Background: Alcoholism is a global problem with 5-10% of the world's population demonstrating alcohol-related diseases. One of the most severe consequences of alcohol dependence is the withdrawal syndrome, for which benzodiazepines are the most popular current treatment. An alternative method to benzodiazepine employs psychotropic analgesic nitrous oxide (PAN).
Objectives: To assess the effects of PAN for treating alcohol withdrawal states
Search strategy: We searched the Cochrane Central Register of Controlled Trials (The Cochrane Library Issue 2, 2005), MEDLINE, EMBASE, CINAHL (all to May 2005). We scanned internet websites, reference lists of relevant articles and abstracts of the international Conferences on Alcoholism. We contacted researchers in the field and industry to identify unpublished trials. No language and publication restrictions.
Selection criteria: Randomised controlled trials including voluntary participants dependent on alcohol. PAN was compared to oxygen and/or benzodiazepine regimens.
Data collection and analysis: Two authors independently assessed the methodological quality of the trials and extracted data.
Main results: Five studies, 212 participants, were included. PAN showed improvement of symptoms (RR 1.35; 95% CI 1.01 to 1.79), of the amount and duration of sedative medication and of psychomotor function (WMD -8.71; 95% CI -13.71 to -3.71). At one hour post intervention, no significant differences were found for depression (WMD -2.40; 95% CI -8.70 to 3.89) and anxiety (WMD -3.70; 95% CI -10.53 to 3.12). None of the included studies reported any significant adverse effects of any treatment.
Authors' conclusions: Results indicate that PAN may be an effective treatment of the mild to moderate alcoholic withdrawal state. The rapidity of the therapeutic effect of PAN therapy coupled with the minimal sedative requirements, may enable patients to enter the psychological treatment phase more quickly than those on sedative regimens, accelerating the patients recovery. Our review does not provide strong evidence due to the small sample sizes of the included trials. Neither does the review indicate any causes for concern that PAN is more harmful than the benzodiazepines. Clinicians wishing to use PAN may initially wish to do so within trial settings. Further high quality trials should be done to confirm these findings and to investigate whether the PAN therapy has fewer adverse effects than other treatments for the alcohol withdrawal states. Studies to investigate the possible cost-effectiveness of PAN by reducing costly hospital admissions and decreasing post administration supervision also need to be performed.