Journal of Analytical Toxicology, Vo]. 27, January/February 2003
Determination of Chromate Adulteration of Human
Urine by Automated Colorimetric and Capillary Ion
ElectrophoreticAnalyses"*
Kenneth E. Ferslew 1,t, Andrea N. Nicolaides 1, and Timothy A. Robert 2
ISection of Toxicology, Department of Pharmacology, East TennesseeState University, Johnson City, Tennessee37614 and
2Aegis Analytical Laboratories, 345 Hill Avenue, Nashville, Tennessee37210
Abstract ]
Variouschemicalscan be added to urine specimenscollected for
drug analysisto abnormallyelevate ionic concentrationsand/or
interfere with either immunoassay urine drug-screeningprocedures
or gas chromatographic-mass spectrometricconfirmation
techniques. One such adulterant, "Urine luck" (formula 5.3), has
been identified in our previous research to contain potassium
dichromate. Screeningof suspected adulterated specimensand
confirmation of the adulterant are important for forensic drug
screening. The application and comparison of automated
colorimetric and capillary ion electrophoretic techniques for the
detection, confirmation, and quantitation of chromate adulteration
of urine specimens were the purpose of this investigation.Thirty-six
urine specimens suspected of adulteration were analyzed for
chromate by colorimetric analysiswith diphenylcarbazide.
Duplicate aliquots were analyzed for chromate by capillary ion
electrophoresis. Resultsof the colorimetric chromate analyses
revealed a mean chromate concentration of 929 pg/mL with a
standard error of 177 pg/m/and a range of 30 to 5634 pg/mL.
Results of the capillary ion electrophoresis chromate analyses
revealed a mean chromate concentration of 1009 pg/ml with a
standard error of 218 pg/mL and a range of 20 to 7501 pg/ml.. The
correlation coefficient between the capillary ion electrophoretic
and colorimetric chromate results was r = 0.9669. Application of
the automated diphenylcarbazide colorimetric technique provides
rapid determination of chromate adulteration of a urine specimen.
Capillary ion electrophoresis offers a separation technique to
confirm the presence of chromate in suspectedadulterated
specimens. The excellent correlation between thesemethods
substantiates their application to forensic testing as screening
and/or confirmation techniques.
for drug analysis to abnormally elevate ionic concentrations,
alter pH, and/or interfere with either immunoassay urine drugscreening procedures or gas chromatographic-mass spectrometric (GC-MS) confirmation techniques (1-5). One such
adulterant, "Urine Luck", in its initial formulations has been
identified by other researchers to contain pyridium chlorochromate (6,7) and from our previous research on formula 5.3 to
contain potassium dichromate (8). The various complex anions
of chromium (chlorochromate, dichromate, and chromate) are
not normal constituents of human urine. The hexavalent chromate anions are very strong oxidizing agents and as such can
interfere with immunoassay drug-screening procedures and
drug extraction and GC-MS analysis for confirmation, particularly cannabinoids. This can result in false-negative urine
drug-screen results. Screening of suspected adulterated specimens for adulterants and confirmation of the adulterant present
in the specimen is important for forensic urine drug screening.
Color formation from the reaction of chromate with diphenylcarbazide in the chromate-carbazide reaction (9,10) is the basis
for many of the spot-test screening procedures for chromates
(6,7). Chromate reactions with hydrogen peroxide and potassium iodide/potassium permanganate solutions have also been
used as spot tests to detect chromate-adulterated urine specimens (11). The application and comparison of automated colorimetric and capillary ion electrophoretic techniques for the
detection, confirmation, and quantitation of chromate adulteration of urine specimens were the purpose of this investigation. Validationdata of the automated colorimetric and capillary
ion electrophoretic techniques as well as the results of the
analysis of 36 urine specimens suspected of adulteration with
chromate are presented.
Introduction
Materials and Methods
Various chemicals can be added to urine specimens collected
9 Presentedal the meeting of lhe Society of Forensic Toxicologists. New Orleans, LA,
October 3, 2001.
t Author to whom correspondence should be addressed: Kenneth E. Fersiew, Ph.D., Section
of Toxicology, Box 70422, EastTennesseeSlate University, JohnsonCity, TN 37614. E-mail:
[email protected].
36
Capillary ion electrophoresis
Urinary chromate concentrations were determined by capillary ion electrophoresis using the method of Ferslew et al. (8).
The capillary ion electrophoresis was performed on a Waters
Reproduction (photocopying) of editorial content of this journal is prohibited without publisher's permission.
Journalof AnalyticalToxicology,Vol. 27, January/February2003
Quanta 4000 capillary electrophoresis system (Waters Corp.,
Milford MA)with a 745 Data Module using an uncoated 75-1am
(i.d.) x 375-1am (o.d.) x 60-cm (length) capillary with a voltage
of 18 kV, a current of 50 laamps, and direct ultraviolet absorption detection at 254 nm (mercury lamp). The Waters Quanta
4000 capillary electrophoresis system has a filter designed detector system thus the 254 nm filter was used as it was the
closest wavelength to the chromate maximum (257 nm). Chromate was detected by direct ultraviolet absorption in aliquots of
urine diluted I to 100 with 18 Mohm deionized water in 0.5 mL
polypropylene sample vials (Waters), vortex mixed, and analyzed
using a run electrolyte of 12.5mM monosodium phosphate/12.5mM disodium phosphate (Fisher Scientific, Fairlawn, NJ)/3.5mM tetradecyltrimethylammonium hydroxide
(OFM-OH) (Waters Corp.) at pH of 7.0. Chromate eluted at
3.12 rain.
Analytical standards of chromate were prepared by making a
stock standard of potassium dichromate (ACS grade, Fisher
Scientific, Fairlawn, NJ) at a concentration of 1000 Hg/mL in 18
Mohm deionized water. The stock standard was serial diluted to
make a standard curve of 100, 50, 20, 10, and 5 lag/mL. The correlation between chromate peak area and concentration was
found to be linear over the concentration range of 5 to 100
~g/mL (r = 0.9982, y = 9.9883x + 1.0735). A two-point linear
calibration was performed using 0 and 100 tag/mL chromate as
standards. A positive control at 35 lJg/mL and a negative control were assayed with each batch of specimens. Repeated analyses of the 35 pg/mL chromate control were performed within
day (n = 15) and between day (n = 5). Within-day analyses revealed a mean plus or minus standard error of 31.2 • 0.63
lag/mL (CV= 7.8%) and between-day analyses revealed a mean
plus or minus standard error of 34.7 • 1.08 lag/mL (CV = 6.9%).
Specimen chromate concentrations were determined by interpolation of peak areas versus the standard curve. Specimens
with concentrations greater than the standard curve were serial
diluted with deionized water and reanalyzed. If a specimen
chromate peak area was too small to integrate (at 1:100 dilution), then the specimen was diluted 1:10 with deionized water
and reanalyzed.
Colorimetric chromate procedure
Urinary chromate concentrations were determined by applying the colorimetric procedure of Bose (9,10). Chromate in
the hexavalent state (Cr § oxidizes diphenylcarbazide to
diphenylcarbazone and is reduced to tfivalent chromium (Cr+S).
The Cr§ ions then react with the enolate form of diphenylcarbazone (III) in a 1:1 molar ratio to yield a colored chromophore
which can be quantitated by measurement of absorbance at 540
nm. Urine chromate concentrations were determined by colorimetric analysis on a Wako 30R random access automated
chemistry analyzer [Wako 30R, (a.k.a. Syva 30R) Syva Co.
(Behring Diagnostics, Inc., San Jose, CA)].Specimen (5 lad was
mixed with 250 laL of diphenylcarbazide reagent [1 g/L 1,5diphenylcarbazide (Sigma Chemical Co., St. Louis, MO) in 50%
deionized water/acetone (ACS grade, Fisher Scientific, Fairlawn, NJ)]. The endpoint reaction was measured at 10 rain.
Analytical standards of chromate were prepared by making a
stock standard of pyridium chlorochromate (DRI, Sunnyvale,
CA) at a concentration of 500 lag/mL in 18 Mohm deionized
water. The stock standard was serial diluted to make a standard
curve of 500, 100, 50, 20, 10, and 5 lag/mL. The method was
found to be linear over the concentration range of 5 to 500
lag/mL (r = 0.9979, y = 1.05gr - 5.04). A two-point linear calibration was performed using 0 and 100 lag/mL chromate as
standards. A positive control [potassium chromate (Fisher Scientific) at 50 lag/mL in deionized water] and a negative control
(blank urine) were assayed. Replicate analyses of the positive
control were performed within day (n = 5) and between day
(n = 13). Within-day analyses revealed a mean plus or minus
standard error of 55.4 • 0.25 lag/mL (CV = 0.99%), and the
between-day analyses mean plus or minus standard error was
58.15 • 0.42 pg/mL (CV = 2.61%). The change in absorbance of
the unknown specimens was compared by linear regression to
the standards to determine their chromate concentrations. The
Table I. Chromate Data by Colorimetric and Capillary
Ion ElectrophoreticAnalyses*
Specimen
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
Colotirnetric
Analysis
IOO1
1t99
1826
1628
192
1728
1078
364
428
1045
488
1045
5634
454
1290
368
1440
75
975
73
1600
136
301
3290
39
323
30
298
1025
1300
350
481
66
234
960
694
CapillaryIon
Eledrophotesis
961
1418
1708
1726
384
1118
386
422
621
911
812
481
7501
270
1187
432
1695
59
903
3O4
1860
172
493
3460
20
330
414
328
1158
1459
581
396
37
191
1049
1077
' Chromateconcentrationsin pg/mL.
37
Journal of Analytical Toxicology, Vol. 27, January/February 2003
assay was linear to 500 lag/mL.Concentrations greater than 500
lag/mLwere serial diluted with deionized water and reanalyzed.
Specimens
Thirty-six urine specimens were included in the study, representing a subset of random voided urine specimens collected
and submitted for substance of abuse testing. The specimens
were suspected of adulteration on the basis of coloration, interference with confirmation analyses, or the generation of a
positive colorimetric test for chromate. All of the specimens
were analyzed for chromate by the automated colorimetric and
capillary ion electrophoretic techniques. These specimens were
analyzed and found not to contain other adulterants (i.e.,
bleach, nitrites, or phosphates).
Results
Results of the colorimetric and capillary ion electrophoretic
analyses of the 36 urine specimens suspected of adulteration are
presented in Table I. Statistical analysis of the chromate results
are presented in Table II. Colofimetfic analyses revealed a mean
chromate concentration of 929 lag/mLwith a standard error of
177 lag/mL,a range of 30 to 5634 lag/mL,and a 99% confidence
interval of 448 to 1410 lag/mL. Capillary ion electrophoretic
analyses revealed a mean chromate concentration of 1009
lag/mL with a standard error of 218 lag/mL, a range of 20 to
7501 lag/mL and a 99% confidence interval of 414 to 1604
lag/mL
Table II. Statistical Analysis of Urine Chromate
Concentrations
Procedure/Statistical
Parameter
Colorimetric
Analysis
Capillary Ion
Electrophoresis
Mean
Standard Deviation
Range
99% Confidence Interval
929 pg/mL
177 pg/mL
30 to 5634 iJglmL
448 to 1410 pg/mL
1009 pglmL
218 pg/mL
20 to 7501 pglmL
414 to 1604 pg/mL
Correlation of Urinary Chromate Concentrations
7000
~ 6000f Y= O'TglSx+ 140"6
~ 5000f r=0.9669
9|
4~176176
f
I 2ooo~
0
o
I000
^ /
2000
3000
4000
5000
6000
7000
8000
Chromate by CIE (pg/mL)
Figure 1. Correlation of urinary chromate concentrations by colorimetric
and capillary ion electrophoretic analysis.
38
Correlation of the urinary chromate concentrations as determined by capillary ion electrophoretic analysis and colorimetric analysis is presented in Figure 1. The correlation
coefficient (r) between the chromate capillary ion electrophoretic and the colorimetric results was 0.9669.
Discussion and Conclusions
Chromium is considered an essential trace nutrient and is
found in biological tissues primarily in the tfivalent form (Cr§
Daily human intake is less than 100 lag from food, water, and
air. Hexavalent chromium (Cr§ readily crosses tissue membranes and is reduced intracellularly to Cr§ There is no evidence that Cr§ is converted to Cr+6 in biological systems (12).
Urinary excretion accounts for at least 80% of chromium excretion from the body and normal urinary chromium concentrations average 0.1 to 0.5 pg/L and range up to 1.0 pg/L
(13,14). Occupational exposure to Cr +3compounds in tannery
workers has resulted in elevated urinary chromium concentrations of up to 62 pg/L (15). Ingestion of a saturated potassium dichromate solution has produced a maximum urine
chromium concentration of 2.4 rag/L, which was reduced to
0.88 mg/L following hemodialysis treatment in which the patient recovered from the ingestion (16). Ingestion of 400 lag/day
of chromium picolinate for three days by eight healthy adults
with background urine chromium levels of < 0.2 to 1.8 pg/L resulted in peak urinary chromium concentrations averaging
11 pg/L at 7.2 h postdosage (17). [t is apparent that urine chromate concentrations from normal daily intake, nutritional supplementation, chronic occupational exposure, or toxic overdose
are relatively minute and are not consistent with urine chromate concentrations greater than 5.4 pg/mL
The Cr§ compounds are much more toxic to humans than
the Cr§ compounds. The Cr § compounds are corrosive and
cause tissue ulceration, oral burns, gastrointestinal bleeding,
and necrosis, tubular damage, and renal failure. The Cr§ compounds exhibit excellent membrane penetration and are reduced intracellularly to less toxic Cr§ The Cr§ compounds are
non-corrosive and non-irritating with low toxicity and poor
membrane permeability (18). The enhanced toxicity of the Cr+6
compounds prevents their intentional oral ingestion to increase urinary chromate excretion, and therefore, elevated
urine chromate concentrations of greater than 10 pg/mL can
only be the result of postcollection urine adulteration.
Chromium in biological specimens can be measured by a
number of sophisticated methodologies such as atomic absorption spectrometry, GC-MS, and neutron activation analysis
(19). Although these methods can detect nanogram-permilliliter concentrations of chromium in biological fluids, they
are very expensive and time consuming and the limits of detection are far lower than the range needed to detect adulteration. The diphenylcarbazide colorimetric technique can detect
urine chromate concentrations of 5 to 500 IJg/mLand is rapid,
relatively inexpensive, and readily automated. This chemistry
uses the same principle methodology as many of the rapid
spot-test procedures (6,7) or the urine dipsticks used to detect
Journal of Analytical Toxicology,Vol. 27, January/February2003
chromates by a color change (20), and as such, one is not an
ideal forensic confirmation for the other. These colorimetric
techniques can also have interference by other chromophores,
positive reaction with other ions such as molybdenum, mercury, or vanadium (21), and may even have altered colorimetric
responses from different forms of chromate anions. The capillary ion electrophoretic methodology offers a separation technique to isolate and a direct spectrophotometric detection to
quantitate chromate anion in urine specimens. The procedure
is direct, rapid, automated, and relatively inexpensive compared to more sophisticated methodologies. The capillary ion
electrophoretic technique offersa procedure with an alternative
methodology to the colorimetric techniques for forensic confirmation of chromate adulteration of urine specimens.
Results of this investigation revealed that both the diphenylcarbazide colorimetric technique and the capillary ion electrophoretic technique can quantitate chromate concentrations
found in adulterated urine specimens. The concentrations of
chromate found in these adulterated specimens ranged from 20
to 7501 pg/mL with all but 2 of the specimens being from 20 to
less than 2000 IJg/mL.These concentrations far exceed both the
endogenous urinary concentrations of chromium from environmental background exposure and those that would result
from nutritional supplementation, chronic occupational exposure, or toxic overdose.Although there was an excellent overall
correlation between the chromate capillary ion electrophoretic
and the colorimetric results (r = 0.9669), there were a few
specimens in which the colofimetric and the capillary ion electrophoretic results varied rather significantly (e.g., specimens
7, 12, 20, and 27). The discrepancies in these results may be due
to the form or forms of chromate which were added to the
specimen and differences in their response in the diphenylcarbazide reaction. The formulation of Urine Luck has changed
from its original content of pyridium chlorochromate (6,7) to
a later content of potassium dichromate (8). Changes in chromate concentration over time due to continual oxidation within
the urine specimen may also explain some of the differences in
analytical results.
These changes underscore the importance of performing
both the screening and confirmation procedures within a relative short period of time. Although the capillary ion electrophoretic methodology separates and detects chromate anion
(CrO4), further research with this technique is needed to determine if separation of the different forms of chromate anions
can be achieved. Capillary ion electrophoresis may also be
useful in determining the ionic ingredients of future formulations of urine adulterants.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
References
1. S.L. Mikkelsen and K.O. Ash. Adulterants causing false negatives
in illicit drug testing. Clin. Chem. 34(11): 2333-2336 (1988).
2. F.M. Urry, G. Komaromy-Hiller, B. Staley, D.K. Crockett,
M. Kushnir, G. Nelson, and R.E. Struempler. Nitrite adulteration of
21.
workplace urine drug-testing specimens I. Sources and associated concentrations of nitrite in urine and distinction between
natural sources and adulteration. J. Anal. Toxicol. 22:89-95
(1998).
J.T. Cody and R.H. Schwarzhoff. Impact of adulterants on RIA
analysis of urine for drugs of abuse. J. Anal. Toxicol. 13:277-284
(1989).
R. Schwarzhoff and J.T. Cody. The effects of adulterating agents on
FPIA analysis of urine for drugs of abuse. J. Anal. ToxicoL 17:
14-17 (1993).
J.S.C. Tsai, M.A. EISohly, T.R Tsai, B. Twarowska, and S.J. Salamone. Investigation of Nitrite Adulteration on the Immunoassay
and GC-MS analysis of cannabinoids in urine specimens. J. Anal.
Toxicol. 24:708-714 (2000).
A.H.B. Wu, B. Bristol, K. Sexton, G. Cassella-McLane, V. Holtman,
and D.W. Hill. Adulteration of urine by "Urine Luck". Clin. Chem.
45(7): 1051-1057 (1999).
B.D. Paul, K.K. Martin, J. Maguilo, Jr., and M.L. Smith. Effects of
pyridium chlorochromate adulterant (Urine Luck) on testing for
drugs of abuse and a method for quantitative detection of
chromium (VI) in urine. J. Anal Toxicol. 24:233-237 (2000).
K.E. Ferslew, A.N. Hagardorn, and T.A. Robert. Capillary ion electrophoresis of endogenous anions and anionic adulterants in
human urine. J. Forensic ScL 46(3): 615-626 (2001).
M. Bose. The reaction of chromate with dipbenylcarbazide. I.
Anal. Chim. Acta 10:201-208 (1954).
M. Bose. The reaction of chromate with diphenylcarbazide. II.
Anal Chim. Acta 10:209-221 (1954).
A. Dasgupta, A. Wahed, and A. Wells. Rapid spot tests for detecting the presence of adulterants in urine specimens submitted
for drug testing. Clin. Chem. 117:325-329 (2002).
R.A. Goyer and T.W. Clarkson. Toxic Effects of Metals in Casarett
and Doull's Toxicology: The Basic Science of Poisons, 6th ed.,
C.D. Klaassen, Ed. McGraw-Hill, New York, NY, 2001, pp
826-829, 839-840.
M. Kiilunen, J. Jarvisalo, O. Makitie, and A. Aitio. Analysis, storage,
stability and reference values fro urinary chromium and nickel. Int.
Arch. Occ. Env. Health 59:43-50 (1987).
A.H. Stem, N.C.G. Freeman, P. Pleban, R.R. Boesch, 1".Wainman,
1". Howell, S.L Shupack, B.B. johnson, and P.J.Lioy. Residential exposure to chromium waste. Env. Res. 58:147-162 (1992).
A. Aitio, J. Jarvisalo, M. Kiilunen, A. Tossavainen, and P. Vaittinen. Urinary excretion of chromium as an indicator of exposure
to trivalent chromium sulphate in leather tanning. Int. Arch. Occ.
Env. Health 54:241-249 (1984).
R.S.Pedersenand P.T.Morch. Chromic acid poisoning treated with
acute hemodialysis. Nephron 22:592-595 (1978).
M.L. Gargas, RI. Norton, D.J. Paustenbach, and BI. Finley. Urinary excretion of chromium by humans following ingestion of
chromium picolinate. DrugMetab. Disp. 22:522-529 (1994).
Ellenhom's Medical Toxicology: Diagnosis and Treatment of
Human Poisoning, 2nd ed., M.J. Ellenhorn, S. Schnold, G. Ordog,
and J. Wasserberger, Eds. Williams and Wilkins, Baltimore, MD,
1997, pp 1551-1553, 1807.
R.C. Baselt. Disposition of Toxic Drugs and Chemicals in Man, 5th
ed. Chemical Toxicology Institute, Foster City, CA, 2000, pp
184-186.
Chromate-DetectTM, Testpackage insert, DRI, Sunnyvale, CA, August, 1999.
Testfor chromates in the presence of permanganate. In Spot Tests
in Inorganic Analysis, E Fiegl, V. Anger, and R.E. Oesper, Eds. Elsevier Science, Amsterdam, The Netherlands, pp 188-192.
Manuscript received November 5, 2001;
revision received November 12, 2002.
39