Simple and Rapid Analysis of Nitrofurazone from Blood, Milk, Urine and

Meat Samples

T. S u o r t t i .1) / K. H e i n o n e n 2)

1) Technical Research Centre of Finland, Food Research Laboratory, Biologinkuja 1, SF-02150 Espoo, Finland
2) College of Veterinary Medicine, Department of Obstetrics and Gynecology, SF-04840 Hautj~rvi, Finland

be obtained with some HPLC methods [1, 10,11], but

Key Words
they require tedious sample preparation including extrac-
Nitrofurazone tions, concentrations and column chromatography puri-
Dairy cows fications.
Blood and milk By combining the rather high selectivity of detection
Meat, fat, kidney, liver attainable at a wavelength of 365 nm with the sensitivity
obtained by trace enrichment, a simple sample preparation
was developed for the analysis of nitrofurazone in low
Summary concentrations at the #g/kg level.
Nitrofurazone is an effective chemotherapeutic drug in
the treatment of infections of the urinary tract. In order
to study its occurrence and metabolism, a simple and Materials and Methods
sensitive method was developed.
The liquid chromatograph consisted of M-6000A pump,
M-710B automatic injector, Nova-Pak Cle column
(8 X 100 mm) in RCM-100 column chamber and M-440
I ntroduction detector monitoring at 365 nm (all from Millipore/Waters).
The mobile phase was 23.5/76.5 acetonitrile/20mM
The method consists of the homogenization of the sample sodium acetate (pH 5.0). The solvents used were of HPLC
with acetonitrile: aqueous buffer, centrifugation, filtration quality and water was purified using a Milli-Q system. All
and dilution of the sample with buffer. The analysis is reagents were of analytical grade. Nitrofurazone and nitro-
carried out on a reverse phase column. Nitrofurantoin is furantoin stock solutions (600 rag/I) were made in dimethyl
used as an internal standard and also as a demonstration sulfoxide and standard solutions were diluted from them
that degradation of nitrofurazone by the action of light in 15/85 acetonitrile/sodium acetate buffer-solution. All
has not occurred. The sensitivity reached by this simple the samples were stored in the dark a t - 18 ~ The sample
preparation is of the order of 5#g/kg for most samples, preparations were carried out in a room illuminated with
with the exception of liver samples, which very quickly incandescent lamps and no fluorescent light or daylight.
degrade small additions of nitrofurazone and thus make
determinations impossible at concentrations below Blood and Milk Samples
40 #g/kg.
To a one milliliter sample, 10~1 10mg/I nitrofurantoin
The method can also applied to the analysis of the other
solution was added as internal standard and one milliliter
nitrofuranes. Nitrofurazone is an effective chemothera-
of acetonitrile. The sample was centrifuged for 25 minutes
peutic drug with a wide therapeutic spectra. In some
at 3000 g, and 4 ml 20 mM sodium asetate buffer was
countries, although not in Finland, it may be added to
added (pH 5.0). The samples were filter through a Millex-
pig and swine feed [1]. Nitrafurazone is used in the local
HV filter (0.45#m pores} and 100 ... 400#1 samples were
treatment of uterine infections of dairy cows. In order
injected into the HPLC instrument.
to study its absorption into systemic blood circulation
after intrauterine application as well as its residues in
Meat, Liver and Kidney Samples
edible tissues and milk, a sensitive and simple method was
developed. Samples were sliced and ground with an ordinary household
Several methods have been described for the analysis of grinder. A five milligram sample was taken, and a 2 5 p l
nitrofurans, ranging from colorimetry [2], through gas 10 mg/I nitrofurantoin solution was mixed into the sample.
chromatography [3, 4] to high performance liquid chro- The sample was then homogenized with 2 0 m l 30/70
matography (HPLC) [ 5 - 9 ] . The required sensitivity can acetonitrile/sodium acetate buffer and centrifuged at

344 Chromatograohia Vol. 24, 1987

0009-5893/87 0344 03 ~03 00.0 9 1987 Eri,~dr.Vieweg & Sohn Verlagsge.sellschaftm~H

 obtained with 400#1 injections. Thus, a simple sample
preparation with only an extraction/protein precipitation
step and a subsequent sample dilution step could be used
(Fig. 2). The recovery for both compounds in blood and
milk samples was 9 8 - 1 0 4 % at the 50 #g/I level and 76 %
in other samples. The standard deviation for the deter-
mination of nitrofurazone was 6.8%, and the standard
deviation using an internal standard for quantitation was
5.8 %. The sensitivity of the determination was in the
range of a few #g/kg for all the other samples except liver.
The beef liver samples contained disturbing impurities
which limited the sensitivity to about 4 0 # g / k g (Fig. 3);
these samples also necessitated frequent changing of the
pre-column. This cannot be regarded as a major drawback
of the method, as the alternative would be far more
complex sample preparation. Furthermore, the lower
sensitivity obtained with liver was less significnat because
it was shown during these studies that the freshly frozen
beef liver samples, in contrast to normal commercial pig

3 0 0 0 g for 15 minutes. Five milliliters of this suspension

was taken and diluted with 2 ml sodium acetate buffer and
centrifuged for 25 minutes at 3000 g. The suspension was
then filtered through a Millex-HV filter and injected into
the HPLC instrument.

Fat Samples

The sample was ground with an ordinary household grinder

and a 5 g sample was taken. 25#1 10 mg/I nitrofurantoin
solution and 6 ml acetonitrile were added and the sample
was melted at 60 ~ in a water-bath. It was then homo-
genized concurrently with the addition of 1 4 m l sodium
acetate buffer. After homogenization, the sample was
centrifuged for 15 minutes at 3000 g. To five milliliter of
this suspension, 2 ml sodium acetate buffer was added
before recentrifuging. Prior to injection, the sample was
filtered through a Millex HV-filter.

Results and Discussion

Nitrofurantoin was chosen as an internal standard because
it is chemically chromatographically and spectrophoto-
metrically suitable and shows a somewaht higher sensitivity
to degradation by light (30% loss cf. 10% for nitro-
furazone after 1.5 h exposure to daylight), which confirms
that degradation of nitrofurazone by light during sample
manupulation in blank samples had not taken place. The
standard curve for nitrofurazone was linear at the con-
centration range studied (10 ... 3000#g/I) and went
through the origin.
No disturbing peaks at the retention times of nitrofurazone
or nitrofurantoin were seen with blank samples. In order to
Fkj. 2
achieve the required sensitivity, trace enrichment was
utilized. As can be seen in Fig. 1, if the standard was in- Effect of sample solvent on chromatography of nitrofurazone [1l
and nitrofurantoin [2J with 200/zl injections.
jected in a solution where the acetonitrile concentration
A: 15/85 Acetonitrile/sodium acetate buffer
was 8 % less than in the mobile phase, no adverse effect on B: 20/80 Acetonitrile/sodium acetate buffer
peak shape was detected with 200 #1 injections. However, C: 25/75 Acetonitrile/sodium acetate buffer
the sensitivity was twenty-fold in comparison with normal Mobile phase 23.5/66.5 acetonitrile/sodium acetate buffer 1 ml/
10#1 injections. A forty-fold increase in sensitivity may be minutes flow-rate, 0.005 aufs at 365 nm.

Chromatographia Vol. 2 ~', 1987 345

 liver samples (which also gave much cleaner chromato-
grams), were capable of rapidly decomposing low levels
of both nitrofurazone and n i t r o f u r a n t o i n . The small sample
requirement of the method made it possible to scale it
d o w n for analysis of n i t r o f u r a z o n e concentrations in
150 mg endometrial biopsies.

Acknowledgements
Deepest thanks are due to H y m y , Sirkku, Lysti, Misa, Oili,
Valko, Nasta, Orko, Osta and L i r k k u , whose contributions
by donations of samples made this study possible.

References
[1 ] E. A. Sugden, A. I. Macintosh, A. B. Vilim, J. Assoc. Off.
Anal. Chem. 66, 874 (1983).
[2] R. J, Herrett, J. A. Buzard, Anal. Chem. 32, 1676 (1960).
[3] A. R. Ritchie, M. H. Clear, S. R. B. Sol/y, N. Z. J. Sci. 20,
225 (1977).
[4] U.R. Cieri, J. Assoc. Off. Anal. Chem. 62, 168 (1979).
(5J A . B . Vilim, A. /. Macintosh, J. Assoc. Off. Anal. Chem.
62, 19 (1979).
I6] V.A. Thorpe, J. Assoc. Off. Anal. Chem. 63, 981 (1980).
[7] M.B. Aufr~re, B.-A. Hoener, M . E . Vote, Clin. Chem. 23,
2207 (1977).
18J H. Roseboom, H . A . Koster, Anal. Cim. Acta 101, 359
(1978).
[9J T. B. Vree, Y. A. Hekster, A. M. Baars, J. E. Damsma, E. van
der Klein, J. Chromatogr. 162, 110 (1979).
[101 M. Petz, Dtsch. Lebensm.-Rundsch. 78, 396 (1982).
[11 I R. Malisch, Z. Lebensm. Unters. Forsch. 183, 253 (1986).

Received: June 29, 1987

Accepted: Aug. 5, 1987
F
Fig. 3
Chromatography of the samples prepared by the method described
with 400 ~ul injections.
A: 13.4/~g/I standard of nitrofuranzone [1] and 13.7/~g/I nitro-
furantoin [2J
B: Meat sample
C: Meat sample spiked with 20/Jg/kg nitrifurazone
D: Kidney sample
E: Kidney sample spiked with 20 p.g/kg nitrofurazone
F: Liver sample spiked with 40 #g/kg nitrofurazone
To all samples 50 jug/kg nitrofuratoin was added as internal standard.

346 Chromatograph~a Vol. 24, 1987