The answer to this depends on 1.) whether it is a standing dx. 2.) whether or not the timing of any new dx was made appropriately. I have seen a lot of people being given a BDI upon admission into treatment (Who wouldn't be depressed when the "hit bottom"?). That that is very poor timing, and almost always erroneous. If they have an existing dx, I recommend 6 months to a year of continued SSRI's. I generally discourage any new medication being introduced during the treatment process unless there are clear indicators of enduring psychopathology (not just anxiety and depression)

Dear Mateju, 

although rodents are very good models for human, they may differ in this respect: 

http://www.ncbi.nlm.nih.gov/pubmed/23551979

Article Increased alcohol consumption in rats after subchronic antid...

Article Alcoholism & Drug Dependence

Article Psychotropic Drug Use in São Paulo, Brazil - An Epidemiological Survey

It is difficult to say as there are no reliable data on the prevalence of actual depression in alcohol dependent pts. In general, in spite of recently rapidly increasing prescription of ADs in many countries, the estimated prevalence of depression is there even higher so that in general, depression is supposed to be rather under-treated. Even if depression is being falsely positively diagnosed in persons "bankrupted" because of alcohol, the overuse of antidepressants in addicted pts might be non-significant if any. CH

Yes, and for at least two reasons. First, the specific effectiveness of antidepressants over placebo is small and getting smaller. (See recent series of papers in the WPA bulletin, among many other articles on the 'medicalisation of unhappiness'.) DSM-5 notwithstanding, 'understandable misery' does not equate with 'depression'. With active placebos (i.e. drugs without known antidepressant effects that cause symptoms like dry mouth, which make it more likely that patients will think they are taking an effective drug and thus increase placebo effects) the difference between placebo and active wings is even less.

Second - as has already been pointed out - patients whose lives have been sufficiently messed up to seek treatment are very probably feeling understandably unhappy, especially if they have been trying and failing to abstain or drink moderately. Antidepressants will not cause them to regard their excessive drinking and other undesirable behaviour as somehow positive and desirable and neither will it have that effect on their families and employers. However, when they have been dry for 2-3 months, much of that unhappiness often improves or diminishes to 'ordinary' human levels. (We all live in an imperfect world.) Families and employers may also feel happier and more hopeful. Conversely, real underlying illnesses - eg real, as opposed to diagnostically fashionable and 'spectrumised', bipolar disorder - is much easier to diagnose with confidence in an abstinent patient when the dust has settled somewhat. So prescribing antidepressants on admission is not only a waste of not very effective medication and a source of needless side effects but can actually obscure the aetiology, assessment and diagnosis of underlying conditions and lead to inappropriate long-term prescribing.

Thirdly, most trials of antidepressants in alcoholism (and heroin abuse) don't seem to have notably positive outcomes compared with appropriate placebos or energetic non-pharmacological alternatives. I would put in a good word for supervised disulfiram as part of the long-term management programme if patients find it impossible to abstain or drink moderately for long enough to benefit from the change but I've probably offended enough people and therapeutic orthodoxies as it is.

Thanks for the answers of all of you. It is true that the efficacy of antidepressants is lowered in this special population, however when we read the results from clinical trials we should be very careful with the treatment duration. In this population often an onset of antidepressants is prolonged and therefore these results are underestimated. The next story is psychopharmacology, which is very important when we select appropriate antidepressants in these patients. Often an antidepressant with a small dopamine activity could be chosen first (e.g. bupropion). The next important issue is biological psychiatry and an impact on hippocampus. It is known that antidepressants can have an impact on higher volume of hippocampus, which is also thinner in patients with alcohol dependence syndrome. The next issue is epidemiology, because ADHD is seen in comorbidity with at least 30 % patients with alcohol dependence syndrome and is often undertreated. Therefore appropriate medications such as bupropion and atomoxetine can be used in these patients (stimulants should be used with a great caution in these patients although their big effect size). The most important think in drug efficacy is drugs' effect sizes and in these patients antidepressants' efficacy are often small-moderate (g=0,25-0,5), however medications for ADHD are more efficacious (moderate to big effect sizes). Many alcohol dependence syndrome are treated with antidepressants, because the symptoms of depression are often similar than those with ADHD, anxiety disorders and bipolar depression, which is seen especially within primary health system and especially then when an efficacy of antidepressants is very small. If I conclude, this topic is clinically very important and is not easy to make a clear conclusions. However, before a 'non-clear' prescribing of antidepressants with small effect sizes to these patients, a brief different diagnosis should be done to exclude several important comorbidities (e.g. ADHD, MDD, anxiety disorders). On another side, after unsuccessful and long-term trial with antidepressants (e.g. 8-12 weeks) a drug discontinuation should be considered to avoid inappropriate prescribing and polypharmacy, which can also have a negative impact on clinical outcomes.

I think the prevalence of ADHD among 30 % of alcohol -dependent patients is magnified by misclassification. The prevalence of anxiety disorders ( social phobia , generalized anxiety ... )  is greater than that of ADHD between alcohol dependents; so that  activator drugs such as bupropion could worsen anxiety in patients who drink just to calm their baseline symptoms.

http://www.ncbi.nlm.nih.gov/pubmed/26269096

You are right! In patients with anxiety symptoms (comorbidity) prescribing of activating antidepressants should be avoided (e.g. bupropion) and also among SSRI it is better to avoid fluoxetine and sertralin (perhaps escitalopram is a good option). Also because many patients have elevated liver enzymes a use of sertraline and bupropion should be avoided. Another story is a titration process, which should be even smaller if there are symptoms of anxiety (and very important is which BDZ is used if should be used; less activating much better). As clinical pharmacist specialist within psychiatry i have made many interventions within this field. A brief different diagnosis should be done to exclude several important comorbidities (e.g. ADHD, MDD, anxiety disorders) before 'appropriate' prescribing of psychopharmaceuticals.

We can continue with this interesting discussion. Which benzodiazepine you use in these patients and which dose and for how long?

The relationship between alcohol and depression is bidirectional; alcohol as we all know is a cerebral depressant and patients with depression self medicate with alcohol.

As Austin mentioned, we have to be very careful about the timing of mental state examination to look for mood symptoms in previously undiagnosed patients. Anyhow we have to see whether the previous diagnosis is accurate. The literature says the examination should be at least 2 weeks after detoxification.

Not treating depression will often result in relapse into alcohol consumption.

Other than prescribing antidepressants for the depressed alcohol dependent patients (who wish to quit or reduce drinking), psycho-social measures are of utmost importance in this patient category (whose life once revolved around alcohol) to prevent relapse to alcohol consumption and improving depression. Not addressing this aspect properly may be one factor for observed reduced efficacy of antidepressants in this patient group (and as Virginia mentioned unrecognized anxiety disorders)

VERY curious question I also very easy or difficult for simple answers. Why? Only be carried forward their observations as long Medicus and practitioner with patients which include this group of patients, depending on the alcohol?

 I think these patients should only initially begin to treat with antidepressants and help has to overcome the initial phase of abstinence of alcohol! Then develop symptoms of dependence: the tension, increased sweating ..tremor, insomnia, general irritability, loss of appetite,   forgetting .... etc.

Then it is necessary to give tranquilizers to neutralize these symptoms.!

My opinion and experience that they should only be used in the beginning of treatment and a limited time to the stabilization of the patient's condition!

 Otherwise, we can create another problem if you give more, because they can lead to habituation, to the creation of dependence on them ... (sedative).

But there is no easy solution to this problem because it should provide strong support to families of addicts, himself the natural and social environment where he is.!

Only then can decide whether to start COMPLEX TREATMENT OF DEPENDENCE ON ALCOHOL!

Dear colleagues Stuhec,

I liked your idea to make the study of the action of bendiazepama and that could help us to unify our doctrinal positions on the dose and length of treatment! But this should include a wide range of diseases.

Normally it depends on all etipathogenesis patients treated because it is not the same physical tension, fear and anxiety when given in low doses has quite a good effect, or some other disease such as mentioned herein syndrome of addiction to alcohol or drugs or the treatment of status epilepticus, tetanus, muscle spasms or premedication and induction of anesthesia, etc. where they need higher doses of the drug!

I think only associated clinical experience MEDIKUS (which I had to use the most) and terapeutic guides based on clinical trials can answer the second part of the question how long it should be given!

a friendly Greetings from Sarajevo

Jasenko Karamehic