When a woman has a history of severe and relapsing mental illness, but is stable on her current treatment, and is planning a pregnancy or is postpartum, what is the best course of action for her and her baby?
In my conventional training, I spent (and continue to spend) countless hours analyzing the available registry, cohort, retrospective, and prospective studies on the safety and efficacy of psychotropic medications in pregnancy and lactation. As one of approximately 300 reproductive psychiatrists in the world, I am privy to the defensive posture of clinicians who deign to help these women, often perceived by general psychiatrists to be too high liability to treat. I understand the feeling that psychiatric medications are held to a standard of obstetrical scrutiny that anti-emetics, antibiotics, steroids, and painkillers are not. I know that there are more than 25,000 cases of SSRIs in the perinatal literature and yet there still isn’t a clear signal of any reproducible, consistent teratogenicity or danger that is definitely not attributable to the underlying illness.
Miscarriage, preterm birth, and neonatal adaptation are all considerations, but are such common and multifactorial outcomes, that it will be some time before clear A-causes-B information surfaces. In many ways, the collective outcome of this data surprises me – we are applying medications (often prepared with unstudied preservatives, dyes, and metals) based on subjective diagnostic parameters, to women who bring other burdens to the table that are largely uncontrolled for in these studies (high glycemic diets, xenoestrogen/endocrine modulator exposure, baseline inflammation). We are rarely certain that the medication, if taken, has treated their symptoms to remission, and study after study fails to show a signal of serious danger on the order of alcohol, Accutane, or even Depakote.
These studies are compromised, fundamentally, by the fact that they are not, and will never be, randomized, double-blind-placebo controlled trials. Analyzing the methodology, reading beyond the abstract, and contextualizing findings for any given patient requires serious dedication of a practitioner’s intellectual resources. How do we personalize these recommendations? After all, studies are studies, and individuals are individuals. In my perspective, we assess for a better treatment path for the patient that addresses root cause hormonal and inflammatory contributors and applies interventions that are low risk and potentially high yield, preferably with some degree of evidence base. We start by looking at interventions that have benefits for mom, mood, and baby (diet, exercise, meditation, fatty acids, l-methylfolate, vitamin D, probiotics, etc).
This process is not for the impatient. It can take upwards of a year to truly rebalance a patient’s system, and to properly taper medication. It is no wonder that data suggests a high risk of relapse. Abrupt discontinuation of antidepressants (and mood stabilizers/benzos/stimulants) places a demand for reequilibration that often overwhelms a system’s ability to achieve that without support, or return to medication. The process of medication weaning is a complex and time-consuming one that, in my experience, requires cell membrane, amino acid, and micronutrient support in the setting of painstakingly slow taper of dosage. I always tell my patients that the ultimate goal is to gently usher the medication out the back door, not push it off the balcony.
When I meet a patient who is already pregnant on medication, wondering what is best for her and her fetus, it is not a black or white decision-making process. We have to consider the central nervous system stimulation of medication withdrawal, potential relapse (often hard to distinguish from the former), safety of supplements in pregnancy, and then the flipside, which is the known and unknown risks of medications, assuming that it is helping her. This web is brilliantly illuminated in this paper, which I would imagine leaves most readers feeling a sense of sky-is-falling nihilism about treating pregnant patients. Oberlander holds all of our assumptions about the interplay between medications, genes, and the epigenome under a microscope, revealing all of the pores and wrinkles we were hoping to obscure with a professional distance. The number of “if, then…” clauses becomes overwhelming.
An example: If the mother carries a short/short copy of the serotonin transporter gene (which we understand to confer “environmental sensitivity” or lack of resilience, and poor response to SSRIs), and she experienced childhood trauma herself, her serotonin trafficking may look very much like that of a woman with long alleles and SSRI exposure. He references an animal study which suggests normalization of the stress response after SSRI exposure in utero, alluding to the potential effects of SSRIs that span well beyond the serotonin modulation and may even be protective in some cases.
Then, once the serotonergic imprinting has occurred in utero, the effects are further modulated by postnatal exposures such as nurturing care and breastfeeding so that cortisol levels may be normalized even if there was alteration of serotonin signaling in pregnancy that messed with the hypothalamic pituitary axis. It appears that genetic variants “matter” only in settings of environmental stress, and maybe only if that stress represents a “mismatch” to the in utero level of stress. After all, epigenetics is a means of preparing the baby for the world it appears to be entering.
The summary of this article reads like a “well, maybe, uhh, we’re not sure” conclusion to the examination of just one proposed system of influence – serotonergic tone in utero as regulated by SRIs, genetic variants, and mood disorders. It renders the reductionist model of depression as a serotonin-deficit disease cured by a reuptake inhibitor, embarrassingly simplistic. In light of this, it is important to feel comfortable sitting with uncertainty, giving wellness and healing a cautious and humble best effort while respecting the vast swaths of knowledge that have yet to be achieved. I don’t waste my patient’s time with hand waving about biochemical imbalances as an explanation for their state (I have been known to gesticulate wildly when talking about gluten, though).
Adding to this complexity, we suspect, based on a study by the same author that maternal antenatal depression may increase methylation of the glucocorticoid receptor (shutting down its on switch), leading to elevated stress response in the child at three months. He discusses some 15% of childhood behavioral problems being attributed to antenatal stress/anxiety. This is a sign of a toxic, inflamed system, and it is no surprise that this maternal inflammation has a negative effect on the fetus.
What else can cause maternal inflammation? Here’s the short list:
- Processed foods/trans fats
- Lack of sleep
- Sugar and associated insulin resistance
- Environmental toxins/industrial chemicals
- Chronic infection
I don’t know of many safety studies controlling for these individual variables, yet there are several that examine the deleterious effects on birth outcomes when a woman has evidence of inflammatory markers.
We are operating in a system of medication-driven treatment, and the only real time that medication risks are taken seriously is when a woman has seen that little plus sign on the stick in her bathroom. Then they are often violently tossed out the window, and deemed contraindicated for the coming 1+ years of gestation and lactation. We need to be thoughtfully examining how to best support a woman’s system from the time that she gets her first period (and even more so if that is before she is 13!), because engaging this line of rhetoric once she is pregnant is often too little to late, and a ones or zeros, meds or no meds approach does not do justice to the complexity of a system that is hard at work building another system.
Mad in America hosts blogs by a diverse group of writers. These posts are designed to serve as a public forum for a discussion—broadly speaking—of psychiatry and its treatments. The opinions expressed are the writers’ own.
Mad in America has made some changes to the commenting process. You no longer need to login or create an account on our site to comment. The only information needed is your name, email and comment text. Comments made with an account prior to this change will remain visible on the site.