A Pilot Study to Determine the Usefulness of the Urinary Excretion of Methadone and its Primary Metabolite (EDDP) as Potential Markers of Compliance in Methadone Detoxification Programs S. George* and R.A. Braithwaite Regional Laboratory for Toxicology, City Hospital NHS Trust, Dudley Road, Birmingham, England, B 18 7QH Abstract ] Fourteen subjects (selected on the basis of compliance with the methadone-maintenance program prescribed by the consultant psychiatrist in charge of their treatment) undergoing opiate detoxification by methadone-replacement therapy were studied to determine if a relationship exists between the dose of methadone prescribed and the urinary excretion of methadone and/or its primary metabolite, 2-ethylidene-l,5-dimethyl-3,3-diphenylpyrrolidine (EDDP). After the derivation of this relationship, it was hoped that the urinary concentrations of methadone and/or EDDP could be used as a noninvasive technique to monitor the methadone compliance of 56 drug abusers. Despite statistically significant correlations (p <.1) between methadone dose and urine concentrations of methadone and EDDP, the large variation in concentrations measured in the urine of drug abusers negated the possibility of any clear-cut relationship being confirmed. However, it may he possible to use excretion data to monitor individual compliance but only through long-term monitoring of individual subjects to establish their own intraindividual variation in excretion patterns. 7 so.4 4.4 3 -I 2 "4 14 o! done and the plasma concentrations achieved (4). Furthermore, it has been suggested that subjects regularly attending clinics are more compliant while undergoing methadone detoxification programs than subjects treated at home (5). However, both of 3 3 4 4 5 5 ~ 55 6 7 7 75 ] I Dally methadone dose (mgl [] Methadone concentration [] EDDP concentration Figure 1. The urinary excretion of methadone and EDDP plotted against the dose prescribed for the 14 compliant subjects. Introduction There is a long history of studies concerning methadone-maintenance programs, methadone excretion (1,2), and therapeutic outcomes of maintenance programs (3). However, none of the research to date has related urine excretion data to compliance with methadone detoxification programs. Previously published data have proposed a relationship between the dose of metha- 3 3 4 4 SO 5 5 $$ 6 7 7 75 I I m Methadone/creatinine ratio [] EDDPIcreatinine ratio Figure 2. The urinary excretion of methadone and EDDP both corrected for urinary creatinine concentration plotted against the dose prescribed for the 14 compliant subjects. "Author to whom correspondence should be addressed. Reproduction (photocopying) of editorial content of this journal is prohibited without publisher's permission. 81
these studies were based on plasma data, which requires the collection of blood specimens from subjects. A recent Canadian study (6) suggested that once steady-state concentrations of methadone and its primary metabolite 2- ethylidene-l,5-dimethyl-3,3-diphenylpyrrolidine (EDDP) are achieved, urine concentrations of methadone and EDDP will reflect a steady state in excretion. Therefore, any changes in ~o,s 2o ~s 3o 3s 4o 4s so s5 6o ~ ~ 7s,o 9 ~ Figure 3. The mean + 1 SD of methadone concentration excreted plotted against the dose prescribed for the 144 specimens from 56 drug abusers. ZS 2o o e is G I L Q Q W 5 IO 15 lo 25 3 35 4 45 5 5S 6 65? 75 lo 9 13 dosage could be detected from altered excretion patterns. The hypothesis was, however, only tested using a single "stable" subject from whom serial specimens were obtained and over a period of 2 days analyzed. Methadone is a weak base (pka 8.62); therefore, urinary ph will have a marked effect on its excretion in urine. Acidifying the urine results in a rise in the amount of unchanged methadone excreted, leading to increased drug clearance. Nilsson et al. (7) found that the plasma half-life of methadone averaged 19.5 + 3.6 h in 5 subjects under acidic (ph 5.2) urine conditions. In the same subjects, the plasma half-life increased to 42.1 8.8 h when the urine was maintained at alkaline ph (ph 7.8). The respective clearances were 22% of the dose prescribed in 24 h for acidic compared with only 5% under alkaline conditions. The effect of urine ph on EDDP excretion is, however, less marked. Urinary excretion of methadone or EDDP could be a simple and noninvasive marker of methadone compliance. However, urinary ph and creatinine concentration (as a marker of urine concentration) also need to be measured to account for unpredicted or unexpected results. In addition, it was hoped that this pilot study could provide a mechanism to identify those subjects who are not compliant and therefore at increased risk of overdosage because of the recent concern over the high incidence of deaths in drug abusers where methadone has been implicated as a cause of death. This investigation was performed to try to resolve the issue of whether urinary excretion could be used in place of plasma concentration as a mechanism to monitor compliance during methadone replacement therapy for opiate addiction. Figure 4. The mean _+ 1 SD of EDDP concentration excreted plotted against the dose prescribed for the 144 specimens from 56 drug abusers. Aims of the Study I 3.5 3 2.5 "~ 1.5 m II o P i I I i i ~ ~ I i i I i I 2 3 4 ~ 6 7 8 9 I I1 12 1t Day of specimen --=- Methadone/creatinine ratio ~ EDDP/creatinlne ratio Figure 5. The historical profile of excretion data from a subject prescribed 3 mg of methadone daily over a period of 13 days. The first intent was to study the relationship between administered daily methadone dose and subsequent urinary concentrations of methadone and EDDP. This would be performed using random urine specimens collected to determine drug abuse patterns in subjects attending clinics and community drug teams. The urine specimens would be studied on a confidential basis to determine the methadone and EDDP excretion in these subjects. The second intent was to explore the various factors that influence the urine excretion of methadone and EDDP such as urine ph, body mass, urine concentration (as determined by urinary creatinine concentration), and methadone dose. 82
The final intent was to determine if quantitative urinary excretion data of methadone and EDDP may be used to distinguish between compliant and noncompliant subjects undergoing methadone detoxification. Specimen collection Random urine specimens were collected into plain, sterile 25-mL universal plastic containers without preservative. These were spot urine collections to mimic routine procedures that are currently followed by clinics and addiction units using the services of the Laboratory. Methods Materials and equipment Methadone and EDDP were purchased from Sigma (Poole, Dorset, U.K.). Prazepam was obtained from Parke-Davis Medical (Eastleigh, Hants, U.K.). Butyl acetate was HPLC grade from Sigma-Aldrich (Gillingham, Dorset, U.K.). The gas chromatography (GC) system consisted of a CTC Analytics A2SE liquid sampler, with an AI GC94 GC (both from AI Ltd Cambridge) linked to a Hewlett Packard (Stockport, Cheshire, U.K.) HP3395 integrator. The analytical column was purchased from Jones Chromatography (Hengoed, Mid Glamorgan, U.K.). Subject selection Following ethical approval (City Hospital NHS Trust Ethics Committee), a well-controlled and regulated population of 14 subjects consisting of both males and females undergoing methadone detoxification for opiate addiction was selected by the consultant psychiatrist at the local addictive behavioral services center. These individuals were monitored in an attempt to derive a relationship between dose and urinary excretion of methadone and EDDP. A questionnaire was developed to collect patient data, including height, weight, age, gender, methadone dose, and other drugs prescribed or taken (Table I). Following this initial phase, the study was extended to monitor the compliance of 56 drug abusers, including those suspected of missing doses or "topping up" from other sources of methadone. O R- 2.5 l.s t o,5 Urine analysis Urinary measurements of methadone and its primary metabolite, EDDP, were performed by capillary gas chromatography using a modified method of Caldwell and Challenger (8). The modifications were as follows: the capillary column used is a 15 m x.53-mm i.d. with a DB-5 phase (1.5-1Jm film thickness). The extraction was performed in a 1.9-mL capped, polypropylene Eppendorf tube. Sodium hydroxide (.1 ml, 5M), a.7-ml urine specimen, and.15 ml of internal standard solution (1 mg/l prazepam in butyl acetate) were placed into an Eppendorf tube. The tube was capped and vortex mixed for 2 s followed by centrifugation at 11, rpm for 5 min. The supernatant organic phase was then transferred to autosampler vials before injection o i ~ i I I I I I i i I I I 2 3 4 5 6 7 9 I 11 12 13 Day of specimen -e- Methadone/creatinine ratio ~ EDDP/creatinine ratio Figure 6. The historical profile of excretion data from a subject prescribed 8 mg of methadone daily over a period of 13 days. Table I. Demographic Data and Analytical Results from Spot Urine Specimens Collected from the 14 Compliant Subjects Methadone Urinary Urinary Urinary Other drugs currently Subjed Age Weight dose Urinary methadone EDDP creatinine taken by or prescribed number (years) Gender (kg) (mg/day) ph (mg/l) (mg/l) (mmol/l) to the subject 1 38 Female 48 3 5.5 3.4 9. 11.3 None 2 2 Female 48 4 5.3 13.9 25. 11.4 Nitrazepam 3 41 Male 83 3 5. 12.5 15.5 24.3 None 4 3 Male 51 4 5.4 3. 28. 2.6 Temazepam 5 28 Male 7 5 6. 1.3 5.1 18.3 Cocaine, Heroin, Zopiclone 6 35 Male 72 5 5.1 33.6 29.2 42.7 Amphetamine, Diazepam, Temazepam 7 36 Male 96 5 5.6 5. 19. 18.3 None 8 27 Male 76 55 5. 24. 58.1 25.9 Nitrazepam 9 36 Male 7 6 6. 7.5 12.8 8.7 None 1 27 Male 6 7 5.1 1. 6.9 13.5 None 11 27 Male 61 7 5.2 11.3 15.7 12.3 None 12 28 Male 75 75 7.2 6.7 15.9 7.8 Diazepam 13 41 Male 83 1 6.4 5. 8.4 1.4 Amphetamine, Cocaine, Diazepam 14 26 Male 6 1 5.2 1.1 1.5 17.7 Cannabis, Heroin 83
of 1 IJL onto the capillary column. The GC operating conditions were as follows: injector 26~ detector 3~ initial column temperature 11~ for I min, then 16~ to 29~ and hold for I min; total cycle time 23 min per specimen. Under these conditions, the retention times for EDDP, methadone, and prazepam were 7.35, 8.3, and 1.78 min, respectively. Standard calibration curves for both methadone and EDDP were prepared in drug-free urine at concentrations of 1, 5, 1, and 2 mg/l and extracted as per specimens. The calibration curves were run for each batch of specimens analyzed, and the derived concentrations of methadone and EDDP were calculated from a regression analysis performed on the standard calibration curve. Results The recoveries of methadone and EDDP were calculated from the standard calibration curves for neat solutions and urine extracts. It was found that the mean recovery for methadone was 86%, and the mean recovery EDDP was 85% using the described method. The between-batch coefficients of variation for the 1-mg/L urine standard were 9.2% for methadone and 11.% for EDDP (n = 6). At a concentration of 1 mg/l, the between-batch coefficients of variation for the urine standard were 8.8% for methadone and 11.% for EDDP (n = 6). Table II. Linear Regression Analyses of the Excretion Data for the 14 Control Subjects Studied" Parameter Correlation coefficient r 2 Significance Dose vs. methadone concentration.1282 p >.1 Dose/kg vs. methadone concentration.917 p >.1 Dose vs. methadone/creatinine ratio.587 p >.1 Dose/kg vs. methadone/creatinine ratio.357 p >.1 Dose vs. EDDP concentration.113 p >.1 Dose/kg vs. EDDP concentration.159 p >.1 Dose vs. EDDP/creatinine ratio.376 p >.1 Dose/kg vs. EDDP/creatinine ratio.42 p >.1 * Doses were measured in milligrams; concentrations were measured in milligrams per liter. Creatinine concentrations were measured in millimoles per liter. Table III. Linear Regression Analyses of the Excretion Data for the 144 Specimens from the 56 Drug Users Studied' Correlation Parameter coefficient r 2 Significance Dose vs. methadone concentration.1452 p <.1 Dose vs. methadone/creatinine ratio,298 p <,5 Dose vs. EDDP concentration,114 p <,1 Dose vs, ED DP/creatinine ratio,145 p >.1 * Doses were measured in milligrams; concentrations were measured in milligrams per liter. Creatinine concentrations were measured in millimoles per liter, Linear regression parameters from the results of the urine analyses performed for the 14 control subjects are presented in Table II. Figures I and 2 illustrate the methadone and EDDP excretion data, both with and without creatinine correction. It can be clearly seen from Table II and Figures 1 and 2 that there was no statistically significant relationship between the dose of methadone prescribed and the resultant excreted urinary concentrations of methadone or EDDP. It can further be seen from Table I! that correcting for body weight made no improvement in the correlation coefficient. The linear regression parameters from the urine analyses performed for the 144 specimens from the 56 drug users are presented in Table III. Figure 3 shows the mean 1 SD for the methadone concentration excreted in this group plotted against the daily dose of methadone. Figure 4 shows the excreted EDDP concentrations (mean 1 SD) against dose for the same group. Despite the strong, statistically significant (p <.1) correlation for the excretion of methadone and EDDP against dose prescribed (Table III), Figures 3 and 4 illustrate the large variation in results obtained from the analysis of the subjects urine specimens. Again, it can be seen that correcting for creatinine concentration did not improve the relationship between the dose of methadone prescribed and the resultant methadone and EDDP concentrations excreted in the urine. The results of urine analyses that were obtained from repeated monitoring of two subjects over a period of 13 days in each case are illustrated in Figures 5 and 6. From these Figures, it can be seen that the excretions of methadone and EDDP (both corrected for creatinine concentration) do indeed show some stability, but only when they are followed over periods of several days. Discussion It was hoped that a useful relationship between the dose of methadone and the excretion of methadone and EDDP in urine could be determined. It was also expected that it would be possible to use this relationship as a noninvasive index against which to monitor compliance in subjects maintained on methadone. The results presented in Table II suggest that there is too large of an interindividual variation to use urinary excretion concentrations of methadone or EDDP as markers of compliance. It can be seen that the dose accounted for only 13% of the variance in methadone concentrations excreted in the compliant group and only 11% of the EDDP excreted. These results were not improved by correcting for creatinine concentration in the urine specimens of the subjects studied. The data presented in Table III were derived from 144 specimens collected from 56 different subjects and yielded statistically significant correlations between dose and methadone concentration excreted and EDDP concentration excreted. However, the dose accounted for only 15% and 1% of the variance (r 2) of the excreted concentrations of methadone and EDDP, respectively, as illustrated in Table III and Figures 3 and 4. These results 84
would again point to the lack of suitability of using urine concentrations of methadone or EDDP as markers of compliance during a program of methadone detoxification. When a single subject is repeatedly monitored over a period of several days (Figures 5 and 6), it may be possible to determine a trend which can be followed for that individual. This data could indeed demonstrate that excretion patterns of both methadone and EDDP may prove to be useful mechanisms for establishing a subject's compliance. However, such patterns would need to be determined over a long time period to determine the variations in the excretion patterns of each individual, thereby distinguishing intra-individual changes from additional dosing or lack of compliance. One possible weakness of this study was that the time of collection of the urine in relation to the last dose was not controlled. However, for the current approach to be useful, it is necessary to be able to use data from randomly obtained urine specimens. In conclusion, it may be possible to use urinary methadone and EDDP excretion patterns to monitor compliance in individuals prescribed methadone for opiate detoxification, but the success of this approach would be based on the quantitation of every specimen received for each subject studied. These data would need to be archived so that historical excretion patterns could be established in preparation for the question: "Is this subject compliant or 'topping up' with methadone?" For practical purposes, it seems that the only reliable method available to monitor methadone compliance is the use of plasma methadone concentrations. Acknowledgment The authors would like to thank Dr. Ian Telfer, consultant psychiatrist, for his help in selecting and recruiting the compliant subject group under his care and Claire Meadway for her assistance in the preparation of this manuscript. References 1. K. Verebely, J. Volavka, M. Salvatore, and R. Resnick. Methadone in man: pharmacokinetic and excretion studies in acute and chronic treatment. Cl in. Pharm. Ther. 18:18-19 (1975). 2. G.D. Bellward, P.M. Warren, W. Howald, J.E. Axelson, and F.S. Abbott. Methadone maintenance: effect of urinary ph on renal clearance in high and low doses. Clin. Pharm. Ther. 22:92-99 (1977). 3. J. Holmstrand, E. Anggard, and L.M. Gunne. Methadone maintenance: plasma levels and therapeutic outcome. Clin. Pharm. Ther. 23" 175-18 (1978). 4. K. Wolff, M. Sanderson, A.W.M. Hay, and D. Raistrick. Methadone concentrations in plasma and their relationship to drug dosage. Clin. Chem. 37:25-29 (1991). 5. K. Wolff, A. Hay, D. Raistrick, R. Calvert, and M. Feely. Measuring compliance in methadone maintenance patients: use of a pharmacologic indicator to "estimate" methadone plasma levels. Clin. Pharm. Ther. 5:199-27 (1991). 6. B.M. Kapur, Z. Verjee, M. Anderson, and E. Geisbrecht. Urinary EDDP (methadone metabolite) as a marker of compliance and diversion of methadone. Ther. Drug Monit. 17:388 (1995). 7. M.I. N ilsson, E. Widerlov, U. Meresaar, and E. Anggard. Effect of urinary ph on the disposition of methadone in man. Eur. J. Clin. Pharmacol. 22:337-342 (1982). 8. R. Caldwell and H. Challenger. A capillary column gas-chromatographic method for the identification of drugs of abuse in urine samples. Ann. Clin. Biochem. 26:43-443 (1989). Manuscript received March 3, 1998; revision received May 28, 1998. 85