"In Utero Exposure to SSRIs and Development of Mental Disorders: A Systematic Review and Meta-analysis"

The title heading this post - "In Utero Exposure to SSRIs and Development of Mental Disorders: A Systematic Review and Meta-analysis" - comes from the findings published by Annemette Halvorsen and colleagues [1] on a topic that has been of some research interest for quite a few years (see here and see here for example).

SSRIs denote a class of medicines called the selective serotonin reuptake inhibitors, principally used to treat/manage various types of depression. Although in quite a few cases a lifesaver, such medicines also come with a risk-benefit profile (same as all medicines), a risk-benefit profile that also extends to their use during important times such as pregnancy. Minus any scaremongering sentiments, for quite a while now, there has been some interest in whether SSRI use during pregnancy *might* have some important implications for offspring health and development [2]. Such *associations* whilst potentially important, need of course to be balanced with the reasons why SSRIs are used, and how for example, depression is not typically something that magically disappears as and when a woman becomes pregnant...

The Halvorsen paper represents a "systematic review and meta-analysis" of the pertinent research literature examining whether "in utero exposure to selective serotonin reuptake inhibitors (SSRIs) is associated with increased risk of developing mental- or behavioral disorders." Such a 'boiling down'  of the research literature was, we are told, "conducted in adherence with the Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) guideline" and specifically covered in connection to various childhood developmental labels such as attention-deficit hyperactivity disorder (ADHD) and autism spectrum disorder (ASD).

Based on their analysis of the available data - "20 studies were included in the review and results from 18 of these were meta-analyzed" - researchers observed "a statistically significant positive association between in utero exposure to SSRIs and mental- or behavioral disorders such as autism spectrum disorder..., attention-deficit/hyperactivity disorder... and mental retardation." I say that with the suggestion that intellectual (learning) disability might be a more appropriate description for that last label mentioned. The authors caution however that their results do not necessarily mean that prenatal SSRI exposure *causes* something like autism or ADHD: "these associations do not necessarily reflect a causal relationship since the results included in this meta-analysis are likely affected by residual confounding by indication, which is likely to account for some (or all) of the positive association."

So what to make of this research? Well 'it's complicated' is the long-and-short of it. Minus any sweeping generalisations I'll take you back to some other research (see here) that did try and disentangle the 'causality' side of this area of study (whether underlying maternal depression or antidepressant use was the most important variable). The findings reported by Rai et al [3] did [cautiously] talk about how "children exposed to antidepressants during pregnancy seemed to be at a higher risk of autism, particularly autism without intellectual disability, than children of mothers with psychiatric disorders who were not treated with antidepressants during pregnancy." Other research however has observed something rather different (see here). Y'see, it's complicated.

So all I can really say is that more research is required, and if in doubt, talk to your prescribing physician.

----------

[1] Halvorsen A. et al. In Utero Exposure to SSRIs and Development of Mental Disorders: A Systematic Review and Meta-analysis. Acta Psychiatr Scand. 2019 Apr 2.

[2] Pedersen LH. et al. Selective serotonin reuptake inhibitors in pregnancy and congenital malformations: population based cohort study. BMJ. 2009;339:b3569.

[3] Rai D. et al. Antidepressants during pregnancy and autism in offspring: population based cohort study. BMJ. 2017;358:j2811.

----------

Psychiatric and seizure medicines for autism: what 'works' best

The paper by Devon Coleman and colleagues [1] represents pretty good scientific value for money by my reckoning. Describing the results of a 'survey' called the "National Survey on Treatment Effectiveness for Autism" created by the researchers, the aim was to provide "separated... scales for overall benefits and overall AEs [adverse effects]" for a wide range of interventions used in the context of autism.

The focus this time around was on "Psychiatric and Seizure medications data" but it looks like there may be quite a bit more to see from this group in future with regards to data on "supplements, diets, therapies, and educational interventions" also collected during this initiative. Before continuing on with this paper I have to hat-tip the authors for mentioning a great and much under-valued resource in autism circles: "the Parent Ratings of Behavioral Effects of Biomedical Interventions Survey,... conducted by the Autism Research Institute (ARI) and published in 2008." Indeed, in other posts talking about the medicines cabinet and autism (see here) I've expressed my positive views of the ARI resource albeit with caveats.

Anyhow: "we report ratings of 26 psychiatric and seizure medications by 505 participants." Researchers actually noted that nearly 900 people completed their survey; most of whom were described as the "Primary caregiver of an individual with autism." Some people might um-and-ah about the lack of 'authentic autistic representation' in this study, but one needs to bear in mind that most participants - about three-quarters - were under 18 years of age, most diagnosed with autism and not autism spectrum disorder (described as "less severe than a diagnosis of autism") and most were currently described as having mild, moderate or severe autism. I know this won't be enough for some people, but there you have it.

Then to the medicines that were 'graded', which fell into "five general categories: stimulants (four medications), SSRIs [selective serotonin reuptake inhibitors] (five medications), antipsychotics (four medications), seizure (nine medications), and other (four medications)." There's a lot of data included in the Coleman paper which really is too much for a blog post. I'll direct you to Figure 8 of the Coleman paper which provides a handy 'net benefit score' taking into account an 'overall benefit score' and an 'overall adverse score' for each medicine. When it came to SSRI medicines - typically indicated for treating depression - sertraline came top. When it came to antiepileptic medicines - primarily used to manage epilepsy and/or seizures - lamotrigine came top. When it came to antipsychotic medicines, aripiprazole came top. I was also interested to see that buspirone, a medicine typically indicated for anxiety, also did pretty well according to the Coleman results, which kinda ties in with some continuing research interest in this medicine with autism in mind (see here). Researchers also provide a handy 'medications for symptoms' overview as a consequence of their results (see Table 7) covering various symptoms from aggression/agitation to tics/abnormal movements. I can see this being particularly useful when it comes to physicians having to make big medication decisions (which should never be entered into lightly).

There's a couple of other details that are also mentioned in the Coleman paper outside of those 'how was medicine rated?' sentiments. Some details are not likely to make many friends in some quarters. So, around 2% of participants were described as follows: "No current diagnosis, but he/she was on the autism spectrum previously." Yes folks, such data once again harks back to the idea that for some people at least, autism is not a lifelong diagnosis (see here and see here). And also: "Thirty-four percent of participants had early onset of symptoms, but 56% had normal development followed by a plateau or regression." Regression accompanying autism is not the 'dirty' concept that it used to be (see here). Indeed, in the few years that I've been blogging about autism research, I've seen it become a lot more commonplace to talk about regression and autism (see here) even to the point that some now talk about it being 'the rule rather than the exception' (see here). Interesting.

And then there's something even more controversial in the Coleman paper: "The perceptions of possible causes of the regression are listed in Table 2." So we have things like high fever, illness, seizure and then... vaccination. I know this takes us into some uncomfortable territory, but the authors report that 51% of respondents to their survey who cited regression as part of the clinical picture mentioned vaccination as the 'perceived cause' whether singly or in conjunction with other factors. Of course I'm going to provide a link to what the population-based science says on this matter (see here) with the caveat that such 'perceived cause' data perhaps needs objective and dispassionate follow-up (see here and see here).

The Coleman results are not without their limitations as per author comments such as: "The survey is retrospective and based on respondent memory which reduces the accuracy" and "The results are subject to “placebo effect” since it represents clinical data without a placebo control, so the real benefit is likely less than the perceived benefit." But let's not take too much away from the findings and how they may, as well as informing clinical practice, also hopefully lead to further inquiry to make medicines safer and more reliable for those on the autism spectrum who access them.

----------

[1] Coleman DM. et al. Rating of the Effectiveness of 26 Psychiatric and Seizure Medications for Autism Spectrum Disorder: Results of a National Survey. J Child Adolesc Psychopharmacol. 2019 Feb 6.

----------

Carefully: effect of SSRI use on "rating-scale-assessed suicidality in adults with depression"

I stress the word 'carefully' in the title of this post discussing the findings reported by Jakob Näslund and colleagues [1] because it covers the very sensitive idea that "selective serotonin reuptake inhibitors (SSRIs) have been claimed to elicit or aggravate suicidal ideation."

I think it's sensible to begin this post by stressing that (a) NO medical or clinical advice is given or intended on this blog, and (b) anyone with concerns about their taking this class of medicines really needs to speak to their physician BEFORE making any changes to their prescribed medication routine. I know that point (b) sounds like me giving medical / clinical advice but it's common sense to talk to your medical professional first who's spent years studying and probably years practising medicine, rather than tinker around yourself...

There is always a balancing act to consider when discussing research such as this. A medicine indicated for various clinical conditions, that is taken by many, many people, and quite successfully treating / treated (nay, very successfully [2]) a condition that can, without treatment, have life-limiting consequences. No-one wants to rock the boat and scare or deter people from accessing such a treatment. At the same time however, one needs to know everything about that medicine; not least whether for some, there may be side-effects to possibly consider...

It's been a quite a long running saga talking about the possible additional effects of SSRI use for some (see here). It's drawn heavily on often harrowing individual stories and perspectives and not also been helped by the seeming (in)actions of some of the manufacturers of such medicines (see here). Näslund et al decided to approach this delicate topic from the point of view of analysing "the effect of [SSRI] treatment on rating-scale-assessed suicidal ideation in individual patients." This is distinct from other work that has focused on actual suicides or "suicide-related adverse events" that have been carried out before. The authors suggested that their approach might have the advantages of measuring the "net influence of treatment on suicidality at a group level" as well as the ability to "detect individual cases of emergence or aggravation of suicidal ideation." To this end, scores on the Hamilton Rating Scale for Depression (HRSD) particularly focused on "item 3 of the HRSD" covering suicidal ideation/attempts, was a core feature of their study covering "young adults (18–24) (n = 537) and adults (≥25) (n = 7725)." Said participants were derived from "all industry-sponsored, HRSD-based, FDA-registered placebo-controlled studies undertaken to explore the effects of citalopram, paroxetine or sertraline in major depression in adults."

Results: "In patients above the age of 24, SSRIs were found to reduce the mean rating of the HRSD suicidality item from week 1 until study end-point and also to reduce the risk for aggravation of suicidal ideation and emergent suicidal behaviour." This is very good news. It provides "strong support for the view that the net effect of SSRI treatment is beneficial rather than harmful" when it comes to suicide ideation/contemplation bearing in mind the specific focus on on item on the HRSD. I will again link to the recent findings by Cipriani and colleagues [2] reporting that: "All antidepressants were more efficacious than placebo in adults with major depressive disorder." It doesn't, as Näslund et al suggest, rule out rare cases of 'adverse effects', but does suggest that any such extreme side-effects are not likely to be encountered by most people who take such medicines.

When however it came to those younger adults (aged 18-24 years), the results were a little less straight-forward. So: "In young adults, those given an SSRI were at enhanced risk for worsening of suicidal ideation (in the unadjusted analysis) or emergent suicidality (loose but not strict definition) during the late (weeks 3–6) but not the early phase (weeks 1–2) of treatment." You'll see from the number of brackets used in that last quote that the authors provide some caveats to such findings; but this shouldn't take away from the trends observed. Indeed, bearing in mind such findings and also that "both SSRIs and placebo resulted in an end-point rating of suicidality equal to that observed in adults given an SSRI and lower than that observed in adults given placebo" you kinda get the impression that further investigations are needed to ascertain for example, whether depression and/or suicidality in the 25 and overs is somehow 'different' from depression in the younger age group. At least, different insofar as what treatment choices might be primarily made available. No, I'm not saying that this is evidence enough that SSRIs should have some sort of age restriction, just that cost/benefit ratios might perhaps have to be a little more 'age-sensitive' as well as individual-sensitive.

And, if anyone needs someone to talk to, there are resources available.

----------

[1] Näslund J. et al. Effects of selective serotonin reuptake inhibitors on rating-scale-assessed suicidality in adults with depression. Br J Psychiatry. 2018 Feb 5:1-7.

[2] Cipriani A. et al. Comparative efficacy and acceptability of 21 antidepressant drugs for the acute treatment of adults with major depressive disorder: a systematic review and network meta-analysis. Lancet. 2018. Feb 21.

----------

Pregnancy antidepressant use and offspring outcomes beyond autism?

I'll say one thing for the British Medical Journal (BMJ), they don't seem to be afraid to cover important issues relating to the *possible* "reproductive safety of drugs" [1] as per their publication of yet another paper looking at antidepressant use during pregnancy and offspring outcomes published by Xiaoqin Liu and colleagues [2]. This follows another paper published in the BMJ and covered on this blog a few weeks back (see here).

The specific class of drugs being examined again are, as I mentioned, the antidepressants and based, yet again, on data from one or more of those wonderful Scandinavian population registries in relation to how pregnancy use of antidepressants *might* impact on risk of risk of various psychiatric disorders in offspring. A diagnosis of autism spectrum disorder (AD) has been a primary focus of previous research in this area; particularly the question of whether the medicines themselves or the condition(s) that the medicines are used for (i.e. maternal psychopathology) might be the more important variables related to enhanced offspring risk of autism. The last paper covered on this blog by Rai and colleagues [3] very cautiously suggested that: "The results of all these analyses, which used different assumptions, seemed to be consistent with each other, suggesting that the association between in utero exposure to antidepressants and autism might not be fully explained by confounding." Cautiously...

This time around the net was widened from just looking at autism as an offspring outcome to "overall risk of psychiatric disorders" including: "autism spectrum disorder... mood disorder... neurotic, stress related, and somatoform disorder... behavioural and emotional disorder... and mental retardation." I might add that these are the authors words not mine and diagnoses were based on ICD-10 criteria and codings.

Looking and following some 900,000 children born between 1998 and 2012 in Denmark until 2014, researchers categorised participants according to their pregnancy antidepressant exposure(s): "unexposed, antidepressant discontinuation (use before but not during pregnancy), antidepressant continuation (use both before and during pregnancy), and new user (use only during pregnancy)" and looked at the frequency of those conditions included for study.

Results: some 2% of children were born to mums who used an antidepressant during pregnancy (n=21,063). Various types of antidepressants were used but the majority were prescribed SSRIs (Selective Serotonin Reuptake Inhibitors) either alone (monotherapy) or in conjunction with other non-SSRI medication.

"We observed increased risks of psychiatric disorders in all three groups of antidepressant users (discontinuation, continuation, and new user groups), compared with the unexposed group." This itself is an important finding but does not yet disentangle whether medicine use or the reason(s) for medicine use might be the more important variable. Then: "we observed an increased risk of psychiatric disorders in children whose mothers continued antidepressant use during pregnancy, compared with mothers who discontinued." Such a statement potentially edges a little closer to some influence of pregnancy antidepressant use on offspring outcomes but, and it is an important but: "These associations could be attributable to the severity of the underlying maternal disorders in combination with in utero antidepressant exposure." In other words, those mums who needed to continue antidepressant use throughout pregnancy probably had to do so because their symptoms either returned or were not controlled properly when medication is not in place. This might imply that more serious maternal psychiatric disorder could be a variable in any enhanced risk of offspring psychiatric disorder.

When it came to looking at specific diagnostic labels for offspring, all but "mental retardation" (I prefer the term learning disability) seemed to be elevated alongside "in utero exposure to antidepressants." The conditions with the highest risk were the mood disorders including diagnoses such as clinical depression and bipolar disorder. Given that antidepressants are typically (but not exclusively) used to treat/manage mood disorders, such a finding of maternal mood disorder potentially transmitting down into offspring mood disorder receives further credence from such results.

The Liu paper and accompanying editorial grapple with the question of whether the *possible* risks from pregnancy use of antidepressants on offspring merit a change to guidance on their use at such a critical time. As I've mentioned on other occasions discussing this topic, antidepressants are not typically just dispensed willy-nilly without appropriate clinical indication. They provide an important service in controlling various types of symptoms; pertinent to the idea that uncontrolled depression during pregnancy for example, can have all-manner of negative outcomes. As all medicines do, yes they may have side-effects but these need to be weighed up against perceived benefits, taking into account any important influence on the developing child. Indeed, the authors note: "any final decision on antidepressant continuation should be individualised and made jointly by health professionals and patients."

And whilst I've mentioned pregnancy antidepressant use and offspring autism risk, yet another review paper enters the scientific fray [4]...

----------

[1] Nordeng H. et al. Prenatal exposure to antidepressants and increased risk of psychiatric disorders. BMJ. 2017; 358: j3950.

[2] Liu X. et al. Antidepressant use during pregnancy and psychiatric disorders in offspring: Danish nationwide register based cohort study. BMJ. 2017; 358: j3668.

[3] Rai. D. et al. Antidepressants during pregnancy and autism in offspring: population based cohort study. BMJ. 2017; 358: j2811.

[4] Andalib S. et al. Maternal SSRI exposure increases the risk of autistic offspring: A meta-analysis and systematic review. Eur Psychiatry. 2017 Jun 20;45:161-166.

----------

Pregnancy exposure to SSRIs and offspring autism risk: debate continues

"It remains unclear whether the association between first trimester SSRI [selective serotonin reuptake inhibitor] exposure and child autism that was present in the case-control studies even after adjustment for MMI [maternal mental illness] is a true association or a product of residual confounding."

So said the results of the systematic review and meta-analysis undertaken by Hilary Brown and colleagues [1] looking at a potentially important association between pregnancy use of a class of medicines typically used as antidepressants (albeit with some caveats [2]) and risk of offspring autism. This topic has previously received some airtime on this blog (see here and see here) and specifically, how maternal mental health - as per the question 'why were mothers taking SSRIs during pregancy?' - might be a rather large confounding variable affecting any possible correlation.

Unfortunately even with the Brown paper, the debates will continue as to whether the SSRI-offspring autism correlation is a 'true' correlation or not. Based on the results of 6 studies - "4 case-control studies and 2 cohort studies" - where MMI was adjusted for/restricted to, authors reported some interesting trends. So in their meta-analysis of the data where results from case-control studies were adjusted for a potential impact from MMI, researchers observed that "first trimester exposure remained statistically significant." In "MMI-restricted analyses" covering the same study type, the collected studies did not show any connection between pregnancy SSRI use and offspring autism during either the first trimester or 'any time during pregnancy'. Similar results were found in the cohort studies included in the Brown paper (although both first trimester and 'any point during pregnancy' SSRI use both showed significant correlations to offspring autism in adjusted studies). I might also add that the Brown meta-analysis on this topic is not the only recent addition to the peer-reviewed literature [3]; indeed, there are several [4] others.

"Future studies require robust measurement of MMI prior to and during pregnancy" said Brown et al. I would agree with this sentiment added to the caveat that we may never truly know whether there is a definitive connection between pregnancy SSRI use and offspring autism risk on the basis of observational studies alone. Yes, I know it is unethical to withhold treatment such as SSRIs when clinically indicated even during pregnancy and so investigations utilising this kind of 'interventionist' study design are not likely to be undertaken anytime soon. But it does strike me that we could do quite a bit more modelling any potential effects (or not) in animal studies for example, as per some investigations with fish a while back (see here) as a start.

And finally, although it is not my place to give clinical or medical advice on this blog, I should point out that much like investigations on another medicine prescribed during pregnancy potentially linked to offspring outcomes (see here), SSRIs are not generally given willy-nilly to pregnant women; there are very valid reasons for managing mum's psychiatric health particularly during pregnancy. If anyone is in doubt, please consult your doctor (and not just Dr Google).

To close, before 'fake news' there was The Day Today (and they did it oh so well)...

----------

[1] Brown HK. et al. The Association Between Antenatal Exposure to Selective Serotonin Reuptake Inhibitors and Autism: A Systematic Review and Meta-Analysis. J Clin Psychiatry. 2017 Jan;78(1):e48-e58.

[2] Jakobsen JC. et al. Selective serotonin reuptake inhibitors versus placebo in patients with major depressive disorder. A systematic review with meta-analysis and Trial Sequential Analysis. BMC Psychiatry. 2017; 17: 58.

[3] Kaplan YC. et al. Prenatal selective serotonin reuptake inhibitor use and the risk of autism spectrum disorder in children: A systematic review and meta-analysis. Reprod Toxicol. 2016 Dec;66:31-43.

[4] Kobayashi T. et al. Autism spectrum disorder and prenatal exposure to selective serotonin reuptake inhibitors: A systematic review and meta-analysis. Reprod Toxicol. 2016 Oct;65:170-178.

----------

Brown HK, Hussain-Shamsy N, Lunsky Y, Dennis CE, & Vigod SN (2017). The Association Between Antenatal Exposure to Selective Serotonin Reuptake Inhibitors and Autism: A Systematic Review and Meta-Analysis. The Journal of clinical psychiatry, 78 (1) PMID: 28129495

Shared genetics? Autism, gastrointestinal issues and serotonin

I note the paper by Kara Gross Margolis and colleagues [1] (open-access available here) has been garnering a few media headlines with the suggestion that: "Gastrointestinal [GI] problems in autistic children may be linked to the same genetic mutations that cause other characteristics of autism spectrum disorder."

The study, focusing on the idea that SERT (the serotonin transporter) encoded by the SLC6A4 gene might show some connection to 'some' autism [2], looked to model gastrointestinal (GI) development in a mouse model where SERT function were affected - SERT variant (Ala56). They compared results on various parameters with "those obtained in mice either lacking SERT or treated from gestation to weaning with the selective 5-HT reuptake inhibitor fluoxetine" and wild-type (WT) mice. I should point out that these results were expected to be published around about now and Dr Margolis, a paediatric gastroenterologist, seems to have quite a bit of interest in this whole area.

Results: bearing in mind this was a study of mice, and included only quite a small number of mice, some interesting results are presented: "The ENS [enteric nervous system] was strikingly hypoplastic in SERT Ala56 mice." This 'under-development' of the 'thinking part' of the gut manifested as less neurons being found in various regions of the gut (yes, your gut does house neurons) compared to the other models looked at. The authors suggest that their findings are "consistent with the ideas that defective 5-HT signaling due to the increased 5-HT clearance of SERT Ala56 mice interferes with enteric neurogenesis." 5-HT by the way, is another name for serotonin. Further: "The data are also consistent with the hypothesis that a defect common to the ENS and CNS [central nervous system] could be responsible in ASD [autism spectrum disorder] for comorbid GI disturbances."

They also reported that gut motility - "GI transit time and colonic transit" - was also affected in the SERT Ala56 mice where "slow GI transit [and] diminished peristaltic reflex activity" were more readily present compared to other models. The press release accompanying the study quotes the authors on these points: "Basically, the gut goes slower and the mice were constipated, which is a common complaint in kids with autism." Indeed it is (see here).

And then to quite an interesting piece of information: "The SERT Ala56 mutation decreases crypt epithelial cell proliferation, stunts growth of villi, and decreases mucosal permeability." Decreases mucosal permeability eh? If I'm reading is right (and I'm not expert in this area), all that talk about increased gut permeability (intestinal hyperpermeability) and [some] autism (see here) might have something of a counter-balance under certain circumstances and serotonin (5-HT) might play something of a role...

Finally: "Administration of a 5-HT4 agonist [prucalopride] to Ala56 mice during development prevented Ala56-associated GI perturbations, suggesting that excessive SERT activity leads to inadequate 5-HT4-mediated neurogenesis." The authors caution that although the drug seemed to affect GI issues in the Ala56 mice, this does not necessarily mean that it will do the same in humans carrying the same genetic issue.

Despite the methodological issues associated with this type of study, there are quite a few avenues of further research indicated. Not least is the requirement to identify those people carrying the gain-of-function SERT Ala56 mutation and compare and contrast with other non-carriers in terms of things like gastrointestinal (GI) functions as well as perhaps on other behavioural parameters. This set in the context of functional bowel issues already being suggested to impact on behaviour (see here). The associated idea that offspring of those mice who were treated with the antidepressant fluoxetine during pregnancy might show a distinct pattern of GI issues is also worthy of further study without any big headlines. Accepting that such medication use and risk of offspring autism is still a bit of a scientific hot potato (see here), it looks like a little more science in this area could be indicated.

Music to close, and when following Sunderland Association Football Club, this piece seems particularly apt... (blood pressure starting to return to normal).

----------

[1] Margolis KG. et al. Serotonin transporter variant drives preventable gastrointestinal abnormalities in development and function. J Clin Invest. 2016 Apr 25. pii: 84877.

[2] Muller CL. et al. The serotonin system in autism spectrum disorder: From biomarker to animal models. Neuroscience. 2016 May 3;321:24-41.

----------

Margolis KG, Li Z, Stevanovic K, Saurman V, Israelyan N, Anderson GM, Snyder I, Veenstra-VanderWeele J, Blakely RD, & Gershon MD (2016). Serotonin transporter variant drives preventable gastrointestinal abnormalities in development and function. The Journal of clinical investigation PMID: 27111230

Pregnancy antidepressant use and risk of offspring autism (again)

"Use of antidepressants, specifically selective serotonin reuptake inhibitors, during the second and/or third trimester increases the risk of ASD [autism spectrum disorder] in children, even after considering maternal depression."

That was the conclusion reported in the study by Takoua Boukhris and colleagues [1] dealing with a topic which has previously graced the autism research landscape (see here and see here). Detailing the results of a "register-based study of an ongoing population-based cohort, the Québec Pregnancy/Children Cohort" covering data on all pregnancies in Québec (Canada) between 1998 and 2009 (N=145 456 singleton full-term infants born alive and whose mothers were "covered by the Régie de l’assurance maladie du Québec drug plan for at least 12 months before and during pregnancy"), the Boukhris results have created quite the media stir.

Among the 140,000+ infants followed up, just over 1000 were eventually diagnosed with an autism spectrum disorder (ASD) equating to 0.7% of the population. When researchers looked at those who were prenatally exposed to antidepressants specifically during the second or third trimester of pregnancy, the rate of ASD was calculated at 1.2%. When also controlling for potentially confounding variables such as a maternal history of depression, the elevated risk of offspring autism persisted. Ergo, there may be more to see when it comes to antidepressant use during pregnancy and offspring developmental outcomes. I might also direct readers to an editorial discussing the findings [2].

Most of the media on this latest paper have been quite sensible about the findings. They've for example, pointed out that other research studies have reported slightly less in the way of any connection between pregnancy antidepressant use and offspring autism (see here) as well as putting the results into some context with the idea of what 'elevated risk' might actually translate into (see here). That also antidepressant use during pregnancy is not normally entered into lightly without good reason is something else that I'd bring into proceedings as per other research talking about other pregnancy medication use and potentially elevated risk to offspring outcomes (see here).

The authors do suggest that more research is required to build on their findings and "to specifically assess the risk of ASD associated with antidepressant types and dosages during pregnancy." I would agree that we do need more data on this possible association (including that from animal models and related studies) in order to ascertain whether specific medicine formulations might be more strongly involved and onwards the possible mechanism(s) of effect. I'm not necessarily sold on the idea that serotonin chemistry is specifically the be-all-and-end-all of any effect on the unborn child given what we are starting to realise about the wide-ranging effects of various medicines outside of that listed on the package insert (see here). I am willing however to entertain the idea that the further reaches of tryptophan metabolism might eventually come into the frame (see here). We await further studies.

Music now, and with the imminent launch of a certain Russian Soyuz rocket, a song for Tim...

----------

[1] Boukhris T. et al. Antidepressant Use During Pregnancy and the Risk of Autism Spectrum Disorder in Children. JAMA Pediatrics. 2015. Dec 14.

[2] King BH. Assessing Risk of Autism Spectrum Disorder in Children After Antidepressant Use During Pregnancy. JAMA Pediatrics. Dec 14.

----------

Boukhris, T., Sheehy, O., Mottron, L., & Bérard, A. (2015). Antidepressant Use During Pregnancy and the Risk of Autism Spectrum Disorder in Children JAMA Pediatrics DOI: 10.1001/jamapediatrics.2015.3356

Autism's environmental exposome (part 2)

Back in June 2012, I posted an entry on this blog titled: 'Autism's environmental exposome: fish and pharmaceuticals' covering some work by Michael Thomas & Rebecca Klaper [1] (open-access). In it, authors suggested that unmetabolized psychoactive pharmaceuticals (UPPs) - residues from certain medicines - present in drinking (or in the case of this work, swimming) water may "induce autism-like gene expression patterns in fish."

The UPPs in question were "FLX [fluoxetine], VNX [venlafaxine], and CBZ [carbamazepine] in a 3-component mixture" and the lucky fish volunteers were fathead minnows who got to swim with those UPPs. The data were interesting insofar as the potential "ability to induce ASD-like gene expression patterns in developing brains" as a function of exposure to those UPPs, albeit with concentrations used in the Thomas/Klaper study "higher than observed environmental concentrations". The idea being that drug residues are present in the environment around us and some, either alone or in combination, may potentially host some important biological effects.

Enter then further work from this group in the form of the paper by Gaurav Kaushik and colleagues [2] (open-access) who undertook some rather interesting network analysis among other things and concluded that: "protein products from gene sets with enriched expression in fish brains and human neuronal cells, due to an exposure of psychoactive pharmaceuticals, were comparatively more inter-connected to other neighboring proteins than protein products of non-enriched gene sets." Further: "these genes are more likely to experience altered expression upon exposure to PPCPs [pharmaceuticals and personal care products], causing further dysregulation of the whole interactome due to a ripple effect."

I'll be honest with you and say that I'm not altogether au fait with all the goings-on reported by Kaushik et al and their bioinformatics approach adopted so you'll have to take my interpretation with a pinch of salt.  What they appear to be suggesting is that the effects of UPP exposure may not be just centred on the gene expression patterns they previously reported but rather having something of a wider knock-on effect on how gene products are expressed and how this might map onto something like autism. Interestingly, this time around researchers also introduced valproate (VPA) into their investigations given the growing evidence base that "VPA is known to induce ASD [autism spectrum disorder]-like phenotypes in mice" (see here for more information) as it might in people [3]. They reported some potentially important connections - "enrichment effects of clinical doses of VPA are similar to those for environmental concentrations of pharmaceutical mixtures."

Accepting how the word 'chemical' has been very wrongly demonised over the years, the idea that environmental 'exposures' either singularly or as combinations, might have important effects on development and behaviour is something that requires quite a bit more investigation when it comes to something like autism. The idea that gene expression for example, can be modified by said exposures adds an extra layer of complexity to the rather too simplistic idea of 'genes vs. environment' when it comes to autism risk. One might also be minded to take into account gender/sex (see here) too particularly in light of some of the findings reported by Werling & Geschwind [4] recently...

Oh, and UPPs might not be the only pharmaceuticals requiring further research attention with wastewater and minnows in mind...

Music: Dinosaur Jr. - Freak Scene.

----------

[1] Thomas MA. & Klaper RD. Psychoactive pharmaceuticals induce fish gene expression profiles associated with human idiopathic autism. PLoS One. 2012;7(6):e32917.

[2] Kaushik G. et al. Psychoactive pharmaceuticals as environmental contaminants may disrupt highly inter-connected nodes in an Autism-associated protein-protein interaction network. BMC Bioinformatics 2015, 16(Suppl 7): S3.

[3] Wood AG. et al. Prospective assessment of autism traits in children exposed to antiepileptic drugs during pregnancy. Epilepsia. 2015. 11 May.

[4] Werling DM. & Geschwind DH. Recurrence rates provide evidence for sex-differential, familial genetic liability for autism spectrum disorders in multiplex families and twins. Molecular Autism 2015, 6:27.

----------

Gaurav Kaushik, Michael A Thomas, & Ken A Aho (2015). Psychoactive pharmaceuticals as environmental contaminants may disrupt highly inter-connected nodes in an Autism-associated protein-protein interaction network BMC Bioinformatics

Serotonin - melatonin (and the in-betweeners) linked to autism

The paper by Pagan and colleagues [1] (open-access) looking at "serotonin, melatonin and the intermediate N-acetylserotonin (NAS) in a large cohort of patients with ASD [autism spectrum disorder] and their relatives" set the old grey-pink matter into action recently. Not only because I have some real interest in the starting material for these compounds - the aromatic amino acid known as tryptophan - but because this research group included some quite important analysis of the enzymes involved in the reaction from serotonin to melatonin with autism in mind.

Just before heading into the paper and the details, I'm gonna draw your attention to the picture shown to the right (hand drawn by yours truly) which was originally included in a blog post on something called 5-hydroxytryptophan (5-HTP). As you can see, the source material tryptophan eventually cascades down into various other compounds with serotonin and melatonin in mind. I might add that this is not the only metabolic fate of tryptophan as, for example, per another important compound set: the kynurenine pathway again talked about on this blog.

Serotonin (5-HT) for those who might not know is a neurotransmitter that represents one of the 'S' in the class of medicines called SSRIs hinting at its relationship to mood regulation among other things. Melatonin by contrast has quite an important role in functions like sleep; although, as has been previously mentioned on this blog, melatonin might be quite the molecular handyperson (see here). Both serotonin and melatonin have some history when it comes to autism research and practice (see here for example).

N-acetylserotonin (NAS) is a slightly less well-known compound when it comes to autism. A quick trawl of PubMed using the search term 'N-Acetylserotonin autism' came up with two other entries at the time of writing. Granted both the Anderson-Maes [2] and Carter and colleagues [3] make for interesting reading for different reasons, but there does seem to be a dearth of research on the possibility of a role for NAS for at least some autism.

Now, back to the Pagan paper and a few pointers even though it is open-access:

• The hypothesis: "that (i) the intermediate NAS might also be altered, (ii) alterations of the serotonin-NAS–melatonin pathway might constitute a possible biomarker for a subgroup of individuals with ASD and that (iii) they would be associated with specific clinical profiles."
• Whilst avoiding foods high in tryptophan and/or serotonin such as bananas and chocolate, morning blood samples were provided by "278 patients with ASD, their 506 first-degree relatives (129 unaffected siblings, 199 mothers and 178 fathers) and 416 sex- and age-matched controls" and various parts of the sample assayed for serotonin, melatonin and NAS. The analytical weapons of choice were HPLC (albeit with a rather antiquated method by today's mass spec / NMR standards) and ELISA among other things. A small number of urine samples were also collected and analysed for 6-Sulfatoxymelatonin.
• Results: on the whole, those with autism presented with "elevated whole-blood serotonin" whilst "Plasma melatonin was significantly decreased in individuals with ASD and their relatives compared with controls." These are not surprising results given the research history in this area. 
• With slightly more novelty: "the intermediate metabolite NAS, measured in blood platelets, was found to be significantly elevated in individuals with ASD and their relatives compared with controls." Further such elevations in platelet NAS "strongly correlated" with the plasma melatonin findings noted in cases of autism. 
• There was potentially also something to see when the results were pooled together in terms of discriminating autism from not-autism but I'll leave it up to you to decide how well their biomarkers functioned.
• Bearing in mind my diagram shown above, the increase in serotonin, increase in NAS but decrease in melatonin might provide some important information about where there may be a metabolic 'block'. In this respect, the authors' analysis of "two enzymes, AANAT [Aralkylamine N-acetyltransferase] and ASMT [N-Acetylserotonin O-methyltransferase] known to form protein complexes with 14-3-3 scaffolding proteins" is also important. Actually, authors looked at 14-3-3 in platelets and reported it/them: "significantly decreased in patients with ASD." Previous work from this group [4] had indicated that ASMT activity to be lower in cases of autism; thus suggesting that problems with this enzyme or the availability of this enzyme converting NAS to melatonin might for example, account for the lower melatonin findings.

I've gone on a little bit in this post but hope that you can see the logic in doing so. Metabolic pathways when it comes to human physiology are pretty complex things affected by all manner of variables including things like enzyme function and the availability of those all-important cofactors (see here for some chatter about BH4 for example). Pagan et al have done a good preliminary job of stitching together some important compounds with autism in mind and particularly their findings in relation to NAS. I personally am looking forward to seeing some independent replication of these findings and perhaps onwards, a little more analysis of some other tryptophan derivatives [5] potentially important to [some] autism...

A little music to close: Mark Ronson - Uptown Funk.

----------

[1] Pagan C. et al. The serotonin-N-acetylserotonin–melatonin pathway as a biomarker for autism spectrum disorders. Translational Psychiatry. 2014. November 11.

[2] Anderson G. & Maes M. Redox Regulation and the Autistic Spectrum: Role of Tryptophan Catabolites, Immuno-inflammation, Autoimmunity and the Amygdala. Curr Neuropharmacol. 2014 Mar;12(2):148-67.

[3] Carter MD. et al. Quantitation of melatonin and n-acetylserotonin in human plasma by nanoflow LC-MS/MS and electrospray LC-MS/MS. J Mass Spectrom. 2012 Mar;47(3):277-85.

[4] Melke J. et al. Abnormal melatonin synthesis in autism spectrum disorders. Mol Psychiatry. 2008 Jan;13(1):90-8.

[5] Anderson RJ. et al. Identification of indolyl-3-acryloylglycine in the urine of people with autism. J Pharm Pharmacol. 2002 Feb;54(2):295-8.

----------

Pagan C, Delorme R, Callebert J, Goubran-Botros H, Amsellem F, Drouot X, Boudebesse C, Le Dudal K, Ngo-Nguyen N, Laouamri H, Gillberg C, Leboyer M, Bourgeron T, & Launay JM (2014). The serotonin-N-acetylserotonin-melatonin pathway as a biomarker for autism spectrum disorders. Translational psychiatry, 4 PMID: 25386956

Zinc and depression

"Low dietary zinc intake is associated with a greater incidence of depression in both men and women, as shown in two prospective cohorts".

At the risk of overdoing the whole 'you are what you eat' sentiment, today I'm addressing a portion of the peer-reviewed research literature linking issues with zinc availability to depression. That opening quote by the way, comes from the paper by Khanrin Phungamla Vashum and colleagues [1] who looked at self-reported dietary intake of zinc based on data derived from "the Australian Longitudinal Study on Women׳s Health (women aged 50-61 years) and Hunter Community Study (men and women aged 55-85 years)". I'll come back to that shortly...

Absinthe is the aphrodisiac of the self

As I've mentioned once or twice before on this blog, the description of 'depression' covers quite a lot of diagnostic ground, with all-manner of correlations put forward to try and account for why depression is seemingly so prevalent in modern society. What does seem to be apparent from the voluminous literature is that various factors, at various times and under various situations seem to contribute to depression as a clinical condition with comorbidity also seemingly playing an important hand. In short, it's very, very complicated and very, very individual.

I've tended to talk more about the physiological correlates linked to various types of depression on this blog as per discussions as diverse as gut bacteria (see here) and autoimmune conditions (see here) and upcoming conversations on something like depression and inflammation. This on top of my borderline obsession with all-things vitamin D (see here). It's not that I'm not interested in the psychological or sociological side of things (as per some chatter about overlapping syndromes) but quantifying such factors is rather more difficult than for example, taking a blood sample and looking at the performance of compound X or gene Y. I'd also drop in the fact that quite a bit of the 'causative' research in the area of psychological and/or sociological factors seems to overlook important factors such as resilience too.

Anyhow, aside from the focus on 'self-report' noted in the Vashum study, which is notoriously problematic when it comes to recording eating habits and patterns, there is actually quite a bit of research already published on the topic of zinc and depression.

• The splendidly named Walter Swardfager and colleagues [2] published a very comprehensive review and meta-analysis on the topic of zinc and depression a little while back (covering the peer-reviewed text up to June 2012). Based on data looking at blood zinc levels in over 1500 people with depression compared against 800 asymptomatic controls, they concluded that: "Depression is associated with a lower concentration of zinc in peripheral blood". Indeed, zinc levels were: "approximately -1.85 µmol/L lower in depressed subjects than control subjects".
• Other studies have complemented the Swardfager findings albeit with zinc deficiency present in a more general context in psychiatry. The findings from Grønli and colleagues [3] (open-access) are a good example, whereby researchers reported: "a significant difference in zinc deficiency prevalence between the control group (14.4%) and the patient group (41.0%)" where the patient group consisted of "psychogeriatric patients" who were "compared with the elderly controls". The findings from Maserejian and colleagues [4] (open-access) also suggested some gender effect mediating the zinc-depression link; specifically: "inadequate dietary zinc intake contributes to depressive symptoms in women".
• Insofar as the reasons for zinc deficiency appearing in cases of depression, science is yet to settle on a definitive answer. The paper by Marcin Siwek and colleagues [5] (open-access) suggested three possible reasons: (i) nutritional deficiencies as per the Vashum findings, (ii) "hyperstimulation of the hypothalamic-pituatary-adrenal (HPA) axis, and the associated hypercortisolism" and/or (iii) the result of an inflammatory response "associated with oxidative stress". That last variable on inflammation and oxidative stress might also tie into other data on zinc supplementation "decreasing oxidative stress and generation of inflammatory cytokines such as TNF-alpha and IL-1beta" in certain patient groups [6]. I might also refer you to an excellent post by Dr Emily Deans titled: 'Zinc! An Antidepressant?' with a more detailed analysis of some of the possible hows and whys.
• The supplementation of zinc in cases of depression has also been covered in the research literature. As an adjunctive therapy, Ranjbar and colleagues [7] (open-access) reported that: "zinc supplementation together with SSRIs antidepressant drug improves major depressive disorders more effectively in patients with placebo plus antidepressants (SSRIs)". A review of some of the controlled trials prior to the Ranjbar results by Lai and colleagues [8] concluded similar things with regards to zinc as an 'add-on' treatment but that: "There is less clear evidence on the effectiveness of zinc supplementation alone on depressive symptoms of non-depressed healthy subjects". 
• Animal studies have complemented this collected literature as for example, reported by Tassabehji and colleagues [9] looking at rats; the authors suggested that: "zinc deficiency leads to the development of depression-like behaviors that may be refractory to antidepressant treatment".

There is quite a compelling scientific case for far greater research inspection of zinc in relation to depression and for example, further working out what it seems to be doing. Important too is the issue of who might be the best-responders to something like zinc supplementation in relation to depression/depressive symptoms.

That being said, I wouldn't want anyone to assume that I'm advocating zinc supplementation as some sort of cure-all for depression or anything else. To repeat myself: depression is a very complex set of conditions combining both biology and psychology. Science is still feeling its way around this area, despite the importance of nutrition to depression being increasingly recognised (see here).

So... Cosmic Girl by Jamiroquai.

----------

[1] Vashum KP. et al. Dietary zinc is associated with a lower incidence of depression: findings from two Australian cohorts. J Affect Disord. 2014 Sep;166:249-57.

[2] Swardfager W. et al. Zinc in depression: a meta-analysis. Biol Psychiatry. 2013 Dec 15;74(12):872-8.

[3] Grønli O. et al. Zinc deficiency is common in several psychiatric disorders. PLoS One. 2013 Dec 19;8(12):e82793.

[4] Maserejian NN. et al. Low dietary or supplemental zinc is associated with depression symptoms among women, but not men, in a population-based epidemiological survey. J Affect Disord. 2012 Feb;136(3):781-8.

[5] Siwek M. et al. Zinc as a marker of affective disorders. Pharmacol Rep. 2013;65(6):1512-8.

[6] Prasad AS. Zinc: role in immunity, oxidative stress and chronic inflammation. Curr Opin Clin Nutr Metab Care. 2009 Nov;12(6):646-52.

[7] Ranjbar E. et al. Effects of zinc supplementation in patients with major depression: a randomized clinical trial. Iran J Psychiatry. 2013 Jun;8(2):73-9.

[8] Lai J. et al. The efficacy of zinc supplementation in depression: systematic review of randomised controlled trials. J Affect Disord. 2012 Jan;136(1-2):e31-9.

[9] Tassabehji NM. et al. Zinc deficiency induces depression-like symptoms in adult rats. Physiology & Behavior. 2008; 95: 365–369

----------

Vashum KP, McEvoy M, Milton AH, McElduff P, Hure A, Byles J, & Attia J (2014). Dietary zinc is associated with a lower incidence of depression: findings from two Australian cohorts. Journal of affective disorders, 166, 249-57 PMID: 25012438

Prenatal SSRI exposure and autistic traits

A quote to start today's post: "Our results suggest an association between prenatal SSRI exposure and autistic traits in children". That was a primary finding reported by Hanan El Marroun and colleagues [1] who looked at whether maternal depressive symptoms or a class of quite commonly used pharmaceutics - the selective serotonin reuptake inhibitors (SSRIs) - used to manage depressive symptoms, during pregnancy might impact on offspring development.

"Everything the light touches is our kingdom" 

Before progressing through some of the details around this area, I'm going to also direct your attention to a couple of important accompanying commentaries on the Marroun findings from Jones & McDonald [2] and Petersen and colleagues [3] (open-access). Both caution about reading too much too soon into the reported association between SSRIs and offspring outcomes, and the very real outcomes that can come about if psychiatric issues such as depression are not properly managed. Something I think most people might have heard about recently.

A few details about the Marroun paper might be useful:

• Following some previous discussions correlating maternal SSRI use during pregnancy and offspring outcome with autism in mind (see here) including the quite recent papers by Harrington and colleagues [4] and Rai and colleagues [5], the authors looked to "prospectively determine whether intra-uterine SSRI exposure is associated with childhood autistic symptoms in a population-based study". 
• "A total of 376 children prenatally exposed to maternal depressive symptoms (no SSRI exposure), 69 children prenatally exposed to SSRIs and 5531 unexposed children were included" for study. The commentary from Petersen et al notes how small a group were actually exposed to SSRIs and how "these numbers rapidly dwindled when it came to the measurement of the outcome".
• The Child Behavior Checklist and Social Responsiveness Scale (SRS) were used to assess "pervasive developmental and affective problems" and "autistic traits" respectively. 
• Results: aside from an association between prenatal selective serotonin reuptake inhibitor (SSRI) exposure and autistic traits in children, researchers also reported that: "Prenatal exposure to maternal depressive symptoms without SSRIs was related to both pervasive developmental (odds ratio (OR) = 1.44, 95% CI 1.07-1.93) and affective problems (OR = 1.44, 95% CI 1.15-1.81)". The suggested link between maternal depressive symptoms and autistic traits was to some degree weaker than the SSRI exposure correlation.
• The authors conclude that: "Long-term drug safety trials are needed before evidence-based recommendations are possible" as once again I'll direct you to the Jones and Petersen commentaries.

In the same way that the emerging data on prenatal valproate exposure *might* link into offspring outcome including the presence of autism (see here), so the Marroun paper potentially adds another medicine to the list. I would perhaps temper that last sentence by adding that the valproate story is perhaps a little further along in terms of rodent models of prenatal valproate exposure mimicking some features of autism (see here) and the data providing something like mechanisms to be looked at with further investigations in mind [6]. Still, the CDC Treating for Two initiative might be once again relevant.

The added complication with the SSRI-autism correlation is the discussion about maternal depressive symptoms also potentially mediating any link with offspring autism or autistic traits. The paper by Sørensen and colleagues [7] (open-access here) kinda hinted that this and other important confounding factors might impact on any studies of association, including details like: "paternal antidepressant use during the time of pregnancy was not associated with an increased risk of autism spectrum disorders, except for a 30% increase when the fathers took SSRI". Even more recently Clements and colleagues [8] talked about how maternal "major depression" confounded any medication relationship with offspring presentation. They also talked about a link with ADHD which brings me back to yesterday's post on comorbidity (see here)...

One would do well not to discount such confounding factors at this stage. Indeed, if one assumes that depression might have a physiological link to something like inflammation for example [9] we then start to arrive at the increasingly important research looking at maternal inflammation as being a risk factor for offspring autism (see here). And before you ask, yes, C-reactive protein (CRP) has been linked to depressive symptoms as per the meta-analysis by Valkanova and colleagues [10].

The Marroun results are interesting and add something to an increasing bank of peer-reviewed literature [11] suggestive of a possible link between SSRI use during pregnancy and offspring outcomes. On the basis of the current existing literature and with my blogging caveat of no medical advice given or intended, I would be minded to conclude that there is quite a bit more experimental investigation to be done on this category of medicines. But I don't yet think there is enough clear evidence to conclusively put an elevated risk of offspring autism on the list of potential side-effects of these medicines.

Music then. Scissor Sisters and Laura.

----------

[1] Marroun HE. et al. Prenatal exposure to selective serotonin reuptake inhibitors and social responsiveness symptoms of autism: population-based study of young children. The British Journal of Psychiatry. 2014; 205: 95-102.

[2] Jones I. & McDonald L. Living with uncertainty: antidepressants and pregnancy. The British Journal of Psychiatry. 2014; 205: 103-104.

[3] Petersen I. et al. Prenatal exposure to selective serotonin reuptake inhibitors and autistic symptoms in young children: another red herring? The British Journal of Psychiatry. 2014; 205: 105-106.

[4] Harrington RA. et al. Prenatal SSRI Use and Offspring With Autism Spectrum Disorder or Developmental Delay. Pediatrics. 2014 Apr 14.

[5] Rai D. et al. Parental depression, maternal antidepressant use during pregnancy, and risk of autism spectrum disorders: population based case-control study. BMJ. 2013 Apr 19;346:f2059.

[6] Bambini-Junior V. et al. Prenatal Exposure to Valproate in Animals and Autism. Comprehensive Guide to Autism. 2014: 1779-1793.

[7] Sørensen MJ. et al. Antidepressant exposure in pregnancy and risk of autism spectrum disorders. Clin Epidemiol. 2013 Nov 15;5:449-59.

[8] Clements CC. et al. Prenatal antidepressant exposure is associated with risk for attention-deficit hyperactivity disorder but not autism spectrum disorder in a large health system. Molecular Psychiatry. 2014. August 26.

[9] Berk M. et al. So depression is an inflammatory disease, but where does the inflammation come from? BMC Med. 2013 Sep 12;11:200.

[10] Valkanova V. et al. CRP, IL-6 and depression: a systematic review and meta-analysis of longitudinal studies. J Affect Disord. 2013 Sep 25;150(3):736-44.

[11] Rais TB. & Rais A. Association Between Antidepressants Use During Pregnancy and Autistic Spectrum Disorders: A Meta-analysis. Innov Clin Neurosci. 2014 May;11(5-6):18-22.

----------

Marroun, H., White, T., van der Knaap, N., Homberg, J., Fernandez, G., Schoemaker, N., Jaddoe, V., Hofman, A., Verhulst, F., Hudziak, J., Stricker, B., & Tiemeier, H. (2014). Prenatal exposure to selective serotonin reuptake inhibitors and social responsiveness symptoms of autism: population-based study of young children The British Journal of Psychiatry, 205 (2), 95-102 DOI: 10.1192/bjp.bp.113.127746