ANNALS OF CLINICAL AND LABORATORY SCIENCE, Vol. 13, No. 1 Copyright © 1983, Institute for Clinical Science, Inc.

Drug Interference with Measurement of in Urine* BERT SPILKER, M.D., PH.D.,f • j BILLIE S. WATSON, B.S.f and JAMES W. WOODS, M.D.f f Department of Medicine, University of North Carolina School of Medicine Chapel Hill, NC 27514 and \Burroughs Wellcome Co., Research Triangle Park, NC 27709

ABSTRACT The influence of 35 commonly used drugs on measurement of metaneph­ rines in urine was evaluated. Two concentrations of drugs were chosen for study based on usual doses and the percent of dose excreted unchanged in the urine. At “medium” drug concentrations, only falsely elevated levels, whereas at a 10-fold higher drug concentra­ tion, guanethidine, hydrocortisone, imipramine, isoetharine, levodopa, phénobarbital, and phenylephrine caused positive interference. Propranol and caused a negative interference at the two concentrations studied. The significance of these results is discussed.

Introduction sociated with elevated urinary metaneph­ rine levels in patients in whom a pheo­ The diagnosis of a chromocytoma was not demonstrated.8 A often rests on the demonstration of ele­ number of these drugs were found to vated urinary metanephrine excretion. cause interference with the quantitative Although a number of methods have measurement of metanephrine and, there­ been used to measure metanephrine fore, potentially to complicate the assess­ levels, the majority utilizes spectropho- ment of the patient’s correct diagnosis. tometric techniques. These techniques This study did not evaluate any in vivo are subject to interference by drugs effects of the drugs that would directly or which the patient may be taking at the indirectly alter amounts of metanephrine time. excreted in urine. The intent of this study was to evaluate the in vitro effect of a number of com­ Methods monly used drugs on the measurement of Urine collections (24 hours) from three urinary metanephrine levels. Many of normal volunteers were pooled to a vol­ these drugs had been previously as­ ume of 2000 ml and divided into 10 ml aliquots and frozen at —40°C. On the day *Send correspondence and reprint requests to of the experiment, a stock solution of d,l Dr. Spilker at Burroughs Wellcome Co. metanephrine • HC1 was dissolved in 0.01 16 0091-7370/83/0100-0016 $00.60 © Institute for Clinical Science, Inc. METANEPHRINES IN URINE 17 N HC1 and diluted in two aliquots of urine to a final concentration of 5 fig per Estimated Concentration of Various Drugs in Urine ml to be used as controls. The latter con­ Amount Estimated centration was chosen since it is compa­ Representative Excreted Concentration D aily Dose Unchanged in Urine* rable to the elevated levels observed in Drug (mg) (percent) (mg/ml) patients with pheochromocytoma. Acetaminophen 2,000 4 0.04 Drugs to be tested did not contain ex­ Acetylsalicylic 2,000 10 to 80f 0.50 acid cipients and were dissolved in distilled Allopurinol 400 10 to 30 0.04 water unless otherwise noted (table II). Alphamethyldopa 2,000 20 to 70f 0.60 Atropine 0..5 45 0.00011 Aliquots of these drug solutions were Chlorpromazine 400 1 0.002 Clonidine 0..4 45 0.00009 added to other samples of urine that had Digoxin 0..25 75 0.0001 50 4 0.001 been frozen and did not contain any Erythromycin 2,000 15 0.15 added metanephrine. Concentrations of Estradiol 0.,05 20 0.000005 Furosemide 100 90 0.045 drug were expressed as mg per ml in Guanethidine 100 60 0.03 Heparin 30,000 ÍI.Ü.) 20 3.00 (I.U.) urine. Two concentrations of each drug 100 3 0.0015 Hydrocortisone 200 1 0.001 and a control urine sample containing Imipramine 200 4 0.004 Indomethacin 100 20 0.01 metanephrine were assayed in duplicate. Isoetharine 3 25 0.0004 300 30 0.045 Drugs that were insoluble in distilled Isoproterenol 1..5 5 0.00004 water were dissolved in pure ethanol. Levodopa 3,000 5 0.075 Phénobarbital 250 25 0.0313 Additional urine-alcohol controls were Phentolamine 100 10 0.005 Phenylephrine 25 12 0.0015 assayed with the latter. The alcohol con­ Phenytoin 400 2 0.004 Probenecid 1,500 5 0.0375 trol did not interfere with the quantita­ Propranolol 500 1 0.0025 Quinidine 1,000 20 to 50f 0.20 tion of metanephrine. The spectrophoto- Reserpine 0..5 100 (est.) 0.00025 metric method of Pisano9 as modified Rifampicin 600 15 0.045 Spironolactone 200 50 (est.) 0.05 by Crout, Pisano, and Sjoerdsma4 was Sulfisoxazole 4,000 60 1.20 Tetracycline 2,000 60 0.60 used to measure metanephrine or its Theophylline 750 8 0.03 equivalent in all samples. Interference was also determined as described by *Obtained using the formula in the methods section. Crout et al.4 fPrecise value depends on the pH. The approximate concentrations of each parent drug estimated to be present “high” concentrations, respectively. in the urine of patients were calculated Thus, two concentrations of each drug from the following formula: Representative Daily Dose of Drug (mg) X Percent Excreted Unchanged in Urine Average 24 Hour Volume of Urine (assumed to be 2000 ml) Percentages of most parent drugs ex­ creted unchanged in urine were obtained were studied representing the upper from the Handbook of Clinical Phar­ range of urine concentrations expected macology3 and, when possible, were to be found in patients receiving that confirmed in Disposition of Toxic Drugs drug. The resulting data are presented and Chemicals in Man.1 Urine concen­ in table I. trations of most drugs studied were de­ termined as described previously using a Results and Discussion representative daily dose of each drug. Urine concentrations 10-fold higher and Results of our experiments are pre­ 100-fold higher were chosen for study. sented in table II, which demonstrate These were defined as “medium” and that at the “medium” concentrations 18 SPILKER, WATSON, AND WOODS T A B L E II interference with measurements of meta- Effect of Various Drugs on Measurement nephrines. of Metanephrine in Urine These data indicate that 20 percent of "Medium" Con­ Equivalent of Meta­ the drugs studied may potentially inter­ centration of nephrine Obtained (mq) * Drug Used Medium Con- ffigh Con­ fere with metanephrine measurements in Drug (mg/ml) centra tion f c e n tr a tio n f urine. Drugs found by other workers to

Acetaminophen 0.4 0.37 0.84 interfere with unspecified Acetylsalicylic 5.0 0.85 1.11 metabolites were , , acid Allopurinol 0.4 0.34 0.57 trihexiphenidyl, and methocarbamol.5 Alphamethyldopa 6.0 0.048 1.02 Atropine 0.0011 0.00 0.00 Metabolites of these drugs were not Chlorpromazine 0.02 0.12 1.72 considered in the present study owing to Clonidine 0.0009 0.08 1.03 Digoxin 0.001 0.16 0.24 the difficulties involved in obtaining and Diphenhydramine 0.01 0.08 0.06 Erythromycin 1.5 0.006 0.022 combining each with urine in the proper Estradiol$ 0.00005 0.18 0.49 Furosemide 0.45 0.06 0.91 proportions. There is little doubt that Guanethidine 0.3 1.52 9.80 some metabolites would have additive or Heparin 30. I.U. 0.60 1.28 Hydralazine 0.015 0.08 0.08 synergistic effects with the parent drug or Hydrocortisone 0.01 1.63 9.02 Imipramine 0.04 0.78 2.44 other metabolites in interfering with this Indomethacin$ 0.1 0.22 0.69 Isoetharine 0.004 0.24 2.72 test. Isoniazid 0.45 0.00 0.012 Another factor that must be considered Isoproterenol 0.0004 0.72 1.24 Levodopa 0.75 0.27 4.0 in interpreting these results is that of Phénobarbital 0.31 0.99 2.63 Phentolamine 0.05 0.00 0.12 metabolic conjugation. Although the par­ Phenylephrine 0.015 2.5 30.§ ent drugs in these experiments were un­ Phenytoin 0.04 0.01 0.54 Probenecidt 0.375 0.22 0.99 conjugated, glucuronide or sulfate conju­ Propranolol 0.025 0.00 0.00 Quinidine 2.0 0.38 1.20 gation occurs in vivo and could therefore Reserpine 0.0025 0.08 0.68 Rifampicin 0.45 0.00 0.10 lead to different results. Other in vivo Spironolactone 0.5 0.47 1.13 factors that may affect metanephrine Sulfisoxazole 12.0 0.00 0.14 Tetracycline 6.0 0.00 0.07 levels include drug effects on metabo­ Theophylline 0.3 0.00 0.00 lism and storage of , the

♦Baseline urine controls ± S.D. without added meta- metabolic fate of the parent drug, and nephrine were 0.18 ± 0.03 mg. Individual control drug interactions. readings for each drug were subtracted before the equivalent of metanephrine was determined. Recovery In order to avoid the problems men­ of added metanephrine in the standard was 77.0 ± 8.7 percent (range was from 65 percent to 90 percent). tioned, theoretically urine could be The mg calculated from 10 mi aliquots are comparable to the number of mg of metanephrine measured in 24 tested from patients free of pheochro- hour urine samples from patients. mocytomas who were taking only one fThese values are the mean of duplicate values minus control. High concentration used is 10 times the drug. This has the advantage of including medium concentration. ^Alcohol vehicle used to dissolve drug did not cause metabolites and conjugation factors de­ interference. scribed as well as possessing direct clini­ §Samples became purple after boiling. cal relevance. It has the disadvantage, however, that doses of drugs which pa­ studied, only one of the 35 drugs (phen­ tients take vary considerably and that a ylephrine) caused a false positive result suitable population taking one drug (i.e., falsely elevated metanephrine would be difficult to locate and study. levels above two mg). At the high con­ There was no correlation between the centration seven drugs caused interfer­ interference observed and the mole ence equivalent to two mg or more of amounts (mg of drug used per molecular metanephrine (table II).Two drugs, pro­ weight) of chemicals used. Large doses of pranolol and theophylline, decreased guanethidine and reserpine have been readings below control at the doses reported7 to decrease metanephrines via studied and, therefore, caused negative in vivo actions. However, in this labora­ METANEPHRINES IN URINE 19 tory investigation, guanethidine marked­ 2. B l u m b e r g , A. G., H e a t o n , A. M., and V a s- s il ia d e s , J.: The interference of chlorprom­ ly increased the level of metanephrine azine metabolites in the analysis of urinary equivalent in urine and reserpine in­ methoxy-catecholamines. Clin. Chem. 12:803— creased it by a small amount. The inter­ 807, 1966. ference observed with chlorpromazine 3. B o c h n e r , F. et al.: Handbook of Clinical Phar­ and imipramine has been previously macology. Boston, Little Brown and Company, reported,2,6 as was that observed with 1978. 4. C r o u t , J. R., P isa n o , J. J., and Sjo e r d sm a , A.: 1-dopa.5 The present study indicates Urinary excretion of catecholamines and their that numerous other drugs may also cause metabolites in pheochromocytoma. Amer. marked interference with urinary meta­ Heart J. 61:375-381, 1961. 5. G it l o w , S. E., M e n d l o w it z , M ., and B e r t a n i, nephrine measurements. L. M .: The biochemical techniques for detect­ It is apparent that urine collection for ing and establishing the presence of a pheo­ metanephrines in patients suspected of chromocytoma. Amer. J. Cardiol. 26:270—279, having a pheochromocytoma should be 1970. 6. H a n s t e n , P. D.: Drug interactions, 3rd ed. carried out in the absence of drugs shown Philadelphia, Lea and Febiger, 1976, pp. to interfere with the method. Since only a 326-331. small number of drugs in clinical usage 7. M a n g e r , W . M ., and G if f o r d , R. W., J r .: have been studied, all medications should Pheochromocytoma. New York, Springer- ideally be withdrawn prior to the col­ Verlag, 1977, pp. 201-222. lection. 8. P e t e r so n , D. D., W o o d s , J. W ., F r ie d , F . H., B r o w n , D. R., and C o l e , A. T.: Pheochro­ mocytoma: Recent experiences with detection References and management. Urology 10:133—138, 1977. 1. B a se lt, R. C.: Disposition of Toxic Drags and 9. P isa n o , J. J.: A simple analysis for normeta- Chemicals in Man. Vol. 2. Biomedical Publica­ nephrine and metanephrine in urine. Clin. tions, Canton, CT, 1978. Chim. Acta 5:406-414, 1960.