Severe Protracted Alcohol Withdrawal Syndrome: Prevalence and Pharmacological Treatment at an Inpatient Detoxification Unit – A Naturalistic Study

This is the first systematic report about i) the prevalence of severe protracted alcohol withdrawal syndrome (PAWS) and ii) the effect of gabapentin on this condition. As a rule, acute alcohol withdrawal syndrome (AWS) lasts up to 14 days [6]. Withdrawal symptoms extending beyond this period have been inconsistently described and have been attributed to adaptations within the central GABAergic, glutamatergic and catecholaminergic systems [1] [7]. There has been no clear delineation between acute and protracted withdrawal syndrome due to a lack of studies, therefore the concept of PAWS has not been included in the DSM-IV and the ICD-10 [8]. Physicians in detoxification units are very famililar with the protracted withdrawal states of restlessness, anxiety, hyperhydrosis, insomnia, dysphoria, tremor and vegetative complaints. These often reappear in the course of a careful tapering of withdrawal medications which had been initially beneficial.

The established and effective medications in the treatment of AWS are benzodiazepines and clomethiazole which both are addictive [6]. The PAWS responds well to a dose-increase of these drugs [8], however this means treatment lasting up to several weeks with addictive drugs, which should be avoided whenever possible. Other drugs were anecdotally applied as an alternative (e.g., acamprosate, clonidine, tiapride, doxepine and sedative anticonvulsants). We chose gabapentin because (1) this medication has an influence on the central glutamatergic and catecholaminergic systems which are presumably involved in the pathophysiology of PAWS (see above), (2) there are data from animal studies that this drug might be successful in PAWS [2] [4], and (3) disadvantageous interactions of gabapentin with benzodiazepines and clomethiazole are not expected due to its pharmacological profile [3].

This naturalistic study was carried out at an inpatient detoxification unit between March and December 2005. A severe PAWS was diagnosed if the following criteria were fulfilled [6] [8]: 1) re-appearance and increase of typical alcohol withdrawal symptoms along with the careful taper of benzodiazepines or clomethiazole after at least 14 days of treatment; 2) immediate reduction of these symptoms after a test increase of benzodiazepine or clomethiazole doses; 3) no relevant psychiatric or somatic co-morbidity, e.g. anxiety, affective, psychotic disorder, substance abuse or deranged metabolic or cardiovascular disease; 4) no use of sedatives or hypnotics in the last 6 months; 5) ≥15 points on the revised Clinical Institute Withdrawal Assessment (CIWA-Ar) [9].

According to these criteria, 3 male and 2 female patients out of 109 inpatients undergoing detoxification from alcohol suffered from severe PAWS. All patients who had given informed written consent to the treatment were between 39 and 66 years old, had no relevant psychiatric and somatic co-morbidity requiring acute intervention, and no other substance use disorders (except from tobacco) as proved by urine screenings. The patients denied the intake of benzodiazepines or hypnotics in the last 12 months, which was confirmed by their partners or general practitioners. Patients also reported that there had never been a period of benzodiazepine or hypnotic abuse previously. There was no current use of the following medications: lithium, disulfiram, β-blockers, neuroleptics, anticonvulsants, antidepressants, or antacids, such as aluminium hydroxide or magnesium hydroxide. Their daily ethanol consumption at admission was between 300 and 640 g, with 10–48 years of drinking. They had only modest hepatic sequels.

When CIWA-Ar reached ≥15 points during gradual tapering of clomethiazole (n=3) or clonazepam (n=2), gabapentin administration began. At the start of gabapentin augmentation (baseline), in the five patients daily doses of clomethiazole were 4, 8, 8 capsules (@ 192 mg each), and of clonazepam 2.0 and 2.5 mg, respectively. Tapering was continued with 2 capsules clomethiazol or 0.5 mg clonazepam per day, respectively. A 600 mg gabapentin capsule was administered at baseline, followed by gabapentin capsules every 6 h. After 3 days with gabapentin 600 mg q.i.d. the taper began with 300–600 mg daily. CIWA-Ar ratings were carried out at baseline, two hours later, at day two, day three, and the end of the gabapentin augmentation by U.B. Mean scores of successive CIWA-Ar ratings were compared using paired-samples t tests (significance level p<0.05).

A significant decrease of CIWA-Ar scores from 18.2±3.8 points to 9.2±3.4 (mean±SD; p=0.009, n=5) was observed two hours after the first gabapentin administration. At days 2 and 3 a further decrease of the scores was observed in all 5 patients (mean, SD: down to 5.2 points±2.8, p=0.014, and 3.4 points±2.2, p=0.021, respectively). The mean duration of treatment with gabapentin was 13.6±5.4 days (n=5). At the end of the inpatient treatment, PAWS was nearly suppressed ([Fig 1]). The scheme of step-wise reduction of clometiazol and clonazepame could be carried out as planned. According to clinical evaluation and patients’ self-report, all patients felt well throughout the treatment without any hints of alcohol craving. The only side effects observed were mild somnolence, dizziness and ataxia in two male patients on the first day of gabapentin treatment.

Fig. 1 Individual courses of PAWS (measured by CIWA-Ar) under augmentation treatment with gabapentin: Note that the time scale is inhomogeneous.

In summary, we report on 3 male and 2 female out of 109 inpatients experiencing a severe PAWS in alcohol detoxification. An alternative explanation for the observed clinical state could be a benzodiazepine withdrawal syndrome, however, this could be ruled out by urine screens and by case history. According to the presented rate of 4.6%, a severe PAWS (CIWA-Ar≥15) requiring pharmacological treatment is infrequent. This might explain the lack of warranted controlled treatment studies in this field. We observed a good response to augmentation with gabapentin 600 mg q.i.d. in a small sample of patients. These data are preliminary and tentative, and should be proven by controlled investigations in larger samples. In addition, it should be investigated whether mild to moderate PAWS might profit from lower doses of gabapentin.

References

• 1 Aston-Jones G, Harris GC. Brain substrates for increased drug seeking during protracted withdrawal.  Neuropharmacology. 2004;  47 ((Suppl. 1)) 167-179
• 2 Bailey CP, Mollemann A, Little HJ. Comparison of the effects of drugs on hyperexitability induced in hippocampal slices by withdrawal from chronic ethanol consumption.  Br J Pharmacol. 1998;  123 215-222
• 3 Bonnet U, Banger M, Leweke FM. et al . Treatment of alcohol withdrawal syndrome with gabapentin.  Pharmacopsychiatry. 1999;  32 107-109
• 4 Dooley DJ, Donovan CM, Pugsley TA. Stimulus-dependent modulation of [(3) H] norepinephrine release from rat neocortical slices by gabapentin and pregabalin.  J Pharmacol Exp Ther. 2000;  295 1086-1093
• 5 Gatch MB, Lal H. Pharmacological treatment of alcoholism.  Prog Neuropsychopharmacol Biol Psychiatry. 1998;  22 917-944
• 6 Kosten TR, O’Conner PG. Management of drug and alcohol withdrawal.  N Engl J Med. 2003;  348 1786-1795
• 7 Krystal JH, Petrakis IL, Mason G. et al . N-Methyl-D-aspartate glutamate receptors and alcoholism: reward, dependence, treatment, and vulnerability.  Pharmacol Ther. 2003;  99 79-94
• 8 Satel SL, Kosten TR, Schuckit MA. et al . Should protracted with-drawal from drugs be included in DSM-IV?.  Am J Psychiatry. 1993;  150 695-704
• 9 Sullivan JT, Sykora K, Schneiderman J. Assessment of alcohol withdrawal: the revised clinical institute withdrawal assessment for alcohol scale (CIWA-Ar).  Br J Addiction. 1989;  84 1353-1357

Correspondence