June 2013

The two years beginning in late pregnancy and continuing post childbirth can be particularly difficult for many women. The risk of serious mental illness is significantly higher than at any other time in a woman’s life. More often than not, however, the mental health issues are attributed to the stress of becoming a mom and though hormones are often in the mix, the consensus about pregnancy and postpartum hormone-related mental health changes is more broad than specific, anecdotal than evidence based. The prevailing hypothesis suggests that puerperal mental illness, commonly referred to and investigated as postpartum depression, is not the result of endocrine pathophysiology, but rather a ‘maladaptive’ response to normal changes in reproductive hormones.  In other words, having babies is a normal function, with normal hormone changes, ‘if you can’t handle it, there must be something wrong with you.’

The fact that there are no norms or even broad reference ranges established for pregnancy/postpartum hormone changes doesn’t seem to enter into many conversations (how can one reasonably say something is normal if it isn’t ever measured?); nor does the fact that ‘abnormal’ hormone changes could easily be causative in mental illness or the fact that ‘normal’ hormone changes, if large enough, such as during pregnancy and following childbirth are likely to impact mood, cognition and behavior in some fairly noticeable ways. Failing to recognize and prepare women and their families for the spectrum of the biochemically mediated  mental health or mental status changes, whether they are ’normal’ or not, is just wrong and potentially dangerous. Many years ago, I set out to change that – to understand how the hormones of pregnancy and postpartum could affect mental health and cognition. What I found was fascinating, but first, a little (OK, a lot of) background.

Fundamental Precepts about Hormones and Behavior

All of my research begins with the basic proposition that hormones affect brain chemistry. We know from animal research that hormone receptors are located all over the brain (and the body for that matter), even in areas not responsible for reproduction. We also know that steroid hormones produced in the body, because they are lipid soluble, easily cross the blood-brain-barrier and bind to hormone and non-hormone receptors to change brain chemistry. And, we know that the brain can and does produce a full complement of steroid hormones by itself, having all of the necessary building blocks to synthesize hormones de novo or from scratch. Since the brain is the control center for mental health, cognition and indeed, everything, it stands to reason that because hormone activity is integral to brain chemistry, hormones are involved in mental health. Indeed, there are no biologically or even logically plausible reasons to remove hormones from any discussion of brain chemistry or subsequent changes in mental status. It just makes no sense.

Pregnancy and Postpartum Hormone Changes Mirror an Addiction Withdrawal Cycle

We see hints, sometimes rather loud hints, of the hormone-brain connection across a woman’s life cycle (puberty and menopause) and across the menstrual cycle, but these are often more gradual and less drastic biochemical changes than those of pregnancy and postpartum. During pregnancy, some hormones increase by over 1000 times their non-pregnant concentrations, only to drop immediately, to nothing or almost nothing following childbirth. Simultaneously, other hormones seem to increase following childbirth, thus, creating the complex chemical cocktail that is postpartum. With these enormous changes in biochemistry, it is truly remarkable that so few women experience difficulties.

From a pharmacological standpoint, the hormone changes across pregnancy and postpartum provide the perfect drug addiction-withdrawal model, where the drug use increases gradually but significantly over an extended period of time only to be eliminated cold turkey over a period of a couple days. From the brain’s standpoint, while there may be differences in specific reactions, there really is no difference, broadly speaking, between compensatory reactions it exhibits relative to increasing concentrations of a drug followed by its abrupt withdrawal and those it exhibits relative to increasing concentrations of hormones followed by abrupt withdrawal. The brain is going to get used to having certain concentrations of chemicals floating around and adapt accordingly. When those chemicals are removed, especially abruptly, there will be hell to pay in the withdrawal syndrome. How that withdrawal syndrome manifests will be contingent on the degree and pattern of biochemical change – which hormones or drug(s) are creating the problems, where and to what degree.

Consider alcohol versus heroin withdrawal as an example. Both withdrawal periods are horrible, but because each drug acts on different neurotransmitters within the brain, each withdrawal syndrome looks a little bit different. It is the same way with hormones. Each elicits a different biochemical reaction in the brain. Some hormones are sedatives, some are stimulants, some are direct, some are indirect; some have a whole bunch of receptors in areas of the brain that control memory, while others have receptors in the emotional centers of the brain. Without measuring the actual hormone changes associated with pregnancy and postpartum and the behavioral symptoms that ensue, there is no way to recognize or to treat a postpartum withdrawal syndrome or syndromes. And as many of you well know, hormone measurement in women’s health is all but ignored.

Pregnancy and Postpartum Mood Changes are Poorly Characterized

Perhaps because of our feminist tendencies (not wanting to admit that hormones affect our moods or our cognitive abilities), perhaps politics (blaming women) or perhaps just poor research (including that which does not consider the role of hormones in the diagnostic criteria), the standard nomenclature and diagnostic parameters for postpartum mental health issues are at best poorly defined and at worst completely incorrect.

According popular perspectives, the three classes of postpartum disease are the baby blues which is said to affect 80% of all new moms, postpartum depression that develops in 10-15% of women and postpartum psychosis, the rare condition that afflicts 1-2 per 1000 pregnant women. What does this mean? It looks like a progression of sadness that leads to psychosis. Is this what postpartum women experience? Well, not really, but the nomenclature stuck and was sufficiently correct that they could characterize some of the symptoms, in some of the women, to make using these terms a useful shorthand. However, because the symptoms associated with each of these conditions were never fully characterized appropriately, they have been repeatedly included or dis-included from diagnostic manuals with varying and even diametrically opposed diagnostic criteria depending upon the political winds of any given generation (the pitfalls of consensus based medicine).

Indeed, in the last iterations (IV, TR) of the DSM manual (the diagnostic bible for mental illness), postpartum was merely a time course specifier. That means, none of these conditions actually existed according to the diagnostic manual. There was no discrete illness or set of illnesses recognized as unique to the postpartum period, and certainly none connected to postpartum hormone changes.  Depression or psychosis, if they happened to arise within 30 days of childbirth, was considered postpartum related.  If these conditions developed during pregnancy or after the 30 day period, then they were not considered postpartum related. In effect, these conditions were just the normal, run-of-the-mill depression or psychosis.  From a purely logical standpoint, it seems difficult to believe that the brain chemistry of a postpartum woman is in any way similar to the brain chemistry of teenager or menopausal, or other non-postpartum woman or to a male depressed or psychotic patient.  If we believe that brain chemistry mediates behavior (and isn’t the entire medical-pharmaceutical establishment built on that presumption), why would we presume that radically different brain chemistries produce the same symptoms or behaviors?  We wouldn’t.

So, on the one hand, we have popular terminology that has done wonders to bring awareness to the potential difficulties some women have following childbirth but whose terms were not consistent with the DSM criteria. On the other hand, we have DSM criteria that really didn’t recognize postpartum as unique condition, but only as a time-frame to be noted and neither set of diagnostic opportunities was based on evidence that truly considered specific hormones changes might impact brain chemistry. Sure, there has always been the tacit – it’s hormonal – and certainly, there has been hormone-mood research but attempting to delineate which hormones, in which women, relative to which symptoms and within what time frame has yet to be fully addressed. And, as one might imagine, it is difficult to bring another set of variables – hormones- into an already poorly defined disease space. Do we measure hormones related to blues, depression and psychosis or are we measuring something else entirely?

Where to Begin

When beginning a research career in area where the data are limited, one has a few choices – ‘don’t’ -being the first and most logical option; take the safe, career boosting-route of replicating someone else’s work or throw all previous assumptions in the garbage can and begin from scratch. Not being the wisest, of course, I chose the third option.

I had a couple operating assumptions. The first was and still is, that certain hormones affect certain neurotransmitters (we know this to be true from animal research). When we radically change the concentrations of those hormones, the behaviors associated with said neurotransmitters (and maybe even some we hadn’t thought of) would become apparent.  Second, the symptoms that were expressed would be related to the particular pattern of hormone change – whatever that pattern may be. Third, the constellation of symptoms that arose would not likely not fall into the current diagnostic categories, but would cluster together in unique, and yet to be determined, ways. In other words, I believed that certain patterns would emerge based on animal research, but because there was so little human research and much of it was limited in scope, I was prepared for the fact that I was wrong. And I was wrong, in some ways, but that willingness to test more broadly and openly is what led to some pretty amazing discoveries.

How I Think about Perinatal Psychiatric Distress

Last bit of background, I promise. Notice that I said perinatal psychiatric distress and not postpartum depression, mood, or blues. Perinatal psychiatric distress and full-blown psychiatric disorders can emerge during either period, pregnancy or postpartum and relative to a myriad of biochemical and psychosocial factors. Limiting the discussion and nomenclature to ‘postpartum’ ignores women who are affected negatively by the pregnancy hormones and whose symptoms arise prior to delivery of the child.

Similarly, the hormone syndromes are not specifically depressive.  Some of the hormones affected by childbirth are clearly anxiogenic (elicit anxiety) and by the nature of where their receptors are located, other hormones can affect memory, decision-making, impulse control, sensory perception and a wide variety of emotions, physiological and cognitive functions. By categorizing and limiting the syndrome to ‘depression’ even an atypical depression, as it is often referred to, fails to recognize the spectrum or severity of symptoms experienced.

Finally, for the same reasons I don’t use the phrase postpartum depression, I don’t ascribe to the characterization of the baby blues. When one thinks of the baby blues, one immediately thinks of a milder form of depression or sadness. Though useful as a popular term, it does nothing to distinguish what, in some cases, may be emotional expressions of the hormone-based, physiological changes occurring postpartum (or during pregnancy – though not often measured) and in other cases early markers for distress. Neither the term nor the scale used to assess the ‘condition’ has any predictive ability and fails to recognize a whole host of symptoms linked to perinatal hormone changes, that cause significant distress for the mom.

Because there are a myriad of hormones involved in carrying a pregnancy to term that are involved in number of physiological systems, and the symptom expression from those interactions is broad, limiting the focus to depressive type symptoms, unnecessarily limits the spectrum and severity of distress that some women experience.  As with everything, if we don’t measure, we cannot manage. Part of measuring is figuring out what to measure.  Depressive symptoms are certainly important, but they do not represent the totality of the symptoms experienced and so, we must expand the symptom base and re-work the diagnostic nomenclature.

Just Get to the Damned Research, Already!

Why have I spent so much time explaining the nature of postpartum research in general and my assumptions and perspectives specifically?  Why haven’t I just told you what I learned?  Well, because where you start determines where you end up, especially in science. Yes, I could have assumed the definitions and the research supporting those definitions of ‘postpartum depression’ were correct and then designed studies to support the appropriate hypotheses. It certainly would have been easier, but I didn’t. There were too many missing pieces and unanswered questions – things that just didn’t fit or make sense for me to go down that route. I had to create a new path – to throw everything in and let the pieces fall where they may.  I had to let the data tell the story. I did and I will, let the data tell story.

Part two: Beyond Depression, Understanding Perinatal Mental Health.

Related posts:

Perinatal Mood, Cognition and Hormones Severe Postpartum Anxiety, Suicidality, and Polypharmacy Reframing Maternal Health: How Do We Know What We Think We Know?