Adventures in Accutane

Jezebel has a post up discussing the link between Accutane and suicide.  I link to the Jezebel post and not the new article because I find the comments interesting.

First, some background.  Accutane (generic name: Izotretinoin) is mostly an anti-acne medication, although it’s occasionally used for other skin disorders and cancers.  Depending on who you ask, it’s either a form of vitamin A or a close relative of vitamin A.  Vitamin A is fat soluble, which means if you take too much it will kill you (the livers of many arctic mammals are toxic for this reason).  One to two courses of Accutane (generally lasting 8-10 months) is usually enough to permanently cure severe acne.

Izotretinoin is a notorious teratogen, meaning it can cause serious birth defects if taken during pregnancy.  The cause appears to be that izotretinoin serves as a signalling molecule in a certain stage of pregnancy, but this is not definite.  In America, the relevant legal authority has responded with the iPLEDGE system.  Only prescribers registered with iPLEDGE can prescribe accutane. They must register recipients (with a detailed sexual and menstrual history) in the system and certify that they’ve been  warned about the risk of birth defects, and promised to take two forms of birth control.  Prescriptions must be picked up in a seven day window, lost prescriptions cannot be replaced, a prescription must be for exactly 30 days.  Recipients must take a blood (not urine) test for pregnancy every month before getting their prescription.  They recently relaxed the latter requirement for men and women of non-child-bearing age to pledge to take birth control, but still require a blood test.

This program represents some of the worst government interventions in health to me.  It imposes huge burdens of time, money, and privacy.  It infantilizes women by requiring a pregnancy test even if they report no sexual activity, yet takes their word on the whole birth control thing.  It very nearly requires hormonal birth control, and while I recognize hormonal birth control as a life changing intervention for many people, I think the dangers are understudied and under reported and dislike anything that dismisses its costs.  The timing restrictions seem to be designed to force users to adhere to pregnancy tests, which are predicated on the idea that you can’t trust women to notice they’re pregnant.  And of course, you have to do all this for people without working uteruses.  It’s also considerably more work than you need to do to take thalidomide, an even more notorious teratogen, which means the decision to impose iPLEDGE is being made based on something other than strict rational consideration of costs and benefits.

No one is quite sure how izotretinoin works.  It may induce cell death in the sebaceous glands.  It may act as an anti-microbial.  It may affect the expression of hundreds of genes.  It has shown promise as a cancer treatment, which means it probably inhibits cell growth.

Some people report that Accutane has psychological side effects, including suicide.  This is really tricky.

A lot of Accutane users are teenagers, who are already an emotionally chaotic group.  You can compensate for differing risk with a control group, but a higher variance is still more likely to lead to a false report of differing means.

Second, suicidality and depression are not as linked as you would think.  People at the absolute pit of depression are often too overwhelmed and apathetic to kill themselves. Suicide risk actually increases when treatment starts working (which may be why anti-depressants are linked to suicide in teenagers, and is definitely why mixed states are so dangerous).  Severe acne is socially isolating, which can lead to depression.  It’s entirely possible that getting better frees up enough emotional energy to feel bad, or that the first set back after an improvement feels unmanageable, and so while the drug has no direct psychoactive effects, it looks like it does through its effects on the skin.

On the other hand, it’s a known biohazard with no accepted explanation for why it works, and almost any disruption in physical systems can manifest as depression and anxiety.  The weird thing would be if it didn’t.  The Jezebel comment thread is full of people saying it induced terrifying psychosis in them.  It’s also full of people saying it saved their lives by clearing up their acne.  A 24 mouse study showed some increase in depression-like behavior in mice following injection.   An even smaller study showed that Accutane reduced cell growth in certain regions of the brain, which is terrifying for a drug we give to teenagers but hard to draw concrete conclusions from.

So what do you do, as a potential patient or parent of same, in the absence of good evidence?  Clearly you need to watch new patients closely for behavioral changes and discontinue if they occur, but what about the risk of subtle long term damage?    I don’t know how to weigh that.

I can fix anything/if you let me near

Health stuff is often cyclic, and I’m in the unpleasant part of the cycle right now.  When the first derivative is negative it’s easy to feel like everything will get worse forever, so here’s a list of things that are objectively better than they were a year ago, even after the recent backsliding.

1.  My overbite.  I’ve had it since childhood.  I was prescribed braces, but my mom eventually convinced my dentist my teeth were so bad that the increase in plaque due to difficulty in brushing would be even worse for me.  Two months ago, it disappeared.  I still pull my lower jaw in when I chew because I ground my back teeth down in the pattern of my overbite, but the resting position is now almost perfect.

2.  My dental health.  I’ve had on 4x/cleanings since childhood, and my last dentist spoke in terms of great Game of Thrones like battles to fight back the plaque.  She noticed a marked improvement as soon as I went on HCl, without me telling her anything had changed.  I just started with a new dentist and he’s less bracing for winter and more trying to make sure things tay as awesome as they are now.

3.  Oranges.  I used to hate oranges.  Two or three drops of orange juice in a full glass of water were all I could stand.  Now I eat them whole, without HCl.

4.  Auditory sensitivity (increased).  When I noticed discontinuities in the sound of my last audiobook (slight changes in the volume or tone) I put it down to cheap recording not standardizing between sessions.  But my current book does it too.  I wonder if every book has these, and I just didn’t notice until now.

When ethics is ugly

I am so tired of medical shows having brave doctors lie to the evil transplant committee that is heartlessly denying a transplant request because the patient is deemed ineligible due to some stupid technicality.  Most recently I saw this on House, where a woman destroyed her heart with bulimia, which is apparently a disqualification for transplant, but it comes up on practically every medical show eventually.  The transplant rules didn’t arise out of some Puritan sense of punishing people for their sins: it’s because eating disorders have a high relapse rate, and a relapse would destroy the new heart too.  By lying and getting her the heart, Dr. House killed whoever would have gotten the transplant instead. 

I’m sure that transplant committees make mistakes, that they’re manipulated by fame and money, and that the guidelines are imperfect.  But they’re not there to be mean, or to “punish” people for their transgressions.  They’re there because there are far fewer donor organs available than are needed, and in light of that scarcity, we want to make sure the organ goes to the person who will get the most additional life out of it.   If a patient has a condition that will lessen the use they get out of the heart, that lowers their priority for a transplant.

Maybe I’m wrong.  I don’t know what went into creating the rules, maybe they’re biased and terrible.  But if so, the appropriate reaction is to say that and fight them.  Not kill someone else because it will make you sad to tell your patient they’re going to die.

Similarly, House, stop lying to get ineligible patients into clinical trials.  You mean well, but there are two options:  the treatment doesn’t work, and you bought your patient nothing, or it does, and you just artificially worsened the numbers, potentially taking it away from everyone else who would have benefited from it.

Prior assumptions

Frequentist statistics are what you learned in school if you’ve only had a little bit of statistics.  You have a null hypothesis, and you attempt to gather data to prove it’s not true.  You do this by showing the odds that the data you collect would occur if the null hypothesis was true.  E.g. the odds of a fair coin coming up heads 10 times out of 10 is 1/(2^10) = 0.0009765625 (this is known as the p value).  Generally we consider the null hypothesis disproven if the odds of the results under the null hypothesis are <= 0.05, so you could safely conclude that this coin was not fair.  Similarly if you have four schools that draw from similar populations and have similar teachers, and one scores three orders of magnitude better than the others on standardized tests, a frequentist will conclude that that school is a lot better than the others.
Bayesian statistics on the other hand asserts that we know something about the likely distribution before we collect the data, and it is good to use that data.  In that school example, a bayesian might say “well, these schools are very similar, so that difference probably mostly represents a statistical fluke.”  To me, who was raised in a frequentist household, this looks like an exercise is assuming what you’re trying to prove.  But I think I finally see its value.

The other day I posted a link to this blog post on facebook.  The post reported on a study that showed a large difference in how law partners reacted to the same paper when told it was written by a black or white student- they wrote more negative comments, were less encouraging, and found more typos.  A friend of mine responded with “I believe there is systemic racism, but…”, which is never a great start.  He went on to make a bunch of criticisms which would have been  extremely fair to make in a journal club discussing a peer reviewed study, but in context came across as reinforcing the cultural pattern of “I refuse to believe in racism is a factor until it’s proven in each specific case.” Forcing victims of systemic discrimination to prove the discrimination gives them an additional burden on top of all that discrimination.  It’s not right and it’s not fair.  It’s also heavily intertwined with the idea that falsely believing something was affected by race is worse than ignoring discrimination.  

At the same time, I dislike the argument “you’re technically correct ” (which he absolutely was- the study was done by some consulting firm and didn’t even include statistics beyond the mean) “but don’t say it because it will encourage racists.”  Nothing good comes from quieting science.

What I eventually realized was that what we really needed was to argue our bayesian priors.  His frequentist criticisms were implicitly asserting a null hypothesis of “no racism”, while I trying to include the that we knew racism existed and shoulder consider that when evaluating the study- e.g. I wanted to do a Bayesian analysis with “some racism” as a prior.  Once we said this we could have a sensible argument about bayesian priors.  We didn’t, because by the time I thought of this we’d already reached the point where the conversation needed to end.  But we could have.
So now I see the value of Bayesianism.  Frequestism is much harder to manipulate, but that also makes it less sensitive to reality.  You make your best guess at which technique works best for your particular question, and then you accept that involving numbers does not magically make you unbiased.

First Impressions: It’s Not Just Who You Know, by Tommy Spaulding.

I have a long and antagonistic relationship with social skill how-to books, starting with my award winning 8th essay about How to Win Friends and Influence People.  The essay was about how it didn’t teach you how to make friends, it taught you how to suck up to people, and my award was that I got to stop reading the book.*  As an adult I realized that HtWFaIP couldn’t possibly have stayed that popular for that long for no reason and gave it another shot.  The most charitable thing I could say was that it was for incompetent or unconfident extroverts.  I was a profound introvert, and Carnegie genuinely did not seem to get the idea that talking to people could be draining.

This would become a theme with me.  I am eternally grateful to the friend who loaned me Crucial Conversationsbecause it immediately improved my life by 15%, and has never stopped paying dividends.  How to Talk so Kids will Listen and Listen so Kids will Talk and John Gottman’s books on marriage** were not quite as revolutionary, but I enjoyed reading them and knew they would have been useful had Crucial Conversations not gotten there first.  But those were books focused on doing emotional work in deep relationships.  Books about professional networking or making friends tended to leave me just as angry as HtWFaIP, for similar reasons.  They assumed that everyone wanted the things extroverts wanted, and that they could get them by acting like extroverts.  I didn’t need someone to encourage me to speak up at corporate dinners, I needed either a list of tricks to make them less energetically costly for me or a less painful way to get the benefits of attending said dinners.***  Or they imply there’s something wrong/selfish/possibly evil about me not wanting to engage with every person I meet.

Along with the rest of the internet, I went through an introverts rights phase, reading Quiet and Introvert Power.  Those made me feel better about myself, and helped me socially by making me feel more comfortable advocating for my needs.  But they were never intended to be how-to books.

Now I am reading It’s Not Just Who You Know. It’s pitched as “How to Win Friends and Influence Peoplebut for emotionally meaningful relationships.”, which so far has been ragingly inaccurate, but I’m only 30% in, and what it’s been so far has been even better.

So far INJWYK has been a memoir, starting in high school, of how Tommy Spaulding’s social skills and resulting social connections have helped him.  For a while I struggled with jealousy at what Spaulding got through socializing and I thought be “earned.”  What I eventually realized was this:

Tommy Spaulding talks to people the way I read books.

That is, sometimes he has a very explicit goal oriented exchange (like, I don’t know, buying a shirt.  He doesn’t devote much book time to these interactions) just like I sometimes read a technical manual to extract a specific fact.  But mostly, he talks because talking is a thing he enjoys doing.  He does some research to direct his efforts to the people he’s most likely to find most interesting, and he’s worked out techniques to make those conversations as interesting and rewarding as possible.   If nothing useful ever came out of it he would probably change his approach.  But he is not approaching anyone with the idea of what they can do for him, and he’s very conscious of the value of other people’s time.  I am never, ever going to talk to as many people as Tommy Spaulding does.  But I respect and value his approach, in a way I never did Dale Carnegie.

Additionally, this knowledge has helped me stop resenting people who get nice things via social interaction.  Those people benefit from their hours of socializing.  I benefit from my hours of reading.  If I’m going to complain about the unfairness of people benefiting from social connections, I’d better be prepared to give up the advantages I gain from reading.

*That was the year I was homeschooled.  The primary benefit of homeschooling was no longer attending a school where I had a perfectly reasonable fear of being stabbed, but the option to reason my way out of dumb assignments was nice too.

**I have never been married and have no kids, but I’d heard really good things and thought the principles might be useful.  I was right.

***For example, suppose your goal was meet new people and advertise your professional skills, with some thought to advancing your career.  You could go to to industry social events and apply HtWFaIP.  But if you’re a programmer you can get many of the same benefits by contributing to open source projects. Yes, it’s a lot of hours of work for the amount of recognition and networking you achieve.  But if you love programming and hate talking to strangers in crowds, it’s worth considering as a partial replacement.

Depression and Inflammation

Now that I know what inflammation is and have given you my personal view of depression, I can talk about depression as an inflammatory disease.  Most specifically, the article “So depression is an inflammatory disease, but where does the inflammation come from?” from BiomedCentral Medicine.  I’ll do some additional research as warranted, but mostly I want to talk about this article.  Prepare for deep dive.

Background: We now know that depression is associated with a chronic, low-grade inflammatory response and activation of cell-mediated immunity, as well as activation of the compensatory anti-inflammatory reflex system. It is similarly accompanied by increased oxidative and nitrosative stress (O&NS), which contribute to neuroprogression in the disorder. The obvious question this poses is ‘what is the source of this chronic low-grade inflammation?’


  • As I painstakingly established last week, inflammation means a chronic dilation of blood vessels allowing more stuff to leak into tissue.
  • cell-mediated immunity involves producing new white blood cells in response to a specific antigen, as opposed to producing new antibodies in response to a specific antigen, or non-specific immunity that is not a response to anything in particular (example: skin).
  • Almost every system in your body is moderated by a complex series of signals that can aid or inhibit one another.  Rather than just produce less of an activator to reduce a reaction, your body instead introduces an inhibitor.   Having high levels of both activators and inhibitors has similar primary effects to having low levels of both, but can be either cause weird secondary effects or be a symptom of something weird going on.  It’s also plain more stressful, just like keeping yourself steady against a strong wind is harder than standing still.  “[A]ctivation of the compensatory anti-inflammatory reflex system” implies that depressed people are falling in the “high levels of both” category.
  • In the course of the day, your body naturally produces many types of molecules.  Some of those molecules contain oxygen, or nitrogen and oxygen, and a net negative charge.  These are the home wreckers of the molecular neighborhood, luring parts of other molecules to dump their other halves and bond with them instead.  This is a problem if that other molecule is doing something important, like being the blueprint for every other protein in your body.  The damage these homewreckers cause is called oxidative stress.  Your body has ways to contain it, but not perfectly.  Oxidative/nitrosative stress can arise either from producing too many of the reactive molecules, or from a deficiency in the post-processors.
  • Fun fact: we don’t know if Nitric Oxide is a pro- or anti-inflammatory messenger.  It looks the answer may be “anti- under normal circumstances, pro- under times of stress”, which implies but doesn’t prove that it’s primary effects are anti-inflammatory, but the containment mechanisms break down under stress and cause a lot of damage.

Discussion: This review explores the role of inflammation and oxidative and nitrosative stress as possible mediators of known environmental risk factors in depression, and discusses potential implications of these findings. A range of
factors appear to increase the risk for the development of depression, and seem to be associated with systemic inflammation; these include psychosocial stressors, poor diet, physical inactivity, obesity, smoking, altered gut permeability, atopy, dental cares, sleep and vitamin D deficiency.


  • Two very common bad things seem to have a lot in common.  I think there’s a lot of merit to their hypothesis, but this venn diagram does not impress me.

Summary: The identification of known sources of inflammation provides support for inflammation as a mediating pathway to both risk and neuroprogression in depression. Critically, most of these factors are plastic, and potentially amenable to therapeutic and preventative interventions. Most, but not all, of the above mentioned sources of inflammation may play a role in other psychiatric disorders, such as bipolar disorder, schizophrenia, autism and post-traumatic stress disorder


  • If we are right, this suggests several really easy ways to treat, cure and prevent depression, and potentially other psychiatric issues.

Oh wow, apparently you can induce depression-like symptoms by activating the immune system, either by triggering it with a toxin or just injecting some cytokines (a broad class of messenger, many of which signal the immune or inflammatory system) directly.  SSRIs (the most common class of anti-depressants) decrease production of proinflammatory cytokines,* and a return of symptoms is often associated with the loss of these effects.

The authors then go on to cover in detail various things you already knew were bad or good for you (Sleep- good.  Chronic stress- bad).  The research isn’t at the stage where it’s generating novel recommendations yet.  Which is not a criticism, it’s an impressive set of basic science work and it takes time to translate that into practice.If you are deeply interested in the biochemistry I encourage you to check it out, but I have no more to add.

*SSRI stands for selective serotonin reuptake inhibitor.  Because serotonin is a neurotransmitter, we originally assumed SSRIs treated depression by making serotonin more available.  Except serotonin levels change within hours of taking an SSRI, but the anti-depressant effects takes weeks to show up (unless you’re bipolar, in which case they can trigger a manic phase almost immediately).

Chronic and Systemic Inflammation

If I’m reading this correctly, there is no bright line between acute and chronic inflammation.  Chronic inflammation is acute inflammation that didn’t go away.  Over time this may lead to accumulated effects, and the distribution of mediators and their sources may shift, but there’s a lot overlap.

There’s a few ways chronic inflammation can hurt you.  One is that is releases immune cells (aka white blood cells)  into your tissues.  Didn’t I list that as one of the features of inflammation last time?  Yes, yes I did.  When you have an infection in the tissue, you want immune cells there to go after it.  But if letting immune cells run around organ tissue didn’t have any side effects, we’d do it all the time.    Immune cells are very very good at distinguishing foreign body from host , but not perfect.  The more time they spend in the trenches, the more likely they are to misidentify a host protein as dangerous and attack.  In the worst case this triggers a really ugly autoimmune disorder.  In a milder case it triggers more inflammation.   I think you can see where this is going. So much like being in a hospital or on crutches, you want your immune cells to be circulating in tissue for as long as you need, but no longer.

Similarly, some of the proteins released during inflammation (aka positive acute-phase proteins), which do useful things like signal inflammation, coagulate blood, or retard microbial growth, can change shape* and become insoluble.  These form fibrous  masses in intracellular space known as amyloids.  Amyloids are or are associated with some of our most terrifying diseases, like Alzheimers and Mad Cow Disease, and the amyloids from inflammation can contribute to these, although they’re not necessarily the most important part.

Chronic inflammation can be self-reinforcing in other ways.  Swelling is just about the least useful reaction to an ingrown toenail, but it’s what we do.

Everything I’ve talked about so far has been local inflammation, where the reaction is contained to one identified area.  There is also systemic inflammation, in which the vasodilation is body-wide.  It’s associated with all sorts of bad things, including overeating and obesity, which immediately makes me think we don’t know the real issue because both those topics are moral panics in our society.

*Quick lesson in proteins.  Proteins are made of up a long string of amino acids.  The order of amino acids is called the primary structure.  Individual bits of those string fold themselves into 3-d structures such as the ß-Hairpin and  Greek Key.  Those structures, also known as motifs, make up the secondary structures.  The motifs in turn interact to form the total three dimensional shape of the protein, which is the tertiary structure.  The tertiary structure is determined by the electrical charges and physical shape of the primary structure.

In general, a given primary structure has a single tertiary structure, because there’s only one island of stability.  However it is possible for a second island to exist, and the protein to convert to it- either it’s more stable in a new environment, or something catalyzed the reaction and it can’t go back.  Very, very rarely, this new tertiary structure is capable of catalyzing other proteins of the same type to its new shape.  These are known as prions (e.g. Mad Cow Disease) and they awesome and terrible.