The aim of the trial was to assess alpidem efficacy in preventing and treating the benzodiazepine (BZ) withdrawal syndrome (WS). A multicentre, double-blind, randomized versus placebo, parallel group study of six-week duration was carried out in outpatients suffering from generalized anxiety or adjustment disorder with an anxious mood and taking non-hypnotic BZ as continuous course of therapy of at least one-year duration. At the entry, the patients abruptly discontinued BZs and were treated with 50 mg/bid/tid of alpidem or placebo. Withdrawal syndrome diagnosis was (regarding treatment allocation) formulated by an independent psychiatrist, according to DSM-III-R and an appropriate scale, the SESSB. One hundred seventy-three patients were randomized and 148 completed the study. Withdrawal syndrome occurred in 27 patients of the alpidem group (31.0%) and in 38 patients of the placebo group (44.2%). A severe WS was diagnosed in 11.1% of the patients in the alpidem group and in 31.6% of the placebo group. If not having been withdrawn from the market, alpidem could have been useful for the prevention of BZ withdrawal syndrome.