Abstract

Introduction
The term postpartum mood disturbance generally refers to maternity blues, postpartum depression and postpartum psychosis (PP). PP is the most serious but rare complication of childbirth that follows 0.1–0.2% of deliveries. The typical clinical picture of PP has been described as a woman with ‘an odd affect, withdrawn, distracted by auditory hallucinations, incompetent, confused and catatonic; or alternatively elated, labile or rambling in speech, agitated or excessively active’ (Brockington et al., 1981). Psychotic features can occur alone or in combination with rapidly fluctuating mood symptoms. Co-occurrence of manic and depressive symptoms with psychotic features has also been noted. Confusion and perplexity are often mentioned as distinctive features of PP but these symptoms may be a consequence of the severity and/or acuity of the postpartum episode. The risk for PP is increased in women with a personal or family history of bipolar disorder (BD) and/or PP. The risk of recurrence of a postpartum episode with psychotic features is between 30% and 50% (American Psychiatric Association, 2013). Owing to its acute onset, rapidly escalating symptoms, and association with maternal suicide and infanticide, PP is a psychiatric emergency that often requires psychiatric hospitalization. For some women the psychotic episode is limited to the postpartum period, but the majority of women have both puerperal and non-puerperal recurrences. Psychosis limited to the postpartum period has a distinct risk profile and phenomenology compared to PP occurring as part of BD. Despite lack of acceptance of PP by the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5) as a separate diagnostic entity and lack of consensus about its definition, the term PP remains widely used by clinicians and researchers alike (American Psychiatric Association, 2013). The debate about its definition has centered on the diagnostic composition of PP, the duration of the postpartum period, and whether or not to include disorders with prepartum onset (Chaudron and Pies, 2003). In this paper, we discuss the limitations of using PP as a diagnostic label. We argue that the term ‘postpartum psychosis’ should be dropped and replaced with more specific diagnostic entities.
Diagnostic status
PP has been variously considered as a distinct diagnostic entity, a childbirth-triggered psychotic episode of BD, an episode of cycloid psychosis, or a psychotic episode of any psychiatric disorder in the postpartum period. It has also been linked to organic factors including thiamine deficiency, postpartum thyroiditis, and infective delirium (Sit et al., 2006). However, the DSM-5 does not consider PP as a separate diagnosis. Considering that PP is usually a mood episode with accompanying psychotic features, DSM-5 allows the use of the specifier ‘with peripartum onset’ to characterize mood episodes with psychotic features. The specifier can be used in individuals with major depressive disorder (MDD), BD-I, or BD-II, but not in patients with schizophrenia and other psychotic disorders, except brief psychotic disorder that can have a postpartum onset. Conversely, the International Classification of Diseases, Tenth Revision (ICD-10) allows the use of the diagnostic category ‘puerperal psychosis not otherwise specified’ to capture severe mental and behavioral disorders associated with the puerperium not elsewhere classified (World Health Organization, 1992). This category includes only psychiatric disorders commencing within 6 weeks of delivery that do not meet the criteria for another disorder because of insufficient information for diagnosis, or because special additional clinical features are present that make their classification elsewhere inappropriate. Owing to its acute onset, limited duration, and recovery between recurrences, PP has been linked to cycloid psychosis – a diagnosis not recognized in DSM-5 or ICD-10 classification systems. Cycloid psychosis is a term used for disorders that are characterized by acute psychotic features, as in schizophrenia, but have an episodic illness course as in manic-depressive disorder.
Definition
PP appears to be a heterogeneous entity and includes various psychiatric disorders that share the common feature of emergence of psychotic symptoms in the postpartum period. The diagnostic breakdown of PP depends on the way the disorder is defined. For example, studies defining PP as occurrence of a psychotic episode in the first 2 weeks postpartum were more likely to include disorders that closely followed childbirth such as manic or mixed episodes, and may have excluded disorders with delayed onset such as MDD. Additionally, it is unclear whether studies excluded cases with onset of symptoms before delivery: approximately 8% of cases have ‘prepartum’ onset of PP. Another commonly used criterion to define PP is the requirement of psychiatric hospitalization. Once again, this requirement may lead to overrepresentation of diagnoses such as manic or mixed episodes, and exclusion of diagnoses that are less likely to require psychiatric hospitalization such as MDD, depressive episode of BD-II, or obsessive compulsive disorder (OCD).
Drawbacks of using the term ‘postpartum psychosis’
Optimal management of psychiatric disorders requires knowledge of the type of the disorder, phase of the disorder (e.g. mania or depression in patients with BD), safety risk, comorbidity, and risk of recurrence. The term PP is applied generically to diagnose psychosis and does not provide specific information to facilitate accurate diagnosis of the underlying psychiatric disorder. Exclusive focus on the assessment of psychotic features may delay the diagnosis and treatment of mood episodes that accompany or follow psychotic symptoms. Symptoms of depression may be considered secondary to psychosis and the consequent delay in diagnosis and treatment may prolong the depressive episode and have an adverse effect on the mother–child relationship. Owing to its emphasis on the cross-sectional symptomatology, clinicians may not give the necessary attention to the longitudinal illness course. The consequences of misdiagnosis of bipolar depression as MDD can be particularly detrimental, including a higher risk of suicide or infanticide. Also, the diagnosis of PP is not informative when assessing safety issues. Certain disorders such as BD might be associated with a higher risk for self-harm or harm to the baby, while the risk may be quite low in disorders such as OCD with psychotic features.
Lack of guidance in the selection of appropriate treatment for the management of the disorder is another concern. For example, antidepressants in combination with neuroleptics are the mainstay treatment of delusional depression or OCD with psychotic features, but antidepressants are contraindicated for women with a manic or mixed episode. Lithium is the most studied drug and has demonstrated efficacy in the prevention of PP; however, whether or not lithium is similarly effective for all the disorders that constitute PP is unknown.
Recommendations
As the name suggests, occurrence of a psychotic episode in the first few weeks after delivery is the defining feature of PP. Thus, psychotic features and their temporal association with childbirth appear to be the only characteristics that unite disparate disorders that constitute PP. To a large extent, the clinical picture of PP, including safety risk, also depends on the accompanying psychiatric disorder, and in the case of BD, on the phase of the illness. The term PP should be abandoned and replaced with more specific diagnostic entities. For example, a woman presenting in the postpartum period with depression and psychosis should be diagnosed as having MDD with psychotic features rather than PP. Similarly, a manic episode with current psychotic symptoms should be classified as BP-I disorder, manic episode with psychotic features. Women with no prior history of psychiatric disorders but presenting with psychosis after delivery should receive the diagnosis of brief psychotic disorder with postpartum onset.
Footnotes
Funding
This research received no specific grant from any funding agency in the public, commercial or not-for-profit sectors.
Declaration of interest
Verinder Sharma has received grant support from, participated on scientific advisory boards for, or served on the speakers’ bureaus of AstraZeneca, Bristol-Myers Squibb, Cephalon, Eli Lilly, Janssen, Lundbeck, the Ontario Mental Health Foundation, Pfizer, Servier, and the Stanley Foundation. Christina Sommerdyk has no conflict of interest to declare. The authors alone are responsible for the content and writing of this paper.
