Journal of Clinical LaboratoryAnalysis 6:190-193 (1992)

A Simple ICP-MS Procedure for the Determination of Total Mercury in Whole Blood and Urine Ramaswami Kalamegham and K. Owen Ash Department of Pathology, University of Utah Medical Center and Associated Regional and University Pathologists Inc., Salt Lake City, Utah A simple and sensitive procedure for total mercury in whole blood and urine using inductively coupled plasma-mass spectrometry (ICP-MS) is described. Specimens are prepared by precipitation-extraction with 50% viv hydrochloric acid containing EDTA and cysteine, centrifuged, and filtered through fritended screening column; the filtrates are directly analyzed by ICP-MS. The method is linear between 2 and 200 pgiL in the specimen with an absolute sensitivity of 0.2 pg/L in the final supernatant. The assay variability at various concentrations (pgiL) of merKey words:

cury are as follows: intra-assay whole blood (n = 20)-4.6 t 0.6(c.v.12.3%),18.3 & 1.1 (c.v. 6.l%), 56.4 & 2.8(c.v. 5.0%); inter-assay whole blood (n = 1 5 b 5 . 7 t 1 .O(c.v. 16.8%), 19.7 t 2.7(c.v. 13.5YO),and50.1 6.9(c.v. 13.7%); urine (n = 20)-9.3 2 1.2 (c.v. 12.9%), 29.6 & 2.2 (c.v. 7.4%). Recovery of organic and inorganic mercury from blood samples ranges from 91.6% to 1 10.2%. The method is suitable for analysis of total mercury, both organic and inorganic, in whole blood and urine. Q 1992Wiley-~iss,Inc.

*

inorganic mercury, methylmercury, inductively coupled plasma-massspectrometry

INTRODUCTION Mercury is widely used in industry and dentistry. In addition to occupational exposure, the general population is exposed to mercury through the diet (fish being a dominant source of human exposure to methyl mercury) and dental amalgam (1). Both organic and inorganic forms of mercury can be toxic to humans (1,2) and the current threshold limit values (TLV) for mercury in air is 0.05 mg/m3 (3). The levels of total mercury in biological fluids of “unexposed” individuals are below 10 kg/L (1,2). Therefore, sensitive procedures are required for monitoring exposure to this metal. Many procedures require oxidative conversion of the organic forms to Hg2+,followed by reduction to the elemental form prior to analysis (4). Inductively coupled plasma-mass spectrometry (ICP-MS) is a highly sensitive technique for elemental analysis which requires simple sample preparation. ICP-MS uses an argon plasma (at around 6,000”K) to generate singly charged ions from the elemental species within a sample and a quadrupole mass spectrometer to separate and quantify the ions. The argon plasma is maintained (inductively coupled) by the interaction of a radio-frequency (RF) field and ionized argon gas. ICP-MS is widely employed for monitoring many elements including mercury in the environment (5,6), but to our knowledge, a procedure has not been described for mercury measurement in clinical specimens.

using an argon plasma, was equipped with a Meinhard nebulizer, type “C,” a glass spray chamber, and an interphase made of platinum cones with orifices of 0.045” (sampler) and 0.035” (skimmer). The parameterdconditions employed are as follows: plasma RF power, 1,250 Watts; auxiliary flow, 1.0 Limin; plasma flow, 14 Llmin; nebulizer pressure, 50 psi; scanning mode, Elemental; measurements/peak, 3; repeats/ integration, 5; Digipot settings Lens B, - 40 V; Lens P, 13 V; Lens E l , - 70 V; Lens S2, 39 V; interval between samples, 3 min.

Reagents Cysteine hydrochloride and EDTA-disodium salt were from Sigma Chemical Co (St. Louis, MO). Yttrium and mercury calibrators (1,000 pg/L in 10% nitric acid) were from SPEX Industries Inc (Edison, NJ) and were further diluted in 5% (v/v) HCI. Hydrochloric acid (E Merck, SUPRAPUR, 30% w/v) was from Curtin Matheson Scientific Inc. (Los Angeles, CA). Methyl mercuric chloride was from Alfa Products (Ward Hill, MA). S/P’” screening columns used for filtering acid extracts were obtained from Baxter Scientific Products Division (McGaw, IL). Pooled human blood was used for recovery studies and controls. Whole blood reference controls were kindly

MATERIALS AND METHODS Instrument

Received November 14, 1991; accepted February 18, 1992.

A PEiSCIEX ELAN 500 ICP-MS (Perkin Elmer, Norwalk, CT) was used for the determination of mercury. The system,

Address reprint requests to K. Owen Ash, ARUPLJUMC Pathology Dept., 500 Chipeta Way, University Research Park, Salt Lake City, UT 84108.

0 1992 Wiley-Liss, Inc.

ICP-MS Procedure for Total Mercury

provided by Dr. Daniel C. Paschal of the Center for Disease Control (Atlanta, GA). Urine reference controls were from the NIST Standard Reference Materials, Gaithersburg, MD.

Procedure While the extraction conditions and the acid concentration and type (nitric/hydrochloric) were varied during the method development, the final standardized procedure involved the following steps:

1. To 0.5 mL of blanWcalibrators/samples (urine or blood) were added and mixed 0.5 mL each of deionized water, 1.75% disodium EDTA solution, cysteine hydrochloride (0.25% w/v) solution in 5% (v/v) HCI), and internal standard solution (50 Kg/L Yttrium in 5% (v/v) HCI). 2. Concentrated HCI' (2.5 mL) was then added as a fine jet to ensure mixing during the addition. This was particularly important in the case of whole blood samples to minimize clumping and was easily achieved with an Eppendorf repipetor. Clumps observed with whole blood samples were broken up by vigorous shaking. 3. After standing at room temperature for 15 min the tubes were centrifuged for 5 min at 2,500 rpm. 4. The supernatants were then filtered through frit-ended screening columns to remove any flocculant precipitate staying at the top (seen often with whole blood) and analyzed for total mercury. When the effect of acid concentration on the recovery was studied, deionized water replaced the acid to make up the total volume to 5 ml.

RESULTS AND DISCUSSION ICP-MS has been employed for the determination of mercury in water (5,6), and in petroleum (7), as well as in the determination of methyl mercury in biological reference materials (8). However, to our knowledge, there has been no report on the use of ICP-MS in the determination of mercury in clinical specimens. Our initial attempts using procedures employed for other heavy metals revealed problems of poor/erratic recovery and low ion counts. While the degree of ionization for most elements is >90%, it is much lower for mercury, being around 38% (9). In view of these considerations, optimal conditions for complete recovery and maximal ionization were investigated.

'Suprapur'" concentrated HCI employed in this study is 30% wiv compared to the commercial reagent grade HCl, which is 36-38% wiv.

191

InstrumentalConditions Ion lenses were optimized for maximum ion counts for mercury. Earlier studies have reported that settings for the ion focus lens affected the signal strength of the analyte ion, with lens B being the most critical for sensitivity and precision (6). Digipot settings found optimal in our hands are described under instrumental parameters/conditions above (see Materials and Methods). Peak measurement and integration repeat settings were selected to improve precision at low ion counts. The assay was not affected by the memory effect, when the interval between samples was increased to 180 s . Initial attempts to monitor all major naturally abundant isotopes of mercury revealed that the background counts also increased and thus offered no improvement over monitoring just one of the more abundant isotopes (either Hg2O0 or Hg202). We have used Hg2O0 in all subsequent experiments as has been done by others (5). Such monitoring would also permit the use of isotope dilution for internal correction, if required. The background ion-counts observed with blank ranged from 90 to 120.

ExperimentalConditions Dilute nitric acid, normally preferred for sample preparation in ICP-MS, and successfully employed in the case of other elements, proved useless in the case of mercury because of poor recovery. Increasing the nitric acid concentration did not improve recovery. Hydrochloric acid, which converts methyl mercury to its chloride form before extraction procedures (lo), provided a good extraction medium for methyl mercury. Using 25% (v/v) HCI, methyl mercury spiked into whole blood at 5 and 25 pg/L was recovered at 91.9% and 95.2%, respectively. The concentration of HCI in the extraction medium was, however, found to be critical for the quantitative recovery of inorganic mercury from blood as shown in Figure 1. A concentration of 45-50% (viv) HCI was required to give complete recovery of inorganic mercury. Higher concentrations (>%% v/v) of HCI solubilized proteins that slowly precipitated on standing. These precipitates tended to clog the nebulizer requiring frequent cleaning of the nebulizer and torch adversely affecting the precision. At the end of mercury runs, the sample aspirator was left in 50% v/v nitric acid for 5 min to clean the line, nebulizer, and torch. The nebulizer and torch were cleaned once a week. Cysteine and EDTA released protein-bound Hg and stabilized it in the supernatant thereby improving the precision.

Assay Linearity, Precision, and Recovery The assay was linear from 2 to 200 pg/L of total mercury (Fig. 2), with an absolute sensitivity of 0.2 Fg/L in the final supernatant. The intra- and inter-assay imprecision data have been illustrated in Tables 1 and 2 . Recovery of both organic and inorganic mercury ranged from 92 to I 10% (Table 3) and total mercury values obtained for reference materials agreed well with reported values (Fig. 3).

192

120 1

loo h

s

80-

Y

*er

Lu

60-

w

Ya

4020

-

01

10

I

I

20

30

40

50

60

CONC HCI (VOLUME PERCENT) Fig. 1. Recovery of inorganic mercury from whole blood as a function of hydrochloric acid concentration.

HgZfwas spiked to whole blood at a final concentration of 20 (Ig/L.

F 2 3

0 0

z 0

0

100

200

Hg CONCENTRATION ( I g l L ) Fig. 2.

TABLE 1. Intra-Assay Variability (n in Blood Analyzed by ICP-MS

=

Linearity of total mercury analyzed by ICP-MS.

20) of Total Mercury

SD (pg/L)

Cv(%)

I

4.6

0.6

12,3

I1

18.3

1.1

6 .I

111

56.4

2.8

5.0

Level

Mean (@L)

Cold vapor atomic absorption spectrometry (CV-AAS) has now become the workhorse for the determination of Hg in biological fluids (1,4), although it may be subject to interference by aromatic hydrocarbons that absorb strongly at 253.7 nm region. The introduction of ICP-MS technique that is free of such spectral interference while still offering high sensi-

ICP-MS Procedurefor Total Mercury

TABLE 2. Inter-Assay Variability of Total Mercury in Whole Blood and Urine Analyzed by ICP-MS

TABLE 3. Recovery of Mercury From Whole Blood

Level

Sample

n

WB- 1 WB-2 WB-ZIORG" WB-3

4

WB-310RG" WB-6

Whole blood (n

Mean (p,g/L) =

=

CV(%)

5.7 19.7 50.1

1.o 2.7 6.9

16.8 13.5 13.7

9.3 29.6

1.2 2.2

12.9 7.4

15)

1 11 111

Urine control (n Low High

SD M I L )

20)

__

.

i m

yY = 1.127

+

193

Amount spiked WIL)

Amount recovered (mean 2 SD)

% recovery (mean ? SD)

8 6 2

5 20 20 50

1 1

100

5.0 2 0.2 20.5 ? 0.5 18.3 2 1 . 1 49.8 47.8 55.1 106.2

100.5 ? 4.4 102.8 ? 2.6 91.6 ? 5.6 99.6 95.6 110.2 106.2

50

aMethyl mercury.

0.960X 0 . 9 6 0 ~ R"2

=

1.000

.

.

100-

3.

v

W

80-

3 J

U

> n W > K

60

-

/

40 40-

W

tn

m 0

20-

,

0

20

-

m

-

40

, 60

-

80

I

100

-

.

120

REFERENCE VALUE (pg/L) Fig. 3. Comparison of total mercury values obtained with ICP-MS procedure with reference values obtained by cold vapor atomic absorption spectrometry and reported by the Center for Disease Control and NIST. ORG designates the assay specimen containing organic mercury.

tivity and excellent recovery should provide researchers and clinical laboratories a good alternative method for total mercury determinations in whole blood and urine. The ICP-MS procedure for total mercury in biological fluids described here is simple and can be modified if required to estimate methyl mercury by first extracting the methyl mercury into benzene or toluene and then back extracting into aqueous cysteine solution before ICP-MS analysis (8).

ACKNOWLEDGMENTS We wish to acknowledge the gift of whole blood reference control samples by Dr. Daniel C. Paschal of the Center for Disease Control, (Atlanta, GA), and the technical help of Bill Gordon and Marilyn Cooper.

REFERENCES 1. WHO: Inorganic Mercury, Environmental Health Criteria 118. World Health Organization, Geneva, 1991. 2. WHO: Methyl Mercury, Environmental Health Criteria 101. World Health Organization, Geneva, 1990.

3. American Conference of Governmental Industrial Hygienists: Threshold limit values for chemical substances and physical agents in the workroom environment with intended changes for 1976. Cincinnati, OH, American Conference of GovernmentalIndustrial Hygienists, 1976, p 94. 4. Boiteau HL, Pineau A: Mercury. In Quantitative Trace Analysis ofBio1ogicalMaterials HA McKenzie, LE Smythe, eds. Elsevier Science Publishers, Amsterdam, 1988, p 553-560. 5. Pruszkowski E, Ediger RD, Denoyer E: ICP-MS: An alternative method for the USEPA contract laboratory program. Am Lab 21 :36-39, 1989. 6. Powell MJ, Boomer DW, McVican RJ: Introduction of gaseous hydrides into an inductively coupled plasma mass spectrometer. Anal Chem 58:2864-2867,1986. 7. Osborne SP: Quantitation of mercury in petroleum hy ETV-ICP-MS. Appl Spectrosc 44: 1044- 1046, 1990. 8. Beauchernin D, Siu KWM, Berman SS: Determination of organomercury in biological reference materials by inductively coupled plasma mass spectrometry using flow injection analysis. Anal Chem 60:25872590, 1988. 9. Houk RS: Mass spectrometry of inductively coupled plasmas. Anal Chem 58:97A-l05A, 1986. 10. AOAC: Mercury (Methyl) in fish and shellfish. In O@cial Methods of Analysis ofthe AOAC, 15th Ed, Hilrich, K, ed. Association of Official Analytical Chemists, Arlington, VA, 1990, p 266-269.