[Therapeutic action lag time and resistance to treatment].

The need for an antidepressant with a short time to action onset is justified by the need to reduce the risk of suicide, the patients suffering and the cost of the disease. How can a difference in action lag time be demonstrated? An appropriate methodology specific to the question must be developed. The selection of sensitive scales is an initial requirement enabling detection of early onset alleviation under treatment. Response criteria must first be defined in order to clearly answer the question "what is a short time to action onset?". Frequent scale determinations (Hamilton Depression Rating Scale, MADRS) at the start of the study is indispensable. The literature recommends at least two determinations per week over the first two weeks of treatment. Statistical methods of the survival curve type are the most appropriate for demonstration of a between-treatment difference in terms of time to action onset. Some authors consider that only placebo-controlled studies enable a difference in terms of time to action onset to be concluded. For other authors, neglecting studies versus reference products seems prejudicial to demonstrate a difference between the times to action onset of reference products and new products (Montgomery, 1997). The literature on time to action onset is not very rich. An initial study comparing amineptine and fluoxetine concluded the time to action onset was shorter for amineptine (Daléry et al., 1992). The value of the booster effect of pindolol in combination with serotonin reuptake inhibitors is controversial. Venlafaxine seems to accelerate the response (day 4) and be superior to a comparator (Clerc et al., 1996; Guelfi et al., 1995). The study conducted by Guelfi, in France, in patients presenting with severe depression was designed to evidence the efficacy of venlafaxine in melancholia. In 1996, Benkert published a double-blind study comparing venlafaxine and imipramine in severe depressions. Both studies stressed the rapid action onset of venlafaxine (between day 4 and week 2) in severe depressions. The number of patients presenting with a characterized state of depression and not responding to initial antidepressant treatment has been estimated at 30%. While there is no formal consensus on the definition of resistance to treatment, certain authors define responders as those showing a 50% reduction in the total score on the Hamilton Depression Rating Scale (HAM-D) or MADRS. Helmchen (1991) considers that resistance can only be suggested after two successive single-agent treatments with antidepressants with different action mechanisms have failed. A number of factors must be considered with respect to the genesis of treatment failure: mainly psycho-organic, psychoaffective and psychosocial factors.