Case History & Initial Impressions
DR. MEL WILCOX: Hello, and welcome to this program on NSAID-Induced GI toxicity. I'm your host, Dr. Mel Wilcox, professor of medicine at the University of Alabama at Birmingham. Joining me in the discussion today are Dr. Jeroan Allison, associate professor of medicine in the Division of General Internal Medicine, also here at the University of Alabama at Birmingham, and Dr. Uri Ladabaum, associate professor of clinical medicine at the University of California at San Francisco.
We will begin this program by considering a case of a 78-year-old man admitted with a history of melena.
You are asked to evaluate a 78-year-old man admitted with a six-hour history of melena. He was in his usual state of health until ten days ago, at which time he began to develop exertional chest pain and underwent cardiac catheterization, demonstrating two-vessel coronary artery disease, requiring stenting. Following the procedure, aspirin at 325 mg a day and Plavix were administered. He was discharged and now presents with two episodes of dark black stool and mild fatigue. He denies nausea, vomiting, abdominal pain or prior history of peptic ulcer disease or any gastrointestinal bleeding.
His past medical history is notable for hypertension, treated with hydrochlorothiazide and beta blockers. He has a distant smoking history and rare alcohol use. His family history is positive for coronary artery disease and diabetes.
Do we have any additional history, Jeroan that you'd like to see that was not mentioned that might be pertinent for his both short-term and long-term management?
DR. JEROAN ALLISON: Well, Mel, certainly, on the surface, it seems like all of the clinical building blocks are here to make a diagnosis in this case. But, in reality, based on my experience and based what the peer review literature tells us, there is something that's really important that I would like to know.
Of all patients who get a prescription for any kind of a nonsteroidal or analgesic, almost half, about 40%, are going to take something over-the-counter. And, actually, usually, not even tell the physician about it. And, in this particular case, you could be lulled into a false sense of security because, perhaps, when those medicines, when the Plavix and aspirin were first started, he didn't have any abdominal pain and just assumed everything is going okay, because there's no abdominal pain, no nausea or vomiting.
So what we do know that, very often, frequently, GI bleeding is asymptomatic and you can't rely on symptoms to predict what's going to happen. So I'd want to ask you, "Does this patient have a history of arthritis? Has he been taking over-the-counter nonsteroidals for arthritis pain or other sorts of musculoskeletal pain?"
DR. MEL WILCOX: Yeah, that's key information. And, sure enough, he was taking over-the-counters three to four days a week. Over-the-counter, usually ibuprofen, which he took for hip pain from a prior injury.
On his physical examination on admission, this is pertinent for his blood pressure of 120 over 70; heart rate 106, but without any orthostatic changes; his stool was dark and Hemoccult-positive.
Uri, we know that clinical presentation is important. In this gentleman, what do we know about his clinical presentation that might predict outcome? In addition to his blood pressure, we know his routine laboratory studies were stable with hematocrit of 31 and a normal MCV. So tell me a little bit about the presentation and what that tells us.
DR. URI LADABAUM: Well, Mel, I think there are several issues. First of all, in terms of the physical assessment, his blood pressure, on the face of it, seems normal, but he is hypertensive and we don't really know what his baseline is. So this may just be relative hypotension for him, we don't know.
Similarly, with a pulse, he's already a bit tachycardic. Granted, he doesn't have orthostatic changes, but this may be somewhat of a blunted response, because he is on a beta blocker. So I think we just need to be careful that he may be a little bit more hypovolemic than we might think, although he doesn't have any orthostatic changes.
In terms of laboratory studies, we know that, oftentimes, the initial hematocrit is not a good reflection of how much blood a patient has lost already. So it may well be that, with some intravenous fluid, further hydration, his hematocrit is actually lower than 31. So, overall, I think we know this is somebody who has lost at least several units of blood.
Now, he is at higher risk of potential complications being an older gentleman who has some comorbidities including coronary artery disease. Now, ideally, he would be somewhat protective now compared to before, because he has a recent stent. But all of those, I think, gives us a gentleman who is at moderate risk here for complications. He is not somebody who's having a massive, relentless, gastrointestinal bleed, most likely. But I do think that prompt endoscopy is something we should consider.
Now, before doing that, a nasogastric lavage would usually be done and that may help us prognosticate a little bit more, but I think, even if we didn't see a bright, fresh, bloody return, this is somebody I would consider for prompt endoscopy for prognosis and also for the possibility of finding a treatable lesion.
Additional Investigations
DR. MEL WILCOX: Okay. Jeroan, any additional laboratory studies or workup you'd like to see I didn't mention?
DR. JEROAN ALLISON: Well, you know, Mel, I'm actually thinking forward to the point, as a general internist, where I'll be seeing this gentleman in clinic after he's been discharged. And I really need to know some additional information to judge the severity of the GI bleed.
He has had the stent placed, and I know that there is an increased risk of subacute closure from the stent, from thrombosis and certainly this guy does have enough underlying coronary artery disease that he could have had heart attack that was not apparent just on the routine lab. So, at this point, I'm really interested in knowing what his EKG showed and knowing if cardiac enzymes were drawn and what the results of those were.
DR. MEL WILCOX: Good. Yes, he did have some changes, left ventricular hypertrophy, no acute changes and his cardiac enzymes were, fortunately, negative. So that was good.
Dr. Ladabaum brought up the issue of endoscopy and that sure is appropriate in this gentleman, based on what we've heard. At the time of the endoscopy, he had a 4 mm benign-appearing antral ulcer without any stigmata bleeding and several scattered antral erosions, which were not bleeding. The esophagus and duodenum were, in fact, normal.
Evaluating for H. Pylori
So let's raise a couple of important questions germane to this case and the issue of NSAID-Induced GI toxicity in general. Uri, at the time of endoscopy, would you evaluate for H. pylori? How would you do it, and do you think it may play any role in this gentleman based on the endoscopic findings?
DR. URI LADABAUM: Mel, I do think we are compelled to look for all the possible risk factors of peptic ulcer disease. Now, the pattern of injury is quite classic for aspirin and nonsteroidals, so it may very well be that, but, unfortunately, we have no way of knowing, once we see an ulcer, whether it was caused by H. pylori, maybe it was there as a silent ulcer and we're now seeing a precipitation of bleeding by the superimposition of aspirin and Plavix or whether this is a de novo ulcer caused by the aspirin.
Because we don't know that, I think we have to test for H. pylori. Now even though there is some decreased sensitivity of rapid urease testing, if there is blood present, I would still do that. One could also consider doing serology.
I think this case is a little bit different from those in which you might see a duodenal ulcer where someone would advocate just treating empirically. But, even there, I think documenting H. pylori will be important, because, if it is present, then, ultimately, we can discuss what treatment and followup testing is appropriate, which becomes one of the major management issues in this case.
DR. MEL WILCOX: Yes, I think I would agree with that. It's important. We don't really know, we don't have a history, we don't know whether it's H. pylori, NSAID and there may be some additive effects for those two.
His physicians did do a biopsy and his rapid H. pylori test was positive. Uri, what would you do with that information now?
DR. URI LADABAUM: I think he needs treatment, Mel, and one can start it right away. Part of the management of the peptic ulcer, even if it had been independent of H. pylori would probably include antisecretory medication; I think most people would do that and that would be included in the H. pylori eradication regimen.
Now, what to give him, most people would go with triple therapy, I think, a proton pump inhibitor, amoxicillin and clarithromycin, for example, for ten to fourteen days in the United States.
Balancing Long-term CV Therapy
DR. MEL WILCOX: If you've eradicated H. pylori, I think we would all agree, it's important that the guy needs to be reevaluated to make sure the H. pylori's eradicated. But would that influence what you're going to do with him on a long-term basis, assuming you did eradicate him?
DR. URI LADABAUM: Well, I think we do need to confirm eradication of H. pylori and the appropriate thing there would be an active test, either a breath test or a stool antigen test.
But I think what you might be getting at, Mel, is the more complicated question of "what's going to be the longer-term cardiovascular treatment for this patient" and "what are going to be his longer-term risk factors of recurrent bleed." Specifically, is he going to be on aspirin longer term? Is he going to continue some of the other agents he was using? And, at that point, we need to think about preventive strategies to try to decrease the risk of recurrence.
DR. MEL WILCOX: Right. So, on a long-term basis, I think he does need something, which we'll address in a minute. But, at least in the short term, you mentioned antisecretory drugs; proton pump inhibitors would be the most likely drugs chosen. We would go ahead and treat his ulcer in the usual treatment regimen. Some might use twice-daily; probably once-daily's enough.
But, Jeroan, this patient has a history of coronary disease, he's on aspirin, he just got his stents. Do we stop the aspirin? Do we not stop the aspirin? That's an important point.
DR. JEROAN ALLISON: And it's really a difficult point. I'm sitting here thinking, Mel, as we're discussing the case, is that you've got me up between a rock and a hard place. There's absolutely good data showing that, after a coronary stent placement, that mortality is reduced by having patients on clopidogrel and aspirin for six months to a year. And the highest risk of restenosis and cardiac complications occurs in the period of time around the stent placement. And so this stent was placed several days ago, but I certainly would not feel comfortable, at this point, stopping the aspirin or actually stopping the clopidogrel.
Now, that decision is reinforced in my mind somewhat because it seems that the endoscopic findings were, I guess, mild to moderate. He was taking over-the-counter nonsteroidals in addition to other agents and he was not on a PPI when he had the bleed. So there's several things that we can do right now to decrease his risk of acute rebleeding.
However, that being said, I want to make the point that he needs to be followed really closely. This is not someone that you can just say, "Oh, continue these medicines, take the proton pump inhibitor and I'll see you back in clinic. Go home from the hospital; I'll see you back in the clinic in three months." I would actually follow up with this gentleman once he was discharged; on a weekly or, perhaps, telephone contact on a daily basis for a period of time.
DR. MEL WILCOX: Yes, I think most of us would use the proton pump inhibitor. His bleeding was modest, endoscopic findings looked pretty good, there's no stigmata bleeding. I think, as you said, we're between a rock and a hard place. We would use the PPI. We would use the aspirin. What about the Plavix? Do we keep that? Or do we not keep that? Would it be safer to stop the aspirin and just use Plavix? Jeroan, that's another tough one.
DR. JEROAN ALLISON: Well, you know, I wish I could say yes. There are studies that show that aspirin alone does not convey the degree of protection that aspirin plus clopidogrel conveys. And then there was the assumption that clopidogrel did not have a lot of GI toxicity. Certainly, we know that it doesn't have the direct irritant effect on the GI mucosa as aspirin or other nonsteroidals. So you cannot just put this guy on clopidogrel and assume that he's not going to be at increased risk for bleeding.
DR. MEL WILCOX: Yes, the proton pump inhibitor, we know it's pretty good at decreasing recurrent ulcer-related bleeding from aspirin, but he's also on the Plavix. We don't have studies of PPI plus Plavix as compared to PPI plus aspirin to see which would be safer in this high-risk patient who really needs these antiplatelet agents.
So, Uri, in the long term, what about the PPI? Do you just treat the ulcer, follow up to make sure it's healed, then stop the PPI? We have a cardiac patient with stents. What are you going to do with him long term?
DR. URI LADABAUM: Well, I think he will end up on aspirin in the longer-term and Plavix, at least for some time, as Jeroan has mentioned. And I think we need to do the best we can to decrease his risk of recurrence. We do have data on aspirin alone and patients who had a bleed, if H. pylori's eradicated, ulcer is healed and they go back on low-dose aspirin, they have, perhaps, a 15% chance of having a repeat event in a year, whereas that is diminished dramatically with concomitant use of a proton pump inhibitor to perhaps a couple of percent a year. This is, again, from one of the groups in Hong Kong a few years ago in the New England Journal. So I think, as long as he is going to be on low-dose aspirin, which he will be, he will require long-term treatment with a proton pump inhibitor.
The other thing I wanted to comment on, Mel, hearing Jeroan's comments is that, sometimes with gastroenterologists, we might be tempted to make a recommendation that is really unrealistic, which is stop these medicines, because we may narrowly focus on the patient's gastrointestinal problems. But, of course, cardiovascular risks are huge issues and we have to realize that many of these patients simply have to continue on these medications. So it becomes an issue of decreasing risk, because we can't completely eliminate it, and balancing the tradeoff between cardiovascular protection versus gastrointestinal toxicity.
DR. MEL WILCOX: Yes, in this fella's clinical presentation, he's more on the mild-to-moderate bleed rather than the severe bleed. I think, all of us, if he had had a severe bleed, require, even perhaps surgery, we would probably try to hold the antiplatelet agents if possible. But, with this gentleman, given the endoscopic findings, I think, if you could, try to continue the antiplatelet agents, aspirin and/or Plavix, get him through the bleed and then, as you said, keep him on the proton pump inhibitor really indefinitely, and we know it is fairly efficacious in decreasing the chance of him having a recurrent ulcer and ulcer-related bleeding.
DR. JEROAN ALLISON: Mel, an additional point I wanted to make, following up on what Uri was talking about, about balancing risk, is that I don't believe there would be any harm in reducing his aspirin from 325 mg a day to 81 mg a day. I know there is still GI risk associated with 81 mg a day, but the data from cardiovascular studies seems to indicate that you don't really get increased protection from thrombotic events by being on a higher dose. So I definitely would decrease the dosage of the aspirin.
And then I would really be looking forward to the day when I could stop the Plavix. This is not someone whom you would just see in clinic as an outpatient and continue the Plavix on for years. I would hope to get him up to at least six months of Plavix, maybe up to a year. But I would have a very low threshold in the future in clinic for stopping the Plavix, although he will be on long-term aspirin really for the rest of his life.
DR. MEL WILCOX: Yes. Uri, Jeroan brings up a good point about aspirin and potential dose. Do we have much data on a dose relationship or dose response, if you will, between aspirin and GI toxicity?
DR. URI LADABAUM: Yes Mel, that's an excellent question and, Jeroan, I think you make an excellent point. There are data to suggest that there is increased risk with increasing dose and so the guiding principle, really, should be to use the lowest dose possible. At the lowest-possible dose, the risk is perhaps twice that of having a complication compared to nonusers.
Now we have a combination of risk factors, because this patient is already at higher risk, already having had an event. But if you talk about higher doses of aspirin, the risk with those is several-fold higher than at the lower doses. So I absolutely agree we should use the lowest dose possible, which still conveys the cardiovascular protection.
DR. MEL WILCOX: You might want to expand on that briefly, Jeroan. Is there really pretty good data, solid data that 85 is just as efficacious as 325?
DR. JEROAN ALLISON: Yes, I think there is solid data that 85 works just as well as 325 mg in a wide variety of circumstances.
DR. MEL WILCOX: Uri, this patient had an upper GI bleed. He was given a proton pump inhibitor. Would you add anything else to the PPI? Some might say, "Well, why don't we add Carafate or antacid?" Are you happy with the PPI alone in standard dose?
DR. URI LADABAUM: I think so. We don't really have studies to guide us for additional therapy. I think standard-dose PPI is reasonable.
DR. MEL WILCOX: Okay. This gentleman also is taking over-the-counter NSAIDs, as Jeroan brought up very nicely. And could it be important in why he had an upper GI bleed, Uri?
DR. MEL WILCOX: Right. So this gentleman does have some hip pain. What might you give him when you see him in clinic?
DR. JEROAN ALLISON: Well, Mel, I can tell you from having seen a lot of patients like this is that he probably will continue taking those medications, and it's really going to require some serious discussion with him, because I have a feeling his pain is real. However, I think there are a lot of alternatives that should be tried. Often, acetaminophen has not been tried in an adequate dose, although it could have been in this situation. For a degenerative arthritis, we should be thinking about weight loss. There's a group of medicines called the nonacetylated salicylic acid derivatives like salicylate that often will work, but do not have the GI toxicity.
There are really a whole host of other physical modalities such as physical therapy and exercise that should be used, so I think that one of the things the general internist would want to address in clinic is some serious effort to pursue approaches to getting him off at least regular nonsteroidals. And then if you can't get him to stop, at least you would be able to minimize his exposure.
DR. MEL WILCOX: Yes, Uri brought up the point about multiple NSAIDs. If you could get patients off and use alternatives. That would clearly be best. Even if he's on a proton pump inhibitor, he's probably still at risk for an additional complication.
Magnitude of NSAID-Induced GI Toxicity
Now, let's discuss NSAID GI toxicity in more detail in a general fashion. And, also, I'd like to, again, bring in my two coworkers here, Jeroan and Uri.
We hear a lot, Jeroan, about NSAID-Induced GI toxicity. Is this really overblown or is this really an important issue?
DR. JEROAN ALLISON: Well, as a matter of fact, I think it's underblown. Depending on which study you look at, the number of deaths attributable to just GI side effects or the toxicity of NSAIDs ranges from around 3000 all the way up to 16,000 and that doesn't even consider the cardiovascular toxicity, which we can talk about a little later on. And, furthermore, I think there's a general under-recognition of this problem in the physician community and certainly in the patient community. There's data from that suggests that patients take these medicines, they take them in combination (over-the-counter, prescription medicines) and are not aware of the risk associated with taking them in combination or even taking them singly. So I think that it's a very important problem.
Furthermore, from a generalist point of view, and we can talk about this more later in the program, I worry about kidneys, and I also worry about heart failure.
DR. MEL WILCOX: Yes, today we focus on GI, but clearly those are very, very important issues.
We've talked a little bit about the over-the-counter NSAIDs, and we know that they do have some toxicity. But we, oftentimes, tend to forget to ask the patient about over-the-counter use. Here in the Southeast, we have several NSAIDs that are used for a variety of other reasons which, oftentimes, don't make much sense, but we have to ask the patient about that.
Uri, in your experience in California, is over-the-counter NSAID use a significant issue?
DR. URI LADABAUM: Absolutely. It's extremely common. I think all the points that both of you have brought are very germane to our discussion. The other thing to point out is that, sometimes, the patient himself or herself does not even know that they're getting a nonsteroidal agent. They might be taking combination preparations for cold symptoms relief or other things, which contain nonsteroidal agents.
And maybe more here in the West than in your part of the country Mel and Jeroan, I don't know. Sometimes people take herbal remedies and other things where we, frankly, have no idea what's in there. And sometimes when these are tested or if you do a drug assay on the patient, they have evidence of salicylates.
Is There a Safe NSAID?
DR. MEL WILCOX: Which NSAID is safer? Is there really one that's safer than not, Jeroan?
DR. JEROAN ALLISON: Well, you have to think about safety along two different dimensions. One would be cardiovascular safety and well, really, three dimensions. Another, GI safety and maybe even from a renal standpoint. I think, from a GI point of view, there is good data, Mel, that we've discussed ourselves that the selective COX-2 agents have a safer but not completely safe GI profile.
DR. MEL WILCOX: What about the older, traditional NSAIDs? Are any of those safer than others? If somebody couldn't afford or didn't want to use a COX-2, what about the whole issue of heart disease? They might not use that. Can we guide anybody about any of the traditional NSAIDs as to more safe for the GI tract?
DR. JEROAN ALLISON: I think that may be a dangerous discussion to enter into comparing the nuances of safety. Certainly there are a lot of studies that are out there that compare the different agents and you could draw different conclusions. But I think the most important point to emphasize, from my point of view, is that all nonsteroidal agents, especially the nonselective agents, carry a substantial risk of GI toxicity.
DR. MEL WILCOX: Yes. We've talked so far about the use of over-the-counters, the spectrum and importance of NSAID-related GI toxicity. When a patient walks into our office or your office, Uri, and you might consider giving an NSAID, are there any risk factors for NSAID-related GI toxicity that you can glean from the patient's history at that time?
DR. URI LADABAUM: Absolutely. I think many of them were illustrated by the case we discussed before. So certainly the most worrisome thing, from the gastrointestinal point of view, is if a patient has had a past history of a complicated ulcer, with a hemorrhage or perforation or obstruction. Those patients are at extremely high risk and I would avoid nonsteroidals at all cost, if possible.
If patients have had a past ulcer that maybe was not as severe as having a severe hemorrhage, that's still a risk factor. If patients are using low-dose aspirin, which becomes a very, very big issue and, for many of them, they are taking it for the appropriate indication, the combination with an NSAID would be an additional risk factor.
Now, once you're thinking about prescribing the NSAID, the issue is what's going to be the dose; what's going to be the duration. We'd like to use the lowest of each of these. Advancing age is a risk factor and then also concomitant medications. If patients are taking steroids, which by themselves don't seem to be a risk factor, but, in combination with NSAIDs, they do seem to increase the risk of an adverse event. Or if patients are on blood thinners such as warfarin, all of these things would place patients at higher risk.
DR. MEL WILCOX: Right. The dose of aspirin may be important and, in your practice, Jeroan, you're seeing older and older patients. As our U.S. population is living longer, the word on the street in the older population is the importance of aspirin, "Aspirin's going to protect my heart." What can you tell your patients about when to use aspirin and what dose?
DR. JEROAN ALLISON: Well, certainly, if someone is older and they have other risk factors, you'd want to use the lowest dose possible, which is probably 81 mg three times a week. I think, often, in practice, we see patients who are younger and who do not appreciable cardiac risk factors and are started on an aspirin a day or take an aspirin a day and really don't need it.
I remember a patient that I saw in clinic just this past week who actually had a history of recurrent GI bleeding, had been worked up, Mel, by your colleagues many times at UAB. And then, in clinic, I just happened to bring up aspirin and GI distress and he said, "You know, I've been taking all along over the past several years," and I really think that was the cause, just an aspirin a day.
And what was even more remarkable about this patient was that he was relatively young and didn't have other cardiovascular risk factors, so there was really no need for the aspirin and he had been taking it just because.
DR. MEL WILCOX: Yes.
DR. JEROAN ALLISON: Word on the street was "Hey, an aspirin a day is good for me."
DR. MEL WILCOX: Right. We talked a little bit about the traditional NSAIDs and GI toxicity, very briefly about the coxibs. Now, of course, we're down to one coxib that we can use. Uri, what data is out there that can help guide us as to: Are these drugs really safer? What patient population may you want to use them in?
DR. URI LADABAUM: Well, Mel, as you know, this area has been evolving. It's confusing, and there are a lot of things that we don't know yet. So I wish that we could have a very simple message. But I think we can talk about a few things that we do now and that is that, although there are probably differences within the coxibs, there may well be a class effect that results in an increased risk of cardiovascular complications.
Now, as you mentioned, the one that we still have available is celecoxib. Part of the issue there may be that celecoxib, at the traditional doses, may not have a significant risk of increasing cardiovascular complications. There's more work that's needed there, but perhaps at the higher doses, it does. So that's one area of concern: Is the patient at higher risk of cardiovascular disease already, based on other risk factors, and should we avoid coxib in that patient? And I think that's definitely something to consider.
In somebody who does not have cardiovascular risk factors, perhaps we can think about selective coxibs with a little bit more comfort and there, as Jeroan has mentioned, the gastrointestinal complications are perhaps 50% as likely as with traditional NSAIDs, so we would be gaining an advantage there.
I do want to make the point, though, that is sometimes overlooked, which is that the combination of low-dose aspirin with a coxib essentially behaves like a nonselective inhibitor. And if you think about the mechanism of action of these medications, it shouldn't be surprising. The COX-2 agents were designed to spare COX-1 and, by doing so, try to avoid the problems with gastrointestinal toxicity. But, when you add low-dose aspirin, which essentially is a nonselective inhibitor, you have a combination that behaves like a nonsteroidal agent.
So the whole question of the coxib use in somebody who's at cardiovascular risk is complicated even more by the fact that that person may well be on aspirin.
Balancing the Risk
DR. MEL WILCOX: Jeroan, in your circles, is there a lot known about the aspirin-coxib relationship as far as GI toxicity, the internist may know "Hey, I heard that coxibs are not good for heart, but I'm going to give you the patient aspirin," is that okay?
DR. JEROAN ALLISON: Well, I think there is an under-appreciation of the fact that when you have that combination, you may not be getting the cardioprotective effect of aspirin and you may not be getting the GI protective effect of the coxib. There may not be a general awareness of that yet in the physician community, but it is important to note, as Uri was saying, that the addition of aspirin plus a coxib may not only negate the beneficial effects of the coxib, but also the addition of a coxib or another nonsteroidal agent to aspirin may diminish the cardioprotective benefit that you would be getting from aspirin.
DR. MEL WILCOX: Right. I hear Uri say "dose of coxib"; we've heard that several times today. Dose response with aspirin, potentially dose response with coxib, multiple NSAIDs, higher doses, GI toxicity. So it's kind of a theme we're hearing is there's GI toxicities related to the dose of these medications.
Jeroan, I'd also like to ask you about cardiovascular toxicity with the traditional NSAIDs. The coxibs are getting all the bad press that they hurt the heart. How much do we really know about the traditional NSAIDs and cardiac risk?
DR. JEROAN ALLISON: I think there definitely is cardiac risk associated with the traditional NSAIDs. I think that it's probably safe to assume, in fact, that all of these agents have risk and that risk can be placed on a spectrum. I think that the COX-2 selective agents may have a bit more risk than the traditional agents, but still, at this point, in the evolution of our knowledge, we have to assume that all of the traditional NSAIDs increase that risk. And then with these several mechanisms operating, one mechanism is that all traditional NSAIDs and selective COX-2 agents increase blood pressure and can also have harmful effects on the kidneys and renal blood flow.
So I think that it's very important to keep this in mind in patients who have cardiovascular risk. And it's also very important to keep this in mind when you're evaluating someone who has hypertension or who has previously uncontrolled hypertension that now is coming in with an elevated blood pressure.
DR. MEL WILCOX: Yes, we tend to focus on the GI toxicity and forget some of the other potential problems with these agents, such as the heart and kidney that you had mentioned, in fact, earlier.
So if you have the coxib and you have the traditional NSAIDs, Uri brought up the issue of GI risk. And it seems to me that the patient needing the coxib the most may be the patient least likely to get it because they're older, maybe they're on aspirin, other risk factors for GI toxicity.
Reducing the Risk
Uri, we talked about GI complications and, say, if your patient is at high risk, what do we know about agents that we can use to reduce risk?
DR. URI LADABAUM: Well, Mel, there are two main classes. I would say one is the antisecretory medications and then the prostaglandin analog, misoprostol. So, taking the second one first, the nice thing about the literature for misoprostol is that we actually have large, prospective outcome trial data with outcomes that matter. And we do know that there is a modest decrease in the risk of complications with concomitant use of misoprostol.
The problem is that a significant number of people, maybe 20, 25% or perhaps even higher numbers, have side effects that are intolerable, including bad diarrhea and other types of side effects. It may also be an issue that younger women with rheumatologic disorders should not be taking this medication if they're considering pregnancy, for instance, so there may be somewhat of a limitation in the real-world applicability of the use of misoprostol. But that is an agent that can decrease complications and can actually also help healing of NSAID-Induced ulcers, even if you continue the NSAID.
The other major class is the antisecretory drugs. The H2-receptor antagonists are of some benefit in duodenal ulcers but not gastric ulcers and, as we know, NSAID toxicity is often associated with gastric ulcer.
So the second antisecretory medication would be a proton pump inhibitor. We've talked about those at length and they may well similar effects as those of misoprostol in the long term in terms of preventing the real-world events that we care about. Unfortunately, we don't have actual data on that, a large outcome trial on the proton pump inhibitor. But the shorter-term studies, the ulcer-healing studies would suggest that, if we were to extrapolate from those, that the long-term benefit of the proton pump inhibitor and misoprostol might be comparable.
DR. MEL WILCOX: The misoprostol, as you said, there are outcome studies that it does work. There's also an important point to be made. You need to take the four times a day, 800 mcg. A little bit less than that, 600, 400, you don't really get the same effect. You brought up the issue of side effects. But, at least with this drug, you take less, it's not going to be as effective, which, as you suggest, is leaning us more to use proton pump inhibitors once a day, relatively safe.
The Role of H. pylori
What about the role of H. pylori in your patient who may have prior history of risk factors for GI toxicity? You're going to consider a COX-2 or a traditional NSAID. Should we be looking for H. pylori in that situation? Uri?
DR. URI LADABAUM: That's another complex question. There is some data that, before initiating NSAID use, if we identify H. pylori and eradicate it, that there might be a decrease in risk. The problem, of course, is that many patients, many millions of patients are starting NSAID use every day. Or every year, rather. What we're facing is really a public health question, to a large question. What's going to be the overall benefit? Do we want to be testing so many people for H. pylori? To a large extent, I think that's unanswered at the population level.
I would point out that there are other theoretical benefits, such as decreasing the risk down the road of gastric cancer, but when we're faced with the question of starting a patient on an NSAID, if they have not had a prior peptic ulcer, do we test them for H. pylori? I think that can be argued both ways. I don't think it's settled now.
DR. MEL WILCOX: Yeah, I think there is, as you said, the data is unsettled. If I saw a patient who had a prior history of a peptic ulcer and the patient was never evaluated for H. pylori, in that situation, you would, particularly, if they're going to use long-term NSAIDs or even if you wouldn't, given the natural history of ulcers and H. pylori.
DR. URI LADABAUM: Absolutely. I think there's no question. With a past ulcer, they need evaluation. If they don't have past ulcer, do we prospectively test all of these people? That's less clear.
DR. MEL WILCOX: Right. Jeroan, in the internal medicine circles, do your colleagues have a search-and-destroy philosophy about H. pylori in the NSAID setting? What's the word on the street in internal medicine circles?
DR. JEROAN ALLISON: Well, I don't know that I can speak for all internists, but I think there really is a propensity for testing for H. pylori and especially basically it's something easy that can be done without much difficulty, although it's very difficult to get patients to complete a prescribed course of treatment.
And so, as I was listening to the discussion unfold, I was thinking how that I would be agreeable with testing patients who are at risk of GI bleeding that are going to be initiating NSAID therapy or who have had a previous ulcer history, but, certainly, I don't think it's wise to go on a seek-and-destroy mission on patients who are currently on NSAIDs and been taking them chronically and there's no evidence that they're having GI intolerance.
Are There NSAID-Induced Intestinal Problems?
DR. MEL WILCOX: There are other potential side effects that we occasionally encounter related to NSAIDs that may raise issues of disease lower in the GI tract. Do we know much about small-bowel disease? Is there such a thing as NSAID-Induced colonic problems? How would you address that question, Uri?
DR. URI LADABAUM: There are certainly complications further down in the GI tract. I think that's been recognized for a long time. There can be ulcers in the small bowel, ulcers in the colon. There's a so-called diaphragm disease of the diaphragm-like strictures that can lead to obstruction. So it's clear that it happens.
What we don't have as good a handle on is how often it happens. It's possible that maybe 15%, 20% of the complications from NSAIDs are actually not in the upper GI tract within the reach of the upper endoscope. But this is really one of those areas that is in need of further research. And, also, one of those areas that is slightly different from the discussion we've had so far in that our usual protective strategies probably do not work further down in the GI tract.
DR. MEL WILCOX: Particularly the proton pump inhibitors. I guess we don't know as much about misoprostol, if that's a systemic effect. One might suggest that that could be potentially beneficial, but I guess you're right. We just don't know.
Seeing those capsule studies that NSAIDs cause a lot of toxicity and erosions in the small bowel, that can be pretty scary. Do you believe that? How do you incorporate that into your management, particularly patients who present with occult bleeding, iron deficiency, what about that type patient, Uri?
DR. URI LADABAUM: I think, in somebody who has already manifested a problem, as you say, occult bleeding, iron deficiency, then, depending on the workup, if you don't find any abnormalities in the standard upper and lower gut testing, they might end up getting a capsule. And if you do demonstrate abnormalities that can be attributable to the NSAID, then that makes it an even more compelling case for trying to stop the medication. Even in the absence of abnormalities in the capsule studies, though, I think you would have enough suggestive evidence to suspect the NSAID and stop it in a patient with unexplained iron deficiency anemia, for example.
The quandary, I think, is how much stake to put in a study that maybe had less of a clinical indication. In other words, if we were to do these capsule studies on all comers who are using NSAID and there are some studies out there like that, you may see a relatively high prevalence of these lesions in the small bowel. Which is, by the way, analogous to what you would see in the scheduled endoscopy studies of upper GI toxicity.
But it's important to remember that the majority of these endoscopically- or capsule-detected lesions will never have an important clinical manifestation. So I think they're a marker of the potential toxicity of the diagnosed, but I would try to evaluate the significance of those in the whole clinical context of whether the patient has actually had a problem, a clinical problem or not.
DR. MEL WILCOX: On the front line, in the internal medicine clinic, you clearly must see patients who are on chronic NSAIDs, somebody checks the stool for occult blood, they're iron-deficient. Is the whole issue of small-bowel disease on the radar of the general internist?
DR. JEROAN ALLISON: Well, from my perspective, I don't start thinking about capsule studies, but, instead, think about trying to get patients to stop those medications. I think, often, patients come in and have anemia without an obvious cause or have evidence of GI bleeding with a negative upper or lower GI evaluation. And we don't think not just about the small bowel, but we don't think that NSAIDs could really be the cause.
So, from my perspective, I think it's really important, when you're faced with these situations, to always consider that the patient could be taking an NSAID. You need to go in and talk to them very carefully to see if they're taking over-the-counter NSAIDs or if another physician has prescribed these types of medications.
Cost-Effectiveness Issues
DR. MEL WILCOX: We haven't talked much about cost-effectiveness issues. So I'd just like to raise the question, first, to Uri, in your patient who you're trying to risk-stratify, where does the cost come in as far as what drug you put them and when you put them on it? Is that an important consideration?
DR. URI LADABAUM: Mel, I think the reality, actually having thought about this a lot, is that it's very difficult and some would argue it doesn't really have a place to think about that when you're facing a patient one-to-one. Those are discussions that one can have when you're talking about approach to a population at large or you're going to have a certain policy in your hospital, in your health system, in your HMO, in your national health system. Who is going to be able to get a certain medication? Then one can try to think about cost considerations and the cost-effectiveness of these things. But, on an individual patient level, it's really very difficult to consider that and I think one has to go with what you think is the safest thing.
Now the cost issue sometimes does enter the equation in a very real way for patients, because they may be able to afford certain medications and not others. But one of the things that has really changed the equation here, I think, is the availability of an over-the-counter proton pump inhibitor. Before, one could talk about the economics of an affordable, generic, nonselective NSAID combined with an expensive proton pump inhibitor versus an expensive COX-2 selective agent. But now that we have an over-the-counter proton pump inhibitor, that may have changed the equation a little bit.
I should say, Mel, sorry to interrupt, that one of the issues that, unfortunately, has not really been addressed in some of the formal decision analyses that have been done because the issue wasn't really known at the time is the whole cardiovascular toxicity issue. And once you start bringing that into the equation, not only in terms of harm but also in terms of economic cost, it becomes a potentially very important issue.
DR. MEL WILCOX: Yeah, I think that's right on target. Jeroan, in your circles, we heard a little bit from Uri, a gastroenterologist, obviously, is gastroenterologists think a little bit different about these kinda issues. What about as an internist, this whole cost-effectiveness and risk stratification, etc.?
DR. JEROAN ALLISON: Well, I certainly think now there may be a shift more towards using a low-cost generic proton pump inhibitor combined with a nonselective agent. But, once again, you know, I just want to emphasize that that regimen does not come free of cardiovascular risk and does not come free of GI risk.
DR. MEL WILCOX: Clearly, the cardiovascular issues, as Uri brought up, if you're doing cost-effectiveness analysis, you throw a few heart complications in there and it sure will sway the data one way or the other.
Uri and Jeroan, I'd like to thank you both for excellent discussion. I hope you, the audience, that this was a very learning a very good learning tool for you and thank all of you for joining us today and we appreciate the opportunity to be here.
