Mixed depression: a farewell to differential diagnosis?

Abstract

J Clin Psychiatry 2015;76(3):e378–e380 (doi:10.4088/JCP.14com09578). © Copyright 2015 Physicians Postgraduate Press, Inc. I this issue of the Journal, Perugi and colleagues1 delve into the murky and controversial diagnostic boundaries between unipolar and bipolar mood disorders. As the authors intimate, the decision by DSM-5’s crafters to loosen the operational definition of mixed episodes adds to ongoing debate over the beleaguered document’s various symptom tweaks and diagnostic reshufflings. DSM-5 rightly acknowledged that many depressed bipolar patients experience subthreshold mania symptoms while depressed and that simultaneous symptoms of both poles can and often do coexist in bipolar II as well as bipolar I mood states. DSM-5 also took notice of empirical findings that high energy states robustly differentiate bipolar from unipolar disorder,2 underscoring the principle that bipolar illness is as much a disorder of energy and psychomotor functioning as mood. But the shortcomings of DSM-5’s revisions involving “mixedness” are important and risk dwarfing these nosologic advances. DSM-5’s decision to apply the “mixed features” specifier to both unipolar and bipolar depressive episodes when subsyndromal mania symptoms also exist harkens to Kraepelin’s view that polarity was not an all-or-none phenomenon and was less important than other characteristics (such as excitation-agitation, or high recurrence) in his original conception of manic-depressive illness. In practice, though, efforts to articulate valid operational criteria for “mixedness” create their own problems—particularly when asking diagnosticians to cherry-pick only some “possible” mania symptoms, side-stepping those that could also reflect depression. Researchers who conduct criteria-based diagnostic interviews may thus agonize all the more over ambiguous cases, while time-pressured everyday clinicians will probably ignore such nuanced distinctions altogether and potentially make “mixed features” (never mind polarity) a wastebasket descriptor for any and all forms of irritable moody people. Throughout its editions and revisions, the DSM has included as its final criterion for virtually every described entity the proviso that observed symptoms should not be attributed to a particular disorder if they can be better accounted for by another condition. This caveat becomes especially difficult when evaluating diagnostically nonspecific symptoms such as irritability or mood instability (or other such broad mental phenomena as “inattentiveness” or “apathy”). When considering depressive mixed states, Perugi et al1 (and other authors3) confront us with the challenge of how best to classify an important nosologic entity without “double counting” symptoms, or conflating 2 or more disease states that share common elements. The BRIDGE data1 invite us to consider whether depressed patients with such diagnostically nonspecific symptoms as irritability, mood instability, distractibility, agitation, and anxiety belong to the bipolar spectrum, or whether they may be better accounted for by other nonbipolar comorbidities. In doing so, they draw our attention to the DSM-5 conundrum of whether to include or exclude transdiagnostic symptoms when characterizing a distinct clinical entity. (By analogy, should a newly recognized inflammatory disease exclude fever as part of its core symptom constellation in order to avoid conflation with a comorbid infectious process? Or, while dyspnea and chest pain do not help to differentiate angina from pneumonia, can those symptoms really be ignored when considering either diagnosis?) If phenomenology studies in psychiatry have taught us anything over the past century, it is that no psychiatric symptom is pathognomonic of any one disorder. Bizarre thought content no longer figures so prominently as it once did in considering a diagnosis of schizophrenia. Nor does catatonia, paranoia, or even Schneiderian first-rank symptoms. In the case of mood disorders, while “mixedness” is no doubt a real phenomenon, its “truest” definition remains messy. Prominent irritability, for example, was found in almost half of unipolar depressed patients in the National Institute of Mental Health (NIMH) Sequenced Treatment Alternatives to Relieve Depression (STAR*D) trial, and was more closely related to anxiety and depression severity than to occult “bipolar spectrum” features.4 Similarly, in the NIMH Collaborative Depression Study, irritability and anger were evident in over half of unipolar depressed patients, and were in turn linked to an array of complex illness characteristics, including more anxiety and substance misuse, greater depression severity and chronicity, and personality disorders.5 While some authors feel that irritable depression is practically synonymous with “bipolarity,”6 it may simply be an indicator of overall severity and complex psychopathology. “Affective instability” also poses a quandary for differential diagnosis. No edition of the DSM has ever identified affective instability as a criterion for mania or hypomania, although popular perception sometimes leads both patients and practitioners to assume that moment-to-moment “mood swings” are a defining element of bipolar disorder. Our own group found that community practitioners often diagnose bipolar disorder based mainly on the presence of mood lability—yet, when compared alongside individual DSM-IV symptom criteria for mania or hypomania, “affective See article by Perugi et al