HIV Management In Depth

Assessing and Acting on Cardiovascular Disease Risk in HIV-Infected Patients

Marshall Glesby, M.D., Ph.D.Jens Lundgren, M.D.
Marshall Glesby, M.D., Ph.D.Jens Lundgren, M.D.
Click here to launch multimedia player
A Podcast Discussion With Marshall Glesby, M.D., Ph.D., and Jens Lundgren, M.D.

December 6, 2010

 < Prev  |  1  |  2  |  3  |  4  |  Next > 

Assessing Risk


Myles Helfand: In the meantime, then, when it comes to the brass tacks for a clinician, what factors should they be taking into account when they try to determine cardiovascular risk in an HIV-positive patient? Is it the same exact risk assessment procedure that any clinician would normally use for any patient? Or should they be factoring in things like previous abacavir experience, or length of HIV infection?

Jens Lundgren: I can start by saying that the prediction of underlying risk is becoming quite an important management tool. So therefore I appreciate the question that you're raising here. At least in D:A:D, our advantage is that we can study the observed risk. So whatever risk prediction that you apply, you can essentially compare that risk prediction to what actually happens to patients.

We just went through that exercise recently and published that. What was interesting was that if you used, for example, the Framingham equation, the traditional equation developed with a population on the East Coast of the United States back in the '60s and '70s, at least for relatively younger people -- so that's people below the age of 50, being HIV infected -- the Framingham equation was over-predicting what was actually observed. That may have a logical explanation, given the fact that it's a very different population from where the Framingham equation was developed.

We actually went through the exercise of developing an HIV-specific risk equation that can be used, and which actually takes into account not only traditional risk factors but, as well, for example, abacavir, which remains an independent predictor of the underlying risk in patients. We put that online, so if people are interested in that, they can at least use that for patient management.

As far as I'm aware, that's the only sort-of-validated risk equation in HIV patients. And by validated, I mean that you develop a risk prediction and then you actually compare that to the observed risk. But I'm not sure. I would be curious to hear what Marshall is using to make risk predictions in his cohorts of patients.

Marshall Glesby: Clinically, when I'm faced with an individual patient -- just getting back to part of the original question, in terms of how to approach them -- I think most expert guidelines suggest that we should be monitoring lipids and glucose probably more aggressively than in someone without HIV of a similar age. So I do tend to check lipids, at least annually, depending on what the values are. And I follow fasting glucose and, in select patients, would consider doing an oral glucose tolerance test. I am concerned about undiagnosed diabetes.

I have used the Framingham risk equation, although there may be some subtle differences, in terms of actual event rates. What you've seen in your cohort, compared to what's predicted, I think, is an approximation that's reasonable. I have not used HIV-specific equations to date, but I look forward to trying those out.

I don't order inflammatory markers, like C-reactive protein. I think the data are fairly limited, in terms of the predictive value of C-reactive protein, and whether it is the same in someone with HIV compared to someone without HIV.

There are, I think, some pretty reasonable data from a number of studies suggesting that [C-reactive protein] levels may remain elevated despite controlling HIV replication. I think Jens alluded to the subanalysis from the SMART study, and there are some data from the ACTG [AIDS Clinical Trials Group] trial, as well, that suggest that levels may end up being persistently elevated in some patients, despite controlling the HIV.

So I'm not sure what the role of further risk stratifying using things like C-reactive protein might be. But I think we need further data on that.

Myles Helfand: Would you extend that to other inflammation markers, like IL-6 or D-dimer?

Marshall Glesby: I don't check those markers clinically. I think it will be interesting to see what happens in some of the ongoing studies, in terms of using those markers. I think there are some interventional studies that are planned that will use some of the data that came from SMART, in terms of cutoffs for some of these values and then potentially intervening to reduce inflammation to see if they affect some markers.

But, in 2010 or 2011, are we there yet, in terms of using these things clinically? I personally don't think so. I'm not sure what Jens and his colleagues do across the Atlantic. I'd be interested to hear.

Jens Lundgren: Well, I can just say that I 100% agree with you that, first, in terms of the risk prediction, I think the Framingham equation is an excellent risk prediction tool, which obviously takes into account lipids and the development of diabetes, etc. -- you know, all the traditional risk factors. And it's an excellent way to stratify a cohort.

I certainly also agree with your point about the measurement of inflammatory and coagulation biomarkers. It's a great research tool at the moment to further understand this, but it's really not ready for clinical routine usage.

I think first and foremost, as Marshall already said, because we don't know actually how to react to this: What is a rational intervention in case you find that these biomarkers are elevated? What do you actually do, if you want to practice evidence-based medicine? It's an area of intense research effort at the moment. We really need to sort this out, so we don't start to measure things in people, and start to react to it by giving more medicine when we are unsure whether it actually provides any benefit to them. So I 100% agree with you on that.

Caring for a New Patient

Myles Helfand: This sounds like a very stressful clinical situation. In the context of, let's say, an HIV clinician, a new patient presents. Let's say it's a 45-year-old man who doesn't smoke, but who has a family history of some cardiovascular disease. You do some testing. You find he's got borderline high cholesterol, borderline high triglycerides. So it's not clear that he's a time bomb, but his CD4 count is low, as frequently happens in clinical settings because patients present late. Let's say his CD4 count is 250.

How do you start, as a clinician, to pull apart the different aspects of potential risk factors for complications, and make the decision about when to start antiretroviral therapy, and what to start with, given the potential cardiovascular risks?

Marshall Glesby: In a patient like that, clearly, treating their HIV should be the priority. If their CD4 count is 250, and that's confirmed, then I think most of us would be more focused on that, as the shorter-term issue, rather than the cardiovascular risk factors that you mentioned. Personally, I would see what evolves with treating their HIV.

If someone does have a number of cardiovascular risk factors, I certainly would keep that in mind in terms of selecting an antiretroviral regimen. But for the most part, I think the regimens that we're thinking about using first-line are relatively lipid friendly, and hopefully not causing major metabolic complications, the way some of the older regimens that we used tended to do.

So I would start out with treating the HIV and then certainly, if the patient persisted, or had a worsening lipid problem, then I would address that as appropriate. But I don't know that I would approach that patient fundamentally differently than someone without HIV. In the back of my mind, I might be thinking that their HIV and perhaps their years of uncontrolled HIV could be predisposing them to atherosclerotic changes. But I have not necessarily had that be the driving force, in terms of my decision making. I would focus on modifiable risk factors and perhaps just keeping in mind that they may be at somewhat higher risk because of their underlying HIV infection.

Myles Helfand: How aggressive would you be about incorporating statins and fibrates into the mix if a person has borderline high cholesterol or triglycerides? Where do you decide to start filtering that stuff in, particularly in light of a couple of potential interaction issues here and there between antiretrovirals and some lipid-fighting meds?

Jens Lundgren: First of all, this patient is in need of effective antiviral therapy to improve his overall health. Cardiovascular disease is not the primary focus here. Also, this person, who supposedly has just entered your clinic, he needs to start being comfortable with his HIV disease, and being comfortable with taking medicine and adherence and all sorts of stuff. That's what you need to focus on initially.

Let's now assume that half a year or a year has gone, and he has been stabilized. He is well-suppressed and no major problems have arisen. Then you need to start to assess his underlying risk. If his underlying risk goes above, say, 15% (a 10-year risk) and he has elevated lipids (total cholesterol or LDL cholesterol), I would then, in that case, consider initiating a statin as my primary focus.

Obviously, if he is diabetic, you need to treat that; if he is hypertensive, you obviously need to treat that, also, earlier. But in terms of lipid management, that would be, at least in my regular practice, the strategy: to control his HIV first, and then see how it goes.

Also, if I can just say: We're talking cardiovascular disease here, right? But many people who initiate [antiretroviral] therapy are also starting drugs that can affect other organ systems -- for example, the kidneys. For example, tenofovir [Viread] can impair renal function. Not in many patients, but it can happen. So there are a lot of issues to deal with.

All these things need to be sorted out before you start to think of what will happen thereafter and whether you need additional medications in the mix. At least, that would be how I would approach it. It's not an emergency. I mean, it's something you need to deal with. But you need to, first of all, get his HIV under control.

Myles Helfand: I'm assuming lifestyle and diet changes would also be pretty high on the priority list before you consider any kind of pharmaceutical intervention, right?

Jens Lundgren: Absolutely. This person needs to deal with a lot of issues, right? He's just been told he has HIV. He's just been told that he needs to be on medication for the rest of his life. Whatever sexual health impact that may have -- there are a lot of issues for that person to deal with.

But gradually, you start to get to know your patient well, and understand his background, and can talk about these other sorts of more subtle issues that he needs to start focusing on. At least, that has been my approach so far.

Marshall Glesby: I agree with that completely. And I think, in terms of drug-drug interactions, there are certainly important drug-drug interactions between some of the statins and protease inhibitors, for example, that we do need to keep in mind. But they are not insurmountable problems. If someone needs a statin, we can pick a statin that will not put the person at higher risk of a clinically important drug-drug interaction.

Jens Lundgren: Or just start with a low dose, and then see how it goes.

Myles Helfand: So in most cases, then, it sounds like you guys would not be basing your decision about what to prescribe, in terms of antiretrovirals, on a person's cardiovascular risk profile. Like, we're not talking about holding back on protease inhibitors. As you mentioned a little earlier, Dr. Glesby, the current preferred first-line antiretrovirals are not really believed to be so much of a problem, when it comes to cardiovascular risk. Some of the older ones, though, might be a little bit different. And there are certainly criteria that come into play that might make those drugs more desirable from the standpoint of treating HIV infection. For instance, there may be complications involving a person's prior drug resistance to a preferred first-line drug, or other confounding factors that might prevent certain antiretrovirals from being used.

Would you make any modifications to what you're prescribing, in terms of antiretrovirals, based on cardiovascular risk? Or is that only after everything else has been exhausted?

Marshall Glesby: In the specific patient that you mentioned, it doesn't sound like that person is going to be in a particularly high-risk category, based on the information that you provided. I would probably try to more formally risk-stratify a patient like that. But given what you said, I don't know that cardiovascular risk issues would be the primary driving force, in terms of selecting a regimen. I mean, there are other issues to factor in, in terms of dosing convenience, as you alluded to -- you know, whether the person has any underlying resistance, etc. So it may be a consideration, but I don't know that it's the most important consideration.

Jens Lundgren: I would agree with that.

Marshall Glesby: I think we do have plenty of options. I mean, there's an integrase inhibitor; protease inhibitors that appear to not have major adverse effects on lipids; and, of course, the non-nucleoside option, if the patient is sensitive to non-nukes.

Myles Helfand: That seems like pretty practical, straightforward information: Take it step by step. Tackle the HIV. If there are any risk factors, then tackle the lifestyle and diet issues first. If the risk appears significant -- high cholesterol, high triglycerides -- if they're above that threshold, then move forward with statins or fibrates. But it sounds like neither of you are suggesting a particularly aggressive take on ameliorating or preventing the development of cardiovascular risk. It's more just, "Let's wait and see. Let's treat the HIV. And then see where things go."

 < Prev  |  1  |  2  |  3  |  4  |  Next > 

This article was provided by TheBodyPRO. It is a part of the publication HIV Management in Depth.

No comments have been made.

Add Your Comment:
(Please note: Your name and comment will be public, and may even show up in
Internet search results. Be careful when providing personal information! Before
adding your comment, please read's Comment Policy.)

Your Name:

Your Location:

(ex: San Francisco, CA)

Your Comment:

Characters remaining: