21 v66, #08 Letters

Sir: Perlis et al. found that irritability was present in 40% of outpatients with major depressive disorder (MDD). The finding mirrors DSM-IV-TR, which lists irritability not only as a core symptom of mania and hypomania, but also as a common symptom of MDD. Kraepelin classified irritability as a manic (excitement) symptom. The bipolar nature of irritability in MDD can have an important impact on treatment of MDD. Although a study on MDD with irritability showed that fluoxetine was effective in reducing irritability, clinical observations suggest adding mood-stabilizing agents to antidepressants in MDD with irritability because antidepressants alone may worsen irritability. To test the bipolar nature of irritability, I scanned a large database of patients who presented to our practice from June 1999 to January 2005. As the present analyses were not planned when data were recorded, an interviewer’s bias is unlikely. Detailed study methods are reported elsewhere. The present sample includes 379 consecutive bipolar II disorder (BPII) and 271 consecutive MDD outpatients who were assessed when they presented to a private practice for treatment of a major depressive episode (MDE) before starting psychopharmacologic treatment. Patients were interviewed by a senior clinical and research psychiatrist using the Structured Clinical Interview for DSM-IV, the Hypomania Interview Guide to assess hypomanic symptoms in the MDE, and the Family History Screen for assessing bipolar (type I and II) family history in probands’ first-degree relatives. In the BPII sample, the mean (SD) age was 41.3 (13.0) years, 67.2% were women, the mean (SD) Global Assessment of Functioning (GAF) score was 50.3 (9.2), the mean (SD) age at onset of first MDE was 22.6 (10.5) years, and 45.2% had a family history of bipolar disorder. In the MDD sample, the mean age was 46.6 (14.6) years, 61.6% were women, the mean GAF score was 50.7 (9.6), the mean age at onset of first MDE was 31.7 (13.7) years, and 16.0% had a family history of bipolar disorder. Irritability was present in 60.6% of BPII patients, compared with 37.9% of MDD patients (χ = 34.6, df = 1, p = .0000) (a figure close to that reported by Perlis et al.). An important external validator of diagnosis of bipolar disorder is bipolar family history. My logistic regression analysis of irritability versus family history of bipolar disorder found an odds ratio (OR) of 2.1 (95% CI = 1.4 to 3.2, p = .000); when the analysis was controlled for the confounding effect of BPII, the OR was 1.7 (95% CI = 1.1 to 2.6, p = .007). Difference in age at onset between bipolar disorders and depressive disorders is also an important diagnostic validator. My logistic regression analysis of irritability versus onset age found an OR of 0.7 (95% CI = 0.6 to 0.8, p = .000); when the analysis was controlled for the confounding effect of BPII, the OR was 0.8 (95% CI = 0.7 to 0.9, p = .006). We also tested whether MDD with irritability, as distinguished from MDD without irritability, shared demographic and family history characteristics with bipolar disorder. Family history of bipolar disorder and onset age were the validators. Logistic regression analysis of MDD with irritability versus family history of bipolar disorder found an OR of 3.2 (95% CI = 1.5 to 7.1, p = .003). Logistic regression analysis of MDD with irritability versus onset age found an OR of 0.7 (95% CI = 0.6 to 0.9, p = .010). These results seem to support the bipolar nature of irritability. This does not mean that irritability is always bipolar (no symptom is 100% specific in psychiatry), but that it is more likely than not to be bipolar. Treatment implications can be important, as the bipolar nature of irritability could require moodstabilizing agents with antidepressants for the treatment of MDD with irritability, as suggested by clinical observations. Controlled studies are required to confirm the findings of the study reported here.