When evidence based medicine results in a government sanctioned lie.

Starring Megan Fox.

The headline says almost everything:  Acadia shares plunge more than 50% on study data for schizophrenia drug.  Turns out the drug didn't work at either of the two doses tested.

They should have called me first: their study was flawed.

Belated, anyway.

You may have seen the advertisements on TV:  "if you've taken Zyprexa and have diabetes, call us, the legal team at..."

 

Before you take your patients off of Zyprexa in a misguided attempt at warding off litigation, consider the following:

1.  You can't be sued if there's no damage. In other words, you can't be sued because of the risk of diabetes, you can only be sued for diabetes.  No damage, no lawsuit.  If a patient gets diabetes, you catch it and act appropriately, you can't be sued.  If you take reasonable care (note the weight every, say, 6 months; follow blood sugars every, say, year-- more frequently if there is weight gain), not only have you shown above standard-of-care practice, but you're going to catch the problem and fix it-- so no lawsuit.

2.  The lawyers in these ads are trying for a class action-- against the company.  Class actions are not about the severity of drug side effects.  The class action requires  that the company (Lilly) knew about the risks, but purposefully hid these risks from doctors and the public.  (This is why there are no class actions against chemotherapy makers.)   But if the company hid the info, then the doctor couldn't be responsible for the diabetes, because the risk was hidden.  So the class action actually protects the doctor, in a sense.

 

3.  Here's a puzzler: consider the following by-product of these advertisements.  By soliciting patients who have taken Zyprexa and gotten diabetes, they are, essentially, telling people about the risk.  So a patient who develops diabetes sometime in the future may not be able to claim he didn't know about the risk, as the risks have now entered the public discourse.

 

Stop worrying about lawyers.  Worry about loose practice.

The authors of this article have an interesting hypothesis, upon which I speculate wildly.  But it is fascinating: 

 

GLUT5, is found primarily in the small intestine (though also in muscle and kidneys.)  What's interesting about it is that it transports fructose, which in turn directly stimulates additional GLUT5 mRNA  expression.  You eat fructose, this increases the expression of GLUT5 in the intestinal villi, which increases the transport of fructose.  So the more fructose you eat, the more readily you can absorb it.

 

Now fructose doesn't stimulate insulin secretion.  Since insulin regulates leptin, fructose actually reduces leptin.  Fructose increases ghrelin. So you get hungry.  Fructose goes to the liver and is metabilized to acyl glycerols, and consequently result in  increased triglycerides.

 

So you have a situation in which Remeron and Zyprexa (and high dose Haldol) cause an increase in GLUT5 expression; if they are also eating fructose (read: high fructose corn syrup) this is causing an additional expression in GLUT5, and hunger, and increased triglycerides...  if one wants to conduct a useful experiment, find out if the people who gain the most weight on Zyprexa are those who consume the most  high fructose corn syrup (and not just those who eat the most.)  In other words, can you gain weight on Zyprexa if you are eating Atkins?

 

(NB: there are many who want to believe that Zyprexa causes weight gain by increasing leptin; and so fructose and GLUT5 lowering leptin seems confusing.  Zyprexa, as shown above,  actually decreases leptin, acutely.  (And clozaril has either no effect, or minimal lowering.)  Letpin only increases with increased fat-- i.e. as a consequence of fat, not as the cause of fat.  Those who have found increases in serum leptin do so only after chronic administration, and resultant weight gain (for example, in a study of 13 schizophrenics on Zyprexa who showed a small increase in leptin after 4 weeks-- and after a 2 kg weight gain; or 6 week animal study finding increased fat and leptin.   The question, as noted by the authors, is whether the acute hypoleptinemia and hypoglycemia is what triggers hunger and an ultimate increase in fat, leptin, glucose and insulin. )

 

In order for this post-- and any discussion on antipsychotic induced weight gain-- to make sense, you have to understand one thing: each antipsychotic seems to cause weight gain by a different mechanism, not varying degrees of the same mechanism.  Because let me tell you right off the bat: researchers here are far from agreed. 

A review of some articles: 

 

In rat pancreatic beta cells,  neither clozapine nor haloperidol had any effect on basal insulin release.  In the presence of high glucose,  haloperidol had no effect on the normal insulin surge, but clozapine inhibited this effect by 40%.  How it did this is not clear, as clozapine, in the presence of glucose, completely suppressed electrical activity by hyperpolarizing the membrane potential (i.e. increased K+ conductance.)  Haloperidol depolarized (inhibited K+ conductance).  Thus, by completely suppressing electrical activity, it should have completely suppressed insulin release-- but it only inhibited 40%.  Similarly, haloperidol should have increased insulin release (via depolarization) but it didn't have any effect.  We don't know what would have happened if the study had been continued for a year; but note here that the effect on insulin is dependent on the presence or absence of glucose, not the other way around.

Most studies focus on the changes in serum parameters (triglyceride, cholesterol, insulin, etc) and not the mechanism for these changes.  

For example, in 112 schizophrenics on meds for 8 weeks, Zyprexa, clozapine, Risperdal, sulpiride all increased insulin and C reactive peptide, as well as insulin resistance; but only clozapine and Zyprexa increased triglycerides and cholesterol, and had a greater impact on insulin, insulin resistance, and  C-peptide. What you can't tell is when this happened and what came first: did the insulin go up as a direct effect of the med, and consequently so did cholesterol, or did insulin resistance happen first, etc?

 

In the first study looking at the drugs' effects on GLUT1-5 mRNA, it was found that  Remeron (mirtazapine)  increases GLUT4 (muscle/fat) and 5(intestine) mRNA, and Haldol and Zyprexa increase GLUT5. No effect on GLUT1-3. (Contrast with Clozaril and Risperdal, below.)

The authors propose something interesting about Remeron: "Therefore, the increasing effects of mirtazapine on GLUT4 mRNA levels in our study might lead to a decrease in blood glucose levels and to an increase in cellular fat deposition, leading to intermittent or continuous lowering of blood glucose levels with a subsequent increased uptake of carbohydrates and other types of nutrients."  In other words, better glucose uptake into cells means more fat inside cells, and less glucose outside cells (hypoglycemia)-- which is a stimulus to eat more.

This is important, so I'll repeat it: the hyperglycemia seen with Zyprexa and Remeron is here proposed to be due to the acute lowering of blood glucose (because of increased transport), and thus an increase in eating and fat deposition, and consequently insulin resistance and hyperglycemia; not a direct affect on glucose metabolism. 

(Consistent  with Zyprexa's effect on GLUT5 (and not on carbohydrate metabolism, per se), metformin did not prevent weight gain in 40 people on 10mg Zyprexa (all gained 5-6kg in 14 weeks.)


In (male C57) mice, over a 6 month period, clozapine, chlorpromazine and quetiapine induced hyperglycemia via effects on glucose transport.  Haldol and amisulpiride have little effect on GLUT, and were found not to induce hyperglycemia.    Risperdal had a medium effect on hyperglycemia, but at the lower doses. 


Using rat pheochromocytoma cells, clozapine and Risperdal both inhibited glucose transport (i.e. GLUT3).

Desmethylclozapine (a metabolite) was an even more potent inhibitor, while clozapine-N-oxide, the other metabolite, had no effect on glucose transport.  Clozapine and fluphenazine also inhibited glucose transport in (rat) muscle cells. The drugs block glucose transport in a non-competitive (i.e. allosteric) manner (and tricyclics appear to work in the same way.)  What is interesting about this is that different people metabolize clozapine differently, and perhaps those who create more desmethylclozapine get more hyperglycemia than those who make less (and/or more clozapine-N-oxide.

A follow-up study tried to correlate the toxicity of these drugs to cells to their inhibition of glucose transport. 

They found that clozapine,  desmethylclozapine, Seroquel and fluphenazine were toxic to cells; Risperdal was minimally toxic; and Zyprexa actually promoted cell growth.

Seroquel, Zyprexa and clozapine all inhibited glucose transport about the same amount, and in a dose dependent manner.  (Remember: Haldol and sulpiride don't.) 

However, if the cells were exposed to drug for a longer time, fluphenazine greatly inhibited glucose uptake, clozapine had no effect, and  Zyprexa increased glucose uptake.  In other words, the toxic typicals only need a sort exposure to kill a cell, while less toxic atypicals need prolonged exposure.  Also, fluphenazine increased GLUT3, and the atypicals had little or no effect (as found above.)

Zyprexa was found not to affect either the basal or the insulin stimulated glucose transport via GLUT1 or 4. (Fun fact: bovine serum albumin (or impurities therein), used to replicate the fact that olanzapine is highly (93%) protein bound, actually increased basal glucose transport, making suspicious all studies previosuly done with BSA.)  This contradicts the findins of the Dwyer articles, above, where antipsychotics had inhibitory effects on glucose transport.  A possible explanation could be dosing: this study used doses comparable to 20mg, while others used 20x that amount.

Another study, in humans, found that neither Zyprexa nor Risperdal affected acute (3 week) insulin sensitivity.  Again, what happens after you get heavy is up for debate.

 

So what we have here is confusion, but:

1. acute, high dose in vitro studies indicate that typicals>atypicals inhibit glucose transport, but Haldol does not.

1b. Typicals are toxic to cells, atypicals less so, and Zyprexa promotes cell proliferation. 

2. Normal dose and human studies show no effect on insulin dependent glucose transport (i.e. GLUT4) but there are effects on small intestine absorption (GLUT5) with Zyprexa and Remeron.

3. Clozapine inhibits insulin release in the presence of glucose, but Haldol doesn't. 

4.  Acute effects may be different than chronic.  i.e. even though antipsychotics may not directly affect insluin resistance or glucose transport, if they make you hungry or increase fat over time, this could result in later insulin resistance, hyperglycemia, etc. 

 

 

 

Glucose is absorbed through the small intestine into the blood.

All glucose is taken into cells via hexose transporters: this is facilitated diffusion (no ATP).  Facilitated diffusion is passive diffusion through a channel made by a transmembrane protein; the proteins are able to open and close this channel. There are many ways channels can be opened/closed: ligand gated (i.e. neurotransmitter receptors), voltage gated (neurons), or, in the case of hydrophilic molecules such as glucose, mechanically gated: the channel is shaped like a closed "V".  Glucose goes to the bottom of the V, causes a conformational change and the "V" opens, but closes at the top (makes an upside down "V".)  Glucose can pass, and the V recloses.  All diffusion is down a concentration gradient.

The hexose transporters are called, randomly, "Glucose transporters 1-5" (GLUT1-5).

GLUT4 is the main transporter in muscle, fat, and the heart.  GLUT4 is insulin-sensitive (though it can also be activated by muscle contraction-- go figure.) In the absence of insulin, GLUT4s are stored in cytomplasmic vesicles floating around in the cytosol.  If insulin binds to the insulin receptor (an ATP dependent tyrosine kinase receptor-- NOT the GLUT4), a signal cascade is activated that causes the cytpolasmic vesicle to go to and bind to the plasma membrane and lodge the GLUT4 there. The GLUT4 then allows glucose to diffuse through.  When insulin disappears, the insulin receptor reconforms, the signal cascade stops, and the GLUT4 pinches off (by clathrin and other contracting proteins in the cell membrane) into a vesicle again (pinocytosis).

Thus, if there is no insulin: even if there is much glucose, there is no signal for the vesicle to go to the plasma membrane and lodge the GLUT4, so there will be no transport of glucose into the cell; so glucose stays high in the blood.  Thus we have Type 1 diabetes.


Insulin also stimulates the creation of glycogen in the liver and muscle.  [Insulin activates hexokinase (1st enzyme in glycolysis) as well as phosphofructokinase and glycogen synthase) and inhibits glucose-6-phosphatase ((opposite direction of hexokinase, same reaction) gluconeogenesis).]

Insulin promotes fatty acid synthesis.
Once glycogen synthesis has maxed out (i.e. about 30g, about 20% of the carbohydrate part of a studied meal, max in 4-6hrs,) then fatty acid synthesis IN THE LIVER takes over.  Glucose is converted to free fatty acids (FFAs) and dumped back into the blood as lipoproteins-- which are then broken up into FFAs.

FFAs go into the adipose cells of the body.  Glucose also goes into adipose cells-- via GLUT-- and are converted into glycerols.  Glycerol+FFAs= triglycerides.

Thus, insulin's role is to store fat and/or oxidize glucose.  Too little insulin will also trigger protein catabolism.

GLUT1 and GLUT3 account for 95% of the glucose transport to the brain. GLUT1 is for the blood brain barrier (the tight junctuions of the BBB are what require these channels), and GLUT3 is in the neurons. (pic here)  GLUT1 is also found in muscle.

These are not insulin dependent (like GLUT4 is) so the brain can continue to get its energy.  Not only does the distribution of GLUT 1 and 3 mirror capillary density and areas of relative glucose utilization, the GLUT1/3 densities can change depending on chronically increased (or decreased) need for glucose.  Interestingly, nicotine, which increases brain glucose utilization, increases GLUT1/3 but not capillary density.


GLUT2 and 7 are in the liver.  GLUT2 can also carry D-fructose.

GLUT5 is in the intestine, and some glial cells of the brain.

Type II diabetes is insulin resistance, not lack of insulin.  There is not, at least initially, a problem with the pancreas's secretion of insulin in response to high glucose.  The problem is at the level of the insulin receptor and/or GLUT, which become insensitive to the effects of insulin-- because there has been so much of it for so long. (For more info, see: News Physiol Sci. 2001 Apr;16:71-6.

Next up: how do antipsychotics affect glucose/insulin/transporters? 

 

(For a review: What We Know About Facilitative Glucose Transporters )

According to USA Today, 2.5 million antipsychotic prescriptions a year are written for kids under 18.  The rate for privately insured kids is 6.5 in 1000-- it has to be easily ten times that for Medicaid kids.

The FDA database has 45 deaths; 6 from diabetes, the rest from CV disease, liver failure, suicide, etc. There were 41 pediatric NMS cases.

According to the article, 13% of antipsychotic prescriptions are for bipolar disorder.

So are antipsychotics being overprescribed?  The answer is yes, but not for the reasons cited in the article.

The article, indeed, all articles about pediatric psychiatry, make a special point about how these medicines are not FDA approved for kids.  This is absolutely meaningless.  FDA approval requires two double blind, placebo controlled studies.  These studies are universally taken on by the drug companies.  No drug company would ever assume the massive risk of such a study-- let a lone two-- in kids. How do you recruit the study subjects?  What parent is going to allow it?  Rich parents?  No chance.  So it will have to be Medicaid parents-- and thus will come the Tuskegee-like charges, dripping with the obvious social and racial implications of pharma testing on poor minorities. Pharma is already loathed; they're not going to take any risks for the sake of a medal from the FDA.  So there will not be any new pediatric indications for psych meds.  Not in this climate.  Think this hurts Pharma?  It's your kids that suffer.

But don't be confused by crypto-socialist hysterics who say that Pharma will do anything for a profit, including peddle drugs to kids.  Drug companies do not market these antipsychotics  for kids.  They are paranoid to a fault about doing this; they know everyone is scrutinizing them, especially lawyers.  If you are a child psychiatrist who sees no adults, reps cannot even call on you.  And if they call on you for other things, they cannot mention the use in kids.  In the past five years, it has never-- never-- happened that a rep detailed me about their use in kids.

The only two reasons these drugs are used in kids is because psychiatrists give them, and parents demand them.

First, the parents. They don't come looking for antipsychotics, specifically.  But my experience is that they are unrealistic about what is going on with their kids; in near denial about the family dynamics impacting on the kid's behavior; and virtually devoid of insight into relatively obvious, though procedurally difficult, maneuvers that could improve the situation. If your kid doesn't sleep enough, and consistently-- if your five year old doesn't nap-- you cannot tell me your kid has ADHD.  Period. Parents demand a diagnosis of bipolar disorder for their kids because it means the divorce had nothing to do with it.  They demand another medication when the first one fails to get the kid to do math homework instead of playing Xbox all day. And their kids' marijuana and alcohol abuse can't possibly have anything to do with their own marijuana and alcohol abuse.   Parents: don't flame me.  Your situation is different, I know.  I know. 

Second, psychiatrists prescribe them because of the pressure to do something, in the face of consistent failure.  They don't start with antispychotics-- they end up with them.  They prescribe them out of desperation.  This is why, in every story about a child getting sick from one of these medicines, they are, in fact, on several medicines.  First they start with Ritalin.  If Ritalin doesn't help, or there is a side effect, or they can't sleep-- then a second drug is added.  Maybe this helps, but after a while something else happens-- and another drug is added to this.  That's why psychiatry's current obssession with the detection of underdiagnosed "bipolar disorder" is so important.  This diagnosis justifies, and encourages, polypharmacy.

It is psychiatry's  ridiculously dangerous, and ultimately doomed, paradigm: if you are not doing well on a medication, you must be so sick that you require two medications.  It seems to have occurred to no one in psychiatry that failure on a medication could mean that it was the wrong medication. 

The reason this polypharmacy madness is even possible is psychiatry's obsession with diagnosis, labels-- with semiotics.

What makes a drug an antipsychotic?  Well, it treats psychosis.  Fine-- but does that exclude its efficacy for something else?  If it is later found to be efficacious in, say, depression, then what do you call it?  Is the drug an antipsychotic that's also good for depression, or an antidepressant that's also good for psychosis?

There's no value in the label "antipsychotic" or "antidepressant" except what we give it.  It's a drug that treats psychosis and depression, not an antipsychotic that treats depression (or the other way around).  If you can't see the difference, stop reading now and go back to watching American Idol.  

For example, why are antipsychotics viewed as "off label" for kids?  The word "antipsychotic" is meaningless. Antipsychotics are tested against a scale, like the Brief Psychatric Rating Scale.  But these scales measure a lot of things, like depression, and not just psychosis.

And at what point did we start making a distinction between psychosis and "dementia related psychosis?"  Or bipolar depression and regular depression?  Why do we need separate FDA approvals?  Does someone know something about the physiology of these disorders that I don't? Do we need to start approvals for "diabetes related depression?"

Saying an antipsychotic is worse than an antidepressant for depression is a valueless statement, especially in the absence of data on this question.  You are actually better off asking, "which is better for depression, blocking the serotonin transporter or blocking 5HT2a receptors?"  See?  Put this way the distinction seems less obvious.  And even that question is valueless, as there is nothing (that we know of at this time) that allows us to say what effect either pharmacologic maneuver actually has.  5HT2A blockade does what again?  Really?  Do you have any evidence for that at all?  And no more post hoc ergo propter hoc nonsense.  David Hume laughs at you.

A Simpson's reference is helpful here:

Homer: Not a bear in sight.  The Bear Patrol must be working like a
       charm.
Lisa: That's spacious reasoning, Dad.
Homer: Thank you, dear.
Lisa: By your logic I could claim that this rock keeps tigers away.
Homer: Oh, how does it work?
Lisa: It doesn't work.
Homer: Uh-huh.
Lisa: It's just a stupid rock.
Homer: Uh-huh.
Lisa: But I don't see any tigers around, do you?
        [Homer thinks of this, then pulls out some money]
Homer: Lisa, I want to buy your rock.

I know.  The FDA, the Scientologists, socialists, the parents at the end of their ropes- the easy thing to do is blame Pharma.  I'm in the strange position of having to be a Pharma apologist, to be the only doctor willing to defend Pharma.  There are plenty things I don't like about the way Pharma conducts business, but I can't voice these complaints because I have to use the time countering these inane attacks.   I know what will happen if the Pharma critics get their way.

You think Pharma should have no sales contact with physicians?  Fine.  Now deal with the consequences.

And enough with the notion that medication compliance is a good proxy for overall efficacy.

 

All of these horrible psychiatry studies-- CATIE, Lamictal and Depakote maintenance trials, etc-- keep telling us how long patients stay on medications, because they say this means the drugs are working.  The authors think that if a drug is working, they patient will stay on it.  But you would think this only if you didn't actually treat many patients.  I can make a similar argument that staying on a medication is inversely related to efficacy-- because when a patient feels better, they simply stop taking their meds.  

 

Think about antibiotics.  People don't finish the full 14 day course, precisely because they feel well.  If they felt sick, they would probably take them longer than 14 days.  In fact, people overuse these antibiotics even when its a virus, despite the antibiotic having no efficacy at all.    They will demand an antibiotic even though know that it shouldn't be doing anything.

 

Same with pain meds.  Oh, that's an acute problem?  How about the chronic problems of diabetes and hypertension.  People will skip/miss/forget doses when they feel asymptomatic, and will be more compliant when they have symptoms associated with these illnesses (e.g. headache, dizziness, etc.)

 

Look, I'm not telling you that compliance and efficacy aren't related.  I am saying that if you want to measure efficacy, don't use compliance as a proxy-- go measure actual efficacy.   And don't tell me it's too hard.  You got $67 million for this study.  Find a way.

if CATIE is so important, and was funded by the NIMH (i.e. taxpayer dollars), why isn't it available free on the CATIE or NIMH websites?  Or anywhere else?  Snakes.  All of you are snakes.

1.  You know, if you're going to be rigorous about BID dosing schedules because the FDA requires it, why so liberal with total dosing for Zyprexa?  A mean dose of Zyprexa is 20.8 is way (150%) above FDA guidelines.  For comparison, that would have meant dosing Geodon at 240mg, Seroquel at 1000mg, and Risperdal at 6mg.  BTW: a mean of 20.8mg means that a lot of people were dosed with MORE than 20.8mg (max=30mg).

2.  The miracle here isn't that Zyprexa won, but that Zyprexa 20mg barely won against Geodon 114mg.

3.  Why Trilafon (perphenazine)?  Originally you thought all conventionals were the same; so why not Haldol?  Or Mellaril?  You say it's because it had lower rates of EPS and TD, which is fine, but then why exclude TD patients from that arm?

4.  So you excluded patients with tardive dyskinesia from the perphenazine group (fine) but then had the nerve to say people tolerated it as well as other meds?  Do you think maybe people who have TD may have different tolerances to meds?  Different EPS?  Different max doses?  That they're just different?

5.  You can't generalize from an obviously slanted "typical" arm to all other typicals.  If you chose Trilafon over Haldol because of better tolerability a priori, you can't now say that "typicals" have equal tolerability to atypicals.  Why not pick two typicals of differing potencies (like Mellaril and Haldol) and infer from there?

6.  Do you actually believe-- does anyone believe-- that any of these patients are compliant with BID regimens?  Especially with sedating meds like Seroquel?


The secret to understanding CATIE 2 is to understand that there are two CATIE 2s.

CATIE2-Efficacy: People who dropped out of CATIE 1 because their med didn't work were randomized to Clozail, Zyprexa, Risperdal or Seroquel.  On average, new Clozaril switches stayed on 10 months, everyone else only 3.  44% of Clozaril stayed on for the whole 18 month study; only 18% of the others completed the study.

CATIE2-Tolerability: People who dropped out of CATIE 1 because of side effects (not efficacy) were randomized to Zyprexza, Risperdal, Seroquel, and Geodon (not Clozaril.)  Risperdal patients stayed on for 7 months, Zyprexa for 6, Seroquel for 4 and Geodon for 3.

CATIE2-Efficacy is fair.  If you fail a drug, you're likely to do better on Clozaril than anything else.

CATIE2-Tolerabilty makes no sense at all.  The reason Geodon was used is because it has "very different" side effects.  Hmm. How?  "In particular, ziprasidone [Geodon] was known not to cause weight gain."  But this assumes that the intolerability of the first antipsychotic was its weight gain.

Most importantly is this: if a patient couldn't tolerate their first antipsychotic, how likely is it that it was effective?  In other words, if it wasn't tolerable, it wasn't efficacious-- these patients could have been in CATIE2-Effectiveness study.  So how did they choose?

Easy: they gave the patient the choice: Geodon or Clozaril?  Out of 1052, half left altogether.  99 went into the Clozaril study (CATIE2-Effectiveness) and 444 went into Geodon (CATIE2-Tolerability.)  Of the 444 in the Tolerability trial, 41% were actually labeled first drug non-responders.  38% were labeled as not tolerating their first drug, but of those, who knows how many were also nonresponders?

And 74% dropped out again.

If you take the 444 in the Tolerability study and divide them into two groups:

• those who left CATIE1 because of lack of efficacy: then switching to Zyprexa or Risperdal kept them on their meds longer.  (Which makes no sense again: this is the same thing as the CATIE2-Effectiveness, where (except for Clozaril) there was no difference between Seroquel, Zyprexa and Risperdal.)
• those who left CATIE 1 because of lack of tolerability, then it made no difference what you switched to.

Sigh.

 

And what's with the blinding?  In every other study with a clozapine arm, you equalize the weekly blood draws by making everyone have to submit to them.  But in this case, they unblinded clozapine so as not to have to subject all these people to blood sticks.  Except they were subjecting them already-- they were checking blood levels.

And where was perphenazine?  "[CATIE1] did not anticipate this unexpected result [that perphenazine would be as efficacious] that challenged the widely accepted (but never proven) belief that the newer atypical antipsychotic medications are better than all older antipsychotic medications" and so was not considered for CATIE2.  Apart from the fact that it is simply untrue that anyone thought the atypicals were more efficacious than the typicals, it is furthermore untrue that that the authors did not "anticipate this unexpected result."  In 2003, after basically doing Medline meta-analysis, they found that "not all of them were substantially different from conventionals such as perphenazine."

What's funny about these guys is how they conveniently lump all typicals together but arguing for differential effects of  individual atypicals; then argue typicals are different from each other to justify picking Trilafon; and then say atypicals are different from each other ("not all of them were different") but typicals are all pretty much the same ("conventionals such as perphenazine.")

 

Bottom line:

 

The stated purpose of CATIE2 was to help clinicians decide which drug to switch to if patients a) failed their first drug; b) couldn't tolerate their first drug.

The divorce rate in America is 40-50%. Say you get divorced, and a friend says, I have two women for you, Jane and Mary.   If the problem with your first wife was that she didn't turn you on, you should marry Jane.  If the problem with your ex was that she was annoying, you should marry Mary.

 

What's going to happen here is that your second marriage, to either girl, is doomed.  Certainly more than the national average of 50%.    How long is it going to take before your second wife doesn't turn you on either?  How long before you find stuff intolerable about her?    The answer is, more likely than your first marriage-- say, 75%-- because the problem isn't your wives, it's you.  You've framed the question in an idiotic and arbitrary manner.  You don't get married to get turned on OR to be with someone who isn't annoying.  You want the marriage to have both simultaneously, and much more.  These things are not separable.  This is CATIE2.  A meaningless dichotomy-- efficacy and tolerability are not separate, let alone opposites-- used to create a false paradigm of medication selection.

As you may know, when prescribing Clozaril (clozapine), a complete blood count with differential (CBC w/ diff) has to be checked every two weeks, because of the risk of agranulocytosis.

The FDA has relaxed these requirements: now, you have to check weekly for the first six months; then every two weeks for six months, then only monthly after that.  You have to show WBC >3500/ml, and ANC>2000/ml.  (That's white blood count and absolute neutrophil count.)

Don't think for a millisecond this was done because the FDA did a rigorous re-evaluation of safety data.   This is the FDA that black boxed antidepressants for suicide and antipsychotics-- oh, sorry, only atypical antipsychotics, even though typicals are as bad, if not worse-- for death in patients with dementia related psychosis.

What's stupid about this is that agranulocytosis is the least of anyone's problems.  In the Clozaril National Database (1990-1994) (1), there were 99,502 patients.  382 (0.4%) got agranulocytosis, and 12 died (that's 0.001%).  The number of clozaril related deaths (all kinds) was more than 400.

In an Italian study, the rates of neutropenia are about 0.9%, and agranulocytosis 0.7% (2)

They are, however, dying in not insignificant numbers by other things.

Consider a Maryland finding: of the 2046 clozaril patients from 1990-2000, three died of new onset diabetic ketpacidosis. (0.15%) None had had diabetes. (3)  Or the Israeli study (4) that found that 4/561 clozaril patients had sudden death--  10 years younger, healthier, and 4 times the rate of  non-clozaril treated sudden deaths.  NB: no one died from agranulocytosis.

How about myocarditis: 8000 patients over 6 years: 15 myocarditis, 8 cardiomyopathy; 6 died.  That's 0.3%  5 of the 6 deaths occurred in first month  (that's right: month). (2)  Given the rapidity of death, the authors speculate it's an acute hypersensitivity reaction (i.e. IgE/Type I).

A review of Pubmed/MEDLINE from 1970-2004  found rates of fatal myocarditis/cardiomyopathy to be between 0.015% and 0.188%.

An oft cited article by Walker examined 67072 clozapine patiens from 1991-1993, and found that of the 396 deaths, the most common cause was pulmonary embolism.  (FYI: Zornberg found that exposure to low potency antipsychotics massively increases the PE risk to (OR 24 for low potency, 3 for high potency; not dose related, usually occurred in first three months.) 

In contrast to Hagg's finding of 12 cases of PE/DVT, and a frequency of about 0.03%, another study of 13000 inpatients over 6 years found 5 PEs, i.e. a rate of 0.038%; but this was no different than typical neuroleptics of non-treated.

Look, I'm not saying to ignore agranulocytosis.  I'm saying that when your patient's heart explodes, you can't say, "but the FDA only said CBCs!"  You need to be checking EKGs.  And when the lawyer asks you how most people die when on clozapine, it'll look really bad when you give the wrong answer.