TCPR: Is psychotic depression difficult to diagnose?

Dr. Swartz: Psychotic depression is often missed and often misdiagnosed. A study of hospital units at four academic medical centers found that one in three cases of psychotic depression were missed (Rothschild AJ et al, J Clin Psychiatry 2008;69(8):1293–1296). This is important because these patients aren’t going to respond very well to common antidepressants like SSRIs, and they may not respond well to medication at all.

TCPR: What gets in the way of detecting psychotic depression?

Dr. Swartz: Number one is subjectivity. All of the symptoms—depression and hallucinations or delusions—depend on patients’ self-reports, and the DSM encourages a kind of bureaucratic accounting of them. But patients cannot reliably identify or describe their own symptoms, ­particularly in psychosis. Even in nonpsychotic disorders where self-awareness is relatively intact, like anxiety disorders, I find that patients are unable to recognize improvements that are visible to me in their mental status—as when they are more relaxed, no longer fidgeting, and no longer hyperventilating.

TCPR: How can we move beyond self-report?

Dr. Swartz: Mental status and behavior. In psychotic depression, you may notice a complete rigidity of thought or poverty of thought in the interview. You may pick up on hallucinatory or delusional behaviors through the relatives.

TCPR: How do you screen for psychotic depression?

Dr. Swartz: I ask, “Do you feel guilty about anything? Do you believe you’re sick, physically sick? Do you believe the symptoms you have are due to a physical illness? Do you think about death a lot, and what are your thoughts about it?” Nihilism, especially in elderly patients, is a common symptom of psychotic major depression with melancholic features. Patients believe they’re sick or dead or guilty of some horrible thing they can’t put into words.

TCPR: Suppose the patient answers, “Yeah, I feel sick. I feel empty. I feel like I may as well be dead.” How would you proceed?

Dr. Swartz: I would tease it out more and look for cognitive rigidity or behavioral evidence of delusions. “You feel sick, but are you sick? And what have you been doing to figure out the cause of the sickness?” Often patients do odd things to prove their sickness. They hound doctors. They collect urine or other bodily substances. I am trying to identify something out of the range of common thought and behavior, so if they feel guilty, I might ask, “What punishment do you deserve?” If the patient says, “I deserve death; I am dead,” that would be a psychotic answer.

TCPR: Which symptoms tend to be more prominent—the psychosis or the depression?

Dr. Swartz: It can go either way, and I’m glad you asked that because there are two common types of psychotic depression. I call them depression-dominant and psychosis-dominant. Both can be misdiagnosed. In the depression-dominant, the psychotic features may not be readily apparent, and the patient is often misdiagnosed as having major depression. In the psychosis-dominant, it’s the mood symptoms that need to be teased out. When they are missed, the patient is usually misdiagnosed with delusional disorder or schizophrenia, or the psychotic symptoms might be misunderstood as dissociation, conversion disorder, or PTSD.

TCPR: Can you describe a psychosis-dominant patient?

Dr. Swartz: A woman believed she was being electrocuted by people who were spying on her. She felt electric shocks in her body, and she complained about this obsessively, so it dominated all conversation with her. But once I reviewed the symptoms of major depression, it became clear she had them. She experienced no pleasure. Her sleep was disrupted. Her appetite and energy were low.

TCPR: What if she was just worn down by the intense delusions—too scared to sleep, too distracted to eat or enjoy anything?

Dr. Swartz: That’s psychological reasoning, which is valid in its place, but it doesn’t apply to psychiatric diagnosis. Signs and symptoms have to be assessed in their own right, regardless of what we may think about their cause or context.

TCPR: Tell us about a typical patient with depression-dominant psychosis.

Dr. Swartz: Here the depression is usually a melancholic type, and the psychosis usually presents as delusions with a depressive theme. Themes of sickness, guilt, and nihilism are common, as in an emaciated young man who quit eating because he believed his body could not metabolize food, or an elderly woman who spent two days in a bathtub believing she was too sick and weak to get out. Hallucinations are less common. Now, when I say “melancholic,” I’m talking about classical melancholic depression, and the DSM criteria do not capture the classical definition very well (Chelminski I et al, J Clin Psychiatry 2000;61(11):874–875).

TCPR: In DSM-5, melancholia means a total loss of pleasure along with early-morning awakening, low appetite, excessive guilt, and psychomotor changes. How is that different from classical melancholia?

Dr. Swartz: What I’m referring to is impairment of problem solving, inability to understand complexity, and visible apathetic behavior, including withdrawal from pleasurable activities. Their faces are unreactive—lacking in spontaneous expression—and they lack spontaneous thought. Sometimes they have a very interesting physical sign: the omega sign. This is the appearance of the Greek letter Ω between the eyes just above the nose.

TCPR: Some of that sounds like cognitive symptoms of psychosis: thought blocking, concrete and distorted thinking.

Dr. Swartz: To clarify, we often see those cognitive symptoms in psychotic patients, but they are not psychotic symptoms, and we see them in nonpsychotic disorders as well. In the past, some psychiatrists used the term “psychosis” to describe those cognitive symptoms, like the Maudsley group in the UK back in the 1970s, but that is not the mainstream view. Psychotic depression has delusions or hallucinations, not simply distorted or impaired cognition.

TCPR: How do those delusions and hallucinations differ from the psychosis we see in schizophrenia?

Dr. Swartz: Actually, schizophrenia can be difficult to distinguish from mood disorders. Every symptom that occurs in schizophrenia can be seen in mood disorders, including unusual mannerisms and inappropriate laughter. The difference is in the course. In mood disorders, the symptoms fully resolve, while in schizophrenia the course is chronic. How you define “chronic” is a matter of experience and judgment. Personally, I favor two to three years rather than six months.

TCPR: Why not six months?

Dr. Swartz: Because mood episodes often last that long or more. The natural history of psychotic and melancholic depression is about one year, and the average manic episode lasts six months if untreated or if nonresponsive to treatment. If these cases are misdiagnosed as schizophrenia and the psychiatrist prescribes an antipsychotic, then it’s likely the patient will stay on that drug for the long term. Few clinicians will stop the drug, because it’s just too risky if the patient carries a schizophrenia diagnosis.

TCPR: OK, so that might be one clue between the two. Does family history ever help you distinguish between schizophrenia and psychotic depression?

Dr. Swartz: I would have to say family history is not reliable. It distinguishes the presence of severe psychiatric disturbances, but it doesn’t help you identify which disturbance it is. I think the twin studies have shown remarkable discordance for identical twins where one has schizophrenia and the other one is seriously ill with a mood disorder and not schizophrenia.

TCPR: Another disorder with psychotic-like symptoms is PTSD: flashbacks, paranoia, and dreams that intrude on waking life. How do you distinguish that from psychotic depression?

Dr. Swartz: We do see psychotic symptoms in PTSD, in an estimated 15%–60% of cases, but I do not think PTSD is a psychotic illness. Mark Zimmerman’s group found that nearly all the psychotic symptoms in PTSD could be better explained by comorbidities that can cause psychotic symptoms, such as schizophrenia, bipolar, substance use, and borderline personality disorder (Gaudiano BA and Zimmerman M, Br J Psychiatry 2010;197(4):326–327). On the other hand, I often see PTSD symptoms in people with psychotic depression, and I think that’s something we need to pay attention to.

TCPR: Tell us about that.

Dr. Swartz: The experience of having a serious mental illness like psychosis is traumatic. Being hospitalized, stigmatized, disempowered, losing your family, losing your job, losing your identity. Childhood trauma is also very common in this population. Patients with psychotic depression are two or three times more likely to have experienced physical or sexual abuse in their childhood than those with nonpsychotic depression (Gaudiano BA and Zimmerman M, Acta Psychiatr Scand 2010;121(6):462–470).

TCPR: How do you address that?

Dr. Swartz: Psychotherapy and antianxiety medication. There is some evidence that psychotherapy helps psychotic depression, but I suspect that it’s treating these PTSD symptoms. Sometimes patients with psychotic depression tell me they are still depressed after a course of ECT, but it’s very different from the depression they came in with. It’s PTSD from the illness.

TCPR: Anything else we should look for when a patient has psychotic depression?

Dr. Swartz: Well, psychotic depression is a marker for bipolar disorder, so you’d want to look carefully for past manias. In the elderly, you should think about vascular depression, particularly if it’s their first episode. Vascular depressions are usually melancholic, though not necessarily psychotic. So, new-onset melancholia is a hint for the clinician to look for cardiac risk factors, blood pressure, and cholesterol. Medical morbidity is high in psychotic depression, and they have twice the rate of death compared with severe nonpsychotic depression (Vythilingam M et al, Am J Psychiatry 2003;160(3):574–576). Their risk of suicide is also much higher (Gaudiano BA et al, Depress Anxiety 2009;26(1):54–64).

TCPR: Why is that?

Dr. Swartz: These individuals are more disturbed, more hopeless, and more desperate than other patients. They can’t see past their delusions, and their judgment is more impaired.

TCPR: What’s your first-line treatment for psychotic depression?

Dr. Swartz: It’s pretty frustrating to treat psychotic depression with medications. Tricyclics are a good choice, and in women I’ll often augment with triiodothyronine (Cytomel) 25 mcg QD. There’s some evidence that thyroid augmentation works better in women, and I’ve seen good results. I also use bupropion (Wellbutrin). If the antidepressants do not work, I’ll add lithium, which has open-label evidence as augmentation in psychotic depression (Birkenhäger TK et al, J Clin Psychopharmacol 2009;29(5):513–515). But the treatment I’ve relied on the most is ECT (Petrides G et al, J ECT 2001;17(4):244–253); most of these patients are receptive and sign consent forms to receive ECT.

TCPR: The 2010 APA guidelines recommend lithium augmentation for psychotic depression, but they seem to favor antipsychotic augmentation.

Dr. Swartz: For me antipsychotics are second or third line.

TCPR: Why is that?

Dr. Swartz: If you read those guidelines closely, they admit the combination may be no better than an antidepressant alone. The only paper they cite in support of the combination is a small randomized controlled trial of 18 patients (Spiker DG et al, Am J Psychiatry 1985;142(4):430–436). The other papers they cite—which include two meta-analyses and a randomized controlled trial—actually concluded that antidepressant monotherapy was just as effective as the antipsychotic combination (Wijkstra J et al, Br J Psychiatry 2006;188:410–415). Now, I think antipsychotics can improve symptoms, particularly symptoms of depression, which they are FDA approved for in low doses. And I will use antipsychotics if there is a risk of violence or suicide and ECT is not available. But my concern is that antipsychotics don’t improve functioning and might even worsen it.

TCPR: What do you mean?

Dr. Swartz: They can cause frontal lobe syndromes: apathy, difficulty solving problems or dealing with complexity in relationships. These patients have trouble multitasking and taking initiative. They become dependent, passive, and quiet. We see these syndromes in case reports, and there are imaging studies showing a reduction in frontal lobe activity with antipsychotic use (Swartz C and Walder M, Ann Clin Psychiatry 1999;11(1):17–19; Cohen RM et al, Arch Gen Psychiatry 1997;54(5):481–486). It’s not something researchers look for in clinical studies, but some antipsychotic trials mention “personality changes,” and this is what I think they are talking about.

TCPR: I’m familiar with those side effects, but I thought they only happened at high doses.

Dr. Swartz: Yes, and the studies are pretty clear that to treat psychotic depression you need high doses of antipsychotics, so not the low doses you use in depression. We’re talking about the doses you’d use in acute schizophrenia. Anyway, frontal impairment is just one reason to avoid them. There’s also the metabolic and cardiovascular risks, which this population is already vulnerable to. Then there’s the possibility of tardive psychosis, which is where chronic antipsychotic use upregulates the D2 receptors and withdrawal of the antipsychotic then causes a rebound psychosis. There are case reports of psychosis developing in people after abrupt withdrawal of an antipsychotic, even in those who were taking metoclopramide and had no history of psychiatric problems (Lu ML et al, Ann Pharmacother 2002;36(9):1387–1390).

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