Acute nickel toxicity in electroplating workers who accidently ingested a solution of nickel sulfate and nickel chloride.код для вставкиСкачать
American Journal of Industrial Medicine 14:257-266 (1988) Acute Nickel Toxicity in Electroplating Workers Who Accidently Ingested a Solution of Nickel Sulfate and Nickel Chloride F. William Sunderman, Jr., MD, Brian Dingle, MD, Sidney M. Hopfer, PhD, and Thomas Swift, PhD Thirty-two workers in an electroplating plant accidently drank water contaminated with nickel sulfate and chloride (1.63 g Ni/liter). Twenty workers promptly developed symptoms (e.g., nausea, vomiting, abdominal discomfort, diarrhea, giddiness, lassitude, headache, cough, shortness of breath) that typically lasted a few hours but persisted 1-2 days in 7 cases. The Ni doses in workers with symptoms were estimated to range from 0.5 to 2.5 g. In 15 exposed workers who were tested on day 1 postexposure, serum Ni concentrations ranged from 13 to 1,340 p,g/liter and urine Ni concentrations ranged from 0.15 to 12 mg/g creatinine. Ten subjects (with initial urine Ni concentrations >0.8 mg/g creatinine) were hospitalized and treated for 3 days with intravenous fluids to induce diuresis, resulting in a mean elimination half-time (T%) for serum Ni of 27 hours (SD +- 7 hour), which was significantly shorter (p < .001) than the mean T% of 60 hours (SD I 1 hours) in 1 1 subjects who did not receive intravenous fluids. Laboratory tests showed transiently elevated levels of blood reticulocytes (N = 7), urine albumin (N = 3), and serum bilirubin (N = 2). All subjects recovered rapidly, without evident sequellae, and returned to work by the eighth day after exposure. * Key words: metal poisoning, acute nickel toxicity, oral INTRODUCTION The literature on acute oral toxicity of nickel in humans is limited to three papers over a span of more than a century. Da Costa [ 18831 administered nickel salts to patients as a treatment for diarrhea and epilepsy. He noted that oral doses of 1-3 grains (65-195 mg) of nickel sulfate were well tolerated and therapeutically beneficial, whereas 5 grains (325 mg) induced giddiness, nausea, variable slowing of the pulse, and slight reduction of body temperature. Assuming that Da Costa used the common hexahydrate salt of nickel sulfate, the 5-grain dose contained 73 mg of nickel. Faichnie and Langrishe [ 19101 reported accidental nickel poisoning of five Departments of Laboratory Medicine (F. W.S., S .M.H., T. S.) and Pharmacology (F.W.S., S.M.H.), University of Connecticut Medical School, Farmington. Department of Medicine, Kitchener-Waterloo General Hospital, Kitchener, Ontario, Canada (B.D.). Address reprint requests to F. William Sunderman, Jr., MD, Northeast Utilities Professor of Toxicology, University of Connecticut Medical School, 263 Farmington Avenue, Farmington, CT 06032. Accepted for publication March 15, 1988. 0 1988 Alan R. Liss, Inc. 258 Sunderman, Jr., et al. British soldiers who drank lime-flavored barley water that had been mixed in a large nickel urn with a central chamber for ice. Within 15 minutes, the soldiers became acutely ill with nausea that lasted 1 day. Qualitative analysis showed that the beverage contained nickel, apparently dissolved as nickel citrate. Daldrup et al. [ 19831 described fatal poisoning of a girl, age 2.5 years, who accidently ingested 10-15 g of nickel sulfate hexahydrate crystals (- 2.2-3.3 g of Ni) from a chemistry hobby set. The child rapidly became stuporous and developed nuchal rigidity, erythema, dilated pupils, tachycardia, and pulmonary congestion. Despite repeated resuscitations after cardiac arrests, the child died 8 hours after the poisoning, and the autopsy revealed acute hemorrhagic gastritis. The present paper describes a recent industrial accident in which 32 electroplating workers inadvertently drank water that was heavily contaminated with nickel sulfate and chloride. DESCRIPTION OF THE EXPOSURE The episode began on Saturday afternoon, June 13, 1987, when the main water supply to a metal-plating factory was temporarily shut down for repairs. The plant, which employs 338 workers, operates several automated electroplating lines used primarily to refinish automobile bumpers. Except for a few maintenance and repair personnel, the plant was unstaffed over the weekend until the evening shift reported for work at 1 P.M. on Sunday the 14th of June. After 3 P.M. several workers on a nickel-plating production line began to feel ill with symptoms that some attributed to bitter-tasting water from a drinking fountain. The supervisor on the night shift tested water samples from seven drinking fountains and found that the sample from the suspect fountain was green and contained nickel (approximately 2 g/liter) while the samples from the other fountains were colorless and uncontaminated with nickel. The water supply to the contaminated fountain, which was adjacent to a nickel-plating tank, was flushed and the fountain was disconnected at 3 A.M. on Monday, the 15th of June. Since the ambient temperature was warm, most workers on the evening and night shifts sweated profusely; the workers who developed symptoms evidently had ingested 0.5-1.5 liters of water from the contaminated fountain. The green tint of the water was inapparent under the factory lighting when viewed against the metallic gray background of the stainless-steel bowl. The nickel contamination of the drinking water was caused by back-siphonage from a water recirculation system that cooled filtration pumps for the nickel-piating tank. Leakage of pump gaskets had allowed the nickel-plating solution (nickel sulfate, NiS04 * 6H20, 310 g/liter; nickel chloride, NiC12 * 6H20, 127 glliter; boric acid, H3B04, 36 g/liter) to seep into the water recirculation system. The recirculation system was connected via an open valve to a freshwater line. When the water main was shut down, nickel-contaminated water was sucked from the recirculation system into the freshwater line, from which the drinking fountain was directly tapped. SUBJECTS AND METHODS Three groups of subjects were studied: Group A consisted of 21 exposed workers, 18 with symptoms, who were first examined on June 15; ten of these workers (group A l ) were admitted to Kitchener-Waterloo General Hospital; the rest (group A 2 ) were followed as outpatients. Group B consisted of 11 exposed workers, Acute Nickel Toxicity in Plating Workers 259 two with mild symptoms, who were not examined until 17 June. Group C (controls) consisted of 11 workers on the same production line, who did not drink from the contaminated fountain on the 14th or 15th of June. The 43 subjects were men, age 17-36 years (mean ? SD, 26 2 5 years), in good health, excepting one with diabetes mellitus, one with bronchial asthma, and one with chronic obstructive lung disease. At least two of the subjects were heavy drinkers of alcoholic beverages and most smoked cigarettes. A sample of water collected from the contaminated fountain on Sunday evening June 14, and 21 1 specimens of urine or serum, collected from the subjects on 15, 16, 17, or 19 June, were promptly flown by chartered airplane to the University of Connecticut Health Cecter where they were analyzed for nickel by electrothermal atomic absorption spectrophotometry with Zeeman background correction [Sunderman et al., 19881. In addition to nickel, the water sample was tested for boron, the urine specimens were examined by routine urinalysis and analyzed for creatinine and albumin, and the serum specimens were analyzed for urea, creatinine, creatine kinase, lactate dehydrogenase, aspartate aminotransferase, alanine aminotransferase, and alkaline phosphatase. Routine clinical laboratory techniques were used for these assays, except boron, which was analyzed by a spectrophotometric method [Baselt, 19801, and albumin, which was analyzed by radioimmunoassay [Horak, 19871. The ten workers who were admitted to Kitchener-Waterloo General Hospital (group A l ) had the following additional tests: chest X-ray, electrocardiogram, creatinine clearance, urine total protein, blood leukocyte, platelet and reticulocyte counts, blood hemoglobin, pH, pC02 and PO*, serum glucose, sodium, potassium, chloride, bicarbonate, bilirubin, calcium, cortisol, prothrombin time, partial thromboplastin time, fibrinogen, and fibrin split products. Follow-up studies , including physical examination, urinalysis, blood hemoglobin, and reticuloccyte count, were performed 6 weeks postexposure. Urine Ni concentrations were factored on the basis of creatinine concentrations as recommended by Sunderman et al. [ 19861. Statistical computations (e.g., linear and exponential regression, analysis of variance, Student’s t-test, Fisher’s exact test, and Mann-Whitney U test) were performed according to Sachs . RESULTS Estimated Exposures The water sample analyzed at the University of Connecticut had a bright green color and contained nickel (1.63 g/liter) and boron (68 mg/liter). The estimated oral intake of nickel by the 20 symptomatic workers ranged from 0.5 to 2.5 g; the estimated oral intake of boron ranged from 20 to 100 mg. Symptoms The following complaints were voiced by the 20 symptomatic workers: nausea (N = 15), abdominal cramps or discomfort (N = 14), giddiness (N = 7), lassitude (N = 6), headache (N = 5), diarrhea (N = 4), vomiting (N = 3), shortness of breath (N = 2), cough (N = l ) , sore throat (N = l ) , heartburn (N = l ) , myalgia (N = l), and dysgeusia (N = 1). The symptoms developed during the work shift and generally lasted only a few hours. However, nausea, abdominal cramps, diarrhea, headache, and/or lassitude persisted 1-2 days in seven of the ten hospitalized subjects (group Al). All subjects became asymptomatic within three days postexposure. 260 Sunderman, Jr., et al. Initial Clinical Observations When first seen, most workers in group A showed anxiety and abdominal guarding, but their physical findings were generally unremarkable. The only positive findings were 1) expiratory wheezing in the subject with bronchial asthma, 2) dyspnea and mild cyanosis in the subject with chronic obstructive pulmonary disease, and 3) borderline hepatomegaly in two subjects who admitted heavy alcohol intake. Workers in group B , first examined 3 days postexposure, were essentially normal upon physical examination. No abnormalities were found by the routine laboratory tests (urinalysis, blood count) that were performed during the initial clinical evaluations. Nickel Concentrations in Body Fluids In specimens collected from 15 subjects in group A at 4 P.M. on 15 June, serum Ni concentrations averaged 286 pg/liter, (range 12.8-1,340 pg/liter); urine Ni concentrations averaged 5.8 mg/liter (range 0.23 -37.1 mg/liter), equal to 2.4 mg/g creatinine (range 0.15-12.0 mg/g creatinine). In 11 control nickel-plating workers (group C) serum Ni concentrations averaged 4.0 pg/liter (SD 2 1.2; range 2.0-6.5 pg/liter); urine Ni concentrations averaged 50 pg/liter (SD 2 13; range 22-70 pglliter), equal to 29 pg/g creatinine (SD k 7; range 17-41 pg/g creatinine). For comparison, the following reference values for healthy hospital workers were obtained in our laboratory: serum Ni = 0.28 pg/liter (SD k 0.24; range <0.05-1.08 pg/liter; N = 30); urine Ni = 2.0 pg/liter (SD 1.5; range 0.5-6.1 pg/liter; N = 34), equal to 2.0 pg/g creatinine (SD k 1.5; range 0.4-6.0 pg/g creatinine; N = 34) [Sunderman, 19871. Ten workers (group A l ) , whose initial urine nickel concentrations exceeded 1.5 mg/liter (0.8 mg/g creatinine) were admitted to the hospital for observation and treatment by diuresis. These patients received iv infusions of sodium chloride solution (0.15 mol/liter) at 150 ml/hour for 3 days, resulting in urine volumes that averaged 3,680 ml (SD k 760 ml) and 3,330 ml (SD 2 1,070 ml) on the second and third hospital days. Serial diminutions of Ni concentrations in serum and urine specimens from the ten individual subjects in group Al are plotted in Figure 1. Mean Ni concentrations in serum and urine specimens from subjects in groups A], A2, B , and C on days 1, 2, 3, and 5 after the exposure incident are listed in Table I and shown in Figure 2. The Ni concentrations in specimens from the second group of exposed subjects (group B ) on days 3 and 5 were slightly higher than those from group A2. As expected, no significant changes of Ni concentrations were noted from day to day in specimens from group C (controls). Close correlation (R = 0.97) was observed between Ni concentrations in the 103 paired specimens of serum and urine from the 43 subjects (groups A], A2, B , and C); the urine Ni concentrations (pg/g creatinine) averaged 8.3 (SE k 0.2) times the corresponding serum Ni concentrations @&liter) (Fig. 3). * Toxicokinetics Elimination half-times (TY2) of serum Ni for individual subjects in groups A I and A2 were computed by a one-compartment mathematical model [Gibaldi and Perrier, 19821. The model adequately described the data as shown by correlation coefficients (R) of 0.963 (SD Ifr 0.038) and 0.946 (SD 0.041) in groups A, andA2, * 261 Acute Nickel Toxicity in Plating Workers 1 1 .c .w a g 1000: 0) \ ?! Y 2 100: Q) c .- ‘ 03 1 0 1 2 3 4 5 Days post-exposure Fig. 1. Serial measurements of nickel concentrations in serum and urine specimens from the ten individual subjects in group A1 during 5 days postexposure. * respectively. The serum Ni T% for subjects in group A] averaged 27 hours (SD 7), which was significantly shorter (p < .001) than the mean TY2 value of 60 h (SD 11) for subjects in group A2, who did not receive intravenous fluids to induce diuresis. The serum Ni T% values were not significantly correlated with the initial concentrations of serum Ni in subjects of group A] or AZ. * Diagnostic Tests Two subjects in group A] and one subject in group A2 had urine albumin concentrations above the upper limit of the laboratory’s reference range (16 mg/g creatinine) on day 2 postexposure. The elevated concentrations of urine albumin (68, 40, and 27 mg/g creatinine, respectively) diminished on day 3 and returned to normal by day 5 , suggesting mild, transient nephrotoxicity. One of the subjects with albuminuria had concentrations of serum urea nitrogen (22 and 26 mg/dl) on days 2 and 3 postexposure that slightly exceeded the upper limit of the laboratory’s reference range (20 mgldl). The nickel-exposed subjects in groups A ] , A2, and B did not have significantly elevated results of the other clinical chemistry assays that were performed at the University of Connecticut Health Center (creatinine, creatine kinase, lactate dehydrogenase, aspartate aminotransferase, alanine aminotransferase, and alkaline phosphatase) on serums collected 1 , 2 , 3, or 5 days postexposure, compared to the results in control subjects (group C ) . 262 Sunderman, Jr., et al. TABLE I. Nickel Concentrations in Serum and Urine Specimens From Workers in the Various Exposure Categories ~~ Category of exposure and No. of subjects Group A , (N = 1O)a Days postexposure 1 2 3 5 Group A2 (N = Group B (N = 11)' Group C (controls) (N = ll)d Ni concentration (mean f SE) Serum (pgiliter) 460 2 310 & 145 t 38 Urine (pg/g creat.) 140 3,760 f 1,230 1,870 f 480 1,290 f 320 304 68 100 41 * *9 27.8 & 15.5 12.2 7.7 r 7.2 269 f 82 138 f 20 90 f 19 60 f 13 * 1.2 * 1.6 3 5 1.0 18.6 f 12.9 14.7 & 3.4 140 f 69 143 f 31 3.8 & 0.4 3.9 k 0.3 2.8 t 0.5 4.5 f 0.8 28 31 23 30 4 2 2 2 5 +- 5 2 aHeavily exposed workers in the initial study group, who were admitted to the hospital and treated by diuresis. bExposed workers in the initial study group, who were followed as outpatients. CExposed workers who were first examined 3 days after the incident. dControl workers on the same production line, who did not drink from the contaminated water fountain. Tests performed at Kitchener-Waterloo Hospital on the ten subjects in group Al showed that the average concentration of blood hemoglobin increased from 15.1 g/dl (SD & 0.7; range 14.0-16.1 g/L) on day 3 postexposure to 16.0 g/liter (SD k 0.6; range 15.3-17.1 g/L) on day 8 (p < .01 by paired-sample t-test). Blood reticulocyte counts increased from an average of 78 X lo9 cells/liter (SD k 44;range 34-139 X lo9) on day 3 postexposure to 106 X lo9 cells/L (SD rt 49; range 48-217 X lo9) on day 8 (p < .01 by paired-sample t-test). Blood reticulocyte counts exceeded the laboratory's upper reference limit (75 X lo9 cells/liter) in seven of nine subjects in group Al on day 8 postexposure. Mild hyperbilirubinemia developed on day 3 postexposure in two of the subjects (serum bilirubin levels = 43 and 30 pmol/liter, respectively, compared to the upper reference limit of 20 pmol/liter). Electrocardiograms showed first-degree heart block on days 3-5 postexposure in one patient in group A l . Results of chest X-rays and the other laboratory tests mentioned under "Subjects and Methods" were normal throughout the patients' hospital stay. Clinical Course Subjects in group Al had slightly diminished body temperatures on days 2-4 postexposure. Sublingual temperatures, measured at 7 A.M. with an electronic digital thermometer, averaged 36.7"C (SD k 0.3; range 36.2-37.2"C) on day 2, 36.5"C (SD 2 0.4; range 36.0-37.1"C) on day 3, and 36.5"C (SD k 0.3; range 36.0-36.9"C) on day 4. Their basal pulse rates at 7 A.M. were normal, averaging 74 (SD & 10) beats/min on day 2, 66 k 9 beatslmin on day 3, and 72 6 beats/min on * Acute Nickel Toxicity in Plating Workers - 1000 263 3 h g) .- 10000 .c (I m g 1000 P 0) 2 v 2 100 a, .-(I 5 lo 1 2 3 5 Days post-exposure Fig. 2. Nickel concentrations (mean -+ SE) in serum and urine specimens from the control workers (group C , solid bars, N = 11) and the three groups of workers who drank nickel-contaminated water (group A,, bars with dark diagonal hatching, N = 10; group Az, stippled bars, N = 11; group B , bars with light diagonal hatching, N = 11) during 5 days postexposure. day 4. The patients’ hospital courses were uncomplicated and all were discharged from the hospital on day 5 postexposure. All subjects in groups A I , A Z , and B returned to work by day 8. No abnormalitiesor sequellae were detected in the workers who returned for follow-up examination. At 6 weeks postexposure, the average hemoglobin concentration in blood specimens from subjects in group Al decreased to 15.3 g/liter (SD k 0.8; range 14.4-16.6 g/liter) and the average reticulocyte count decreased to 81 X lo9 cells/liter (SD % 41; range 30-168 X lo9 cells/L). DISCUSSION The acute gastrointestinal and neurological symptoms that developed in the workers who drank nickel-contaminated water were generally similar to those described by previous authors [Da Costa, 1883; Faichnie and Langrishe, 19101. Slowing of the pulse, mentioned by Da Costa  as a sign of nickel toxicity, was not observed in the present study, except for one subject with first-degree heart block. Diminished body temperature, another sign mentioned by Da Costa, was noted in the hospitalized patients in this study, consistent with findings of Gordon and Stead [ 19851 that NiCI,-treated mice develop hypothennia and impaired thermoregulation. The symptoms of acute nickel toxicity in this study resemble those noted by Webster et al.  in 23 hemodialysis patients, when leaching of nickel-plated stainless 264 Sunderman, Jr., et al. 1000 7 10 100 1000 10000 Urine Ni (pg/g creatinine) Fig. 3 . Correlation of nickel concentrations in 103 pairs of serum and urine specimens from subjects in groups A,, A2, B , and C . The regression equation is Y = 0.14X - 3.90, where “X” denotes urine Ni concentration (pg/g creatinine) and ‘ ‘Y” denotes serum Ni concentration (pgiliter). The correlation coefficient (R) is 0.97. steel in a water-heater tank caused nickel contamination of dialysis fluid. Their symptoms (nausea, vomiting, weakness, headache, palpitations) developed during and after dialysis and typically lasted 3-13 hours but occasionally persisted up to 48 hours. In the present study, the ten exposed workers whose initial nickel concentrations in urine exceeded 0.8 mg/g creatinine were administered intravenous fluids to induce pronounced diuresis. This therapeutic approach was based on the recommendation of Mertz et al. 119701 that active renal elimination of nickel by maximal diuresis could be applied therapeutically in nickel poisoning. As predicted by Mertz and co-workers, induction of diuresis reduced the mean elimination half-time (T%) of serum Ni to 27 hours in group A l , compared to 60 hours in group AZ, which did not receive intravenous fluids. Although chelating agents were ready for use in the event that a patient’s condition deteriorated, chelation therapy did not become necessary. Since the estimated oral intake of boron by the 20 symptomatic workers was only 20-200 mg, boric acid poisoning probably did not contribute significantly to the present episode. The boron content of drinking water ranges from 0.4 to 5 mg/liter and the boron intake in food and water ranges from 0.3 to 20 mg/day for adult persons [Larsen, 19871. Adults can ingest 4 g of boric acid (- 0.7 g of boron) without incident, whereas a single oral dose of 18-20 g (- 3.2-3.5 g of boron) is probably lethal; chronic intoxication may occur after daily consumption of 4-5 g of boric acid (- 0.7-0.9 g of boron) for 3-4 weeks [Valdes-Dapena and Arey, 19621. The mild erythrocytosis and reticulocytosis that developed in nickel-exposed subjects in group Al is noteworthy since cobalt salts are known to stimulate erythropoiesis in humans [Duckham and Lee, 19761 and intrarenal injections of nickel sulfides and oxides cause erythrocytosis in rats [Sunderman et al., 1982, 19871. These hematological effects of cobalt and nickel are mediated by enhanced renal production of erythropoietin [Hopfer et al., 1984; Beru et al., 19861. The transient hyperbilirubinemiathat occurred in two nickel-exposed subjects may be secondary to nickel induction of microsomal heme oxygenase activity in liver and kidney, a phenomenon that has been extensively studied in rodents [Sunderman, 1987bl. Acute Nickel Toxicity in Plating Workers 265 Nickel concentrations in serum and urine specimens from the control electroplating workers (group C) are comparable to those previously observed in other nickel-plating operations [Bernacki et al., 1978, 1980; Tola et al., 19791. For further information on biological monitoring of industrial exposures to nickel compounds and detailed discussions of nickel toxicology in humans and experimental animals, readers may consult recent monographs and reviews [Sunderman, 1984, 1987a; Brown and Sunderman, 1985; Norseth, 19861. ACKNOWLEDGMENTS We thank Tom T. Liu, M.D., Bernard A. Haberstroh, M.D., and H. John Lachner, M.D., for assistance with medical care, Eva Horak, Ph.D., for assistance with radioimmunoassays of urine albumin, and Kevin R. Sweeney, Ph.D., for assistance with toxicokinetic computations. This study was supported in part by the U.S . Environmental Protection Agency, the Nickel Producers’ Environmental Research Association, and the Northeast Utilities Company. REFERENCES Baselt RC (1980): Borate. In Baselt RC (ed): “Biological Monitoring Methods for Industrial Chemicals.” Davis, CA: Biomedical Publications, pp 50-52. Bernacki EJ, Parsons GE, Roy BR, Mikac-Devic M, Kennedy CD, Sunderman FW, Jr (1978): Urine nickel concentrations in nickel-exposed workers. Ann Clin Lab Sci 8: 184-189. Bernacki EJ, Zygowicz E, Sunderman FW, Jr (1980): Fluctuations of nickel concentrations in urine of electroplating workers. Ann Clin Lab Sci 10:33-39. Beru N, McDonald J, LaCombe C, Goldwasser E (1986): Expression of the erythropoietin gene. Mol Cell Biol 6:257 1-2575. Brown SS, Sunderman FW, Jr (eds) (1985): “Progress in Nickel Toxicology.” Oxford: Blackwell. Da Costa JM (1883): Observations on the salts of nickel, especially the bromide of nickel. Med News 431337-338. Daldrup T, Haarhoff K, Szathmary SC (1983): Toedliche nickel-sulfate-intoxikation.Bericht Gerichtl Med 41:141-144. Duckham JM, Lee HA (1976): The treatment of refractory anaemia of chronic renal failure with cobalt chloride. Q J Med 45:277-294. Faichnie N, Langrishe JP (1910): Nickel poisoning from drinking barley water. J R Army Med Corps 15484. Gibaldi M, Perrier D (1982): “Pharmacokinetics.” 2nd Ed. New York: Marcel Dekker. Gordon CJ, Stead AG (1985): Effect of nickel and cadmium chloride on autonomic and behavioral thermoregulation in mice. Neurotoxicology 7:97-106. Hopfer SM, Sunderman FW, Jr, Reid MC, Goldwasser E (1984): Increased immunoreactive erythropoietin in serum and kidney extracts of rats with Ni&-induced erythrocytosis. Res Commun Chem Pathol Pharmacol 43:299-305. Horak E (1987): Radioimmunoassay of urinary albumin for detection of nephrotoxicity. In Sunderman FW (ed): “Manual of Procedures for Clinical and Analytical Toxicology. ” Philadelphia: Institute for Clinical Science, pp 207-216. Larsen LA (1987): Boron. In Seiler HG, Sigel H (eds): “Handbook on Toxicity of Inorganic Compounds.” New York: Marcel Dekker, pp 129-141. Mertz DP, Koschnick R, Wilk G (1970): Renale Ausscheidungs-bedingungen von Nickel beim Menschen. Z Klin Chem Klin Biochem 8:387-390. Norseth T (1986): Nickel. In Friberg L, Nordberg GF, Vouk VB (eds): “Handbook on the Toxicology of Metals.” Amsterdam: Elsevier, Vol 11, pp 462-481. Sachs L (1984): “Applied Statistics: A Handbook of Techniques.” New York: Springer-Verlag. 266 Sunderman, Jr., et al. Sunderman FW, Jr (ed-in-chief) (1984): “Nickel in the Human Environment.” Lyon: International Agency for Research on Cancer. Sunderman FW,Jr (1987a): Nickel. In Seiler HG, Sigel H (eds): “Handbook on Toxicity of Inorganic Compounds.” New York Marcel Dekker, pp 454-468. Sunderman FW,Jr (1987b): Metal induction of heme oxygenase. Ann NY Acad Sci 514:65-80. Sunderman FW, Jr, Hopfer SM, Reid MC, Shen SK, Kevorkian CB (1982): Erythropoietin-mediated erythrocytosis in rodents after intrarenal injection of nickel subsulfide. Yale J Biol Med 55: 123-1 26. Sunderman FW, Jr, Aitio A, Morgan LG, Norseth T (1986): Biological monitoring of nickel. Toxicol Ind Health 2:17-78. Sunderman FW,Jr, Hopfer SM, Knight JA, McCully KS, Cecutti AG, Thornhill PG, Conway K, Miller C, Patierno SR, Costa M (1987): Physicochemical characteristics and biological effects of nickel oxides. Carcinogenesis 8:305-3 13. Sunderman FW, Jr, Hopfer SM, Crisostomo MC (1988): Nickel analysis by atomic absorption spectrometry. Methods Enzymol 158:382-39 1. Tola S, Kilpio J, Virtamo M (1979): Urine and plasma concentrations of nickel as indicators of exposure to nickel in an electroplating shop. J Occup Med 21:184-188. Valdes-Dapena MA, Arey JB (1962): Boric acid poisoning: Three fatal cases with pancreatic inclusions and a review of the literature. J Pediatr 61531-546. Webster JD, Parker TF, Alfrey AC, Smythe WR, Kubo H, Neal G, Hull AR (1980): Acute nickel poisoning by dialysis. Ann Intern Med 92631-633.