TCPR: You’ve written that we should think about whether the medications we use are “disease modifying.” Tell us about that.

Dr. Ghaemi: Disease-modifying drugs address the underlying pathophysiology and improve the overall course of the illness. That’s in contrast to symptomatic treatments, which reduce symptoms temporarily but don’t change the course of the disease because they don’t address the underlying cause. This idea is common in other fields of medicine. For example, Tylenol is a symptomatic treatment for fever, but penicillin is disease modifying.

TCPR: Which psychiatric medications are disease modifying? 

Dr. Ghaemi: You could argue that lithium qualifies, and possibly other mood stabilizers like valproate, carbamazepine, and lamotrigine. Lithium normalizes the expression of the “CLOCK genes” that code for circadian rhythms and are strongly linked to manic depression (Mishra HK et al, Mol Psychiatry 2021;26(7):3383–3394). It also reverses the hyperexcitability seen in young nerve cells derived from patients with bipolar illness (Mertens J et al, Nature 2015;527(7576):95–99).

TCPR: You left antipsychotics off the list of “mood stabilizers.”

Dr. Ghaemi: Right. Antipsychotics (I prefer the term “dopamine blockers”) are not mood stabilizers because they have not been shown to prevent mood episodes. Now, there are trials where they look preventative, but those were designed in a way that favors the antipsychotic. They withdrew the medications too soon and used “enriched samples” of patients who already responded to them, which gives false-positive results.

TCPR: What about quetiapine? Putting aside the problem with study design, most antipsychotics only prevented mania—not depression—but quetiapine prevented both poles in its long-term studies. 

Dr. Ghaemi: That is true, but I don’t believe those results because the company won’t provide the actual data for researchers like me to analyze independently.

TCPR: How would you analyze it differently?

Dr. Ghaemi: I would be interested in the initial polarity (eg, mania vs depression), the polarity of relapse, and how long it took for the relapse to occur after stopping the drug. That’s how we can tell if a truly new episode has been prevented, as opposed to just withdrawal relapse into the same acute episode that the patient had a few weeks or months earlier. 

TCPR: What is different about lithium’s preventative effects?

Dr. Ghaemi: For one thing, lithium is the only medication that’s been proven to prevent suicide—completed and attempted—in randomized trials, and it prevents suicide in both unipolar and bipolar disorders. Clozapine is a close second. It prevented suicide attempts, but not completed suicides, in a randomized trial of schizophrenia. Lithium also prevents hospitalizations, even in unipolar depression. There was a large Finnish study that followed 123,712 unipolar patients for about eight years after hospitalization for depression. Lithium was the only med associated with a lower risk of rehospitalization (Tiihonen J et al, Lancet Psychiatry 2017;4(7):547–553).

TCPR: Not antidepressants?

Dr. Ghaemi: No. Antidepressants and antipsychotics were associated with higher rehospitalization rates. Even though this was a non-bipolar group, lithium’s preventative effects were greater when used on its own than when it was combined with an antidepressant (70% vs 50% rehospitalization rate). This study wasn’t randomized—each patient served as their own control—but we do know from 39 controlled trials that lithium prevents depression in unipolar disorders, and it has one of the largest preventative effects in bipolar disorders (Undurraga J et al, J Psychopharmacol 2019;33(2):167–176). But I don’t think of mood disorders in terms of bipolar and unipolar.

TCPR: What? You’ve developed diagnostic tools to separate bipolar from unipolar. 

Dr. Ghaemi: That was when I thought the DSM was valid. In 1980, DSM-III separated mood disorders into bipolar and unipolar, using “major depressive disorder” (MDD) for “unipolar.” For nearly a century before that change, psychiatrists lumped severe, recurrent mood disorders into a single category of “manic-depressive illness” regardless of whether the episodes were manic or depressive. 

TCPR: Tell us more about this idea of a single manic-depressive illness.

Dr. Ghaemi: In my view, which is the same as Kraepelin’s, mania and depression are part of one illness, and cycling of mood episodes is the core feature of this illness. Those episodes may be manic, depressive, or mixed (and mixed is actually the most common), and they may last three months, six months, or a year. They are severe. But the key feature is the cycling: They come and go every couple of years. Manic-depressive illness is a biological disease with genetic underpinnings, and it is best treated with lithium and mood stabilizers—especially for the long term, for prevention, but also in the short term in many people. 

TCPR: Why did DSM-III split this into bipolar and unipolar?

Dr. Ghaemi: The idea was that the two poles differed in their genetics and course of illness, but that has not held up so well. In genome-wide association studies, the genetics of bipolar and so-called MDD do not separate out. In terms of course of illness, that idea was based on data showing that bipolar tends to start at ages 15–20 while unipolar MDD begins around age 30. But that has also changed—at minimum, a lot of children have depression without manic episodes and receive a diagnosis of MDD. If the distinction is valid, then we should stop diagnosing MDD in all or most children. Also, many of those children with “depression” have family histories of bipolar illness, further blurring the bipolar vs unipolar distinction. The DSM also had pragmatic considerations, because it was convenient to say “We’ll create a category for tricyclics (MDD) and a second category for lithium (bipolar).” But many MDD patients respond to lithium.

TCPR: What did we lose in the transition from manic depression to bipolar-unipolar?

Dr. Ghaemi: The bipolar side shrank, while the unipolar “MDD” side widened. The bipolar criteria in DSM are very strict, which means that a lot of patients with manic symptoms that don’t meet the full DSM criteria for bipolar disorder will receive antidepressants. About one in four patients with MDD have manic symptoms (“mixed features”) during their depression (Vázquez GH et al, J Affect Disord 2018;225:756–760). 

TCPR: How would you classify depressions that don’t fit into your concept of “manic-depressive illness,” those that don’t cycle in and out of recurrent mood episodes?

Dr. Ghaemi: Outside of manic depression, I see three types: 1) neurotic depression, 2) vascular depression, and 3) existential depression. Melancholic and psychotic depression are also valid categories, but I would include them under manic-depressive illness. By “neurotic depression” I’m referring to patients with the neurotic temperament, not the Freudian sense of neurosis. Neuroticism is one of the best-validated temperaments. These patients tend to be anxious all the time, and they react more to stress than the average person. When stress is high, their symptoms might meet criteria for mild MDD for a few weeks or a few months at the longest. These patients are common in practice, and with today’s DSM they are usually diagnosed with MDD, sometimes with generalized anxiety disorder as well, and treated with SSRIs.

TCPR: How would you treat neurotic depression?

Dr. Ghaemi: That’s difficult. This is actually why there’s a vigorous debate about whether modern antidepressants even work. They do work, but mainly in severe depression. In mild depression, there is little separation from placebo, and most of these patients with neurotic depression have mild depression. This is something we didn’t realize until 20 years ago when the unpublished data were released, and it is why the UK’s NICE guidelines do not recommend antidepressants in mild depression (Fournier JC et al, JAMA 2010;303(1):47–53). So that leads me to suggest psychotherapy, or at least a therapeutic, supportive relationship for these patients. 

TCPR: Tell us about vascular depression.

Dr. Ghaemi: Vascular depression typically begins in midlife in patients with cardiovascular risks, like long-standing diabetes or hypertension. It is associated with white matter abnormalities on the MRI, so there is an actual biological marker that is diagnostic and useful. You have to read the full MRI report because the summary might read “normal for age” but the narrative will mention periventricular or deep white matter hyperintensities.

TCPR: If they haven’t had a brain MRI, what symptoms distinguish vascular depression?

Dr. Ghaemi: These patients may have more cognitive problems, like trouble with task completion, trouble with decision-making, and slow processing speed, but symptoms don’t distinguish vascular depression very well. Vascular depression can also overlap with manic depression, and it often does because vascular disease begins earlier and is much more common in this population. If we look at all cases of depression, about one in five have evidence of vascular depression after age 50, one in two after age 65, and nearly 100% by age 75 (Taylor WD et al, Am J Psychiatry 2018;175(12):1169–1175).

TCPR: How do you treat vascular depression?

Dr. Ghaemi: Vascular depression does not respond well to antidepressants (Aizenstein HJ et al, BMC Med 2016;14(1):161). The most important thing is to address the cause—make sure their metabolic and cardiovascular risks are being addressed, particularly hypertension. I would also consider low-dose lithium, like 150 mg/day, because these patients are at high risk for dementia and low-dose lithium has evidence to prevent that (see the September 2022 issue of The Carlat Psychiatry Report: “Low-Dose Lithium to Delay Dementia?” by James Phelps, MD). Lithium is neuroprotective, and it improved motor recovery after a stroke in a small randomized, placebo-controlled trial at a medium dose of 600 mg/day (Mohammadianinejad SE et al, Clin Neuropharmacol 2014;37(3):73–78). Just watch for drug interactions with lithium and antihypertensives.

TCPR: Tell us about your final category: existential depression.

Dr. Ghaemi: These are depressive states that are caused by problems that are common to human existence, such as life crises or transitions. These crises can include divorce, loss of a relationship, starting a new relationship, losing a job, getting a job, grief, death of a loved one, or one’s own medical illnesses. Existential depression can happen in manic-depressive illness, and often does—because the illness itself can rob people of a meaningful role in life. But I would reserve this fourth diagnosis for people who do not have the other three depressions I’ve discussed: manic depression, neurotic depression, and vascular depression. One way of defining existential depression is Viktor Frankl’s concept that it is a normal psychological response to abnormal experiences or environments. Another way is to see it as just part of normal life, of the “limit situations” of life, as Karl Jaspers put it. And psychotherapy is the best treatment—not medication.

TCPR: Let’s get back to your concept of manic depression. What is the role of antidepressants there?

Dr. Ghaemi: Antidepressants are not disease modifying, but they do give short-term symptomatic benefit for some, particularly those with more severe episodes who lack a history of manic or mixed symptoms. Even then, it is a pretty limited benefit. The problem is that they can worsen the course of manic-depressive illness, particularly in those with a history of rapid cycling, and cycling is the core feature of this illness (Ghaemi SN, J Clin Psychiatry 2021;82(1):19m13136). 

TCPR: How do you tell the difference between rapid cycling and someone with a neurotic temperament who keeps having major stress?

Dr. Ghaemi: Well, it does take some clinical nuance, and indeed patients with manic-depressive illness also have temperamental problems that lead them into these reactive depressions. About half of patients with manic ­depression are basically normal in between their mood episodes. The other half have mood temperaments like cyclothymia (chronic mild symptoms of mania and depression), dysthymia (chronic mild depressive symptoms), and hyperthymia (chronic mild hypomanic symptoms) (Vöhringer PA et al, J Affect Disord 2012;136(3):577–580). What separates these patients from neurotic depression is their family histories and the severity of their mood episodes. While some of their episodes are brief and mild, they usually have some more severe episodes lasting months. Whether or not they are triggered by life stressors is irrelevant to the diagnosis. They will usually have evidence of manic-depressive illness in close relatives, either of bipolar disorder or severe unipolar depression.

TCPR: Just to poke at this concept a little more, neuroticism is also heritable, so what if the patient says “My mom was often anxious and depressed”? How do we know if the mom had neuroticism or manic-depressive illness?

Dr. Ghaemi: Neuroticism is not as heritable as manic-depressive illness but is still about 50% genetic. In manic-depressive illness, that mother would be more likely to have had severe episodes that put her out of work or required hospitalization, but that can be difficult to tell without interviewing the relative.

TCPR: Thank you for your time, Dr. Ghaemi.