The risk of relapse of bipolar affective disorder during the perinatal period is becoming better established, with studies estimating the proportion of women experiencing relapse at that time to be in the region of 35% (Wesseloo Reference Wesseloo, Kamperman and Munk-Olsen2016). The majority of research in this area has focused on bipolar I disorder, resulting in a growing evidence base to guide its management, which includes the prophylactic use of an antipsychotic or mood stabiliser and psychological interventions (National Institute for Health and Care Excellence 2014). However, very few studies have specifically examined the effect of pregnancy and childbirth in women with bipolar II disorder, leading to confusion about the optimal management of such patients during this high-risk period.
ICD-10 (World Health Organization 1992) does not distinguish between bipolar I and bipolar II disorder, in that an episode of mania or hypomania (plus another mood episode) is sufficient for a diagnosis of bipolar affective disorder. However, DSM-5 (American Psychiatric Association 2013) differentiates between the two, with a requirement for a full manic episode in order to meet the criteria for bipolar I disorder. In this reflection we will consider the two separately.
Background
There are no clear guidelines for clinicians regarding the care and treatment of pregnant women who are referred to psychiatric services with an existing diagnosis of bipolar II disorder, despite evidence suggesting a risk of relapse similar to that for bipolar I disorder during the perinatal period. Di Florio et al (Reference Di Florio, Forty and Gordon-Smith2013) examined 424 live births to women with bipolar II disorder and found that 42.2% of pregnancies or postpartum periods (up to 12 months) were associated with a major mood episode, of which 18.4% occurred during the pregnancy. From the total sample, 4% experienced a hypomanic episode, 2.8% a psychotic depression and 35.4% a non-psychotic depressive episode. Only 53% of postpartum mood episodes were within the first 4 weeks after childbirth, indicating that on average the onset of symptoms was later than in women with a diagnosis of bipolar I disorder (in which 80% occurred within 4 weeks). Similarly, the onset of depressive symptoms in individuals with a diagnosis of bipolar II disorder was also significantly delayed compared with women with a diagnosis of major depressive disorder (P = 0.004).
These results suggest that mood episodes in the perinatal period are more likely to be depressive rather than hypomanic in women with bipolar II disorder, and that these symptoms are likely to occur later. However, on wider evaluation, Di Florio et al (Reference Di Florio, Forty and Gordon-Smith2013) found that for women with bipolar II disorder, the incidence of mood episodes during the perinatal period (whether defined as the first 4 weeks or 6 months) did not significantly differ from that during other periods of the woman's life, and therefore concluded that the relationship between mood episodes and pregnancy/childbirth was not as clear as with bipolar I disorder.
Additionally, Sharma et al (Reference Sharma, Xie and Campbell2004) conducted a prospective study to evaluate whether women with a diagnosis of major depressive disorder or bipolar II disorder would change diagnosis following pregnancy and childbirth. They found that, out of 54 women with an initial diagnosis of bipolar II disorder, only one had a change in diagnosis to bipolar I disorder during the 3 months after childbirth. This suggests that women with bipolar II disorder are unlikely to be at risk of a full manic episode in the perinatal period.
Taken together, this evidence highlights important differences in the clinical presentation of bipolar I and bipolar II disorders in the perinatal period.
Current management
There are few studies examining the mental healthcare of women with an existing diagnosis of bipolar II disorder who present during pregnancy, with most research focusing on bipolar I disorder only, or combining the two groups. Therefore, the risk/benefit ratio of prescribing prophylactic medication in bipolar II disorder is less clear, which may make it difficult for clinicians to advise women on the best management of this condition during pregnancy and the postnatal period.
Sharma et al (Reference Sharma, Sommerdyk and Xie2013) studied 37 pregnant women with an existing diagnosis of bipolar II disorder, and found that during pregnancy just over half of participants (54%) were not taking any psychotropic medication, approximately one-third (32%) received monotherapy and the rest were taking combination therapy. In comparison, during the postpartum period only 14% of participants were not on any psychotropic medication. In the same sample, 51% of the women had a mood episode during pregnancy, compared with 70% in the postpartum period. These figures were comparable to those seen for bipolar I disorder in the same study.
Viguera et al (Reference Viguera, Whitfield and Baldessarini2007) conducted a prospective observational clinical cohort study of the treatment of women with bipolar I or bipolar II disorder during pregnancy. They found that women with bipolar II disorder were more likely to discontinue medication (including lithium, anticonvulsants, antidepressants and antipsychotics) during the pregnancy, and that women who discontinued mood-stabilising treatment were more likely to experience a mood episode during pregnancy (85%, 2.3 times the rate in women continuing medication). Women who discontinued medication spent, on average, 40% of their pregnancy in an illness phase, compared with 8% for those continuing medication. However, it is noted that these results did not distinguish between bipolar I and II disorder. Again, in both groups, the vast majority of mood episodes were depressive rather than manic or hypomanic, both of which were rare.
Case study review
To place this topic in a clinical context, we reviewed three case studies and evaluated general themes. The cases were all women who were referred to our specialist perinatal mental health team during pregnancy, and who had an existing diagnosis of bipolar II disorder. All of the women were stable in mental state at the point of referral.
In two cases a prophylactic low-dose antipsychotic was recommended but declined, and both women remained well during follow-up. In one case the woman was taking lamotrigine at the point of referral and continued this during the pregnancy, following the advice of the perinatal psychiatrist. She remained well until 10 months postpartum, at which point she experienced a depressive episode which responded to sertraline. Anecdotally, these cases seem to confirm the background research in this area.
It is noteworthy that two out of the three women remained well despite being medication free, supporting the proposal that relapse is less common in bipolar II disorder than in bipolar I disorder. Additionally, where relapse did occur it was a depressive episode, in line with the evidence that mood episodes are less likely to be hypomanic or manic. The reluctance of women to take psychotropic medication during pregnancy was also evident.
Discussion
Despite it being a relatively common occurrence, there remains little or no guidance for clinicians on how best to care for women with an existing diagnosis of bipolar II disorder during pregnancy and the postpartum period. A point to be considered is whether bipolar II disorder ought to be seen as a distinct clinical entity, separate from bipolar I disorder, or whether the two should be managed in accordance with the same guidelines.
Another aspect of this topic, and beyond the scope of this reflection, is that there is often an element of diagnostic uncertainty or overlap when women present to perinatal services with an existing diagnosis of bipolar II disorder. For example, in some circumstances it may be felt that an alternative diagnosis (such as emotionally unstable personality disorder) may be more appropriate, in which case the management may be different.
It is generally felt that most women would prefer to avoid medications during their pregnancy, making it important that clinicians are able to comprehensively describe the risks and benefits of choosing to continue or start medication, compared with being medication free. The limited available evidence suggests that there may be an increased chance of relapse during this period, but this appears less clear cut for bipolar II disorder than for bipolar I disorder. The risk of a manic or hypomanic episode appears to be small; rather, it seems more common for a woman to experience an episode of depression. In view of this, it may be more appropriate for women to be closely monitored for signs of an emerging depressive episode, which is likely to be gradual, rather than initiating prophylactic medication during the pregnancy.
Best practice would be for management to be tailored to the individual, taking into account their psychiatric history, for example the number and severity of previous mood episodes. Other aspects influencing the risk should also be considered, such as the level of social support, attitude towards the pregnancy, comorbidities, drug and alcohol use and level of engagement. Even if one puts the issue of medication aside, it seems evident that this patient group would benefit from specialist perinatal psychiatric support, in the form of regular monitoring and reviews, psychological input if appropriate, referral to specialist midwifery teams, and social support as required.
Box 1 summarises the evidence on risk and key management pointers.
BOX 1 Bipolar disorder in the perinatal period: some conclusions suggested by the evidence
The evidence suggests that:
• the risk of relapse of bipolar I disorder is slightly increased
• the risk of a manic or hypomanic episode is small
• an episode of depression is more likely than a manic or hypomanic episode
Management pointers:
• regularly monitor, particularly for signs of depression
• offer medication on a case-by-case basis
• involve specialist perinatal psychiatric support
Further research in this area examining women specifically with bipolar II disorder would be beneficial in creating clearer guidance for clinicians and patients.
eLetters
No eLetters have been published for this article.