Chapter 8

OPIATE CRAVING AND PRESENCE OF PSYCHOPATHOLOGY

J.W. de Vos¹, W. van den Brink², R.S. Leeuwin¹.

¹ Department of Pharmacology, Academic Medical Centre, University of Amsterdam, Meibergdreef 15, 1105 AZ Amsterdam, The Netherlands.

² The Amsterdam Institute for Addiction Research, Jacob Obrechtstraat 92, 1017 KR Amsterdam, The Netherlands.

Printed in:

European Addiction Research

Finding the adequate methadone dosage for opiate addiction has been a problem since the introduction of methadone as a maintenance treatment. Although methadone maintenance treatment (MMT) has been the pivot of opiate addiction therapy for many years, not all participants are equally satisfied with methadone as a replacement for or adjuvant to their heroin use. Their dissatisfaction or persistent opiate craving is expressed by continued additional heroin use, continued requests for higher dosages of methadone or low treatment retention rates. Several studies have been performed in order to establish an explanation for the difficulties encountered in the clinical practice of maintaining unsatisfied opiate addicts in methadone maintenance. Individual differences in pharmacokinetics and pharmacodynamics and the co-occurrence of psychiatric disorders have most frequently been mentioned and investigated as potential sources for this dissatisfaction.

Pharmacokinetic studies have focused both on oral dose requirements and on a possible aberrant methadone metabolism in unsatisfied MMT clients. When MMT started, high doses (between 80 and 120 mg/day) have been propagated (Dole and Nyswander, 1965; Dole, 1994). In the following 3 decades several studies have appeared which concluded that a dose of 50 or 80 up to 100 mg per day methadone gives the best results in MMT when decrease of illicit heroin use and/or program retention are considered. This lead to the now advised 80, plus or minus 20 mg, by the American Methadone Treatment Association (Parrino, 1993). However, reports of individual MMT clients who could not be satisfactorily stabilized on a high daily methadone dose, have been presented throughout the history of MMT (Whitehead, 1974; Horns et al., 1975; Goldstein et al., 1975; Bell et al., 1990; Loimer and Schmid, 1992). In order to control for individual differences in bioavailability and metabolism of methadone, plasma concentration - effect studies have been performed. Again clients in MMT were found who were persistently not-satisfiable even with high plasma methadone levels (Horns et al., 1975; Bell et al., 1990; Loimer and Schmid, 1992; de Vos et al., 1996a; de Vos et al., 1996b).

Research outcomes are mixed regarding the effect of MMT on the use of alcohol. In the 12-year DARP follow-up study, "heavy drinking" was reported by 21 percent of the sample in the month before treatment; it rose to 31 percent during the first year afterwards and then declined to 22 percent by year 12. Half the patients reported substituting alcohol for opiates after stopping daily illicit opiate use (Lehman et al., 1990).

The prevalence of psychiatric comorbidity among heroin addicts is widely described and acknowledged (Weissman et al., 1976; Khantzian and Treece, 1985; Rounsaville et al., 1982; van Limbeek et al., 1992). Indications are found that the individual dose requirements differ in relationship with the presence of both state (axis I pathology in DSM IIIR) and trait (axis II pathology in DSM IIIR) psychopathology. Higher methadone doses are given in cases where personality disorders, marked by odd or eccentric behaviour and social withdrawal or isolation (cluster A, axis II in DSM III), are found (Treece and Nicholson, 1980). In a large (n=106) study higher prevalence of emotional distress and anxiety was found in patients with a higher methadone dose (Roszell and calsyn, 1986). It has also been postulated that addicts who show, besides certain psychopathological symptoms, aggressiveness require a higher methadone dose (Maremmani et al., 1993).

This article presents the results from a study designed to explore the possible explanations for the unresponsiveness towards methadone maintenance in individuals currently in methadone maintenance treatment. In an effort to elucidate some of the above mentioned questions, we have investigated simultaneously the pharmacokinetics, the levels of craving and the presence of psychopathology in twenty opiate addicts who where admitted to a closed metabolic ward. The results from the pharmacokinetic part of the study, which were previously published, showed a poor relation between methadone dose and methadone plasma concentration (de Vos et al., 1995). The craving study, previously published, showed no correlation between craving level and plasma methadone concentration on group level. However, a weak positive (!) relation between oral methadone dose and craving level was found (de Vos et al., 1996a).

In this presentation the same group of twenty opiate addicts is examined for presence of psychopathology, alcohol dependence and craving. The predictive potential of psychopathology and alcoholism on craving is investigated.

Methods

Subjects

Twenty long-term opiate addicts, mean MMT duration 6.9 years (range 0.33-13 years), joined the study on voluntary basis after giving written informed consent. A consecutive series of patients was recruited from new admissions to the clinic. The study was performed in a closed metabolic ward of the Crisis, Observation and Detoxification Department of the Jellinek Clinic, Amsterdam. The admittance criteria to the clinic are of somatic, psychiatric or social origin. The clinic can also be entered as a pathway to other methadone reduction or maintenance programs. All patients who where asked to enter the study did so with the exception of one who refused to participate due to a reluctance to intravenous blood sampling. The mean age of the patient sample was 30 years (SD 4.5) with 45 % being female. There is great variation in the range of doses in the sample, from 10 to 225 mg daily, the mean daily methadone dose in the patient sample is 60 mg. Six subjects were tapering their methadone dose during the study. Urine samples showed use of heroin (in 80 % of the cases), benzodiazepines (50 %) and cocaine (60 %), however no methaqualone, amphetamine or barbiturates were found. In one case no methadone was found in the urine sample (Table 1).

Craving

The measurement of the individual daily craving levels has been conducted with the Experience Sampling Method (ESM) (Csikszentmihaly and Larson, 1987; de Vries, 1987; Kaplan, 1992). At 10 random moments during the day (from 8:00 to 22:00) a signal from a wrist watch prompted a self-report. Six items were used to assess craving, each item was scored on a 7-point Likert scale, which ranged from 'no at all' to 'very much'. The items include: "Did you think about using?", " Did you feel stoned?", "Where you in control of yourself?", "Did you feel restless?", " Did you need dope quickly?", "Did you feel the need to use dope?". Principal components analysis produced a single factor. The loading of each item was used to construct a scale by multiplying the raw item score by the respective item loading. The scale ranges from non-craving (score = -0.25) up to a maximal craving (score = 20.57). A maximum of 40 (4 days) craving scores for each subject are used to calculate the mean craving level for each subject.

Table 1 Client addiction history and current status

sub.	sex	age	time in MMT	methadone use in last month	current dose‡	tapering	drugs in urine*			alcohol
no.	m/f	y	y	days	mg/day	y/n	Op	Bz	Co	DIS
1	m	29	7	30	70	n	+	+	+	I
2	f	31	5	9	40	n	+	-	+	I
3	m	32	12	30	55	y	+	+	-	IV
4	m	24	9	20	40a	y	+	+	+	IV
5	m	33	4.25	30	60	n	+	+	+	IV
6	f	26	3	30	30	n	+	-	-	IV
7	f	32	6.5	21	50	n	+	+	-	I
8	M	28	7.5	5	30	n	+	-	-	IV
9	F	39	1	30	50	n	+	-	+	I
10	M	39	10.5	30	65b	n	-	+	-	IV
11	M	24	7	6	70	n	+	-	+	IV
12	F	21	0.33	30	30	y	-	-	-	I
13	F	31	6	30	60c	n	+	-	+	I
14	F	28	6	25	20	n	+	+	+	III
15	M	30	6	20	10	y	+	-	+	I
16	M	34	9	27	60	n	+	-	+	I
17	F	31	10	30	225	y	-	+	-	IV
18	F	28	13	30	70	n	+	+	+	II
19	M	27	9	30	70	y	-	+	-	IV
20	M	28	6	30	90	n	+	-	+	I

‡Current dose: a - 20 mg at 11:23 and 20 mg at 12:30; b - schedule last 4 days: 80, 100, 75, 65 mg/day; c - dubious compliance (no methadone in urine). *Drugs in urine: Op - opiates; Bz - benzodiazepines; Co - cocaine. Alcohol depicts the level of alcohol addiction; I - no abuse and no dependency; II - yes abuse but no dependency, III - yes dependency but no abuse, IV - yes both abuse and dependency.

Psychopathology

The Addiction Severity Index (ASI) is a semi-structured interview that collects data in seven problem areas: medical, employment/support, alcohol related, drug related, legal, familial/social, and psychiatric. In each problem area, objective information on client background and current status is collected as well as the client's subjective estimate of the seriousness of his problems on a 5-point scale. From these two sources of data, the interviewer provides an estimate of problem severity on a 10-point scale (McLellan et al., 1992). For use in The Netherlands a validated translation has been developed and used in this study (Hendriks, 1987; Hendriks et al., 1989). In this study only the objective information of the psychiatric problem area scale has been used. The items of this subscale include the number of psychiatric hospitalizations ever, and the presence of the following psychiatric symptoms in the last 30 days: depression, anxiety or tension, concentration or memory problems, hallucinations, control over violent behaviour, presciption of psychiatric medication, suicidal thoughts and suicide attempts. The composite score (CSs) is arithmetically derived from the set of items in the psychiatric problem area and ranges from .00 (no problem) to 1.00 (severe problem) (McGahan et al., 1986). The subjects estimate and the interviewer's interpretation of the objective information combined with the subjects estimate, leads to an 'interviewer severity rating' of the subjects addiction problem. This rating was not used in this study. The ASI was taken on the third participation day of each study subject.

The General Health Questionnaire (Goldberg, 1972) is a self report instrument for the detection of psychopathology in the community and among primary care patients. The original instrument contains 60 items that refer to the severity of psychological complaints during the last 4 weeks relative to the person's normal situation. In the present study, we used the scaled 28-item version of the GHQ. The GHQ-28 contains four 7-item scales: somatic complaints, anxiety and sleeping disorders, social dysfunctioning and severe depression. The response format is 0 = better than usual, 1 = same as usual, 2 = worse than usual, 3 = much worse than usual. In this study item responses 0 and 1 are rated as 0 (symptom not present) and item responses 2 and 3 are rated as 1 (symptom present).The total score (all scales) of the GHQ-28 ranges from 0 to 28. A total score of 17 or more has a chance of .54 or higher, using the Present State Examination as validation (Ormel et al., 1989), for presence of a psychiatric disorder. A score of 4 on a scale is used as a threshold for presence of pathology. The GHQ-28 was taken on the second participation day of each study subject.

The Symptom Check List - 90 item version (Derogatis et al., 1973), is a self report questionnaire with 9 scales: agoraphobia, anxiety, depression, somatization, insufficiency of thought and action, suspicion and interpersonal sensitivity, hostility, sleeping disorders and psychoneuroticism (= total score). The items refer to psychopathological symptoms during the past week. The distribution of the items over the scales in the Dutch version of the SCL-90 is not completely identical to the American version. The anxiety and phobic anxiety scales are identical but the depression scale has been modified (Arrindell and Ettema, 1986; Koeter, 1992). The total score ranges from 0 to 360. The SCL-90 was taken on the second participation day (in randomized alternating order with the GHQ-28) of each study subject.

Alcohol dependence in this study was measured with the Diagnostic Interview Scheme (van Limbeek et al., 1986). Alcohol dependency and abuse is present in 10 subjects.

Statistics

Spearman rank order correlation (r_s) was used in comparing the ordinal measurements (craving and psychopathology). The internal consistency of the various subscales has been described using Cronbach's alpha coefficient (α). The significance level was set at 5 %. SPSS statistical package was used to calculate the various statistics (Norusis, 1988).

Results

Table 2 presents the means and standard deviations of the independent variables, the internal consistencies and the correlations between the independent variables and the level of craving. The mean craving level for all subjects is 3.8 (SD 2.5). The range is from 0.15 to 8.58 across the individuals. The mean ASI psychiatric composite score is high (.36, SD .18), comparable with a psychiatrically ill substance abuse group (.37(McLellan et al., 1992). The GHQ results are consistent with a high level of psychopathology as indicated by a mean GHQ total score of 14.3. A total GHQ score ³ 17 is seen in 50 % of the patients (n = 10), indicating a high chance of an existing psychiatric disorder. A depressive mood disorder is possible in 9 subjects (item score ³ 4). Analysing the SCL-90 data, the mean scores on the anxiety, depression and psychoneuroticism subscales in this study are high, compared with a normal control group (Arrindell and Ettema, 1981).

Table 2 Independent variables and correlations with craving

ASI subscale	α	_	SD	Max	rs - Craving	P
psychiatric composite score	.69	0.36	0.18	1	.07	.77
GHQ-28 subscales
somatic complaints	.57	3.9	1.7	7	.04	.88
anxiety and sleeping	.75	4.3	2.2	7	.19	.43
social dysfunctioning	.81	2.7	2.3	7	.07	.75
severe depression	.77	3.3	2.3	7	.07	.77
total-score	.90	14.3	6.8	28	.05	.84
SCL-90 subscales
agoraphobia	.87	6.5	7.1	28	.21	.37
anxiety	.91	15.9	10.8	40	.13	.57
depression	.92	31.2	17.1	64	.03	.89
somatization	.86	17.8	9.9	48	.35	.13
insufficiency of thought and action	.88	13.2	8.6	36	.39	.09
suspicion and interpersonal sensitivity	.91	22.3	15	72	.23	.33
hostility	.74	6.5	4.9	24	.32	.17
sleeping problems	.90	6.4	4	12	.28	.24
psychoneuroticism	.98	131	70.5	360	.17	.47

The insufficiency of thought and action, somatization and the hostility SCL-90 subscales showed a substantial correlation with craving, which however did not reach statistical significance. The insufficiency of thought and action subscale uses most of the original obsessive-compulsive items used in the American version of the SCL-90 (Derogatis, 1977). No significant correlation was found between the individual level of craving and the Addiction Severity Index psychiatric composite score. No correlation was found between the level of craving and the alcohol dependency status (t = .53, P = .60).

Discussion

The results from our - previously published - craving study showed that no correlation exists between the plasma methadone trough level and the mean level of craving in these 20 long-term opiate addicts currently in MMT. In this study, using the same relatively small sample, a significant correlation between the level of opiate craving and the presence of general psychopathology could not be established. Because the plasma methadone trough concentration and the mean daily craving level are not correlated, plasma levels can not confound the relationship between general psychopathology and craving level and, therefore, control for this factor is not necessary. A non-significant correlation between craving and three SCL-90 subscales is observed. Although the SCL-90 is a mental state examination, these associations may indicate that related personality traits are predictive of craving. It is known that certain personality traits may predispose towards, or coexist independently with specific state psychopathology (Docherty et al., 1986). The association of craving with these subscales points to a possible presence of a personality disorder - not measured in this study - influencing the experience of craving.

The absence of a clear correlation between the ESM craving measurement and general psychopathology also implies a distinction between opiate craving and general discomfort or general psychopathology.

Together these findings suggest that neither individual methadone pharmacokinetics nor the prevalence of general psychopathology, can explain the presence of high craving in these long-term MMT patients.

A significant increase in craving level with a higher methadone dose, can be explained by a higher dosage demand among high cravers due to the use of comedication (de Vos et al., 1996a). Although the use of dose manipulations in MMT either up or down is not recommended in the American State Methadone Treatment Guidelines (Parrino, 1993), studies exist that examine the influence of both positive and negative dose contingencies on MMT retention in the maintenance phase (McCarthy and Borders, 1985; Stitzer et al., 1986). Environmental factors such as cue exposure, research setting and anticipatory conditioned responses seem important in the experience of craving (Powell, 1995). A previous study, using the same study sample as presented here, showed a time related habitual increase of the level of opiate craving among hospitalised methadone maintenance treatment patients around their usual methadone intake time although plasma methadone levels are high (de Vos et al., 1996a). Even objective withdrawal signs have been observed in MMT clients with (extremely) high plasma methadone levels (Loimer and schmid, 1992; de Vos et al., 1996b). Future research should investigate the influence of pathological personality traits, which where not included in this study.