A SOLID PHASE EXTRACTION/REVERSED-PHASE LIQUID
CHROMATOGRAPHIC METHOD FOR THE QUANTITATION OF THE
RODENTICIDE DIFETHIALONE IN BIOLOGICAL MATRICES
David Arthur Goldade
B.S., South Dakota School of Mines and Technology, 1991
A thesis submitted to the
University of Colorado at Denver
in partial fulfillment of the
requirements for the degree of
Master of Science
This thesis for the Master of Science
David Arthur Goldade
has been approved
Goldade, David Arthur (M.S., Chemistry)
A Solid Phase Extraction/Reversed-Phase Liquid Chromatographic Method
for the Quantitation of the Rodenticide Difethialone in Biological Matrices
Thesis directed by Donald C. Zapien
Reversed-phase ion-pair high-performance liquid chromatographic
methods were developed for the determination of residue levels of the
anticoagulant rodenticide difethialone in laboratory rats (Rattus
norvegicus), Pedigree dog food, and the livers of black-billed magpies
(Pica pica) and European ferrets (Mustela putorius furo). Difethialone was
extracted from fortified samples and samples containing incurred residues
with a solution of chloroform, acetone, and formic acid. The extracts were
cleaned-up by a solid phase extraction (SPE) procedure using
aminopropyl and either silica or florisil SPE sorbents, concentrated, and
constituents of interest isolated by reversed-phase ion-pair high-
performance liquid chromatography. Difethialone and the surrogate
compound, brodifacoum, were quantitated via ultraviolet absorbance at
262 nm. Brodifacoum responses were used to correct for method
This abstract accurately represents the content of the candidates thesis. I
recommend its publication.
I dedicate this, with much gratitude, to my wife Mary. Without her constant
support and encouragement, I would never have had the endurance to
undertake this feat.
My thanks to John Johnston and Peter Savarie for allowing me the
opportunity to work on this project. I would like to thank LiphaTech, Inc.
for sponsoring this study. My thanks also to Eric Petty for his help with
Quality Control issues, and to Thomas Primus and Doreen Griffin for
performing method validations. I also owe an additional thanks to Tom
Primus for his guidance and advice during the majority of this work.
1. Introduction ...................................................... 1
2. Apparatus ......................................................... 8
3. Reagents ......................................................... 10
4. Procedure ........................................................ 12
4.1 Fortification of Controls......................................... 12
4.1.1 Ground Rodent Tissue .......................................... 12
4.1.2 Pedigree Dog Food ............................................ 13
4.1.3 Ferret and Magpie Liver ....................................... 13
4.2 Extraction Procedure ............................................ 14
4.3 SPE Clean-up .................................................... 16
4.3.1 Phase I........................................................ 16
4.3.2 Phase II ...................................................... 17
4.4 Fligh-Performance Liquid Chromatography.......................... 18
4.5 Response Linearity .............................................. 21
4.6 Thin-Layer Chromatography Procedure...............................24
5. Results and Discussion ........................................... 25
5.2 SPE Clean-up ................................................... 26
5.2.1 Ground Rodent and Dog Food................................... 28
5.2.2 Ferret and Magpie Liver ...................................... 32
5.3 Thin-Layer Chromatography Results .............................. 35
5.4 Recovery Data................................................... 37
5.4.1 Ground Rodent................................................. 37
5.4.2 Dog Food...................................................... 40
5.4.3 Ferret and Magpie Liver ...................................... 43
5.5 Method Limit of Detection....................................... 47
5.6 Storage Stability............................................... 48
5.7 Incurred Residue Results ....................................... 55
6. Conclusion ..................................................... 59
1.1 Structure of Difethialone........................................ 1
2.1 Solid Phase Extraction Manifold .................................9
4.1 Chromatogram of a Standard Solution.............................20
4.2 Linear Regression Plots ........................................ 22
5.1 Chrom. Demonstrating Clean-up of Rodent Samples ..................30
5.2 Chrom. Demonstrating Clean-up of Dog Food Samples ................31
5.3 Chrom. Demonstrating Clean-up of Ferret Liver Samples...........33
5.4 Chrom. Demonstrating Clean-up of Magpie Liver Samples...........34
5.5 Graph of Mean Recoveries in Fortified Rodent Samples............38
5.6 Chrom. of Control and Fortified Rodent Tissue ..................39
5.7 Graph of Mean Recoveries in Fortified Dog Food Samples..........41
5.8 Chrom. of Control and Fortified Dog Food Samples................42
5.9 Graph of Mean Recoveries in Fortified Ferret Liver Samples ... 44
5.10 Chrom. of Control and Fortified Ferret Liver Samples.............45
5.11 Chrom. of Control and Fortified Magpie Liver Samples.............46
5.12 Graph of Storage Stability Recoveries .........................49
5.13 Dog Food Storage Stability Results (Refrigerated) .............50
5.14 Dog Food Storage Stability Results (Frozen) ...................52
5.15 Dog Food Storage Stability Results (Frozen/Refrigerated) .... 53
5.16 Dog Food Storage Stability Results (Room Temperature)........54
5.17 Incurred Residue Results in Rodent Tissue .......................57
5.18 Linear Regression of Incurred Residue Results ...................58
4.1 Linear Regression Results .................................... 23
4.2 Linear Regression Results (log/log)...........................23
5.1 Thin-Layer Chromatography Results ............................ 36
5.2 Method Limit of Detection Values .............................47
5.3 Results of Incurred Residue Analyses..........................56
tetrahydro-1-naphtyl]-4-hydroxy-1-benzothiin-2-one) is a second generation
anti-vitamin K anticoagulant rodenticide of the chemical class hydroxy-4-
benzothiopyranones (Lechevin and Poche 1988; Marshall 1992).
Difethialone has reported LD50 values between 0.29 and 0.51 mg/kg in 3
wild warfarin-resistant and non-resistant rodent species (Lechevin and
Poche 1988). Baits containing 0.0025% difethialone are registered with
the US Environmental Protection Agency for the control of rodents in and
around buildings in urban and non-urban settings. A registration for
broadcast application of difethialone treated baits for the control of field
rodents is being pursued. One registration requirement (EPA Pesticide
Assessment Guidelines Subdivision E: Wildlife and Aquatic Organisms) is
a secondary hazard study to determine the risk to non-target species
consuming the carcasses of target rodent species which had consumed
difethialone treated baits.
The National Wildlife Research Center is conducting a secondary
hazard study in cooperation with LiphaTech, Inc. (Millwaukee, Wl). A vital
portion of the secondary hazard study requires the quantitation of
difethialone residues in the carcasses of the target rodent species. Rats
will be fed difethialone treated baits and their carcasses analyzed for
difethialone residues. The half-life of 14C labeled difethialone in livers and
sera of laboratory rats was 108 and 2.3 days, respectively (Lechevin and
Poche 1988). Even though anticoagulants accumulate in the liver of the
target species, difethialone residues in other portions of the carcass may
pose a secondary hazard to wildlife. This requires that the entire body of
the target animal be analyzed for difethialone residues. These incurred
residue levels will be used to determine bait fortification levels for acute
toxicity studies on non-target scavenger species (ferrets and magpies) in
later stages of the secondary hazard study.
Ideally, the secondary species would be fed a diet composed of the
carcass of the target species. However, the production of large quantities
of fortified whole body rodent is prohibitive to a quality study due to both
the expense of purchasing the rodents and to the difficulty in producing a
uniform, homogeneous end-product. Therefore, difethialone fortified
Pedigree dog food will be substituted for ground rodent in feeding the
secondary species: ferrets and magpies. The livers of these secondary
animals will be analyzed to confirm the presence of difethialone residues.
To satisfy these study requirements, methods were also developed to
quantify difethialone in dog food, ferret liver, and magpie liver.
Comparison of the incurred levels of difethialone in the rat study along
with the ferret and magpie toxicity data will permit the estimation of
secondary hazards of difethialone poisoned rodents to both mammalian
and avian predators.
Few methods have been reported for the analysis of anticoagulant
rodenticide residues in whole body rodent (Koubek et al., 1979; Odam and
Townsend, 1976). Methods exist for serum (Breckenridge et al., 1985;
Corn and Berberich, 1967; deVries and Schmitz-Kummer, 1994; Felice
and Murphy, 1989; Kaiser and Martin, 1974; Kieboom and Rammell, 1981;
Lamiable et al., 1993; Lee et al., 1981; Midha et al., 1974; Mundy and
Machin, 1977; Mundy and Machin, 1982; Mundy et al., 1976; O'Bryan and
Constable, 1991), liver (Hoogenboom and Rammell, 1983; Hunter, 1983
(I); Hunter, 1983 (II); Hunter, 1984; Hunter, 1985; Hunter and Sharp, 1988;
Hunter et al., 1988; Jones, 1996; Kieboom and Rammell, 1981; Langseth
and Nymoen, 1991; Mundy and Machin, 1977; Mundy and Machin, 1982;
Mundy et al., 1976; Odam and Townsend, 1976; OBryan and Constable,
1991; Ray et al., 1989), muscle (Hoogenboom and Rammell, 1983;
Hunter, 1984; Mundy and Machin, 1982), and fat (Hoogenboom and
Rammell, 1983). These methods employ a variety of analytical detection
methods. Gas chromatographic methods lack the required sensitivity and
precision (Kaiser and Martin, 1974; Midha et al., 1974; Odam and
Townsend, 1976). Thin-layer chromatographic methods are not well
suited to quantitation of residues (Caissie and Mallet, 1976; Mallet et al.,
1973; Opong-Mensah and Porter, 1988). Spectrophotometric methods
exist for anticoagulants, but lack specificity in the presence of large
quantities of matrix components (Caswell, 1959; Corn and Berberich,
1967; Kawano and Chang, 1980). Reversed-phase high-performance
liquid chromatography (HPLC) methods are sensitive, but chromatographic
resolution tends to be poor (Breckenridge et al., 1985; de Vries and
Schmitz-Kummer, 1994; Felice and Murphy, 1989; Hoogenboom and
Rammell, 1983; Hunter, 1983 (I); Hunter eta/., 1988; Jones 1996;
Kieboom and Rammell, 1981; Koubek eta/., 1979; Lamiable eta/., 1993;
Langseth and Nymoen, 1991; Lee eta/., 1981; Mundy and Machin, 1977;
Mundy and Machin, 1982; Mundy et al., 1976; OBryan and Constable,
1991; Ray et al., 1989; Yeun, 1978). Ion-pair reversed-phase HPLC has
been used for a number of anticoagulant rodenticides and has proven to
be adequately selective and sensitive (Hunter, 1983 (II); Hunter, 1984;
Hunter, 1985; Hunter and Sharp, 1988; Hunter eta/., 1988). Additionally,
the use of column washes overcomes the shortened column lifetime
associated with the use of ion-pair chromatographic methods by removing
a significant portion of the ion-pair reagents from the column. In addition,
the above methods were developed for coumarin and indanedione derived
anticoagulants and were not entirely applicable to the benzothiopyranone,
A wide variety of clean-up procedures have been cited for the extraction
of anticoagulants in various matrices. Gel-permeation chromatography
(GPC) provides adequate clean-up of the sample, but is time consuming
and requires large quantities of solvents (Hunter, 1983 (I); Hunter, 1983
(II)). A method has also been reported which uses a combination of GPC
and Sep-Pak solid phase extraction (SPE) cartridges (C-18), but this
method also requires large volumes of solvents (Hunter, 1984; Hunter,
1985; Hunter and Sharp, 1988; Hunter et al., 1988; Koubek et a/., 1979).
Liquid-liquid clean-up procedures are used for the quantitation of
anticoagulants in plasma and urine samples but proved to be
unacceptable for difethialone due to the mid-polarity of difethialone and its
associated solubility in a variety of solvents (Breckenridge et al., 1985;
Caswell, 1959; Corn and Berberich, 1967; Felice and Murphy, 1989;
Kaiser and Martin, 1974; Lamiable et al., 1993; Lee et al., 1981; Midha et
al., 1974; Mundy et al., 1976; OBryan and Constable, 1991). SPE using a
single sorbent has been successfully employed for the clean-up of
anticoagulant rodenticides in various matrices (de Vries and Scmitz-
Kummer, 1994; Hoogenboom and Rammell, 1983; Jones, 1996; Kawano
and Chang, 1980; Kieboom and Rammell, 1981; Langseth and Nymoen,
1991; Mundy and Machin, 1982; Ray et al., 1989; Yuen, 1978).
We found all of these methods to provide insufficient clean-up and/or
recovery for our purposes, presumably due to the complex nature of the
matrices involved. In addition, the mid-polarity of difethialone made use of
a single sorbent problematic at best. The liver methods found in the
literature did not yield sufficient recovery of difethialone residues, the
whole body rodent methods did not provide adequate clean-up of the
samples, and no methods were found for the analysis of anticoagulants in
dog food. To overcome these limitations, we developed a two-step solid
phase extraction clean-up procedure which utilized aminopropyl and either
florisil or silica sorbents.
A SPE manifold (Jones Chromatography, Lakewood, CO) was used for
the SPE clean-up steps. As seen in Figure 2.1, the apparatus consisted
of a glass tank (A) equipped with a plastic rack (B) and a lid (C) with teflon
needles (D). To this were attached the SPE cartridges (E) which were
packed with 20 //m polyethylene frits and silica, florisil, or aminopropyl
sorbents. Glass test tubes (F) were used to collect waste and sample
eluents. The manifold was connected to a 400 mbar vacuum source (G)
to facilitate complete elution of eluates from the sorbents.
Figure 2.1: Solid Phase Extraction Manifold.
A horizontal shaker (Eberbach, Model 6550, Ann Arbor, Ml) was used to
agitate the samples. A centrifuge (Fisher Scientific, Denver, CO) was
used to separate the liquid extracts from the samples. A water bath (N-
Evap, Model 115, South Berlin, MA) was used to evaporate samples.
Acetone, chloroform, ethyl acetate, ethyl ether, hexane, isopropanol, and
methanol were liquid chromatography grade (Fisher Scientific, Denver,
CO). Formic acid (88%) and glacial acetic acid were high purity grade
(Fisher Scientific, Denver, CO). Deionized water was purified using a Milli-
Q water purification system (Millipore, Bedford, MA). Phosphoric acid
(85%, Fisher Scientific, Denver, CO) was used to make the 4N phosphoric
acid in water solution. Reagent grade ascorbic acid (99%, Fisher
Scientific, Denver, CO) was used to prepare the aqueous solution of 1.0 M
ascorbic acid. A solution of 50% ammonium hydroxide in water (Fisher
Scientific, Denver, CO) was used to prepare the 5% ammonium hydroxide
in methanol solution.
Technical grade difethialone (100.45%, LiphaTech, Milwaukee, Wl) and
brodifacoum (98%, Zeneca Agrochemicals, Kent, England) were dried for
4 hours at 110C. A concentrated stock solution of difethialone was
prepared by dissolving 2.5 mg difethialone in 25 ml_ methanol. A
concentrated stock solution of brodifacoum was prepared by dissolving 10
mg brodifacoum in 10 ml_ acetone. Working standards ranging from 0.1 to
30 //g/mL were prepared by adding 500 yuL of brodifacoum stock solution
to a 10-mL volumetric flask and evaporating the acetone under a gentle
stream of nitrogen. An appropriate aliquot of difethialone concentrated
stock solution was then added to the flask and diluted to volume with
mobile phase. All standard solutions were stored at 5C.
An extraction solvent was prepared by mixing 500 ml_ of chloroform with
500 ml_ of acetone and 5 ml_ of concentrated formic acid in a 1000-mL
graduated cylinder. The solution was mixed by mechanical stirring and
stored in a glass bottle until used.
The ion-pairing reagent tetrabutylammonium dihydrogen phosphate
(97%, Aldrich, Milwaukee, Wl) was used to prepare a 5 mM solution in
methanol. An aqueous solution of 5 mM tetrabutylammonium dihydrogen
phosphate was also prepared. The aqueous solution was buffered with a
0.1 M potassium dihydrogen phosphate and 0.1 M potassium hydroxide.
4.1 Fortification of Controls
Fortification of ground rodent tissue, magpie and ferret liver
homogenates, and dog food with both difethialone and brodifacoum was
required for method development and method validation studies.
4.1.1 Ground Rodent Tissue
The head, feet, tail, and pelt were removed from five euthanized control
white laboratory rats (Rattus norvegicus). The carcasses were ground
with a variable speed batch sample processor (Robot Coup U.S.A., Model
RSI 6V, Jackson, MS) and the resulting ground tissue was stored in glass
jars, purged with nitrogen, and stored at -25C. Portions (2.00-2.10 g) of
tissue were fortified with 40 or 200 //L of a 10 //g/mL, or 400 //L of a 100
//g/mL standard solution in acetone to produce the 0.2 //g/g, 1.0 //g/g, and
20 //g/g difethialone fortification levels, respectively.
4.1.2 Pedigree Dog Food
The dog-food was placed in a Waring blender (Waring Products
Division, Model 31BL91, New Hartford, CT) and blended at high speed
until liquified. The samples were placed in glass jars, purged with
nitrogen, and stored at -25C. Portions (2.00-2.10 g) of dog food were
fortified with 20 /yL of a 10 /yg/mL, 200 /yL of a 100 /yg/mL, or 200 /yL of a
1000 /yg/mL standard solution in acetone to produce the 0.1 //g/g, 10 /yg/g,
and 100 /yg/g fortification levels, respectively.
4.1.3 Ferret and Magpie Liver
The ferret and magpie livers were removed from the carcasses, placed
in 50-mL plastic tubes and homogenized using a Brinkman homogenizer
(Brinkman, Kinematica AG, Polytron PT MR-3000, Littau, Switzerland) at
high speed (approximately 15,000 rpm) until liquified. The samples were
placed in glass jars, purged with nitrogen, and stored at -25C. Portions
(1.00-1.10 g) of liver were fortified with 20 /yL of a 10 /yg/mL, 20 /yL of a
100 /yg/mL, or 200 /yL of a 1000 /yg/mL standard solution in acetone to
produce the 0.2 /yg/g, 2.0 /yg/g, and 200 /yg/g fortification levels,
4.2 Extraction Procedure
Ground samples were accurately weighed in 1.00 or 2.00 g portions into
50-mL screw-cap glass tubes. An aliquot of a 100 /ug/ml_ or 1000 yug/mL
brodifacoum in acetone solution was added to each sample tube as a
surrogate. The volume of the aliquot added was adjusted to produce a
target concentration of 5 //g/mL in the final extract. Each rodent and dog
food sample was acidified by the addition of 100 /jL of 1.0 M ascorbic acid
solution. Each ferret and magpie liver sample was acidified by the
addition of 2.00 mL of 1.0 M ascorbic acid solution. The sample tubes
were vortex mixed for 30 seconds and then allowed to stand for 5 minutes.
Between 10.0 and 10.5 grams of sodium sulfate (Fisher Scientific, Denver,
CO) were added to each sample tube containing rodent or dog food and
between 15.0 and 15.5 grams of sodium sulfate were added to each ferret
or magpie liver sample and vortex mixed for 30 seconds. Then 15.0 mL of
extraction solvent were added immediately to each sample tube. The
mixture was vortex mixed for 30 seconds and placed on a horizontal
shaker (Eberbach, Model 6550, Ann Arbor, Ml) at high speed (250 cycles
per minute) for 20 minutes. The sample tubes were then centrifuged
(Centrific Centrifuge, Fisher, Denver, CO) at approximately 2500 rpm for 5
minutes. The extracts were decanted into a 30-mL plastic syringe fitted
with a 0.45 //m teflon syringe filter (Scientific Resources Inc., Eatontown,
NJ). The syringes were positioned over clean 50-mL screw-cap glass
tubes and allowed to filter by gravity. The extraction procedure was
repeated with two additional 10 mL aliquots of extraction solvent. For
ferret or magpie liver analysis, an additional 10 mL aliquot was added to
each sample tube, the samples were vortex mixed and centrifuged for 5
minutes. The extracts were combined in the syringes, and the remaining
extract solution forced through the filter by manually applying pressure to
the plunger. The filtered extracts were placed in a 65C water bath (N-
Evap, Model 115, South Berlin, MA) and the solvent evaporated under a
gentle stream of nitrogen. The samples were redissolved in 10.0 mL of
hexane and vortex mixed for 30 seconds. Next, the sample tubes were
placed in a sonicating bath (Sonicor, Model SC-100, Copiague, NY) for 15
minutes. Finally, they were mixed again using the vortex mixer for 30
4.3 SPE Clean-up
A two-phase clean-up was required for all samples prior to
chromatographic analysis. Phase I clean-up of dog food and whole body
rodent samples used a silica column, while a florisil column was used for
ferret and magpie liver samples. In phase II clean-up, an aminopropyl
SPE column was used for all sample extracts.
4.3.1 Phase I
A 2 g 1ST Si02 (silica) SPE column with 6-mL reservoir (Jones
Chromatography, Lakewood, CO) was conditioned for each rodent and
dog food sample extract by addition of 5 mL of hexane. A 6-mL reservoir
packed with 3 g of florisil (Alltech Associates, Inc., Deerfield, IL) which had
been dried in an oven at 110C overnight was conditioned for each liver
sample extract by addition of 5 mL of hexane. Without allowing the
packing material to go dry, the reconstituted tissue extracts were eluted
through the conditioned columns via gravity. Each column was rinsed with
10 mL of hexane followed by 10 mL of 20% ethyl ether in hexane solution.
The wash eluate was discarded. A 25-mL screw-cap glass tube was
placed under each SPE column. The analytes were eluted from each
column by the addition of 22.5 ml_ of 90% ethyl ether in hexane for rodent
and dog food samples or 22.5 ml_ of 1:1 methanol:isopropanol for liver
samples. After the final aliquots passed through the columns, gentle
vacuum (400 mbar) was applied to remove all of the eluting solvent.
Each sample tube was placed under a gentle stream of nitrogen at room
temperature until approximately half of the solvent had evaporated. Tubes
were then immersed in a water bath at 65C and evaporated to dryness
under a gentle stream of nitrogen. The samples were redissolved in 5.0
ml_ hexane and vortex mixed for 30 seconds. Next, the sample tubes
were placed in a sonicating bath for 15 minutes. Then they were mixed
again using the vortex mixer for 30 seconds.
4.3.2 Phase II
A 1 g 1ST NH2 (aminopropyl) SPE column with 6-mL reservoir (Jones
Chromatography, Lakewood, CO) was conditioned for each sample extract
by addition of 2.5 mL of hexane. The packing material was not allowed to
go dry. The reconstituted extracts were eluted through the conditioned
SPE columns via gravity. Each column was sequentially washed with 5
mL hexane, 10 mL 2:1 chloroform:2-propanol, 10 mL each of chloroform
and ethyl acetate, and 6 mL 2% acetic acid in ethyl ether. The ethyl
acetate wash was not required for the dog food samples. After the final
aliquot had passed through the columns, gentle vacuum (400 mbar) was
applied to remove all of the wash solvent, and the wash eluates were
discarded. A 15-mL screw-cap centrifuge tube was then placed under
each SPE column. The analytes were eluted from the columns by addition
of four 2 mL aliquots of 5% ammonium hydroxide solution in methanol.
After the final aliquot had passed through the columns, gentle vacuum
(400 mbar) was applied to remove any remaining solvent.
The eluate from each SPE column was placed in a 65C water bath and
evaporated to dryness under a gentle stream of nitrogen. The sample
residues were reconstituted with 2.00 mL of mobile phase and sonicated
for 15 minutes. The reconstituted samples were filtered through a 0.45
/vm teflon filter prior to injection into the HPLC.
4.4 High-Performance Liquid Chromatography
The HPLC system consisted of a Hewlett-Packard 1090 liquid
chromatograph (Hewlett-Packard, Palo Alto, CA) equipped with a column
oven and a diode-array detector. Injections of 100 /jL or 200 /vL were
made automatically by the pneumatically controlled injector valve. Analyte
separation was achieved on a 250 mm x 4.6 mm i.d. stainless steel
analytical column packed with 5 jum Keystone ODS/H (Keystone,
Bellefonte, PA). The column temperature was 40C and the diode array
detector was operated at 262 nm. The analytical column was fitted with a
guard column of the same packing material. The system was operated at
a flow rate of 1.0 mL/minute. The mobile phase consisted of 77% 5 mM
tetrabutylammonium dihydrogen phosphate in methanol and 23% 5 mM
tetrabutylammonium dihydrogen phosphate with 0.1 M dihydrogen
phosphate buffer in water. The pH of the mobile phase was adjusted to
8.5 by the addition of 4N phosphoric acid. The mobile phase was filtered
through a 0.45 yc/m nylon membrane filter (Alltech Associates, Inc.,
Deerfield, IL) prior to use. The mobile phase was degassed by sparging
with helium. Following each set of sample analyses the HPLC system
was flushed with a mixture of 1:1 methanokwater for 40 minutes at a flow
rate of 1.0 mL/minute.
Brodifacoum and difethialone were quantitated by monitoring the UV
absorbance at 262 nm and comparing the ratio of difethialone to
brodifacoum response to the ratio of responses observed in a calibration
standard. The retention times of brodifacoum and difethialone were
approximately 18 and 20 minutes, respectively, for the conditions listed
above. A chromatogram of a 1.05 //g/ml_ difethialone standard solution
with brodifacoum added at the concentration of 5 //g/ml_ is shown in
Figure 4.1: Chromatogram of a Standard Solution of 1 //g/mL
Difethialone and 5 /yg/ml_ Brodifacoum.
4.5 Response Linearity
Two sets of five difethialone standard solutions were prepared and
injected into the HPLC in duplicate. The concentrations of the solutions
ranged from 0.10 to 32 yug/mL difethialone. Each standard solution also
contained 5 //g/mL brodifacoum. To determine if the relationship between
difethialone response and difethialone standard concentration was linear, a
plot was constructed of difethialone chromatographic peak response
versus difethialone concentration. To determine if difethialone response
was proportional to difethialone standard concentration, a second plot was
constructed of the log of difethialone chromatographic peak response
versus the log of difethialone concentration. To assess the validity of the
assumption that using brodifacoum as a surrogate standard would not
affect the linearity, this procedure was repeated using ratio of difethialone
peak response to brodifacoum peak response versus difethialone
concentration. The plots of response versus concentration and ratio of
responses versus concentration can be seen in Figure 4.2 A and 4.2 B,
respectively. A linear regression was performed on each data set (SAS
Data Manager, Version 2.14, August 1988, SAS Institute Inc., Cary, NC).
The regression statistics are shown in Table 4.1 and Table 4.2.
Area Response (mAU)
Ratio of Responses (mAU/mAU)
Figure 4.2: Plots of Difethialone Area Response versus Difethialone
Concentration (A) and Ratio of Difethialone Area
Response/Brodifacoum Area Response versus Difethialone
Table 4.1. Results of Linear Regression Analysis
Data Set R2 Slope Y-intercept
Area Responses 1.0000 254.52 2.370
Ratio of Area Responses 1.0000 0.2298 0.004454
Table 4.2. Results of Linear Regression Analysis (log/log)
Data Set R2 Slope Y-intercept
Area Responses 0.9993 0.974493 2.428194
Ratio of Area Responses 0.9992 0.971180 -0.610267
A linear relation was observed between the peak responses or the ratio of
peak responses and difethialone concentration. In both cases, the
response was directly proportional to concentration over the range of
interest. This was demonstrated by the slopes of the log/log plots not
being significantly different (p = 0.0001) from one and the intercepts of the
linear plots not being significantly different from zero (p = 0.3080 and p =
0.1884). This indicated that a single standard could be used to calculate
observed difethialone concentrations.
4.6 Thin-Layer Chromatography Procedure
Experiments were conducted in an attempt to classify the lipids removed
during the SPE clean-up steps. A whole body rodent tissue sample was
extracted and fractions of the wash eluates collected. These were
evaporated to dryness under a gentle stream of nitrogen and reconstituted
in 1000 //L of 1:1 chlorofornrmethanol. They were spotted on silica thin-
layer chromatography (TLC) plates and developed in either 25:15:2,
65:25:2, or 80:20:10 chloroform: methanol:water. The resulting plates
were allowed to dry and treated with either ninhydrine, 30% sulfuric acid,
or molybdenate solution. The resulting spots were compared to the height
of standard solutions of lipids spotted on the same plate.
5. Results and Discussion
Solutions of acetone and chloroform have been used to extract warfarin-
derived anticoagulants from tissues with good results (Hunter, 1983 (I);
Hunter, 1983 (II); Hunter, 1984; Hunter, 1985; Hunter et al., 1988; Jones,
1996). Therefore these solvents were logical choices for the extraction of
the newer warfarin-based rodenticides difethialone and brodifacoum from
biological matrices. Ascorbic acid was added directly to the sample prior
to extraction and the extraction solvent was acidified to keep heme in its
reduced state. It has been reported that the oxidized form of heme binds
anticoagulant rodenticides leading to reduced analyte recoveries (Schulert
and Weiner, 1953). Preliminary experiments with whole body rodent
demonstrated that solutions lacking these acids yielded significantly
(T=14.49, p=0.0007) reduced recoveries (X4=2.6%) compared to the
corresponding acidified samples (X4=80.4%).
Sodium sulfate was used to dry the samples. An excess amount was
used to ensure complete absorption of water from the samples.
Experiments with whole body rodent have shown that omitting the use of
sodium sulfate resulted in a cloudy hydrophilic layer that yielded reduced
analyte recoveries (X5 = 108% with 8 grams of sodium sulfate and X5 =
103% without sodium sulfate) This difference was demonstrated to be
significant with a T-test (T=2.67, p=0.0282).
5.2 SPE Clean-up
These methods employed a wide spectrum of chromatographic
separations to successfully quantitate difethialone in a variety of complex
biological matrices. The use of a single separation step did not sufficiently
clean-up the extracts to permit quantitation of difethialone residues. A
method employing several different separation techniques provided the
necessary clean-up and separation needed for these applications. During
Phase I of the SPE clean-up the separation mechanism was purely
normal-phase chromatography. Samples were loaded in a non-polar
solvent (hexane) on a polar sorbent (silica or florisil) and eluted with an
increasingly polar solvent (ether or methanol:isopropanol). During Phase
II of the SPE clean-up the separation mechanism was more complex.
Samples were again loaded with a non-polar solvent (hexane). This time
the separation mechanism was not strictly polar as the aminopropyl
sorbent offers several possible retention mechanisms. It can provide
either a polar, non-polar, or ionic retention mechanism depending on the
analyte and solvent being used. With analytes dissolved in hexane,
aminopropyl sorbents retain by polar interactions. In aqueous solutions at
pH less than the pKa of the aminopropyl functional group (pKa=10),
anionic analytes can be retained via ion interactions while non ionic
analytes are retained primarily by polar interactions. In aqueous solutions
at pH greater than 10, the nonpolar interactions of the isopropyl backbone
become a more significant retention mechanism.
In these methods the aminopropyl SPE column retained the analytes via
polar interactions. However, the presence of water and the high pH in the
later wash and elution stages: 1) decreased the strength of ionic
interactions by assuring that virtually all of the amino groups were free
bases, 2) increased the polarity of the elution solution, and 3) eluted the
analytes difethialone and brodifacoum.
5.2.1 Ground Rodent and Dog Food
The use of two separate SPE cartridges was required for several
reasons. Although the NH2 (aminopropyl) SPE column was the primary
column for sample clean-up, it could not be used without initially
employing the Si02 (silica) column. The Si02 column, serving as a pre-
clean-up step, allowing a significant portion of lipophilic material to pass
through without being retained. This lipophilic material would have
decreased the capacity of the aminopropyl SPE column by binding the
active sites and would have been incompatible with the aqueous mobile
phase. As lipids in the sample matrix are potential chromatographic
interferants, a process which removed lipids while leaving the analytes
behind was essential. Samples processed without the use of the Si02
column yielded a bi-phasic end product with multiple matrix components.
The resulting chromatography did not allow resolution of the analytes (see
Figures 5.1 and 5.2) and yielded reduced, highly variable analyte
In addition to not retaining a significant portion of the lipophilic material,
the Si02 column retained some of the more polar matrix components
which also would have interfered with the analytes ability to bind to the
stationary phase of the NH2 column. These polar components became
irreversibly adsorbed onto the stationary phase of the Si02 column. By
using the Si02 column to remove these components, more active sites
were available on the NH2 column. This greatly increased the effective
capacity and separation efficiency of the NH2 column. The benefits of the
clean-up procedure are clearly observed in Figures 5.1 and 5.2 which
present chromatograms of extracts obtained with and without the SPE
clean-up procedure. Both samples were fortified at 20 /jg/g with
difethialone and analyzed on the same liquid chromatographic instrument
equipped with identical analytical columns. A dramatic increase in
resolution and sensitivity of the surrogate and analyte compounds are
clearly evident. Because the SPE clean-up reduced the amount of matrix
components injected into the chromatographic system, it greatly reduced
the need to replace guard columns.
0 5 10 15 20 25 min
Figure 5.1: Chromatograms of Fortified Rodent Tissue
Samples without the Clean-up (A) and with the Clean-up (B).
0 5 10 15 20 25 min
0 5 10 15 20 25 min
Figure 5.2: Chromatograms of Fortified Dog Food Samples
without Clean-up (A) and with Clean-up (B).
5.2.2 Ferret and Magpie Liver
Analysis of difethialone residues in livers required more clean-up than
was required for whole body rodent. The silica column used as a pre-
clean-up column in the analysis of dog-food and whole body rodent
samples provided insufficient retention of brodifacoum and difethialone for
liver samples (X3 = 4% 2.5%). A more polar sorbent, florisil (FL), was
selected as a pre-column. Because florisil is more polar, it retained
brodifacoum and difethialone more strongly than silica. However, a
stronger eluting solvent (1:1 methanokisopropanol) was required to
remove the analytes from the columns. The benefits of this procedure can
be seen in Figures 5.3 and 5.4. Both samples were fortified at 2 yr/g/g with
difethialone and analyzed on the same liquid chromatographic instrument
equipped with identical analytical columns. An increase in resolution and
sensitivity of the surrogate and analyte are realized for both matrices.
0 5 10 15 20 25
Figure 5.3: Chromatograms of Fortified Ferret Liver Samples
without Clean-up (A) and with Clean-up (B).
Figure 5.4: Chromatograms of Fortified Magpie Liver without
Clean-up (A) and with Clean-up (B).
5.3 Thin-Layer Chromatography Results
The results of the TLC experiments are provided in Table 5.1. The silica
SPE column retained the analytes while a variety of lipid materials were
removed by the 20% diethylether in hexanes wash during Phase I. Some
of these same compounds show up again during Phase II of the clean-up
procedure. This is due to the relative separation power of the two
sorbents being used. The analytes have a stronger attraction to the
aminopropyl sorbent than to the silica sorbent. Therefore, stronger wash
solvents can be used during Phase II than during Phase I allowing a more
refined clean-up of the samples. In effect, the silica SPE column is acting
as a preclean-up step to the aminopropyl SPE column.
Table 5.1: Lipid Classes Removed During SPE Wash Steps
SPE Fraction Tested Lipid Class Observed
Phase I: Silica SPE Column
20% Diethylether in Hexanes Cholesterol Stearate, Triolein, Diglycerides, Sphingomyelin
Phase II: Aminopropyl SPE Column
2:1 Chloroform:2-Propanol Phosphatidyl Ethanolamine, 1,3 Di- Olein, Monolein, Cholesterol, Diglycerides, Sphingomyelin
Ethyl Acetate Phosphatidylserine
2% Acetic Acid in Diethylether Cholesterol, Sphingomyelin, Diglycerides
5.4 Recovery Data
A two factor analysis of variance (ANOVA) was performed on each
matrix with factors day and fortification level and percent recovery as the
response. Type I sums of squares were used for whole body rodent.
However, because not all days were represented at each fortification level,
type III sums of squares were used for all other matrices.
5.4.1 Ground Rodent
The mean surrogate-corrected recoveries of difethialone in rodent tissue
are shown in Figure 5.5. The bars indicate the standard deviation within
each fortification level. Seven replicates were analyzed at each
fortification level. The analyses were performed over a 12 day period with
no statistically significant difference in recoveries observed from day to
day. There was no significant effect due to either fortification level (p =
0.0941) or date of analysis (p = 0.5166).
Control tissue samples were treated according to the procedures
previously outlined in this method. A representative chromatogram is
shown in Figure 5.6. A slight chromatographic interference was observed
at the retention time of difethialone, however this did not significantly affect
method limit of detection or low-level fortification recoveries. A
chromatogram of a tissue sample fortified at 1.03 //g/g is shown in Figure
Figure 5.5: Mean Recovery in Fortified Rodent Samples.
0 5 10 15 20 25 30 min
0 5 10 15 20 25 30 min
Figure 5.6: Chromatograms of Control (A) and Fortified (B)
5.4.2 Dog Food
The mean surrogate-corrected recoveries of difethialone in dog food are
shown in Figure 5.7. The bars indicate the standard deviation within each
fortification level. Seven replicates were analyzed at each fortification
level. The analyses were performed over a 22 day period. There was no
significant effect due to fortification level (p = 0.2963). There was a
significant effect due to date of analysis (p = 0.0082). This could be due
to the fact that the majority of the 0.1 /jglg samples were analyzed on a
single day, which might introduce a bias into the sample recoveries.
Control samples were treated according to the procedures previously
outlined in this method. A representative chromatogram is shown in
Figure 5.8. No chromatographic interferences were observed at the
retention times of brodifacoum or difethialone. A chromatogram of a
sample fortified at 10.1 /yg/g is shown in Figure 5.8.
Figure 5.7: Graph of Mean Recoveries in Fortified Dog Food
Figure 5.8: Chromatograms of Control (A) and Fortified (B)
Dog Food Samples.
5.4.3 Ferret and Magpie Liver
The mean surrogate-corrected recoveries of difethiaione in ferret and
magpie liver tissues are shown in Figure 5.9. Seven replicates were
analyzed at each fortification level. The analyses were performed over a
73 day period with no statistically significant difference in recoveries
observed from day to day. There was a significant effect due to
fortification level for both magpie liver (p = 0.0103) and ferret liver (p =
0.0254) where recovery of the 0.2 /jg/g level was higher than that for the
other levels. There was no significant effect due to date of analysis for
either magpie liver (p = 0.6207) or ferret liver (p = 0.3793).
Control tissue samples were treated according to the procedures
previously outlined in this method. A slight chromatographic interference
was observed in both matrices. Numerous attempts to remove this
interferant were not successful. Therefore, recoveries were higher at the
lower fortification level. The presence of this interferant is also reflected in
the elevated Method Limit of Detection (MLOD) values for these matrices.
A chromatogram of a ferret liver tissue sample fortified at 2.25 yug/g is
shown in Figure 5.10. A chromatogram of a magpie liver tissue sample
fortified at 2.01 /yg/g is shown in Figure 5.11.
Figure 5.9: Graph of Mean Recoveries of Fortified
Ferret Livers (top) and Magpie Livers (bottom).
0 5 10 15 20 25 30
Figure 5.10: Chromatograms of Control (A) and Fortified (B) Ferret
Figure 5.11: Chromatograms of Control (A) and Fortified (B)
5.5 Method Limit of Detection
The MLOD was estimated from the chromatographic responses of five
control tissue extracts and seven tissues fortified at a low fortification level.
The MLOD was calculated as the concentration of difethialone required to
generate a signal equal to 3 times the baseline noise (measured peak-to-
peak) plus the peak height of any chromatographic interference which was
observed in the control tissue chromatograms. The MLOD values are
listed in Table 5.2.
Table 5.2: Method Limit of Detection Values
Matrix Fortification Level MLOD
Rodent Tissue 0.2 fjglg 0.054 fjglg
Dog Food 0.1 fjglg 0.085 fjglg
Ferret Liver 0.2 fjglg 0.091 fjglg
Magpie Liver 0.2 fjglg 0.16 fjglg
5.6 Storage Stability
Storage of whole body rodent samples as well as ferret and magpie liver
samples was necessary due to the large number of animals involved in
the study. Therefore, it was necessary to determine the stability of the
analyte during storage prior to analysis. Whole body rodent was the first
matrix tested. Several preservative methods were evaluated in an attempt
to determine which was the best, both in terms of ease-of-use as well as
prevention of analyte loss and/or degradation. The methods selected
included: addition of ascorbic acid, formic acid, or BHT; nitrogen purging
of sample headspace; and no preservative at all. All of the samples were
stored frozen (-5C) for the entire study period. The study was conducted
over a four week period of time. Samples were analyzed at three
intervals. The results can be seen in Figure 5.12.
Figure 5.12: Storage Stability of Difethialone in Whole Body Rodent
Fortified at 5 /jg/g.
As a result of these experiments, nitrogen purging of the headspace in the
sample container was selected as the preservative method to be used in
all future experiments. Although comparable results were obtained with
the other preservative techniques, nitrogen purging was the easiest to use,
making it preferable to the others. Also, addition of a chemical to the dog
food formulations could have changed the palatability and affected the
results of the study.
Storage of difethialone treated dog food formulations was required to
facilitate feeding of the magpies and ferrets during the acute toxicity
portion of the secondary hazard study. The first preservation approach
evaluated consisted of headspace nitrogen purging and storage at 5C for
Observed Difethialone (pg/g)
0 1 2 3 4 5 6
1 pg/g 20 pg/g
Figure 5.13: Storage Stability of Refrigerated Formulations of
Difethialone in Dog Food.
As shown in Figure 5.13, the 1 //g/g and 20 //g/g formulations lost 11.9%
and 18.6% of the active ingredient while in storage, respectively. Three
replicates were analyzed at each time interval for each fortification level.
The vertical lines indicate the standard deviation within each data set.
This loss was excessive and not suitable for our purposes. Therefore, a
second study was established in which the samples were frozen (-5C) for
the entire storage period of six days. The results of that study are shown
in Figure 5.14.
As shown in the graph, the 1 //g/g formulation gained 2.6% and the 20
//g/g formulation lost 5.5% of the active ingredient while in storage,
respectively. A student T-test proved that there was no significant
difference in the observations from day 0 to day 6 with 99% confidence.
Therefore, it appeared that freezing the formulations would be necessary
to prevent loss of the active ingredient difethialone. Discussions with the
study director revealed that this would not be practical given the large
quantities of dog food needed to complete the study. As a result, a third
study was conducted. In this study, the formulations were frozen (-5C)
for 5 days and then stored in a refrigerator (5C) for 1 day. The results of
that study are shown in Figure 5.15.
Observed Difethialone (pg/g)
0 1 2 3 4 5 6
1 pg/g 20 pg/g
Figure 5.14: Storage Stability of Frozen Formulations of Difethialone
in Dog Food.
As seen in Figure 5.15, loss of difethialone was observed at both
formulation levels. However, the percentage lost was substantially lower
than that observed in the refrigerated samples. While the change was
found to be statistically significant (99% C.I.), the amount of active
ingredient lost was determined to be an acceptable trade off for the easier
handling afforded by the frozen/refrigerated storage method.
Observed Difethialone (|jg/g)
1 pg/g 20 Mg/g
Figure 5.15: Storage Stability of Frozen/Refrigerated Formulations of
Difethialone in Dog Food.
The observed loss of active ingredient seen in the refrigerated
formulations prompted us to investigate the potential loss that could be
expected while feeding the test animals the dog food formulations. To
facilitate this experiment, a 50 gram sample of dog food was formulated at
the 20 jjglg level with difethialone. It was then placed in a glass bowl and
left undisturbed at room temperature for 24 hours. Sub-samples were
taken at various times during the 24 hour period. The results of that study
are shown in Figure 5.16.
Observed Difethialone (pg/g)
Figure 5.16: Degradation of a Difethialone Formulated Dog Food
Sample Over 24 Flours.
An ANOVA was performed on the data. No significant difference (p =
0.0910) was found between the concentrations observed at time 0, time 2,
and time 4 hours. The concentrations observed at 24 hours differed
significantly (p = 0.0001) from all other observations. In discussing the
results with the study director, it was decided, based on observations of
the animals behavior, that most of the dog food would be consumed by
the test animals during the first 4 hours it is presented to the animals.
Therefore, the quantity of active ingredient present should be sufficient to
cause no significant affect to the overall study.
5.7 Incurred Residue Results
Six rodents were fed a diet of 0.0025% difethialone on grain. Animals 1
through 3 were allowed to feed for a single day and then euthanized.
Animals 4 through 6 were allowed to consume the bait until death
occurred. The carcasses were processed according to the procedures
previously outlined. The results of residue analysis are tabulated in Table
5.3. In all of the carcasses analyzed, the residue levels of difethialone
observed were higher for those rodents which consumed the bait until
death occurred. The data was graphed as a function of animal number
and it appeared that a significant relationship existed between observed
residue level and difethialone consumed. To further illustrate this point, a
plot was constructed of observed residue level as a function of difethialone
consumed. A linear regression was performed on the data set. A
significant model was generated (p = 0.0100) and the relationship was
linear (R2 = 0.8416). This demonstrated that the method was effective
when used to analyze incurred residues.
Table 5.3: Tabulated Results of Incurred Residue Analyses
Animal # 1 2 3 4 5 6
Observed Residue Og/g) 0.094 0.014 1.58 0.013 1.89 0.059 3.67 0.28 3.67 0.27 3.85 0.30
Consumed 0.0014 0.232 0.253 0.376 0.645 0.715
1 2 3 4 5 6
Figure 5.17: Results of Incurred Residue Analysis.
Observed Difethialone Residue (gg/g)
0 0.2 0.4 0.6 0.8
Difethialone Consumed (mg)
Figure 5.18: Plot of Observed Difethialone Residue Levels as a
Function of Difethialone Consumed for Incurred Rodent Analyses.
The use of a two-step SPE clean-up was demonstrated to be an
effective technique for the determination of difethialone residues in rodent
tissue, dog food, ferret liver, and magpie liver. This is the first reported
use of a two-step clean-up using both silica and aminopropyl or florisil and
aminopropyl sorbents. It is also the first reported use of SPE for the
determination of difethialone residues.
Most importantly, the use of these methods will enable researchers to
estimate the secondary hazard associated with the use of difethialone as a
rodenticide and permit regulatory agencies to make an informed decision
regarding the registration of this new pesticide.
Breckenridge, A.M.; Cholerton, S.; Hart, J.A.D.; Park, B.K.; Scott, A.K. A
study of the relationship between the pharmacokinetics and the
pharmacodynamics of the 4-hydroxycoumarin anticoagulants warfarin,
difenacoum and brodifacoum in the rabbit. Br. J. Pharmac. 1985, 84, 81-
Caissie, G.E.; Mallet, V.N. The fluorometric detection of pesticides on
aluminum oxide layers. J. Chromatogr. 1976, 117, 129-136.
Caswell, R.L. Report of Rodenticides. J. Assoc. Off. Anal. Chem. 1959,
deVries, J.X.; Schmitz-Kummer, E. Determination of the coumarin
anticoagulant phenprocoumon and metabolites in human plasma, urine
and breast milk by high-performance liquid chromatography after solid-
phase extraction. J. Chromatogr. B 1994, 655, 63-71.
Felice, L.J.; Murphy, M.J. The determination of the anticoagulant
rodenticide brodifacoum in blood serum by liquid chromatography with
fluorescence detection. J. Anal. Toxicol. 1989, 13, 229-231.
Hoogenboom, J.J.L.; Rammell, C.G. Improved HPLC method for
determining brodifacoum in animal tissues. Bull. Environ. Contam. Toxicol.
1983, 31, 239-243.
Hunter, K. Determination of coumarin anticoagulant rodenticide residues in
animal tissue by high-performance liquid chromatography, I. Fluorescence
detection using post-column techniques. J. Chromatogr. 1983, 270, 267-
Hunter, K. Determination of coumarin anticoagulant rodenticide residues in
animal tissue by high-performance liquid chromatography, II. Fluorescence
detection using ion-pair chromatography. J. Chromatogr. 1983, 270, 277-
Hunter, K. Reversed-phase ion-pair liquid chromatographic determination
of chlorophacinone residues in animal tissues. J. Chromatogr. 1984, 299,
Hunter, K. High-performance liquid chromatographic strategies for the
determination and confirmation of anticoagulant rodenticide residues in
animal tissues. J. Chromatogr. 1985, 321, 255-272.
Hunter, K.; Sharp, E.A. Modification to procedures for the determination of
chlorophacinone and for multi-residue analysis of rodenticides in animal
tissues. J. Chromatogr. 1988, 437, 301-305.
Hunter, K.; Sharp, A.E.; Newton, A. Determination of diastereoisomers of
bromadiolone, an anticoagulant rodenticide, in animal tissues by high-
performance liquid chromatography. J. Chromatogr. 1988, 435, 83-95.
Jones, A. HPLC determination of anticoagulant rodenticide residues in
animal livers. Bull. Environ. Contam. Toxicol. 1996, 56, 8-15.
Kaiser, D.G.; Martin, R.S. GLC determination of warfarin in human plasma.
J. Pharm. Sci. 1974, 63, 1579-1581.
Kawano, Y.; Chang, W. Spectrophotometric determination of rozol in
paraffinized formulations. J. Assoc. Off. Anal. Chem. 1980, 63, 996-998.
Kieboom, A.J.; Rammell, C.G. Determination of brodifacoum in animal
tissues by HPLC. Bull. Environm. Contam. Toxicol. 1981, 26, 674-678.
Koubek, K.G.; Ussary, J.P.; Haulsee, R.F. High pressure liquid
chromatographic determination of the rodenticide brodifacoum in rat tissue.
J. Assoc. Off. Anal. Chem. 1979, 62, 1297-1301.
Lamiable, D.; Vistell, R.; Trenque, T.; Fay, R.; Millart, H.; Choisy, H.
Sensitive high-performance liquid chromatographic method for the
determination of coumarin in plasma. J. Chromatogr. 1993, 620, 273-277.
Langseth, W.; Nymoen, U. Determination of coumarin anticoagulant
rodenticide residues in animal liver by high-performance liquid
chromatography. Fresenius J. Anal. Chem. 1991, 339, 249-252.
Lechevin, J.C.; Poche, R.M. Activity of LM2219 (Difethialone), a new
anticoagulant rodenticide, in commensal rodents. Proc. Verte. Pest Conf.
1988, 13, 59-63.
Lee, S.H.; Field, L.R.; Howald, W.N.; Trager, W.F. Fligh-performance liquid
chromatographic separation and fluorescence detection of warfarin and its
metabolites by post-column acid/base manipulation. Anal. Chem. 1981, 53,
Mallet, D.; Surette, D.; Brun, G.L. Detection of naturally fluorescent
pesticides on silica gel layers. J. Chromatogr. 1973, 79, 217-222.
Marshall, E.F. The effectiveness of difethialone (LM2219) for controlling
norway rats and house mice under field conditions. Proc. Verte. Pest Conf.
1992, 15, 171-174.
Midha, K.K.; McGilveray, I.J.; Cooper, J.K. GLC determination of plasma
levels of warfarin. J. Pharm. Sci. 1974, 63, 1725-1729.
Mundy, D.E.; Machin, A.F. Determination of the rodenticide difenacoum in
biological materials by high-pressure liquid chromatography with
confirmation of identity by mass spectrometry. J. Chromatogr. 1977, 139,
Mundy, D.E.; Machin, A.F. The multi-residue determination of coumarin-
based anticoagulant rodenticides in animal materials by high-performance
liquid chromatography. J. Chromatogr. 1982, 234, 427-435.
Mundy, D.E.; Quick, M.P.; Machin, A.F. Determination of warfarin in
animal relicta and feeding-stuffs by high-pressure liquid chromatography
with confirmation of identity by mass spectrometry. J. Chromatogr. 1976,
OBryan, S.M.; Constable, D.J.C. Quantification of brodifacoum in plasma
and liver tissue by HPLC. J. Anal. Toxicol. 1991, 15, 144-147.
Odam, E.M.; Townsend, M.G. The determination of warfarin in animal
tissues by electron-capture gas-liquid chromatography. Analyst 1976, 101,
Opong-Mensah, K.; Porter, W.R. Separation of some rodenticides and
related compounds by thin-layer chromatography. J. Chromatogr. 1988,
Ray, A.C.; Murphy, M.J.; DuVall, M.D; Reagor, J.C. Determination of
brodifacoum and bromadialone residues in rodent and canine liver. Am. J.
Vet. Res. 1989, 50, 546-550.
Schulert, A.R.; Weiner, M. The physiologic disposition of
phenylindanedione in man. J. Pharmacol. Exp. Ther. 1953, 110, 451-457.
Yuen, S.H. Determination of the rodenticides difenacoum and brodifacoum
in finished baits by reversed-phase liquid chromatography. Analyst 1978,