There has been a lot of talk in psychiatric literature over the past 20 or so years about “treatment resistant depression” (TRD) and finding the “right” antidepressant. However, as you will see below, this concept may be an artifact of how “depression” is defined in the DSM. In this blog, I propose that TRD may be distracting doctors away from a re-evaluation of the diagnostic process which needs to be determined before prescribing. 

The current definition of Major Depressive Disorder (MDD) created controversy since its inclusion in 1980. MDD as it is broadly defined, can not be considered a disease, i.e. it does not exist in nature with proven genetics like the historically validated disease entity of Manic-Depressive Illness (of which bipolar is a smaller subset). As a wise psychiatrist once said “your treatment is only as good as your diagnosis.”  

 In the DSM-I (1952) and the DSM-II (1968) there were essentially two categories of depression: Manic-Depressive Illness (biologic etiology) and Depressive Neurosis (Freudian etiology). Robert Spitzer MD, the driving force behind the revolutionary changes in the DSM-III still with us today, proposed in 1975 new categories of depression, Major and Minor, and also distinctions of unipolar and bipolar espoused by an east German Psychiatrist Karl Leonhard, in the early 1950’s based on preliminary family history studies which did not pan out in later studies. At this stage of the process, Spitzer was following the lead of the biologically oriented researchers at Washington University in St. Louis and Leonhard.  The minor category was intended to be atheoretical, as a substitute for the beloved Depressive Neurosis of the Freudian psychoanalytic hegemony of psychiatrists of the day. 

[side note: Leonhard decided to break with Kraepelin and promote polarity, i.e., unipolar & bipolar, superseding Kraepelin’s basic concept of recurrence for depression or mania. Long term family genetic studies do not support the bipolar-unipolar distinction as a basis for separate disorders but do support Kraepelin’s original comprehensive diseased entity of Manic-Depressive Illness. Ghaemi N. comprehensive Psychopharmacology 2019]

 The new term of Major Depressive Disorder was created by Spitzer to assuage the ICD-9 committee and the Freudians by dropping the term unipolar which some implied a biological origin.  

Originally, Spitzer had promoted the terms Major Depressive Disorder & Minor Depressive disorder, but due to concerns that anything called minor would not be reimbursed by insurance companies it was eliminated. He then promoted Major Depressive Disorder as a separate stand-alone diagnostic category which included severe and non-severe, (i.e., minor depressions). But to mollify the outrage of the psychoanalysists, he found the term dysthymia (dys-bad or difficult & thymia or mood) in a dictionary as a substitute for Depressive Neurosis which was listed in parenthesis in the final version of DSM-III. Again, more sausage making than science. 

[side note: Spitzer apparently did not realize that dysthymia, i.e., mild chronic depression as part of personality, had already been described by Kraepelin and Kretschmer in the late 19th century as one of the three basic mood temperaments:  hyperthymia, cyclothymia and dysthymia. However, the mood temperaments did not cross the Atlantic to the U.S. as readily as did Freudian theories.]

In Ed Shorter’s history, Before Prozac documenting of the machinations in the development of new DSM-III category of “MDD ”, he quotes Donald Klein MD, 1928-2019, (a prominent psychiatric researcher and figure of that era) on Spitzer’s decision: “I think Dr. Spitzers [sic] suggestion fails on every account. First it is clearly a response to political pressure, rather than a conceptual advance…embarrassingly transparent.” 

 “To respond to this sort of unscientific and illogical, but sociologically understandable pressure in the fashion that Dr. Spitzer suggests, is unworthy of scientists who are attempting to advance our field via clarification and reliable definition.”

I agree with Dr. Klein here. Political pressure does not lead to good science or a medical model of classification. 

The term unipolar referred to severe (non-neurotic depressions), and was classified under Manic Depressive Illness prior to DSM-III in 1980. 

Historically, before 1980, the psychiatric world accepted that there were only two types of depression. The terms endogenous and exogenous depression were used briefly in the 1970’s but were abandoned since the creators DSM-III thought those terms implied a biological vs. non-biological etiology,  and they wanted to be “atheoretical. ” Endogenous was associated with the Manic-Depressive Illness category i.e. recurrent or episodic depression, sometimes with mania, and exogenous with the Depressive Neurosis category. 

The consequence of the overlapping criteria between broadly defined MDD and the newly created Dysthymia (i.e., chronic “minor” depression) skewed the treatment strongly in favor of antidepressants with an increase of MDD diagnosis that doubled from 3.3% in the 1990’s to 7.1% in 2001-2. 

The severe depression of classic melancholia (psychomotor retardation, little or no reactivity of mood and profound anhedonia), described from antiquity, has gotten lost in the DSM grab bag of MDD. It responds best to the classic tricyclic antidepressants or MAOI’S (monoamine oxidase inhibitors) and preferably to ECT but it does not respond well, if at all, to SSRI’s. This type of depression can occur with or without any history of mania, i.e., it is part of the full disease entity of Manic-Depressive Illness. 

Nassir Ghaemi MD, in his Advanced Psychopathology CME course makes the distinction between depression as disease (recurrent severe episodes depressive or manic and depressive episodes, aka Manic-Depressive Illness) and depression as not a disease (continuous mild depression and anxiety formerly called depressive neurosis). DSM criteria does not require recurrence to quality for diagnosis of MDD. To require recurrence would imply a concept of disease which the DSM wanted to avoid so it requires only current symptoms and a two-week time frame.

The release of Prozac (fluoxetine) in 1989 was perfect timing for the amorphous MDD category.  A senior Harvard psychopharmacologist, Ross Baldessarini, is quoted as saying, “The whole pharmaceutical industry is premised on broad definitions that allow big markets. It’s hard to reverse that trend. For example, some people have argued that the SSRI’s could have just as well have been developed as anti-anxiety drugs and nobody would have known the difference. (See Shorter’s Before Prozac in resources). Per Shorter: DSM defined “MDD” doesn’t exist in nature. 

 Bottom line MDD is not a true disease entity:  it was created by compromise rather than empirical studies. That’s why the kappa scores (agreement between doctors) for it are lower now than in 1980 and why we have close to 30, mostly unnecessary antidepressants on the market. 

Do we really need over 30 antidepressants?

(We do not have or need 30 cholesterol or hypertensive medications)

1. citalopram (Celexa)
2. escitalopram (Lexapro)
3. fluoxetine (Prozac, Sarafem, Selfemra, Prozac Weekly)
4. fluvoxamine (Luvox)
5. paroxetine (Paxil, Paxil CR, Pexeva)
6. sertraline (Zoloft)
7. vortioxetine (Trintellix, formerly known as Brintellix)
8. vilazodone (Viibryd)
9. duloxetine (Cymbalta)
10. venlafaxine (Effexor)
11. desvenlafaxine (Pristiq, Khedezla)
12.  levomilnacipran (Fetzima)
13.  amitriptyline (Elavil and Endep are discontinued brands in the US)
14.  amoxapine
15.  clomipramine (Anafranil)
16.  desipramine (Norpramin)
17.  doxepin (Sinequan and Adapin are discontinued brands in the US)
18.  imipramine (Tofranil)
19.  nortriptyline (Pamelor; Aventyl is a discontinued brand in the US)
20.  protriptyline (Vivactil)
21.  trimipramine (Surmontil)
22.  mirtazapine (Remeron)
23.  bupropion (Wellbutrin)
24.  trazodone, (Desyrel)
25.  trazodone extended release tablets (Oleptro)
26.  vilazodone (Viibryd)
27.  isocarboxazid (Marplan)
28.  phenelzine (Nardil)
29.  selegiline (Emsam), which comes as a transdermal patch
30.  tranylcypromine (Parnate)
    
    

This plethora of medicines in part the result of the DSM’s broad definition of depression called Major Depressive disorder (MDD) from 1980 to the present culminating in the so called “drug wars” of the 1990’s and early 2000’s introducing many unnecessary “me too” drugs, mostly SSRI’s. Except for the tricyclics and MAOI’s from the 1950’s, over half the above list was introduced since the late 1980’s. Many of the senior experts in psychopharmacology today believe the pre-1980 antidepressants, TCA’s and MAOI’s are more effective than SSRI’s for severe, recurrent depressive episodes. 

The post 1990 additions (“me too” drugs) are considered to be more effective for milder non-episodic depression with anxiety. Studies show psychotherapy is equally effective for these milder chronic forms of depressive and anxious symptoms when compared to the SSRI medications. 

Fact of Interest: Zoloft (sertraline) was FDA approved in 1999 on the basis of only outpatient studies because it failed studies for more severe depression in the hospital,  i.e. it was less effective than the tricyclic antidepressants like Imipramine  and amitriptyline or the monoamine oxidase inhibitors or ECT for hospitalized patients.  

Part II – Clinical Aspects

Thinking beyond the “MDD” as defined  in the DSM:

1. Neither Psychiatric nor primary providers are trained to know when not to give an antidepressant and to first ask: “what type of depression does this person have”?

    

2. The vast majority of bipolar patients 70-80%, whether they have been previously diagnosed or not, present to the provider in the depressed state. Without specific questioning to elicit prior subsyndromal or syndromal hypomanic symptoms or full manic episodes, that patient usually ends up with an antidepressant rather than a mood stabilizer and the results are usually less than beneficial.

    

3. Kraepelin’s original research over 100 years ago has been confirmed by modern research showing that 60% of all mood disorders are of the mixed type which worsen with antidepressants and only 20% are either pure  depression or pure mania. So up to 60% of patients are not likely to improve on antidepressants, usually SSRI’s.

    

4. This exclusive focus by the mental health industry on a broad and vaguely “depression” has resulted in the use of the overuse PHQ-9 (Patient Health Questionnaire for depression) which preordains the outcome of any depressive complaints and virtually guarantees the patient will receive an antidepressant since there will be no consideration of subsyndromal or syndromal hypomanic symptoms or the more common mixed symptoms.

    

5. The DSM requires only symptoms for a diagnosis and omits the classic diagnostic validators of illness course, i.e., when did it start, are there distinct episodes with intervals and length of each, family history (grandparents, great grandparents, siblings and second degree relatives), response to treatment, and biological markers.

    

6. For Bipolar depression:  Findings from STEP-BD studies indicate that antidepressants should be avoided in most, if not all, cases when treating acute depression in patients with mixed or bipolar disorder.

    

7. Rather, one should choose from among 6 medications with a reasonable evidence-base for effectiveness and safety in treating or at least preventing bipolar depressions, namely: lithium, quetiapine, lamotrigine, lurasidone, and cariprazine, lumateperone (approved in 2022) or combinations of these. 

Is there depression without mania or mixed features?

(From Clinical Psychopharmacology Nassir Ghaemi – 2019)

• • 50% of all DSM defined MDD persons have mild or greater manic symptoms including agitation and irritability, i.e., mixed states (multiple researchers over 20 years)
    
    
  • Combined with 15% of classic manic-depressive cycles in bipolar illness, 65% of all depressive episodes so far have a manic/agitated core. 
    
    
  • Other studies show that hyperthymic or cyclothymic temperaments are present in a third of the 35% of remaining persons with depressive episodes (about 12%).
    
    
  • We have now explained 77% of all persons traditionally diagnosed with severe clinical depressive episodes. (50% + 15% +12%) about 4/5 of such persons. 
    
    
  • This leaves only 23% of people with a pure unipolar Major Depression who may benefit from antidepressant medication.
    
    

So, 77 % will not be helped with antidepressants because their symptoms may worsen, suicidal thoughts may increase or the medicine will simply do nothing. 

So how can practitioners find this 23%,  since antidepressants are not recommended for 77% of all patients who present with depressive complaints? 

It is my experience that patients can talk at length about feeling depressed but rarely will report mild manic or the validated mixed symptoms of irritability/rage, agitation or distractibility. 

The following quotes are from Goodwin and Jamison (see references):

• “False unipolar depression, sometimes called pseudo-unipolar depression characterized by non-euphoric psychomotor agitation, irritability and distractibility, i.e., mixed states, occurs in at least 50% of supposed MDD (higher on inpatient) since it is well known that patients rarely volunteer these symptoms but will call themselves “depressed” because they feel unhappy.  Relatives however do report these symptoms as well as subsyndromal hypomanic symptoms.” 
  
  
• “Patients cannot always distinguish the hypomanic state, since they may not view it as behaviorally different from their usual personality—a phenomenon recognized by DSM suggesting a 3rd party reports for the diagnosis of hypomania.”
  
  
• “The patient’s self-report is not sufficient for a dependable denial of past hypomanic or manic episodes, due to the problem of lack of insight, considering any mood that is not depressed as “normal”. 

• “Also, patients are often in such psychic pain during a depressive episode that they can be unmotivated to search their memory carefully for times when they may have felt better (albeit pathologically).” 

• “It is strongly recommended by many experts that before a patient be given a diagnosis of “MDD”, i.e., pure unipolar depression, a close family member or spouse needs to be contacted to elicit subsyndromal or syndromal hypomanic or manic and/or mixed symptoms, if any before prescribing antidepressants.”

I tell patients we need to choose the right “antibiotic” because depression is not simply a depression, just as an infection is not just an infection, what type is what matters for the best outcome. All infections are not the same. 

Without a familiarity of the research over the past 25 years on this issue, practitioners will go on a quixotic search for the “right” antidepressant with multiple medication trials, so-called “treatment resistant depression, TRD”.

Rather, the search should be for the right diagnosis before prescribing antidepressants.  

Using only the PHQ-9 is like seeing only one side of a coin or never seeing the dark side of the moon. I believe it should always be paired with the instruments listed below. (There are others but these have been around the longest).

• Validated and widely used for over 20 years to elicit hypomanic symptoms is the self-administered Mood Disorders Questionnaire for adults (Hirschfeld, Robert M.A., M.D. et.al. “Development and Validation of a Screening Instrument for Bipolar Spectrum Disorder: The Mood Disorder Questionnaire.” American Journal of Psychiatry 157:11 (November 2000) 1873-1875)

   

• Mood Disorder Questionnaire for Parents of Adolescents (Wagner et al., Journal of Clinical Psychiatry 2006, 67(5): 827-30).

No questionnaire can “make a diagnosis.” Other diagnostic validators such as family history of parents, siblings, grandparents, aunts, uncles and cousins need to be ascertained. In my experience, asking more questions can frequently help people remember.  If they don’t know their diagnosis then ask,

• “did anyone have erratic or irritable moods or make impulsive choices? 
• did you hear any stories about your relatives, did anyone cause a problem in the family?
• did anyone isolate or stay in bed for long periods?

Quite often when I review charts, if a family history is even recorded, the eval will say “non-contributory” or “negative.” which usually means the evaluator did not ask. 

Course of moods also needs of be obtained, i.e.:

• when did you first feel depressed or had suicidal thoughts in your life?
• what was that like, and how long did it last?
• how often have you had episodes like that?
  
  

It has been known for many decades that patients with pure,  episodic, unipolar, “major” depressions usually have a first episode in late 20’s to early 30’s. But a depression related to a bipolar illness most often begins in late teens or the early 20’s.

Another classic diagnostic validator is Response to Treatment, although it is considered to be the weakest and nonspecific. However if a patient says they or a relative had a good response to lithium, that is a helpful sign.  If a patient has undiagnosed Mixed Features (DMX) or history of past hypomania (i.e mild manic symptoms that often go unremembered):

1. Antidepressants do nothing –most common outcome (patients often say “it stopped working”)
2.  2nd most common outcome: Antidepressants may cause destabilization or instability manifesting at dysphoria, anxiety, irritability, volatility and impulsivity that does not fit neatly into hypomania.
3. Antidepressants most often induce “mixity” rather than hypomania.
4. They may increase suicidal ideation and/or attempts
5. Lastly, they may precipixtate a full manic episode.
   
   

This is why we don’t want to give the wrong “antibiotic.” The patient’s history of response is a clue as to what the diagnosis may not be, (i.e., a pure unipolar depression), but is not diagnostic in itself. 

The majority of patients seen by PCPs, and all mental health professionals who do not have any history on close examination of mild manic symptoms, mixed depressive states or mood temperaments of Hyperthymia or Cyclothymia, present with relatively mild degrees of depression and anxiety related to life stressors and/or well documented personality traits of anxiousness and depressivity.  (See pages 761-781 of DSM-5) 

This was called Depressive Neurosis or Neurotic Depression pre-DSM-III in 1980 until the word neurosis was deleted from DSM-III-R in 1987. Many experts in the field believe this category should not have been deleted in spite of its Freudian origins, since it represents a preponderance of depressive presentations. 

These patients have been shown to respond equally well to psychotherapy as they do to antidepressants, usually SSRI meds. In fact, there seems to be a consensus amongst prominent leaders in the field that the SSRI’s are really better for mild degrees of depression and anxiety rather than severe melancholic depressions that respond better to the tricyclic antidepressants, MAOI’s or ECT (no SSRI has ever surpassed the efficacy of these treatments). 

Course

In patients with severe, recurrent melancholic depression (psychomotor slowing, little or no reactivity of mood, and profound anhedonia), it usually begins around age 30.  The typical duration of the episode is about  6-12 months  (according to data from the pre-antidepressant era) and may occur once a year or once every few years. 

Bipolar depressions, by contrast, usually last 3-6 months, and usually begin in the late teens or early 20’s.