationinhibits binding of GTP to tubulin in rat brain:similarity to a molecular lesion in Alzheimerdiseased brain. Neurotoxicology 1997; 18: 315±24

125 Gala-Gorchev H. Dietary intake, levels in foodand estimated intake of lead, cadmium andmercury. Food Addit Contam 1993; 10: 115±28

126 Weiner JA, Nylander M. An estimation of theuptake of mercury from amalgam ®llings basedon urinary excretion of mercury in Swedishsubjects. Sci Total Environ 1995; 168: 255±65

127 Barregard L, Sallsten G, Jarvholm B. People withhigh mercury uptake from their own dentalamalgam ®llings. Occup Environ Med 1995; 52:124±8

128 Halbach S. Combined estimation of mercuryspecies released from amalgam. J Dent Res1995; 74: 1103±9

129 Clausen J. Mercury and multiple sclerosis. ActaNeurol Scand 1993; 87: 461±4

130 Fung YK, Meade AG, Rack EP, Blotcky AJ.Brain mercury in neurodegenerative disorders. JToxicol Clin Toxicol 1997; 35: 49±54

131 Fung YK, Meade AG, Rack EP, Blotcky AJ,Claassen JP, Beatty MW, et al. Determination of

blood mercury concentrations in Alzheimer'spatients. J Toxicol Clin Toxicol 1995; 33: 243±7

132 Eley BM. The future of dental amalgam: a reviewof the literature. Part 6: possible harmful effectsof mercury from dental amalgam. Br Dent J 1997;182: 455±9

133 Henriksson E, Mattsson U, Hakansson J. Heal-ing of lichenoid reactions following removal ofamalgam. A clinical follow-up. J Clin Periodontol1995; 22: 287±94

134 Kanerva L, Komulainen M, Estlander T, JolankiR. Occupational allergic contact dermatitis frommercury. Contact Dermatitis 1993; 28: 26±8

135 Handley J, Todd J, Burrows D. Mercury allergyin a contact dermatitis clinic in Northern Ireland.Contact Dermatitis 1993; 29: 258±61

136 Echeverria D, Heyer NJ, Martin MD, NalewayCA, Woods JS, Bittner AC. Behavioural effectsof low-level exposure to elemental mercuryamong dentists. Neurotoxicol Teratol 1995; 17:161±8

137 Gallagher JD, Noelle RJ, McCann FV. Mercurysuppression of a potassium current in human Blymphocytes. Cell Signal 1995; 7: 31±8

138 Goodwin CA, Wheeler-Jones CP, Namiranian S,Bokkala S, Kakkar VV, Authi KS, et al.Increased expression of procoagulant activity onthe surface of human platelets exposed to heavymetal compounds. Biochem J 1995; 308: 15±21

139 Franchi E, Loprieno G, Ballardin M, Petrozzi L,Migliore L. Cytogenetic monitoring of ®shermenwith environmental mercury exposure. Mutat Res1994; 320: 23±9

140 Jansson G, Harms-Ringhdahl M. Stimulatingeffects of mercuric and silver ions on the super-oxide anion production in human polymorpho-nuclear lymphocytes. Free Radic Res Commun1993; 18: 87±98

141 Herrstrom P, Holmen A, Karlsson A, Raihle G,Schutz A, Hogstedt B. Immune factors, dentalamalgam and low dose exposure to mercury inSwedish adolescents. Arch Environ Health 1994;49: 160±4

142 Levin M, Jacobs J, Polos PG. Acute mercurypoisoning and mercurial pneumonitis from goldore puri®cation. Chest 1988; 94: 554±6

143 Waldron HA. Did the Mad Hatter have mercurypoisoning? BMJ 1983; 287: 1961

144 Buxton JT, Hewitt JC, Gadsden R, Bradham GB.Metallic mercury embolism. Report of cases.JAMA 1965; 193: 573±5

145 Auer C, Ducrey N, Uffer S, Othenin-Girard P,Herbot CP. Self-mutilating intraocular injectionof metallic mercury. Arch Ophthalmol 1997; 115:556±7

146 Giombetti RJ, Rosen DH, Kuczmierczyk AR.Repeated suicide attempts by the intravenousinjection of elemental mercury. Int J PsychiatryMed 1988; 18: 153±67

147 Wright N, Yeoman WB, Carter GF. Massive oralingestion of elemental mercury without poisoning[letter]. Lancet 1980; i: 206

148 Henningsson C, Hoffman S, McGonigle L,Winter JSD. Acute mercury poisoning



(acrodynia) mimicking phaeochromocytoma inan adolescent. J Paediatric 1993; 122: 252±3

149 McNeil NJ, Issler HC, Olver RE, Wrong OM.Domestic metallic mercury poisoning. Lancet1984; i: 269±71

150 Campbell JR, Clarkson TW, Omar MD. Thetherapeutic use of 2,3-dimercaptopropane-1-sul-phonate in two cases of mercury poisoning.JAMA 1986; 256: 3217±30

151 Toet AE, van Dijk A, Savelkoul TJ, Meulenbelt J.Mercury kinetics in a case of severe mercuricchloride poisoning treated with dimercapto-1-propane sulphonate. Hum Exp Toxicol 1994; 13:11±6

152 Shibata A, Ohneseit PF, Tsai YC, Spruck CH,Nichols PW, Chiang HS, et al. Mutationalspectrum in the p53 gene in bladder tumoursfrom the endemic area of blackfoot disease inTaiwan. Carcinogenesis 1994; 15: 1085±7

153 Chatterjee A, Das D, Mandal BK, ChowdhuryTR, Samanta G, Chakraborti D. Arsenic inground water in six districts of West Bengal,India: the biggest arsenic calamity in the World.Part 1. Arsenic species in drinking water and urineof affected people. Analyst 1995; 120: 643±50

154 Das D, Chatterjee A, Mandal BK, Samanta G,Chakraborti D, Chanda B. Arsenic in groundwater in six districts of West Bengal, India: thebiggest arsenic calamity in the World. Part 2.Arsenic concentration in drinking water, hair,nails, urine, skin-scale and liver tissue (biopsy) ofthe affected people. Analyst 1995; 120: 917±24

155 Stoeppler M, Vahter M. Arsenic. In: HerberRFM, Stoeppler M, eds. Trace Element Analysisin Biological Specimens. Amsterdam: Elsevier,1994: 291±320

156 Wilkinson SP, McHugh P, Horsley S, Tubbs H,Lewis M, Thould A, et al. Arsine toxicity aboardthe Asiafreighter. BMJ 1975; 3: 559±63

157 Clarkson TW. Molecular targets of metaltoxicity. In: Brown SS, Savory J, eds. ChemicalToxicology and Clinical Chemistry of Metals.London: Academic Press, 1983: 211±26

158 Lafontaine A. Health effects of arsenic. In: DiFerrante E, ed. CEC Trace Metals: Exposure andHealth Effects. Oxford: Pergamon Press, 1978:107±16

159 Heaven R, Duncan M, Vukelja SJ. Arsenicintoxication presenting with macrocytosis andperipheral neuropathy, without anaemia. ActaHaematol 1994; 92: 142±3

160 Chen CJ, Kuo TL, Wu MM. Arsenic and cancers[letter]. Lancet 1988; i: 414±5

161 Le XC, Cullen WR, Reimer KJ. Human urinaryarsenic excretion after one-time ingestion ofseaweed, crab and shrimp. Clin Chem 1994; 40:617±24

162 Vahter M. What are the chemical forms of arsenicin urine and what can they tell us about exposure?Clin Chem 1994; 40: 679±80

163 Koons RD, Peters CA. Axial distribution ofarsenic in individual human hairs by solidsampling graphite furnace AAS. J Anal Toxicol1994; 18: 36±40

164 Moore DF, O'Callaghan CA, Berlyne G, Ogg CS,Davies HA, House IM, et al. Acute arsenicpoisoning: absence of polyneuropathy aftertreatment with 2,3-dimercaptopropanesulphonate(DMPS). J Neurol Neurosurg Psychiatry 1994; 57:1133±5

165 Richardson BA. Sudden infant death syndrome: apossible primary cause. J Forensic Sci Soc 1994;34: 199±204

166 Fleming PJ, Cooke M, Chantler SM, Golding J.Fire retardants, biocides, plasticisers and suddeninfant deaths [editorial]. BMJ 1994; 309: 1594±6

167 Richardson BA. Cot mattresses and suddeninfant death syndrome [letter]. Lancet 1995; 345:1045

168 Howatson AG, Patrick WJA, Fell GS, LyonTDB, Gibson AAM. Cot mattresses and suddeninfant death syndrome [letter]. Lancet 1995; 345:1044±5

169 De-Wolff FA. Antimony and health [editorial].BMJ 1995; 310: 1216±7

170 Warnock DW, Delves HT, Campbell CK,Croudace IW, Davey KG, Johnson M, et al.Toxic gas generation from plastic mattresses andsudden infant death syndrome. Lancet 1995; 346:1516±20

171 Dezateux C, Delves HT, Stocks J, Wade A,Pilgrim L, Costeloe K. Urinary antimony ininfancy. Arch Dis Child 1997; 76: 432±6

172 White GP, Mathias CG, Davin JS. Dermatitis inworkers exposed to antimony in a smeltingprocess. J Occup Med 1993; 35: 392±5

173 Thomas DW, Sobecki S, Hartley TF, Coyle P,Alp MH. Variable absorption and excretion ofbismuth and its potential for toxicity. In: BrownSS, Savory J, eds. Chemical Toxicology andClinical Chemistry of Metals. London: AcademicPress, 1983: 391±4

174 Playford RJ, Matthews CH, Campbell MJ,Delves HT, Hla KK, Hodgson HJ, et al. Bismuthinduced encephalopathy caused by tripotassiumdicitratobismuthate in a patient with chronicrenal failure. Gut 1990; 31: 359±60

175 Kubulla J, Wilken RD, Jantzen E, Kwan KK,Chau YK. Speciation and genotoxicity of butyltincompounds. Analyst 1995; 120: 667±73

176 Walker AW, ed. Clinical and Analytical Hand-book, 3rd edition. Guildford, UK: SAAS TraceElement Laboratories, 1998

177 Schramel P, Wendler I, Angerer J. The determi-nation of metals (antimony, bismuth, lead,cadmium, mercury, palladium, platinum, tell-urium, thallium, tin and tungsten) in urinesamples by inductively coupled plasma-massspectrometry. Int Arch Occup Health 1997; 69:219±23

178 McElvaine MD, Harder EM, Johnson L, BaerRD, Satzger RD. Lead poisoning from the use ofIndian folk medicines. JAMA 1990; 264: 2212±3

179 Sanchez-Guerrero J, Colditz GA, Karlson EW,Hunter DJ, Speizer FE, Liang MH. Siliconebreast implants and the risk of connective-tissuediseases and symptoms. N Engl J Med 1995; 332:1666±70



Appendix 1: Units in common useInterpreting the range of units used for reportingresults and reference ranges requires great care;the Control of Lead at Work Code of Practice(Control of Lead at Work; Code of Practice)expresses blood lead concentrations in mg/dL;toxicologists frequently use mass units, clinicalscientists use mmol/L and the Ministry ofAgriculture, Fisheries and Food and the Water

Authorities use parts per million and billion. Forbiological monitoring of occupationally exposedsubjects, 24 h urine collections are not practic-able and random urine samples are used. Theresults are often expressed in terms of creatinine(unless this is extremely low) to compensate fordilution effects. Table 13 shows how thecommonest units may be interconverted orapproximated.



TABLE 13. Interconversion of units most frequently used

From To Calculation

ng/L nmol/L Divide by atomic weightmg/L mmol/L Divide by atomic weightmg/dL or mg/100mL mmol/L Multiply by 10, divide by atomic weightppm mg/L Equivalentppm mg/mL Equivalentppm mg/L Multiply by 1000mg/L* mmol/mmol creatinine Divide by atomic weight, divide by 10mg/g creatinine* mmol/mmol creatinine Divide by atomic weight, divide by 10

*Approximate conversions; normal creatinine concentration being assumed as 1 g/L, which is roughly equivalent to10mmol/L.

Documents

Heavy metal poisoning and its laboratory investigation · Heavy metal poisoning and its laboratory investigation Dianne R Baldwin and William J Marshall From the Department of Clinical