Antipsychotic polypharmacy in inpatient setting

This forum made possible through the generous support of SDN members, donors, and sponsors. Thank you.

flash

Senior Member
15+ Year Member
Joined
Oct 26, 2004
Messages
390
Reaction score
0
Hey all,

So we have 2 attendings with somewhat different approaches. One left for vacation, and the older (40+ years of experience) is covering. Today he adds yet another antipsychotic (Zyprexa) after just doubling Abilify yesterday to 10mg BID for a patient with Schizoaffective D/O, bipolar type, manic. Pt has at least partially responded in the 10-12 days or so he's been here (no more aggression/threatening, reduced hyperactivity, voices better and no longer driving behavior it seems) but he continues to remain delusional, paranoid. Regimen now includes Risperdal, Lithium, Abilify, and now Zyprexa.

Do other see this sort of approach at times - throw everything at it? I know some would do conventional + atypical combo, +/- traditional mood stabilizer perhaps, but 3 atypicals? Aren't receptors maximally occupied? If we do see a clinical change, isn't the most likely reason just more time has elapsed?

I'm just starting my review of literature here, for those interested:
http://www.ncbi.nlm.nih.gov/pubmed/15056517

Members don't see this ad.
 
Okay, first off disclaimer: I can only speak in generalities. There may be things specific to this particular patient that cannot be addressed by me because I haven't talked to him/her in person.

I see no need to give Abilify at a BID dosage at 10mg BID. Again this is with the disclaimer that there may be something about this patient that makes them an exception to the rule.

1) Abilify pills are, for all intents and purposes, priced per pill. In effect, this doctor has doubled the price of the patient's med regimen with no reasonable medical evidence as far as I know to increase it's efficacy by giving them 10 mg QBID instead of just 20 mg QAM.

2) The manufacturer recommends once daily dosing, and the half life of the med and it's metabolites is relatively long (75 hrs for the med, 146 hours for it's metabolites) leaving no reason for me to believe that BID dosing will be better than once-daily dosing.

3) The med, per the manufacturer, because of it's partial dopamine agonism actually can cause wakefulness. Stepping aside from what the manufacturer says because sometimes what happens in real-life doesn't match the textbooks, I have found this to be the case with most patients. Some doctors I know tell their patients to take it at night then the patients don't know why they can't fall asleep. When I get the patients and tell them it's supposed to be given in the morning, all of a sudden they get upset and wonder why that doc told them to do that. I can only answer I don't know.

The manufacturer did mention that in cases where the med is given above it's guidelines (>30 mg QAM), it does cause sedation in the majority, but that's not what's going on here.

Giving it at BID dosing could lead the to the patient taking the med at night.

Again this is all with the disclaimer I mentioned. Several patients don't follow the norm, but they should not be put on a medication regimen first-line as if they don't follow the norm.

the older (40+ years of experience) is covering.

Older doesn't mean better. Trust me on this. I've see a lot of old doctors that don't know much. Several doctors turn off the light-bulb after residency. Some even turned it off during residency but were able to get away with it. The longer it's been turned off, the worse the doctor.

Now if the light-bulb's been on longer that's a different story...

In general treatment recommendations in most texts recommend to increase a dosage to the manufacturer's maximum before going on to another medication. An exception is if the patient had some benefit to one but cannot tolerate it at a dose that's lower than the manufacturer's maximum, but is still in need of more treatment.

The way most psychotropics work, you got to reach at least a particular dosage before you give up or add augmentation with another med. E.g. In Seroquel D2 blockage doesn't even start until about 150mg+ per day, you will likely not even reach any benefit until 400 mg +, and in many patients you will have to go to 800 mg (if not even more!) If you get no antipsychotic response at 100 mg Q daily it's idiotic in terms of psychopharmacology to believe that you now got to add another med. You got to increase the dosage before you give up on it as a monotherapy. With each med that dosage differs because most of them bind to other receptors before they bind to the D2 receptors, and the molecular weights of each med influence the actual mg of dosing needed.

This all leads back to what I said before, don't try another med until the maximum's been reached in one or the patient got a benefit with the med but cannot tolerate it at the maximum dosage.

There was a recent AJP article showing the using two meds could be better than one, but while that may be true, it's not true in every case (though in a significant number), and monotherapy should still be the preferred goal because it's cheaper and if it works, less side effects for the patient, and easier for compliance.

http://ajp.psychiatryonline.org/article.aspx?Volume=168&page=667&journalID=13

Do other see this sort of approach at times - throw everything at it?

Unfortunately most of the doctors that do this, more often than not, IMHO don't really know what they're doing.

A problem with making blanket assumptions is there are so many exceptions to the rule. I've had my fair share of polypharm patients, but I did follow the guidelines of only going to 2nd, 3rd, or even 4th meds only if one gave a benefit and it could not be raised higher. Most of these patients in the polypharm category were patients I had in forensic settings where the patients in general were much more treatment resistant, and far more dangerous if not on meds.

I can certainly say that some doctors I've known well enough to know they don't know squat. E.g. I know one doctor that starts every patient on about 4-8 meds the first visit. I'm not joking: an antidepressant, antipsychotic, mood stabilizer, stimulant, benzo, cogentin, and a sleep med plus a few more. Almost every single patient I've known who saw this guy thought he was terrible, and most of them were far worse with treatment with this guy than even without it.

I don't want to judge the doctor you mentionedr, but any doctor should have at least some type of sound model based on evidenced-based research or at least very good anectdotal experience as why they're doing what they're doing. Also if a doctor engages in polypharmacy, they should at least document the reasoning and benefit that occurred with each med. I've seen too many doctors throw meds around as if they're throwing splotches of paint on a canvass. It's based on an "inner feeling" and not on psychopharmacology, evidenced-based medicine, or even common sense.

E.g. I had a guy I discharged on Risperdal 4 mg QBID and Zyprexa 15 mg QHS. In short, he got no benefit from Risperdal at higher than 4 mg QBID (he was titrated up to 8 mg QBID), but was still having hallucinations at 4 mg QBID (though his paranoia and hostility were greatly reduced), and when Zyprexa was added he was very much better. I told him that I could try to shoot for monotherapy and see if the Zyprexa by itself could do the trick, but the guy was so happy that he finally was free from psychotic symptoms that he didn't want to mess with his current regimen. He opted to stay on both. Given that it took us 2 months to get him better, he understood the risks and benefits of the polypharmacy, he had the capacity to make a decision, I let him take this regimen, and I documented all of it, including summarizing it on his discharge summary. Personally I think Zyprexa may have been the best med for him and monotherapy was worth a shot, but it was his life and he has capacity, so he should've had the final decision.
 
Last edited:
Agreed nobody does BID dosing with Abilify. Also I've always been a fan of typicals, who the heck can afford 3 atypicals. On rare occasion ( I.E. chronically state hospitalized patients) I've used 2 antipsychotics and a mood stabilizer. I've never seen a need for more than that. Sounds like he is just throwing meds at the pt and seeing what sticks but in the end if you stabilize him you arent going to know whats helping and what isin't doing a thing.

Just a side note if the patient can follow and be compliant with that complicated of a regimen (I couldn't) then Clozaril should be no problem instead. If he can't follow that regimen then why prescribe it. Also if the mania is really a problem see if u can slow it down with some klonopin instead of 2 extra antipsychotics. I feel it's as if 1/4 of the doctors out there are benzo crazy and the other 3/4 are afraid to use them for anything but withdrawal.
 
Members don't see this ad :)
I only do polypharm if the max dose of one med was tried and the patient got some benefit to it but is still psychotic and in need of help or wants more medication.

And when I add another med, I usually try a med that's known to augment the original (e.g. Depakote to antipsychotics), or another antipsychotic of a very different structure if the first med showed no benefit whatsoever, and if that med showed no benefit whatsoever, I take them off of it completely (after a taper down).

Several antipsychotics have similar structures. E.g. Seroquel, Clozapine, Loxapine, and Zyprexa are all benzo-ring based meds.

If the first med showed at least some benefit, I'll either add another med, or consider a different med completely. It all depends. E.g. I had a guy that got almost completely better on Zyprexa, and that dosage reached the maximum. He still had some residual paranoia. Since Loxapine is almost the same med as Zyprexa, I added just a few mgs of that med and the guy then completely stabilized. My reasoning was if Zyprexa worked so well, his receptors were likely compatible with the structure of that medication.

Attendings are people and individuals have their own strengths and weaknesses. maybe this old dude can teach you something about the psychodynamic perspective or the history of psychiatry (or just what NOT to do yourself)

First year of residency I certainly did not know more than any of my attendings in psychiatric knowledge, though my medical knowledge was certainly far better than many of them (e.g. one guy ordered med consults for BPs on the order of 135/78 saying the guy had HTN). By 2nd year I suspected I knew more than some of them but couldn't fathom this could be the case and chalked my feeling up to narcissim. By fourth year I was convinced I was actually quite better than some of them after sitting in discussions with the program director and department chair, where they mentioned the performances of several of the attendings that were actually quite worse than some of the residents and that some of these attendings were in the same program as residents and they were the ones that blew chunks, but the program took them as attendings because of the shortage of psychiatrists.

If you ever notice yourself in the same boat be very diplomatic. Many doctors have fragile and narcissistic egos. Don't ever challenge an attending unless you know you will win that fight (and in most cases I can't see how that will happen), and If the attending is a good one, asking questions won't be interpreted as a challenge. A good doctor will admit he doesn't know the answer, admit to his limitations, and ask you to find the answer with him. I've seen some of the world's top doctors just say they didn't know something to a medstudent that the student knew. Medicine, after all, is something no one can completely master, and it's an everquest in terms of gathering more knowledge and understanding.

Other doctors on the other hand get-red faced and feel humiliated if they are asked a question where they do not know the answer.
 
Last edited:
Also I've always been a fan of typicals,

CATIE caused a paradigm-shift, leading several doctors to think, for a few years, that typicals really weren't worse vs. atypicals in terms of efficacy though they do put patients at higher risk for D2 blockade-related side effects such as EPS and TD. I even cited this paradigm-shift in a legal investigation of a psychiatrist who had a patient that tried to defame him because he switched her Abilify (that was not fully stabilizing her and she was frequently noncompliant and trying to kill people while off of meds) to Haldol Decanoate that actually stabilized her. She accused that doctor of mistreatment because he switched her from a newer expensive med to an older one.

I testified to the state medical board that with the CATIE trial, the current paradigm are that the older typicals actually aren't worse, and given her history, it was justified in terms of safety because she clearly lacked capacity (she denied she was mentally ill, was quite paranoid and delusional) and was very dangerous with the medication regimen she wanted to be on.

That said, in the last few years, there has been research suggesting that only atypicals, not typicals, slow the neurodegeneration related to psychosis.
www.medscape.org/viewarticle/569521

So for patients that had problems affording atypicals, I'd put them on atypical and try a low-dose generica atypical (E.g. Risperdal 4 cut in 1/2 daily so they wouldn't have to shelve out as much money) in the hope that I was at least doing my part to slow the neuronal degeneration.

I've long avoided atypical antipsychotic injections because they are far more expensive than good old Haldol IM.

While sitting pretty in my paradigm, just a few days ago, Nasrallah sent a dept-wide email in my neck of the woods recommending we no longer give out IM typical meds because he's recently discovered data showing these meds actually cause neuronal damage. The data as far as I know is not yet published and he wrote he will present this data to the faculty in the near future.

It's funny how little we really do know, and getting just a little more of the picture can completely change our outlook. I'm now having second thoughts of ever giving out a typical unless an atypical is given first and the patient is still raging. I'm also wondering WTF will I do for patients on a typical that can't afford a typical at all. Tell them they're brains are going to degenerate but too bad because they can't afford the right meds?
 
Last edited:
I'd add that you get somewhere around 90% dopamine saturation with just 5mg abilify, and I rarely see benefit above that.

My experience is less that people "turn off the light" as stop learning from reliable sources. Risperdal, for example, should technically be dosed BID. If you don't keep up on the literature you wouldn't know what we learn along the way. Older psychiatrists especially have less experience with newer meds. Similar to how I'm not especially adept at dosing MAOI's.

I've found high doses of abilify do nothing but cause akathisia.

In this case 3 atypicals is usually a poor call. The only exception involves hospital formulary issues of only certain antipsychotics being allowed for a prn, but titrating up a standing dose of something else.

"Rational polypharmacy" speaks to selectively using 2 antipsychotics, such as abilify/risperdal or abilify/haldol to titrate the level of dopamine blockade. The only other explanation might be getting caught in the middle of a cross-titration.

All of this being said, manic patients can chew up a lot of medications, and should preferentially be maxed out on individual medications, use behavioral interventions (sleep meds/benzo's temporarily to get them sleeping), check levels on lithium/VPA to make sure they're therapeutic, and sometimes just wait. I also have become a fan of thorazine, as I've had a few manic pt's who have had response to this after failing depakote PLUS 40mg of Zyprexa.

Very very few psychiatrist also know about the concept of the inverted U curve in dopamine blockade, and that at a certain point patients get worse, not better.

But the bottom line is to learn from your attendings (including as examples of what not to do). Then you can make your own informed decisions in the future.

I'd echo Scorcher that this is likely an appropriate case for using benzo's.
 
Last edited:
To summarize it for you:

No you are not wrong in thinking that multiple antipsychotics for unclear cause is very bad form. It's likely a form of unmanaged countertransference. Nevertheless, this is your attending of record, so you must be somewhat careful in dealing with this. That said, remember patient care is your top priority, and while deferring to authority is usually judicious, at times you do need to have some independence in inquiry and trust your own judgement.

Here are some practical ways you can think about in terms of dealing with this situation.

(1) Gently discuss with the attending in terms of his thought process, how does he justify these medications, if there's any evidence.

(2) Do a focused literature search with the goal of better care. Is there evidence for antipsychotic polypharmacy? What are they? What about in acute mania?

(3) Suggesting consult an expert or do a case conference in a patient such as this who's refractory to multiple medications. Is cloazpine a better option. Is ECT?

(4) If you suspect genuinely that optimal patient care is not delivered, consider going to your program director. Generally speaking if your PD is supportive that would be the person to negotiate with in terms of how to deal with this and get into the right framework (i.e. director of clinical services)

Remember, as a resident always frame everything in terms of education, learning, deference, seeking expert opinion, collaboration, etc.. Do not label it as incompetence, adversarial processing, etc. even if you genuinely think it's a case of that. There is a time and a place to make an argument for that if the appropriate channels are notified.


I always think that just as the key to being a great ID doc is knowing when to STOP abx, the key to being a great psychiatrist is knowing when to take people OFF meds. In many public psychiatry systems around the country new initiatives are popping up to systematically reduce antipsychotic polypharmacy. For instance the NYSOMH has a new initiative called SHAPEMEDs which attempts to use ideas made famous in Atul Gwande's article about checklists. If you are interested, PM me so I can get you in touch with the leading experts in this field at my institution.
 
Last edited:
http://schizophreniabulletin.oxfordjournals.org/content/35/2/443.full.pdf

There's a link for the meta analysis comparing mono therapy to antipsychotic polypharamcy. Its the full article which is cited in the summary article posted by Whopper. It's a pretty lengthy and stat heavy read, but one thing that stood out was, "When grouping combinations by antipsychotic class, only SGA + FGA combinations, either compared with SGA or FGA monotherapy (N = 8, n = 428), were statis- tically superior to antipsychotic monotherapy". The issue I have is most studies utilized Clozaril, which I think the consensus is that it has already been established that polypharamcy including Clozaril is superior to monotherapy. This is not very practical because, as at least in my experience, the vast majority of people I have seen on antipsychotic polypharamcy are not on Clozaril, which is used more of a last resort. I think more relevant studies would be to examine antipsychotic polypharamcy with drugs other than Clozaril, since prescribers tend to use it as a last resort, in my experience. And the authors touched on that as well in the discussion.
 
On our superchronicresistant floor, probably about 1/4 - 1/3 of the patients are on a combination of clozaril, risperdal consta, depakote, zoloft, and klonopin. Our general use of depot shots was for compliance issues, i.e. a patient is doing okay on clozaril, but keeps going off of it because of various issues, and with the lower dosed consta still on board, they would not drop off as far, and it might prevent a hospitalization (which then pays for the consta/sustenna dosing). That's pretty accepted logic in these parts, though using haldol or prolixin is becoming more common for lots of reasons instead of consta or sustenna. I don't know if I think this is a good idea, but I'm a child psychiatrist now, what do I care? :D
 
Somewhat off on a tangent, if you only stick in one place, it leads to intellectual incest as one forum member put it.

For example, where I did my training, most of the people graduated from that program that were in it, and their psychopharmacology knowledge was based on that which was taught to them by the same group.

Getting to my point (and hoping to keep it much more short-winded than my prior posts on this thread where I was on mega-over-write mode), in this area, the better psychiatrists have instructed me that a patient on Clozaril above 400-500 mg a day should be on Depakote. Reason being is that aside that it augments antipsychotic treatment and a patient on Clozaril could benefit from that, the risk of seizures is dramatically higher in that dosage range. I never specifically saw the studies backing this up but from what I've heard this was specifically endorsed by Paul Keck. Further, several hospitals in this area have this in their Clozaril protocol and to not put someone on Depakote at Clozaril at 400-500 mg+/day requires the physician document why.

(Something I was never taught in residency).

Another thing I learned in these parts is the phenomenon of benign ethnic neutropenia, where men of African ethnicity appear to have a much lower absolute neutrophil count
http://www.ncbi.nlm.nih.gov/pubmed/10385477
And many have ANCs to the degree where you can't even start them on Clozaril. Further Depakote lowers ANC and the mixture of Depakote and Clozaril can be devastating to it.

And yet another thing learned here and not at home was that due to the reduced ANC from Clozaril, someone with an acute bacterial infection while on that med might not have classic signs to it such as inflammation naked to the eye or an acute high fever. The IM doctor covering the psych unit told me to beware of psychotic patients on Clozaril and not being able to communicate their symptoms well, they may have an infection and the Clozaril could be masking it.
 
From what I can tell, clozapine is used a lot less in the US (although I am sure there are wide variations) than where I was before. As Black patients were overrepresented +++ in our schizophrenia cohorts many were on clozapine. It seems likely that actually this group may be discontinued more commonly, but more unnecessarily, due to benign ethnic neutropenia. My experience elsewhere in the world, using chloramphenicol which also causes neutropenia and agranulocytosis is that we never really saw it in melanesians and it is rare in africa i hear. This suggests huge ethnic differences in ADRs to drugs. One clinical pharmacologist who does Phase I and II trials told me he learned a lot about ethnic differences in physiological parameters from measuring various things in healthy subjects that might be seen as 'abnormal' in caucasian patients who typically make up the majority of those enrolled in clinical trials. I think we are wrong to use the term 'benign ethnic neutropenia' - it is not neutropenia at all for these individual - it is entirely normal and we need to develop (as they have elsewhere in the world) reference ranges for black patients that are more valid in the same way we have done for eGFR etc.

I am probably a bit less risk averse than americans, and just worry less about rashes with lamictal or neutropenia with clozapine etc, and I have had patients with leukemia with a neutrophil count of 0.0-0.3 for years without any ill effects and without the need for antibiotic/antifungal prophylaxis. Which just goes to show that a critical level of neutropenia does not necessarily lead to life-threatening infection (although of course certainly increases the risk).
 
There are only two situations where antipsychotic polypharmacy is acceptable:

1) the patient has failed multiple truly adequate monotherapy trials (that is, 4-8weeks of risperdal monotherapy, then 4-8 weeks of abilify monotherapy, then 4-8 weeks of olanzapine monotherapy. . . ). Stahl suggests that an adequate trial of antipsychotic monotherapy can be as long as 16 weeks (but who actually waits that long?)

2) you're in the middle of a cross-titration

http://onlinelibrary.wiley.com/doi/10.1034/j.1600-0447.2002.2e011.x/fullhttp://onlinelibrary.wiley.com/doi/10.1034/j.1600-0447.2002.2e011.x/full
 
Last edited:
Top