Interterence by the 4-Hydroxylated Metabolite of Propranolol with ...

2 downloads 0 Views 457KB Size Report
Interterence by the 4-Hydroxylated Metabolite of Propranolol with. Determination of Metanephrines by the Pisano Method. David Chou,' Masanobu Tsuru,2 ...
CLIN. CHEM. 26/6, 776-777 (1980)

Interterence by the 4-Hydroxylated Metabolite of Propranolol with Determination of Metanephrines by the Pisano Method David Chou,’ Masanobu Tsuru,2 Jordan L. Holtzman,2 and John H. Eckfeldt”3 Measurements of urinary metanephrines by the Pisano procedure (C/in. Chim. Acta 5: 406, 1960) are unreliable in patients who are taking propranolol, owing to the presence of 4-hydroxypropranolol in the urine. Three properties of this propranolol metabolite lead to interference: (a) it is absorbed and eluted from ion-exchange resins under the conditions Pisano used for metanephrine isolation, (b) at high pH it absorbs at 350 to 360 nm, and (c) it is oxidized by periodate to a substance with negligible absorption in this region. Because 350 to 360 nm is the wavelength used to quantitate vanillin, the product formed from periodate oxidation of metanephrines, and because the unoxidized eluate is used as a specimen blank, the presence of 4-hydroxypropanolol spuriously decreases the

measured

urinary

metanephrines;

this

has special

significance because patients being tested for increased metanephrines are also likely to be receiving propranolol for hypertension. The Pisano procedure for urinary metanephrines (1) has been widely used as a simple and highly effective method of screening patients for increased catecholamine production. Recently we noticed that many urines being tested for metanephrines had higher blank absorbances than after periodate oxidization. We found that the urines from patients showing this phenomenon had all been receiving the antihypertensive/antiarrhythmic drug propranolol (Inderal#{174}; Ayerst, New York, NY 10017) in doses of 160 to 320 mg/day. Consequently, we examined the effects of propranolol and its principal metabolite (2, 3), 4-hydroxypropranolol, on the Pisano method for urinary metanephrines. Our results indicate that the interference is ascribable to the presence of 4-hydroxypropranolol in the urine.

Materials and Methods Propranolol and 4-hydroxypropranolol hydrochloride were obtained through the courtesy of Ayerst Research Laboratories, Montreal, P.Q., Canada H3C-3J1. The chromatographic columns and normetanephrine standards used for the assay of metanephrines were obtained from Bio-Rad Labs., Richmond, CA 94804. All other chemicals used were of reagent grade. The spectra of propranolol and 4-hydroxypropranolol were determined with an Aminco DW-2 (American Instruments, Silver Spring, MD 20910) and a Cary 219 (Varian Instruments, Palo Alto, CA 94303) scanning spectrophotometer using quartz-halogen lamps. The urinary metanephrine analysis was done according to the procedure provided with the Bio-Rad columns (4). After initial filtration and acid hydrolysis at pH 0.5 to 0.9, pass 5 mL

Veterans

Administration

Medical

Center,

Minneapolis,

and the

Departments of’ Laboratory Medicine and Pathology, 2 Medicine and Pharmacology, University of Minnesota, Minneapolis, MN 55455. ‘ Direct reprint

requests to Dr. Eckfeldt, Laboratory

Service (113A),

VA Medical Center, 54th Ave. and 48th St. S., Minneapolis, 55417. Received Jan. 31, 1980; accepted Feb. 22, 1980. 776

CLINICAL CHEMISTRY. Vol. 26, No. 6, 1980

MN

of the urine adjusted to pH 6.5 through the column and wash with boric acid, 40 g/L, then with ammonium hydroxide, 4 mol/L. To a 4-mL aliquot of the animonium hydroxide eluate, add 100 1zL of sodium metaperiodate, 2 g/L solution; 2 mm later add 100 ML sodium metabisulfite (10 g/L solution) to stop the reaction and decrease the absorbance of the periodate. Prepare a blank by taking a second 4-mL aliquot of the eluate and adding 100 ML of the sodium metabisulfite, followed by 100 ML of the sodium periodate 2 mm later. The blank preparatioi’l in the original Pisano procedure involved 100 ML of water in place of the sodium periodate, but results by these two procedures do not differ notably in our hands. Measure the blank and sample absorbances at 350 nm with a spectrophotometer (we used model 300N; Gilford Instruments, Oberlin, OH 44074) and compare results with values on a normetanephrine standard curve.

Results Under both acidic and basic conditions, 4-hydroxypropranolol-but not propranolol-absorbed substantially between 310 and 370 nm. The basic and acidic spectra were rapidly reversible by adding a strong acid or base, suggesting that the differences were due to ionization of the phenolic group. Acidic solutions of 4-hydroxypropranolol could be stored at room temperature for as long as four weeks with little degradation, which is consistent with the findings of others (5). Solutions at pH >10.5 lost about half of their absorbance in 4 h, while solutions at pH between 9 and 10 lost absorbances within minutes. Addition of periodate rapidly oxidized 4-hydroxypropranolol in basic solutions with the disappearance of the 343-nm maximum. When sodium metabisulfite was added first, the sodium periodate had essentially no effect. Attempts to elute the 4-hydroxypropranolol from the columns with boric acid (670 mmol/L, pH 3.8) or sodium borate (250 mmol/L, pH 8.5) were unsuccessful. When ammonium hydroxide (4 mol/L, pH 10.8) was used, however, both the 4-hydroxypropranolol and metanephrines were eluted. Addition of propranolol to urines showed no effect on the results for metanephrine. However, when 4-hydroxypropranolol hydrochloride, 20 mg/L, was added to urines of patients known not be taking propranolol and previously analyzed for metanephrines without difficulty, all showed the characteristic reversal previously described. Addition of 5, 10, and 20 mg of 4-hydroxypropranolol hydrochoride per liter of urine from a patient with pheochromocytoma progressively increased the absorbance of the blank and sample while decreasing the difference between the blank and sample. This led to the progressively decreasing apparent metanephrine concentrations: from 2.4 mgfL (4.3 mg/24 h) before 4-hydroxypropranolol addition, to 2.1 mg/L (3.7 mg/24 h), 1.5 mg/. (3.1 mg/24 h), and finally 0.1 mg/L (0.3 mg/24 h), respectively.

Discussion Propranolol was originally approved in 1964 for use as an antiarrhythmic drug. In the early to mid-1970’s its use expanded to include therapy for hypertension, at doses two- to six-fold those used for arrhythmias. Because the incidence of

many respects, as evidenced by its ion-exchange

side effects has been lower than with any other antihyper-

tensive medications, propranolol has become a second-line drug in patients refractory to diuretics alone. At the lower doses administered for arrhythmias, propranolol was not recognized to affect the Pisano procedure (6). The effect we describe appears to be related to the higher doses administered to hypertensives. This same population is frequently screened for pheochromocytomas. Propranolol metabolism has been extensively studied (5, 7-9). More than 90% of the drug is excreted in the urine as the glucuronides of propranolol and its metabolities. In two studies (5, 9), the concentration of 4-hydroxypropranolol in serum was found to be between 5 and 25% that of propranolol. Although no figures are given in the literature, we estimate that 5 to 25% of the daily dose of propranolol appears in the urine as 4-hydroxypropranolol glucuronide. Interference problems can be partly identified by careful observations of the resulting spectra. Crout et al. (10) suggested that absorbance at 333 and 347 nm should be measured for all samples with total metanephrmnes >1 mg/day, to check for substances causing positive interference. Johnson et al. (11) suggested making a reading at 300 nm in the case of samples suspected of having interference from certain radiographic dyes containing methylglucamine. Because in-

creased absorbances in both the blank and sample fractions accompany high concentrations of 4-hydroxypropranolol, one useful method of detecting this interference is to read the nonoxidized blank eluate against a water reference at 333,347,

and 360 nm in samples with high absorbances. The presence of a peak at 347 nm in the metanephrine blank strongly suggests 4-hydroxypropranolol interference. To circumvent this problem of interference, a 24-h urine should be collected before the patient starts to take the drug. We have begun this practice since discovering this interference. Alternatively, for patients already on propranolol, it may be possible to withdraw them from the drug before performing the test; both propranolol and 4-hydroxypropranolol appear to have relatively short biological half-lives in the serum, and three to five days without the drug should be sufficient to eliminate interference. Be cautious in withdrawing patients,

however, because a rebound hypersensitivity of receptors to catecholamines may result in myocardial infarcts (12), although this phenomenon has not been well documented (13). Finally, metanephrmnes may be assayed by a more specific method such as liquid chromatography (14). Compared with tests for urinary catecholamines, serum catecholamines, and vanillylmandelic acid (4-hydroxy-3methoxymandelic acid), the Pisano test appears to be still the most reliable and widely used screening test for above-normal

catecholamine production. Catecholamines are subject to large physiological variations, whereas tests for vanillylmandelic acid suffer from interference problems. 4-Hydroxypropranolol appears to mimic metanephrine and normetanephrine

in

resin binding and elution characteristics, acid stability, and relative base lability. Unfortunately, it absorbs in the 350- to 360-nm region used to quantitate the vanillin formed from metanephrines, and it is oxidized by the periodate to a product without much absorbance in this region. As seen in the case of the sample from the patient with pheochromocytoma, this interference is sufficient to result in falsely normal values in a pa±ient with highly increased metanephrines. The negative interference caused by 4-hydroxypropranolol suggests that urinary metanephrines determined by the Pisano procedure should be interpreted with caution in patients receiving propranolol.

References 1. Pisano, J. J., A simple analysis for normetanephrine and metenephrine in urine. Clin. Chim. Acta 5, 406-414 (1960). 2. Bond, P. A., Metabolism of propranolol (‘Inderal’), a potent, specific beta-adrenergic receptor blocking agent. Nature 213, 721 (1967).

3. Patterson, J. W., Conolly, M. E., Dollery, C. T., et al., The pharmacodynamics and metabolism of propranolol in man. Eur. J. Clin. Pharmacol. 2, 127-133 (1970). 4. The determination of metanephrines. In Catecholamines by Column Test Instruction Manual, Bio-Rad Labs., Richmond, CA

94804,1977,pp 17-22. 5. Rao, P. S., Quesada, L. C., and Mueller, H. S., A simple micromethod for simultaneous determination of plasma propranolol and 4-hydroxypropranolol. Clin. Chim. Acta 88, 355-361 (1978). 6. Young, D. S., Pestaner, L. C., and Gibberman, V., Effects of Drugs on Clinical Laboratory Tests. Clin. Chem. 21, 328D (1975). Special issue. 7. Walle, T., and Gaffney, T., Propranolol metabolism in man and dog: Mass spectrometric identification of six new metabolites. J. Pharmacol. Exp. Ther. 182,83-92 (1972). 8. Walle, T., Fagan, T. C., Conradi, E. C., et al., Presystemic and systemic glucuronidation of propranolol. Clin. Pharmacol. Ther. 26, 167-172 (1979). 9. Wong, L., Nation, R., Chiou, W. L., and Mehta, P. K., Plasma concentrations of propranolol and 4-hydroxypropranolol during chronic propranolol

therapy.

Br.

J. Clin. Pharmacol. 8,

163-167

(1979).

10. Crout, J. R., Pisano, J. J., and Sjoerdsma, A., Urinary excretion of cateeholamines and their metabolites in pheochromocytoma. Am. Heart J. 61,375-381(1961). 11. Johnson, L. R., Reese, M., and Nelson, D. H., Interference in Pisano’s urinary metanephrine assay after use of x-ray contrast media. Clin. Chem. 18,209-211(1972). 12. Alderman, E. L., Coltart, J., Weittach, G. E., and Harrison, D.C., Coronary artery syndrome after sudden propranolol withdrawal. Ann. Intern. Med.. 81, 625-627 (1974). 13. Shiroff, R. A., Mathis, J., Zelis, R., et al., Propranolol rebound-a retrospective study. Am. J. Cardiol. 41, 778-780 (1978). 14. Shoup, R. E., and Kissinger, P. T., Determination of urinary normetanephrine, metanephrine, and 3-methoxytyramine by liquid chromatography with amperometric detection. Clin. Chem. 23, 1268-1274 (1977).

CLINICALCHEMISTRY,Vol. 26, No. 6, 1980 777