Things You Learn When Making Psychopharm Lectures

Some people do monthly newsletters with interesting links, I do sporadic newsletters with psychiatry trivia.

Sorry for the lull in writing, August and September have been really busy. Most of the time that I would spend writing essays has been spent doing a large amount of reading to prepare basic psychopharmacology lectures that I’m giving to our PGY-1s and PGY-2. This is just a collection of some unanswered questions I have, pieces of historical trivia, clinical pearls, and other assorted minutiae that maybe you will find interesting..

General/Non-Psychiatric Things

1. Beware of drawing too many conclusions about how a drug functions based off of its inhibitory constant (Ki), especially for experiments in vitro. The first three paragraphs of the discussion section in this paper looking at the NET occupancy of venlafaxine (Effexor) highlights just how huge the measurement variation can be.

2. The more you read about the history of drug development, the more you realize how much we’ve discovered through dumb luck. Antipsychotics? Lucky that chlorpromazine causes a mild hypothermia. Lithium? Lucky that lithium makes uric acid very soluble. Valproate? Lucky that the Germans tried turning coal into margarine. (Expanded stories on all of these below)
   
   The point is, don’t assume that the medications we have are the product of some well-thought out, first principles based approach to drug development.

Antipsychotics

1. Chlorpromazine (Thorazine), the first antipsychotic, was originally used in the 1950s as an adjunct to surgical anesthetics for its ability to lower body temperature. At the time, inducing a mild hypothermia with cold water was used as a method to control agitation and so the logic was that chlorpromazine’s ability to induce hypothermia would do the same.

2. Ever notice that the side-effects of the TCAs and the more sedating antipsychotics (e.g. clozapine, olanzapine, chlorpromazine) are suspiciously similar? Well, imipramine — the first TCA — is a derivative of chlorpromazine!

3. Most antipsychotics cause some amount of weight gain, but the rate to which they cause other metabolic problems like high cholesterol, elevated blood glucose, and insulin resistance is highly variable. Generally, the so-called ‘-pines’ (olanzapine, quetiapine, and clozapine) seem to have the highest metabolic risk of them all. You might think that this is purely a product of excessive weight gain, but a study in healthy adults showed that olanzapine significantly increased the AUC on an oral glucose tolerance test, elevated fasting leptin and triglycerides, and decreased HDL after just 3 days!
   
   Some recent research has discovered that D2-like and D3-like receptors in pancreatic islet cells are responsible for modulating insulin release in response to elevated blood glucose, and that blocking those receptors increases insulin release which could lead to the increased insulin resistance and T2DM. Interesting… but if this mechanism is important, how can a drug like haloperidol — one of the most potent D2/D3 antagonists — be one of the few antipsychotics that is clearly not associated with weight gain whatsoever?
   
   My original guess was that there’s a sort of two-hit model, where the increased appetite (and thus weight gain) caused by H1 antagonism — which is seen most heavily in the -pines — is the second piece of the puzzle. Of course, this is ruined by data from the latest and most rebellious of the -pines, asenapine (Saphris). Which, despite being quite a quite potent antagonist of D2, D3, and H1, does not appear to cause weight gain relative to placebo. Agh!

Antidepressants

1. Trazodone was developed to target the α1-receptor by two Italians who believed that psychic pain was caused through the same neural mechanisms as physical pain. They did manage to make an antidepressant, but probably only because Trazodone is a weak SSRI. I wrote about this bit of trazodone’s history, as well as its current use as a sleep aid in my very first essay on Substack: Wot’s… Uh The Deal With Trazodone?

2. The modern dietary restrictions for MAOIs¹ are probably much less restrictive than you think they are. These dietary restrictions are built around limiting ingestion of tyramine — a product of fermentation and food spoilage — to prevent hypertensive crises since MAO isn’t around to break it down. Classically, medical students are taught that patients have to scrupulously avoid common foodstuffs like wine, beer, aged meat and cheeses, hotdogs.
   
   You may be surprised to learn that — since the MAOIs hit the market in the 1950s — there has been considerable progress made in food preparation and preservation techniques that have driven down tyramine levels in most foods to the point where this is a total non-issue. Dr. Ken Gillman, an authority on the MAOIs, has both a long and short overview of modern MAOI dietary restrictions at his invaluable website Psychotropical.

3. Venlafaxine, at doses <75mg, is basically just a SSRI. It doesn’t become a true SNRI until north of 150mg, which might explain why it’s one of the few antidepressants that shows clear benefit with doses towards the higher end of its dosing range.

   1. This is probably also the case for duloxetine too, but the dosing starts at 60mg which is right around where it hits 50% NET occupancy.

4. The two MAOIs we’re most familiar with in the US — tranylcypromine and phenelzine — are the two survivors of more than a dozen MAOIs that made it to market and were subsequently withdrawn. Why? Well, you might be clued in by the fact that most of them belonged to a class called the hydrazines, because they are derivatives of common rocket-fuel compound hydrazine. Turns out that many of the MAOIs inherited their parent compound’s fondness for destroying livers, and all except for phenelzine — and the rarely used isocarboxazid — have been pulled from the market.

5. Iproniazid — the first MAOI — was originally used as a treatment for tuberculosis. Doctors eventually noticed that their patients were quite happy, despite being alive in the 1950s prior to the discovery of the now famous emotion, Love, in the summer of 1967.

Benzodiazepines

1. Alprazolam (Xanax) is a benzodiazepine that is traditionally discussed as having rapid onset and a very short duration of effect. The typical explanation is that this is because it has a very short half-life. In reality, its half-life is pretty much the same as lorazepam’s (Ativan) - about 12h in the average individual. I’m working on an essay in which I try to chase down some sort of plausible explanation for its reputation. (Edit: Here’s the essay)

2. In 1989, New York State created a perfect natural experiment to test whether or not benzodiazepines increased the risk of falls in the elderly. They made benzos so hard to prescribe that the number of prescriptions dropped by 55%. Neighboring New Jersey — conveniently home to a demographically very similar population — didn’t change a thing and their rates of benzodiazepine prescribing stayed the same. A group of researchers published a paper in 2007 analyzing the rates of falls in Medicaid enrollees between the two states before and after the NY policy was enacted and found no difference in the rate of hip fractures.

3. Lorazepam, oxazepam, and temazepam² are favored in patients with liver failure, but this is incorrectly attributed to the idea that these drugs are “not hepatically metabolized.” The truth is that they are hepatically metabolized; it’s just that they undergo hepatic glucuronidation which is not substantially impaired by liver disease or old age.

Lithium

1. In the 1800s lithium was attempted as a treatment for gout. Lithium urate is highly soluble, so the assumption was that it would help dissolve crystalized uric acid in the joint spaces. It did not. In the mid 1800s, some dude named Sir Alfred Garrod proposed that mood disorders were caused by “gout retroceding to the head,” and some of his contemporaries speculated that mania was caused by high uric acid levels. They were wisely ignored until the 1940s when an Aussie psychiatrist by the name of John Cade decided that maybe those guys in the 1800s were on to something and started injecting guinea pigs with lithium urate. Instead of causing manic symptoms, the guinea pigs appeared sedated. Cade (wrongly) hypothesized that the lithium counteracted urate-induced ‘activation’ and then tested this hypothesis in 10 manic patients of his. Somehow, it worked.

2. Did You Know? People have very strong opinions about lithium and how efficacious it is in preventing suicide. I find myself persuaded by the statisticians who point out that we literally cannot know the answer to this question based on current data, because our N is too small to detect the magnitude of effect proponents of lithium claim it has, due to how rare suicide events are in studies.

Depakote

1. Despite it being thought of as a prophylactic for mood episodes in bipolar disorder, there is only one placebo controlled trial that examined valproate’s efficacy as in this context… and it was negative. Not a very good mood stabilizer, is it?
   
   Except, as Dr. Ghaemi points out in his book, Clinical Psychopharmacology, that study is better for valproate than it seems. First, it also found that lithium did not have any prophylactic efficacy, which suggests that finding an effect in the studied population might be quite hard indeed. Second, the study did something relatively unusual - it did not use an enriched study design³. When a secondary analysis was done in individuals who had previously responded to valproate, it performed better than both lithium and placebo as a prophylactic.

2. In the 1940s, the Germans were getting a bit desperate with the whole “losing the war” thing and were looking for food substitutes. Where did they turn? To their absolutely enormous coal reserves, obviously. Which they turned into a mix of triglycerides that tasted vaguely like butter, obviously. A tree died 300 million years ago and we turned it into margarine. Amazing.
   
   In analyzing ersatz butter, as they called it, the Germans isolated valproic acid. It was not nutritionally useful, but turned out to be an excellent solvent. In the 1960s it was used as a vehicle for injection of an otherwise quite insoluble compound called khelline. The khelline-valproate mixture was noted to produce significant sedation in the human subjects it was administered to, leading to the discovery of valproate’s pharmacologic properties.

1. Monoamine oxidase inhibitors

2. Commonly remembered with the acronym LOT, though I don’t know anyone that actually uses oxazepam or temazepam

3. Enriched studies use patients who are previously known to respond to the drug in question. It gives better clarity when asking the question “when this drug works for someone, how well does it tend to work?” but obviously has its downsides when trying to figure out how likely it is to help a population of the treatment naïve.