THE STIMULATION OF ALBUMIN SYNTHESIS BY METHADONE

Transcription

1 GASTROENTEROLOGY 71: by The Williams & Wilkins Co. Vol. 71. No.2 Printed in U.S.A. THE STIMULATION OF ALBUMIN SYNTHESIS BY METHADONE MARCUS A. ROTHSCHILD, M.D., MARY JEANNE KREEK, M.D., MURRAY ORATZ, PH.D., SIDNEY S. SCHREIBER, M.D., AND JOSEPH G. MONGELLI, M.S. Radioisotope Service, New York Veterans Admin~tration Hospital, The Rockefeller University, Departm~nt of Medicine, New York University School of Medicine, and Department of Biochemistry, New York University Col/ege of Dent~try. New York, New York Elevated levels of serum albumin have been noted in patients on chronic methadone maintenance and in heroin addicts. This observation was investigated in rabbits maintained on daily methadone 4 mg per kg of body weight after a period of 3 months on increasing dosage to assure drug tolerance. Albumin distribution and metabolism were measured with tested lots of U51 rabbit albumin. Studies were made before and again after the attainment of the methadone maintenance state. Albumin distribution was altered markedly with a shift of intravascular albumin to extravascular sites. Associated with this change, the serum albumin level rose by an average of 0.5 g per 100 ml. Albumin degradation increased by 32% from 248 to 327 mg per kg per day. The total exchangeable albumin pool increased 35%, or 3.6 g. Since the exchangeable albumin pool increased in the face of an increment in albumin degradation, albumin synthesis must have increased even further to account for this change. Although the specific factors responsible for these alterations in albumin metabolism and distribution are not known at present, to date, this hyperalbuminemic hypercatabolic state is not produceable in any other clinical or experimental situation. Clinically, hyperalbuminemia is usually not seen, except perhaps in acute dehydration, or after infusions of serum or albumin. These changes in the serum albumin level are rapidly returned toward normal levels by means of hydration or by increasing the rates of albumin degradation.' Recently, however, it was observed that the serum albumin level was increased above the normal mean in a group of patients on chronic methadone maintenance. I. I This observation was also made in heroin addicts at time of admission to methadone treatment, when the patients were on a relatively poor nutritional intake, a situation most frequently related to a decrease in the level of serum albumin.' Two-thirds of these patients, both before and during methadone maintenance, had evidence of chronic liver disease, usually chronic persistent hepatitis. I. I The possible changes in albumin metabolism due to these drugs were investigated in 8 rabbits maintained on methodone for a period of 2 to 3 months, and the results indicate a marked increment in exchangeable albumin, a significant shift in distribution of albumin from plasma to extravascular sites, and augmented albumin synthesis. Received December I, Accepted February 20, Presented in part at the annual meeting of the American Associa tion for the Study of Liver Diseases, Chicago. lllinois. November This work was supported in part by grants from the United States Public Health Service AA HL 09562; the Bear Foundation, and by the New York State Office of Drug Abuse Services (ODAS-84854). The conclusions reached in this report are not necessarily those of that office. Dr. Kreek is a recipient of a Career Scientist Award of the Health Research Council of the City of New York. Methodology Male rabbits weighing 3.3 to 4.1 kg were used in all studies. The rabbits were kept in metabolism cages and complete urine and stool collections were made daily. All rabbits were fed standard rabbit ration containing 17% protein, 2% fat, and 15% crude fiber (Wayne Rabbit Ration obtained from Allied Mills, Inc., Chicago, 111.). Two to 3 drops of Lugol's solution were administered daily in the rabbits' drinking water to inhibit the thyroidal uptake of lui released from degraded labeled protein. The distribution and metabolism of albumin were measured by means of rabbit serum albumin labeled with The albumin used in these studies was isolated by preparative acrylamide gel electrophoresis. The pooled albumin fraction was examined by qualitative polyacrylamide electrophoresis and immunoelectrophoresis at a 6% protein level. 1 Albumin was labeled at approximately 1 atom of iodine per protein molecule by the ICI method. 1 The lots of iodinated albumin were tested in control rabbits to guard against the use of a preparation that contained a significant fraction of denatured material. '-1 Between 75 and 150 I'C of "'I-albumin was injected into the ear vein. Heparin-treated blood samples were obtained from the contralateral ear 6 and 10 min after the injection and at daily intervals thereafter. Control observations were made daily on the excretion of radioactivity in the urine and stool for 11 to 15 days. At the end of the control period, methadone hydrochloride, dissolved in saline, was injected intramuscularly daily at increasing dosages starting at 0.5 mg (methadone-free base) per kg of body weight. The period of time at each step varied from 3 to 6 days in each rabbit, depending on the sensitivity of the rabbit to the higher dose. These injections were continued until the rabbits were receiving 4 mg (methadone-free base) per kg of body weight per day, and this level was then maintained. Although tolerance continued to develop during the period of increasing doses of methadone, at times, the animals, would exhibit severe lethargy, and the food intake over the 1st month 214

2 August 1976 METHADONE AND ALBUMIN METABOUSM 215 of treatment averaged less than 50% of the control nitrogen intake. On achieving the maintenance dose of 4 mg per kg of body weight per day, food intake approximated the control values. After at least 3 weeks of methadone maintenance at 4 mg per kg per day, a second injection of lui-albumin was made, and albumin metabolism and distribution were remeasured during another 11 to 15 days while methadone was continued. The calculations are shown in (table O. Plasma volume was determined from the mean space of distribution of uoi-albumin at 6 and 10 min after injection. Total exchangeable albumin was determined from the product of the space of distribution of the injected ""I-albumin at distribution equilibrium and the plasma albumin concentration. The space of distribution of "'I-albumin was determined by dividing the amount of ""I-albumin remaining in the body (per cent dose) by the plasma level of lui-albumin (per cent dose per milliliter). At any time (T) the dose of '''I-albumin remaining in the body is the U6I-albumin administered less that excreted in the urine and stool in time (T). Because under similar conditions, as used in these studies, we have consistently accounted for 95% or more of the administered dose in the urine and stool, other significant routes of excretion can be neglected." t. " The total exchangeable albumin is determined by assuming that the specific activity is uniform throughout the body after distribution. It has been shown that this assumption is in error by an insignificant fraction. unless the rate of loss from the body is extremely excessive.' The fraction of the total exchangeable albumin located intravascularjy was determined from the values for total plasma albumin and total exchangeable albumin. The rate of albumin degradation during the control and experimental periods was determined from the product of the renal clearance (microcuries excreted per day divided by mean micro-. curies of "'I-albumin per milliliter of plasma) and the plasma albumin concentration. Fecal excretion of "'I during the periods of the study did not amount to more than 4 to 5% of the administered dose. This amount was included in the values for urinary excretion. This procedure, the metabolic clearance method, has the advantage of being valid even in nonsteady states and has been discussed in the past.' In order to obtain valid estimates for degradation of albumin using iodine-labeled albumin, it is necessary to use a material as near to native albumin as possible. If the labeled albumin is denatured, then the damaged protein will be degraded rapidly and the iodide so released will also be cleared rapidly from the body. Thus the clearances obtained with denatured albumin after the 1st day after injection, would be high when compared to clearances after 3 or 4 to 9 or 10 days after the injection of TABLE 1 I. Plasma volume = Space of distribution of '''I-albumin 6-10 min after injection 2. Total exchangeable albumin space (TEAS) - Space of distribution of "'I-albumin at distribution equilibrium 3. Total exchangeable albumin (TEA) = TEAS x [SA]" 4. Total intravascular albumin = Plasma volume x [SA] 5. Extravascular albumin = TEA - intravascular albumin 6. Renal clearance of plasma "'I = Urinary excretion of "'I (/,c/day) Mean plasma '''I-albumin (/,c/day) 7. Metabolic clearance (g/day) = Renal clearance (ml/day) x serum albumin (g/m\) " Serum albumin concentration (grams per milliliter). the labeled albumin. In the present studies, the clearances on days 2 and 3 averaged 98 ± 2% and 95 ± 4% of the clearance rates for days 4 to 10 for the control and experimental periods, respectively. These data indicate the relative un denatured state of the iodine-labeled albumin used. The clearance rate on the 1st day was between one-third and one-half of the clearance rates obtained on subsequent days, owing to the fact that the urine was collected after only 18 to 20 hr initially, and that the iodide released from degraded albumin must mix in the iodide pool before excretion. 11 After methadone administration some 45 to 60 days after the start of the daily methadone injections, another injection of UtI-albumin was made to remeasure albumin distribution and the total exchangeable albumin pool. Thus, albumin metabolism was first studied during the steady state control period, and again after the attainment of a new steady state condition after methadone administration. Plasma, urine, and stool samples were assayed for " 1 in a wen type scintillation counter with a sensitivity of 2 x 10" counts per min per IiC of " 1 above a background of 100 counts per min. Total plasma protein was determined by a biuret procedure,l2 and the concentration of plasma albumin and protein partition was determined by a boundary electrophoresis employing a Kern microelectrophoretic unit as previously described.' Some of the plasma samples, both control and experimental, were subjected to immunological analysis for the determination of serum albumin according to the method of Mancini et al.13 Plasma levels of methadone were determined by gas-liquid chromatography. "-n For these determinations, blood samples were drawn at 1 and 24 hr after the parenteral dose of methadone was stabilized at 4 mg per kg for 2 weeks. The blood was drawn into oxalated tubes, promptly centrifuged, separated, and either analyzed immediately or quick frozen and stored in a freezer. Methadone will break down into various degradation products, especially nitrogen oxide, if it is left in solution at room temperature for extended time periods." ['H \Methadone (10}.l1) (DL-, 0., or L-methadone with a specific activity of 92 me per mm) containing 1500 to 1800 counts per min, were added to 2 ml of plasma. The plasma was blended on a Vortex mixer and allowed to equilibrate for 1 hr at room temperature. Then 2 ml of 0.1 N ammonium hydroxide and 8 ml of ethyl acetate were added while continually shaking the mixture, and then it was centrifuged. The upper phase (ethyl acetate) was transferred to a tapered conical tube and taken to dryness with a stream of nitrogen at room temperature. The residue was dissolved in 1 ml of ethyl acetate containing 0.121ig of docosane (as an internal standard). A 500-1i1 aliquot was transferred to a second conical tube. Then, loo-1i1 aliquots, in triplicate, of the remaining solution were transferred to counting vials for liquid scintillation counting, to calculate recovery through the extraction procedure. At this time "'I from the initial isotope injection was negligible. The 500-1i1 aliquot was taken to dryness with a stream of nitrogen at room temperature and redissolved in 10 iii of carbon disulfide. Two microliters (equivalent to 0.2 ml of plasma) were injected onto the gas chromatography column. Standard calibration curves were constructed and are linear within the range of 0.02 to 2.0 IJg per ml. Gas-liquid chromatography was performed with a Hewlett Packard research chromotograph model 7620-A or 5700-A. A 6-foot column, with an internal diameter of 1;. inch. packed with 10% methyl vinyl silicone gum rubber coating (VCC WW980) on 80 to 100 mesh high performance Chromosorb W was used. The injection port temperature was 220 C, column temperature was 210 C, and flame detector was 225 C. Nitrogen carrier gas was adjusted to a flow of 25 ml per min, and

3 216 ROTHSCHILD ET AL. Vol. 71, No.2 hydrogen and air were adjusted to give maximal sensitivity for flame detection (replicate analyses ±3.6%). Results Peak plasma levels of methadone at 1 hr after the single daily parenteral 4-mg per kg dose, ranged from 0.08 to O.38lJg per ml with a mean of ± These values are less than those observed in patients on the usual dose of methadone, 80 to 100 mg per day or 1.1 to 1.3 mg per kg per day, U-li suggesting a more rapid rate of methadone metabolism in rabbits. Methadone levels at 24 hr after this dose were below the limits of detection of the method used (less than O.OllJg per ml). Data pertaining to albumin distribution during the control and experimental periods are shown in table 2 and figure 1. It can be seen. that the weight of the animals changed only slightly during the methadone period, and likewise that there was no significant alteration in hematocrit value or in plasma volume, although it should be noted that the plasma volume decreased by about 10%, and that this decrease occurred in 7 of 8 rabbits. However, the exchangeable albumin space rose by 17% (P < 0.01), and the fraction of the total exchangeable albumin located within the plasma volume decreased from 36.5 to 28.2% (P < 0.01), indicating a shift in intravascular albumin to extravascular sites. Animal TABLE 2. Effects of methadone on albumin distribution" Weight Hematocrit Plasma volume Total exchangeable yalue albumin Ipace C E C E C E C E kg % ml ml Mean ±SEM P< 0.01 C Intravaocular albumin % E P < 0.01 C, control; E, experimental (methadone); number of studies, 8.! 1 D.i. 100 ~ 10 ~ 60 L i! 40 -.:: ;/ j '" j '" '" 20 D CONTROL TEAS' [SA] 0 343ml 32gl100ml '::::- : Q.-..,--Q.-oo--O'- 'Q o METHAOONE TEAS I [SA) ml ' ~ """,... A.. CONTROL 'II... METHADONE! i. L '--'"... '--L--'--.L.-..L IS Ja " o,,51z I 11 hlglkgldo1 FIG. 1. Rabbit 5. Solid symbols represent control study March 11 to March 24. Total exchangeable albumin space (TEAS) e; plasma disappearance curve, ~. [SAl, serum albumin concentration. Open symbols represent the study performed from May 27 to June 9, while the rabhit was maintained on methadone, 4 mg per kg per day. 0, TEAS; 11, plasma disappearance curve. The TEAS on methadone maintenance rose from 343 to 450 mi. The product of this expanded space and the elevated serum albumin level resulted in an increase in the exchangeable albumin pool to However, the fractional rate of loss determined solely from the plasma disappearance curves did not change. This figure again illustrates that if the fractional rate of loss from the plasma is used as the sole index of the effects of a specific perturbation, marked changes may be minimized or missed altogether. 11.0g TEA 171g TEl

4 August 1976 METHADONE AND ALBUMIN METABOUSM 217 The effects of methadone on albumin metabolism are shown in table 3 and figure 2. Total protein increased from 6.3 to 6.8 g per 100 ml, and the mean plasma albumin concentration increased by nearly 0.5 g from 3.26 ± 0.07 to 3.73 ± 0.09 g per 100 ml.* An even greater increment of 38% took place in the exchangeable al bumin pool, due to the fact that the exchangeable albumin space increased. as well as the plasma albumin level. Total intravascular albumin, however, remained essentially unchanged after methadone administration, and the increment in the exchangeable albumin pool was solely due to the increment in exchangeable albumin located in extravascular sites. Albumin degradation also increased some 22 to 32 %, expressed either as mg per day or mg per kg per day, respectively. Assuming that steady state conditions existed with respect to albumin metabolism, after the animals reached the methadone maintenance level of 4 mg per kg per day and had been kept at this level for at least 3 weeks, then these values for degradation, together with the increase in total exchangeable albumin located intravascularly, would indicate an increment in albumin synthesis as well. The fractional rate of albumin degradation (albumin degraded per day per total exchangeable albumin) did not show any significant increment, 8.6 ± 0.3% in the control period, to 7.9 ± 0.3% after methadone administration. Alternative methods of expressing the fractional rate of albumin degradation are based on using the plasma albumin mass as the denominator. When these techniques are used (table 4), the fractional rate of albumin degradation (albumin degraded per day divided by the plasma albumin mass) showed an increment from 23.6 ± 1.1 to 28.1 ± 1.3% of the plasma albumin per day (clearance technique), or from 24.3 ± 1/3 to 29.7 ± 1.7 of plasma albumin per day by the equilibrium time method. Proteinuria was not observed in any animal. Using plasma albumin mass as the basis, the fractional catabolic rate was determined by two methods. First, the absolute amount of albumin degraded per day was divided by the plasma albumin mass. This technique used the clearance method to obtain the absolute amount of albumin degraded per day. A second technique employs the relationship between the intravascular and extravascular masses, as well as the fractional rate of disappearance of radioactive albumin from the plasma pool Because the basic data are the same in both methods, only minor variations in values would be expected when the data are subjected to both techniques. In the clearance procedure, the assumptions are that iodide excretion and albumin distribution are rapid in comparison to degradation, and these conditions are present unless massive albumin loss is present. In the equilibrium time method it is necessary to assume further, that "the pool masses, and rates of exchange Serial immunological measurements of albumin levels were made in rabbits and 7. During methadone maintenance the plasma albumin concentration was higher by 0.6 g in rabbit 5, 0.7 g in rabbit 6. and 0.95 in rabbit 7. These increments in plasma albumin concentration were nearly identical to those observed by the electrophoretic technique <table 3). and catabolism are constant and that newly synthesized protein is discharged into and catabolism occurs in the extravascular pool or in a pool which is in rapid exchange with it."" Because the two methods agree, the validity of these various assumptions can be assumed to be correct. Discussion The validity of the use of iodinated albumin to trace albumin degradation and albumin distribution has been discussed repeatedly. As long as adequate controls are maintained to ensure that the iodinated albumin does not contain excessive amounts of material which are degraded rapidly. and the studies are conducted for a long enough period of time to assure adequate distribution equilibrium, then valid estimates for albumin degradation and albumin distribution are obtained. During the control periods in both groups, albumin degradation averaged 248 mg per kg per day, values which agree with those that have been reported previously There are many methods available for the calculation of albumin metabolism. The use of the metabolic clearance technique has the advantage of being valid in nonsteady states. The amount of radioiodine excreted in the urine is related to the mean plasma concentration of 125I albumin for that day. Because the rate of distribution of the radioiodinated albumin is rapid in comparison to the rate of degradation, only minimal errors are incurred, owing to the small difference in specific activity from intravascular to extravascular albumin. Further, inasmuch as both the control and experimental studies were performed at a time when the serum albumin levels remained constant, or had attained new staple levels and the weights of the animals were not changing, it may be assumed that steady state conditions with respect to albumin metabolism were at least approximated. Thus, the values attained for albumin degradation are probably very close to those for albumin synthesis. t. t. I. There were marked changes in albumin distribution after methadone administration. Although the plasma volume decreased only slightly, the total exchangeable albumin space increased by over 16%. These changes in the opposite direction resulted in a marked shift in albumin distribution from plasma to extraplasma sites. In fact, the increment in total exchangeable albumin pool was solely the result ofthe increase in the exchangeable albumin that was located extravascularly. There was also marked changes in albumin metabolism with albumin degradation increasing in seven of the eight studies. The absolute amount of albumin degraded per day increased, as did the total exchangeable albumin pool, indicating that at some point between the time when methadone administration was started to the time of distribution equilibrium of the second tracer dose of l2&i-albumin, albumin synthesis must have exceeded albumin degradation to account for this finding. The mechanisms responsible for these shifts in albumin distribution and the discrepancy between degradation and synthesis are not clear. Of the many factors which may influence albumin metabolism and distribu

5 218 ROTHSCHILD ET AL. Vol. 'II, No.2 Animal Total protein Plasma albumin TABLE 3. Effects of metluldone on albumin metabolism" Total exchangeabl. albumin (TEA) Total plasma albumin (intravascular alb) Total tklrevabc:ular albumia Albumin degradation C E C E C E C E C E C E,/l00ml,/l00ml,,, mg/ltg/day Mean ,",SEM P < 0.01 P < 0.01 P < 0.01 P< 0.01 P< 0.01 C, control; E, eiperimental (methadone); number of studies, 8. TEA - serum albumin concentration ([SAil x total exchangeable albumin space. Intravascular albumin - plasma volume x [SA). Extravascular albumin - TEA - intravascular albumin. to '11$1 au 'lui ALI,. I. 3.5 ", ~.--. I~ " l:ii,,,." -' t" i~ n"" ::,.,.::,.", ~! I I I I I I 1'" "'" laoo IS 1600!! 1400 ~! 1200.' 1000 ;.00 METHADONE 05~"q/4ay iller""", t. h./k./4ay-' " IIAICII 13 20!lAY 2a3013~TII.lJ1I[ FIG. 2. Rabbit 5. The serum albumin roae from 3.2 to an average of 3.8,per 100 ml. The product of the serum albumin level and the slightly elevated renal clearance resulted in a marked increase in albumin degradation during the methadone maintenance period, June 3 to June 9. tion, hormone levels, colloidal accumulation and nitrogen intake have been shown to be the most specific.' Cortisone and thyroid hormone stimulate albumin synthesis and degradation but do not cause an accumulation of extraplasma albumin. t. 0 Endocrine function in patients maintained on methadone has been studied, although the effects of methadone on the unbound hormone levels are not known. It has been shown that during the first 2-month period of development of tolerance to this narcotic, in maintenance patients, metapirone tests of hypothalamic reserve have been abnormal. After that time, however, all parameters studied, defining the functioning of the hypothalamic-pituitaryadrenal axis have been normal, including plasma cortisol and urinary 17-0H and ketosteroid levels. Similarly, with the exception of elevated levels of thyroid-binding globulin, all other parameters of thyroid function studied were normal in methadone maintenance patients. Extravascular colloid accumulation has been reported after immunologically stimulated hypergamma-

6 August 1976 METHADONE AND ALBUMIN METABOLISM 219 Animal Control TABLE 4. Albumin degradation Methadone TEA IV albumin Albumin degradation TEA IV albumin Albumin deiliadation g g mg/day %of TEA/day % of IV alb/day g g mg/day %01 TEA/day % of IV alb/day ' 19.5' ' 22.5' Mean ±SEM IV albumin, intravasculaj' albumin; TEA, total exchangeable albumin. 'Determined from the clearance method: Albumin degraded/day. Plasma albumin mass, Determined by the method of Matthews. II. " globulinemia or after infusions of dextran or -y-globulin. In these situations, albumin levels are low, not high, and albumin synthesis is depressed, not elevated. 2 '... Further, the shifts in albumin distribution in these instances are small in comparison to the large shift that took place after the institution of methadone treatments. Amino acid deficiency and plasmaphoresis result in shifts of albumin from extra plasma sites into the plasma. Hypercatabolism, however, is not observed, and albumin synthesis is depressed 10 nutritional deficiency." 1., 2' Thus, these factors which have been shown to influence albumin distribution and metabolism do not explain the present findings with methadone maintenance. With the data that are now available, it is not possible to define the specific etiological mechanism responsible for the alterations in albumin metabolism seen after methadone administration. Nevertheless, these studies show that an increase in the exchangeable albumin pool has been produced by the administration of methadone in experimental animals, and that albumin synthesis must have exceeded albumin degradation during methadone administration to account for these findings. This hyperalbuminemic hypercatabolic state is not seen in any other clinical or experimental situation to date. REFERENCES 1. Rothschild MA, Oratz M, Schreiber SS, et 01: Alterations in albumin metabolism after serum and albumin infusions. J Clin Invest 43: , Kreek MJ, Dodes L, Kane S, et 01: Long-term methadone maintenance therapy: effects on liver function. Ann Intern Med 77: , Kreek MJ: Medical safety and side effects of methadone in tolerant individuals. JAMA 223: , Rothschild MA, Dratz M, Schreiber SS. et ai : Albumin synthesis. N Engl J Med 286: ; Rothschild MA, Orotz M, Mongelli J. et ai: Alcohol induced depression of albumin synthesis: revenal by tryptophan. J Clin Invest 50: IS12-1S1S, Bale WF, Helmkamp RW, Davis TP, et al : High specific activity labeling of protein with '''I by the iodine monochloride method. Proc Soc E~p Bioi Med 122: , Berson SA, Yalow RS: Distribution and metabolism of 1111 labeled proteins in man. Fed Proc 16: 13S-1SS, Berson SA, Yalow RS, Schreiber SS, et al: Tracer experiments with '"l labeled human serum albumin: distribution and degradation studies. J Clin Invest 32: , Rothschild MA, Schreiber SS, Oratz M, et 01: The effects of adrenocortical hormones on albumin metabolism studied with albumin-lilt. J Clin Invest 37: , Rothschild MA, Bauman A, Yalow RS, et al: The effect of large doses of desiccated thyroid on the distribution of albumin-"'i in euthyroid subjects. J Clin Invest 36: , McFarlane AS, Koj A: Short term measurements of catabolic rates using iodine labeled plasma proteins. J Clin Invest 49: Gomall AG. Bardawill CJ, David NM, et 01: Determination of serum proteins by means of the biuret reaction. J Bioi Chern 177: Mancini JJ, Carbonara AO, Heremans JF, et 01: Immunochemical quantitation of antigens by single radial immunodiffusion. Immunochemistry 2: Sullivan HR, Blake DA: Quantitative determination of methadone concentrations in human blood, plasma and urine by gas chromatography. Res Commun Chern Pathol PharmacoI3:467-47S, Kreek MJ: Plasma and urine levels of methadone. NY State J Med 23: , Inturrisi CE, Verbely K: Disposition of methadone in man after a single oral dose. Clin Pharmacal Ther 13: , Kreek MJ, Schecter A, Gutjahr CL, et 01: Analyses of methadone and other drugs in maternal and neonatal body fluids: use in evaluation of symptoms in a neonate of mother maintained on methadone. Am J Drug Alcohol Abuse 1: , Matthews CME: Effects of plasmaphoresis on albumin pools in rabbits. J Clin Invest 40: , Matthews CME: The theory of tracer experiments with 'OIl labeled plasma proteins. Phys Med Bioi 2:36-53, Cushman P. Sordier B. Hilton JG, et al : Hypothalamic-pituitaryadrenal axis in methadone treated heroin addicts. J Clin Endocrinol Metab 30:24-29; Cushman P, Kreek MJ: Some endocrinologic observations in narcotic addicts. In Narcotics and Hypothalamus. Edited by E

7 220 ROTHSCHILD ET AL. Vol. 71, No.2 Zimmerman and R George. New York, Raven Press, 1974, p Shenkman L, Massie B, Mitsuma T, et aj: Effects of chronic methjldone administration on the hypothalamic-pituitary-thyroid axis. J Clin Endocrinol Metab 35: Azizi F. Vagenakis AG, Portnay GL. et al: Thyroxine transport and metabolism in methadone and heroin addicts. Ann Intern Med 80: , Rothschild MA. Oratz M. Franklin EC. et al: The effect of hypergammaglobulinemia on albumin metabolism in hyperimmunized rabbits studied with albumin J Clin Invest 41: , Rothschild. MA, Oratz M, Wimer E, et aj: Studies on albumin synthesis: the effects of dextran and cortisone on albumin metabolism in rahbits studied with albumin "'I. J Clin Invest 40: , Waterlow JC: Observations on the mechanism of adaptation to low protein intakes. Lancet 2: