According to Thase M., et al., Treatment resistant depression (TRD) - "unipolar" one - could be staged using 0-5 levels.

Stage 0: "psudoresistance"=any drug accounted as "inadequate" to date. Stage1: failure to respond to an adequate trial (e.g. see the definition by Fava M. et al., 2003 for the adequacy of a trial with SSRIs in case of MDD) of a major antidepressant class (still SSRIs in modern days), stage2: failure to respond to 2 distinct trials with 2 distinct antidepressant classes. Stage3: stage2+failure to respond to an additional trial with a different antidepressant class (including TCAs). Stage4: stage3+failure to respond to an additional trial with MAO-Is. And Stage5: Stage 4+failure to ECT.

I suppose you were referring to this classical definitions. Yet the STAR*D levels and the classical Texas algorithms for TRD.

Personally, I would just be sure about the following: 1) actual compliance (but you talk about stage 5, so I submit this exists since the patient underwent at least an ECT trial), especially in the presence of a history of multiple failure, recurrences/chronic (DSM-5 should use the term "persistent") depressive episodes... 2) evaluate the HPA axis, with a special emphasis towards THYROID functioning and and consider boosting it and or adding folate or s-adenosinademetionine 3) be sure he/she can tolerate it, and introduce lithium, especially in case of familial predictors of bipolarity. 4) But the real answer is CONSIDER sub-threshold bipolarity (and that some issues as recurrence, stagionality, post-partum or even previous full-blown mania are usually neglected or go underscored by spontaneous reports of our patients!!!!!). Take advantage also of the MDQ or HCL-32 or similar tools. CONSIDER BIPOLARITY: also consider switching from traditional antidepressants to Second Generation Antipsychotics (especially olanzapine+fluoxetine combination or quetiapine RP or very low dose aripiprazole with lithium). 5) Last but not least, consider MIXED features, especially those the current DSM-5 would describe as NOS or NEC.

By the way: lithium should be used at THERAPEUTICALLY meaningful levels (not

I think that generally a resistant depression is due to an incorrect diagnosis or imperfect treatment.

it's now clear that there is no only one depression but there are at least three: anxious depression, asthenic depression and anhedonic depression. the first pharmacologically responds to a prevailing serotonergic antidepressants, the second one to prevailing noradrenergic antidepressant and the third one to those with prevailing dopaminergic action.

it should also be pointed out that a careful analysis of the depressive episode could highlight a type of depression in act secondary to a primary different one. For example, you may find a depressed patient with a strong component of anxiety as a result of a primary untreated anhedonic depression. This requires careful therapeutic decision according to the predominant in act symptoms: for example using before a serotonergic antidepressant to turn off anxiety symptoms and then switch to a dopaminergic antidepressant to treat the trigger anhedonic symptoms.

furthermore, we must remember the importance of adequate psychotherapy.

in principle once, fired two antidepressants and created the conditions for a treatment-resistant depression the association between a serotonine selective antidepressant with a new generation antipsychotic with a strong noradrenergic action at low doses provides great results. the good results with this strategy are also often due to an underestimated bipolar disorder.

In these types of scenarios, I usually start all over again, and revaluate the patient, preferably in an inpatient setting, where he/she can be directly observed. There needs to be a re-evaluation of all available data, and I, with my junior staff, usually do the following

1. File review. Yes all twenty or so volumes! Hopefully there are some students around. What medications have been tried, what doses, what duration were the trials. Equally important, what clearly had no benefit and what medications might have had a partial benefit.

2. Apart from medications, there may also be psychotherapy notes and/or correspondence. Were there previous courses of therapy and were they beneficial? Were they prematurely terminated, despite early gains?

3. Organic work up. Thyroid levels, especially as one option might be thyroid augmentation with run thyroid levels at the upper range on normal.

4. Consider co-morbidities such as bipolar disorder and a personality disorder. If elderly, dementia needs to be considered.

5. Scales to measure severity and benefits from interventions

After reviewing all the information there may be several options, though it does depend on factors such as ease of administration, patient preference, availability and cost. So, these aren’t necessarily in a pre-ordered preference, just options

1. Use the best antidepressant previously used, even though the response was only partial.

2. Consider another antidepressant even though others in the same class have been tried. For example a patient may have tried paroxetine, sertraline and citalopram. Would fluoxetine be an option? Remember there are several classes of tricyclics.

3. Then consider augmentation with thyroxine, T3, lithium, second generation antipsychotics, or methylphenidate (up to 60mg per day)

4. Lithium augmentation is usually at a low dose, but if there is a personal or family history of bipolarity, you may want to run lithium levels high, 0.8-1.2

5. Second generation antipsychotics such as quetiapine or olanzapine may be helpful irrespective of bipolarity

6. Sometimes I use methylphenidate or dexamphetamine as monotherapy

7. Combination antidepressants may be a bit heroic- such as a SNRI or SSRI, combined with mirtazapine

8. Reconsider another course of therapy, such as long term CBT, or even dynamic therapy. This is usually limited by availability and patient preference

9. What course of ECT was done? Were enough treatments given? Were the seizures satisfactory? And what type of electrode placement- bitemporal or unilateral? Ultra brief or brief pulse? Many patients undergo unilateral ECT with no response and then respond to bilateral ECT.

10. Novel neurostimulation techniques such as repetitive Transcranial Magnetic Stimulation, Direct Cranial Stimulation or Direct Brain Stimulation. However, access to these can be very limited.

Hope this helps

Bradley Ng

Sorry, forgot to answer the actual question! Chances of success overall, modest. But not zero!

I agree with Prof. Papagni in that TRD is probably due to an incorrect diagnosis and treatment, and with Prof. Fornaro in that, after assuring compliance and organic conditions, sub-threshold bipolar disorder may be behind this resistance.

In TRD patients many times I see Koukopoulos mixed state features:

Mixed Depression without psychomotor agitation diagnostic criteria,proposed by Koukopoulos et al. (21)

Along with Major Depression, at least three of the following symptoms must be

present:

1. Inner tension/agitation

2. Racing or crowded thoughts

3. Irritability or unprovoked feeling of rage

4. Absence of signs of retardation

5. Talkativeness

6. Dramatic description of suffering or frequent spells of weeping

7. Mood lability and marked emotional reactivity

8. Early insomnia

A. Koukopoulos, G. Sani. DSM-5 criteria for depression with mixed

features: a farewell to mixed depression. Acta Psychiatr Scand 2013: 1–13

The term TRD puts the blame on the patient for the treatments not working when the most parsimonious explanation for TRD is that the fault lies with the treatments.

Several recent papers show that the response rate to the commonly prescribed antidepressants is little better than 40% (see Fig 2 of the paper linked below for e.g.) - which is about the same rate seen with placebo in clinical trials.

The same applies to behavioural therapies and so the suggestion has to be made that all we're seeing with any treatment is a placebo effect

http://www.plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pone.0041778

I see some excellent contributions to this topic already, and forgive me if I leave someone out here, but Sergio's and Bradley's responses tie up a lot of territory that matches our experience. First, I do not see logic through the prism of Drug A or Drug B to overcome Stage 5 Depression--something is wrong with this picture when this is the question in such serious cases. What does psychotherapy show? Too often it has not been done. What body/mind factors of health have been found (substance abuse, food sensitivities, heavy metals, enviornmental toxins, caffeine/HFCS intake, nutrition deficiencies, hydration)? Even more often, these are forgotten and not examined closely. Misdiagnoses permeate the field, as does mistreatment with powerful drugs that merely fluff over symptoms but fail to address anything causal. So many vested interests have gone after the instant solution, while real therapy is not instant. The worse thing about using neurochemistry to try and resolve a complex problem is that it can unwittingly negate the other more therapeutic approaches that can help the individual back to homeostasis and healing. Thanks, all, for you fine input on this most important topic!

Thanks. All coments were useful. Any experience about augmentation wiith dopamonergic agonists?

The comments by all were very useful. Quetiapine and other antipsychotics work well when combined with antidepressants in treatment resistant depression. In fact this can be tried as 3rd line after combining antidepressants have not yielded desired results. Thorough organic workup, before, is necessary though...

Yes, Milena, we have helped a lot of patients untangle the mess created with dopamine agonists like Requip and Mirapex when used in a milieu of other side-effect laden meds that over time only plunged the patient deeper into new problems. Of course, we are are talking of the really sick ones, which would be the treatment resistant cases.

Some of the troubling side effects from these included psychosis, anorexia, syncope, narcolepsy (causing auto accidents), Raynaud's, pericardial effusion, etc. Often they were loaded up with addictions to the opioid pain killers, including high doses of morphine. We do not consider these good solutions to clinical depression, but create new problems on top of the old ones. Most of the ones we have examined also suffer from problems such as the CARB Syndrome (see Wilson, W. (2013). Autism and Diet: Is there a connection? North American Journal of Medicine and Science, 2013; 6(30):158-162).

If first, all causal factors (lifestyle, trauma, dehydration, nutrition, heavy metal, environment toxins, food additives, gluten, substance abuse, polypharmacy, etc.) are addressed the need for side-effect laden meds disappears. When we speak of "treatment resistance" I find that only applies when the treatments have little or nothing to do with the disease. In the event a simple pharmacological approach is effective, things will most certainly change over time, as the body finds ways to get around the effects and the side effects finally begin. It is long term, life-changing approaches that address underlying causes and drivers of such disease that are really sustainable in my opinion.

You are all a hard act to follow! I have had success with reducing dosages of any SSRI that is being taken. Sometimes less is more.

I also increase the dose of self care, family involvement and shared responsibility . Often non-compliant.

If you have tried several antidepressant + psychotherapy as I understand:

1. revise diagnosis. Consider bipolar disorder, personality disorder and alcohol/substance.

2. check lack of compliance. More frequent than we think.

3. I have had some good surprise by stopping all medication with a couple of patients combined with physical activity.

4. otherwise, try potentialisation with:

- lithium, can have incredible effect. My experience is it works quickly or not. Tolerance may be an obstacle though.

- pramipexole, no great experience yet. Some small data in literature.

- T3. Interesting.

- Aripiprazole up to 5mg. Can be very interesting.

- psychostimulants for motor retardation and fatigue.

5. rTMS: some improvement in one of my patient even after ECT, lithium and many other trials. Worth trying.

Our experience is similar to Lisa's (keeping in mind that these drugs are threshold drugs and reductions may effectively be reduced to placebo effect).

Another consideration on the SSRI front: SSRI-induced Indifference. We are seeing this when patients come from other clinics loaded down with Buprion and 2 or more SSRI/SNRIs--bipolar behaviors, dramatic shifts in mood that did not exist before treatment began.

A good primer on the topic:

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2989833/

I see there have been posted a lot of remarkable comments, mostly focusing on the pharmacological management. Unfortunately, "stage 5" TRD cases would already failed most combinations (2 or more antidepressants from the same class) or augmentations (a conventional antidepressant plus an non conventional one; e.g. lithium, T3, pindolol, second generations antipsychotics [SGAs], ketamine, or else). I see that most of the comments here were made upon pharmacodynamics considerations, which is fine. But almost invariably, they do not account for the "stage 5" condition, which as I stated, implies most alternatives to be already failed (including bupropion and more). Also, TRD may include very heterogeneous clinical pictures (e.g. atypical presentations, severe anxiety or other multiple comorbidities, which apart from psychopathological considerations which further point out towards soft or hard bipolarity and mixed specifiers, also require flexible algorithms: just to mention bupropion, this failed to provide any robust additional benefit on stage >3 TRD cases of MDD, or it just alleviated the emotional blunting induced by the high doses of SSRIs (not a good idea to to manage the side effects of an ineffective drug by adding an not effective more). Also, one should bare in mind that while the meta-analysis by Shelton and the one by Papakostas, did not provide any conclusive evidence in support of SGAs augmentations (Shelton) or for switching antidepressants (Papakostas), yet SGAs augmentations provided slightly favorable results compared to antidepressants switching (should this indicate a sort of ex adiuvantibus confirmation of a sub-threshold bipolarity or mixed features, at least in some cases finally going to respond?). By the way, the meta-analysis by Papakostas showed a slightly favorable profile for across class vs. within class switching... But virtually NONE of the studies covered so far by such meta-analyses included STAGE 5 TRD cases... This is a core issue most people is missing here... Stage 5.... So this is why psychopathological considerations may prevail towards mere pharmacodynamic ones at this stage... Finally, TRD is not just a MDD outcome, as it may regard also some BD depressed patients (either type I or type 2), and while there is no conclusive guideline for such cases (as for unipolar TRD), one may just refer to algorithms which are unfortunately made based on the number of previous failed trials (!!!), not based on pharmacological or psychopathological rationales!!! For example, the STAR-D study arbitrary started by citalopram just because it was the cheapest, most broadly prescribed antidepressant of the most commonly prescribed class (the SSRI one) at protocol approval date - about 10 years ago... So once more, when talking of STAGE 5 TRD, one should not mess with "other/previous???" than 5 cases. Noteworthy, the best and most recent algorithm proposed for BIPOLAR TRD depressions (please refer to Pacchiarotti et al in Acta Psych Scand) shares "ECT" as ultimate step, but indicates quetiapine or lithium or lamotrigine as first choices. The same alternatives that most clinicians account for stage FIVE "unipolar" TRDs nowadays (despite the evident publication bias: absence of evidence does not necessarily mean evidence of absence in support or discharge of a given clinical practice), just because they are not accounted in previous steps... So whenever HIGH recurrence (no matter the predominant or sole depressive polarity), RAPID POP-UP responses on antidepressants - or else, especially NMDA antagonists - (actually too rapid responses/remissions to represent clinically meaningful outcomes) occur, or mood switches and INCREASED ENERGY OR SUICIDAL IDEATION do, sub-threshold bipolarity or mixed features may underpin a stage 5 TRD presentation actually being the "dark" side of bipolarity, and I would recommend to avoid any further harm with additional trials with standard antidepressants because they already failed to provide any benefit at stage 5. Quetiapine XR, Symbiax and lurasidone, with or without valproate, carbamazepine or lithium will do the job in most stage 5 cases (thyroid functioning should also be monitored or even slightly boosted if tolerated) in my clinical experience with refractory TRD "DSM-defined unipolars".

It is comfortable for me to read that Lisa noted that some TRD cases benefited from a dose REDUCTION of the SSRIs, or even stopping of standard antidepressants (Fabrice): often TRD is (also) a iatrogenic outcome, since what Klerman called "the antidepressant view of the world" is still very popular among many clinicians struggling with (too many?) TRD "unipolars". Real stage 5 TRD are also difficult to observe and to steady follow-up: they really must trust in their psychiatrists to stay on so many "adequate", repetitive, trials with different antidepressants and also accept to undergo ECT... In most of the cases they are actually not (fully?) adherent (how to blame them...) or they already committed suicide (I know this should sound provocative, but in a minor number of these cases too much adherence is hazardous even if they are unaware of this obviously...). Also, those receiving an adequate number of bifrontotemporal ECT sessions usually respond to that nonetheless, especially considering that in some cases they are required to stop their current pharmacotherapy (a confounding bias in the interpretation of what was really effective???... Good for them in any case, bad for their doctors if this was an unaware serendipitous outcome...).

I and my collaborators just submitted a RCT on STAGE 3 TRD with atypical features receiving duloxetine plus bupropion vs. duloxetine plus placebo: definitely, this "superiority" trial failed to provide any support for adding the NDRI drug to the SNRI one. This was a very "special" TRD population... So just imagine how unlikely it would be to expect any additional benefit for stage FIVE "unipolar" TRD cases... at least perseverating with standard antidepressants either as monotherapy or combination attempts...

It was stage FIVE in this forum... The (few) evidence for DA-agonists for such refractory cases says it is not a worth option...

Michele Fornaro, MD, PhD.

Maybe I missed it, but I don't see any mention of Ketamine in the discussion.