— Aims: About 50% of alcoholic patients relapse within 3 months of treatment. Previous studies have suggested that cyanamide may help to prevent such relapse. The aim of our study was to assess the efficacy and safety of long-term cyanamide treatment in alcohol dependence of adolescents. Methods: In this, double-blind, placebo-controlled study, we recruited 26 patients, aged 16–19 years, with chronic (frequent and regular) or episodic (frequent, but irregular) alcohol dependence. Patients were randomly allocated treatment with cyanamide (200 mg daily) or a placebo for 90 days. Patients were assessed on the day the treatment was started, and on days 30 and 90, by interview, self-report, questionnaire and laboratory screening. Patients were classified as abstinent, relapsing or non-attending. Time to first treatment failure (relapse or non-attendance) was the primary outcome measure. Results: The cyanamide (n = 13) and placebo (n = 13) groups were well matched in terms of baseline demographic and alcohol-related variables. Mean cumulative abstinence duration was significantly greater in the cyanamide group than in the placebo group. Apart from occasional diarrhoea, there was no difference in side effects between groups. Conclusions: Cyanamide seems to be an effective and well tolerated pharmacological adjunct to psychosocial and behavioural treatment programmes for the treatment of some adolescent alcohol-dependent patients. Because of reported hepatotoxic, haematological and dermatological side effects, patients should be observed continuously by experienced clinicians. Further studies are necessary to prove the efficacy of cyanamide in adolescents. Treatment for chronic alcoholism.
About 50% of alcoholic patients relapse within 3 months of completion of treatment ( Feuerlein, 1986 ). Several studies have reported various results with opiate antagonists. O’Malley et al.(1992), Volpicelli et al.(1992) and Mason et al.(1994) reported a significant reduction of relapse severity, whereas Krystal et al.(2001) reported no significant effect. With drugs that affect transmission of serotonin: Naranjo and Kadlee (1991) reported a significant effect, dopamine: Shaw et al.(1994) and Borg (1983) reported a significant effect, whereas Walter et al.(2001) reported an increase in relapse rates caused by D1 and D2 antagonists, γ-aminoburyric acid (GABA): Gallimberti et al.(1992) reported a significant effect, and acamprosate: Lesch et al.(2001) reported a good effect only in special subgroups of patients.
Cyanamide is an aldehyde dehydrogenase (ALDH) inhibitor used as a pharmacological adjunct in the aversive treatment of chronic alcoholism. Its elimination half-life and total plasma clearance values range from 42.2 to 61.3 min and its oral bioavailability is 70 at a 1.0 mg/kg dose ( Colom et al., 1999 ). Cyanamide blocks ethanol metabolism by inhibition of both the low- and high-K
forms of ALDH (Loomis and Brian, 1983a,b, Cederbaum and Dicker, 1985 ) in a pH-dependent manner [at pH <, 7.5 formation of an irreversible form, at pH >, 8.5 formation of a reversible form ( DeMaster et al., 1998 )]. As a result of this inhibition an increase is induced in acetaldehyde concentrations in blood and liver. This is responsible for the alcohol deterrent activity of cyanamide, because it induces a severe reaction (the toxic acetaldehyde syndrome) characterized by tachycardia, hypotension, flushing and dyspnea ( Brien et al., 1978, 1979 ).
The mechanism of the inhibition of ALDH by cyanamide remains unclear. Some authors postulate that this inhibition is achieved by one active metabolite ( DeMaster et al., 1979 ). Pruflonosa et al.(1989 Pruflonosa et al.(1991) suggested that the metabolic conversion of cyanamide to an active inhibitory form does not take place, and that ALDH activity thus appeared to be irreversibly inhibited in vitro in the presence of catalase and NAD
Treatment for severe alcoholism
. So far, cyanamide pharmacokinetic studies have been carried out in experimental animals ( Deitrich et al., 1976 , Obach et al., 1986 ) and in most of these studies the doses administered were higher than the therapeutic doses.
Cyanamide may also have some side effects, which fortunately could be observed in only a few cases: Bruguera et al.(1987), Yokoyama et al.(1995), Suzuki et al.(2000) and Tamai et al.(2000) reported the possible development of ground-class inclusion bodies in the hepatocytes, Ajima et al.(1997) observed cyanamide-induced granulocytopenia, Yerro et al.(2000) even reported cyanamide-induced pancytopenia. Rios-Herranz et al.(1992) reported the development of aplastic anaemia. Dermatological side effects of cyanamide have also been observed: Abajo et al.(1999) reported cyanamide-induced eczematous erythroderma, Kawana et al.(1997) reported exfoliative dermatitis and Torrelo et al.(1990) reported lichen-planus-like eruptions.
No data reporting the benefit of cyanamide for alcohol-dependent adolescents have been reported. That is why we undertook a double-blind, placebo-controlled trial of 90 days’ treatment with cyanamide, the endpoint for analysis was continuous abstinence. Because of the reported side effects, a continuous observation of the patients included in our study was essential.
SUBJECTS AND METHODS
Eligible patients were those who presented to our hospital that treats in-patients with alcohol dependence of chronic (frequent and regular) or episodic (frequent but irregular) type (DSM-III criteria) ( American Psychiatric Association, 1980 ) because of national documentation rules. Patients had to be aged 16–19 years, to have been abstinent for at least 5 days before the study, to have a γ-glutamyl transferase (γGT) value of at least twice the upper limit of the normal range or a mean corpuscular volume (MCV) of 93 fl or more, or both, and their parents gave written informed consent. We used the CAGE questionnaire’s ( Ewing, 1984 ) four clinical interview questions on cutting down, annoyance by criticism, guilty feelings, and eye-openers to assess the severity of patients’ alcoholism.
We excluded patients with serious coexisting disease (inadequately controlled juvenile diabetes mellitus, hypertension, cardiac failure, septicaemia, active tuberculosis, neoplastic disease, renal failure with a serum creatinine concentration of 120 μmol/l or more and hypercalcaemia of all aetiologies, epilepsy unrelated to alcoholism, and psychiatric disorders that might necessitate specific drug treatment, polytoxikomanic patients, patients with severe personality disorders). In 2000, we screened 36 patients, of whom three were excluded because of coexisting disease. Thus, 33 patients were recruited. The study was conducted according to the European Good Clinical Practice Guidelines and the Declaration of Helsinki under the auspices of an Ethic Commission.
Under the intention-to-treat principle, all randomized patients are eligible for analysis irrespective of whether they fulfil the conditions of the protocol. We used a slightly modified approach, in that we excluded seven patients who had been randomized but who did not attend the assessment on day 0 and, therefore, did not receive any medication. Lehert (1993) proposed this modification for alcohol studies because it is more practicable than the standard intention-to-treat approach for studies of patients with very high withdrawal rates and low motivation.
The remaining 26 patients underwent an in-patient, pharmacologically supported, alcohol-withdrawal treatment. After abstinence for at least 5 days they were reassessed and baseline measurements for safety and efficacy calculations were made. Patients were then randomly assigned to cyanamide or a placebo. In our assessment, day 0 was the day when cyanamide or placebo treatment was started. Cyanamide and placebo tablets were identical in appearance. Patients received 200 mg daily (two tablets in the morning, one at midday, and one in the evening). Patients in the placebo group took the same number of tablets. Patients were classified as abstinent or relapsed on day 0 and on days 30 and 90 according to their self-reports. The investigator recorded their judgements of whether the self-report was likely to be true and biological markers (γGT, MCV). The duration of double-blind treatment was 90 days. All patients received additional psychosocial and behavioural treatment. Patients who missed a visit, but attended the next one, were not withdrawn. Patients who relapsed during treatment were able to continue in the study on an out-patient basis, or were admitted to hospital for alcohol withdrawal where they continued to take their coded medication, however, if such patients could not be returned to the community within 15 days they were withdrawn from the study. The variables used in assessment were: red and white blood cell count, prothrombin time (70–120%), and electrolytes, blood urea nitrogen, creatinine, uric acid, fasting blood glucose, total bilirubin, triglycerides and albumin, as well as the alcoholism-related variables (Table 1 ). Mean cumulative abstinence duration was significantly greater in the cyanamide group than in the placebo group (Table 2