effects ofwelding on healthvii - american welding society · publication of the first report...
TRANSCRIPT
Effects of Weldingon Health VII
Keywords — Welding, health, cancer, disease, exposure, fumes, gases, literature review, noise,radiation, toxicology
Effects of Weldingon Health—VII
Research performed by Biomedical Toxicology Associates, Rockville, Maryland, under contract with theAmerican Welding Society and supported by industry contributions.
An updated (January 1986-December 1987) literature survey and evaluation of the data recorded since thepublication of the first report (1979). This series of reports is intended to aid in the understanding of thehealth effects of welding.
Performed by:
Winifred Palmer
July 1989
Abstract
This literature review, with 194 citations, was prepared under contract to the American Welding Societyfor its Safety and Health Committee. The review deals with studies of the fumes, gases, radiation, and noisegenerated during various arc welding processes. Section 1 summarizes recent studies of the occupationalexposures, while Section 2 contains information related to the human health effects of exposure to byprod-ucts of welding operations. Section 3 discusses studies of the effects of welding emissions on laboratoryanimals and in vitro cell systems. Referenced materials are available from Biomedical Toxicology Associates.
Prepared forSafety and Health Committee
American Welding Society550 N. W. LeJeune Road, P.O. Box 351040
Miami, Florida 33135
International Standard Book Number: 0-87171-344-6
American Welding Society,550 N.W. LeJeune Road, P.O. Box 351040, Miami, Florida, 33135
© 1990 by American Welding Society. All rights reservedPrinted in the United States of America
This report is published as a service and convenience to the welding industry and is the product of an inde-pendent contractor (Biomedical Associates) which is solely responsible for its contents. The materials in thisreport have not been independently reviewed or verified and are offered only as information. AWS assumesno responsibility for any claims that may arise from the use of this information. Users should make indepen-dent investigations to determine the applicability of this information for their purposes.
Personnel
AWS Safety and Health Committee
K. L. Brown, ChairmanK. A. Lyttle, 1st Vice Chairman
R. J. Simonton, 2nd Vice ChairmanM. E. Kennebeck, Jr., Secretary
O. J. Fisher*J. F. HinrichsD. G. Howden
W. S. HowesA. F. Manz
M. T. NeuG. R. Spies, Jr.*
M. J. Vasquez
The Lincoln Electric CompanyUnion Carbide CorporationINCO Alloy International, IncorporatedAmerican Welding SocietyConsultantA. O. Smith CorporationOhio State UniversityNational Electrical Manufacturers AssociationA. F. Manz AssociatesCaterpillar Tractor CompanyConsultantGeneral Motors Corporation
* Advisor
111
Foreword
(This Foreword is not a part of Effects of Welding on Health VII, but is included for informational purposesonly.)
This literature review was prepared for the Safety and Health Committee of the American Welding Societyto provide an assessment of current information concerning the effects of welding on health, as well as to aidin the formulation and design of research projects in this area, as part of an ongoing program sponsored by theCommittee. Previous work consists of the reports Effects of Welding on Health (I through VI) each coveringapproximately 18 months to two years. Conclusions based on this review and recommendations for furtherresearch are presented in the introductory portions of the report. Referenced materials are available fromBiomedical Toxicology Associates.
Comparative Listing — Welding Processes
Explanatory Note: Terms used in the technical literature sometimes do not correspond to those recom-mended by AWS in its publication ANSI/AWS A3.0, Standard Welding Terms and Definitions.
Accordingly, the following list may aid the reader in identifying the process in use.
EWH — VII Preferred AWS TermGas or Flame Cutting
Gas Welding
MAGMIG, GMAMMA, SMATIGWire
(OC)
(OFW)
(GMAW)(GMAW)(SMAW)(GTAW)
Oxygen Cuttingor (OFC) Oxyfuel Gas Cutting
Oxyfuel Gas Weldingor (OAW) Oxyacetylene Welding
— (with specified shielding gas)Gas Metal Arc WeldingShielded Metal Arc WeldingGas Tungsten Arc WeldingElectrode
Acknowledgments
Funds for this project were provided by the American Welding Society.The American Welding Society gratefully acknowledges the financial support of the program by industry
contributions.
Supporting Organizations
Air Products and Chemicals, IncorporatedAirco Welding ProductsAllis-ChalmersAlloy Rods Division, The Chemetron CorporationAWS Detroit SectionAWS New Orleans SectionArcos CorporationThe Binkley CompanyCaterpillar Tractor CompanyChicago Bridge and Iron CompanyGrove Manufacturing Company, Division of Kidde, IncorporatedGeneral Electric CompanyThe Heil CompanyHobart Brothers CompanyINCO Alloys InternationalLincoln Electric CompanyMiller Electric Manufacturing CompanyNational-Standard CompanyA.O. Smith CorporationTeledyne-McKay, IncorporatedTrinity Industries, IncorporatedTruck Trailer Manufacturers AssociationWalker Stainless Equipment CompanyWeld Tooling Corporation
Many other organizations have also made contributions to support the ongoing program from May 1979 tothe present.
VI
Table of Contents
Page No.
Personnel iiiForeword ivComparative Listing vAcknowledgments viList of Tables ixList of Figures ixIntroduction 1Executive Summary 3Technical Summary 5Conclusions 12
Section One — The Exposure
1. Fumes 151.1 Effects of Electrode Composition 161.2 Chromium 161.3 Nickel 161.4 Analytical Methods 161.5 Analysis of Metals in Biological Tissue 171.6 Particles 17
2. Ozone 19
3. Electromagnetic Radiation 19
4. Production Coatings 20
5. Noise 20
6. Hygiene and Work Practices 216.1 Accidents 216.2 Clothing 236.3 Protective Gear 236.4 Training 24
Section Two — Effects of Welding on Human Health
7. Respiratory Tract 247.1 Alveolar Macrophages 247.2 Estimation of Retained Particles in the Lungs 257.3 Pulmonary Function and Bronchitis 267.4 Acute Respiratory Tract Disease — Case Report 297.5 Production Coatings 30
vn
8. Cancer 308.1 Epidemiologic Studies 308.2 Metal Carcinogens 338.3 Screening for Cancer 34
9. Metal Fume Fever 34
10. Effects on the Ear and Hearing 35
11. Effects on the Eye and Vision 35
12. Effects on the Nervous System 37
13. Effects on the Skin 37
14. Sensitivity to Fume Components 37
75. Effects on the Musculoskeletal System 38
16. Effects on the Urogenital Tract 38
/ 7. Effects on the Teeth and Oral Cavity 39
18. Effects on the Voice 40
19. Biological Monitoring 4019.1 Aluminum 4019.2 Manganese 4019.3 Chromium 4019.4 Analytical Methods 41
Section Three — Investigations in Animals and Cell Cultures
20. Animal Studies 4120.1 Inflammation and Fibrosis 4120.2 Clearance 4220.3 Toxicity 4420.4 Allergic Sensitivity 4520.5 Biochemical Studies 4520.6 Barium 45
21. In Vitro Studies 4621.1 Bacterial Assays 4621.2 Mammalian Cell Studies 4621.3 Hyperbaric Pressure 48
References 49
List of Tables
Table Page No.
1 — Morphology and Chemistry of the Major Particle Populations in Welding Fumes 182 — Permissible Exposure for UV Light From Diverse Welding Processes 203 — Chromium Concentrations in Erythrocytes, Plasma and Urine of Welders Using SMAW or
GMAW, or both, of Stainless Steel 424 — Toxicity of Welding Fumes in Rats 43
List of Figures
Figure Page No.
1 — Ozone Concentration and Arcing Time During GMAW of Aluminum 172 — The A-weighted Sound Pressure Measured With a Stationary Microphone and Transient
Recorder 223 — Effect of Laundering on the UV Transmission Through Fabrics used in Work Clothing 234 — Alveolar Macrophage (AM) Counts in 9 Nonsmoking and 11 Smoking Welders at a Ship-
Building Yard 255 — Amounts of Iron, Chromium and Nickel in Rat Lungs after Intratracheal Instillation of
Activated Fumes Generated by SMAW and GMAW of Stainless Steel 446 — Effects of Fumes Generated by SMAW of Stainless Steel 47
IX
Introduction
The health of workers in the welding environment is a major concern of the American Welding Society. Tostay abreast of this subject, the health literature is periodically reviewed and published in the report Effects ofWelding on Health. Six volumes have been published to date; the first covered data published before 1978,while the latter five covered time periods between 1978 and December 1985. The current report includesinformation published between January 1986 and December 1987. It should be read in conjunction with theprevious volumes for a comprehensive treatment of the literature on the Effects of Welding on Health.
Included in this volume are studies of the characteristics of welding emissions that may have an impact oncontrol technologies necessary to protect the welder (Section 1). In keeping with previous volumes, the healthstudies are organized according to the affected organ system. The respiratory tract, the primary route ofexposure to welding emissions, is also a major target organ of many components of these emissions. Acute(e.g., metal fume fever, cadmium poisoning) as well as potential chronic respiratory effects (e.g.,emphysema,cancer) of welding emissions are of concern. However, chronic effects are not as well defined or understood,and the risk of cancer from these exposures has not been clearly established. Continued research in the formof epidemiologic studies, investigations with laboratory animals, and in vitro genotoxicity studies will help toresolve this question.
Executive Summary
The Respiratory Tract
Much of the recent research performed on thehealth effects of welding involved the respiratorytract. Of major concern was whether exposure towelding emissions induces conditions such as re-duction in pulmonary function, bronchitis, andemphysema. Essential to this question is whetherpulmonary impairments are associated with expo-sure to specific component(s) of welding emissionsand the identification of welding electrodes thatproduce less hazardous emissions.
Rounded opacities or shadows are frequently ob-served in chest X-rays of welders. These opacitiesrepresent deposits of fume particles in the respira-tory tract and are not usually associated with lossof pulmonary function. Larger and coalescentshadows reflect fibrotic processes and do not usu-ally result from welding exposures. X-ray shadowsoften regress after welding exposure ceases. A recentstudy by Desmeules and Tardif indicated that thecontinued progression of X-ray opacities after weld-ing exposures ceased is associated with exposureto lung irritants from sources other than welding(Ref. 33).
Some investigators have reported that a mild re-duction in pulmonary function accompanies theappearance of nodular shadows in X-rays. How-ever, there are reports to the contrary and other in-vestigators consider pneumoconiosis in welders tobe a benign condition. The literature shows manydisparities among the results of lung function testsin welders. During this report period, deficits inpulmonary function among welders were reportedin two studies, while two others found no correla-tion between welding exposures and abnormalitiesin pulmonary function. These disparities may resultfrom the use of tests that measure the function ofthose parts of the respiratory tract least likely to beaffected by welding fumes. Other disparities arisefrom studies of heterogeneous populations. The
quantities and chemical characteristics of gaseousand particulate components of welding emissionsvary widely with the welding method. Since the ef-fects on the respiratory tract can be expected to varyas well, it is important to correlate observed effectson the respiratory tract with the welding methodused.
Cancer
Thirteen new cancer epidemiology studies wereconducted which included welders among the studypopulations. In most of these, the inclusion of weld-ers was incidental and little information could beobtained about the duration of exposure, type ofwelding process used, or possible important non-welding exposures. Only one study specifically ex-amined a cohort of welders; no excess cancer of anytype was found among the 3000 welders examined(Ref. 25). Of the eight studies which evaluated lungcancer rates, four found a significant association be-tween welding and lung cancer. In three of these,exposures to carcinogens from nonwelding pro-cesses could not be ruled out (Refs. 90, 152, 158,161, and 162).
An elevated risk for leukemia was noted in a ship-yard study performed by F. B. Stern (Ref. 169). Incontrast, R. M. Stern found no association betweenwelding exposures and leukemia in his analysis of15 epidemiologic studies which considered the inci-dence of leukemia in welders (Ref. 170). Similarly,an elevated risk for brain tumors was found by Mc-Laughlin in a population-based study in Sweden(Ref. 97), but no brain tumors were identified in adeath- certificate study by Thomas in the U.S. (Ref.175), or in a case-control study conducted by Olinin Sweden (Ref. 123). Isolated studies reported ele-vated risks for cancer of the kidney (Refs. 158 and161), gastrointestinal tract (Ref. 141), nasal sinus(Ref. 52), and pancreas (Ref. 109).
A major argument supporting the probability ofan increased cancer risk in some welders is the pres-ence of nickel and hexavalent chromium in fumesfrom some welding processes. The mechanism ofaction of metal carcinogens remains unknown andis an important area for further research. Epidemio-logic studies differ widely in their conclusions con-cerning the cancer risk of welders. Most of the 13cancer epidemiology studies reviewed in this vol-ume of Effects of Welding on Health were of generalpopulations, making it difficult to rule out non-welding exposures. More carefully controlled stud-ies of welding populations are critically needed tohelp understand the risk of cancer in welders. Stud-ies of large populations of welders, in which expo-sures to specific welding processes can be carefullydelineated are necessary to provide a much neededanswer to the question of whether exposures to spe-cific fumes account for the elevated cancer risk ob-served among some welding populations.
Effects on the Urogenital Tract
With the exception of cadmium-induced kidneydamage, kidney disease is not generally associatedwith welding. This was confirmed by two epidemio-logic studies which found no relationship betweenwelding and the incidence of chronic kidney disease(Refs. 51 and 193). However, chronic exposure tolow concentrations of cadmium fumes can causekidney damage, which is first manifested by urinaryexcretion of low molecular weight serum proteins.Signs of kidney tubular dysfunction were foundamong workers exposed to cadmium in welding andsoldering fumes and aerosols from cadmium-con-taining paints (Refs. 164 and 184). A study of sol-derers with a work history of high cadmiumexposures and signs of kidney tubular dysfunctionsuggested that mild tubular proteinuria will notprogress to more severe kidney dysfunction in theabsence of further exposure to cadmium (Ref. 164).
Biological Monitoring
Because hexavalent chromium passes morereadily through cell membranes, trivalent chro-mium tends to be transported by plasma while hex-avalent chromium is transported mainly inerythrocytes (Ref. 3). Based on the possibility thatthis difference could be used to distinguish betweenexposures to trivalent and hexavalent chromium,
two research groups examined chromium levels inplasma and red blood cells of stainless steel welders.Angerer et al. (Ref. 4) concluded that simultaneousdetermination of chromium levels in plasma anderythrocytes may give a useful picture of overall ex-posure to both total chromium and hexavalentchromium. In contrast, Gustavsson and Welinder(Ref. 50) concluded that determination of blood orerythrocyte chromium levels would not add impor-tant information to that gained from biologicalmonitoring of urinary chromium concentrations.Further studies are needed to establish the utility ofchromium distribution between plasma and redblood cells in biological monitoring.
Stability of Welding Fumes
Recent work revealed that particles in some weld-ing fumes undergo chemical and morphologicalchanges in the lung. Physicochemical studies byStern et al. (Ref. 172) indicated that substitutedmagnetite is the primary magnetic component infreshly formed shielded metal arc welding (SMAW)fumes whereas a less substituted form of magnetiteappeared to be the predominant magnetic compo-nent of fumes that have been retained in the lungs.Similar findings were reported by Kalliomaki et al.(Ref. 76). They conducted an in vivo rat studywhich showed that the remanent magnetic fields offumes generated by SMAW of stainless steel and gasmetal arc welding (GMAW) of mild steel are not sta-ble in the lungs.
Other in vivo studies showed that there is amarked difference in pulmonary clearance rates offumes from SMAW and GMAW of stainless steel(Refs. 70, 71, and 73). Anttila demonstrated thatthese differences are related to differences in thesolubility of particles in the lung (Refs. 6 and 7). Themorphology and chemical composition of particleswere examined periodically in rat lungs after expo-sure to welding fumes by inhalation. The majorityof particles in SMAW fumes had irregular shapesand were enriched with sodium and potassium.These particles changed with time, losing practi-cally all sodium, potassium, and chromium, andalso some manganese. Eventually, so much materialwas lost that the particles became transparent. Incontrast, the particles from GMAW fumes showedno signs of dissolution. Thus, the solubility of parti-cle populations may be an important determinantin the pulmonary clearance of welding fumes.
Technical Summary
The Exposure
Fumes
A major objective of research performed on thecomposition of emissions from different electrodesand welding processes is to identify those methodswhich generate the least harmful pollutants. Suchinformation is also useful for selection of electrodesfor use in the workplace. Many studies of the com-position of fumes, and the rate of fume emissionfrom different welding processes, were published in1986 and 1987.
An industrial hygiene survey conducted in sixDutch factories showed that the fume and gas expo-sures from plasma welding of stainless steel werecomparable to the low exposures produced by gastungsten arc welding (GTAW). Exposures fromplasma cutting of stainless steel were comparable tothose produced by SMAW (Ref. 183). Henderson etal. showed that the fume generation rate (FGR) waslower for gas shielded flux cored electrodes than forthose used without gas shields (Ref. 58). With resis-tance welding, the quantity of emissions increasedmarkedly during "splashing" and also when weld-ing greased surfaces (Ref. 85). Studies of the effectsof hyperbaric pressure on fume emission demon-strated that the ratio of manganese to iron in-creases, and the FGR is 5 to 10 times greater whenperforming GTAW at a pressure of 32 bar than at 1bar (Ref. 16).
Mcllwain and Neumeier determined FGRs andfume constituents generated by selected coatedelectrodes commonly used in mines (Ref. 96). Olahand coworkers quantified gases (ozone, hydrogenfluoride, and nitrogen oxides) and light and heavyelements [Fe, Cr, Cr(VI), Ni, Mn, Mo, Si, C, andothers] in fumes produced by coated electrodes,flux cored hardfacing electrodes (Refs. 115, 116,
118, and 119), and by electrodes used for welding ofhigh alloy steels (Refs. 117 and 177). They foundthat exposures resulting from hardfacing are com-parable to those from SMAW of high alloy steels(Ref. 116). Tandon et al. determined 19 elements inthe flux and fume from three types of hardfacingand two types of high strength, low alloy steelcoated electrodes. An inverse relationship betweenthe fluoride concentration in the fume and the FGRwas found. There was a positive relationship be-tween sodium and potassium concentrations in theflux and the ratio of water- soluble hexavalent chro-mium to total chromium in the fume (Ref. 174).
The concentration of hexavalent chromium inGMAW fumes is substantially greater in fume sam-ples collected by wet impingement than in thosecollected by traditional filtration methods (Ref.137). A study by Hewitt and Madden of GMAWfumes collected by wet impingement suggested thatreactions between trivalent chromium oxides andozone may be responsible for the formation of hexa-valent chromium (Ref. 59). Wiseman and Chap-man (Ref. 190) found traces (0.1 ppb) of nickelcarbonyl in fumes produced by GTAW of nickel al-loys with shield gases containing carbon monoxideor carbon dioxide (CO2).
Several research groups described methods fordetermination of chromium and other elements inwelding fumes by X-ray fluorescence analysis (Ref.106, 120, and 186). Sawatari and Serita (Ref. 148)designed a method using a series of leaching solu-tions to differentiate between chromium of differ-ent valences and solubilities. Methods employinghigh pressure liquid chromatography (Ref. 94) andatomic absorption spectrometry (Ref. 21) to ana-lyze elements in welding fumes were described byother investigators. Three research groups pub-lished methods for measuring trace quantitiesof metals in biopsied lung tissue (Refs. 7, 10, 88,and 89).
Particles
Welding fumes are composed of microscopic par-ticles which tend to be of respirable size. The outersurface of particles from SMAW fumes are com-posed primarily of lighter elements derived fromthe flux (e.g., Na, K, Ca, F), while heavier elementsderived from the filler metal (e.g., Fe, Cr, Ni, andMn) tend to predominate in the particle core (Refs.48 and 187). Kalliomaki et al. (Refs. 47,49, and 72)classified particles in fumes from SMAW, GMAW,and flux cored arc welding (FCAW) of mild steelinto four groups on the basis of their morphology:clusters of globules, chains, granular networks, andlarge spheres. The distribution of the particleclasses varied with the welding process.
Ahlberg et al. determined the mass median aero-dynamic diameter (MMAD), the equivalent mobil-ity diameter (DME), and the growth in particle sizeas a function of relative humidity for particles infumes from SMAW of mild steel (Ref. 2). They con-cluded that hydrated particles could collapse tospheres after entering the respiratory tract whichwould cause the MMAD to increase and the DMEto decrease. This would result in increased sedi-mentation and diffusion rates, and would therebycause greater particle deposition in the alveoli.
Gases
Engstrom and Virtamo determined ozone levelsin the breathing zone of welders from 24 weldingshops using different gas shielded welding processes(Ref. 39). The mean concentrations for all weldingprocesses was less than the Finnish standard of 0.1ppm. GMAW of aluminum produced the highestozone concentrations with a geometric mean of0.06 ppm. The addition of nitric oxide to the shieldgas was less effective in reducing ozone levels thanpredicted by the laboratory studies of Sipek andSmars(Refs. 132 and 157).
Electromagnetic Radiation
Okuno estimated permissible exposure times at 1meter from the source for the ultraviolet (UV) andblue light generated by various welding processes(Refs. 113 and 114). The permissible exposuretimes for UV light varied from several seconds toone minute for most of the welding processes stud-ied. For each welding process, the intensity of theUV, but not the blue-light radiation, increased with
the diameter of the wire, arc current, and voltage.As expected, the UV hazard was greater withGMAW than with SMAW. The greatest blue-lighthazard was found for CO2-shielded GMAW of mildsteel, for which a permissible daily exposure time of42 seconds was estimated. In similar work, Salsiand Barlier measured the spectral emissions fromUV to far infrared light generated by 47 diversecoated electrodes (Ref. 9).
Production Coatings
Heikila et al. found that concentrations of benzo(a)pyrene as high as 0.64 micrograms/m3 were pres-ent in the vicinity of welders during SMAW of rail-road ties (Ref. 53).
Noise
Impulse noise may be more harmful to hearingthan continuous steady state noise of the same en-ergy level (Ref. 133). The impulse noise level hasbeen defined as the difference between the peakvalue and the root mean square of the slow time-weighted value. Noise is considered to be impulsiveif the difference between peak levels and the slowtime-weighted value is greater than 15 dB. A studyconducted in a shipyard assembly hall indicatedthat the impulsiveness for different processes in-creases in the following order: gouging < carbon arcgouging < chiseling < welding (process not speci-fied) < GMAW (Ref. 168). Further tests indicatedthat earmuffs effectively attenuate impulsiveness(Ref. 138).
Hygiene and Work Practices
Accidents, the primary health risk associatedwith the use of robots, occur most often duringteaching, testing, and maintenance. Accidents mayresult from pushing the wrong button in the robotcontrols, by improper programming of robot move-ment, or by difficulties in perceiving or anticipatingthe movement of the robot arm during work opera-tions (Refs. 56, 57, and 134).
A fatal accident occurred in which a welder wasworking inside a vapor degreaser pit filled with1,1,1-trichloroethane vapors. While welding, thevapors ignited and the welder fell into the pit, wherehe died. Tests of the 1,1,1-trichloroethane remain-ing after the accident indicated that it containedtwice the quantity of corrosion inhibitors normally
present, which may have contributed to the flam-mability of the vapors (Ref. 30).
Protective Gear. A study by Sliney et al. (Refs. 12and 163) indicated that most fabrics used in stan-dard work clothing provide adequate attenuation ofthe UV radiation generated by welding. Launderingdecreased the UV transmission through mostfabrics tested. This was thought to be due to shrink-age of the fabric, with a consequent decrease in thesize of the openings between fibers.
A study by Beckett et al. (Ref. 11) demonstratedthat wearing vapor-barrier garments in the work-place can cause heat stress. The authors noted thatwhen heavy muscular work is performed in a warmenvironment, impairment of the dissipation ofsweat by vapor-barrier garments can result in signif-icantly higher body temperatures with a high riskfor heat-related disorders, including heat stroke.
Jakubcik (Ref. 66) described a lightweight, main-tenance-free, disposable respirator used in a steelworks in Czechoslovakia to provide protectionwhen welding high alloy steels with electrodes con-taining chromium, nickel, and manganese. Severalmethods for extracting welding fumes from thework area (Refs 67 and 145) and from hyperbaricchambers (Ref. 35) were published.
Training. Rosen and Lundstrom coupled videotaping of workers and direct reading instrumen-tation of breathing zone exposures (Ref. 146). Theresultant video tape of work practices with a super-imposed presentation of real-time exposure datawas used to evaluate the relationship between expo-sures to fumes, dusts or solvents with individualwork patterns, and to train persons who work withhazardous substances.
Effects of Welding onHuman Health
Respiratory Tract
Alveolar Macrophages. Pierre et al. (Ref. 140) ex-amined the interactions of welding fumes with twolung antiproteinases. They found that fume parti-cles bind reversibly to the antiproteinases and pre-vent them from inhibiting the activity of proteolyticenzymes.
Mylius and Gullvag showed that sputum fromcoke plant workers, aluminum plant potroom work-ers, asphalt workers, and shipyard welders contain
more alveolar macrophages than controls. Com-bined smoking and occupational exposures pro-duced a marked increase in the number of macro-phages (Ref. 104).
Estimation of Retained Particles in the Lungs.Magnetopneumography utilizes the magnetic prop-erties of inhaled particles to estimate the burden ofoccupational dusts in the respiratory tract. Kal-liomaki et al. developed a magnetopneumographthat measures pulsed coercive force which can dis-criminate between remanent fields produced by dif-ferent types of fumes (Ref. 75). An innovationintroduced by Drenck and Stern (Ref. 36) used anAC susceptibility bridge to provide informationabout the total content of ferromagnetic, paramag-netic, and diamagnetic material. Using this system,Stern et al. (Ref. 172) obtained evidence for in vivochanges of welding fume particles from the substi-tuted magnetite present in freshly formed weldingfumes to unsubstituted magnetite. These findingswere supported by studies with X-ray diffractionand Mossbauer spectroscopy which indicated thatsubstituted magnetite is the primary magnetic com-ponent in freshly formed welding fumes, whereas aless substituted form of magnetite is the predomi-nant magnetic component in inorganic dust ob-tained from ashed lungs of deceased welders.
A study in which fumes were administered to ratsby intratracheal instillation also indicated that themagnetic properties of some welding particles arenot stable in the lungs (Ref. 76). The specific mag-netic moment of welding fumes generated bySMAW of mild steel and GMAW of stainless steelwere stable, while the magnetic moment of weldingfumes generated by SMAW of stainless steel andGMAW of mild steel decreased by about 50% dur-ing the 106-day period following the introductionof fumes into the lungs.
Pulmonary Function and Bronchitis. Chest X-rayabnormalities, and deficits in lung function, excesschronic bronchitis, or both, among welders wereexamined in 13 studies. Of these, two were con-founded by exposure to asbestos or other nonweld-ing exposures in the workplace (Refs. 34 and 122).
Nodular opacities are frequently observed inchest X-rays of welders. These opacities are not usu-ally associated with loss of lung function or diseasesof the respiratory tract. During the current reportperiod, two studies indicated that siderosis or pneu-moconiosis in welders is not associated with a re-duction of lung function (Refs. 26 and 166) whiletwo other studies (Refs. 84, 150, and 151) con-cluded that such an association exists. Chest X-ray
shadows often regress after welders retire or are re-moved from further exposure. Desmeules et al.found that workplace exposures other than weldingexisted in cases where opacities increased or pro-gressed after welding exposures ceased (Ref. 33).
While bronchitis and loss of pulmonary functionwere attributed to welding exposures by several in-vestigators, the reports differed in their conclusionsconcerning the contribution of tobacco smoke tothe development of respiratory impairments. Fivestudies found that nonsmoking welders have an ex-cess risk of bronchitis (Refs. 81, 82,84, 92, 102, and149) while two found an elevated incidence of bron-chitis only among welders who smoked (Refs. 26and 166).
Similar disparities exist among reports of the in-fluence of tobacco smoke on the induction of im-pairments in lung function. Two studies founddeficits in pulmonary function only in smokingwelders (Refs. 26 and 166) while three other studiesfound an elevated incidence of pulmonary deficitsin both smoking and nonsmoking welders. In asmall cohort of shipyard welders, Kalliomaki founddeficits in pulmonary function only among retiredshipyard welders and not among the younger, activewelders (Ref. 74).
Although there are conflicts among these reports,the results suggest that nonsmoking welders maydevelop bronchitis and deficits in pulmonary func-tion. Welders who smoke clearly have an elevatedrisk for both bronchitis and pulmonary deficits.Whether the combined effects of smoking and weld-ing exposures are additive or synergistic is not es-tablished. The study of shipyard welders byLyngenbo et al. (Ref. 92) carefully controlled for ex-posure to cigarette smoke, asbestos and exposure toother lung toxicants. This study demonstrated thatwelding alone can cause bronchitis, upper respira-tory tract symptoms, and reduced lung function.The effects of cigarette smoke on the developmentof these symptoms were additive.
The influence of constituents of welding fumeson the development of pulmonary impairments wasexamined by two research groups. Krasnyuk et al.found that welders of high alloy steel were morelikely to have extensive pneumoconiosis with moresevere restrictive disturbances in pulmonary func-tion than welders of low alloy steel (Ref. 84). Ulrichet al. found no differences in pulmonary functionbetween welders with high and low chromium expo-sures (Ref. 179).
Acute Respiratory Tract Disease — Case Report.The case of a welder who developed a sudden, work-related onset of pulmonary disease was described
by Silberschmid (Ref. 156). Examination revealedswollen and red bronchial mucosa, irregular opaci-ties in chest X-rays, and deficits in pulmonary func-tion. Extreme inflammatory changes were found inbiopsied lung tissue. Symptoms of exertional dys-pnea, bronchial hyperreactivity, and recurrentpneumonias persisted for 7 years. Prior to the onsetof disease, he had been working without properlyfunctioning local ventilation and was exposed tohigh concentrations of fume.
Production Coatings. Emmerling et al. (Ref. 38)found no significant abnormalities in pulmonaryfunction among 35 workers who welded metalsprimed with zinc chromate, polyvinylbutyral coat-ings or alkyl paints. In addition, there was nocorrelation between work place exposure and con-centrations of zinc, hippuric acid, methylhippuricacid, formic acid, or carboxyhemoglobin in bloodor urine samples. The number of participants in thestudy was small, and measurements of airbornechemicals indicated that exposures were low.
Cancer
Thirteen new cancer epidemiology studies wereconducted which included welders among the studypopulations. Eight of these were case-control stud-ies, while the remainder were population-basedmorbidity and mortality studies. Only one studyspecifically examined a cohort of welders. In thatstudy, no excess cancer of any type was foundamong the cohort of 3000 welders (Ref. 25). Of theeight studies which evaluated lung cancer rates,four found a significant association between weld-ing and lung cancer. In one of these, there was also ahigh incidence of mesothelioma and, thus, asbestosexposure could not be ruled out (Refs. 158,161, and162). A second study found an excess of lung canceramong shipyard welders, but not among welderswho did not work in shipyards (Ref. 152), whichsuggests the effects of exposures other than welding.The remaining two were case-control studies inwhich the total number of welders with lung cancerwere 28 (Ref. 83) and 19 (Ref. 90). In the latter,Lerchen cautioned that "the welders in our studyhad worked in diverse industries that may have in-volved exposures other than welding that also in-creased the lung cancer risk" (Ref. 90). Because ofthe possibilities of confounding exposures, it is dif-ficult to draw firm conclusions concerning the lungcancer risk of welders on the basis of studies de-scribed above.
Other tumors reported to be associated with
welding include leukemia, cancer of the stomach,brain, nasal sinus, and pancreas. An elevated riskfor leukemia was noted in a shipyard study per-formed by F. B. Stern (Ref. 169). In contrast, R. M.Stern found no association between welding expo-sures and leukemia in his analysis of 15 epidemio-logic studies which considered the incidence ofleukemia among welders (Ref. 170). Similarly, anelevated risk for brain tumors was found by Mc-Laughlin in a population-based study in Sweden(Ref. 97), but no brain tumors were identified in adeath-certificate study by Thomas in the U. S. (Ref.175), or in a case-control study conducted by Olinin Sweden (Ref. 123). One study found an elevatedrisk for renal tumors in welders (Ref. 161), whileisolated studies also reported elevated risks for can-cer of the gastrointestinal tract (Ref. 141), nasal si-nus, (Ref. 52) and kidney (Ref. 158). Norrellconducted a case control and morbidity and mortal-ity study of pancreatic cancer (Ref. 109). An ele-vated incidence of pancreatic cancer in welders wasobserved in the case-control but not in the mortalitystudy.
Screening for Cancer. A nontraumatic method forscreening workers at risk of nasal cancer was re-ported by Reith et al. This cytologic screeningmethod examines a representative population ofcells obtained by brushing the surface of the middlenasal turbinate. In a study of nickel refinery work-ers with known nasal dysplasia, more cases of pre-neoplastic nasal lesions were detected by cytologicscreening than by biopsy (Ref. 144).
Metal Fume Fever
Vogelmeier et al. (Ref. 185) observed changes inpulmonary function and marked pulmonary in-flammation in a welder who had recurring symp-toms of metal fume fever after welding zinc-coatedmaterials, but not other metals. Abnormalities inpulmonary function and a marked increase in poly-morphonuclear leukocytes in bronchoalveolar la-vage fluid were associated with symptoms of metalfume fever.
A case of metal fume fever with a severe, concom-itant allergic reaction to zinc fumes was reported byFarrell (Ref. 40). After welding galvanized metal,the patient experienced classic symptoms of metalfume fever associated with an immediate anaphy-lactoid reaction to zinc fumes as well as a delayedreaction consisting of angioedema and urticaria.
Effects on the Eye and Vision
Keratoconjunctivitis, a painful inflammation ofthe cornea, is the eye condition most frequently en-countered by welders. Although this condition usu-ally lasts for up to 2 days, Sharir and Ben-Davidobserved severe symptoms of keratoconjunctivitiswhich persisted for several months in four workersemployed in plasma arc thermal spray coating(Ref. 153).
Two cases of photic maculopathy, in which visualloss was noted soon after exposure to the weldingarc, were reported (Refs. 20 and 180). In one of thecases, some visual loss and signs of retinal injurywere present 6 months after the incident (Ref. 180).
An unusual case of retinal damage in a man with alens implant was reported by Turut et al. (Ref. 178).Four years after the surgery, he welded withoutwearing eye protection and experienced an immedi-ate loss of visual acuity. A small lesion, apparentlycaused by radiation from the arc, was present on theretinal fovea of the eye with the lens implant. Therewere no indications of damage or loss of vision inthe eye with the normal lens.
Effects on the Skin
A case of a welder with recurrent, severe facialdermatitis was described by Shehade et al. (Ref.154). Allergic dermatitis was ruled out by skin patchtesting with a battery of allergens. The welder hadroutinely performed spot welding without a pro-tective face mask. His case was diagnosed asphotodermatitis, caused by UV exposure duringwelding.
Sensitivity to Fume Components
A case of a welder who had anaphylactoid reac-tions following exposure to chromates was reportedby Moller et al. (Ref. 101). The symptoms includedurticaria, difficulty breathing and chest tightness.Nordman et al. examined 58 stainless steel welderswho had asthmatic reactions when welding stain-less steel. Some of these welders responded posi-tively to the histamine provocation test prior towelding and had significant reductions in pulmo-nary function following welding of stainless steel(Ref. 108).
10
Effects on the MusculoskeletalSystem
Two techniques for analyzing causes of posturalstress were described. One was a computer-aidedsystem used to analyze working postures of thetrunk and shoulders on repetitive assembly linejobs (Ref. 78). Both systems were applied to im-proving postural stress incurred during welding op-erations (Refs. 55 and 78).
Effects on the Urogenital Tract
Chronic exposure to low concentrations of cad-mium fumes can cause kidney damage, which isfirst manifested by urinary excretion of low molec-ular weight serum proteins such as beta-2-microglobulin and retinol-binding protein. Signs oftubular dysfunction, but not glomerular dysfunc-tion, were found among a group of 26 cadmium-exposed workers by Verschoor et al. (Ref. 184).Smith et al. found low molecular weight proteins(typical of tubular dysfunction), but not albumin(characteristic of glomerular dysfunction) in theurine from 20 of 51 solderers with up to 25 years ofuncontrolled exposure to fumes from cadmium-containing alloys. No significant changes in beta-2-microglobulin were found in 19 solderers followedfor 2 years after cadmium exposure ceased, sug-gesting that mild tubular proteinuria will not prog-ress to more severe kidney dysfunction in theabsence of further exposure to cadmium (Ref. 164).
Two epidemiologic studies showed no associa-tion between renal impairment and welding. In thefirst, elevated levels of nickel and chromium werefound in the urine of stainless steel welders, but kid-ney function, assessed by measurement of total pro-tein and enzyme activity in urine, was comparableto that of the control group (Ref 193). The secondstudy examined kidney function in 143 males whohad received kidney biopsies as a result of suspectedor clinically established glomerulonephritis or ob-scure renal disease. There was no relationship be-tween welding and the incidence of severe chronickidney disease nor was there an increase in the riskof death from nephritis among welders (Ref. 51).
Effects on the Teeth
To investigate the cause of a metallic taste sensa-tion and loosening dental amalgams among under-water welders, Ortendal et al. investigated the
intraoral electrical and magnetic activity during un-derwater welding. Humidity was the predominantfactor in the generation of intraoral electrical activ-ity. However, even when water was deliberately in-troduced into the diver's suit, the intra- oral electricfield was too small to damage dental amalgam. Nocorrelation was found between the magnetic fieldand a metallic taste in the mouth. In vitro studiesindicated that a slight degeneration of dental amal-gams may be caused by magnetic fields equivalentto those generated by welding. However, chemicalchanges were insufficient to account for a metallictaste sensation.
Biological Monitoring
Sjogren et al. (Refs. 159 and 160) examined therelationship between urinary aluminum concentra-tions and exposure to fumes generated by GMAWand GTAW of aluminum. Urinary aluminum levelswere dependent on the most recent daily exposuresas well as on the total duration of exposure. The re-sults indicated that aluminum is probably stored intwo or more compartments in the body, with widelydifferent clearance rates.
Two studies were based on findings that red bloodcells are permeable to hexavalent, but not trivalent,chromium (Refs. 4 and 50). Angerer et al. (Ref. 4)determined chromium concentrations in thebreathing zone and in erythrocytes, plasma andurine from 103 male welders who performedSMAW or GMAW, or both, of stainless steel. Chro-mium levels in erythrocytes were too low to provideaccurate estimates of exposure below airborne con-centrations of 100 micrograms/m3. However, theinvestigators concluded that simultaneous determi-nation of chromium levels in plasma and erythro-cytes may give a useful picture of exposure to totaland hexavalent chromium.
Different conclusions were drawn by Gustavssonand Welinder (Ref. 50) who observed no obviousdifferences in clearance rates of erythrocyte andplasma chromium in studies performed with fourstainless steel SMA welders and six electroplaters.These investigators concluded that determinationof blood or erythrocyte chromium levels would notadd important information to that gained from bio-logical monitoring with urinary chromium concen-trations.
Zschiesche et al. (Ref. 194) determined levels ofmanganese in the breathing zone and body fluids ofeight welders from two factories who worked withfiller materials containing about 0.8% manganese.Levels of manganese in blood and urine tended to
11
be low and, for the most part, did not vary substan-tially from values obtained for the general popula-tion. There were no correlations between externalexposures and internal manganese levels, nor withthe time of day or the day of the week that the bio-logical samples were collected.
Effects on the Voice
A study by Ohlsson indicated that voice andthroat problems were much more frequent amongwelders than office workers. The investigators con-cluded that vocal problems among welders are dueto high noise levels which force a strain on the lar-ynx during vocalization, whereas welders believedtheir problems resulted from fume exposure (Ref.112).
Toxicologic Investigations inAnimals and Cell Cultures
Animal Studies
Inflammation and Fibrosis. The development ofpractical methods for assessing the fibrogenicity offumes and gases from welding operations is impor-tant for the determination of risk associated withdifferent electrodes and welding methods. Hicks etal. (Ref. 62) compared the irritant and fibrogenicproperties of fumes from different electrodes by in-jecting fume samples into rat and guinea pig muscletissue. The inflammation produced by weldingfumes decreased in the following order: SMAW ofstainless steel >SMAW of mild steel >GMAW ofstainless steel.
Studies of the effects of fumes on the respiratorytract were conducted by the American Welding So-ciety (Ref. 23). Rats were exposed by inhalation for6 hours to fumes from six welding processes and thehistopathologic effects on the respiratory tract weredetermined. Other assays examined viability andphagocytic activity of alveolar macrophages fromexposed animals. Extracts from urine were not ge-notoxic in the Salmonella/Ames test nor were anychromosome aberrations seen in bone marrow cellsrecovered from animals 20 hours after exposure. Onthe basis of these studies, the electrodes were classi-fied into three categories: highly toxic (E308-16),moderately toxic (E7018, E6010, and E70T-1) andnontoxic (E70S-3 and E5356).
Wilmer et al. conducted 14-day inhalation stud-
ies of fumes from ten electrodes in Wistar rats (Ref.188). Fumes from SMAW of stainless steel were themost toxic, causing severe dyspnea, some mortality,and reduction in body weight gain. GMAW fumeswere less toxic and induced only slight signs of re-spiratory distress. Fumes from SMAW of mild steelwere only slightly toxic. The fumes from most elec-trodes were cytotoxic to alveolar macrophages.None of the fumes caused sister chromatid ex-changes (SCE) or structural chromosome aberra-tions in bone marrow cells.
Clearance. In vivo studies conducted by Kal-liomaki et al. demonstrated marked differences inpulmonary clearance rates of fumes from SMAWand GMAW of stainless steel (Refs. 70, 71, and 73).Anttila showed that this was due to differences inthe solubility of particles in the lung (Refs. 5 and 7).The morphology and chemical composition of par-ticles were examined periodically after exposure ofrats to welding fumes by inhalation. The majority ofparticles in SMAW fumes had irregular shapes andwere enriched with sodium and potassium. Theseparticles changed with time, losing practically allsodium, potassium, chromium, and some manga-nese. In contrast, the particles in GMAW fumesshowed no signs of dissolution.
Another study focused on clearance of manga-nese from the respiratory tract (Ref. 77). The resultssuggested that manganese was cleared from thelungs as intact particles and that little manganeseclearance resulted from in situ dissolution ofparticles.
Toxic Effects. A long-term inhalation study of theeffects of nickel oxide and arsenic trioxide in Wistarrats was conducted by Glaser et al. (Ref. 42). Arse-nate exposure had no apparent effects. Rats treatedwith nickel oxide had an increased mortality rateand severe alveolar proteinosis. No treatment-re-lated tumors were seen.
Gorban et al. examined the effects of exposure tofumes generated by CO2-shielded GMAW of me-dium and high alloy steel with two types of austen-itic electrodes (Ref. 45). The LD50s did not differsignificantly for fumes from the two electrodes, butdifferences were noted in the effects on levels of nu-cleic acid and alanine and aspartate aminotransfer-ase in the liver and lung.
Geleskul et al. (Ref. 41) used the formation ofmalondialdehyde in the presence of ascorbic acid asa measure of lipid peroxidation produced by weld-ing fumes. The investigators concluded that lipidperoxidation can be used for the assessment of thetoxicity of welding fumes.
12
Very low concentrations of hexavalent chromiumcan activate the enzyme beef heart phosphodiester-ase, which is regulated by calmodulin, a calcium de-pendent protein. MacNeil et al. suggested that theeffect of chromium on this regulatory protein maybe related to its toxic effects (Refs. 29 and 93).
Allergic Sensitivity. Caldas and Hicks reportedthat pulmonary exposure to chromate or extracts ofchromium and nickel-rich fumes from SMAW ofstainless steel provokes a specific tolerance to theimmunologic effects of chromate in the skin (Refs.18 and 60).
In Vitro Studies
Bacterial Assays. Biggart et al. reported that the gasphase and particulates of emissions from SMAW ofmild steel are mutagenic in a modified Salmo-nella/Ames assay. The fumes contained both direct-and indirect-acting mutagens which inducedframeshift mutations. The results suggested thatfume particles become degraded at the cell surface,permitting particle components to be taken up bythe bacteria (Ref. 14). Unlike the fume particles, ex-posure to the gas phase caused base pair substitu-tions but not frameshift mutations. (Ref. 13).
Ong et al. demonstrated that the SOS test, whichdetects gene damage in Salmonella typhimurium bya colorimetric reaction, can be adapted for use inoccupational settings (Refs. 124 and 125). Stainlesssteel welding fumes induced a high response withthe SOS test, and the effects were greater withoutmicrosomal enzymes than in their presence. TheSOS test was much more sensitive to welding fumesthan the Ames test. This may be related to the ob-servations of Pedersen et al. (Ref. 137) of higherconcentrations of hexavalent chromium in fumescollected by impingement than in those collected byfiltration.
Mammalian Cell Studies. Two studies examinedthe induction of SCE by welding and cutting aero-sols. Baker et al. showed that SCE activity, but notmitotic delay, was proportional to the hexavalentchromium content of the welding fume in culturedChinese hamster lung cells (Ref. 8). The solublefume fraction was much more active than the insol-uble fraction in both the mitotic delay and SCEassays, presumably because of the greater bio-availability of the soluble components of the fumeparticles.
Valerio et al. showed that fumes emitted by gas-cutting during refitting operations in oil tanks cause
SCE in murine bone marrow cells (Ref. 182). Thesefumes are known to contain high levels of benzo(a)pyrene and other polycyclic aromatic hydrocar-bons which may be responsible, in part, for theirmutagenic activity.
Hooftman et al. reported that the cytotoxicity ofwelding fume particles to bovine alveolar macro-phages decreases in the following order: SMAW ofstainless steel >GMAW of stainless steel >SMAWof mild steel >GMAW of mild steel (Ref. 64).Pasanen et al. (Ref. 135) found that the cytotoxicityof welding fumes to rat alveolar macrophages de-creases in the following order: SMAW of stainlesssteel = SMAW of mild steel >GMAW of stainlesssteel >GMAW of mild steel. Similar to the findingsof Hooftman et al., Pasanen showed that the viabil-ity of cells treated with fumes from GMAW of mildsteel differed little from control cells. Removing wa-ter-soluble materials by washing with phosphate-buffered saline substantially reduced the cytotoxic-ity of fumes from SMAW of stainless steel but hadlittle effect on the other fumes. The effect of washedand unwashed fumes on the release of lactic dehy-drogenase followed the same pattern (Ref 135). Theresults of both Hooftman and Pasanen indicatedthat hexavalent chromium may be responsible forthe cytotoxic activity of welding fumes to alveolarmacrophages.
In vitro studies indicated that combined expo-sures to chromate and hyperbaric pressures can actsynergistically to disrupt the organization of actinfilaments, which are responsible for cell shape andmotility and may prolong the Gl phase of the cellcycle (Refs. 68 and 173).
Conclusions
Welding emits fumes and gases which may causeadverse health effects. Emissions vary widely withthe process and can be controlled to some degree bychoice of method, electrode, and filler material. Theuse of proper ventilation, protective gear, andclothing can keep exposures to a minimum. Train-ing programs may be needed for management andwelders alike, demonstrating the economic andhealth advantages of reducing exposures. Newtraining techniques, employing video tapes andcomputer systems, have been developed which mayhelp the welder to a better understanding of expo-sures and to adjust work practices accordingly.
Unresolved questions concerning health issuespersist. The most important of these concern the ef-fects of welding on the respiratory tract and cancerrates among welders.
13
The Respiratory Tract. Exposure to some compo-nents of welding emissions, including ozone, nitro-gen oxides, chromium, and nickel, may injure therespiratory tract. However, the relationship be-tween welding exposures and deficits in pulmonaryfunction, or respiratory diseases such as bronchitisand emphysema, is not well understood. The gener-ation of injurious pollutants varies with the weldingmethod and the metal welded. The influence of thespecific chemical constituents of welding fumes onthe development of pulmonary impairment is animportant area that has not been adequately inves-tigated.
Inconsistencies in reports of the chronic effects ofwelding fumes on the lungs may be due, in part, tostudies of populations exposed to different weldingmaterials. Therefore, studies of the incidence of re-spiratory diseases in welders should be carefully de-signed and should concern populations selected in amanner that will allow correlation of results withexposures to specific welding processes. In addi-tion, a variety of pulmonary function tests are avail-able. Questions have been raised as to whethersome widely used tests are likely to detect the typesof lesions that could potentially be produced bywelding fumes. Thus, the applicability of differentpulmonary function tests for research purposes orfor general medical screening of welders should befurther examined.
Further development of practical techniques forassessing the fibrogenicity of fumes and gases fromwelding operations is important for the assessmentof risk associated with different electrodes andwelding methods. In vivo and in vitro methods forstudying the potential of occupational pollutants tocause irritation, inflammation, or fibrosis in thelung and other tissues should be explored. A com-prehensive literature search to determine whethermethods exist that are useful for studying weldingemissions might prove fruitful.
Several research groups reported that exposure tosome welding fumes can affect the function and via-bility of alveolar macrophages, which are importantfor maintaining the health of the lungs. However,
since their activity may stimulate inflammatoryand fibrotic processes, alveolar macrophages mayalso contribute to the development of chronic lungdisease. Mylius and Gulvag showed that cigarettesmoke and welding exposure combine to produce amarked increase in the number of alveolar macro-phages in the lung (Ref. 104). Vogelmeier et al.observed a tenfold increase in the number of leuko-cytes obtained by bronchoalveolar lavage in a pa-tient with recurrent metal fume fever (Ref. 185).Recognizing the potential importance of macro-phage activity in the development of chronic lungdisease, both of these investigators questionedwhether such impressive numbers of inflammatorycells can be present in the lungs without causingdamage. It may be of interest to focus more atten-tion in the future on the effects of welding fumes onalveolar macrophages and to determine what theseeffects mean in terms of potential health effects.
Cancer. Fumes generated by welding stainless steelcontain potentially carcinogenic chemicals, in par-ticular hexavalent chromium and nickel com-pounds. The presence of these metals in fumes fromsome welding processes provides a major argumentin support of the probability of an increased cancerrisk in some welders. The mechanism of action ofmetal carcinogens remains unknown and is an im-portant area for continued research. Epidemiologicstudies differ widely in their conclusions concern-ing the cancer risk of welders. In many of the cancerepidemiology studies reviewed in this volume of Ef-fects of Welding on Health, it is difficult to rule outnonwelding exposures among the welding popula-tions. More carefully controlled studies of weldingpopulations are critically needed to help under-stand the cancer risk of welders. Studies of largepopulations of welders, in which exposures to spe-cific welding processes can be accurately deter-mined, are necessary to provide a much neededanswer to the question of whether exposures to spe-cific fumes account for the elevated cancer risk ob-served among some populations of welders.
Effects of Welding on Health VII
Section OneThe Exposure
Hazardous fumes, gases, and electromagnetic ra-diation can be generated during welding. Organicvapors may be present in welding emissions if thebase metal is coated with organic products such aspaints or oils. The composition of the emissionsvaries with the welding method. Of the most com-mon welding processes, fume generation is lowestwith gas tungsten arc welding (GTAW) and highestwith shielded metal arc welding (SMAW) and flux-cored arc welding (FCAW). Ozone levels are great-est with gas metal arc welding (GMAW) of alumi-num. High levels of nitrogen oxides (NOX) areproduced by all processes that involve intense heat.Gas welding and plasma cutting tend to producemore NOX than other welding methods.
An industrial hygiene survey conducted in sixDutch factories showed that fume and gas expo-sures from plasma welding of stainless steel arecomparable to the low exposures produced byGTAW, while exposures from plasma cutting ofstainless steel are comparable to those produced bySMAW (Ref. 183). The Dutch occupational healthstandard for welding fumes (5 mg/m3) was exceededwhen plasma cutting of stainless steel was con-ducted without local exhaust ventilation; fume con-centrations were below the standard when localexhaust was used and also during cutting under wa-ter. Hexavalent chromium was found in only twoplants and its concentration never exceeded 40ug/m3. Total chromium concentrations did not ex-ceed 400 ug/m3. In no instances were the Dutchstandards for nitrogen dioxide (NO2: 5 ppm), nitricoxide (NO: 25 ppm) or nickel (1 mg/m3) exceeded,even during the process of cutting alloys containing30 to 35% nickel. None of the samples containedwater-soluble nickel. Breathing zone levels of ozonewere excessive because of the proximity of thewelder's face to the torch. This problem was re-
solved by placing a glass shield on the work tablebetween the workpiece and the welder.
1. FumesWelding fumes are composed of microscopic
metal oxide particles which originate primarilyfrom the filler metal and the electrode coat or corematerials. The concentrations of fume constituentsvary with the welding process, electrode, basemetal, and welding parameters (voltage and cur-rent). The fume generation rate (FGR), a measureof the quantity of fume emitted per unit time, isused to compare fume emissions from differentelectrodes. The importance of this value for assess-ing relative hazards associated with different elec-trodes was stressed by Moreton who urged thatstandard methods for measurement of FGR beadopted (Ref. 103).
FGRs have been determined for a large numberof electrodes used in different welding processes.Recently, Henderson et al. showed that the FGRwas lower for gas shielded flux cored electrodesthan for those used without gas shields (Ref. 58).Krause and Press reported that the quantity ofemissions generated during resistance welding in-creased markedly during "splashing". Resistancewelding of greased metals also increased fume for-mation. In the absence of splashing, a layer of grease(1.3 mg/cm2 metal surface) increased fume forma-tion by 30%. The emissions increased with thethickness of the grease (Ref. 85).
The effects of hyperbaric pressure on weldingfume emissions is receiving attention in NorthernEurope because of the importance of offshore oiloperations in that area. In this regard, Bjoerseth etal. showed that for GTAW, fume production is 5 to10 times greater at a pressure of 32 bar than at 1 bar.The ratio of manganese to iron was also greater athigher pressures, presumably due to heat distribu-tion effects (Ref. 16).
15
16
1.1 Effects of Electrode Composition. In the inter-est of identifying electrodes that generate the leasthazardous emissions, Olah and coworkers mea-sured gases (ozone, hydrogen fluoride, NOX) andlight and heavy elements [Fe, Cr, Cr (VI), Ni, Mn,Mo, Si, C, and others] in fumes produced by coatedelectrodes, flux cored hardfacing electrodes (Refs.115, 116, 118, and 119), and by electrodes used forwelding high alloy steels (Refs. 117 and 177). Analy-ses were conducted by neutron activation analysis,X-ray fluorescence spectroscopy, and atomic ab-sorption spectrometry (AAS). The results showedthat the exposures resulting from hardfacing arecomparable to those from SMAW of high alloysteels (Ref. 116).
Tandon et al. determined 19 elements in the fluxand fumes from three types of hardfacing electrodesand two types of coated, high-strength, low alloysteel electrodes. Magnetite (Fe3O4), potassium chro-mate (K^CrO-t), calcium fluoride (CaF2) and sodiumfluoride (NaF) were the only compounds detectedin the fumes. The water solubility of the fumes var-ied from 22.5% to 3,7.5% for the five electrodes. Aninverse relationship between the fluoride concen-tration in fumes and the FGR was noted. The con-tent of sodium and potassium in the flux was highlycorrelated with the ratio of water-soluble hexava-lent chromium to total chromium in the fume.(Ref. 174).
In order to establish exposure guidelines, Mcll-wain and Neumeier of the Bureau of Mines deter-mined FGRs and fume constituents generated byselected welding electrodes commonly used inmines. From these data, exposure indices were de-termined which provide a relative measure of thepotential health hazards associated with emissionsfrom the electrodes (Ref. 96).
1.2 Chromium. In 1983, Thomsen and Stern re-ported finding a short-lived hexavalent chromiumspecies in freshly formed welding fumes collectedby impingement in an aqueous medium (Ref. 133).GMAW fumes collected by traditional methods ondry membrane filters generally contain little hexa-valent chromium, whereas Thomsen (Refs. 137 and176) reported that GMAW fumes collected by liq-uid impingement contain 6 to 7% hexavalent chro-mium. The same investigators examined the effectsof sample collection method, storage time, and ana-lytical procedure on the concentration of hexava-lent chromium in welding fumes (Ref. 137).Hexavalent chromium was reduced to the trivalentstate when SMAW fume samples were stored for upto 15 days on cellulose acetate, but not on polyvinylchloride (PVC) membrane filters, and reduction of
hexavalent chromium was observed with some ana-lytical procedures. Pedersen et al. suggested thatchromium measurements made on samples col-lected by wet impingement may reflect actual .expo-sures better than those made on samples collectedon dry membrane filters. However, other investiga-tors have argued that the short-lived chromium spe-cies may be an artifact resulting from chemicalreactions in the wet collection medium (Ref. 133)and that much more information concerning thegeneration of hexavalent chromium is needed be-fore traditional methods of fume collection can bereplaced, or supplemented, by wet impingement(Ref. 46).
Hewitt and Madden (Ref. 59) used the wet im-pingement collection method described above (Ref.176) during a study of the effects of various weldingparameters on hexavalent chromium concentra-tions in fumes generated by GMAW of stainlesssteel. Hexavalent chromium levels were greatestwith a shield gas of pure argon and decreased withincreasing concentrations of carbon dioxide in theshield gas. Their finding that ozone increased theconcentration of hexavalent chromium led to theconclusion that the formation of hexavalent chro-mium in GMAW fumes is due, in part, to reactionsbetween trivalent chromium oxide and ozone. Inhis critique of this work, Gray argued that since theincrease in hexavalent chromium was not propor-tional to the ozone concentration, ozone reactionscannot account for hexavalent chromium forma-tion (Ref. 46).
1.3 Nickel. Nickel carbonyl is a severe pulmonaryirritant and has caused cancer in animal studies.Wiseman and Chapman (Ref. 190) determinednickel carbonyl levels in emissions produced byGMAW and GTAW of Nickel 200, stainless steel,and other alloys (Inconel 600, Incoloy 800, andMonel 400). Traces (0.1 ppb) of nickel carbonylwere found in emissions from 3 of the 16 weldingcombinations tested, all of which involved GTAWof Nickel 200 with shield gases containing carbonmonoxide (CO) or carbon dioxide (CO2).
1.4 Analytical Methods. The utility of X-ray fluo-rescence analysis for determination of componentsin welding fumes was described by Nemcova (Ref.106). Voitkevich and Fedorina compared differentmethods for determining hexavalent chromiumconcentrations in stainless steel welding fumes(Ref. 186). Olah and Pivoluska described a rapidmethod using X-ray fluorescence spectrometry toquantify chromium in fumes from welding of stain-less steel. The limit of detection was 3 ug. The corre-
17
lation between results obtained with X-rayfluorescence spectrometry and the diphenyl carba-zide method was poor (Ref. 120).
A method employing high pressure liquid chro-matography to separate beryllium, cobalt, nickel,and chromium was described by Maiti and Desai(Ref. 94). The elements were separated as beta-isopropyltropolone complexes on a reversed-phasecolumn. The method was used to determine chro-mium in air samples from welding operations. Theprocedure did not distinguish between trivalentand hexavalent chromium.
Sawatari and Serita (Ref. 148) examined chro-mium species in metal fumes generated from purechromium powders using a plasma metal sprayer. Aseries of leaching solutions was used to differentiatebetween chromium of different valences andsolubilities. The procedure was rather time-con-suming, but the authors indicated that it could besimplified to allow determination of the solubilityand valence of chromium with only three steps(leaching with sodium carbonate, digestion of un-dissolved fumes with heated dilute sulfuric acid,and fusion of the undissolved portion with a boilingmixture of sulfuric acid and perchloric acid).
Chong described a method using AAS to deter-mine concentrations of Ag, Bi, Cd, Cu, Fe, Ni, Pb,and Sn in lead- and zinc-based solders and metalalloys. The technique prevented loss of metals byco-precipitation and allowed all of these elements tobe analyzed from a single sample (Ref. 21).
Ingle developed a device for automatically re-cording the time and duration of arcing periodsduring electric welding (Ref. 65). A reed relay wasused as a detector of the circumferential magneticfield around the cables carrying the welding cur-rent. The relay was placed next to the cable in thedirection of the field; it opened and closed as thestrength of the field rose and fell. Ingle reported thatthe device could be used to record arcing patterns ofGMAW, GTAW, and SMAW, with both direct andalternating current. The pattern obtained for thevariation of ozone concentrations with arcing timeduring GMAW of aluminum is shown in Figure 1.
1.5 Analysis of Metals in Biological Tissue. Ant iliaet al. found that transmission electron microscopyand X-ray microanalysis could successfully identifysome mineral and metal particles in autopsied lungtissue. However, occupational histories or addi-tional chemical analyses were necessary for accu-rate identification of particles from metal grindingor welding (Ref. 6). Landsberger and Simsonsshowed that epithermal neutron activation analysiscould be used for the nondestructive determination
a.a.
O
I0.04-
0 .03-
LU
2 0.02-
0.01 -
0.00-1
ARCING-*.
NOT •*>ARCING
i i r4 5 6
i r7 8
TIME, MINUTES
Figure 1 — Ozone Concentration andArcing Time During GMAW of Aluminum
Ingle, Ref. 65
of chromium, nickel, and zinc in biological tissuesand for the determination of nickel and arsenic inurine. These techniques were used to analyze metal-lic deposits in lung tissue from a welder who diedfrom cancer (Refs. 88 and 89). Baumgardt et al. de-scribed a method using AAS for measuring tracequantities of metals in biopsied lung tissue. Ele-vated levels of chromium, cobalt, and manganesewere found in lung tissue from a welder (Ref. 10).
1.6 Particles. Microscopic particles comprise thebulk of welding fumes. The particles tend to be ofrespirable size (0.1 to 10 um in diameter) and aregenerally spherical, although the regularity of thesurface may vary with the welding process and elec-trode. Particles may be present in fumes as singleentities or as agglomerates or chains of varyinglengths. The particles are chemically heteroge-neous. Their composition is determined primarilyby the content of the filler metal but, for SMAW,elements from the flux also contribute to the fumes.
The outer surface of particles from SMAW tendsto be composed of lighter elements (e.g., Na, K, Ca,F) while heavier elements (e.g., Fe and Mn) arefound in the particle core (Refs. 48 and 187). Thesurface and bulk elemental composition of particlesfrom SMAW of stainless steel were determined byGrekula et al. (Ref. 48) using scanning electron mi-croscopy (SEM) in conjunction with energy-disper-sive X-ray analysis (EDXA). The surface andsubsurface chemical composition (as revealed byargon-ion sputtering) differed for large and smallparticles. The results suggested that formation of
18
WeldMethod
GMAW/MS(CO2)
GMAW/SS(Ar/O2)
SMAW/MS
SMAW/SS
FCAW/MS(Ar/O2)
FCAW/MS(withoutgas)
Morphology and Chemistry of the
ParticlePopulation
Granular networkClusters of globChains
Granular networkClusters of glob
Granular networkClusters of globBlobs
Granular networkClusters of globBlobs
Granular networkClusters of globChains
Granular networkClusters of globChains
PopulationDiam. (um)*
31
10
up to 101
up to 101
Similar toSMAW/MSfume particles
415
10.55
Table 1Major Particle
ParticleSize (um)
0.020.10.05
0.020.1
0.030.20.5
0.020.10.03
0.020.20.03
Fe
747082
4140
8622
9
4891
754976
**1186
Populations in
Si
201310
1810
3102
523
2
201114
44
Mn
6178
1024
173
151
53010
32
Cr
2322
1152
Ni
84
111
Welding
Na
37
71829
6
K
102771
173565
2
Fumes
Ca Mg Ba
307
1 26 483 5
Other
1
4
4
5
SS - Stainless steelMS - Mild steel* - Size of particle cluster** - "Composition varies greatly; the main component can be either Fe, Mn, Al, Mg, or Ni"Data from Kalliomaki et al., Ref. 72
particles is partially dependent on vapor pressures.Metals with lower vapor pressures (e.g., Fe, Cr, andNi) are enriched in the core, whereas those withhigher vapor pressures (Na, K, F) condense on theparticle surface.
In other studies, Grekula et al. examined thephysical characteristics of particle agglomerates infumes from SMAW, GMAW, and FCAW of mildsteel (Refs. 47, 49, and 72). Particles were groupedinto four classes on the basis of their morphology:clusters of globules, chains, granular networks, andlarge spheres (Table 1). The finest metal-rich parti-cles formed granular networks which representedmost of the particles in Cd-shielded GMAW fumesand a minor fraction of the particle agglomerates inSMAW fumes. Particle agglomerates produced byGMAW and gas shielded FCAW of mild steel weresimilar in morphology. Clusters of globules pre-dominated in fumes from SMAW and FCAW (with-out gas), whereas particles in GMAW and gasshielded FCAW fumes were present primarily as
granular networks. Individual semitransparent par-ticles, referred to as blobs, were unique to SMAWfumes. They were composed primarily of potassiumand sodium and apparently arose from the slag. Themorphology of particles from SMAW was the samefor rutile and for basic electrodes. Most of the ag-glomerates had a characteristic chemical composi-tion; however,"the composition of individualparticles within the agglomerates varied.
The size of particles in welding fumes is generallyexpressed as the mass median aerodynamic diame-ter (MMAD), which is used to estimate the likeli-hood of their deposition in the lungs upon inhala-tion. According to Ahlberg (Ref. 2), the MMAD isnot completely valid for welding fume particlessince they tend to be agglomerated into chains andother nonspherical masses. The equivalent mobilitydiameter (DME) better accounts for the sedimenta-tion and diffusion forces to which welding agglom-erates are subjected. In addition, some weldingfume particles are hygroscopic and may become hy-
19
drated in the respiratory tract. This hydration couldtheoretically alter their deposition rates in thelungs. Ahlberg et al. determined the MMAD, theDME, and the growth in particle size as a functionof relative humidity for particles in fumes fromSMAW of mild steel. They concluded that hydratedparticles could collapse to spheres after they enterthe respiratory tract which would cause the MMADto increase and the DME to decrease. This wouldresult in increased sedimentation and diffusionrates causing greater particle deposition in the alve-oli. The investigators concluded that the actual lungdeposition is less than the deposition predictedfrom the MMAD for nonhygroscopic weldingfumes and more than that predicted for hygroscopicwelding fumes (Ref. 2).
2. Ozone
Ozone, a severe respiratory irritant, is generatedby reaction of ultraviolet (UV) light with oxygen.Ozone is unstable in air and its decomposition isaccelerated by metal oxide fumes. Because of this,welding processes such as SMAW and FCAW,which generate large quantities of fumes, are notusually associated with significant quantities ofozone.
Engstrom and Virtamo determined ozone levelsbehind the welding helmet in the breathing zones ofwelders from 24 welding shops (Ref. 39). A total of118 measurements, 5 to 15 minutes in duration,were taken with several different gas-shielded weld-ing processes. Ozone concentrations fluctuatedwidely during individual measurement periods.Peak levels as high as 7 ppm were recorded duringGMAW of aluminum in the absence of local venti-lation. Ozone concentrations were markedly re-duced by the use of local ventilation and the meanconcentration for all welding processes was lessthan the Finnish standard of 0.1 ppm. GMAW ofaluminum produced the highest concentrationswith a geometric mean of 0.06 ppm. The addition ofnitric oxide to the shield gas was less effective inreducing ozone levels than was predicted by thelaboratory studies of Sipek and Smars (Refs. 132and 157).
3. Electromagnetic Radiation
Electromagnetic radiation in the UV, visible, andinfrared (IR) regions of the spectrum is producedduring welding. Overexposure to radiation in any of
these categories can damage the eyes. In addition,exposure to UV radiation can injure the skin (e.g.,erythema or sunburn, skin cancer). The biologicaleffects of electromagnetic radiation vary substan-tially with the wavelength. Many factors affect theintensity and spectral distribution of the UV lightgenerated by welding equipment. These include thewelding process, base metal, electrode, voltage, cur-rent, and arc length.
Guidelines for determining permissible exposuretimes to electromagnetic radiation that relate thebiological effectiveness to the wavelength (effectiveirradiance) have been established by the AmericanConference of Governmental Industrial Hygienists(ACGIH) (Ref. 1). Using these guidelines, Okunoestimated permissible exposure times at 1 meterfrom the source for the UV light generated by vari-ous welding processes (Ref. 114). The results areshown in Table 2.
The permissible exposure times varied from sev-eral seconds to one minute for most of the weldingprocesses studied. For each welding process, the UVradiation increased with the diameter of the wire,arc current, and voltage. As expected, the UV haz-ard was greater with GMAW than with SMAW.Okuno concluded that permissible exposure timescould readily be exceeded by welders, as well as byother persons in the vicinity.
Using the same welding conditions, Okuno mea-sured the blue light radiation at a distance of 1meter from the arc (Ref. 113). No correlation wasfound between the blue light irradiance and thediameter of the wire, arc current, or voltage. Thegreatest blue light effective radiance (2.4 W/cm2)was found for CO2-shielded GMAW of mild steel,for which a permissible daily exposure time of 42seconds was estimated. In similar work, Salsi andBarlier measured the spectral emissions from UV tofar IR (200 to 3000 nm) generated by 47 diversecoated electrodes. The ACGIH guidelines for UV,visible light, and IR radiation were used to calculatedaily permissible exposure levels and to evaluatethe shade numbers of filters to be used with the elec-trodes studied (Ref. 9).
Proper precautions to protect the eye must be ex-ercised during laser welding, as the cornea and ret-ina can be permanently damaged by exposure tolaser beams. The extent of the hazard varies withthe characteristics of the laser beam and, thus, selec-tion of protective eyewear must address the specificlaser process used. According to Winburn (Ref.189), goggles supplied or recommended by lasermanufacturers often provide more protection thanneeded and, in so doing, do riot promote optimalvisibility. For example, colored plastic goggles are
20
Permissible Exposure
Process
SMAWilmenite
SMAWilmenite
SMAWilmenite
SMAWlow hydrogen
SMAWlow hydrogen
GMAWsolid wire
GMAW
GMAWsolid wire
GMAW
GMAW
GMAW
Self-shielded
N/A - Not applicableSS - Stainless steelMS -Mild steel
Metal
MS
MS
MS
MS
MS
MS
MS
MS
MS
SS
Aluminum
MS
Self-shielded - 'Self-shielded arc welding"Data from Okuno, Refs. 113 and 114
forUV
Diam.(mm)
2.6
4.0
8.0
4.0
8.0
1.2
1.2
1.6
1.6
1.2
1.6
3.2
Table 2Light From Diverse
Current/Voltage
70A20V AC
170A30V AC
400A36V AC
170A23V AC
400A32V AC
150A23V DC(+)
150A215V DC(+)
300A35V DC(+)
300A33V DC(+)
270A35V DC(+)
250A27V DC(+)
400A25V AC
Welding Processes
ShieldGas
N/A
N/A
N/A
N/A
N/A
CO2
5Ar-CO2
CO2
5Ar-CO2
CO2
Argon
N/A
PermissibleExposure TimePer Day (sec)
300
51
33
36
17
23
14
16
4.9
21
31
250
sometimes provided by manufacturers for use withCO2 lasers. Wavelengths produced by these lasersare readily absorbed by almost all transparent mate-rials and any clear goggles would serve as efficientprotective devices without reducing visual acuity.
4. Production Coatings
Toxic substances may be released when weldingmetals coated with paints, primers, greases, andother materials. Heikila et al. (Ref. 53) showed thatphenanthrene, anthracene, pyrene, benzo(a)pyreneand other polycyclic aromatic hydrocarbons are re-leased during SMAW of railroad ties. Benzo(a)pyrene, a potential human carcinogen, was mea-sured in concentrations as high as 0.64 ug/m3 in thevicinity of the welders.
5. Noise
Impulse noise may be more harmful to hearingthan continuous steady state noise of the same en-ergy level (Ref. 133). The impulse noise level maybe defined as the difference between the peak valueand the root mean-square (RMS) of the slow time-weighted value; this difference is also referred to asthe crest factor. Noise is considered to be impulsiveif the difference between peak levels and the slowtime-weighted value is greater than 15 dB.
Starck and Pekkarinen described a telemetricprocedure for measurement of industrial impulsenoise (Ref. 168). The method provides data for bothequivalent noise level and impulsiveness. Measure-ments were made with both a stationary samplingdevice, which provides only a limited estimate of aworker's noise exposure, and with a miniature mi-
21
crophone placed in the worker's ear. Earlier tests,performed at a shipyard plate welding workshop,indicated that grinding and carbon arc work pro-duce minimally impulsive noise, whereas weldingprocesses, especially GMAW, generated high im-pulse noise levels (Ref. 133). A more recent study(Ref. 168) was conducted in a shipyard assemblyhall. The impulsiveness for different processes in-creased in the following order: grinding < carbonarc gouging < chiseling < welding (process not spec-ified) < GMAW. Noise generated by GMAW hadthe highest impulsiveness and relatively low energyequivalent levels, while that generated by gougingand grinding had the lowest impulsiveness and thehighest equivalent levels.
Impulse noise generated by different processeswas analyzed with a transient recorder; the resultsfor grinding, chiseling, and GMAW are shown inFigure 2. Chiseling generated recurrent impulsenoise, with peaks at intervals of about 20 milli-seconds, whereas grinding produced random,nonimpulsive noise. GMAW produced short im-pulses that occurred randomly at intervals of tens ofmilliseconds. The impulsive noise peaks fromGMAW had average equivalent levels of 90 dB andmaximum levels of 120 dB. While technical mea-surements proved noise from GMAW to be highlyimpulsive, noise generated by GMAW does not ap-pear to be impulsive to the human ear because thepeaks are too short to be transmitted through theauditory nerve to the brain.
Tests designed to determine whether hearing pro-tectors effectively attenuate impulsiveness (Ref.138), were conducted at 21 different workplaces.The average equivalent noise varied between 85and 98 dB outside the earmuffs and between 69 and83 dB inside the earmuffs. The average attenuationof equivalent noise levels was 17 dB. Earmuffs re-duced the equivalent noise levels below 85 dB(A) in90% of the measurements. The percent of the totalmeasurement time during which the peak soundlevel exceeded the RMS level by at least 15 dB, was1.6% outside the earmuffs and 0.8% inside the ear-muffs. Thus, earmuffs were effective in reducingimpulse noise. It was noted that the age of the ear-muff and problems with individual fit strongly af-fect the efficiency of these hearing protectors.
6. Hygiene and Work Practices
6.1 Accidents. Accidents are among the most im-portant health risks for welders. A survey in Wash-ington State of accidental fatalities that occurred inthe construction industry between 1973 and 1983
indicated that welders ranked third among occupa-tional groups. (Ref. 17). Accidents are generallyconsidered to be the primary health risk associatedwith the use of robots. Currently, welding opera-tions represent the most common use of robots inindustry. Although published statistics on robot-re-lated accident rates are limited, a study performedin Japan indicated that the majority of accidents oc-cur during teaching, testing, and maintenance. Ac-cidental injuries and deaths mainly involve impact,pinch, or release of an object from the robot's arm.
A common cause of accidents is pushing thewrong button in the robot controls. Other accidentsare caused by improper programming of robotmovement. Some accidents occur because of failureor difficulties in perceiving or anticipating themovement of the robot arm. Few standards exist forcontrolling the maximum speed of arm movements.The robot arm should move slowly enough to givesufficient time for operator reaction in case ofemergency. However, arm movement should be fastenough to be readily perceptible by the operator orother personnel in the area. Pauses in work opera-tions should be short enough to prevent the mis-taken perception that robot operation has stopped(Refs. 56, 57, and 134).
Some robots are being equipped with auxiliarysafety devices which monitor robot operations. Ifpersons or extraneous objects intruding in the robotwork environment are detected, the operation ofthe robot is automatically terminated. It has beensuggested that, in addition to this "watchdog" de-vice, personnel who must function within the robotenclosure should wear a device that transmits a sig-nal to the robot's safety sensor, indicating humanpresence (Ref. 57).
De Nevers described an accident that occurred ina facility where solid rocket boosters from the spaceshuttle program are cleaned and inspected (Ref. 30).The welder was in a basket suspended from anoverhead crane inside a vapor degreaser pit filledwith 1,1,1-trichloroethane vapors. When he startedwelding, he noticed "long hydrocarbon-typeflames" around the arc. He resumed welding afterallegedly being informed by the facility engineerthat it was safe to proceed because 1,1,1-trichlor-oethane would not burn. During the welding opera-tion, a ball of flame developed near his chest andrapidly expanded; the welder fell into the pit as ahuge fireball erupted. His body was severelyburned, but it was not clear whether his death wasdue to burns or to asphyxiation by 1,1,1-trichlor-oethane fumes. Although 1,1,1-trichloroethane hasno flash point, and the liquid solvent will not burn,according to De Nevers, the vapors can burn in the
22
- 1 0 -
(A)
0 10 90 100 (ms)
(C)
Figure 2 — The A-Weighted Sound Pressure Measured With a Stationary Microphoneand Transient Recorder for (A) Chiseling (B) GMAW, and (C) Grinding. Starck and
Pekkarinen, Ref. 168
23
absence of the liquid phase. Commercial formula-tions frequently contain corrosion inhibitors whichincrease the flammability of the solvent. Tests of the1,1,1-trichloroethane remaining after the accidentindicated that it contained twice the quantity ofcorrosion inhibitors normally used, which mayhave contributed to the flammability of the vapors.
6.2 Clothing. To determine their effectiveness inprotecting welders from actinic UV radiation,Sliney et al. measured the attenuation of UV byfabrics commonly used in industrial work clothing(Refs. 12 and 163). The direct and diffuse transmis-sion of UV light were measured separately for 45different fabrics at five wavelengths between 220and 340 nm. Most fabrics transmitted less than0.01% of the incident light, which was considered tobe adequate for protection against UV radiationfrom the welding arc. However, two lightweightfabrics had transmission values close to 1%, whichpermitted transmission of sufficient UV light to ex-ceed current occupational exposure limits.
The attenuation of transmitted light varied littlewith the wavelength, which indicated that lightpassed through openings between the fibers. Laun-dering decreased the UV transmission through allbut one of the fabrics. This was thought to be due toshrinkage of the fabrics with a consequent decreasein the size of the openings between fibers. (The ef-fect of laundering on UV attenuation for severalfabrics is shown in Figure 3.) This study indicatedthat most fabrics used in standard work clothingprovide adequate attenuation of the UV radiationgenerated by welding arcs; thick, dense fabrics arebetter than lightweight materials. The investigatorsrecommended the use of long-sleeved clothes andgloves made of thick, dense material with an opticaldensity of 4 during welding operations with cur-rents up to 400 A.
Although work clothes should be impermeable toUV and should provide some protection againstburns, they should also allow dissipation of sweatand body heat. A study by Beckett et al. demon-strated that vapor-barrier garments in the work-place can cause heat stress (Ref. 11). Plumbers,insulators, and welders, wearing vapor-barrier suitsfor protection against asbestos, were examinedwhile working. One subject had a heart rate of 164beats per minute and an oral temperature of100.6°F while wearing a vapor-barrier suit overconventional clothing. When wearing conventionalclothing only, his heart rate was 112 beats per min-ute, and his oral temperature was 98.3°F. The au-thors noted that when heavy muscular work isperformed in a warm environment, impairment of
the dissipation of sweat by vapor-barrier garmentscould result in significantly higher body tempera-tures with a high risk for heat-related disorders, in-cluding heat stroke.
6.3 Protective Gear. Jakubcik (Ref. 66) described alightweight (16 grams), maintenance-free, dispos-able respirator used in a steel works in Czechoslova-kia to provide protection when welding high alloysteels with electrodes containing chromium, nickel,and manganese. The RVD-F respirator, developedin Carlsbad, covers the mouth and nose, fits readilyunder the welder's mask, and has an expected life-time of about one week. Particles retained on thedust mask were analyzed by SEM and EDXA. Parti-cles as small as 0.2 um were retained by the maskand the elemental composition varied with the sizeand shape of the particle.
Several different methods for extracting weldingfumes from the work area (Refs. 67 and 145) andfrom hyperbaric chambers (Ref. 35) were pub-lished. Olander developed equations that relatedplume height and shape to temperature gradients.With this work he demonstrated that local exhaustventilation provides the most efficient removal offumes from the welder's work area (Ref. 121).
o<oQ.
o
10.0
8.0
6.0
4.0
2.0
0.0
WASHED
UNWASHED
WASHED
UNWASHED
200 250 300WAVELENGTH (nm)
350
Note: Fabrics were washed ten times under standardconditions at 140°C. Lines 1 and 2 represent resultsobtained with different fabrics.
Figure 3 — Effect of Laundering on theUV Transmission Through Fabrics Usedin Work Clothing Sliney et al., Ref. 163
24
6.4 Training. To facilitate the identification ofwork practices that can lead to excessive exposuresto pollutants, Rosen and Lundstrom coupled videotaping of workers and direct-reading instrumenta-tion of breathing zone exposures (Ref. 146). Theresultant video tape of work practices, with a super-imposed presentation of real-time exposure data,was used to evaluate the relationship between expo-sures to fumes, dusts, or solvents with individualwork patterns and to train persons who work withhazardous substances. The authors suggested thatthis method could be augmented to include IR radi-ation and light scattering techniques to enable visu-alization of gaseous emissions and particulates,respectively.
Section TwoEffects of Weldingon Human Health
Welding generates physical and chemical agents,including gases, fumes, radiation, and noise whichcan produce adverse health effects. If welders arenot properly protected, exposure to welding emis-sions may harm the lungs, eyes, ears, and skin.Overexposure to fumes containing certain nonfer-rous metals or metal alloys may also injure the ner-vous system, kidney, and possibly other internalorgans. While a great deal of research has been per-formed on the health effects of welding, much is stillunknown about how the complex emissions gener-ated by welding affect the body. A problem inherentin research on the health effects of welding is thegreat variability in welding processes and in work-ing conditions. This makes it difficult to conductepidemiologic studies on homogeneous popula-tions of sufficient size to permit statistical analysis.Understanding health effects is essential to the de-sign of protective clothing and equipment and tominimizing hazardous emissions from welding pro-cesses. Described in this chapter are health reportswhich appeared in the published literature fromJanuary 1986 through December 1987.
7. Respiratory Tract
Welding emissions may contain irritant particlesand gases such as nitric oxide (NO), nitrogen diox-ide (NO2), hydrogen chloride (HC1), ozone, andphosgene which can elicit an inflammatory re-sponse in the lung. The intensity of the response isdependent on the dose, the physical/chemical prop-erties of the irritant, and the overall health status of
the exposed person. Macrophages migrate into thelung in response to the presence of irritants. Thesecells proliferate mediators, causing inflammation ofrespiratory tissues which may result in cough, chesttightness and pain. Severe respiratory irritants,such as ozone, phosgene, and cadmium, can causeaccumulation of fluids in the lung, followed bychemical pneumonitis or pulmonary edema withinhours after exposure. Chronic exposure to lowlevels of some irritants may result in loss of pulmo-nary function, emphysema, and bronchitis. The as-sociation between exposure to welding fumes andthe development of chronic lung diseases, such asbronchitis, emphysema or lung cancer, has not beenclearly established.
Some of the particulate matter that enters thelungs is engulfed by macrophages (phagocytosis).Many of the free particles and particle-laden macro-phages are removed from the lungs by coughing andthe natural movement of fluids lining the air pas-sages. The remainder of the particles may be depos-ited in the lining of the lungs where they can becomesurrounded by fibrous material. These pockets ofparticles appear as opacities on chest X-rays, a con-dition known as pneumoconiosis. The term "sider-osis" refers specifically to deposits of iron in thelungs. Although some particles (e.g., iron and alu-minum oxide) are relatively inert and can remain inthe lungs for years without adverse effects, com-bined exposure to these metals with other agentsmay lead to fibrotic processes and ventilatory im-pairment (Ref. 31). Other particles such as asbestosand quartz are strong irritants and can induce in-flammatory processes and, eventually, scarring ofthe lungs. Inhalation of sufficient quantities of suchmaterials can lead to extensive deposition of fi-brous or scar tissue, appearing in X-rays as massiveor coalescent opacities instead of isolated spots.Such scarring can cause loss of elasticity in the lungand reduced pulmonary function.
7.1 Alveolar Macrophages. Some of the lung dam-age that results from chronic exposure to low levelsof irritants is thought to be caused by proteolyticenzymes released from macrophages at the site ofinflammation. These enzymes hydrolyze proteinsand hence may damage cells and tissues in their vi-cinity. Antiproteinases, molecules that inhibit theaction of proteolytic enzymes, are normally presentin the lungs and are thought to provide some protec-tion against the destructive activity of proteolyticenzymes. Based on the assumption that an-tiproteinases are necessary for protection againstthe deleterious effects of inflammatory processes inthe lung, Pierre et al. (Ref. 140) examined the inter-
25
actions of welding fumes with two lung an-tiproteinases alpha-1-proteinase inhibitor andbronchial inhibitor. They found that welding fumeparticles bound reversibly to these antiproteinasesand prevented them from inhibiting the activity ofproteolytic enzymes.
The number of macrophages in the lungs may in-crease substantially in response to inhalation of irri-tants. Mylius and Gullvag used the number ofalveolar macrophages in sputum as a measure of theinflammatory response elicited by foreign materialsin the lung (Ref. 104). They showed that sputumfrom coke plant workers, aluminum plant potroomworkers, asphalt workers, and shipyard welders hadmore alveolar macrophages than controls. Smokingand occupational exposures combined to give amarked synergistic effect (Figure 4). The investiga-tors reasoned that the continuous release of proteo-lytic enzymes from alveolar macrophages inchronically exposed persons may represent a haz-ard to lung tissue.
7.2 Estimation of Retained Particles in the Lungs.Some of the particles in welding fumes containFe3O4 or other minerals (e.g., maghemite or gammaFe2O3, and some sulfides of iron, nickel and cobalt)that can be magnetized. Magnetopneumography,developed in 1973 by Cohen (Ref. 24), uses thisproperty to estimate the burden of occupationaldusts in the respiratory tract. A magnetic field is ap-plied to the chest which magnetizes and rotates theparticles into a common alignment. When the ex-ternal magnetic field is removed, the remanent fieldof the magnetized particles is determined and usedas a measure of the amount of dust in the lungs. Theprinciples involved in this technique, its applicabil-ity, advantages, and limitations were reviewed byLippmann (Ref. 91).
A difficulty associated with the application ofmagnetopneumography to welding is that the mag-netic properties of fumes produced by different pro-cesses vary. For example, fumes from GMAW ofstainless steel have a high specific remanence whilethat of fumes from SMAW of stainless steel is low. Ifa welder is exposed to both types of fumes, the re-manent field generated by GMAW will dominate,and the dust burden from SMAW fumes will be un-derestimated. To overcome this problem, Kal-liomaki et al. (Ref. 75) developed a system whichmeasures the pulsed coercive force. The coerciveforce is the reverse magnetic field which must beapplied to just cancel the remanent field. It has acharacteristic value for individual industrial aero-sols and can be used to discriminate between rema-nent fields produced by different fumes.
30 - i
25 -
20 -
ccLUm
15 -
10 -
5 -
SMOKERS
NONSMOKERS
iMEDIAN
CONTROL
0.1 0.8
WELDERS
0.5 30.5
Figure 4 — Alveolar Macrophage (AM)Counts in 9 Nonsmoking and 11 SmokingWelders at a Ship-Building Yard. Mylius
and Gullvag, Ref. 104
An innovation introduced by Drenck and Stern(Ref. 36) combined AC susceptibility bridgemagnetopneumography with a DC method similarto that originally designed by Kalliomaki et al. Thesystem provides information about the total con-tent of ferromagnetic, paramagnetic, and diamag-netic material. Using this system, Stern et al. (Ref.
26
172) characterized the magnetic properties offumes from a variety of welding processes and otherindustrial aerosols. X-ray diffraction and Moss-bauer spectroscopy indicated that substituted mag-netite is the primary magnetic component in freshlyformed welding fumes, whereas a less substitutedform of magnetite appeared to be the predominantmagnetic component in inorganic dust obtainedfrom ashed lungs of deceased welders. Magneto-pneumographic studies on live welders also pro-vided limited evidence for in vivo changes ofwelding fume particles to the magnetic constituentmagnetite. The investigators concluded that thesechanges may result from loss, with time, of the moresoluble components of welding fume particles.
A preliminary report by Kalliomaki et al. (Ref.76) also suggested that the remanent magnetic fieldof welding particles in the lungs may not be stable.Four different types of welding fumes were intro-duced into the respiratory tract of rats by intratra-cheal instillation and the specific magnetic moment(magnetic moment divided by iron concentration)was followed for the succeeding 106 days. The spe-cific magnetic moment of welding fumes generatedby SMAW of mild steel and GMAW of stainlesssteel were stable, while the magnetic moment ofwelding fumes generated by SMAW of stainlesssteel and GMAW of mild steel decreased by about50% during the 106-day follow-up period. Littlechange was seen in the remanent magnetism ofwelding fumes during a 55-day incubation in hotwater (70°C) or physiological saline (70°C). On thebasis of these data, it was postulated that the insta-bility of the remanent moment of some weldingfumes could be due to the slow transformation ofgamma-Fe2O3 to alpha-Fe2O3 or to differences insolubility of particulate material. In addition, dif-ferences in clearance rate among different sized par-ticles could contribute to changes in the specificmagnetic moment.
7.3 Pulmonary Function and Bronchitis. Pulmo-nary function tests are used to detect disease pro-cesses, such as fibrosis and emphysema, whichrestrict lung expansion or reduce pulmonary elas-ticity, or both. These tests measure the air volumewhich can be inhaled or expelled either forcefully orunder normal breathing conditions. Pulmonaryfunction tests are often used to monitor industrialworkers for damage to the respiratory tract. How-ever, these measurements are not always sensitiveto early changes in the lungs, and irreversible dam-age may occur before noticeable reduction in pul-monary function is detected.
Disparities have appeared in the literature con-
cerning the effects of welding on pulmonary func-tion. It has been suggested that differences in resultsof pulmonary function studies in welders may stemfrom the use of tests that measure those parametersthat are least likely to be affected by welding (Refs.74 and 81). Upon inhalation, the minute particlesin welding fumes would tend to be deposited in thedistal airways and alveolar spaces of the lung, wheremost pathological changes would be expected. Inthis regard, it has been argued (Refs. 74 and 81) thatpulmonary function tests which measure impair-ment in the small airways e.g., forced expiratoryflow from 25 to 75% (FEF25-75) and FEF75-85 aremore appropriate for measuring pulmonary func-tion in welders than are the more conventional spi-rometric indices [e.g., vital capacity (VC), forcedexpiratory value in the first second (FEV1) andFEV%]. Additional inaccuracies in the results ofpulmonary function studies in welders may resultfrom reports of respiratory symptoms among weld-ers with confounding exposures such as asbestos orcigarette smoke. Finally, it is not known how fumesfrom different welding processes vary with respectto chronic effects on the lungs and, thus, compari-sons of welders working with different processesmay be misleading.
In their survey of welders from nine Danish ship-yards, Lyngenbo et al. (Ref. 92) examined the inci-dence of bronchitis and upper respiratory tractsymptoms (colds, sore throats, hoarseness) in 2660welders and 881 electricians. The welders were di-vided into two groups: 1516 had high exposures(greater than 10 hours per week) and the rest hadlower exposures (10 hours per week or less). Most ofthe welders had performed SMAW of mild steel forat least 20 years, and both groups were comparablein respect to asbestos exposure and tobacco use.Chronic bronchitis and upper respiratory tractsymptoms occurred significantly more often inwelders than in engineers, and the frequency of re-spiratory symptoms was related to the extent ofwelding exposure. There was an additive relation-ship between smoking and welding in regard todevelopment of respiratory symptoms. The investi-gators estimated that 8 hours of welding is equiva-lent to smoking 25 cigarettes.
Pulmonary function tests (Ref. 92) and magneto-pneumographic measurements (Ref. 171) were per-formed on 74 welders from the high-exposure groupwho had never smoked and had never worked withasbestos or other materials known to produce lungdamage. The welders had significantly impairedvalues for VC, forced vital capacity (FVC), totallung capacity (TLC), diffusion capacity (TCO),peak expiratory flow (PEF), FEV1, and the maxi-
27
mum expiratory flow/75% (MEF75) relative to con-trols. There was no association between thequantity of metal contaminants in the lung and theincidence of chronic bronchitis. The major conclu-sions drawn from this study were that (1) SMAwelders have a significantly elevated incidence ofupper and lower respiratory tract symptoms; (2) theincidence of respiratory tract symptoms correlatespositively with the extent of exposure; and (3) weld-ers with high levels of exposure and no history oftobacco use have an elevated incidence of deficits inlung function (Ref. 92).
Kilburn et al. (Refs. 81 and 82) examined lungfunction and respiratory symptoms in 148 weldersfrom a Los Angeles shipyard. The study group con-sisted of 73 nonsmokers with 11.3 years' weldingexperience, 18 ex-smokers who had welded for 16.5years, and 57 current smokers who had welded for10.5 years. The frequency of respiratory illness,shortness of breath, and chest pain or heavinessduring exercise was more prevalent in all threewelding groups than in controls. The prevalence ofbronchitis was lowest in nonsmoking welders, butwas still substantially higher in this group than in acontrol population stratified according to tobaccouse. Of the spirometric measurements, the majorfinding was a 20% reduction in FVC and FEV1 inwelders who smoked. Nonsmokers had significantreductions in FVC (4.3%), FEV1 (9%), FEF25-75(4.4%) and FEF75-85 (14.4%). The reduction ofthese parameters in nonsmoking welders was lessthan half that in cigarette smokers, suggesting a syn-ergism between welding fumes and cigarette smoke.The investigators recognized that prior exposure toasbestos may have confounded pulmonary functiondata obtained in this study.
As a continuation of their studies of the long-termeffects of welding exposure (Ref. 133), Spacilovaand Hykes followed lung function and chest X-raypatterns in 74 welders for 8 years (Ref. 166). In non-smoking welders, lung function and chest X-rayswere normal at the start of the study period andshowed no statistically significant changes after 8years. Welders who smoked showed initial stages ofrespiratory obstruction and even more frequentlysuffered from chronic bronchitis. The X-ray pat-terns progressed in 24% of the smokers versus 4% ofnonsmokers and regressed in 6% of the smokers and14% of nonsmokers.
Similar results were obtained when the welderswho ceased welding at the start of the 8-year follow-up period were studied independently from thosewho continued welding. As for the entire cohort,there was no statistical difference in pulmonaryfunction or in the degree of nodulation evident in
chest X-rays at the beginning and end of the follow-up period. Spacilova and Hykes concluded that X-ray findings of nodular opacities in arc welders arenot necessarily indicative of a pathological condi-tion and that welding exposures alone do not lead toserious pulmonary impairment. However, smokingin combination with welding presents a much moresevere hazard than welding alone (Ref. 166).
In another study by these investigators, the gen-eral health status of 73 of the welders was comparedwith that of 52 persons with a history of occupa-tional exposure to silica. The incidence of chronicbronchitis and ischemic heart disease did notdiffer significantly between the two groups. In addi-tion, there were no significant differences in theincidence of hypertension, obesity, or diabetes(Ref. 165).
Kalliomaki et al. (Ref. 74) followed changes overa 6-year period in pulmonary function and the pul-monary dust load of shipyard welders. The 47 SMAwelders participating in the study worked mostlywith mild steel. They were divided into four groupsaccording to their total welding experience. (Activewelders had approximately 4, 18, or 26 years weld-ing experience, while retired welders had approxi-mately 25 years welding experience.) In general,lung function values of retired welders were signifi-cantly lower than normal while those of youngerwelders were within normal limits. The mean VC ofretired welders was significantly lower than that ofmen who had welded for about 4 years while themean carbon monoxide diffusion capacity of re-tired welders was significantly lower than that of allactive welders. Values for CV did not differ betweenany of the groups.
Among welders with 4 years experience at thestart of the study, the quantity of metals retained inthe respiratory tract increased linearly with expo-sure time during the first 5 years, with an average of70 mg dust being retained per year. The quantity oflung contaminants was greatest in persons who hadwelded for 18 to 26 years. The lung contaminationamong these welders decreased slowly during the 6-year follow-up period, reflecting engineering im-provements in the workplace. In retired welders,metals were cleared at a mean rate of 18% per year.
Schneider et al. previously reported that the inci-dence of bronchitis and nodular lung opacities wassignificantly greater in 433 SMA welders of mildsteel than in 421 nonwelding controls from thesame factory (Ref. 133). In that report, deficits inFEV were related to the duration of welding experi-ence. More recently, these values were related towelding fume exposure as measured by personal airmonitoring. The prevalence of impaired FEV1/VC,
28
decrease in FEV1 following acetylcholine chal-lenge, chronic cough, X-ray opacities, and pain dur-ing exercise increased with exposure to weldingfumes. Manganese levels in hair were positively as-sociated with spirometric test results, nasal sinusopacities, and pain during exercise (Ref. 149).
The prevalence rates of large airway obstructions(FEV1) and pneumoconiosis were the same among158 GTA and GMA welders of stainless steel andaluminum and 159 controls. The welders had a sig-nificantly higher rate of disturbed ventilatory dis-tribution (V2/V1) and chronic phlegm production.The frequency of impaired FEV1(VC,V2)V1,MEF50, and chronic bronchitis varied positivelywith chromium levels in hair (Ref. 149).
Schneider et al. also performed a cross sectionalstudy of shipyard welders who had worked withmild steel for about 18 years. Siderosis was evidentin chest X-rays from 15% of the welders and 8% ofnonwelding controls who worked at the same ship-yards (Ref. 150). More extensive clinical tests, in-cluding spirometry, flow-volume curves, bodyplethysmography, single-breath diffusion capacity,pulmonary compliance and ergooxytensiometry,were performed with 56 of the 64 welders with sid-erosis. The results indicated that siderosis is associ-ated with a slight deterioration of pulmonaryfunction (Refs. 150 and 151).
Krasnyuk et al. (Ref. 84) examined the effects ofexposure to high and low alloy welding fumes on therespiratory tract and cardiovascular system. Thestudy group consisted of 500 welders, about half ofwhom were exposed to fumes containing manga-nese, chromium, and nickel; 200 engineers from thesame company served as controls. Chronic bronchi-tis and pneumosclerosis, the major pathologic re-spiratory conditions noted, were more frequent inwelders than in controls. Chronic bronchitis wastwice as prevalent in low alloy steel welders whilepneumosclerosis and pneumoconiosis occurredmore frequently in welders of high alloy steel. TheVC was decreased in welders with overt pathologicpulmonary processes. The degree of this impair-ment increased with the length of employment inwelding. Restrictive disturbances were more severein welders of high alloy steel and smoking exacer-bated impairment in pulmonary function. Hyper-tension and electrocardiographic changes weremore common in high alloy than in low alloy steelwelders.
Cotes and El-Gamal (Ref. 26) conducted clinicaland spirometric tests on 596 welders from a ship-yard in northeast England. Welding exposures wereassociated with siderosis, but not respiratory im-pairment. No ill effects were associated with burn-
ing. Welding aggravated chronic bronchitis insmokers, while both welding and burning reducedthe FEV1 and increased the index of airflow ob-struction in smokers. The investigators concludedthat fumes from welding and burning interact withtobacco smoke to cause respiratory symptoms andsome impairment of lung function. Tests conductedwith 345 of the younger shipyard workers (ages23 to 47 years) showed that exercise ventilation in-creased in association with welding or caulker/burning in confined or semiconfined spaces; ageand smoking enhanced this effect. The investigatorsconcluded that welding in confined or semicon-fined spaces causes respiratory symptoms, someimpairment of lung function, and an increased ven-tilatory cost of exercise in smokers (Ref. 26).
Ulrich et al. (Ref. 179) compared chest X-raysand pulmonary function in 79 persons exposed tochromium during welding of austenitic steel, 109current welders of mild steel, and 93 former weldersof mild steel. Austenitic steel welders were exposedto an average total chromium concentration of0.152 mg/m3. The incidence of chronic bronchitiswas 53% for welders of austenitic steel, 62% forwelders of mild steel, and 36% for retired mild steelwelders. The prevalence of positive X-ray findingsand results of pulmonary function tests (FVC andFEV1 were normal; MEF50 was reduced by 13 to17%) were similar in all welders.
Moravik (Ref. 102) examined lung function andthe incidence of respiratory symptoms in a cohortof 121 welders in Czechoslovakia. About 70% of thewelders worked with CO2-shielded GMAW whilethe remainder worked with SMAW. Welders tookan average of 15 days of sick leave per year as com-pared with an average of 4 days per year taken bythe 42 office worker controls. The incidence ofbronchitis was significantly greater in welders whosmoked than in those who did not. In addition, nineof the nonsmoking welders had bronchitis whilenone of the nonsmoking controls had this disorder.Deficits in pulmonary function increased with theduration of welding experience; the relationshipbetween pulmonary function and smoking wasnot discussed. The incidence of pneumoconiosiswas directly related to the duration of weldingexperience.
Oleru and Ademiluyi (Ref. 122) noted changes inpulmonary function in 16 welders in a metal cuttingand welding factory in Nigeria. Dust levels were es-timated to be above 5 mg/m3 in welding areas. Thewelders, as well as workers in the same factory withother exposures (13 paint dippers, 7 metal cutters,18 aluminum workers), had reductions in FEV1and FVC relative to controls. Acute 40-hour
29
changes in FEV1 and FVC were observed in eightwelders when measurements taken at the beginningand the end of the work week were compared. Acute8 hour, but not 24 hour, changes in peak expiratoryflow rate were also observed in the welders.
Dolan and Dolan (Ref. 34) studied the prevalenceof pulmonary disease in 104 welders and pipefitterswith 20 or more years experience in diverse indus-tries from three population centers in California.Pulmonary function and chest X-rays were normalin only 12% and 16%, respectively, of the welders.About 95% of the welders had worked with oraround asbestos. The prevalence of asbestosis pre-cluded the evaluation of the effects of weldingfumes on pulmonary function.
Desmeules et al. reviewed medical histories of 68welders with siderosis (Ref. 32). Mean pulmonaryfunction values were normal but 31 had chronicbronchitis. Ten of 17 persons with inspiratory ralesreported exposures to dusts other than weldingaerosols. The progression of lesions was followed in27 welders with siderosis. Follow-up X-ray exami-nations were performed a mean of 8.8 years afterinitial diagnosis. Lesions regressed in 14 cases, 5cases remained stable, and opacities increased in 8cases. Exposure to nonwelding dusts was reportedfor 6 of 13 cases with stable or progressing opacities,but in none with regressing lesions (Ref. 33). It wasconcluded that dust exposures other than weldingmay be responsible for progression of pulmonaryabnormalities in welders with siderosis.
Based on an extensive review of the literature,Zober concluded that siderosis is a benign condi-tion and pulmonary fibrosis is a rare disease inwelders (Ref. 192). His analysis indicated that thereis no statistical evidence for an association betweenobstructive lung disease and welding. Accordingly,for purposes of worker compensation in West Ger-many, fibrotic lung conditions can be considered asoccupational diseases only if there are proven expo-sures to known fibrogenic substances such as asbes-tos, silica, or certain metals (chromium, beryllium,cadmium).
In summary, the majority of the 13 studies de-scribed above found deficits in lung function or ex-cess chronic bronchitis among welders, or both. Thereports differed on the contribution that tobaccosmoke made to the development of these impair-ments. In one study (Ref. 34), there was a high prev-alence of asbestosis among welders, while inanother study the number of welders was small andthe potential for nonwelding exposures appeared tobe high (Ref. 122).
Of the remaining eleven studies, five found thatnonsmoking welders have an excess risk for bron-
chitis (Refs. 80, 81, 84, 92, 102, and 149) while two(Refs. 26 and 166) found a high incidence of bron-chitis only among welders who smoked. Two stud-ies found deficits in pulmonary function in smokingwelders, but none in welders who did not use to-bacco products (Refs. 26 and 166). Three otherstudies found an elevated incidence of pulmonarydeficits in both smoking and nonsmoking welders.In a small cohort of shipyard welders, Kalliomaki(Ref. 74) found deficits in pulmonary function onlyamong retired shipyard welders and not among theyounger, active welders. Finally, the investigatorsfrom two studies (Refs. 26 and 166) concluded thatsiderosis or pneumoconiosis in welders is not asso-ciated with a reduction in lung function while thosefrom two other studies (Refs. 84,150, and 151) con-cluded that such an association exists.
Although there is some conflict among these re-ports, the results suggest that bronchitis and deficitsin pulmonary function among nonsmoking weldersis possible. Clearly, welders who smoke have an ele-vated incidence of both bronchitis and pulmonarydeficits. Whether smoking and welding exposureshave additive or synergistic effects is unknown. Thestudy of shipyard welders by Lyngenbo et al. (Ref92) carefully controlled for exposure to cigarettesmoke, asbestos, and other lung toxicants. Thatstudy demonstrated that welding alone can causebronchitis, upper respiratory tract symptoms, andreduced lung function.
The effects of cigarette smoke on the develop-ment of these symptoms were additive rather thansynergistic. The influence of the chemical constitu-ents of welding fumes on the development of pul-monary impairments is still not established.Krasnyuk (Ref. 84) found that welders of high alloysteel were more likely to have extensive pneumoco-niosis with more severe restrictive disturbances inpulmonary function than welders of low alloy steel,while Ulrich (Ref. 179) found no differences in pul-monary function between welders with high andlow chromium exposures.
7.4 Acute Respiratory Tract Disease — Case Re-port. The case of a man who developed a sudden,work-related onset of acute pulmonary disease 5months after the start of his employment as a full-time welder was described by Silberschmid (Ref.156). At first, his symptoms (coughing, chills withfever, tightness in the chest, and shortness ofbreath) occurred only at work. The symptoms be-came persistent during the next 2 months and head-aches, general weakness, and weight loss developed.Clinical examination showed swollen, red bron-chial mucosa, irregular opacities in chest X-rays,
30
and deficits in pulmonary function. Extreme in-flammatory changes were found in biopsied lungtissue. Steroid treatment produced some immedi-ate improvement, but exertional dyspnea, bron-chial hyperreactivity, and recurrent pneumoniaspersisted for 7 years.
Before the start of his employment, the patienthad only occasional experience welding and had nohistory of pulmonary disorders. At the factorywhere his disease developed, he worked with mildsteel, using mostly SMAW with basic and rutileelectrodes; occasionally GMAW of stainless steel oraluminum was performed. Inspection of the workplace indicated that he had been working withoutproperly functioning local ventilation. Fume con-centrations were high (about 20 mg/m3) and manga-nese and copper levels varied from 2 to 4 mg/m3 andfrom 0.3 to 1.2 mg/m3, respectively. Silberschmidcommented that the development of chronic fibro-sis following acute pulmonary reactions to weldingfume components is uncommon.
7.5 Production Coatings. Emmerling et al. (Ref.38) examined effects of welding primed metals in 35persons. Metals primed with zinc chromate wereroutinely welded by ten welders, nine worked withmetals primed with alkyd paints and six weldedmetals primed with polyvinylbutyryal coats. Con-trols were flame cutters, welders of unprimed met-als, and persons without welding experience.Breathing zone concentrations of Zn, Cr, Ni, Fe,Cd, NOX, CO, formaldehyde, toluene, xylene, andethylbenzene were all below the respective GermanMAKs.
There was no correlation between workplace ex-posure and concentrations of zinc, hippuric acid,methylhippuric acid, formic acid, and carboxyhe-moglobin in blood and urine samples. No changesin pulmonary function occurred during the shift orduring the work week which could be related towelding of coated metals. On the other hand, non-specific bronchial hypersensitivity, as measured bymethacholine provocation, was two times higherbefore, and three times higher after the work shift inwelders of primed metals than in controls. The dif-ferences were not statistically significant, which wasattributed to the small number of participants in thestudy.
8. Cancer
8.1 Epidemiologic Studies. Although numerousstudies have considered the incidence of cancer inwelders, the cancer risk associated with welding is
still uncertain. An elevated cancer incidence hasbeen reported at different sites in the body, includ-ing the lung, nasal sinus, bladder, kidney, and diges-tive tract. Of these, the most frequently reportedtumor is cancer of the lung. However, the associa-tion between lung cancer and welding is not clearlyestablished. It has been suggested that focusing epi-demiologic studies on populations of welders mostlikely to be exposed to carcinogens (e.g., chromiumor nickel exposures in stainless steel fumes) mayyield more meaningful results.
In his recent review of published cancer epidemi-ology data, Peto concluded that the association be-tween cancer and welding exposures has not beenadequately investigated (Ref. 139). According toPeto, many epidemiologic studies of welders havenot considered the duration of the welding experi-ence and the types of welding processes used. Theconfounding effects of tobacco smoking and otheroccupational and nonoccupational exposures haveoften been insufficiently addressed. In addition, inmany studies the numbers of cancer cases amongwelders were too small to provide an accurate esti-mate of the relative risk (RR).
Peto described conditions for an epidemiologicstudy that would provide answers to many ques-tions now left open by previous work. This retro-spective cohort study would include 1000 men, halfof whom had already died. The subjects would havebeen exposed to welding fume, with no significantoutside occupational exposures, for at least 5 yearsor, preferably, all their working lives. To ensure suf-ficient time for the development of tumors, thewelding cohort would have begun their employ-ment as welders at least 20 years before the start ofthe study. Information on the welding processesused would be carefully collected so that a subset ofwelders with specific exposures, such as chromiumand nickel in stainless steel, could be readily identi-fied and studied.
Several epidemiology studies published during1986 and 1987 addressed the lung cancer incidencein welders. A population-based case-control studyof the association between occupation and lungcancer was conducted in New Mexico by Lerchen etal. (Ref. 90). The study group included 506 patients,ages 25 to 84 years, with primary lung cancer, otherthan bronchioloalveolar carcinoma. The controlswere 771 persons, frequency-matched to the casesfor sex, ethnicity, and 10-year age category. Partici-pants in the study group were identified by the NewMexico Tumor Registry and controls were obtainedthrough the Medicare Program. The participants(or surrogates) were interviewed to obtain informa-tion on occupational exposures, smoking habits,
31
and employment histories. Welders had a signifi-cantly elevated lung cancer risk, with an odds ratio(OR) of 3.2 (p<0.1). The increased lung cancer riskpersisted after adjusting for cigarette smoking andalso when those with shipyard experience were ex-cluded to reduce the number of participants withasbestos exposure. The authors cautioned that the"welders in the study had worked in diverse indus-tries that may have involved exposures other thanwelding that also increased the lung cancer risk."
The association between occupation and lungcancer risk in a case-control study in three munici-pal areas of New Jersey was examined by Schoen-berg et al. (Ref. 152). The study included 763 whitemales with primary cancer of the trachea, bronchus,and lung; the controls were 900 white males selectedfrom driver's license and death certificate files inthe general population. Workers in the shipbuildingindustry had a significantly elevated risk for lungcancer. Although the excess risk for all shipbuildingworkers was primarily among those with reportedexposure to asbestos, the risk was also high amongwelders and other workers with no reported asbes-tos exposure. The lung cancer risk was not elevatedin welders who had no experience in the shipbuild-ing industry.
Sjogren et al. performed a morbidity and mortal-ity study based on the 1960 Swedish National Cen-sus in which over 20 000 males, aged 20 to 69 years,stated their occupation as welders or gas cutters(Refs. 158 and 162). The 1960 census data werelinked with the Register for Statistics of Causes ofDeath and the Cancer Register for 1961 to 1970.Data on tobacco use were taken from a survey ofSwedish males in different occupational groups.Pulmonary tumors were more frequent than ex-pected among welders and gas cutters. After cor-recting for smoking habits, the rate-ratio was 1.30(p=0.02). In addition, the incidence of renal tumorswas significantly elevated in welders (rate-ratio =1.30; p=0.04) while that of malignant melanomawas decreased (rate-ratio = 0.58; p=0.01). Theprevalence of pleural mesothelioma was also ele-vated, suggesting cross exposures to asbestos in thewelding cohort. No difference was observed in therate of nasal tumors between welders and controls.Sjogren commented that a weakness inherent inthis type of study is that the general classificationwelder included those with only brief exposures(Ref. 158).
To examine whether the increased lung cancerrate was associated with exposure to chromium,Sjogren et al. (Ref. 161) compared the mortality andthe lung cancer incidence in welders exposed tofumes containing high and low levels of hexavalent
chromium. The study groups consisted of 234 weld-ers who had worked primarily with stainless steelfor at least 5 years and 208 railway track welderswho were exposed to only low levels of chromium atsome time between 1950 and 1965. Although asbes-tos exposures could not be totally eliminated, weld-ers selected for the study had not worked in areaswhere asbestos dusts were generated. Five welderswith high chromium exposure and one with lowchromium exposure died from lung cancer. Theserates did not differ significantly from those of thegeneral population, but the difference between thetwo groups of welders was statistically significant.
A case-control study of the association betweenlung cancer and occupation was conducted in twoindustrial areas of Norway by Kjuus et al. (Ref. 83).The study included 176 lung cancer cases and anequal number of referents without lung cancer orobstructive lung disease who were identifiedthrough hospital records. The data were adjustedfor cigarette smoking. An increased lung cancer riskwas seen in welders (28 cases of lung cancer; RR =1.9). The risk was higher for a subset of 16 stainlesssteel welders (RR = 3.3).
Corey et al. (Ref. 25) published a brief report of astudy in which the number of deaths from cancer in3000 welders was compared with death rates in acohort of 6500 plumbers and with the Ontario malepopulation. The lung cancer rate of welders did notdiffer significantly from that of the two controlpopulations.
Pearce and Howard (Ref. 136) studied the associ-ation between occupation and cancer mortality inNew Zealand. Demographic data and statistics ontobacco use stratified according to social class wereobtained from the 1976 New Zealand Census; can-cer mortality data were gathered from the NationalHealth Statistics Center's records of deaths regis-tered in New Zealand between 1974 and 1978.Among the male population, 5356 cancer deaths oc-curred. Only tumor sites with a relative risk greaterthan 1.5 were considered to be excessive and werecited in the study. Stomach cancer mortality (12 ob-served deaths in welders; RR = 1.95) was elevatedin welders, while mortality from cancers of the co-lon, leukemia, bladder, liver, lung, and buccal cav-ity in welders did not differ from that of the generalpopulation.
Puntoni et al. conducted a mortality study of2702 dockyard workers in Genoa who died between1960 and 1979 (Ref. 141). Of these workers, 147were gas welders and 141 were electric arc welders.Causes of death were compared with those of thegeneral population of Genoa. A significant increasein relative risk was found for total deaths, for diges-
32
tive tract cancer (7 cases observed versus 3 casesexpected; RR = 2.33), and for cardiovascular dis-ease among gas welders. No cancer excesses wereseen among electrical arc welders.
Some recent studies of electrical and electronicsworkers have indicated that exposure to extremelylow frequency electromagnetic radiation (ELF) maybe associated with an excess leukemia risk. Weldersare exposed to ELF magnetic flux densities that arefrom 2 to 200 times those generally experienced inthe household or by persons in other electricaltrades. To better assess the risk of leukemia in weld-ers, R. M. Stern examined pooled data from 15 pub-lished epidemiology studies that includedinformation on the incidence of leukemia in weld-ers (Ref. 170). The studies included a total of 146cases of observed leukemia versus 159.46 expected,with a risk ratio of 0.92. For acute leukemia, 40cases were observed versus 43.39 expected, with arisk ratio 0.92. Thus, Stern's calculations indicatedthat welders do not have an increased risk for leuke-mia. Stern concluded that the absence of an in-creased leukemia risk among welders does notsupport the hypothesis that the observed excess riskof leukemia among workers in the electrical tradesis due to their ELF exposure.
A fivefold excess leukemia rate among workers atthe Portsmouth (New Hampshire) Naval Shipyardwas reported in 1978 (Ref. 105). At that time, theexcess risk was thought to be due to radiation expo-sure during manufacture and maintenance work onnuclear submarines. As a follow-up to this report,the National Institute for Occupational Safety andHealth conducted a matched case-control study ofthe leukemia incidence among the 24 545 onshoreworkers in all job categories employed between Jan-uary 1952 and August 1977 at the Portsmouth ship-yard (Ref. 169). Included in the study were 53 casesof deaths from leukemia and 212 controls. Thestudy sought to ascertain whether the leukemiadeaths were associated with occupational exposureto either ionizing radiation or to organic solvents.Radiation dose histories and detailed work histo-ries by job and shop were evaluated for each sub-ject.
Contrary to expectations, there were no signifi-cant associations between either ionizing radiationor presumed solvent exposure and myelogenous orlymphatic leukemia. However, an elevated inci-dence of leukemia was found among electriciansand welders. For electricians, the Mantel Haenszelodds ratio was significantly elevated for all leuke-mias (ORMH = 3.00), and particularly for lym-phatic leukemia, (ORMH = 6.00) while for welders,the odds ratio was significantly elevated for my-
eloid leukemia (ORMH = 3.83), but not for lym-phatic leukemia. (Ref. 169)
As part of a cooperative Scandinavian investiga-tion, Hansen et al. (Ref. 52) conducted a case-con-trol study of the incidence of cancer of the nasalcavity and paranasal sinus. The 49 patients withprimary nasal tumors were identified through therecords of Danish Cancer Centers. Patients withcolorectal cancer served as controls and were iden-tified through the Cancer Registries in Denmark,Sweden, and Finland. Occupational histories wereobtained by interview for each subject and coveredthe period between the start of employment and 10years before the diagnosis of disease. Because of thelong latent period of most cancers, exposures in-curred during the 10-year period immediately pre-ceding disease development were considered to beunrelated to the etiology of the cancer. Exposure towelding, cutting, and soldering fumes was found tobe associated with an excess of nasal cancer. Thedata were not adjusted for asbestos exposure,which, according to the authors, may be a contrib-uting factor to the development of both nasal andcolorectal cancer.
Using both case-control and retrospective cohortapproaches, Norell et al. (Ref. 109) examined therelationship between occupational factors and pan-creatic cancer in Sweden. The case-control studyused all newly diagnosed cancers of the exocrinepancreas in patients between 40 and 79 years old atthree hospitals. Age- and sex-matched controls werepatients with inguinal hernias and subjects chosenfrom the general population. Information on occu-pational history and exposures was obtained foreach participant. The retrospective cohort mortal-ity study was based on data from the Swedish Can-cer Registry for 1961 to 1979 and from the 1960census. The observed number of tumors in variousjob categories was compared with numbers ex-pected from the cumulative incidence for all em-ployed males.
An excess risk for pancreatic cancer among weld-ers was suggested by the case-control study. Pancre-atic cancer was observed in 13 patients withwelding experience. The relative risks calculatedfrom hospital controls and population controlswere 1.7 and 2.0, respectively. In the retrospectivecohort study, welders were not considered sepa-rately, but rather were included in the occupationalcategories "iron and steel works" and "iron andsteel manufacture." Men in these categories showedno excess risk for pancreatic cancer. The standard-ized mortality ratios (SMR) for pancreatic cancerwere 1.1 and 0.9 for "iron and steel works" and"iron and steel manufacture," respectively.
33
McLaughlin et al. conducted population-basedassessments of the occupational risks for brain tu-mors (Ref. 97) and meningiomas of the central ner-vous system (Ref. 98). Data for these studies wereobtained from the Swedish Cancer/EnvironmentRegistry. The risk of intracranial gliomas for vari-ous occupations, industrial groups, and regions wasestimated with a 19-year cumulative standardizedincidence ratio (SIR). The SIR is the ratio of thenumber of tumors observed to those expected in aparticular employment category. Occupationalrisks were evaluated for 3394 cases of intracranialglioma (the most frequently encountered brain tu-mor in adults), and 1092 cases of meningiomaamong employed Swedish men for the years 1961 to1979. Welders and metal cutters had a significantlyelevated risk of developing brain cancer (p < 0.05)but not central nervous system meningiomas.
In contrast to the study of McLaughlin et al. (Ref.97), studies by Thomas et al. (Ref. 175) and Olin etal. (Ref. 123) did not find an elevated risk of braincancer in welders. Thomas et al. (Ref. 175) con-ducted a death certificate-based case-control studyof workers in the electrical and electronics industry.This study included 435 white males from northernNew Jersey, southern Louisiana, and Philadelphiawho died from primary brain tumors between 1979and 1981, and 386 controls who died from othercauses. The case-control study of Olin et al. (Ref.123) compared 78 astrocytoma patients, includingfive welders from two Swedish hospitals, with 197clinical and 92 population controls. Risk estimatesfor astrocytomas in welders of 0.6 and 0.2 werebased on hospital controls and population controls,respectively.
In summary, an elevated lung cancer risk in weld-ers was found in four of the studies described above.In the study conducted by Sjogren et al., the in-creased lung cancer incidence among welders wasindependent of tobacco use, but asbestos exposurecould not be ruled out (Refs. 158, 161, and 162).The studies by Sjogren et al. (Ref. 161) and Kjuus etal. (Ref. 83) suggested that stainless steel weldershave a higher risk of lung cancer than mild steelwelders. However, the number of stainless steelwelders with lung cancer was too small to allow de-finitive conclusions. An elevated risk for lung can-cer was also observed by Schoenberg et al. (Ref.152) in a mortality study conducted in metropoli-tan New Jersey. However, in this case, only ship-yard welders had an elevated lung cancer risk, eventhough the study controlled for asbestos exposure.In contrast, Lerchen (Ref. 90) observed an elevatedlung cancer risk among welders even when shipyardwelders were excluded. Neither Pearce (Ref. 136)
nor Puntoni (Ref. 141) observed an elevated lungcancer risk among welders in the general populationof New Zealand or in Italian dockyard workers, re-spectively. However, Pearce observed an elevatedrisk for stomach cancer while in Puntoni's study gaswelders, but not arc welders, had an elevated risk ofdigestive tract cancer.
An elevated risk for leukemia was noted in a ship-yard study conducted by F. B. Stern (Ref. 169). Incontrast, R. M. Stern found no association betweenwelding exposures and leukemia in his analysis of15 epidemiologic studies which considered the inci-dence of leukemia in welders (Ref. 170). Similarly,an elevated risk for brain tumors was found by Mc-Laughlin in a population-based study in Sweden(Ref. 97), but no brain tumors were identified in adeath-certificate study by Thomas conducted in theU. S. (Ref. 175), or in a case-control study con-ducted in Sweden by Olin et al. (Ref. 123). Onestudy found an elevated risk for renal tumors inwelders (Ref. 161). The discrepancies between re-sults of studies that appeared during the current re-port period do not help clarify the actual cancer riskof welders, especially at sites other than the respira-tory tract.
8.2 Metal Carcinogens. A major argument sup-porting the probability of an increased cancer riskin stainless steel welders is the presence of nickeland hexavalent chromium in fumes from somewelding processes. The carcinogenicity of chro-mium and other metals was recently reviewed byNorseth (Refs. 110 and 111). According to Norseth,the epidemiologic evidence for an increased risk ofcancer in workers exposed to arsenic, chromium,and nickel in some industries (other than welding)is extensive, whereas that for exposure to berylliumand cadmium is more limited. The epidemiologicdata receive some support from tests in animals andin vitro systems, but the mechanism of action ofmetal carcinogens remains unknown and is an im-portant area for further research.
Factors such as solubility, particle size, crystal-line structure, chemical species, and surface chargemay be important in determining the biological ac-tivity of metals. Generally, only slightly solublecompounds of nickel (e.g., sulfides and oxides) or ofchromium are thought to cause cancer. Norseth(Ref. 110) suggested that the solubility of these com-pounds may be important because compounds withlimited solubility would be more readily incorpo-rated into cells by phagocytosis than would themore soluble compounds. Once inside the cells,some particulate material could gradually becomesolubilized, bringing about a continuous interac-
34
tion of the metal with the cells and possibly achronic interference with normal cellular processes.
Norseth cited limited evidence implicating triva-lent chromium in the carcinogenic process and sug-gested that intracellular reduction of hexavalent totrivalent chromium may be important to the mech-anism of chromium carcinogenicity. He furtherpostulated that the activity of a short-lived interme-diate, perhaps in the pentavalent state, may be re-sponsible for the carcinogenic or transformingactivity of chromium.
According to Norseth, the experimental evidenceindicates that all nickel and hexavalent chromiumcompounds should be considered to be carcino-genic and should be treated as such by regulatoryagencies (Ref. 111). This view is disputed by Mas-tromatteo (Ref. 95) who noted that, in the recenthealth effects document for nickel (Ref. 181), the U.S. Environmental Protection Agency concludedthat only the following three forms of nickel shouldbe classified as known or probable human carcino-gens: nickel refinery dust (from pyrometallurgicalsulfide nickel matte refineries), nickel subsulfide,and nickel carbonyl. Mastromatteo stressed that acancer risk is associated with specific nickel speciesand indicated that air sampling methods should de-termine not only the total amount of nickel, but alsothe species present.
8.3 Screening for Cancer. Because respiratorytract cancer is generally diagnosed late in develop-ment, the death rate is generally high for this dis-ease. Thus, reliable and noninvasive screening testsfor detection of early stages of lung cancer havebeen sought for use in industries in which the cancerrate may be elevated. Two methods, chest X-raysand sputum cytology, are currently available forscreening large groups of workers for lung cancer.Chest X-rays are less useful for detecting early tu-mors than sputum cytology which can distinguishbetween precancerous and cancerous lesions. Ac-cording to Kilburn (Ref. 80), sputum exams shouldbe performed every four months to be effective. Un-fortunately, sputum samples do not generally con-tain cells from the more distal parts of therespiratory tract and so some lung tumors can bemissed with this technique.
Methods are currently available for screeningworkers at risk for development of nasal cancer byperforming biopsy of the nasal mucosa. However,obtaining tissue samples can be painful, and prene-oplastic lesions may be missed since they tend to bepresent as small foci. To overcome these problems,Reith developed a cytologic screening method inwhich a more representative population of cells is
less painfully obtained by brushing the surface ofthe middle nasal turbinate. In a study of nickel re-finery workers with known nasal dysplasia, Reithshowed that more preneoplastic nasal lesions couldbe detected by cytologic screening than by biopsy(Ref. 144).
In an effort to identify a substance in sputum thatcould serve as a marker for cancer risk, Ebeid stud-ied the relationship between thiocyanate levels insputum and welding or smoking, or both. Sputumsamples were obtained from 60 persons employedin a metal forming factory. Thirty-two subjects werewelders and the remainder were office workers whoserved as controls. Ten welders and sixteen officeworkers were nonsmokers. Mean sputum thiocya-nate concentrations in the nonsmoking welders andoffice workers were 27.77 and 32.92 ug/ml, respec-tively, while concentrations in the sputum of smok-ing welders and office workers were 59.27 and110.42 ug/ml, respectively. Ebeid concluded thatexposure to welding fumes may decrease the thiocy-anate content of sputum, while smokers have ele-vated levels relative to nonsmokers. He suggestedthat determining the thiocyanate content of sputummay be of value as an index of certain exposures inthe work environment. However, it would appearthat, in the absence of further information, thiocya-nate levels do not promise much utility for estima-tion of disease risk in welders (Ref. 37).
9. Metal Fume Fever
Metal fume fever is a flu-like disease caused byinhalation of oxides of various metals includingzinc, copper, aluminum, antimony, iron, manga-nese, nickel, and cadmium. The symptoms, whichinclude fever, chills, general malaise, joint pains,cough, sore throat, chest tightness, and fatigue, usu-ally appear 4 to 12 hours following exposure andlast from 1 to 2 days. A short-lived tolerance tometal fume fever can develop, which is lost during ashort absence from exposure. The name "Mondayfever" has been conferred because symptoms mayrecur on return to work after a weekend or holiday.McMillan suggested that the tolerance is a practical,rather than a physiological phenomenon resultingfrom the exercise of less caution when welding aftera period of absence from work. Conversely, after about of the disease, the welder may more carefullyavoid fume exposure or his supervisor may bemore likely to place him in a better ventilated area(Ref. 100).
The pathogenesis of metal fume fever is poorlyunderstood. It is frequently attributed to an im-
35
mune reaction to metals, resulting in inflammationof respiratory tract tissue and release of histamineor histamine-like substances. Vogelmeier et al. (Ref.186) observed changes in pulmonary function andmarked pulmonary inflammation in a 26 year oldmale welder who had recurring symptoms of metalfume fever after welding zinc-coated materials butnot other metals. When the welder was asymptom-atic, clinical findings, chest X-ray, and spirometricdata were all within normal limits, but inhalationchallenge with methacholine revealed a low gradenonspecific bronchial hyperreactivity.
Two tests were conducted in which the subjectwelded a zinc-coated tube for 1 hour under con-trolled conditions. Airway resistance increased,while inspiratory vital capacity, single-breath trans-fer factor, and arterial oxygen partial pressure weresharply reduced immediately after welding. Chills,fever, and an increased peripheral blood leukocytecount occurred 2 to 6 hours later.
A second test, performed after an exposure-free 6-month period, yielded similar results. Bronchoalve-olar lavage was performed 24 hours after the secondcontrolled welding exposure. The total number ofcells was ten times greater than normal, and thenumber of polymorphonuclear leukocytes wasmarkedly increased in lavage fluid. The cell popula-tion in bronchoalveolar lavage fluid was normal af-ter a 7-week exposure-free period. Thus, exposureto zinc fumes produced a pronounced pulmonaryreaction, with a severe acute inflammation of pe-ripheral bronchoalveolar structures. The investiga-tors remarked that metal fume fever has not, thusfar, been known to cause pulmonary fibrosis; theyquestioned whether the marked increase of leuko-cytes can really be harmless to pulmonary tissues.This question is similar to that raised by Mylius andGullvag (see above) who observed a marked excessof alveolar macrophages in welders who smoke(Ref. 104).
A case of metal fume fever with a severe concomi-tant allergic reaction to zinc fumes was reported byFarrell (Ref. 40). After welding galvanized metal, a34 year old male patient experienced classic symp-toms of metal fume fever associated with an imme-diate anaphylactoid reaction and a delayed reactionconsisting of angioedema (swelling of the face, lips,and throat) and urticaria (an allergic skin reactionwith hives and intense itching). The symptoms re-curred after further exposures to zinc fumes eventhough a protective face mask was used. To confirmthat the anaphylactic reaction was associated withexposure to zinc fumes, the subject welded galva-nized steel under medical supervision. Four hourslater he developed symptoms of metal fume fever.
As on previous occasions, angioedema and urti-caria developed 12 hours after exposure. Theseclinical symptoms were suggestive of an immediateand late immunoglobulin-E-mediated immunologi-cal reaction to zinc fumes. Although metal fume fe-ver is thought by some investigators to have animmunological basis, Farrell does not believe that adirect association between immediate and late ana-phylactoid reactions and metal fume fever can befully substantiated by this case.
Shusterman reported the case of a man who expe-rienced signs of metal fume fever and injury to myo-cardial and skeletal muscle after welding galvanizedmetal for 4 successive days. Welding was performedoutdoors with no respiratory protection. On 2 of the4 days, he ran for 2 to 3 miles after the completion ofwork. After the fourth day, he had chest pains andwas hospitalized for 10 days. Clinical tests indi-cated that there might have been injury to skeletalmuscle as well as the heart. The authors suggestedthe possibility of a causal relationship between in-halation of zinc oxide fumes and muscle injury(Ref. 155). However, this would be difficult to sub-stantiate on the basis of the data presented in thiscase.
10. Effects on the Ear and Hearing
The case of a welder who became permanentlydeaf in one ear as a result of a metal spark burningthe middle and inner ear was reported by Stage andVinding (Ref. 167). The incident occurred while hewas lying on his side welding underneath a car. Hewas wearing a welding mask which afforded no pro-tection to the ears.
11. Effects on the Eye and Vision
Exposure to electromagnetic radiation in the IR,visible, or UV range generated by arc welding caninjure the eye. Shortwave UV light (270 to 290 na-nometers) is absorbed by the outer layers of the eye;this can result in keratoconjunctivitis (arc eye,welder's flash, photokeratitis, or photophthalmia),and pterygia (membranous growths which extendacross the outer eye from the conjunctiva to the cor-nea). Near UV light, in the range 295 to 400 nm,penetrates more deeply into the eye where it candamage the corneal epithelium and possibly thelens. Infrared radiation of longer wavelengths maycause thermal damage to the cornea and aqueoushumor and has also been associated with the devel-opment of lenticular cataracts. Brief and intense
36
exposure to blue light in the visible range (approxi-mately 400-500 nm) or to near IR radiation mayproduce photochemical lesions on the retina. Thepotential effects of welding and electromagnetic ra-diation on the eye were reviewed by McMillan(Refs. 69 and 99).
Keratoconjunctivitis, a marked inflammation ofthe cornea, is the eye condition most frequently en-countered by welders. It is characterized by blurredvision, tearing, acute pain, and headache. Symp-toms normally last up to 2 days and there are usu-ally no sequelae. Most pharmaceutical preparationsthat provide relief of symptoms require prescrip-tions and close medical monitoring because theycan produce permanent corneal damage and loss ofvisual acuity. Nonprescription drugs are available,but they are generally less effective and require fre-quent application. An investigation of the efficacyof nonprescription drugs was conducted byWittgens (Ref. 194). Over-the-counter preparationsfor the treatment of keratoconjunctivitis were dis-tributed to welders through occupational physi-cians and other medical professionals in theworkplace. A survey of the welders who used theproducts indicated that an eyedrop preparation(Chibro-Uvelin), developed in France in 1953, wasthe most satisfactory of the products tested. It pro-vided fast relief from pain and return of visual acu-ity. Unlike some other drugs tested, it was effectivefor a long time and did not require repeated applica-tion. According to Wittgens, the effects of this prep-aration have been thoroughly researched, and it hasno known side effects.
Sharir and Ben-David observed severe and per-sistent symptoms of keratoconjunctivitis amongfour workers who performed plasma arc thermalspray coating (Ref. 153). Excessive tearing, foreign-body sensation, and dull pain persisted for severalmonths. Two of the patients had mild to moderatesigns of nonspecific irritation, while the other twopatients had severe acute inflammation with pro-nounced swelling of the conjunctiva and eyelids.The disorders were attributed to exposure to thehigh levels of UV and visible radiation generated byplasma arc thermal spray coating.
While the retina is the part of the eye most sensi-tive to UV radiation, injury to the retina from expo-sure to the welding arc is rare and usually resultsfrom improper eye protection (Ref. 99). A case of a22 year old man who developed retinal injury withsubstantial bilateral vision loss soon after weldingwithout appropriate eye protection was reported byUniat et al. (Ref. 181). Some visual loss and signs ofretinal injury were still present 6 months after theincident. Cellini et al. described a similar case of
photic maculopathy in a young worker who wassaid to have been wearing "protective eyeglasses"while welding (Ref. 20). Visual loss was noted soonafter exposure to the welding arc; the progressionof the disorder followed a favorable course withtreatment.
An unusual case of retinal damage in a man with alens implant was reported by Turut et al. (Ref. 179).The subject was a 36 year old welder in whom a cat-aractous lens had been replaced by a lens implant.Four years after the surgery, an incident in which hewelded without wearing eye protection was fol-lowed immediately by loss of visual acuity. Exami-nation revealed a small lesion, apparently caused byradiation from the arc, on the retinal fovea of theeye with the lens implant. There were no indica-tions of damage or loss of vision in the eye with thenormal lens. The investigators noted that the nor-mal lens provides some protection to the retina byfiltration of potentially harmful IR and visible radi-ation. Persons with lens implants are more sensitiveto the effects of radiation and should use extremecaution when performing operations, such aswelding, which generate intense electromagneticradiation.
Retinitis pigmentosa is a hereditary condition in-volving progressive atrophy of the retina. The firstsymptom of the disease is night blindness (reducedability to adapt to low intensity light). As the diseaseprogresses, affected persons have increasing diffi-culty adjusting to bright light. Two welders with ret-initis pigmentosa sine pigmento were described inthe Yugoslavian literature by Krizmanic-Mrkoci etal. (Ref. 86). After two years of welding, one weldercomplained of vision problems and photophobia(abnormal sensitivity to light). The problems weregreatest when he left the plant at the end of the workday. Adaptometry and electroretinography indi-cated that he, and a second welder with similar vis-ual difficulties, had retinitis pigmentosa sinepigmento. This condition makes it difficult forwelders to use protective eye wear during work andaffects their ability to adapt to bright sunlight uponleaving the workplace. Krizmanic-Mrkoci recom-mended that peripheral vision tests, which detectretinitis pigmentosa, be included in regular screen-ing exams given to all welders before employment.
The debate persists in the open literature on thepotential hazards of wearing contact lenses duringwelding (Ref. 19). Contact lenses are generally con-sidered to be inadvisable in situations or environ-ments where there may be exposure to chemicalfumes, vapors, splashes, intense heat, or high con-centrations of particulates. Most authorities dis-count reports that radiation emitted by the welding
37
arc can cause adhesion of the contact lens to the cor-nea. In fact, some experts believe that, with the useof appropriate eye safety equipment, there may beadvantages to wearing contact lenses in hazardousindustrial settings. Callier (Ref. 19) would encour-age persons who regularly use contact lenses to wearthem in hazardous industrial situations becausethey frequently see better with them than with eye-glasses.
Randolph and Zavon (Ref. 142) reviewed the lit-erature on contact lenses and found that contactlens wearers exposed to high levels of atmosphericparticulates do not have excess subjective or objec-tive symptoms of eye damage. Contact lenses do notworsen the corneal damage resulting from exposureto chemicals and in some cases may provide protec-tion for the eye. In fact, contact lenses may be bene-ficial because they allow a closer fitting ofprotective eye wear. The authors surveyed the cor-porate policies among 67 large American firms andfound that 86% had some restrictions on the use ofcontact lenses. While most of the restrictions in-volved areas where irritant chemicals were present,52% of the companies restricted contact lens use inwelding operations. Analysis of the survey re-sponses indicated that most corporate policies werebased on perceived risk rather than on facts re-vealed by recent research.
12. Effects on the Nervous System
Chronic exposure to manganese can cause a pro-gressive disease with neurological and psychologi-cal manifestations. Early symptoms include apathy,anorexia, spasm, and irritability. Signs of psychosismay develop with euphoria, impulsive behavior,confusion, and aggression. Advanced stages resem-ble Parkinson's disease, with muscle weakness,muscle rigidity, tremors, and impaired gait. Twocases of welders who had advanced stages of manga-nese poisoning with symptoms of Parkinsonismwere reported by Rasmussen and Jepsen (Ref. 143).Both welders had performed SMAW in a boiler fac-tory, one for 31 years and the other for 17 years.Hygienic conditions in the factory were poor andthe authors presumed that exposure to manganesewas high.
Rudell (Ref. 147) reported two cases of welderswith neurological symptoms. The first was a 29 yearold man who, after 8 years employment in SMAWof mild steel, became dizzy when welding. The sec-ond was a 43 year old man employed in GTAW ofNimonic alloys (75% nickel, 25% chromium) for 14years. He reported dropping objects and had diffuse
neurological symptoms in his left side. Physicalexamination revealed no other neurologicaldisorders.
13. Effects on the Skin
A case of a welder with recurrent, severe facialdermatitis was described by Shehade et al. (Ref.154). Allergic dermatitis was ruled out with skinpatch testing, using a standard battery of allergens,preservatives, potential sensitizers, photoallergens,a welding antispatter formulation, and weldingflux. The case was diagnosed as photodermatitis,caused by UV exposure during welding.
The affected welder performed spot and seamwelding with CO2-shielded GMAW of mild steel.During spot welding, he seldom wore a protectiveface mask and, instead, turned his face to the left toprotect his eyes from the arc. His UV exposure wasmeasured with polysulfone badges; 20 badges wereused over a 3-week period. In all cases, the mea-sured exposure was equal to or greater than themaximum permissible exposure (MPE) for an 8-hour day, established by the ACGIH in 1976. In theworst case, the exposure was 128 times the MPE.When the badge was placed underneath the protec-tive face mask, the MPE was still exceeded by 4 to 9times.
14. Sensitivity to Fume Components
Allergic contact dermatitis in response to chro-mate salts is well documented among welders, butanaphylactoid reactions are rare. A case of a welderwho developed severe systemic reactions after ex-posure to chromates was reported by Moller et al.(Ref. 101). The welder had 10 years' exposure to va-pors from chromium trioxide baths and stainlesssteel welding fumes. Symptoms first appeared afterseveral hours exposure to chromium trioxide mistduring an accidental discharge from an adjacentchrome-plating plant. At this time, he developedswelling and a diffuse pruritic papular rash diag-nosed as urticaria. This reaction to chromium oc-curred twice more over the next 7 months. Withtime, he also experienced difficulty breathing andchest tightness. Challenge with 29 ug/m3 sodiumchromate aerosol induced a delayed anaphylactoidresponse with urticaria, angioedema, severe bron-chospasm, and a threefold increase in plasma hista-mine levels. Skin tests and in vitro immunologicaltests indicated that this was probably not a classicIgE-mediated allergic response, but rather may
38
have been a cell-mediated response involving sensi-tized lymphocytes.
Nordman et al. (Ref. 108) examined 58 welderswho had asthmatic reactions when welding stain-less steel. Of these welders, 13 had significant re-duction in pulmonary function following welding ofstainless steel but not mild steel. Seven of the 13welders responded positively to the histamine prov-ocation test prior to welding.
Exposure to high levels of soluble beryllium com-pounds can cause acute berylliosis characterized byinflammation of the nasopharynx, trachea, andbronchi with possible development of pulmonaryedema. Repeated exposure to low levels of slightlysoluble beryllium compounds may induce chronicberylliosis. This condition is often thought to havean immunologic basis and is typified by a distinc-tive granulomatous process in the lung which maynot become apparent until months to years after ex-posure ceases. Because of similarities in pathologicprocesses in the lung, chronic beryllium disease isreadily mistaken for sarcoidosis. Cullen et al. exam-ined five precious metal refinery workers withsymptoms of chronic berylliosis and demonstratedthat the etiology of the disease could be confirmedby an in vitro immunologic test (Refs. 27 and 28). Inthis test, alveolar lymphocytes from persons withberylliosis, but not from normal controls, undergodivision when exposed to beryllium. The investiga-tors concluded that their results provide furtherevidence for an immunologic basis for chronic be-ryllium disease.
15. Effects on the MusculoskeletalSystem
Awkward work postures of the trunk and shoul-der may cause fatigue and musculoskeletal disor-ders. For example, prolonged elevation of the arms(shoulder flexion or abduction) can produce ex-treme muscle fatigue and, in some cases, acutetendonitis. Evaluation of postural stress in theworkplace is difficult because a continuous recordof the time spent in various postures during a workoperation must be obtained and analyzed. To iden-tify the causes of postural stress, Keyserling devel-oped a computer-aided system to analyze workingpostures of the trunk and shoulders on repetitiveassembly line jobs (Ref. 78). With this system, ananalyst watches a continuous videotape record ofwork operation and enters a symbol into the com-puter for each assumed posture. The time durationof each posture is automatically measured by thecomputer which generates a posture profile for each
joint (trunk, right shoulder, left shoulder). The pos-ture profile is then used to determine whether thereis an excessive frequency of posture change orwhether non-neutral postures must be maintainedfor excessive time periods. The system was tested ata manual spot welding operation in an automotiveassembly plant. Based on the results of the com-puter-aided analysis, work stations were redesignedto eliminate awkward postures and to incorporateseveral new safety features.
A system for analyzing working posture, devel-oped by the Finnish steel industry, was tested byHeinsalmi (Ref. 55). A positive relationship wasfound between poor working postures and sickleave due to musculoskeletal disorders, especiallylower back pain. Sick leave due to neck and shoul-der disorders also correlated with poor arm pos-tures. A prototype wooden manikin was used tostudy work postures in welding operations. A weld-ing unit, developed on the basis of that study, im-proved welding posture and enabled operationaltime savings.
Bjelle et al. (Ref. 15) examined the influence ofergonomic factors on shoulder and neck complaintsin 26 Swedish industrial workers who performed as-sembly, pressing, or welding during the manufac-ture of truck cabs. Arm movements during workwere filmed, and shoulder strain was estimated bymeasuring the duration and frequency of shoulderabduction or forward flexion. Over a 12-month pe-riod, two thirds of the workers complained of mus-culoskeletal injuries; nine of these were shoulderand neck complaints. Tests of muscle strengthshowed that the strength of the shoulder muscles,but not elbow flexion or hand grip measurements,was significantly lower in workers with shoulder-neck complaints. The authors concluded that sinceshoulder muscles, but not elbow or hand muscles,were weaker in workers with shoulder-neck com-plaints, shoulder muscle weakness may be a second-ary phenomenon rather than a reflection of apredisposition to shoulder and neck disorders.
16. Effects on the Urogenital Tract
Chronic exposure to low concentrations ofcadmium fumes can cause kidney damage, whichis first manifested by urinary excretion of lowmolecular weight serum proteins, such as beta-2-microglobulin and retinol-binding protein. Theproteinuria is generally thought to result from dam-age to the kidney tubules. Whether or not cadmiumalso affects glomerular function remains unre-solved.
39
Smith et al. (Ref. 164) examined the cadmiumbody burden and renal function in 53 solderersfrom five firms, who had up to 25 years of uncon-trolled exposure to fumes from cadmium-contain-ing alloys. Elevated cadmium levels in the kidneyand liver were found in 22 of 25 subjects, most ofwhom had more than 5 years' exposure to cadmiumsolders. Cadmium concentrations were elevated inblood and urine in all 32 workers with more than 5years' exposure, and in 11 of 21 workers with lessthan 5 years' exposure. Low molecular weight pro-teins, typical of tubular dysfunction, were presentin urine from 20 of 51 subjects tested, while albu-min, characteristic of glomerular dysfunction, wasfound in only 3 of 51 subjects. No significantchanges in urinary cadmium or beta-2-microglobu-lin were found in 19 workers followed for 2 yearsafter cadmium-free soldering rods were substitutedfor cadmium-containing soldering rods. The latterobservations suggested that mild tubular protein-uria does not progress to more severe kidney dys-function in the absence of further accumulation ofcadmium within the body.
Verschoor et al. (Ref. 185) compared urinary pro-tein and blood and urinary cadmium concentra-tions in 26 cadmium-exposed workers and eightnonexposed controls. Nineteen of the cases were ex-posed to cadmium in plastic pigments and the re-mainder were welders in a radiator shop. The meanvalues of urinary retinol-binding protein and uri-nary and serum beta-2-microglobulin were signifi-cantly higher in the cadmium-exposed welders thanin controls, while total urinary protein and urinaryalbumin did not differ. Thus, signs of tubular dys-function, but not glomerular dysfunction, werepresent in this group of cadmium-exposed welders.
Prompted by observations of five patients withglomerulonephritis or interstitial nephritis at theUniversity Hospital in Umea, Hagberg et al. con-ducted a negative case-referent study to determineif there is an association between chronic renal dis-ease and welding. The study group consisted of 143living males who had undergone kidney biopsies asa result of suspected or clinically established glo-merulonephritis or obscure renal disease. Occupa-tional and vocational exposures were comparedwith age- and residence-matched controls chosenfrom the general population. No relationship be-tween welding and the incidence of severe chronickidney disease was found. In addition, there was noincrease in the risk of death from nephritis amongwelders. (Ref. 51) In this study, no distinction wasmade between welders who had cadmium expo-sures and those who did not.
Similarly, no association between renal impair-
ments and stainless steel welding was found in across-sectional field study performed by Zschiescheet al. (Ref. 196). Kidney function, assessed by mea-surement of total protein and specific enzymes inurine, was compared in 118 stainless steel weldersand an equal number of age-matched controls.Breathing zone measurements during welding re-vealed substantial exposures to nickel and chro-mium during welding. The concentrations of nickeland chromium were significantly higher in urinefrom welders than from controls. However, the re-sults of kidney function tests did not differ betweenwelders and controls or among welders using differ-ent welding processes. The authors noted that, al-though urinary chromium and nickel levels wereelevated in the welders, they were not excessivelyhigh. They recommended that additional studies beconducted to examine these parameters in welderswith greater exposures to nickel and chromium.
17. Effects on the Teeth and OralCavity
Underwater welders frequently sense a metallictaste while welding and have a higher than normalreplacement rate of dental amalgam (Ref. 133). Tofurther examine the association between changes indental amalgam and underwater welding, Ortendaland Holland measured the intraoral electrical activ-ity while divers were welding underwater (Refs.127, 130, and 131). They noted that during under-water welding or cutting, the electrical activity inthe mouth is normally too low to cause a metallictaste or corrosion of dental amalgam (Ref. 130).During three test dives, leaks in the dry suit alloweda small amount of water to enter the gloves whichresulted in an increase in the intraoral electrical ac-tivity. In one such instance, a diver experienced ametallic taste. This could not be explained by dete-rioration of dental amalgam since the electrical ac-tivity was theoretically too low to damage theamalgam. Further studies with divers showed thatthe metallic taste sensation was unlikely to be aresult of magnetic field effects (Refs. 127, 128, and129).
An in vitro system was designed in which expo-sure of amalgams to magnetic fields simulated theenvironment of divers when using a mean cuttingcurrent of 65OA DC (Refs. 127 and 129). Studieswith this system demonstrated that a slight rough-ening of amalgam surfaces could result from mag-netic field exposures. However, chemical changesin the amalgam were insufficient to account for ametallic taste sensation. Thus, neither in vivo nor in
40
vitro studies supported the view that this subjectivesymptom is related to magnetic field effects. Basedon an earlier report (Ref. 107), the investigatorspostulated that the metallic taste resulted from elec-trical stimulation of the taste buds.
18. Effects on the Voice
Voice quality was compared in eight shipyardwelders and eight office workers; none of the parti-cipants smoked. A questionnaire indicated thatvoice and throat problems were more frequentamong welders than office workers; tests indicatedthat the welders' voices were more hyperfunctionaland unstable. The investigators concluded thatvoice problems among welders are due to high noiselevels which cause a strain on the larynx during vo-calization. The welders thought their voiceproblems to be due to fume exposures (Ref. 112).
19. Biological Monitoring
19.1 Aluminum. A study of aluminum excretionamong 23 GMA and GTA welders by Sjogren et al.revealed that aluminum levels in urine are depen-dent on the most recent daily exposures, as well ason the total duration of exposure (Refs. 159 and160). Aluminum levels, measured in the breathingzone of 16 aluminum welders at the start of thestudy indicated that most exposures were greaterthan 1 mg/m3. Aluminum concentrations were de-termined in urine from 16 welders before and after a16 to 37 day exposure-free vacation period. Biologi-cal half-times for urinary aluminum were calcu-lated from samples collected from eight of thewelders in the middle of the exposure-free period.The half-time for welders with less than 1 year expo-sure was about 9 days, whereas it was 6 months ormore in welders who had been exposed to alumi-num fumes for more than 10 years. Half-times foraluminum calculated from urine collected at theend of the exposure-free period were 27 days forwelders with less than 1 year's experience and morethan 2 years for the more experienced welders.
The investigators remarked that, even thoughthere was a wide variation in urinary aluminumlevels among individuals, the large difference in uri-nary half-times between more and less experiencedwelders indicated that aluminum is probably storedin two or more compartments in the body, withwidely different clearance rates. They speculatedthat the compartment with the slow clearance rateprobably includes the lungs and the bones.
19.2 Manganese. Zschiesche et al. determinedlevels of manganese in the breathing zone and bodyfluids of eight welders from two factories whoworked with filler materials containing about 0.8%manganese (Ref. 197). SMAW was the predominantwelding process, but GMAW was also used. Manga-nese was measured in air samples collected for 2hours each day during the work shift for a period of1 week and in urine collected 3 times daily (before,during, and after the work shift) for 1 full week andthe following Monday. Blood samples were col-lected before and after each work shift during thesame time period.
Median breathing zone concentrations were 11.5mg/m3 for total fume and 257.7 ug/m3 for manga-nese. Even though fume levels tended to be high,only one sample had a manganese concentration inexcess of the German MAK of 5 mg/m3. Manganeselevels in blood and urine tended to be low and, forthe most part, did not vary substantially from val-ues obtained for the general population. There wasno correlation between external exposures and in-ternal manganese levels, or with the time of day orthe day of the week that the biological samples werecollected. It was concluded that, under the workingconditions at the two plants, biological monitoringof manganese is not necessary.
19.3 Chromium. Because hexavalent chromiumpasses more readily through cell membranes, triva-lent chromium tends to be transported in the bloodby plasma while hexavalent chromium is trans-ported mainly in erythrocytes (Ref. 3). Two researchgroups examined whether this difference may beused in biological monitoring to distinguish be-tween exposures to trivalent and hexavalent chro-mium. Angerer et al. (Ref. 4) determined chromiumlevels in erythrocytes, plasma and urine obtainedfrom 103 male welders who performed SMAW orGMAW, or both, of stainless steel. Median breath-ing zone concentrations of chromium were 4 ug/m3
for SMAW and 10 ug/m3 for GMAW.The median chromium concentrations in eryth-
rocytes, plasma, and urine of all welders were lessthan 0.60, 9.0 and 32.5 ug/1, respectively. The con-centrations of chromium in erythrocytes, plasma,and urine for individual welding processes areshown in Table 3. In all cases, levels were higher inwelders performing GMAW, or a mixture ofGMAW and SMAW, than in those performingSMAW. Urinary chromium levels among the weld-ers ranged from 5.40 to 229.4 ug/1, which is 5 to 200times greater than average levels in the general pop-ulation.
The correlation between chromium concentra-
41
tions in erythrocytes, plasma, and urine was highlysignificant (p<0.0001). Chromium levels in eryth-rocytes, plasma, and urine of 0.6, 10 and 40 ug/1,respectively, corresponded to an external chro-mium trioxide exposure of 100 ug/m3. Erythrocytechromium concentrations were close to the limit ofdetection and were thus too low to provide accurateestimates of exposure below airborne concentra-tions of 100 ug/m3 (the German "technical guide-line concentration" for chromium trioxide). It wasconcluded that plasma or urine levels are more suit-able for monitoring low level chromium exposures.At higher exposures, the erythrocyte concentrationsmay be useful as a measure of hexavalent chromiumexposures; simultaneous determination of chro-mium levels in plasma and erythrocytes may give auseful picture of both overall exposure to chro-mium and exposure to hexavalent chromium.
Conflicting conclusions were drawn from a sim-ilar, but smaller, study conducted by Gustavssonand Welinder (Ref. 50). Blood plasma, erythrocyte,and urinary chromium concentrations were mea-sured in samples collected from four stainless steelSMA welders on a Thursday morning before work,on Friday before and after work, on the followingTuesday, and after 31 days of vacation. A decreasein chromium levels was seen in all fluids during thevacation period and there was a significant correla-tion between chromium levels in the various fluidsat each time point. Because there were no obviousdifferences in clearance rates of erythrocyte andplasma chromium, it was concluded that determi-nation of blood or erythrocyte chromium levelswould not add important information to thatgained from biological monitoring with urinarychromium concentrations.
19.4 Analytical Methods. Methods for determin-ing low concentrations of metals in biological fluidsfor use in biological monitoring were described bythree groups of investigators (Refs. 22, 54, and 79).Heinrich-Ramm and Angerer described a methodfor determining concentrations of nickel and chro-mium in urine, plasma, and erythrocytes (Ref. 54).Christensen and Kirchoff (Ref. 22) developed amethod using electrothermal atomic absorptionspectroscopy (AAS) for analysis of chromium,nickel, and cadmium in blood. Kiilunen et al. (Ref.79) developed a method using AAS for determininglow concentrations of nickel and chromium inurine. Using this method, they found a mean nickelconcentration of 70 nmol/1 in urine from 299 Finn-ish persons who had no known occupational expo-sures to nickel. A mean value of 1.5 nmol/1 (0.078
ug/1) was determined for urinary chromium concen-trations using a study population of 155 persons.
Section ThreeInvestigations in Animals
and Cell Cultures
20. Animal Studies
20.1 Inflammation and Fibrosis. The developmentof practical methods for evaluating the fibrogenic-ity of fumes and gases from welding operations isimportant for the assessment of risk associated withdifferent electrodes and welding methods. Hicks etal. used the inflammatory response of rat andguinea pig muscle tissue (Ref. 62) to compare theirritant and fibrogenic properties of several differ-ent welding fumes. Intramuscular injection of parti-cles from SMAW of stainless steel caused severelocal inflammation and fibrosis and was cytotoxicto muscle fibers. Fumes from SMAW of mild steelcaused similar, but less severe effects. Fumes fromGMAW of stainless steel caused only mild irritationand fibrosis and were not cytotoxic.
The effects on rats of single 6-hour inhalation ex-posures to fumes from six welding processes (Table4) were investigated by the American Welding Soci-ety (Ref. 23). The major histopathologic changes re-sulting from fume exposure were seen in therespiratory tract. Fumes from electrode E308-16produced the most severe effects, causing pulmo-nary edema and pneumonitis. Electrodes E7018,E6010, and E70T-1 caused upper respiratory tractirritation while the remaining two produced no ap-parent changes in the respiratory tract (Table 4). Ac-companying these effects were changes in thenumber of macrophages that could be retrievedfrom the lungs by lavage. Decreased phagocytic ac-tivity was observed in macrophages from all test an-imals except those treated with fumes from E5356.Extracts of urine were not genotoxic in the Salmo-nella/Ames test nor were there any chromosomeaberrations in bone marrow cells recovered fromanimals 20 hours after exposure.
Alveolar macrophages from untreated rats, wereincubated with eight concentrations of weldingfume particulates. Viability, measured in terms ofthe concentration of fumes that killed 50% of thecells (EC50), was lowest with fumes from electrodeE308-16. On the basis of histologic assessment ofthe respiratory tract and in vitro assays of macro-phage activity, the electrodes were classified into
42
ChromiumWelders
WeldingProcess
SMAW(n=39)
GMAW(n=14)
SMAW and GMAW(n=50)
Total(n=130)
Table 3Concentrations in Erythrocytes, Plasma andUsing SMAW or GMAW, or both, of Stainless
Chromium Concentration (ug/1)
EiyjhrocytesX + SD(median)
1.18 + 1.35(<0.6)
1.44+1.72(<0.6)
2.51 +5.82(<0.6)
1.86 + 4.21(<0.6)
PlasmaX + SD(median)
9.0 + 7.81(6.4)
13.56+ 12.34(9.6)
14.70+ 10.92(12.25)
12.39+10.33(9.00)
Urine ofSteel
UrineX + SD(median)
35.14 + 40.11(22.85)
64.46 + 65.31(37.60)
59.67 + 53.63(40.0)
51.19 + 57.75(32.5)
Data from Angerer et al., Ref. 4
three categories: highly toxic (E3O8-16), moderatelytoxic (E7018, E6010, and E70T-1), and nontoxic(E70S-3 and E5356).
Wilmer et al. exposed rats for 14 days by inhala-tion to fumes generated by SMAW and GMAW ofstainless and mild steel and by SMAW of cast ironusing ten different electrodes (Ref. 188). Fumesfrom SMAW of stainless steel were the most toxic,causing severe dyspnea, some mortality, and reduc-tion in body weight gain. GMAW fumes, especiallythose from mild steel, were less toxic than SMAWfumes. They induced slight signs of respiratory dis-tress, increased lung weight and morphologicalchanges in the lungs, including bronchiolar con-striction, focal interstitial pneumonia and emphy-sema. Fumes from SMAW of mild steel were justslightly toxic, causing only a minimal increase inlung weight in males. The fumes from all electrodesexcept one used in SMAW of mild steel were cyto-toxic to alveolar macrophages. None of the fumescaused SCE or structural chromosome aberrationsin peripheral blood lymphocytes.
Gorban et al. exposed rats by intratracheal instil-lation to particles of magnesium, and three alumi-num/magnesium alloys. All particle types causedhistopathologic changes in the lungs. Alloys of mag-nesium and aluminum were more fibrogenic thanpure magnesium. It was concluded that the crystallattice of the alloy was important in determining itsfibrogenic properties (Ref. 44).
20.2 Clearance. Kalliomaki et al. compared ratesof excretion and pulmonary clearance of fumes
from GMAW and SMAW of mild and stainless steel(Refs. 70, 71, and 73). Fume samples were adminis-tered to rats by intratracheal instillation. Concen-trations of nickel, iron, and chromium wereperiodically determined in the lungs, urine and fe-ces from serially sacrificed rats. About 10% of theadministered fumes from all welding processeswere recovered in the feces during the first threedays after exposure. The ratio of elements in the fe-ces was similar to that in the original fume sample,suggesting that material derived from the fume waseliminated in the feces as intact particles. Afterthree days, fecal excretion rates were similar to lungclearance rates for SMAW and GMAW of mildsteel.
Chromium and nickel from fumes generated bySMAW of stainless steel were excreted rapidly intothe urine during the first few days after instillation,and then at a much slower and gradually diminish-ing rate. The rapid urinary excretion during the firstday was attributed to material that became solubi-lized in the saline solution used as the vehicle forfume particles during the instillation procedure. Af-ter the first day, there was a marked difference inclearance rates of fumes from SMAW and GMAWof stainless steel. SMAW fumes cleared from thelungs with half-times of 73 days for iron, 53 days forchromium, and 49 days for nickel. In contrast, theconcentrations of GMAW fumes in the lung re-mained practically constant throughout the 2-month observation period following exposure (Fig-ure 5). The investigators concluded that a substan-tial part of the fumes from SMAW of stainless steel
43
Table 4Toxicity of Welding Fumes in Rats
Electrode
Shield gasWorkpiece
IN VIVO:Pathology /lungPathology/kidneyMacrophages:Deer, phagocytosisNumber of cellsin lavage fluid
S. typhimuriumChromosome aberration
IN VITRO:Macrophages:ViabilityDeer, phagocytosis*ATP*
E7OS-3
CO2Steel
--
+
incr--
highN.E.359
E308-16
Steel
+++-
+
incr--
very low134
E7018
Steel
+-
+
deer
-
low394
E6010
Steel
+-
+
?
-
-
low1089359
E70T-1
CO2Steel
++
+
deer
-
low118359
E5356
ArgonAluminum
--
N.E.
incr
-
high359
1089
? = Test not performedN.E. = No effect+++ = Severe response* Lowest fume concentration (ug/ml) that depressed phagocytosis or ATP levelsData from Coate, Ref. 23
are dissolved in the lung and eventually excretedinto urine, whereas fumes from GMAW are lessreadily dissolved and remain in the lungs for an ex-tended period of time.
Another study by the same research group fo-cused on clearance of manganese from the respira-tory tract (Ref. 77). Rats were exposed by inhalation(1 hour/day, 5 days/week for 1 to 4 weeks) to fumesfrom SMAW of stainless and mild steel and GMAWof stainless steel; the fumes contained 2.2, 2.8, and7.9% manganese, respectively. Manganese in fumesfrom SMAW of mild steel cleared rapidly from thelung during the first week after exposure (half-time= 0.5 day) and much more slowly thereafter (half-time = 5 days). Clearance of manganese from fumesgenerated by SMAW and GMAW of stainless steelwas much slower. Manganese from SMAW fumescleared from the lungs with a half-time of 5 daysduring the first day and a half-time of 40 days there-after. The half-time of manganese from GMAWfumes was 107 days throughout the observation pe-riod. The clearance of manganese from SMAWfumes paralleled the clearance of iron and the ratioof iron to manganese remained constant through-out the 106-day post-exposure observation period.
This suggested that manganese was cleared fromthe lungs as intact particles and that little manga-nese clearance resulted from in situ dissolution ofparticles.
Using transmission electron microscopy and X-ray microanalysis, Anttila demonstrated that themarked differences between pulmonary clearancerates of fumes generated by GMAW and SMAW ofstainless steel are related to differences in the solu-bility of particles in the lung (Refs. 6 and 7). Themorphology and chemical composition of particleswere examined periodically after exposure of rats towelding fumes by inhalation. Shortly after expo-sure, all particles were found in alveolar macro-phages and type-I epithelial cells; none were foundfree in the interstitium. Two types of SMAW parti-cles were seen in freshly collected fumes and in thelung. The major population consisted of irregularshaped particles, enriched with sodium and potas-sium; the minor population contained a higher per-centage of iron, nickel, and chromium. The majorparticle population changed with time, losing prac-tically all of its sodium, potassium, chromium, andalso some manganese. By 57 days, so much materialwas lost that the particles became transparent. In
44200 —i
1 0 0 -
5 0 -
2 0 -
5 0 -
1̂2 0 -
§10^o§ 5-
2 -
1 -
0 . 5 -
0.2
MMA/SS
[ I40 80
TIME ( DAYS)
53±5d
Ni
T1/2 = 49±6d
I120
1000 -,
500 -
^ 2 0 0 -
| 5 0 0 - ^
O
o20<Hz^ 100 -
50 -
MIG / SS
20
Fe
-* C = 500±100ng
Cr
240±50ng
T1/2 = 170±50d
r \40 80
TIME ( DAYS)
I120
Note: For elements whose levels were decreased in the lung during the 120 day observations period, CO = initialconcentration and Tl/2 = half time. The average lung content (C) is given for elements that underwent no apparentchange in concentration.
Figure 5 — Amounts of Iron, Chromium and Nickel in Rat Lungs After IntratrachealInstillation of Activated Fumes Generated by SMAW and GMAW of Stainless Steel
(MMA/SS and MIG/SS, respectively). Kalliomaki et al., Ref. 73
contrast, the minor population was quite stable andlost only sodium and potassium. The particles inGMAW fumes resembled the minor populationfrom SMAW fumes and, like those in SMAW fumes,showed no signs of dissolution during the 10-daystudy period. The rate of dissolution of the twopopulations of particles in the lungs reflected thedifferences in clearance rates described in the abovestudies. Thus, the solubility of particles may be animportant determinant of the pulmonary clearanceof particles from welding fumes.
20.3 Toxicity. A long-term inhalation study of theeffects of nickel oxide and arsenic trioxide was con-ducted by Glaser et al. (Ref. 42). Male Wistar ratswere exposed continuously for 18 months to aero-sols of nickel oxide and arsenic trioxide (particlesize < 0.3 MMAD). Two dose levels, 60 ug/m3 and200 ug/m3, were used. Exposure to arsenic trioxide
had no apparent effects; there were no significantdifferences in body weight, hematology, clinicalchemistry, gross pathology, or histopathology be-tween treated and control animals. Rats treatedwith nickel oxide had an increased mortality rateand severe alveolar proteinosis. The red blood cellcount, hematocrit, and hemoglobin level were sig-nificantly elevated after 9 months of exposure. Notreatment-related tumors were seen in either group,but the investigators concluded that survival wastoo short to assess the carcinogenicity of nickeloxide.
Gorban et al. examined the effects of exposure tofumes generated by CCVshielded GMAW of me-dium and high alloy steel with two types of austen-itic electrodes (Ref. 45). The concentrations ofchromium, nickel, and manganese were 22%, 7%,and 1.5% for the first electrode and 20%, 9%, and7% for the second electrode. The LD50s, deter-
45
mined by intraperitoneal injection of fumes, didnot differ significantly for the two electrodes (elec-trode 1: 2167 mg/kg; electrode 2: 2017 mg/kg). Dif-ferences noted in the effects of the two weldingfumes on levels of alanine and aspartate amino-transferase and nucleic acid levels in the liver andlung were attributed to differences in the contentand chemical characteristics of manganese andchromium in the welding fumes.
In other studies by this research group, Gorban etal. (Refs. 43 and 87) determined the acute toxicityof welding fume containing 29.9% manganese,3.4% nickel, 7.5% trivalent chromium, 0.03% hexa-valent chromium, and 24% iron. LD50s, deter-mined by intraperitoneal injection in rats and mice,were 1008 and 1450 mg/kg, respectively. No signsof fibrosis resulted from intratracheal instillation of50 mg welding fume. Dose-related changes in serumtransaminases, hepatic succinate dehydrogenase,and liver and kidney weights were observed in ratsfollowing exposure by inhalation for 4 months.From these studies, a maximum permissible expo-sure of 0.3 mg/m3 was recommended based on totalfume or 0.1 mg/m3 based on the manganese contentof the fume.
20.4 Allergic Sensitivity. Repeated skin contactwith chromium salts can produce allergic dermati-tis. Caldas and Hicks reported that exposure of therespiratory tract to chromate or extracts of fumesrich in chromium and nickel from SMAW of stain-less steel can inhibit this reaction (Refs. 18 and 60).Guinea pigs receiving a series of dermal injectionsof chromate developed dermal hypersensitivity re-actions in response to further dermal challenge withchromate. This response was significantly inhibitedin animals given repeated intrapulmonary injec-tions of potassium chromate or potassium dichro-mate before the dermal sensitization procedure.Pretreatment of the respiratory tract with nickelsulfate did not alter the dermal response tochromate indicating that pulmonary exposure tochromate provokes a specific tolerance to the im-munologic effects of chromate.
20.5 Biochemical Studies. Geleskul et al. used theformation of malondialdehyde in the presence ofascorbic acid as a measure of lipid peroxidationproduced by welding fumes (Ref. 41). SMAW fumesamples were administered to rats by intraperito-neal injection or intratracheal instillation. Fumesamples contained 19.8% silicon, 17.3% manga-nese, and 24.2% iron. Malondialdehyde concentra-tions were elevated in the lung, kidney, and liver.The effects were greater after intratracheal instilla-
tion than after intraperitoneal injection, and the re-sponse in the lung was greater than that in the othertwo organs. The response reached a peak within 24hours in the lung and within 7 days in the liver andkidney. The investigators concluded that lipid per-oxidation can be used to assess the toxicity of weld-ing fumes.
Calmodulin, a calcium-binding protein whichregulates diverse cellular processes, is normally ac-tivated by calcium, but can also be activated byother heavy metals with a similar ionic radius (e.g.,lanthanum, cadmium, mercury, and lead). MacNeilet al. explored the possibility that chromium mayexert its toxicity at the subcellular level by interact-ing with calmodulin (Refs. 29 and 93). Interactionswith calmodulin were tested by assaying the effectof metals on beef heart phosphodiesterase, acalmodulin-dependent enzyme. With the exceptionof chromium, micromolar doses of calcium and theheavy metals tested were required to achieve maxi-mal activation of calmodulin. In contrast, trivalentchromium was only slightly effective in micromolardoses, but nanomolar concentrations of chromiumwere uniquely effective in activating calmodulin.The investigators suggested that the sensitivity ofcalmodulin to these low concentrations of chro-mium could be relevant to the mechanism of chro-mium toxicity. Because hexavalent chromium isconverted to the trivalent form once inside the cell,a similar effect could ultimately be produced byboth valence states of chromium.
Red blood cell chromium levels were determinedin 22 stainless steel welders who reported no phys-ical symptoms of chromium toxicity and in 11controls who had no occupational exposure to chro-mium. The average red blood cell chromium con-centration for welders was 28.2 nM compared with7.5 nM for controls. Because red blood cell concen-trations of chromium were of the same order ofmagnitude as those which activate calmodulin invitro, the investigators suggested that chromiummay exert toxic effects within the cell by interactingwith regulatory proteins. However, as the investiga-tors pointed out, chromium in very low levels is anessential element. Thus, it would also seem possiblethat activation of calmodulin by chromium may bea normal cell function.
20.6 Barium. In 1984, Dare etal. reported that wa-ter-soluble barium salts can be extracted fromfumes from three types of electrodes containing bar-ium fluxes (Ref. 133). Using the same extractionprocedures, Hicks et al. later showed that water-sol-uble components of fumes from the Soudefonte Blflux-coated electrode and aqueous solutions of pure
46
barium chloride produce similar physiologicaleffects (Ref. 61). Bronchoconstriction, cardiac ar-rhythmias, and transient hypertension were ob-served when solutions of fume extracts or bariumchloride were administered to guinea pigs either byintravenous injection or by aerosols introducedinto the lungs through tracheal cannulae.
21. In Vitro Studies
21.1 Bacterial Assays. In the past, most studies ofthe mutagenicity of welding fumes indicated thatonly stainless steel welding fumes are mutagenic inthe Salmonella/Ames test. Unlike previous reports,Biggart et al. reported that the paniculate fractionof emissions from SMAW of mild steel are muta-genic when the Salmonella/Ames assay is modifiedto include preincubation of tester cells in liquid me-dium with the fume (Ref. 133). No mutagenic activ-ity was observed when the tester bacterial cells wereplated with fume particles according to standardprotocols in the absence of a preincubation period.The particles contained both direct-acting andpromutagenic compounds (promutagens requireactivation by mammalian metabolic enzymes be-fore mutagenicity is expressed) which inducedframeshift mutations.
The direct-acting mutagenic activity was nonspe-cifically inactivated by microsomal enzymes (i.e.,both native and heat-denatured microsomal en-zymes inhibited mutagenic activity). Based onthese data, and evidence that the mutagenic activityis (1) not altered by incubation with ethylene di-amine tetraacetic acid (EDTA), is (2) not extracta-ble with organic solvents, and (3) resides in particlestoo large to pass through the bacterial cell wall, Big-gart et al. concluded that the mutagen(s) associatedwith welding fume particles are not present in ionicforms and are strongly associated with the particles.They postulated that the fume particle is degradedat the cell surface enabling particle components tobe taken up by the bacteria (Ref. 14).
Mutagenic agents were demonstrated in the gasfraction by passing filtered welding aerosol into achamber containing agar plates seeded with testercells (gas protocol). This method was far more effec-tive than bubbling the filtered gas through a liquidmedium containing bacterial cells (aqueous proto-col). Exposure of the tester strains to less than 0.5%of the gas generated from one welding rod induced a103-fold increase in base-pair substitution muta-tions with the gas protocol, but only a two-fold in-crease with the aqueous protocol. Unlike theparticulate fraction of welding fumes, exposure to
the gas fraction greatly increased base-pair substitu-tions, but not frameshift mutations. The mutagenic-ity of the gas fraction was, at most, only partiallydue to NO2 since quantities of NO2 comparable tothat present in the welding aerosol produced a sub-stantially lower mutation rate (Ref. 13).
Ong et al. demonstrated that the SOS test, whichdetects gene damage in Salmonella typhimurium bya colorimetric reaction, can be adapted for use inoccupational settings (Refs. 124 and 125). Fumesfrom SMAW of stainless steel, diesel exhaust, andoccupational agents known to be mutagenic in theSalmonella/Ames test were collected by impinge-ment directly into medium containing Salmonellatyphimurium tester cells. After several hours of ex-posure in the collection medium, tester cells wereremoved from the impinger and subjected to stan-dard SOS test procedures. Diesel exhaust and stain-less steel welding fumes induced a high responsewith the SOS test, and the effects were greater with-out microsomal enzymes than in their presence.The SOS test was slightly more sensitive to dieselfumes, but much more sensitive to welding fumesthan the Salmonella/Ames test. Ong et al. did notattempt to explain this finding. However, their ob-servations may have resulted from differences inthe collection methods used for the two tests. Fumesamples were collected by impingement into aque-ous medium for the SOS test but by filtration for theAmes test. As discussed earlier, Pedersen et al. (Ref.137) have shown that greater quantities of hexava-lent chromium are present in fumes collected by im-pingement than by filtration. This observationcould explain the findings of Ong et al. if hexavalentchromium is responsible for activity in the SOStest.
21.2 Mammalian Cell Studies
21.2.1 Sister Chromatic! Exchange (SCE). Twostudies examined the induction of SCE in cell cul-tures by welding and cutting aerosols (Refs. 8 and182). Water-soluble and insoluble fractions ofSMAW fumes generated by five electrodes weretested in mitotic delay and SCE assays with cul-tured Chinese hamster lung cells by Baker et al.(Ref. 8). In all cases, SCE activity was proportionalto the hexavalent chromium content in the fumes;trivalent chromium, fluorides, nickel, manganese,or other fume constituents made, at the most, mi-nor contributions to the induction of SCE. How-ever, the hexavalent chromium content could notaccount for the mitotic delay caused by fume sam-ples, indicating that other components of the fumefractions produced this effect. Soluble fume frac-
47
tions were much more active than insoluble frac-tions in both the mitotic delay and SCE assays,presumably because of their greater bioavailability.
In the second study, Valerio et al. tested fumesreleased during gas cutting of oil tanks. Oil tank sur-faces are frequently covered with thick layers of taror crude oil. When repairs are made by gas cutting,thick smokes are released with high levels of benzo(a)pyrene and other polycyclic aromatic hydrocar-bons. The study showed that smokes emitted bygas cutting during refitting operations in oil tankersproduce SCE in murine bone marrow cells (Ref.182).
21.2.2 Cytotoxicity. Alveolar macrophages playan important role in the defense of the lung againstforeign materials. However, their activity may alsobe an important factor in the development ofchronic fibrotic lung disease and emphysema. Be-cause of this, there has been considerable interest inthe effects of welding fume particles on macrophageviability and phagocytosis (Refs. 23, 64, and 188).Hooftman et al. showed that the cytotoxicity ofwelding fume particles generated by different weld-ing processes to bovine alveolar macrophages de-creases in the following order: SMAW of stainlesssteel > GMAW of stainless steel > SMAW of mildsteel > GMAW of mild steel. The toxicity of SMAWfume particles was comparable to that of equivalentconcentrations of hexavalent chromium in K2CrO4,suggesting that hexavalent chromium may be re-sponsible for the cytotoxicity of welding fumes toalveolar macrophages (Ref. 64).
The cytotoxicity of fumes from SMAW andGMAW of mild and stainless steel to rat alveolarmacrophages was studied by Pasanen et al. (Ref.135). Macrophages, obtained by lung lavage, wereincubated overnight with intact fume particles ortheir water-insoluble components. Cell viability, asdetermined by trypan blue exclusion, and release ofthe enzymes lactate dehydrogenase (LDH) and ly-sosomal beta-glucuronidase from macrophageswere measured. Fumes from SMAW of mild andstainless steel were equally effective in reducing cellviability and were much more effective than thosefrom GMAW. Fumes generated by GMAW of mildsteel were less cytotoxic than those from GMAW ofstainless steel. Similar to the findings of Hooftmanet al., Pasanen found that the viability of cellstreated with fumes from GMAW of mild steel dif-fered little from control cells. Removing water-solu-ble materials by washing with phosphate-bufferedsaline substantially reduced the cytotoxicity offumes from SMAW of stainless steel but had littleeffect on the other fume particles.
The effect of washed and unwashed fumes on therelease of LDH followed the same pattern (Figure6), whereas the effect on beta glucuronidase wassimilar for washed and intact particles from allwelding processes. Potassium chromate and parti-cles from SMAW of stainless steel had equivalenteffects. This suggested that the hexavalent chro-mium present in fumes from SMAW of stainlesssteel may be responsible for the toxic effects of theseparticles.
Two research groups compared the cytotoxic ef-fects of fumes from different welding processes oncultured mammalian cells. Hildebrand and Collyn-d'Hooghe examined the effects of welding fumeparticles on cultured human embryonic pulmonaryepithelial cells (Ref. 63). These investigators foundthat fumes from SMAW of stainless steel were morecytotoxic than fumes from GMAW of stainlesssteel. Onoprienko (Ref. 126) compared the cyto-toxic effects on mouse fibroblast and human epithe-lial cell cultures of fumes generated by "powdered"electrodes with those generated by basic and acidelectrodes. The relative toxicities of five Russianelectrodes was determined on the basis of this ex-periment.
SS/MMA SS/MIG MS/MMA MS/MIGV77A iiilil
TOTAL DUST WASHED
Note: The concentration of welding fumes was 50 ug/mlculture medium. ***p<0.001; **p<0.01
Figure 6 — Effects of Fumes Generated bySMAW of Stainless Steel (SS/MMA),GMAW of Stainless Steel (SS/MIG),
SMAW of Mild Steel (MS/MMA), andGMAW of Mild Steel (MS/MIG) on the
Release of Cytoplasmic LDH FromCultured Rat Alveolar Macrophages,Before and After Extraction of Water-Soluble Compounds From the Fume
Samples Pasanen et al., Ref. 135
48
21.3 Hyperbaric Pressure. Because of the impor-tance of offshore oil fields in northern Europe,much attention is being paid there to the effects ofhyperbaric pressures on divers performing func-tions such as underwater welding and cutting. Ad-verse physiological effects can result from exposureto hyperbaric conditions in the absence of chemicalexposures. Concern for the effects of fumes createdby underwater welding on already stressed physio-logical systems prompted the study of the combinedeffects of chromium and high pressures on thecytoskeletal systems of fibroblast cultures (Refs. 68and 173). The cytoskeletal system, responsible formaintaining cell shape and motility, is composed ofchains or filaments of the protein actin. High pres-sure can disrupt these filaments and alter the shapeof the cells causing them to become spherical.
Syversen et al. (Refs. 68 and 173) showed thatwhile 1-hour exposures to either high pressure (100or 150 bar) or chromate (1 to 5 uM K^C^O?) causedcells to round up and changed the appearance of theF-actin filaments, the effects were not additive.Thus, the same number of abnormal cells were seenin cultures exposed for one hour to elevated pres-sures in the absence or presence of chromium. How-ever, the results obtained during 18-hour exposures
were quite different. In this case, exposure to 1 uMchromium caused a marked decrease in the numberof normal cells exposed to 100 bar pressure,whereas only a small number of cells were affectedat normobaric pressures. The synergistic effectsseen with 18-hour exposures to 1 uM chromate and100-bar pressure suggested that the toxicity of chro-mate may be potentiated by elevated pressures.
In a related study, Jenssen and Syversen studiedthe effects of chromium and high pressure on cellcycle kinetics in cultured glioma cells (Ref. 68). Thecell cycle is divided into four phases in regard toDNA synthesis and cell division: cell division takesplace during mitosis (M); the first gap phase (Gl)precedes DNA synthesis (S); the second gap phase(G2) precedes cell division. Exposure of cells to 1uM K2Cr207 at normobaric pressures (1 bar) causeda prolongation of the S phase, whereas cells exposedto 5 uM K2Cr 2O7 did not pass through mitosis.High pressure did not alter these effects. However,the combined exposure to chromate and 100-barpressure resulted in the prolongation of the earlypart (Gl phase) of the cell cycle. This effect was notseen if cells were exposed to chromate or elevatedpressure alone.
References
(1) ACGIH, Threshold limit values and biologi-cal exposure indices for 1986-1987. American Con-ference of Governmental Industrial Hygienists,Cincinnati, OH, (1986).
(2) Anlberg, M. S., Hansson, H. C, Fangmark, I.,The growth of a welding fume as a function of rela-tive humidity. In: Health Hazards and BiologicalEffects of Welding Fumes and Gases. ProceedingsInternational Conference, Copenhagen, 18-25 Feb.1985. Eds: R. M. Stern, A. Berlin, and J. Jarvisalo.Elsevier Science Publishers, Amsterdam, 103-106(1986).
(3) Aitio A. and Jarvisalo J., Levels of weldingfume in tissues and body fluids. In: Health Hazardsand Biological Effects of Welding Fumes and Gases.Proceedings International Conference, Copenha-gen, 18-25 Feb. 1985. Eds: R. M. Stern, A. Berlin,and J. Jarvisalo. Elsevier Science Publishers, Am-sterdam, 169-179.(1986).
(4) Angerer, J., Amin, W., Heinrich-Ramm, R.,Szadkowski, D., and Lehnert G., Occupationalchronic exposure to metals. I. Chromium exposureof stainless steel welders — biological monitoring.Int. Arch. Occup. Environ. Health, 59, 503-512(1987).
(5) Anttila, S. Dissolution of stainless steel weld-ing fumes in the rat lung: An X ray microanalyticalstudy. Br. J. Ind. Med, 43, 592-596 (1986).
(6) Anttila, S., Grekula, A., Sivonen S. J.,Alapieti, T., Kalliomaki, P. L., and Sutinen, S.,Application of transmission electron microscopyand x-ray microanalysis in studying inhaled inor-ganic particles in-situ in lung tissue. Third Inter-national Conference on Environmental LungDisease, Montreal, 15-18 Oct. 1986. Chest, 91(2), 313(1987).
(7) Anttila, S., Sutinen, S., Kalliomaki, P. L.,Nickels, J., Grekula, A., and Sivonen, S. J., Struc-ture and composition of inhaled welding fumes inrat lung. In: Health Hazards and Biological Effectsof Welding Fumes and Gases. Proceedings Int. Con-ference, Copenhagen, 18-25 Feb. 1985. Eds: R. M.
Stern, A. Berlin, and J. Jarvisalo. Elsevier SciencePublishers, Amsterdam, 353-356 (1986).
(8) Baker, R. S., Arlauskas, A., Tandon, R. K.,Crisp, P. T, and Ellis, J., Toxic and genotoxic actionof electric-arc welding fumes on cultured mamma-lian cells. J. Appl. Toxicol, 6(5), 357-362 (1986).
(9) Barlier, A., Guelin, J. C , Jeandel, B., andSalsi, S., Optical radiation in arc welding. Dangerand protective measures. DVS Ber., 105, ISSSchweissen Schneiden '86, 70-75 (1986).
(10) Baumgardt, B., Jackwerth, E., Otto, H., andTolg, G., Trace analysis to determine heavy metalload in lung tissue. Int. Arch. Occup. Environ.Health, 58, 27-34(1986).
(11) Beckett, W. S., Davis, J. E., Vroman, N.,Nadig, R., and Fortney, S., Heat stress associatedwith the use of vapor-barrier garments. J. Occup.Med, 28, 411-414(1986).
(12) Benton, R. E., Bonney, T. B., Cole, H.M.,and Epstein, S.G., Studies of chemical and physicalemissions from welding aluminum alloys. Alumi-num Joining Seminar, Metairie, Louisiana, 25-27Feb. 1986. Publ: The Aluminum Assoc, Washing-ton, D.C., 375-382 (1986).
(13) Biggart, N. W. and Rinehart, R. R., Compari-son between aqueous-phase and gas-phase exposureprotocols for determining the mutagenic potentialof nitrogen dioxide and the gas fraction of weldingfumes. Mutation Res., 188, 175-184 (1987).
(14) Biggart, N. W., Rinehart, R. R., and Verfail-lie J., Evidence for the presence of mutagenic com-pounds other than chromium in particles from mildsteel welding. Mutation Res., 180, 55-65 (1987).
(15) Bjelle, A., Hagberg, M.,and Michaelson, G.,Work-related shoulder-neck complaints in indus-try: A Pilot Study. Br. J. RheumatoL, 26, 365-369,(1987).
(16) Bjoerseth, O., Boerresen, E., Malvik, B., andJakobsen, K., Measurement of welding fume athyperbaric conditions. In: Health Hazards and Bio-logical Effects of Welding Fumes and Gases. Pro-ceedings Int. Conference, Copenhagen, 18-25 Feb.
49
50
1985. Eds: R. M. Stern, A. Berlin, and J. Jarvisalo.Elsevier Science Publishers, Amsterdam, 159-162(1986).
(17) Buskin, S. E. and Paulozzi, L. J., Fatal inju-ries in the construction industry in WashingtonState. Am. J. Ind. Med., 11(4), 453-460, (1987).
(18) Caldas, L. Q. A. and Hicks, R., Intrapulmo-nary pretreatment by metal-fume componentscausing inhibition of delayed hypersensitivity.Arch. Toxicoi, Suppl. 11, 223-226, (1987).
(19) Callier, A., The truth about contact lenses.Operations Forum, 4(2):14; (1987).
(20) Cellini, M., Profazio, V., Fantaguzzi, P.,Barbaresi, E., Longanesi, L., and Caramazza, R.,Photic maculopathy caused by arc welding. A casereport. Int. Ophthalmoi, 10, 157-9 (1987).
(21) Chong, C, Determination of silver, bismuth,cadmium, copper, iron, nickel and zinc in lead- andtin-base solders and white metal bearing alloys byatomic-absorption spectrophotometry. Talanta, 33,91-94(1986).
(22) Christensen, J. M. and Kirchhoff, M. Deter-mination of nickel, cadmium and chromium inwhole blood and urine by Zeeman absorption spec-trophotometry. In: Health Hazards and BiologicalEffects of Welding Fumes and Gases. ProceedingsInternational Conference, Copenhagen, 18-25 Feb.1985. Eds: R. M. Stern, A. Berlin, and J. Jarvisalo.Elsevier Science Publishers, Amsterdam, 181-184(1986).
(23) Coate, W. B., Toxicity of Welding fume inrats. Reports: 2236-100/101. Publ: Miami, FL,American Welding Society, (1987).
(24) Cohen, D., Ferromagnetic contamination inthe lungs and other organs of the human body. Sci-ence, 750,745-748(1973).
(25) Corey, P. and Broder, I., Lung cancer inwelders. Third International Conference On Envi-ronmental Lung Disease, Montreal, 15-18 Oct.1986. Chest, 91(2), 301 (1987).
(26) Cotes, J. E. and El-Gamal, F. M. Cardio-re-spiratory associations with shipyard welding andburning. In: Health Hazards and Biological Effectsof Welding Fumes and Gases. Proceedings Interna-tional Conference, Copenhagen, 18-25 Feb. 1985.Eds: R. M. Stern, A. Berlin, and J. Jarvisalo. El-sevier Science Publishers, Amsterdam, 395-398(1986).
(27) Cullen, M. R., Cherniack M. G., and Komin-sky, J.R. Chronic beryllium disease in the UnitedStates. Seminars in Resp. Med., 7, 203-209 (1986).
(28) Cullen, M. R. Kominsky, J. R. Rossman, M.D., Cherniack, M. G., Rankin, J. A., Balmes, J.,Kern, J. A., Daniele, R. P., Palmer, L., Naegel, G. P.McManus, K., Cruz, R. and Cruz, R., Chronic be-
ryllium disease in a precious metal refinery.Clinical epidemiologic and immunologic evidencefor continuing risk from exposure to low level beryl-lium fume. Am. Rev. Resp. Dis., 135, 201-208(1987).
(29) Dawson, R., Lakey, T., Morris, B., and Neil,S. M., Activation of calmodulin by chromium itspossible relevance to chromium toxicity in man.5th Joint Meeting of British Endocrine Societies,Sheffield, England, Apr. 7-10 1986. /. Endocrinol,70S (1986).
(30) De Nevers, N. A., Fatal fire with "nonflam-mable" methyl chloroform. Arch. Environ. Health,41,279-281 (1986).
(31) De Vuyst, P., Dumortier, P., Rickaert, F.,Van de Weyer, R., Lenclud, C , and Yernault, J. C.Occupational lung fibrosis in an aluminum pol-isher. EurJ. Respir. Dis., 68(2), 131-140 (1986).
(32) Desmeules, M. and Cloutier, L. G., Welderpneumoconiosis; clinical functional and pathologi-cal correlations. Third International Conference onEnvironmental Lung Disease, Montreal, 15-18 Oct.1986. Chest, 97(2), 311, (1987).
(33) Desmeules, M. and Tardif, A., Long-termevolution of pulmonary infiltrates in welders' pneu-moconiosis. Third International Conference On En-vironmental Lung Disease, Montreal, 15-18 Oct.1986. Chest, 91(2), 311, (1987).
(34) Dolan, B. P. and Dolan, D. C , Pulmonarydisease survey among union pipefitters/welders inCalifornia. In: Health Hazards and Biological Ef-fects of Welding Fumes and Gases. Proceedings In-ternational Conference, Copenhagen, 18-25 Feb.1985. Eds: R. M. Stern, A. Berlin, and J. Jarvisalo.Elsevier Science Publishers, Amsterdam, 387-390(1986).
(35) Donker, B., Gelhaus, J. and Wiesner, P., Hy-perbaric welding gas absorber — design and testoperation up to 600 m. Meerestechnik/MarineTechnology, 17(4), 133-137 (1986).
(36) Drenck, K. and Stern, R. M. Alternating cur-rent susceptibility bridge magnetopneumography.In: Health Hazards and Biological Effects of Weld-ing Fumes and Gases. Proceedings InternationalConference, Copenhagen, 18-25 Feb. 1985. Eds: R.M. Stern, A. Berlin, and J. Jarvisalo. Elsevier Sci-ence Publisher, Amsterdam, 219-222 (1986).
(37) Ebeid, N. Thiocyanates in Sputum: Relationto welding fumes, tobacco smoking and carcinoge-nicity. J. Soc. Occup. Med., 37, 59-61 (1987).
(38) Emmerling, G., Zschiesche, W., Schaller, K.H., Weltle, D., Valentin, H. and Zober, A., Occupa-tional-health study of bronchial hypersensitivityduring electric arc welding of surface-coated steels.DVSBer., 97,(1987).
51
(39) Engstrom, B. and Virtamo, M., On siteozone measurements in welding shops. In: HealthHazards and Biological Effects of Welding Fumesand Gases. Proceedings International Conference,Copenhagen, 18-25 Feb. 1985. Eds: R. M. Stern, A.Berlin, and J. Jarvisalo. Elsevier Science Publish-ers, Amsterdam, 163-166 (1986).
(40) Farrell, F. J., Angioedema and urticaria asacute and late phase reactions to zinc fume expo-sure, with associated metal fume fever-like symp-toms. Am. J. Ind. Med., 12, 331-337 (1987).
(41) Geleskul, Y. F., Gorban, L. N., Baida, N. A.,Voitkevich, V. G. and Kornuta, N. A., Peroxidationof lipids in animal tissues as an index of biologicalagressiveness of welding aerosols. Vrach. Delo, 6,108-110(1987).
(42) Glaser, U., Hochrainer, D., Oldiges, H., andTakenaka, S., Long-term inhalation studies withNiO and As203 aerosols in Wistar rats. In: HealthHazards and Biological Effects of Welding Fumesand Gases. Proceedings International Conference,Copenhagen, 18-25 Feb. 1985. Eds: R.M. Stern, A.Berlin, and J. Jarvisalo. Elsevier Science Publish-ers, Amsterdam, 325-328 (1986).
(43) Gorban, L. N., Prilipkov, A., Cered-nicenkov, M., and Kireev, V. I., Bases for a MACfor welding fume containing manganese, chromiumand nickel in workplace air. Gig Tr. Prof. Zaboi, 5,26-29(1986).
(44) Gorban, L. N. and Borodiuk, T.M., Hygieniccharacteristics of exposure of the body to magne-sium and aluminum magnide dust and to the aero-sol formed during the welding of an aluminum-magnesium alloy. Vrach Delo, ISS 2, 105-108(1986).
(45) Gorban, L. N., Prilipko, V. A., Voitkevich, V.G., and Fedorina, E. E., Particle structure of a weld-ing aerosol and its biological activity. Gig. Sanit., 723-26(1986).
(46) Gray, C. N., Hexavalent chromium in weld-ing fume and the role of ozone [letter]. Ann. Occup.Hyg., 31,269-70(1987).
(47) Grekula, A., Peura, R., and Sivonen, S. J.,Quantitative energy dispersive x-ray microanalysisof welding fumes. In: Health Hazards and Biologi-cal Effects of Welding Fumes and Gases. Proceed-ings International Conference, Copenhagen, 18-25Feb. 1985. Eds: R. M. Stern, A. Berlin, and J.Jarvisalo. Elsevier Science Publishers, Amsterdam,85-88(1986).
(48) Grekula, A., Ristolainen, E., Tanninen, V. P.,Hyvarinen, H. K., and Kalliomaki, P. L., Surfaceand bulk chemical analysis on metal aerosols gener-ated by manual metal arc welding of stainless steel./. Aerosol Science, 17, 1-9 (1986).
(49) Grekula, A., Sivonen, S. J., Hyvarinen, H.K., Tanninen, V. P., and Kalliomaki, P. L., Analyti-cal electron microscopy of welding fumes. Univer-sity of Oulu, Reports of the Institute of ElectronOptics, Report 1, Oulu, Finland, (1986).
(50) Gustavsson, C. and Welinder, H., Stainlesssteel manual metal-arc welding and chromium inblood. In: Health Hazards and Biological Effects ofWelding Fumes and Gases. Proceedings Interna-tional Conference, Copenhagen, 18-25 Feb. 1985.Eds: R. M. Stern, A. Berlin, and J. Jarvisalo. El-sevier Science Publishers, Amsterdam, 185-188(1986).
(51) Hagberg, M., Lindqvist, B., and Wall, S., Ex-posure to welding fumes and chronic renal diseases,a negative case-referent study. Int. Arch. Occup. En-viron., Health, 58(3), 191-195 (1986).
(52) Hansen, H. S., Sogaard, H., and Jurgensen K.S.-L., The significance of occupational exposure forcancer in the nasal cavity and paranasal sinuses. Astudy with reference to cases in Denmark, Swedenand Finland. Ugeskr. Laeg., 148(34), 2185-2187(1986).
(53) Heikkila, P. R., Hameila, M., Pyy, L., andRaunu, P., Exposure to Creosote in the impregna-tion and handling of impregnated wood. Scand. J.Work Environ. Health, 13, 431-437 (1987).
(54) Heinrich-Ramm, R. and Angerer, J., Deter-mination of nickel and chromium in body fluids.Trace elements in human health and disease, Sec-ond Nordic Symposium, Odense, Denmark, 1987.70-73(1987).
(55) Heinsalmi, P., Method to measure workingposture loads at working sites (OWAS). The Ergo-nomics of Working Postures. Models, Methods andCases, N. Corlett, J. Wilson, and I. Manenica, Edi-tors, London, Taylor and Francis, 100-104 (1986).
(56) Helander, M. G., Karwan, M. H., andEtherton, J. E., A model of human reaction time todangerous robot arm movements. Proceedings ofthe Human Factors Society, 31st Annual Meeting,191-195(1987).
(57) Helander, M. G. Ergonomics, workplace de-sign. International Encyclopedia of Robotics Appli-cations and Automation, John Wiley and Sons, Inc.,477-488(1988).
(58) Henderson, I. D., Fume generation andchemical analysis of fume for a selected range offlux-cored structural steel wires. AWRA documentsP9-44-85 (revised 15/9/86) Aust. Weld. Res., 15,4-11 (1986).
(59) Hewitt, P. J. and Madden, M. G., Weldingprocess parameters and hexavalent chromium inMIG fume. Ann. Occup. Hyg., 30, 427-434 (1986).
(60) Hicks, R. and Caldas, L. Q. A., Immunologi-
52
cal unresponsiveness of chromium or nickel in theguinea-pig induced by stainless steel welding fumecomponents. Arch. ToxicoL, Suppl. 9. 421-422(1986).
(61) Hicks, R., Caldas, L. Q. A., Dare, P. R., andHewitt, P. J., Cardiotoxic and bronchoconstrictoreffects of industrial metal fumes containing barium.Arch. ToxicoL, Suppl. 9, 416-20 (1986).
(62) Hicks, R., Oshodi, R. O., and Pedrick, M. J.,Cytotoxic, irritant and fibrogenic effects of metal-fume particulate materials investigated by intra-muscular injection in the rat and guinea pig. Arch.ToxicoL, Suppl. 11, 223-226 (1987).
(63) Hildebrand, H. F., Collyn-d'Hooghe, M.,and Stern, R. M. Cytotoxic effects of welding fumeson human embryonic epithelial pulmonary cells inculture. In: Health Hazards and Biological Effects ofWelding Fumes and Gases. Proceedings Interna-tional Conference, Copenhagen, 18-25 Feb. 1985.Eds: R. M. Stern, A. Berlin, and J. Jarvisalo. El-sevier Science Publishers, Amsterdam, 319-324(1986).
(64) Hooftman, R. N., Roza, P., and Arkesteijn,C. W. M., In vitro cytotoxicity of welding fume par-ticles to bovine alveolar macrophages In: HealthHazards and Biological Effects of Welding Fumesand Gases. Proceedings International Conference,Copenhagen, 18-25 Feb. 1985. 1985. Eds: R. M.Stern, A. Berlin, and J. Jarvisalo. Elsevier SciencePublishers, Amsterdam, 311-314 (1986).
(65) Ingle, J., Device for automatic recording ofarcing periods during electric welding. Ann. Occup.Hyg., 30, 123-124(1986).
(66) Jakubcik, L., Use of RVD-F respirators inwelding of high-alloy manganese steels. Zvaranie,36( 12), 374-376(1987).
(67) Jiao, C , Mao, X., Zhang, Z., and Bai, J., De-velopment of model ALJB 1.5-1 welding fume puri-fier. Gongye Weisheng Yu Zhiyebing, 12(3), 179-82(1986).
(68) Jenssen, J. and Syversen, T., High hydro-static pressure and chromate-induced effects on cellcycle kinetics. Undersea Biomed Res, 14(1), 21-9(1987).
(69) Jordan, D. R., The potential damaging ef-fects of light on the eye (Part I). Canadian J.Opthalmology, 21, 216-224 (1986).
(70) Kalliomaki, P. L., Aitio, A., Hyvarinen, H.K. Kalliomaki, K., and Lakomaa, E. L., Lung clear-ance, transportation, and excretion of metals in ratsafter intratracheal installation of activated weldingfumes. In: Health Hazards and Biological Effects ofWelding Fumes and Gases. Proceedings Interna-tional Conference, Copenhagen, 18-25 Feb. 1985.
Eds: R. M. Stern, A. Berlin, and J. Jarvisalo. El-sevier Science Publishers, Amsterdam, 345-348(1986).
(71) Kalliomaki, P. L., Aitio, A., Lakomaa, E. L.,and Kalliomaki, K., Kinetics of the metal compo-nents of intratracheally instilled mild and stainlesssteel welding fumes in rats. /. Aerosol Sci., 75(16),737-740(1987).
(72) Kalliomaki, P. L., Grekula, A. Hagberg, J.,and Sivonen, S., Analytical electron microscopy ofwelding fumes. J. Aerosol Sci., 18(6), 781-784,(1987).
(73) Kalliomaki, P. L., Hyvarinen, H. K., Aitio,A., Lakoma, E. L., and Kalliomaki, K., Kinetics ofthe metal components of intratracheally instilledstainless steel welding fume suspensions in rats. BrJ. IndMed, 43(2), 112-119 (1986).
(74) Kalliomaki, P. L., Kalliomaki, K., andKorhonen, O., Lung retention of welding fumes andventilation lung functions — a follow-up study ofshipyard welders. In: Health Hazards and Biologi-cal Effects of welding Fumes and Gases. ProceedingsInternational Conference, Copenhagen, 18-25 Feb.1985. Eds: R. M. Stern, A. Berlin, and J. Jarvisalo.Elsevier Science Publishers, Amsterdam, 409-412(1986).
(75) Kalliomaki, K., Kalliomaki, P. L., and Moi-lanen, M., A mobile magnetopneumorgraph withdust quality sensing. In: Health Hazards and Bio-logical Effects of Welding Fumes and Gases. Pro-ceedings International Conference, Copenhagen,18-25 Feb. 1985. Eds: R. M. Stern, A. Berlin, and J.Jarvisalo. Elsevier Science Publishers, Amsterdam,215-218(1986).
(76) Kalliomaki, K., Kalliomaki, P. L., and Moi-lanen, M., Stability of remanent magnetism of dif-ferent welding fumes. In: Health Hazards andBiological Effects of Welding Fumes and Gases. Pro-ceedings International Conference, Copenhagen,18-25 Feb. 1985. Eds: R. M. Stern, A. Berlin, and J.Jarvisalo. Elsevier Science Publishers, Amsterdam,223-226(1986).
(77) Kalliomaki, P. L., Lakomaa, E. L., Aitio, A.,and Kalliomaki, K., Manganese distribution in ratsafter three types of welding fume exposures. In:Health Hazards and Biological Effects of WeldingFumes and Gases. Proceedings International Con-ference, Copenhagen, 18-25 Feb. 1985. Eds: R. M.Stern, A. Berlin, and J. Jarvisalo. Elsevier SciencePublishers, Amsterdam, 349-352 (1986).
(78) Keyserling, W.M., A Computer-aided systemto evaluate postural stress in the workplace. Am.Ind. Hyg. Assoc. J., 47, 641-649 (1986).
(79) Kiilunen, M., Jarvisalo, J., Makitie, O., and
53
Aitio, A., Analysis, storage stability and referencevalues for urinary chromium and nickel. Int. Arch.Occup. Environ. Health, 59, 43-50 (1987).
(80) Kilburn, K. H., Medical screening for lungcancer: perspective and strategy. J. Occup. Med.,25,714-718(1986).
(81) Kilburn, K. H., Respiratory functional im-pairment in metal welders: An interpretation ofcurrent data. In: Health Hazards and Biological Ef-fects of Welding Fumes and Gases. Proceedings In-ternational Conference, Copenhagen, 18-25 Feb.1985. Eds: R. M. Stern, A. Berlin, and J. Jarvisalo.Elsevier Science Publishers, Amsterdam, 363-378(1986).
(82) Kilburn, K. H., Warshaw R., Boylen C. T.,Airway obstruction, volume loss, and respiratorysymptoms in shipyard welders in Los Angeles:Baseline values and cross shift changes. In: HealthHazards and Biological Effects of Welding Fumesand Gases. Proceedings International Conference,Copenhagen, 18-25 Feb. 1985. Eds: R. M. Stern, A.Berlin, and Jarvisalo. Elsevier Science Publishers,Amsterdam, 383-386 (1986).
(83) Kjuus, J., Skjaerven, R, Langard, S., Lien, J.T., and Aamodt, T., A case-referent study of lungcancer, occupational exposures and smoking.Scand. J. Work Environ. Health, 12, 193-202(1986).
(84) Krasnyuk, E. P., Lubyanova I. P., andTimofeeva, N. X, Changes in the respiratory andcardiovascular systems of alloyed steel welders. In:Health Hazards and Biological Effects of WeldingFumes and Gases. Proceedings International Con-ference, Copenhagen, 18-25 Feb. 1985. Eds: R. M.Stern, A. Berlin, and J. Jarvisalo. Elsevier SciencePublishers, Amsterdam, 419-421 (1986).
(85) Krause, H. J. and Press, R.H., Measurementof the welding fume quantities and analysis of thefume composition during resistance welding. In:Health Hazards and Biological Effects of WeldingFumes and Gases. Proceedings International Con-ference, Copenhagen, 18-25 Feb. 1985. Eds: R. M.Stern, A. Berlin, and J. Jarvisalo. Elsevier SciencePublishers, Amsterdam, 65-75 (1986).
(86) Krizmanic-Mrkoci, D., Lovric, M., Brkic,K., Ilic, Z., and Mitrecic, A., Dystrophia retinaepigmentosa sine pigmento as a cause of early work-ing disability. Arh. Hig. Rada Toksikol (Yugoslavia),38(2), 159-162(1987).
(87) Kundiev, Y. and Gorban, L. N., Some ap-proaches to the establishment of permissible levelsof welding aerosols in the air. In: Health Hazardsand Biological Effects of Welding Fumes and Gases.Proceedings International Conference, Copenha-
gen, 18-25 Feb. 1985. Eds: R. M. Stern, A. Berlin,and J. Jarvisalo. Elsevier Science Publishers, Am-sterdam, 583-586 (1986).
(88) Landsberger, S., Non-destructive determina-tion of zinc in biological certified reference materi-als. / . Radioanal. Nucl. Chem., Letters, 127(5),319-323(1988).
(89) Landsberger, S. and Simsons, A., Chro-mium, nickel, and arsenic determinations in hu-man samples by thermal and epithermal neutronactivation analyses. Biol. Trace Elem. Res., 13,357-362(1987).
(90) Lerchen, M. L., Wiggins, C. L., and Samet, J.M., Lung cancer and occupation in New Mexico. J.Natl. Cancer Inst., 79, 639-645 (1987).
(91) Lippmann, M., Magnetopneumography as atool for measuring lung burden of industrial aero-sols. In: Health Hazards and Biological Effects ofWelding Fumes and Gases. Proceedings Interna-tional Conference, Copenhagen, 18-25 Feb. 1985.Eds: R. M. Stern, A. Berlin, and J. Jarvisalo. El-sevier Science Publishers, Amsterdam, 199-213(1986).
(92) Lyngenbo, O., Groth, M. V., Groth, S., Ol-sen, O., and Dirksen, H., Respiratory symptomsand pulmonary function among Danish welders. In:Health Hazards and Biological Effects of WeldingFumes and Gases. Proceedings International Con-ference, Copenhagen, 18-25 Feb. 1985. Eds: R. M.Stern, A. Berlin, and J. Jarvisalo. Elsevier SciencePublishers, Amsterdam, 399-402 (1986).
(93) MacNiel, S., Dawson, R., Lakey, T., andMorris, B., Activation of calmodulin by the essen-tial trace element chromium. Cell Calcium, 8(3)207-216(1987).
(94) Maiti, B. and Desai, S. R., High-perfor-mance liquid chromatographic separation of beryl-lium, cobalt, nickel and chromium as the beta-isopropyltropolone complexes and its applicationto the determination of chromium in air. Analyst,111(7), 809-811 (1986).
(95) Mastromatteo, E., Nickel. Am. Ind. Hyg.Assoc. J., 47(10, 589-601 (1986).
(96) Mcllwain, J. F. and Neumeier, L. A., Fumesfrom shielded metal arc welding electrodes. GovtReports Announcements & Index (GRA&I), Issue 07(1987).
(97) McLaughlin, J. K., Malker, H. S. R., Blot, W.J., Malker, B. K., Stone, B. J., Weiner, J. A., Erics-son, J. L. E., and Fraumeni, J. F., Occupationalrisks for intracranial gliomas in Sweden. J. Natl.Cancer Inst., 78(2), 253-257 (1987).
(98) McLaughlin, J. K., Thomas, T. L., Stone, B.J., Blot, W. J., Malker, H. S. R., Stone, B. J., Weiner,
54
J. A., Ericsson, J. L. E., and Malker, B.K., Occupa-tional risks for intracranial gliomas in Sweden. J.Occup. Med., 28(1), 66-68 (1987).
(99) McMillan, G. H. G., Foiling flashes. Occup.Health (Lond.), 38(1), 20-22 (1986).
(100) McMillan, G. H. G., Metal fume fever. Oc-cup. Health (Lond.), 38(5), 148-149 (1986).
(101) Moller, D. R., Brooks, S. M., Bernstein, D.I., Cassedy, K., Enrione, M., and Bernstein, I. L.,Delayed anaphylactoid reaction in a worker ex-posed to chromium. J. Allergy Clin. Immunol., 77(3), 451-456(1986).
(102) Moravik, J., Malik, E., Morihladkova, V.,and Papayova, A., Influence of welding fumes onthe respiratory tract of welders. Cesk. Hyg., 31(2),77-83 (1986).
(103) Moreton, J., The case for standardizationof tests for welding fume emission rate and chemi-cal composition determination. Ann. Occup. Hyg.,50,435-444(1986).
(104) Mylius, E. A. and Gullvag, B., AlveolarMacrophage Count as an Indicator of Lung Reac-tion to Industrial Air Pollution. Acta Cytologica, 30,157-162(1986).
(105) Najarian, T. and Colton, T., Mortality fromleukemia and cancer in shipyard nuclear workers.Lancet, 1, 1018-1020(1978).
(106) Newcova, A., Use of x-ray fluorescenceanalysis in the study of welding aerosols. Zvaranie,35(1), 215-219(1986).
(107) Nilsson, B., Taste acuity of the human pal-ate. Acta Odontol. Scand., 37, 217-234 (1979).
(108) Nordman, H., Keskinen, H., Savolaninen,A., and Rahkonen, E., Asthmatic reactions in stain-less steel welders. Third International Conference onEnvironmental Lung Disease, Montreal, 15-18 Oct.1986. Chest, 91(2), 311 (1987).
(109) Norell, S., Ahlbom, A., Olin, R., Erwald, R.,Jacobson, G., Lindberg-Navie S., and Wiechel, K.L., Occupational factors and pancreatic cancer. Br.J. Ind. Med., 43, 775-778 (1986).
(110) Norseth, T., The carcinogenicity of chro-mium and its salts. Br. J. Ind. Med., 43( 10), 649-651(1986).
(Ill)Norseth,T., Metal carcinogenicity. In: Liv-ing in a Chemical World. Ann. N. Y. Acad. Sci., 534,377-385 (1988).
(112) Ohlsson, A. C , Jarvholm, B., Lofqvist, A.,Naslund, P. E., and Stenborg, R., Vocal behaviourin welders-a preliminary study. Folia Phoniatr (Ba-sel), 39(2), 98-103 (1987).
(113) Okuno, T, Measurement of blue-light effec-tive radiance of welding arcs. Ind Health, 24(4),213-226(1986).
(114) Okuno, T., Measurement of ultraviolet ra-
diation from welding arcs. Ind. Health, 25, 147-156(1987).
(115) Olah, L., Hygienic aspects of chromiumand nickel occurrence in the welding working envi-ronment. Cesk. Hyg., 32(4), 225-231 (1987).
(116) Olah, L., Hygienic characteristics of hard-facing consumables. Zvaracske Spravy, 37(3), 64-72(1987).
(117) Olah, L., The use of nuclear analyticalmethods for the determination of heavy metals inthe working environment of welding operators.Prac. Lek, 38(1), 9-14(1986).
(118) Olah, L.,and Guldan, P., Hygienic study ofwelding with VUZ TNE 800 and VUZ A6 hardfac-ing electrodes. Zvaranie, 35(5), 156-159 (1986).
(119) Olah L. and Mihalkovic, K., Hygienic eval-uation of flux-cored welding rods for surfacing.Zvaranie, 36(2), 57-60 (1987).
(120) Olah, L. and Pivoluska, J., Method for de-termination of chromium (VI) in the working envi-ronment in industrial processing of high-alloysteels. Hutn. Listy, 41(4), 287-289 (1986).
(121) Olander, L., Welding fume plumes in indus-trial workrooms. Chem. Eng. Monogr., 24, ISS Ven-tilation'85, 153-158(1986).
(122) Oleru, G. and Ademiluyi, S. A., Some acuteand long-term effects of exposure in welding andthermal-cutting operations in Nigeria. Int. Arch.Occup. Environ. Health, 59, 605-612 (1987).
(123) Olin, R., Anlbom, A., Lindberg-Navie, S.,Norell, S., and Spannare, B., Occupational factorsassociated with astrocytoma: a case control study.Am. J. Ind. Med., 11, 615-625 (1987).
(124) Ong, T., Stewart, J., Wen, Y. F., and Whong,W. Z., Application of sos umu-test for the detectionof genotoxic volatile chemicals and air pollutants.Environ. Mutagenesis, 9(2), 171-176 (1987).
(125) Ong, T., Stewart, J. D., and Whong, W. Z.,Induction of sos responses in SalmonellaTyphimurium by volatile chemicals and environ-mental pollutants. Environ. Mutagen, 8(6), 62(1986).
(126) Onoprienko, E. N. and Suprun, S.A., Hygi-enic characteristics of flux-cored welding rods.Metall. Tekhnol. Probl. Svarki Poroshk. Provolokoi,Dokl. Mezhdunar. Shk. Stran-Chlenov SEV, 104-108(1986).
(127) Ortendahl, T., Oral changes in divers work-ing with electrical welding/cutting underwater.Swedish Dental J., Suppl. 43, 51 (1987).
(128) Ortendal, T. W. and Hogstedt, P., Magneticfields effects on dental amalgam in divers weldingand cutting electrically underwater. Undersea Bi-omed. Res., in press (1988).
(129) Ortendal, T. W., Hogstedt, P., Odelius, H.,
55
and Noren, J. G., Effects of magnetic fields fromunderwater electrical cutting on in vitro corrosionof dental amalgam. Undersea Biomed. Res., in press(1988).
(130) Ortendahl, T. W. and Holland, R.I., Influ-ence of submerged electrical welding/cutting ondental amalgam in Swedish divers. Scand. J. Dent.Res., 94(6), 545-52 (1986).
(131) Ortendahl, T. W. and Holland, R.I., Electri-cal activity in dental amalgam of submerged diversduring welding. Ada Odontol. Scand ., 45, 313-20(1987).
(132) Palmer, W. G., Effects of Welding on HealthIV. Miami, FL, American Welding Society, 1983.
(133) Palmer, W. G., Effects of Welding on HealthVI. Miami, FL American Welding Society, 1989.
(134) Parsons, H. M., Robotics and the health ofworkers. In: Occupational Health and Safety in Au-tomation and Robotics, Proceedings 5th UOEH In-ternational Symposium, Kitakyushu, Japan, 20-21Sep. 1985. Ed.: K. Noro, New York, Taylor andFrancis, 196-211 (1987).
(135) Pasanen, J. T., Gustafsson, T. E., Kal-liomaki, P. L., Tossavainen, A., and Jarvisalo, J. O.,Cytotoxic effects of four types of welding fumes onmacrophages in vitro: A comparative study. J. Tox-icol. Environ. Health, 18, 143-152 (1986).
(136) Pearce, N. E. and Howard, J. K., Occupa-tion, social class and male cancer mortality in NewZealand, 1974-78. Int. J. Epid., 15, 456-462 (1986).
(137) Pedersen, B., Thomsen, E., and Stern, R.M., Some problems in sampling, analysis and evalu-ation of welding fumes containing Cr(VI). Ann. Oc-cup. Hyg., 31, 325-338 (1987).
(138) Pekkarinen, J., Industrial impulse noise,crest factor and the effect of earmuffs. Am. Ind. Hyg.Assoc. J., 48, 861-866 (1987).
(139) Peto, J., Cancer morbidity and mortalitystudies of welders. In: Health Hazards and Biologi-cal Effects of Welding Fumes and Gases. Proceed-ings International Conference, Copenhagen, 18-25Feb. 1985. Eds: R. M. Stern, A. Berlin, and J.Jarvisalo. Elsevier Science Publishers, Amsterdam,423-434(1986).
(140) Pierre, F., Boudier, C , Pelletier, A., Salsi,S., Vigneron, J. C , and Bieth, J.G., Effect of weld-ing fumes on lung antiproteases. Inst. Natl. SanteRech. Med., [Colloq.], 155, ISS Silicose Pneumoco-nioses Poussieres Mixtes, 129-33 (1987).
(141) Puntoni, R., Vercelli, M., Digiorgio, F.,Valerio, F., Bonassi, S., Ceppi, M., and Santi, L.,Mortality study among autogenous and electricalwelders [gas and arc welders] in the port of Genoa(Italy). In: Health Hazards and Biological Effects ofWelding Fumes and Gases. Proceedings Interna-
tional Conference, Copenhagen, 18-25 Feb. 1985.Eds: R. M. Stern, A. Berlin, and J. Jarvisalo. El-sevier Science Publishers, Amsterdam, 469-472(1986).
(142) Randolph, S. A. and Zavon, M. R., Guide-lines for contact lens use in industry. / . Occup.Med., 29(3), 237-242 (1987).
(143) Rasmussen, K. H., and Jepsen, J. R., Theorganic psychosyndrome in manual metal arc weld-ers: A possible consequence of manganese toxicity.UGESKR LAEG, 149(51), 3497-3498 (1987).
(144) Reith, A., Voss, R., and Reichborn-Kjen-nerud, S., A simple atraumatic sampling method forscreening of precancerous and toxicological lesionsby cytology in the upper respiratory tract. In:Health Hazards and Biological Effects of WeldingFumes and Gases. Proceedings International Con-ference, Copenhagen, 18-25 Feb. 1985. Eds: R. M.Stern, A. Berlin, and J. Jarvisalo. Elsevier SciencePublishers, Amsterdam, 591-594 (1986).
(145) Roeder, M., Development of a mobile ap-paratus for detection and removal of welding fumesin a plant with recirculating air. Tech. Umwelt-schutz, 33, ISS Umweltgestaltung produktionsber-eichen, 116-19(1986).
(146) Rosen, G., and Lundstrom, S., Concurrentvideo filming and measuring for visualization of ex-posure. Am. Ind. Hyg. Assoc. J., 48,688-692 (1987).
(147) Rudell, B., Kolmodin-Hedman, B., andWenngren B. I., Controlled welding in patients withneurological symptoms. In: Health Hazards and Bi-ological Effects of Welding Fumes and Gases. Pro-ceedings International Conference, Copenhagen,18-25 Feb. 1985. Eds: R. M. Stern, A. Berlin, and J.Jarvisalo. Elsevier Science Publishers, Amsterdam,413-414(1986).
(148) Sawatari, K. and Serita, F., Determinationof chromium speciation in fumes prepared by aplasma metal sprayer as a model of actual weldingfumes. Ind. Health, 24, 51-61 (1986).
(149) Schneider, W. D., Liebich, R., Maintz, G.,Schmidt, G., and Wiesener, W., Exposure and stateof health in electric arc welders using different tech-nologies. In: Health Hazards and Biological Effectsof Welding Fumes and Gases. Proceedings Interna-tional Conference, Copenhagen, 18-15 Feb. 1985.Ed: R. M. Stern, A. Berlin, and J. Jarvisalo. ElsevierScience Publisher, Amsterdam, Netherlands,391-394(1986).
(150) Schneider, W. D., Maintz, G., Reimer, W,Schmidt, G., and Tittelbach, U., Are there func-tional hints of fibrosis in arc welders' siderosis.Clin. Respir. PhysioL, 23(12), 3498 (1987).
(151) Schneider, W. D., Maintz, G., Reimer, W.,Schmidt, G., and Tittelbach, U., Functional signifi-
56
cance of lung siderosis in electrode welders. ZGesamtelnn. Med., 42(5), 126-130 (1987).
(152) Schoenberg, J. B., Stemhagen, A., Mason,T. J., Patterson, J., Bill, J., and Altman, R., Occupa-tion and lung cancer risk among New Jersey whitemales./. Natl. CancerInst., 79(1), 13-21 (1987).
(153) Sharir, M. and Ben-David, A., Severe Kera-toconjunctivitis in plasma flame-spraying workers[letter]. J. Occup. Med., 29, 639 (1987).
(154) Shehade, S. A., Roberts, P. J., Diffey, B. L.,and Foulds, I. S., Photodermatitis Due to SpotWelding. Br. J. Dermatology, 117, 117-119 (1987).
(155) Shusterman, D. and Neal, E., Skeletalmuscle and myocardial injury associated withmetal fume fever. J. Fam. Pract., 23(2), 159-160(1986).
(156) Silberschmid, M., Acute pneumonitis pro-gressing to pulmonary fibrosis in a welder. In:Health Hazards and Biological Effects of WeldingFumes and Gases. Proceedings International Con-ference, Copenhagen, 18-25 Feb. 1985. Eds: R. M.Stern, A. Berlin, and J. Jarvisalo. Elsevier SciencePublishers, Amsterdam, 415-418 (1986).
(157) Sipek, L. and Smars, E., Ozone and nitro-gen oxides in gas shielded arc welding. In: HealthHazards and Biological Effects of Welding Fumesand Gases. Proceedings International Conference,Copenhagen, 18-25 Feb. 1985. Eds: R. M. Stern, A.Berlin, and J. Jarvisalo. Elsevier Science Publish-ers, Amsterdam, 107-110(1986).
(158) Sjogren, B. and Carstensen, J., Cancer mor-bidity among Swedish welders and gas cutters. In:Health Hazards and Biological Effects of WeldingFumes and Gases. Proceedings International Con-ference, Copenhagen, 18-25 Feb. 1985. Eds: R. M.Stern, A. Berlin, and J. Jarvisalo. Elsevier SciencePublishers, Amsterdam, 461-463 (1986).
(159) Sjogren, B., Elinder, C. G., Lidums, V. andChangs, G., Uptake and urinary excretion of alumi-num by welders. Arbete Och Halsa, 16 (1986).
(160) Sjogren, B., Elinder, C. G., Lidums, V., andChangs, G., Uptake and urinary excretion of alumi-num among welders. Int. Arch. Occup. Environ.Health, 60, 77-79(1988).
(161) Sjogren, B., Gustavsson, A., and Hedstrom,L., Mortality in two cohorts of welders exposed tohigh and low-levels of hexavalent chromium.Scand. J. Work Environ. Health, 13, 247-251(1987).
(162) Sjogren, B., Weiner, J., Horte, L. G., andCarstensen J. Mortality amongst Swedish weldersand gas cutters. In: Health Hazards and BiologicalEffects of Welding Fumes and Gases. ProceedingsInternational Conference, Copenhagen, 18-25 Feb.1985. Eds: R. M. Stern, A. Berlin, and J. Jarvisalo.
Elsevier Science Publishers, Amsterdam, 457-459(1986).
(163) Sliney, D. H., Benton, R. E., Cole, H. M.,Epstein, S. G., and Morin, C.J., Transmission of po-tentially hazardous actinic ultraviolet radiationthrough fabrics. Appl. Ind. Hyg., 2, 36-44 (1987).
(164) Smith, N. J., Topping, M. D., Stewart, J. D.,and Fletcher, J. G., Occupational cadmium expo-sure in jig solderers. Br. J. Ind. Med., 43(10), 663-666(1986).
(165) Spacilova, M. and Hykes, P., Generalhealth state of electrowelders and persons at risk offree silicon oxide with similar X-ray picture. Prac.Lek., 38(9), 389-392 (1986).
(166) Spacilova, M. and Hykes, P., The evolutionof pulmonary changes of electrical welders in rela-tion to work risk and smoking habits. Prac. Lek., 38(6), 234-239 (1986).
(167) Stage, J. and Vinding, T., Metal spark perfo-ration of the tympanic membrane with deafnessand facial paralysis. /. Laryngol. Otol., 699-700(1986).
(168) Starck, J. and Pekkarinen, J., Industrial im-pulse noise: crest factor as an additional parameterin exposure measurements. Appl. Acoustics, 20,263-274(1987).
(169) Stern, F. B., Waxweiler, R. A., Beaumont,J., Lee, S. T., Rinsky, R. A., Zumwalde, R. D.,Halperin, W. E., Bierbaum, P. J., Landrigan, P. J.,and Murray, W. E., A case-control study of leuke-mia at a naval nuclear shipyard. Am. J. Epid., 123(6), 980-92 (1986).
(170) Stern, R. M., Cancer incidence amongwelders: possible effects of exposure to extremelylow frequency electromagnetic radiation (ELF) andto welding fumes. Environ. Health Perspect., 76,221-9(1987).
(171) Stern, R. M., Drenck, K., Lyngenbo, O.,Dirksen, H., and Groth, S., Health, occupationalexposure, and thoracic magnetic moment of ship-yard workers. In: Health Hazards and Biological Ef-fects of Welding Fumes and Gases. ProceedingsInternational Conference, Copenhagen, 18-25 Feb.1985. Ed: R. M. Stern, A. Berlin, and J. Jarvisalo.Elsevier Science Publishers, Amsterdam, 403-408(1986).
(172) Stern, R. M., Lipka, J., Madsen, M. B.,Morup, S., Thomsen, E., and Koch, C. J. W., Insta-bility of lung retained welding fumes: A slow activeNi and Cr compartment. Occupational and Envi-ronmental Chemical Hazards, Eds.: V. Foa, E. A.Emmett, M. Maroni, and A. Colombi. Chichester,Ellis Horwood Limited, 444-460 (1987).
(173) Syversen, T. and Jenssen, J., High hydro-static pressure potentiation of the toxic effects of
57
chromate in cell culture. Undersea Biomed. Res., 14(1), 11-19(1987).
(174) Tandon, R. K., Ellis J. Crisp, P. T., Baker, R.S., and Chenhall, B. E., Chemical investigation ofwelding fumes from hardfacing and HSLA-steelelectrodes. Welding J., 65, 231s-236s (1986).
(175) Thomas, T. L. Stolley, P. D., Stemhagen, A.,Fontham, E. T. H., Bleecker, M. L. Stewart, P. A.,and Hoover, R. N., Brain tumor mortality riskamong men with electrical and electronics jobs: Acase-control study. /. Natl. Cancer Inst., 79233-238(1987).
(176) Thomsen, E., Stern, R. M., and Pedersen,B., Exposure monitoring and chemical analysis ofwelding fume. In: Health Hazards and BiologicalEffects of Welding Fumes and Gases. ProceedingsInternational Conference, Copenhagen, 18-25 Feb.1985. Eds: R. M. Stern, A. Berlin, and J. Jarvisalo.Elsevier Science Publishers, Amsterdam, 47-50(1986).
(177) Tolgyessy, J., Olah, L., and Klehr, E. H.,Contribution to the radioanalytical determinationof heavy metals in the working environment ofwelders. J. Radioanal. Nucl. Chem., 114(1), 101-104(1987).
(178) Turut, P., Isorni, M. C , Sicard, C , andMalthieu, D., Macular photoinjury caused by awelding arc on an eye with an implant. Bull. Soc.Ophtalmol. Fr., 86(6-1), 857-859 (1986).
(179) Ulrich, L., Malik, E., and Neumanmova,E., Some effects of exposure to chromium andnickel in the aerosol form. Prac. Lek., 39(9), 390-395(1987).
(180) Uniat, L., Oik, R. J., and Hanish, S. J.,Welding arc maculopathy. Am. J. Ophthalmol, 102(3), 394-395 (1986).
(181) United States EPA, Health assessment doc-ument of nickel. U.S. Environmental ProtectionAgency, Office of Research and Development, En-vironmental Criteria and Assessment Office, Re-search Triangle Park, NC 27711 (1986).
(182) Valerio F., Puntoni, R., and Lazzarotto, A.,Mutagenic activity of fumes produced by gas-cut-ting during refitting operations in oil tankers. In:Health Hazards and Biological Effects of WeldingFumes and Gases. Proceedings International Con-ference, Copenhagen, 18-25 Feb. 1985. Eds: R. M.Stern, A. Berlin, and J. Jarvisalo. Elsevier SciencePublishers, Amsterdam, 315-318 (1986).
(183) Van der Wai, J. F., Further studies on theexposure of welders to fumes, chromium, nickeland gases in Dutch industries: plasma welding andcutting of stainless steel. Ann. Occup. Hyg., 30(2),153-161 (1986).
(184) Verschoor, M., Herber, R., Van Hemmen,
J., Wibowo, A., and Zielhuis, R., Renal function ofworkers with low-level cadmium exposure. Scand.J. Work, Environ. Health, 13, 232-238 (1987).
(185) Vogelmeier, C , Konig, G., Bencze, K., andFruhmann, G., Pulmonary involvement in zincfume fever. Chest, 92, 946-948 (1987).
(186) Voitkevich, V. G. and Fedorina E. E., De-termination of the concentrations of chromium andits compounds in the welding dust from stainlesssteel welding. Avtom. Svarka, 6, 52-55 (1987).
(187) Voitkevich, V. G., and Senkevich, A. I.,Study of the heterogeneity of the composition ofwelding dust particles by x-ray photoelectron spec-troscopy. Avtom. Svarka, 3, 34-38 (1987).
(188) Wilmer, J. W. G., Reuzel, P. G. J.,Hooftman, R. N., and Dreef-Van der Meulen, H. C ,The assessment of the carcinogenicity of chromiumand nickel containing welding fumes. Phase I: In vi-tro and in vivo genotoxicity and cytotoxicity stud-ies. Rapp. - TNO, Hoofdgroep VoedingVoedingsmiddelen TNO, ISS V 86.362/130370,(1986).
(189) Winburn, D. C, Laser protective eyewear:A critique. Welding J., 65, 33-37 (1986).
(190) Wiseman, L. G. and Chapman, E. T., Testprogram to determine whether nickel carbonylforms during the welding of nickel-containing al-loys. A progress report. In: Health Hazards and Bio-logical Effects of Welding Fumes and Gases.Proceedings International Conference, Copenha-gen, 18-25 Feb. 1985. Eds: R. M. Stern, A. Berlin,and J. Jarvisalo. Elsevier Science Publishers, Am-sterdam, 497-533 (1986).
(191) Wittgens, H., Eyedrops for arc welders.Arbeitsmed. Sozialmed. Praventivmed. (Germany,West), 21(2), 49-50(1986).
(192) Zober, A., Expert reporting on airway andlung diseases in welders. Arbeitsmed. Sozialmed.Praventivmed. (West Germany), 21(4), 89-92(1986).
(193) Zschiesche, W., Emmerling, G., Schaller, K.H., and Weltle, D., Investigations on renal impair-ments of high alloy steel welders. In: Occupationaland Environ. Chemical Hazards, Eds. V. Foa, E. A.Emmett, M. Maroni, and A. Colombi. Chichester,Ellis Horwood Limited. 339-343 (1987).
(194) Zschiesche, W., Wilhelm, E., Zober, A.,Schaller, K. H., Weltle, D., and Valentin, H., Man-ganese in stainless steel welding fumes: external ex-posure and biological monitoring. In: HealthHazards and Biological Effects of Welding Fumesand Gases. Proceedings International Conference,Copenhagen, 18-25 Feb. 1985. Eds: R. M. Stern, A.Berlin, and J. Jarvisalo. Elsevier Science Publish-ers, Amsterdam, 193-196 (1986).
