Animal & Veterinary
ANALYTICAL METHODS FOR THE DETERMINATION OF DIURETICS IN BOVINE MILK
by Badar Shaikh, Ph.D.
FDA Veterinarian Newsletter March/April 1999 Volume XIV, No II
Diuretics are chemically heterogeneous compounds used as therapeutic agents in certain pathological conditions to eliminate bodily fluids. They not only promote renal excretion of water and salt but also affect the renal absorption and excretion of other ions, e.g., potassium, calcium, and magnesium. The diuretics, due to their variety of chemical structures, have different pharmacological properties and accordingly are classified into four groups: carbonic anhydrase inhibitors, loop, thiazide and thiazide type, and potassium sparing diuretics. The carbonic anhydrase inhibitors, e.g., acetazolamide, decrease the conversion of bicarbonate ion to carbonic acid resulting in an increase in sodium, potassium, and bicarbonate renal excretion. This also increases the pH of the urine. Furosemide has been reported to be the most potent and short-acting loop diuretic. It is widely used to treat edematous states of hepatic, cardiac, and renal origin. It is also a common drug of abuse in livestock shows since it reduces tissue water resulting in improved muscle tone appearance. The thiazide diuretics, chlorothiazide (CTZ), hydrochlorothiazide (HCTZ), and trichlormethiazide (TCMTZ) increase the excretion of potassium and can cause hypokalemia during long-term maintenance therapy. Therefore, thiazides are often given with potassium sparing diuretics such as amloride, in order to maintain electrolyte balance in the body. Trichlormethiazide is given in combination with dexamethasone. The combined diuretic activity of TCMTZ and the specific anti-inflammatory activity of dexamethasone are complementary in the reduction of physiological parturient edema of the mammary gland and associated structures. Because the two drugs are complementary in their action, effects are achieved with a minimum dosage of TCMTZ. The clinically determined saluretic potency of TCMTZ is estimated to be 10-20 times that of HCTZ and 100-200 times that of CTZ, resulting in a decrease in the incidences of hypokalemic manifestation (ref. 1). Diuretic toxic effects include bone marrow depression, altered carbohydrate metabolism, and elevated levels of urea, uric acid and sugar.
Furosemide and thiazide diuretics, CTZ, HCTZ, and TCMTZ are approved for use in dairy cattle for the treatment of post-parturient edema of the mammary gland and associated structures (ref. 2). Furosemide is also a common drug of abuse in livestock shows since it reduces tissue water resulting in improved muscle tone. The unauthorized use of these diuretics, or the failure to follow label directions for approved use in cattle, could lead to unacceptable residues in meat or milk destined for human consumption. Therefore, monitoring of the residues of these diuretic drugs in food is part of a general policy to prevent unapproved uses of diuretics. While there are no official tolerances for these drugs in milk, CVM has tentatively established the safe levels for furosemide, CTZ, HCTZ, and TCMTZ to be 10, 67, 67, and 7 ppb, respectively. The withdrawal time for furosemide is 48 hours and 72 hours for CTZ and HCTZ (ref. 2). Although not listed in the Code of Federal Regulations, the withdrawal time for TCMTZ is considered to be 72 hours, similar to other thiazides (ref. 3).
Accurate, precise and sensitive analytical methods for the determination of diuretics, furosemide, HCTZ, CTZ, and TCMTZ, in bovine milk were developed. These methods are specific and distinguish these compounds from each other and from other diuretics, drugs, and antibiotics used in dairy cattle. The diuretics were isolated from milk employing liquid-liquid extraction and cleanup procedures, followed by quantitative determination by high performance liquid chromatography using fluorescence (furosemide) and ultraviolet (UV) absorption detectors (thiazides). The initial procedure to remove the fat from the milk for the four diuretics was the same. It involved centrifugation of an aliquot of milk in a centrifuge set at 4 0C followed by removal of the top fat layer or bottom skim milk. Further cleanup of milk for furosemide was less rigorous than for the other three thiazides. This is because furosemide is naturally fluorescent, while most endogenous interfering compounds in milk do not fluoresce.
For furosemide, the defatted milk was further deproteinated by mixing with acetonitrile and centrifugation. The supernatant was evaporated to remove acetonitrile and the remaining aqueous layer was directly analyzed by HPLC. The HPLC conditions included a reversed phase silicon column, a fluorescence detector set at 272 and 410 nm excitation and emission wavelengths, respectively, and an isocratic mobile phase of 30% acetonitrile in 0.01-M phosphate buffer, pH3 (ref. 4). The average recovery of furosemide at fortification levels of 5-20 ppb was 95% with 9% coefficient of variation (CV), reflecting good accuracy and precision. Furosemide incurred milk samples obtained from a cow administered intravenously with a single dose of 500-mg furosemide was analyzed to validate the overall method. Furosemide concentration in 8 and 24-hour post-dose milk samples were determined to be 150 and 5 ppb, respectively. No furosemide was detected in 32 and 48-hour milk samples.
For the three thiazides, the defatted milk was further treated with 5% lead acetate and acetonitrile. The sample was vortex mixed and centrifuged. The supernatant was removed and ethylacetate was added, mixed and centrifuged to separate the layers. The organic layer was removed and 10% sodium tungstate solution was added, vortex mixed and centrifuged again. The upper organic layer was removed and evaporated to dryness with nitrogen. The resulting residue was reconstituted in mobile phase and an aliquot injected into the HPLC column. The HPLC conditions included reversed phase polymer column, UV detector set at 225 nm, and a mobile phase mixture consisting of acetonitrile (ACN), tetrahydrofuron (THF), and 0.05 M phosphate buffer, pH 3. The thiazides, CTZ and HCTZ are more polar than TCMTZ and required less organic modifier in the mobile phase for their elution from the HPLC column. However, the lower percentage of organic solvent increased the retention of TCMTZ on the HPLC column. Therefore, two isocratic mobile phases, 14% ACN:THF (1:1) in 0.05M phosphate buffer for CTZ and HCTZ; and 30% ACN:THF (2:1) in 0.05 M potassium phosphate buffer, pH3, for TCMTZ were used for their analysis in the milk. Although a gradient mobile phase, consisting of ACN/buffer resolved the three thiazides, the addition of THF was essential to resolve CTZ from the endogenous interfering compounds in milk. However, the three phase mobile phase caused increased baseline shift during the gradient run, therefore, the use of gradient was abandoned. The average recoveries for CTZ and HCTZ at fortification levels of 35-140 ppb were 97 and 89%, respectively, with corresponding average CV’s of 6 and 5% (ref. 5). The average recovery for TCMTZ at fortification levels of 7-140 ppb was 101% with 13% CV (ref. 6). These recoveries and CV’s reflect good accuracy and precision for the overall method. Thiazide incurred milk samples obtained from cows treated with therapeutic doses (ref. 2) of the thiazides were analyzed to validate the overall methods. Both the HCTZ and TCMTZ were detected only in 8-hour post-dose milk samples at concentrations of 47 and 6 ppb, respectively. Chlorothiazide concentrations in 8, 24, and 32-hour post-dose samples were determined to be 430, 86, and 22 ppb.
The development of the above analytical methods allows for the first time the monitoring of residue levels of diuretics, furosemide, CTZ, HCTZ, and TCMTZ in bovine milk. These diuretics are approved for use in dairy cattle. The methods are accurate, precise, sensitive and specific enough to distinguish these compounds from each other, and other drugs and antibiotics used in dairy cattle. Additionally, the methods are simple, fast, and can be adopted easily in most laboratories, including state regulatory laboratories. These methods were applied to detect their concentrations in incurred milk collected from cows treated with therapeutic doses of these diuretics. The preliminary results suggest that these diuretics are rapidly depleted from cows, well before their withdrawal times. However, it must be noted that these data reflect the results from a single cow for each diuretic, and may not represent the rate of depletion from dairy cattle in general. A detailed study to ascertain the withdrawal time of furosemide, using a larger number of cows, is being undertaken to confirm the above preliminary results.
Veterinary Pharmaceuticals and Biologicals, 8th ed., Veterinary Publishing Company, Lenexa, KS, USA, 1993/1994, p. 601.
Code of Federal Regulations , Food and Drugs, parts 520.420, 520.1010b, and 522.1150, U.S. GPO: Washington, DC, 1997.
Talbot, R.B., Fernandez, A.H., Melendez, L.V. (Eds.), List of FDA Approved Animal Products, Laboratory of Veterinary Medical Informatics, Information Series 87-1, VA-MD Regional College of Veterinary Medicine, Blacksburg, VA, 1987, 2.47.
Shaikh, B., Development and validation of a liquid chromatographic method for the determination of furosemide, a diuretic, in bovine milk. J. Agric. Food Chem. 1995, 43, 2117-2121.
Shaikh, B. and Rummel, N. Liquid chromatographic determination of chlorothiazide and hydrochlorothiazide diuretic drugs in bovine milk. J. Agric. Food Chem. 1998, 46, 1039-1043.
Shaikh, B. and Rummel, N. Determination of trichlormethiazide in bovine milk by high performance liquid chromatography. J. Chromatogr. B., 1998, 709, 137-143.
The author was CVM’s 1998 nominee for the FDA Excellence in Analytical Science Award