FOOD COMPOSITION AND ADDITIVES

Gas Chromatographic Method for Putrescine and Cadaverine in

Canned Tuna and Mahimahi and Fluorometric Method for
Histamine (Minor Modification of AOAC Official Method
977.13): Collaborative Study
ROGERS & STARUSZKIEWICZ: JOURNAL OF AOAC INTERNATIONAL VOL. 80, NO. 3, 1997
PATRICIA L. ROGERS AND WALTER STARUSZKIEWICZ
U.S. Food and Drug Administration, Office of Seafood, 200 C St, SW, Washington, DC 20204
Collaborators: A. Adams; B. Atienza; R.J. Berg; H. Chin; C. DesJardins; M. Gilgan; L.B. Hansen; W. Hummer; M. Jahncke; E.
Jester; K. Kennedy; J. Krzynowek; D. McLachlan; D. Mowdy; R.T. Newton; J.C. Personeau; J.A. Rorberg; G. Shum; T.S.
Thomas; R. Wagner; C. Whitby; G. Yuen

A collaborative study was conducted to test a modification to the AOAC fluorometric method for histamine (AOAC Official Method 977.13) that substitutes 75% methanol as the extracting solvent. All
other steps remain unchanged. The extracts prepared with 75% methanol were also used to collaboratively test a gas chromatographic (GC) method
for determination of putrescine and cadaverine in
seafood. In the GC method, the extracted diamines
are converted to fluorinated derivatives, the reaction mixtures are passed through solid-phase extraction columns, and the derivatives are quantitated by electron capture GC after separation on an
OV-225 column. Fourteen laboratories using the
GC method for putrescine and cadaverine and
16 laboratories using the fluorometric method for
histamine analyzed 14 canned tuna and raw mahimahi (including blind duplicates and a spike) containing 0.22.6 ppm putrescine, 0.69.1 ppm
cadaverine, and 0.6154 ppm histamine. At the
5 ppm level, recoveries ranged from 71 to 102% for
putrescine and 77 to 112% for cadaverine; the respective repeatability relative standard deviations
(RSDr) were 5.2 and 15%, and the respective reproducibility relative standard deviations (RSDR)
were 8.8 and 18%. At the 50 ppm level, histamine recoveries ranged from 84 to 125%, RSDr was 3.6%,
and RSDR was 9.4%. The GC method for determination of putrescine in canned tuna and cadaverine in
canned tuna and mahimahi has been adopted first
Submitted for publication February 16, 1996.
The recommendation was approved by the Methods Committee on
Commodity Foods and Commodity Products, and was adopted by the
Official Methods Board of the Association. See Official Methods Board
Actions (1996) J. AOAC Int. 79, 76A, and Official Methods Board
Actions (1996) The Referee, June issue.

action by AOAC INTERNATIONAL, and the AOAC

Official Method 977.13, Histamine in Seafood,
Fluorometric Method, has been modified.

or many years considerable interest has focused on finding reliable chemical indicators of decomposition to
characterize spoilage in seafood. There are several potential areas for biochemical evaluation to assess quality. For example, metabolic products from actions of enzymes that remain
active in fish after death have been investigated as indicators of
decomposition. These metabolites, which include amines, lactic acid, and hypoxanthine, support growth of bacteria that have
invaded muscle tissue from the gills, intestines, and surface of
the fish.
Metabolites from this microbial action also are candidate
indicators of decomposition. Microbial spoilage produces enzymes called decarboxylases. Decarboxylation of the amino
acids ornithine and lysine produces the diamines putrescine and
cadaverine, respectively. Quantitative analyses by gas chromatography (GC) of a variety of seafood (cod, sole, mahimahi,
and tuna) have shown low levels of putrescine and cadaverine
in fish of acceptable quality and increasing concentrations as
spoilage proceeds. Histamine is formed from decarboxylation
of free histidine. Because there is essentially no free histamine
in fresh tuna, the presence of histamine indicates decomposition and, at sufficiently high levels, has the potential for scombroid poisoning. The U.S. Food and Drug Administration
(FDA) has applied a defect action level (DAL) for histamine as
an index of decomposition and scombroid poisoning since 1978.
Formation of specific decomposition products depends on
the microbial flora present and the marine species. The products formed are a result of storage time and temperature. In
many cases, either cadaverine or putrescine is the major product.
The relation between changes in amines and decomposition
of marine products has been studied by many investigators.
Cadaverine, putrescine, tryptamine, and agmatine have been

identified as decomposition products due to timetemperature

abuse of seafood. A number of workers have reported increases
in putrescine and cadaverine in salmon, tuna, rockfish, shrimp,
lobster, cod, halibut, and Spanish mackerel that correlate with
progressing decomposition (15). In addition to cadaverine and
putrescine, tryptamine was identified during low-temperature
spoilage of saury pike, sardine (6, 7), salmonoid fish (8), octopus, and short-neck clams (9). Marked increases in levels of
agmatine, putrescine, and cadaverine were noted as decomposition progressed in squid (10). Because significant amounts of
cadaverine and putrescine are formed in a wide variety of seafood as a result of decomposition, they are useful indicators of
decomposition. This collaborative study was initiated as a step
in the process of establishing DALs for cadaverine and putrescine.
There is compelling evidence that histamine is the principal causative agent in scombroid food poisoning (11, 12).
High histamine levels are found in most cases of scombroid
poisoning associated with eating fish belonging to the families Scomberesocidae and Scombridae, such as tuna and
mackerel, and with nonscombroid fish, such as mahimahi
(13), bluefish (14), sardines, pilchards, anchovies, herring,
black marlin, and kahawai (12). Evidence indicates that
cadaverine and putrescine are possible potentiators of histamine toxicity caused by competitive inhibition of intestinal
histamine-metabolizing enzymes (12).
The GC method for determination of putrescine and
cadaverine developed by Staruszkiewicz and Bond (3) has
been modified to reduce the number of steps and the time of
analysis. The original method involved 4 steps: extraction of
the test portion, synthesis of a fluorinated derivative of
diamines with pentafluoropropionic (PFP) anhydride, purification of the reaction mixture by alumina column chromatography, and determination of the diamines by GC with
electron capture detection (ECD). Extraction has been modified by using 75% methanol in water instead of 100%
methanol, evaporation step after the PFP anhydride reaction
has been eliminated, and column chromatography has been
replaced by solid-phase extraction (SPE) alumina tubes. The
same extract (75% methanol in water) was used to determine
histamine in seafood by the fluorometric Method 977.13 (13).
Collaborative Study

Study Design
This study involved determination of putrescine, cadaverine, and histamine in canned tuna and raw frozen mahimahi.
The test samples contained putrescine levels ranging from 0.1
to 10 ppm, cadaverine levels ranging from 0.1 to 15 ppm, and
histamine levels ranging from 5.0 to 100 ppm (ppm = g/g =
10 mg/100 g). There were 7 blind duplicates (including a
spike), 1 blank, and 1 practice test sample, for a total of 16 determinations. Except for the spike, analytes were naturally incurred. Laboratories were asked to analyze all test samples in
duplicate and report all data to 0.01 ppm. They were asked to
send their results, including all calculations and chromatograms, with their comments to the study organizer.

Preparation of Test Samples

For canned tuna, 2030 cans from each of several lots of
varioius qualities were opened, mixed, and homogenized in a
commercial-size mixer. Mahimahi samples were prepared by
grinding raw fillets of mahimahi with a meat grinder and then
mixing and homogenizing the ground tissue in a Cuisinart food
processor. Test samples of approximately 15 g were put in
small Twirl-pac bags labeled with random numbers. Each labeled bag was placed in a larger bag identified with the same
number and frozen at 40C. Before shipment to collaborators,
the homogeneity of each batch was verified by taking 6 test
samples at random from each batch and determining putrescine, cadaverine, and histamine according to the prescribed
procedures. The averages of the 6 random test samples were
consistent with collaborative study averages.
A spiking solution accompanied each individual test sample. Collaborators were instructed to pipet exactly 1.0 mL of
the corresponding spiking solution into the blender after the
test portion was transferred and the 75% methanol was
added. The spiking solution for test sample 6 contained
49.2 g putrescine/mL, 49.7 g cadaverine/mL, and
501.5 g histamine/mL in 0.01N HCl. The spiking solution
for all other test samples was 0.01N HCl. The 14 test samples and the spiking solutions were packed in dry ice and
sent by overnight delivery to the participating laboratories.
Each laboratory received the methods under investigation,
instructions for participants, and data summary sheets.
Laboratories were instructed to keep test samples frozen until analysis and to perform analyses within 3 weeks of arrival. Each laboratory was asked to first analyze the practice
test sample to demonstrate the ability to obtain accurate results and then to proceed with a single measurement per test
portion and a reagent blank. Participants were instructed to
determine histamine in each extract by the fluorometric
method as described in Method 977.13 (13), with the modification in section D, Determination, reading: Extract prepared sample with 75% MeOH as in 957.07C, paragraph 1,
and to determine putrescine and cadaverine in the same extract by the GC diamine method described below.

Statistical Calculations
The conventional AOAC (IUPAC-87) Components of
Variance with standard outlier tests were used for analyses of
the data (15, 16).
996.07 Putrescine in Canned Tuna, and Cadaverine
in Canned Tuna and Mahimahi,
Gas Chromatographic Method
First Action 1996
(Applicable to determination of 0.3 g putrescine/g in canned
tuna and 0.5 g cadaverine/g in canned tuna and mahimahi.)
Caution: See Appendix B, safety notes on evaporators,
blenders, flammable solvents, toxic solvents, ethyl acetate, hydrochloric acid, hexane, methanol, and petroleum ether. Pen-

tafluoropropionic anhydride causes severe burns. Avoid contact with skin and eyes. Wear gloves and use effective fume
removal device. Dispose of waste solvents in an appropriate
manner compatible with applicable environmental rules and
regulations.
Method Performance:
See Tables 996.07A and B for method performance data and
Tables 996.07C and D for recovery data.

A. Principle
Putrescine and cadaverine are extracted from test sample
with 75% methanol and then converted to fluorinated derivatives. Reaction mixture is passed through solid-phase extraction column, and derivatives are quantitated by electron capture
gas chromatography (GC) after separation on an OV-225 column.

B. Apparatus
(a) Gas chromatographic (GC) sysem.Equipped with
Ni pulsed electron capture detector. Operating conditions: injection port, 210C; detector, 320C; GC column, 170180C;
nitrogen flow, 25 mL/min; electrometer range, 1010 (full
scale). Detector makeup purge gas as needed (e.g.,
40 mL/min).
(b) GC column.1800 2 mm id, glass column, packed
with 3% OV-225 (50% cyanopropylphenyl and 50% methyl)
on 100120 mesh Gas-Chrom Q, or equivalent capillary column. Condition 2 h with gas flow at room temperature, increase temperature gradually (ca 6C/min) to 240C, and hold
16 h. Retention times of pentafluoropropionic derivatives of
putrescine, cadaverine, and hexanediamine should be 7, 9, and
12 min, respectively.
(c) Solid-phase extraction (SPE) tubes.3 mL, SUPELCLEAN LC-Alumina-N SPE tubes (available from Supelco
Inc., Bellefonte, PA 16823-0048, USA, or equivalent). Check
activity of each lot of SPE tubes as follows: Pass 2 mL petroleum ethertoluene solvent, C(k), through SPE tube and then
add to tube 50 L Sudan I dye solution, C(l). Add a few drops
of petroleum ethertoluene solvent, C(k), to remove dye from
top frit. Pass 3 mL solvent through column. If band of dye stays
in the top 4 mm of SPE tube, the batch is satisfactory for purification of calibration standard and samplePFP reaction mixtures.
(d) Analytical balance.
(e) Rotary evaporator.Capable of maintaining 50
5C.
(f) Water bath.Capable of maintaining 50 5C and
60 5C.
(g) Glassware.Volumetric flasks, 100 mL (glass-stoppered) and 1 L; round or flat bottom 24/40 flasks, 100 mL;
graduated cylinders, 100 and 500 mL (glass-stoppered).
(h) Blender.For test sample preparation.
(i) Pipets.Capable of accurately delivering 1, 2, 3, 5, 10,
and 25 mL.
(j) Micropipets.Capable of accurately delivering 50,
150, and 500 L.
(k) Folded filter paper.
(l) Glass syringe.1.0 mL, calibrated in 0.01 mL.
63

C. Reagents
(a) Pentafluoropropionic (PFP) anhydride.Refrigerate
and protect from moisture (available from Pierce Chemical
Co., Rockford, IL 61105, USA).
(b) Ethyl acetate.Distilled-in-glass. Suitable for GC.
(c) Toluene.Distilled-in-glass. Suitable for GC.
(d) Hexane.Suitable for GC.
(e) Hydrochloric acid (HCl).1 and 0.1N.
(f) Ethyl acetatetoluene solvent.30% Ethyl acetate in
toluene. Transfer 150 mL ethyl acetate to stoppered graduated
cylinder, add 350 mL toluene, and mix.
(g) Methanol.75%, v/v. Place 750 mL MeOH (distilledin-glass) into 1 L volumetric flask. Dilute to volume with H2O.
Swirl flask while adding H2O.
(h) Hexanediamine internal standard solutions.(1)
Hexanediamine standard stock solution.Dissolve 163.0 mg
hexanediamine dihydrochloride in 0.1N HCl in 100 mL volumetric flask and dilute to volume with 0.1N HCl. (2)
20 g/mL.Pipet 2 mL hexanediamine standard stock solution, (1), into 100 mL volumetric flask and dilute to volume
with 0.1N HCl. (3) Hexanediamine internal standard working
solution.5 g/mL. Pipet 25 mL 20 g/mL hexanediamine
standard solution into 100 mL volumetric flask and dilute to
volume with 0.1N HCl.
(i) Putrescinecadaverine standard stock solution.Dissolve 91.4 mg putrescine dihydrochloride and 171.3 mg
cadaverine dihydrochloride in 0.1N HCl in 100 mL volumetric
flask and dilute to volume with 0.1N HCl.
(j) Putrescinecadaverine standard working solutions.
(1) Solution A.10 g cadaverine/mL and 5 g putrescine/mL as free base. Pipet 1 mL putrescinecadaverine standard stock solution, (i), into 100 mL volumetric flask and dilute
to volume with 0.1N HCl. (2) Solution B.1 g cadaverine/mL and 0.5 g putrescine/mL as free base. Pipet 10 mL solution A into 100 mL volumetric flask and dilute to volume
with 0.1N HCl.
(k) Petroleum ethertoluene solvent.4 + 1, v/v. For
checking activity of SPE tubes.
(l) Sudan I dye solution.For checking activity of SPE
tubes. Dissolve 0.01 g Sudan I dye (97%) in 8 mL petroleum
ethertoluene solvent, (k).

D. Preparation of PutrescineCadaverine Calibration

Standard Solutions
Prepare putrescinecadaverine calibration standard solutions IIV (see Table 996.07E) as follows: Pipet 1 mL aliquots
of hexanediamine internal standard working solution, C(h)(3),
and indicated volumes of putrescinecadaverine standard
working solutions (see Table 996.07E) into separate 100 mL
round- or flat-bottom 24/40 flasks. Add ca 0.5 mL 1N HCl into
each flask and evaporate to dryness on rotary evaporator at ca
5060C. (Note: Provide adequate chilling and vacuum to obtain evaporation.) Rinse each flask with 12 mL H2O and
evaporate to dryness again to remove last trace of HCl.
To dried residue, add 1 mL ethyl acetate, C(b), and 300 L
PFP anhydride, C(a), using glass syringe. Close flask with stop-

per, mix, and heat 30 min at 50C in water bath. (Note: Because
of pressure increase as PFP reaction is heated, stoppers should
be secured with restraining clip.)
Swirl solution at least once during reaction. After reaction
of PFP anhydride with residue of putrescinecadaverine calibration standard solution, the resulting putrescinecadaverine
calibration standardPFP reaction mixture should remain clear.
Within 2 h after removal from water bath, proceed to next step
to purify putrescinecadaverine calibration standardPFP reaction mixture using SPE tube.
Add 2 mL toluene, C(c), to putrescinecadaverine calibration standardPFP reaction mixture. Add 2 mL hexane, C(d),
to each SPE tube and let flow through by gravity. Discard hexane. Add 150 L putrescinecadaverine calibration standard
PFPtoluene reaction mixture to top of SPE tube. Start collecting effluent when mixture is added. Add 3 or 4 drops 30% ethyl
acetatetoluene solvent, C(f), to tube. After mixture passes into
frit, add 2 mL 30% ethyl acetatetoluene solvent. Add additional 6 mL 30% ethyl acetatetoluene solvent (total of 8 mL)
and collect entire effluent. Effluent is stable at least 1 week
when stored in refrigerator in the dark.

E. Preparation of Test Sample

Extract product with 75% methanol as follows: Transfer
10.00 g product to blender bowl and add ca 60 mL 75% methanol, C(g). Blend ca 2 min at high speed. Transfer to 100 mL
glass-stoppered volumetric flask, rinsing lid and blender jar
with 75% methanol and adding rinses to flask. Place flask in
60C water bath and maintain 15 min. Cool to room temperature and dilute to volume with 75% methanol. Mix by inverting
and filter through folded filter paper. Methanol extracts are stable at least 4 months when refrigerated.
Pipet 10 mL methanol extract into 100 mL round- or flatbottom flask, add 1 mL hexanediamine internal standard working solution, C(h)(3), and ca 0.5 mL 1N HCl. Evaporate to dryness on rotary evaporator at ca 5060C. (Note: Extracts must
be evaporated to dryness. Provide adequate chilling and vacuum to obtain evaporation.) After all solvent is evaporated, add
23 mL H2O, swirl, and evaporate again. (Note: This will
eliminate the last trace of HCl and ensure that extract is dry.)
Dried residues of methanol extracts (before reaction with PFP
anhydride) are stable at least 3 days.
To dried residue of methanol extract, add 1 mL ethyl acetate
and 300 L PFP anhydride using glass syringe. Close flask
with stopper, mix, and heat 30 min at 50C in water bath. (Note:
Because of pressure increase as PFP reaction is heated, stoppers
should be secured with restraining clip.)
Swirl solution at least once during reaction. After reaction
of PFP anhydride with residue of methanol extract, the resulting samplePFP reaction mixture will turn yellow with most
fish products. (Note: Clear samplePFP reaction mixture indicates presence of H2O in residue of methanol extract. In such
case, repeat evaporation step and then proceed with analysis.)
Within 2 h after removal from water bath, proceed to next step
to purify samplePFP reaction mixture using SPE tube.

Proceed as in D, beginning at Add 2 mL toluene, using

samplePFP reaction mixture instead of putrescinecadaverine
calibration standardPFP reaction mixture.

F. GC Determination
Adjust GC system to give full-scale recorder response for injection of 1 L derivatized putrescinecadaverine calibration
standard solution IV from D. For products containing lower levels
of analytes (e.g., 05 g cadaverine/g), set derivatized putrescine
cadaverine calibration standard solution III (5 g/g) full scale.
Correlation coefficient of standard curve should be 0.9.
Inject 12 L effluent of sample from E onto GC system.
(Note: Full calibration of GC system with duplicate injections of each of putrescinecadaverine calibration standard solutions is required only at the beginning of analysis, unless instrument is shut down between analyses. When analyzing test
portions on succeeding days, rerun derivatized putrescine
cadaverine calibration standard solutions II and III through GC
system to ensure that calibration of instrument is stable.)

G. Calculations
Perform following calculations:
(1) Calculate response ratio of cadaverine (RSc) in each putrescinecadaverine calibration standard solution as follows:
RSc =

PSc
PSh

where PSc = peak height of PFP cadaverine in putrescine

cadaverine calibration standard solution, and PSh = peak height
of PFP hexanediamine in putrescinecadaverine calibration
standard solution.
Plot RSc vs g cadaverine derivatized (Wc).
(2) Calculate response ratio of putrescine (RSp) in each putrescinecadaverine calibration standard solution as above, using peak height of PFP putrescine (PSp).
Plot RSp vs g putrescine derivatized (Wp).
(3) Calculate response ratio of cadaverine in test sample
(RTc) as follows:
RTc =

PTc
PTh

where PTc = peak height of PFP cadaverine in test sample, and

PTh = peak height of PFP hexanediamine in test sample.
(4) Calculate response ratio of putrescine in test sample
(RTp) as above, using peak height of putrescine in test sample
(PTp).
(5) Determine Wc and Wp for test samples from calibration
graph.
(6) Calculate amount of cadaverine in test sample as follows:
Cadaverine, g/g =

Wc F 10
Wt

where F = attenuation (or dilution) factor, if required, and Wt =

weight of test sample, g.

(7) Calculate amount of putrescine in test sample as above,

using Wp value.
Report all results to 0.1 g/g.
If values obtained for putrescine and cadaverine in test sample are higher than those obtained for the most concentrated
putrescinecadaverine calibration standard solution, quantitate
putrescine and cadaverine levels in test sample by analyzing
smaller aliquot of 75% methanol extract (e.g., 1 mL) and multiply by appropriate factor to calculate values in g/g. Smaller
volume of test solution may be injected or instrument attenuation may be adjusted to estimate putrescine and cadaverine
content.
In addition to the above method, Method 977.13, Histamine
in Seafood, Fluorometric Method, was modified to include
75% methanol as the extraction agent.
Ref.: J. AOAC Int. 80, 591602(1997)
Results and Discussion
The composition of the 7 collaborative test samples is
shown in Table 1. The samples were prepared from seafood
of commercial origin and were selected to contain levels of
the amines associated with commercial grade products of
passable quality through levels representing decomposed
seafood. Very fresh tuna and mahimahi contain <0.1 ppm
cadaverine and putrescine. Commercial grade products frequently contain 0.20.3 ppm of the amines; organoleptic
classification of decomposed material generally correlates
with levels of either cadaverine or putrescine above
0.5 ppm. The amount of each amine (cadaverine, putrescine,
or histamine) that forms during spoilage depends on the type
of bacterial contamination and the particular conditions of
decomposition to which the product is subjected. The general association of scombrotoxic fish such as tuna and mahimahi with high histamine and cadaverine levels if spoilage
occurs before chilling is of particular significance, whereas
elevated histamine and putrescine levels may be due to decomposition caused by unsanitary practices in food service
operations. Test sample 1 represented commercial product
usually graded acceptable; the remaining test samples
contained various levels of decomposition product. The
mean value for each analyte in each test sample obtained in
the collaborative study (Tables 996.07A, 996.07B, and 2)
was compared with the mean value for each analyte in the
original 6 randomly selected test samples (Table 1). This
comparison shows that all values for putrescine and cadaverine agree within 1 ppm. Average values for histamine are
within 1 ppm for 5 test samples and within 5 ppm for 2 test
samples containing higher levels of histamine. This comparison indicates that the collaborative test samples were homogeneous.
Sixteen laboratories participated; 2 determined histamine
only. The data from 14 laboratories for putrescine and cadaverine are presented in Tables 3 and 4, and the histamine data from
16 laboratories are presented in Table 5. Data from 14 laboratories were used in the statistical analysis. Results for the 7 levels were paired for all laboratories and examined for outliers by

the Cochran, Grubbs, and paired Grubbs tests. Outliers are

identified in Tables 35. The statistical results were calculated
from 90% of the data reported by 14 laboratories.
The data from 2 laboratories were not included. There appeared to be an error in the preparation of the standard solutions
in one laboratory because the ratio of the putrescine peak to the
internal standard peak was low by a factor of 4 to 5. A nonlinear
standard curve appeared to be the result of extending the concentration beyond the range recommended in the method;
when the chromatograms and results were received, it was too
late to ask for the analyses to be repeated. A second laboratory
experienced wide variations in the height of the internal standard peak, which may have been due to analyst error in addition
of internal standard. The collaborators were contacted and an
attempt was made to find the cause of the unacceptable results,
without success. Constraints on time and resources of the laboratories concerned did not allow repetition of analyses. Therefore, their data were not included in the statistical analysis.
The histamine data from laboratory L were discarded by one or
more tests for outliers. Laboratory L experienced problems in
maintaining stable fluorescence readings. The collaborators calculated histamine in mg/100 g. To avoid confusion in units in discussing the 3 analytes, the histamine values were converted to
ppm in this report. All 3 analytes are reported in ppm (ppm = g/g
= 10 mg/100 g). The statistical data are presented in Tables 996.07A, 996.07B, and 2. The mean, the repeatability relative
standard deviation (RSDr), the reproducibility relative standard
deviation (RSDR), and the Horwitz ratio were calculated. The Horwitz ratio is the ratio of the relative standard deviation among laboratories (RSDR) to the relative standard deviation predicted by the
Horwitz curve for a given concentration (17, 18). The Horwitz equation assesses the acceptability of collaborative values representing a
wide range of analytes, matrixes, and measurement techniques. Values within the range of 0.52 times the predicted value are considered acceptable.
Mean putrescine values (Table 996.07A) ranged from 0.18
to 4.54 ppm. RSDr values ranged from 4.4 to 17.1%, RSDR
values ranged from 8.0 to 44.4%, and Horwitz ratios ranged
from 0.56 to 3.0. The test sample with the lowest putrescine
concentration (0.18 ppm) had a Horwitz ratio of 2.1, and test
sample 7 (mahimahi) had a Horwitz ratio of 3.0. The other
5 levels all had Horwitz ratios of <2.
The Horwitz ratio for putrescine values reported for the mahimahi sample was >2 because of high results from 3 laboratories and low results from 2 laboratories.
The high results could be due to deterioration in the packed
column. Collaborator C, who shared an instrument with other
operators, noted a decrease in column performance after the
column had been left inadvertently in the GC while other analyses were being done at a higher temperature. Sample 7 was run
after this, and baseline separation of putrescine from an adjacent peak was not achieved. Packed columns can deteriorate
over several months of continuous use or from too high an oven
temperature (probably because of loss of the liquid phase) and
result in a decrease in resolution. Chromatograms from laboratory G reporting high results appeared to have late-eluting
peaks from a previous injection eluting with the sample. This

can occur with autoinjections, when the time lapse between

samples is too short. Laboratory K did not comment on any
problems nor did it send the chromatograms. Any explanation
of their results would be a guess.
Two laboratories reported very low results. Laboratory N
reported low values for most samples. This could have been
due to not following the evaporation procedure as outlined in
the method. There was no obvious explanation for the low result from laboratory A. Nine of 14 laboratories showed very
good agreement for putrescine in mahimahi. The agreement
among laboratories was acceptable for putrescine levels in tuna
and mahimahi when there was good separation of the putrescine peak from adjacent peaks (see sample chromatograms).
Mean cadaverine values (Table 996.07B) ranged from 0.56
to 9.12 ppm. RSDr values ranged from 3.0 to 15.1%, RSDR
values ranged from 4.3 to 18.4%, and Horwitz ratios were <2,
ranging from 0.31 to 1.42 and indicating very good agreement
among laboratories for these samples.
Mean histamine values (Table 2) ranged from 5.63 to
158.4 ppm. RSDr values ranged from 3.6 to 21.4%, RSDR values ranged from 8.9 to 23.0%, and Horwitz ratios were <2,
ranging from 0.82 to 1.87 and indicating acceptable agreement
among laboratories for all samples.
Recovery data are presented in Tables 996.07C, 996.07D,
and 6. Recoveries were calculated on the basis of the amount
determined for each compound in test sample 6 (the spike)
compared with the average of 2 determinations for test sample 1 (the background for the spike).
Recoveries of putrescine ranged from 70.7 to 102.4% and
averaged 87.1%. Data from laboratories J, L, M, and N were
not used in calculating recoveries because either the background result or the spike result was an outlier. Recoveries of
cadaverine ranged from 76.7 to 111.9% and averaged 96.6%.
Recovery data were not calculated from laboratories K and M
because of outliers in background results. Results for the
2 spiked test portions analyzed by laboratory L were so far
apart that a recovery calculation would be meaningless. If such
results occurred in regular practice, the test would have to be
repeated. The very low result from laboratory L may have been
caused by incomplete addition or mixing of the spiking solution with the test portion. Laboratory N had low results for putrescine and cadaverine in most test samples. They might be
attributed to the fact that the extract was dried with heat and N2
and the dried extract was transferred to a different flask for the
reaction with PFP anhydride. This was not part of the method
and could have caused loss of analyte. For this reason, recoveries of putrescine and cadaverine were not calculated for data
from laboratory N. Recoveries of histamine ranged from 84.1
to 125.1% and averaged 96.4%.
Three collaborators determined putrescine and cadaverine
with a DB-225 megabore column in addition to the packed column. The mean, standard deviation, repeatability and reproducibility relative standard deviations of the results from the
3 laboratories using the megabore column are reported in Tables 7 and 8 for each test sample along with the statistical values of the results from all laboratories using the packed column.
There was excellent agreement between the packed column re-

sults and the megabore column results for both putrescine and
cadaverine in all test samples. Retention times obtained with
the megabore column (7, 10, and 12 min; 180C; 25 mL
N2/min) were comparable with those obtained with the packed
column, and the peaks were well separated. The method will
work equally well with either column.
Recommendation
The data indicate good within-laboratory and among-laboratories agreement for the test samples, despite instrument
problems in one laboratory and possible analyst error in 2 laboratories. On the basis of the results of this study, we recommend
that the GC method for determining putrescine and cadaverine
in tuna and mahimahi be adopted first action. We also recommend that the fluorometric Method 977.13 be modified in section D, Determination to read, Extract prepared sample with 75%
MeOH as in 957.07C, paragraph 1, and that the calculation be
changed to report histamine in ppm = g/g = 10 mg/100 g.
Acknowledgments
We thank the following collaborators for their participation:
Roberta J. Berg, U.S. Food and Drug Administration,
Atlanta, GA
Gary Yuen, U.S. Food and Drug Administration, Brooklyn, NY
Richard T. Newton and Julie A. Rorberg, U.S. Food and
Drug Administration, Seattle, WA
Roberta Wagner and Christine Whitby, U.S. Food and Drug
Administration, Baltimore, MD
William Hummer, Terry S. Thomas, and Jennifer C. Personeau, U.S. Food and Drug Administration, Denver, CO
Lydell B. Hansen, U.S. Food and Drug Administration,
Dallas, TX
Don Mowdy and Ed Jester, U.S. Food and Drug
Administration, Dauphin Island, AL
Beth Atienza, Starkist, Terminal Island, CA
Judy Krzynowek, National Marine Fisheries, Gloucester, MA
Mike Jahncke, National Marine Fisheries, Pascagoula, MS
Dave McLachlan and Mike Gilgan, Department of Fisheries
and Oceans, Halifax, Nova Scotia, Canada
Henry Chin, National Food Processors Association, Dublin, CA
Karen Kennedy, Department of Fisheries and Oceans, St.
Johns, Newfoundland, Canada
Claude DesJardins, Department of Fisheries and Oceans,
Quebec, Canada
Gerald Shum, Department of Fisheries and Oceans,
Toronto, Ontario, Canada
Abe Adams, Bumble Bee Seafood, Santa Fe Springs, CA
The statistical analysis was provided by Richard H. Albert
and Richard F. Newell, Division of Mathematics, U.S. Food
and Drug Administration, Washington, DC.
References
(1) Meitz, J.L., & Karmus, E.J. (1978) J. Assoc. Off. Anal.
Chem. 61, 139145

(2) Karmus, E. (1981) Food Sci. Technol. 14, 273276

(3) Staruszkiewicz, W.F., Jr, & Bond, J.F. (1981) J. Assoc. Off.
Anal. Chem. 64, 584591
(4) Farn, G., & Sims, G.G. (1986) in Developments in Food Science, Vol. 15, Proceedings of an International Symposium
Coordinated by the University of Alaska Sea Grant College
Program, Anchorage, Alaska, D.E. Kramer & J. Liston (Eds),
Elsevier Science, New York, NY, pp. 175183
(5) Middlebrooks, B.L., Toom, P.M., Douglas, W.L., Harrison,
R.E., & McDowell, S. (1988) J. Food. Sci. 53, 10241029
(6) Yamanaka, H., Shimakura, K., Shiomi, K., & Kikuchi, T.
(1986) Nippon Suisan Gakkaishi 52, 127130
(7) Yamanaka, H., & Matsumoto, M. (1989) J. Food Hyg. Soc.
Jpn. 30, 396400
(8) Yamanaka, H., Shiomi, K., & Kisuchi, T. (1989) J. Food
Hyg. Soc. Jpn. 30, 170174

(9) Nagayama, T., Tamura, Y., Maki, T., Kan, K., Naoi, Y., &
Nishima, T. (1985) J. Hyg. Chem. 31, 362370
(10) Yamanaka, H. (1990) Food Rev. Int. 6, 591602
(11) Morrow, J.D., Margolies, G.R., Rowland, J., & Roberts, J.L.,
II (1991) N. Engl. J. Med. 324, 716720
(12) Taylor, S.L., Hui, J.Y., & Lyons, D.E. (1984) in Seafood Toxins, E.P. Ragelis (Ed.), ACS Symposium Series 262,
American Chemical Society, Washington, DC, pp. 417430
(13) Baranowski, J.D., Frank, H.A., Brust, P.A., & Premaratne,
R.J. (1990) J. Food Prot. 53, 217222
(14) Karolus, J., Leblanc, D., March, A.S., Mshan, R., & Furgalack, T.H. (1985) J. Food Prot. 48, 166168
(15) Horwitz, W. (1988) J. Assoc. Off. Anal. Chem. 71, 160172
(16) Horwitz, W. (1988) Pure Appl. Chem. 60, 855864
(17) Horwitz, W., Kamps, L.R., & Boyer, K.W. (1980) J. Assoc.
Off. Anal. Chem. 63, 13441354
(18) Horwitz, W. (1982) Anal. Chem. 54, 67A76A

Table 996.07A. Method performance for determination of putrescine in canned tuna by gas chromatographya
Sample
1
2
3
4
5
6
a
b
c

Mean, g/g

sr, g/g

sR, g/g

RSDr, %

RSDR, %

rb

Rc

0.323
0.184
0.524
0.778
2.606
4.545

0.053
0.028
0.089
0.058
0.196
0.234

0.064
0.080
0.141
0.072
0.209
0.399

16.49
15.57
17.09
7.51
7.51
5.17

19.75
43.6
26.86
9.23
8.03
8.77

0.148
0.078
0.249
0.162
0.549
0.655

0.179
0.224
0.395
0.202
0.585
1.117

Based on results from 14 laboratories.

r = 2.8 sr.
R = 2.8 sR.

Table 996.07B. Method performance for determination of cadaverine in canned tuna and mahimahi by gas
chromatographya
Sample
Canned tuna
Canned tuna
Canned tuna
Canned tuna
Canned tuna
Canned tuna, spiked
Mahimahi
a
b
c

Mean, g/g

sr, g/g

sR, g/g

RSDr, %

RSDR, %

rb

Rc

0.564
1.064
1.118
2.573
9.120
5.171
8.390

0.056
0.082
0.099
0.076
0.515
0.779
0.418

0.104
0.104
0.117
0.111
0.648
0.914
1.239

9.91
7.71
8.88
2.98
5.65
15.06
4.45

18.40
9.80
10.49
4.55
7.1
17.68
14.77

0.157
0.230
0.277
0.213
1.442
2.181
1.170

0.291
0.291
0.328
0.311
1.814
2.559
3.469

Based on results from 14 laboratories.

r = 2.8 sr.
R = 2.8 sR.

Table 996.07C. Recovery of putrescine added to canned tunaa

Laboratoryb

Background, g/g

Found, g/g

Recovered, g/g

Recovery, %

0.35

0.31

0.35

0.32

0.33

0.42

0.24

0.31

0.26

0.40

4.60
4.80
5.35
4.82
4.78
4.65
4.88
4.83
4.83
4.51
4.00
4.15
4.66
4.74
4.71
4.59
4.57
3.74
4.61
4.42

4.25
4.45
5.04
4.51
4.43
4.30
4.56
4.51
4.50
4.18
3.58
3.73
4.42
4.50
4.40
4.28
4.31
3.48
4.21
4.02

86.38
90.45
102.44
91.67
90.04
87.40
92.68
91.67
91.46
84.96
72.76
75.81
89.84
91.46
89.43
86.99
87.60
70.73
85.57
81.71
87.05

Average recovery, %
a
b

Added 4.92 g putrescine/g.

Data from laboratories J, L, M, and N were excluded because results for the background or the spike were outliers.

Table 996.07D. Recovery of cadaverine added to canned tunaa

Laboratoryb

Background, g/g

Found, g/g

Recovered, g/g

Recovery, %

0.53

0.54

0.46

0.57

0.50

0.58

0.59

0.72

0.60

0.71

5.10
5.40
5.33
5.48
5.35
5.33
5.40
5.30
5.33
4.58
5.38
5.32
5.43
5.50
5.38
5.66
5.52
4.41
6.13
6.27

4.57
4.87
4.79
4.94
4.89
4.87
4.83
4.73
4.83
4.08
4.80
4.74
4.84
4.91
4.66
4.94
4.92
3.81
5.42
5.56

91.95
97.99
96.38
99.40
98.39
97.99
97.18
95.17
97.18
82.09
96.58
95.37
97.38
98.79
93.76
99.40
98.99
76.66
109.05
111.87
96.58

Average recovery, %
a
b

Added 4.97 g cadaverine/g.

Data from laboratories K and M were excluded because results for the background were outliers. Laboratory L had very large variation in
duplicate spiked samples, and laboratory N did not follow the procedure. Therefore, their recovery data were excluded.

Table 996.07E. Preparation of putrescinecadaverine calibration standard solutions

Putrescinecadaverine
calibrating standard solution
I
II
III
IV
a

Putrescinecadaverine
standard working solution

Volumea, mL

Equivalent cadaverine
level in test sample, g/g

Equivalent putrescine level in

test sample, g/g

B
B
B
A

0.5
1.0
5.0
1.0

0.5
1
5
10

0.25
0.5
2.5
5

Added to 1 mL hexanediamine standard working solution.

Table 1. Composition of test samplesa

Test sample

Description

Putrescine content, ppm

Canned tuna

0.31 0.04

0.5 0.10

8.96 0.76

Canned tuna

0.2 0.05

1.04 0.05

12.08 0.35

Canned tuna

0.51 0.07

1.03 0.08

4.78 0.39

Canned tuna

0.71 0.08

2.47 0.09

7.41 0.29

Canned tuna

2.57 0.16

9.41 0.38

20.35 1.54

Canned tuna, spiked

4.67 0.34

5.03 0.49

56.98 2.50

Mahimahi

1.54 0.13

8.14 0.30

153.43 10.67

Cadaverine content, ppm

Histamine content, ppm

Each value is the mean standard deviation from analyses of 6 test portions of each collaborative test sample to determine homogeneity.

Table 2. Statistical analysis of collaborative results for fluorometric determination of histamine (ppm) in canned
tuna and raw frozen mahimahi
Test sample
Parameter
No. of laboratories
No. of results
No. of accepted results
Mean, ppm
sr, ppm
RSDr, %
sR, ppm
RSDR , %
Horwitz ratio

16
32
28
9.896
0.884
9.01
0.915
9.32
0.82

16
32
28
13.0
1.71
13.17
1.71
13.17
1.21

3
16
32
26
5.627
1.205
21.42
1.295
23.03
1.87

16
32
26
7.831
1.084
13.84
1.084
13.84
1.18

16
32
28
20.28
1.140
5.62
2.077
10.24
1.01

16
32
28
58.02
2.111
3.64
5.466
9.42
1.08

7
16
32
30
158.4
6.575
4.15
14.05
8.87
1.19

Table 3. Collaborative results for determination of putrescine, ppm

Test sample
Laboratory
A
B
C
D
E
F
G
H
I
J
K
L
M
N

Average
a
b
c
d

0.40
0.30
0.31
0.31
0.33
0.37
0.31
0.32
0.33
0.32
0.44
0.39
0.20
0.27
0.33
0.29
0.23
0.28
b,d
0.94b,d
b,d
0.54b,d
0.50
0.30
0.28
0.35
d
0.70d
d
0.59d
0.30
0.30

0.20
0.20
0.11
0.08
0.19
0.21
0.24
0.20
0.20
0.17
0.11
0.16
0.18
0.23
0.12
0.13
a
NDa
0.08
0.35
0.29
0.17
0.25
0.17
0.14
0.33
0.37
0.10
0.10

0.50
0.60
0.45
0.48
0.54
0.54
0.45
0.45
0.57
0.55
0.54
0.46
0.62
0.40
0.48
0.57
0.46
0.49
0.76
0.88
0.60
0.465
0.50
0.29
0.85
0.58
0.30
0.30

0.75
0.80
0.69
0.77
0.81
0.77
0.64
0.65
0.81
0.77
0.88
0.79
0.73
0.79
0.72
0.76
0.88
0.72
d
1.25d
d
1.12d
0.84
0.67
0.79
0.85
0.86
0.89
d
0.40d
d
0.35d

2.60
2.60
2.68
2.46
2.58
2.60
2.44
2.48
2.72
2.59
2.66
2.71
2.21
2.77
2.94
2.94
2.55
2.47
c,d
4.02c,d
c,d
3.66c,d
3.14
2.56
2.66
2.34
2.57
2.27
b
3.15b
b
1.45b

4.60
4.80
5.35
4.82
4.78
4.65
4.88
4.83
4.83
4.51
4.00
4.15
4.66
4.74
4.71
4.59
4.57
3.74
b
7.60b
b
10.80b
4.61
4.42
b,d
1.93b,d
b,d
5.09b,d
3.91
3.84
c,d
2.10c,d
c,,d
1.55c,d

0.80
0.90
1.90
1.84
3.01
2.96
1.41
1.46
1.70
1.60
1.60
1.63
2.55
2.50
1.65
1.40
1.04
1.15
c,d
6.39c,d
c,d
7.66c,d
3.23
3.03
1.51
1.50
1.27
1.27
0.70
0.70

0.32

0.18

0.52

0.78

2.61

4.55

1.70

ND = not determined.
Removed as outlier by Cochrans test.
Removed as outlier by Grubbs test.
Removed as outlier by paired Grubbs test.

Table 4. Collaborative results for determination of cadaverine, ppm

Test sample
Laboratory
A
B
C
D
E
F
G
H
I
J
K
L
M
N
Average
a
b
c

0.55
0.50
0.55
0.52
0.47
0.45
0.55
0.58
0.48
0.52
0.61
0.54
0.67
0.51
0.77
0.66
0.52
0.67
0.72
0.69
a, b
1.16a, b
a,b
0.73a,b
0.62
0.63
C
1.33c
C
0.82c
0.35
0.40
0.56

1.00
1.10
1.10
1.00
1.03
1.02
1.08
1.05
1.03
1.00
1.05
1.10
1.20
0.96
1.11
1.14
1.10
1.07
1.25
1.25
a, b
0.99a, b
a,b
2.67a,b
1.20
1.11
0.77
1.04
1.00
0.90
1.06

1.10
1.20
1.09
1.05
0.98
0.99
1.07
1.15
0.94
1.04
1.06
0.93
1.09
1.12
1.16
1.30
1.20
1.22
1.17
1.35
1.34
0.97
1.14
1.18
b
1.33b
b
1.65b
a
4.20a
a
1.10a
1.12

2.60
2.60
2.61
2.55
2.56
2.42
2.50
2.53
2.53
2.47
2.74
2.66
2.53
2.59
2.77
2.77
2.63
2.53
2.47
2.55
a,b
5.72a,b
a,b
3.28a,b
2.56
2.73
2.30
2.54
b
1.30b
b
1.65b
2.57

9.30
9.20
9.56
9.64
8.99
8.91
9.40
9.27
8.89
9.30
9.45
9.54
7.74
9.01
9.58
9.80
9.51
9.25
9.71
9.59
a,b
11.60a,b
a,b
3.02a,b
8.73
8.37
8.82
7.05
9.90
8.60
9.12

5.10
5.40
5.33
5.48
5.35
5.33
5.40
5.30
5.33
4.58
5.38
5.32
5.43
5.50
5.38
5.66
5.52
4.41
6.13
6.27
5.02
7.58
2.50
5.40
4.79
4.71
3.60
3.60
5.17

7.05
7.40
8.66
8.56
9.01
8.67
8.32
8.32
8.20
7.70
8.43
8.95
7.61
8.31
11.7
10.89
5.88
6.37
9.49
10.66
8.64
8.10
8.10
8.38
7.74
7.79
8.00
8.00
8.39

Removed as outlier by Cochrans test.

Removed as outlier by Grubbs test.
Removed as outlier by paired Grubbs test.

Table 5. Collaborative results for determination of histamine, ppm

Test sample
Laboratory
A
B
C
D
E
F
G
H
I
J
K
L
M
N
O
P
Average
a
c
b

1
10.0
10.0
9.5
10.2
10.1
9.2
9.2
8.7
9.8
9.2
9.5
9.8
10.3
12.8
9.8
8.6
a,c
9.6a,c
a,c
16.6a,c
9.4
9.3
9.0
11.5
a,b,c
52.2a,b,c
a,b,c
107.7a,b,c
7.7b,c
b,c
5.6b,c
10.4
9.7
9.0
11.0
10.9
10.4
9.9

15.0
13.0
13.0
13.9
11.4
14.9
12.1
11.9
11.7
12.2
10.9
13.5
11.3
16.9
12.1
13.6
16.3
11.5
11.9
12.0
b,c
9.0b,c
b,c
7.0b,c
a,b,c
75.5a,b,c
a,b
196.9a,b,c
12.0
12.5
12.5
12.4
14.0
15.0
13.3
13.2
13.0

5.0
8.0
4.9
4.6
6.2
6.8
4.7
4.4
5.1
4.9
a,b,c
5.2a,b,c
a,b,c
15.6a,b,,c
5.4
6.1
7.3
4.8
b,c
19.9b,c
b,c
6.9b,c
5.1
4.8
6.5
4.5
a ,b,c
11.8a,b,c
a ,b,c
84.0a,b,c
6.7
4.3
5.2
6.0
6.5
9.5
3.8
5.2
5.6

8.0
8.0
8.4
7.4
5.2
9.4
7.3
7.6
7.1
7.8
9.2
7.8
8.0
7.8
7.1
9.1
9.6
8.6
7.5
7.8
b,c
4.5b,c
b,c
5.5b,c
a,b,c
51.7a,b,c
a,b,c
98.0a,b,c
8.6
6.8
7.9
6.9
c
12.5c
c
8.0c
7.4
7.3
7.8

23.0
23.0
22.0
20.5
24.4
22.4
21.8
21.9
18.7
19.5
19.8
20.0
20.6
20.0
17.9
17.2
a,c
21.0a,c
a,c
9.1a,c
20.0
19.5
20.0
21.0
b,c
80.4b,c
b,c
73.1b,c
18.2
16.8
17.2
20.3
21.0
24.0
17.3
19.9
20.3

69.0
66.0
63.3
62.2
67.0
72.4
55.0
58.1
58.4
55.1
58.1
51.8
61.4
57.8
54.1
54.1
58.3
45.3
55.6
56.1
53.0
53.0
b,c
111.0b,c
b,c
104.2b,c
53.8
52.8
54.3
57.6
59.0
59.0
53.3
53.3
58.0

151.0
156.0
159.8
161.1
153.8
157.3
141.9
146.8
163.0
162.0
149.6
146.3
157.5
148.0
167.6
154.1
131.6
143.1
173.6
168.9
144.0
144.0
b,c
278.8b,c
b,c
296.7b,c
175.3
194.5
184.1
172.5
178.0
161.0
152.5
154.3
158.4

Removed as outlier by Cochrans test.

Removed as outlier by paired Grubbs test.
Removed as outlier by Grubbs test.

Table 6.

Recovery of histamine added to tunaa

Laboratoryb

Background, ppm

Found, ppm

Recovered, ppm

Recovery, %

10.00

9.85

9.65

8.95

9.50

9.65

11.55

9.20

9.30

10.25

10.10

10.00

10.70

69.00
66.00
63.30
62.20
67.00
72.40
55.00
58.10
58.40
55.10
58.10
51.80
61.40
57.80
54.10
54.10
55.60
56.10
53.00
53.00
54.30
54.30
59.00
59.00
53.30
53.30

59.00
56.00
53.45
52.35
57.35
62.75
46.05
49.15
48.90
45.60
48.45
42.15
49.85
46.25
44.90
44.90
46.30
46.80
42.75
42.75
44.20
44.20
49.00
49.00
42.60
42.60

117.65
111.67
106.58
104.39
114.36
125.12
91.82
98.01
97.51
90.93
96.61
84.05
99.40
92.22
89.53
89.53
92.32
93.32
85.24
85.24
88.14
88.14
97.71
97.71
84.95
98.89
96.41

Average recovery, %
a
b

Added: 50.15 g histamine/g.

Data from laboratories I, L, and M were excluded because results for the background or the spike were outliers. See Table 5.

Table 7. Putrescine values obtained by using megaborea column, ppm

Test sample
Laboratory
G
H
I
b

Mean, ppm
b
sr , ppm
b
RSDr , %
b
sR , ppm
b
RSDR , %
c
Mean, ppm
c
sr , ppm
c
RSDr , %
c
sR , ppm
c
RSDR , %
a
b
c

0.28
0.31
0.29
0.29
0.28
0.30
0.29
0.01
5.05
0.01
5.05
0.32
0.05
16.49
0.06
19.75

0.17
0.14
0.13
0.12
0.13
0.25
0.16
0.05
32.3
0.05
32.3
0.18
0.03
15.57
0.08
43.6

0.57
0.38
0.46
0.52
0.54
0.56
0.51
0.08
16.2
0.08
16.2
0.52
0.09
17.09
0.14
26.85

0.73
0.82
0.75
0.88
1.14
0.74
0.84
0.18
20.8
0.18
20.8
0.78
0.06
7.51
0.07
9.23

2.90
2.73
2.50
2.61
2.80
2.80
2.72
0.08
3.04
0.16
5.77
2.61
0.20
7.51
0.21
8.03

5.21
4.90
4.59
4.44
4.92
4.06
4.69
0.38
8.07
0.42
8.89
4.55
0.23
5.17
0.40
8.77

1.35
1.86
1.52
1.44
1.47
1.57
1.54
0.21
13.99
0.21
13.99
1.70
0.08
4.41
0.76
44.37

J & W DB-225, 0.54 mm 30 m, 1 m; 180C; N2 at 25 mL/min.

Statistics of results obtained by the 3 laboratories using the megabore column.
Statistics of results obtained by all laboratories using the packed column.

Table 8. Cadaverine values obtained by using megaborea column, ppm

Test sample
Laboratory
G
H
I
b

Mean, ppm
b
sr , ppm
b
RSDr , %
b
sR , ppm
b
RSDR , %
c
Mean, ppm
c
sr , ppm
c
RSDr , %
c
sR , ppm
c
RSDR , %
a
b
c

0.46
0.49
0.56
0.49
0.47
0.64
0.52
0.08
14.7
0.08
14.7
0.63
0.05
9.91
0.10
18.40

1.21
1.10
1.06
1.24
1.25
1.06
1.15
0.12
10.1
0.12
10.1
1.06
0.08
7.71
0.10
9.80

1.25
1.38
1.13
1.47
1.29
1.23
1.29
0.15
11.7
0.15
11.7
1.12
0.09
8.88
0.12
10.49

2.53
2.44
2.67
3.00
2.70
2.59
2.66
0.15
5.52
0.20
7.67
2.57
0.08
2.98
0.11
4.33

9.74
9.53
9.57
9.48
9.24
9.28
9.47
0.09
1.00
0.20
2.15
9.12
0.52
5.65
0.65
7.10

5.68
5.23
5.64
5.35
5.64
4.31
5.31
0.59
11.0
0.59
11.0
5.17
0.78
15.06
0.91
17.68

17.22
19.31
18.40
18.10
18.10
18.64
18.30
10.89
10.7
10.89
10.7
18.39
10.42
14.45
11.24
14.77

J & W DB-225, 0.54 mm 30 m, 1 m; 180C; N2 at 25 mL/min.

Statistics of results obtained by the 3 laboratories using the megabore column.
Statistics of results obtained by all laboratories using the packed column.