This page has moved. Click here to view.


Drug Interactions

Clinically important drug/drug interactions. A drug/drug interaction is a pharmacologic response which cannot be explained by the action of a single drug alone, but rather it is due to two or more drugs acting simultaneously. I made absolutely no comments as to whether or not it was beneficial or harmful. The reason for that is because there are situations actually where we do use drugs in combination, because they interact and they do have a beneficial effect to our patients. It really demonstrates this point very clearly, and that is the use of probenecid which actually inhibits the excretion

The ones that we are most concerned about and the ones that I will spend the most time talking about today are those that are harmful. Obviously we are not in the business of trying to do harm to our patients, and when itís the harmful ones itís the ones that we donít anticipate. Because they are

So what types of drug/drug interactions are there? Well, thereís the pharmaceutical, pharmacodynamic and pharmacokinetic. Let me explain a little bit further. The pharmaceutical drug/drug interactions simply are nothing more than when you take two clear solutions, mix them together, and you end up with a solution thatís developed a precipitate. This is the type of interaction that we repeatedly get asked questions about from the

What we are going to spend the rest of the morning talking about are the ones that are more clinically oriented. If you want to take a look at the right had side of the slide here, it is the pharmacodynamic drug/drug interactions. And pharmacodynamic drug/drug interactions happen when you have

What about enzyme inhibition? This is where things get really complicated, unfortunately. Enzyme inhibition basically involves inhibition of the enzyme system. Now this can happen within 24 hours. We are not requiring protein synthesis. We are simply blocking the enzyme system. There are numerous enzymes within the liver. When I went through pharmacy school we talked about the cytochrome P-450 system and it was the cytochrome P-450 system and it was homogenous. There was no difference. All the enzymes were the same. Well, now itís been determined that there are many different enzymes within the liver and they are all affected differently by different drugs, depending on the structure of the drug. So this is where we get into things like, why verapamil will affect one, will have a drug interaction, but nifedipine will not. Because they are different

First of all, benzodiazepines can be affected by induction. For example, rifampin. It has been shown that patients who are receiving rifampin and come in for outpatient surgery and they plan to use midazolam, that there has been complete absence of sedative effect by typical doses of midazolam because rifampin causes midazolam to be metabolized so quickly and so elevated doses need to be used in that situation. Also, inhibition of metabolism: we have fluoxetine, fluvoxamine and nefazodone are the serotonin re-uptake inhibitor type drugs. Grapefruit juice is listed there and

The calcium channel blockers: again, rifampin has resulted in increases of metabolism of verapamil primarily. Verapamil is being used basically for slowing down AB conduction and when the patient is also on rifampin it can abolish that effect and result in problems for the patient necessitating dosing increases or changes in the therapy. Again, as far as inhibition, erythromycin and clarithromycin can inhibit liver enzymes resulting in increases in felodipine and also in nifedipine. Nifedipine serum concentration with the patient complaining of peripheral edema and decreased effectiveness in blood pressure control.

Cisapride with induction: there really are not any clinically significant type side effects that need to be concerned about when a drug that induces liver enzymes is induced simultaneously. However, when drugs that inhibit cisaprideís metabolism are administered - drugs like ketoconazole, itraconazole, fluconazole, metronidazole, erythromycin and clarithromycin - a reaction very similar to those with the non-sedating antihistamines occurs, with potential for ventricular tachycardia and ventricular fibrillation. There have also been some reports of death as a result of 

Cyclosporine: we worry about the use of things like phenytoin, carbamazepine, Phenobarbital, also rifampin in a patient thatís receiving cyclosporine and the potential for increasing itís metabolism to the point that an inadequate immunosuppressive response would be seen, necessitating dosage increases of cyclosporine. Now we have used drugs to inhibit cyclosporineís metabolism on purpose, because cyclosporine is such an expensive

With the oral contraceptives and estrogens, all of the anticonvulsants - with the exception of sodium valproate and the newer ones, lamotrigine and gabapentin - will induce the metabolism of the oral contraceptives for all practical purposes making them ineffective. There was one study that looked at a patient who was receiving phenytoin and they administered Ortho-Novum 150 - which isnít even available anymore - to the patient and they had non-detectable estradiol levels. It took up to the equivalent of Ortho-Novum 1-100 before estradiol levels were such that the patient would not become pregnant. So the use of a different form other than oral contraceptives is suggested in those patients. Also rifampin and griseofulvin can have similar effects. Under inhibition, I have not affected. What I mean by non-affected is that contraception will still be achieved. That does not mean that the patient may not complain of some type of side effects, like maybe increased nausea. May complain of breast tenderness and those sorts of things. So you may see exacerbation of side effects but you will not see decreased efficacy of the therapy.

The HMG-CoA reductase inhibitors, primarily lovastatin - although it can happen with all of them - the big problem is with inhibition. If we inhibit its metabolism through itraconazole or erythromycin we can end up with problems with increases in creatine kinase, with the progression on to rhabdomyolyses and those type of problems.

The protease inhibitors: one problem that we run into it with it is in a patient receiving rifampin of the treatment of TB, and TB is becoming a bigger problem with our AIDS population and they need to be treated for tuberculosis. The CDC has actually published a report with their recommendations of how to treat patients in this situation. Because rifampin can cause sub-therapeutic levels of the protease inhibitors and we know that part of the reason for resistance to the protease inhibitors is sub-therapeutic levels. The last thing we want to do is treat the patient for TB and then make their protease inhibitors ineffective, or for them to develop resistance to that. So if you ever have the need, the CDC does have very specific recommendations. Also with inhibition, if the patient needs fluconazole or ketoconazole, dosage decreases of the protease inhibitors may be needed.

Phenytoin: phenytoin is one of those drugs that is an inducer on its own so you donít typically think about it being affected by inducers or inhibitors. It can, with rifampin, you can have near - or you can have therapeutic failure of phenytoin because it will be metabolized away so quickly that the serum concentrations will be diminished. Also with inhibition, any of those drugs that I have listed there can result in significant increases. For example fluoxetine can result in 100 to 200% increases in serum concentration of phenytoin. So itís something you need to pay attention to and make dosage adjustments.

Terfenadine and astemizole: Iím sure all of you have heard and seen all of the information on this. Received many "Dear Dr." letters about it. As far as induction, the use of a drug that induces liver enzymes is not a problem with these agents. The place that we get in trouble is with drugs that inhibit liver enzymes. I will make the comment that loratadine, which is the other non-sedating antihistamine, is inhibited. The metabolism of it is inhibited by all of these drugs. The difference is the parent compound does not have cardiotoxic properties. So it is an option in this situation. But all of these drugs that I have listed there can result in syncopal episodes, torsades have been reported and there also have been deaths reported. In this situation grapefruit juice actually down there at the bottom, has resulted in Q-t prolongation. So there actually hasnít been a fatality reported but thereís been some things suggesting that it could be a problem.

Theophylline is affected both by inducers and inhibitors resulting in changes in serum concentrations, necessitating both or either dosage increases or decreases. Warfarin also is one of those drugs that can be significantly affected resulting in potential problems, potentially significant problems, for the patient. So close monitoring is needed. The one that I will point out, that Iíve actually seen problems happen with, is the trimethoprim

Weíll move on now and get out of the metabolism area and weíll talk about altered excretion. Diuretics basically increase sodium resorption and lithium always follows sodium in the kidney. If you wonder about lithium, just think about what happens to sodium and thatís what happens to lithium. When a patient takes a diuretic we have increase in sodium re-absorption, so you also have increased lithium re-absorption with increases in lithium serum concentrations happening. Also, with non-steroidal antiinflammatory agents, this is an example of indomethacin. This is a kind of confusing slide. The top curve here with the open circles is urinary lithium levels. The bottom line is plasma lithium levels, and this center section that is kind of striped is the time period that the patients received indomethacin. So if we look here at the time they began taking indomethacin, plasma lithium levels rose that directly correspond to urinary lithium excretion going down. When the indomethacin was discontinued the urinary excretion went up and the plasm lithium levels came back down corresponding to that. It appears as though sulindac is a non-steroidal that does not have affects on the prostaglandins in the kidney. So it appears as though it does not - Clinoril or sulindac would not have this affect on serum lithium levels. But all the other non-steroidals will have this effect to varying degrees. So you would want to monitor lithium levels in those patients who you end up

Miscellaneous: I hate to throw things in miscellaneous. It always appeared to me that when somebody says miscellaneous that is was an afterthought. They didnít think well enough to put it into the handout where it actually belongs. I promise you, this is not the case here. I just honestly could not find a spot that it fit anywhere else so I put it into miscellaneous. And that is: the combination of monoamine oxidase inhibitors, both the type A and the type B, when used with meperidine - and I donít know if any of you remember the case that hit the lay press, the Libby Zion case. This is the case that kind of led to the big discussion of residents are overworked, they spend too much time in the hospital. Their term of duty or the length of