Concluding remarks and recommendations for future research
The aim of this study was to investigate the relevance of each of the following aspects which involve the clinical performance and the individual experience of craving of heroin-dependent drug users in methadone maintenance therapy (MMT).
The first study described in this thesis is, to our knowledge, the first effort to systematically measure methadone plasma levels simultaneously with its primary metabolite, 1,5-dimethyl-3,3-diphenyl-2-ethylidene-pyrrolidine (EDDP), in opiate dependent subjects, currently under MMT in a closed metabolic ward during a 24hour period. For methadone and EDDP measurement, a highly sensitive and reliable reverse-phase high-performance liquid chromatography (HPLC) technique was developed.
It was established that the steady-state pharmacokinetics of methadone can best be described by a two-compartment model. This observation has been suggested previously by others (Nilsson et al., 1982b), but was questioned recently (Wolff et al., 1993). The latter study, howeve4r, was performed under different circumstances than our first study. Our study did not provide much clarity with regard to correlations between the daily dose and plasma concentrations of methadone. In contrast to other studies (e.g. Wolff et al.,1991), poor correlations were found between the daily dose, on the one hand, and steady-state plasma concentrations or AUC(^ 24h)'s on the other. As discussed earlier bio-availability of methadone could not be assessed in this study.
The elimination rate constants (and half-lives) showed a wide range of variation, which is in agreement with other studies. A feasible factor which might affect methadone pharmacokinetics is the concomitant use of other drugs, i. e. enzyme inducing compounds. One subject using the known enzyme inducing drugs, isoniazid, rifampicin and azidothymidine, had a methadone elimination half-life of 13 h. The cytochrome p450 subtype responsible for metabolisation of methadone p450 3A4 (Iribarne et al., 1996) - can be induced by rifampicin as well, which explains the shortened elimination half-life in this subject. Another point is that we observed (weak) differences between pharmacokinetics of male and female subjects. The mean elimination half-life of methadone appeared to be 28 h for men and 35 h for females respectively.
This study indicated that plasma concentrations-time curves of EDDP could be measured in each of the 20 test subjects. Interidividual variation in pharmacokinetic parameters, e.g. of AUC(024h)'s were considerable, indicating a variable metabolic activity. The maximum plasma concentrations ranged from 10 to 301 ng-ml-'. The peak plasma concentrations (Cm",) of EDDP in most cases occurred ahead of the methadone peak: means 01:49 h:m and 02:20 h:m, respectively. This can be explained by a first-pass effect.
The second study (chapters 3 and 6), which was restricted to 6 heroin-dependent subjects, was performed in a closed metabolic ward under strictly controlled conditions, and using a newly developed analytical procedure. The pharmacokinetics of simultaneously administered dextromoramide and methadone were investigated. In contrast to methadone, the pharmacokinetics of dextromoramide behave according to a one-compartment model system. The mean elimination half-life time of dextromoramide found in this investigation amounts to 01:11 h:m, a considerably shorter value than that found by others (Lanšon et al., 1989; Pagani et al., 1989). It is well known that intensive polydrug use - as was the case with participants in this study, as compared to other investigations - may be an important factor in the induction of microsomal liver enzymes, resulting in decrease of elimination half-lives, in this case of dextromoramide.
The pharmacokinetics of l-methadone and d-methadone in opiate-dependent patients undergoing MMT with either l-methadone or d,l-methadone were compared in our third study (chapter 7). The l-methadone and d-methadone plasma concentration ratio showed a consistently higher l-methadone concentration. A lasting question concerning the side effects and the effectiveness of l-methadone as compared to d,l-methadone, has been investigated as well in this study. No significant difference between the two groups was seen with regard to dose adjustment requests, number of illicit drugs in urine and opiate craving.
The three studies described in this thesis started with the assumption that symptoms of craving among heroin-dependent patients in MMT can be influenced by their methadone dose, the presence of psychopathology and additional drug use. As the correlation between oral methadone dose and the plasma methadone concentration proved to be poor, both were evaluated separately with regard to their influence on craving.
If one assumes a negative relationship between the plasma concentration of methadone and the level of craving (and dysphoria), one would expect an increase of the patient's comfort and satisfaction with increasing plasma methadone concentrations. The results from the first study did not show a significant correlation between plasma methadone concentration (range: 69 - 630 ng-ml-') and the degree of opiate craving in the twenty MMT patients. But, in eleven of the twenty patients a significant increase of craving between 14:00 and 22:00 was observed. However, it must be argued that the first study, reporting on patterns of craving related to methadone pharmacokinetics (chapter 4) has its weaknesses in so far that the patient sample is small (n = 20) and heterogeneous, in a rather specific setting, with a lack of outcome parameters. On the other hand, the Experience Sampling Method used is a dynamic measure of craving which enabled the construction of daily craving patterns, which can be related to plasma methadone concentrations. The craving as reported by the study subjects, seemed to be more associated with anticipatory conditional responses in reaction to environmental cues, e. 9. habitual time of self-administration of drugs and the customary dispensing time of drugs in MMT or during admission to a clinic. This finding has been found by other researchers both among opiate-dependent patients (Wikler, 1952; Powell et al., 1993) and cocaine users (Ehrman et al., 1996). A pattern showing an anticipatory increase of craving towards the moment of methadone administration, is also suggested by observations made in an individual case (chapter 5). In spite of extremely high plasma concentrations of methadone (range: 1950 - 3421 ng-ml~'), marked increase in self-reported craving was observed around the expected daily methadone ingestion time. After the Ingestion of 250 mg methadone, the craving dropped to a low in a period of 4 hours. The opiate withdrawal signs, as measured in this case-report, although starting simultaneously with the craving, persisted throughout the measurement period. The findings in this case report suggest a distinction between psychological craving as measured with ESM and opiate withdrawal signs. Other studies have found poor relationships between plasma methadone and objective withdrawal ratings as well (Loimer et al., 1991b; Hiltunen et al., 1995).
In spite of the absence of a negative correlation between craving and plasma methadone concentration, a positive correlation was found between the increase of craving with the increment of the methadone dose in the range of 10 to 90 mg methadone per day among the twenty subjects in the first study. This can be explained by a higher methadone dose demand from poly-drug using MMT patients, who do not experience a diminution of their craving, in spite of the dose increase. However, the mean daily methadone maintenance dose prescribed by the Amsterdam Municipal Health Service in 1990, the year the first study started, was 34 mg. The daily MMT dose being based on the patients need and desire, this finding suggests that an increase of the daily methadone dose does not guarantee a diminution of the craving.
Not all participants are equally well satisfied with the use of methadone while undergoing MMT. Several patients undergoing MMT continue to experience discomfort, dysphoria and experience opiate craving, despite receiving what seemed optimal dosing of methadone. It has already been discussed that high plasma levels of methadone do not warrant suppression of these experiences. In the Amsterdam Municipal Health Service the opiate drug dextromoramide was introduced as an adjuvant to methadone, current rumour predicting that dextromoramide would be similar to methadone in producing euphoria. Another argument for this introduction was the suitability of this orally ingestable opiate for the - mostly from Surinam originating - non-injecting drug users in Amsterdam. In the second study presented in this thesis, in spite of the small patient sample (n = 6), observations from the heroin-dependent volunteers who received dextromoramide as an adjuvant to methadone revealed some interesting craving patterns. A clear relationship, although not significantly affirmed (r, = -0.32, P = .08, n = 30) due to small numbers of observations, was observed between plasma concentrations of dextromoramide and craving level in all subjects. A comparable distinct relationship has not been described before, neither was it found in the first study performed by us, using methadone only. The clear relationship between craving and plasma dextromoramide was not observed with the plasma methadone concentration. An explanation for this distinction can be the steeper rise of the dextromoramide plasma concentration (mean tmax 01:16 h:m) as compared with the methadone plasma concentration (mean tmax 02:18 h:m), enabling quicker occupation of CNS opiate receptors.
Significant correlations between the level of opiate craving and the presence of current psychopathology could not be established in this study, implying a clear discrimination between opiate craving and current psychopathology.
4. Urine monitoring
No relationship between the plasma methadone concentrations or the oral methadone dose and the number of illegal drugs in the urine was found among the twenty study subjects from the first study. Both low and high cravers, as well as subjects with a high or low plasma methadone concentration, showed positive urine samples for illegal drugs, a finding which is in accordance with other studies ( Loimer and Schmid, 1992; Wolff et al., 1996). Although urine monitoring for illicit drug use was performed in the second study, which evaluated the therapeutic effectiveness of d,l- and l-methadone, these results were used only for comparison purpose. In this study the group which was continued on l-methadone, showed an increase of positive urine samples for heroin.
Recommendations for future research and for methadone maintenance therapy
In this study the plasma EDDP concentrations were shown to appear in relatively high amounts, it is as yet premature to exclude any psychopharmacologic activity for EDDP. Considering the continuous presence of EDDP in long-term methadone users, and the close chemical resemblance of the two compounds, it would be worthwhile to submit the metabolite to a closer pharmacodynamic examination.
The wide variation of the pharmacokinetic parameters found in this study, call for further investigations with regard to the bioavailability of methadone in long-term methadone users, under strictly controlled experimental conditions. The gender differences found in our first study, with regard to the methadone elimination rate constant, seems worthwhile to investigate more closely.
The interpretations of the' results from the combined craving and plasma methadone and dextromoramide studies suggest that craving persists throughout methadone maintenance, despite adequate plasma methadone concentrations. Further investigations discerning craving patterns while using methadone and dextromoramide or dextromoramide exclusively seem to be indicated. However, the results from the methadone study suggests that other factors than plasma methadone level play a part in the craving level, which of course does not exclude the significance of assessing methadone levels, if only to establish pharmacokinetic causes of failure of MMT. It seems worthwhile to investigate those environmental or time-related cues, which seem to evoke craving, more closely. Furthermore, it seems that more investigations are indicated in relation to pathological personality traits or learned behaviour aspects influencing the outcome of MMT.
In any case, the study made it clear that plasma methadone levels cannot be used for clinical adjustment of clients undergoing MMT, who are harassed by opiate craving. A logical conclusion from the findings in this study is that further involvement of the MMT patients in the decision-making with regard to the methadone dose, the dosage schedules, and the effectiveness of treatment seems relevant for influencing the experience of continued craving during MMT. Moreover, an individual patient's MMT status, co-medication, polydrug use should be taken into account.