February 4, 2008
A poster presented at CROI found a higher risk of heart attack (myocardial infarction) in people using the anti-HIV drugs Ziagen (abacavir, and in the fixed-dose combination pills Epzicom and Trizivir) and Videx (didanosine, Videx EC).
These findings came from an analysis from the Data Collection on Adverse Events of Anti-HIV Drugs, or D:A:D study. D:A:D is a multi-center study that follows several groups of people, mostly in Europe, and examines the unintended side effects of HIV drugs. Earlier research from D:A:D found that HIV drugs on their own increased a person's risk of heart attack by 26%. The study has previously reported on a higher risk related to taking HIV drugs from a specific class -- protease inhibitors.
The D:A:D study is not the kind of controlled study typically used to determine the effects of a drug, in which all elements of treatment are carefully controlled to eliminate any possible biases in the evaluation of the drug. In D:A:D, patients were not randomly assigned to receive any particular treatment. Instead, it is what is called a large, simple trial or observational study. In this kind of study, patient data is simply collected from large numbers of hospitals and health care facilities, without regard for what treatments are used. This type of study provides less specific and less precise conclusions, but it is believed that this is compensated for by the fact that they follow much larger numbers of patients. Observational studies simply look at what happens to patients over time and then attempt to statistically analyze the factors that may have contributed to the result.
The current report from D:A:D examined whether another class of HIV drugs, called NRTIs, also affected the risk of heart attacks. Of the five drugs studied, D:A:D found a statistically significant increase in the risk of heart attacks in people using Ziagen or Videx, but not Retrovir (zidovudine, AZT), Epivir (lamivudine, 3TC) or Zerit (stavudine, d4T). The study found that the risk of heart attack increased by 90% for people on Ziagen and 49% on Videx.
While these numbers sound alarming, it's important to note that they still represent a relatively small change in the absolute number and risk of heart attacks. When the numbers of an event, such as heart attacks, are small to begin with, and the number of patients being studied is very large, even a few additional events might represent a large percentage of change. For example, if the number of events in a study was just one to begin with, adding a second event would be a 100% increase. In this particular study, there were 2.6 heart attacks per 1,000 patient years among people not using abacavir vs. 6.1 per 1,000 patient years. This represents a 90% increase, but the absolute risk of heart attack, whether on or off the drug, was still low.
The absolute risk of heart disease (meaning the likelihood a person is to be diagnosed with heart disease) was strongly associated with known risk factors for heart disease, such as smoking, family history, high blood pressure, cholesterol and the presence of any other form of heart disease. The effects of Ziagen and Videx were present in people regardless of their absolute risk of heart disease, but they were most pronounced in people at the highest underlying risk of heart disease.
Putting all of this in perspective, the study investigators pointed out that the level of increased risk from smoking, for example, was far greater than the increase risk from the use of the two drugs. Thus, a patient who was a smoker could make a much greater reduction in the risk of heart attack by stopping smoking than by stopping the use of the drugs.
Interestingly, the study found that the higher risk was only present when people were actually taking the drugs. If they stopped taking the drugs, the increased risk disappeared, regardless of how long they had used the drug. This contrasts to the protease inhibitors, where the risk grows over time and is not quickly reversed. In the D:A:D study, the length of time a person used abacavir or ddI seemed unrelated to the risk of heart attack.
There is no known biological mechanism that explains this NRTI finding, nor any known explanation why this effect would be seen with these two particular NRTIs but not the other drugs of this class. Protease inhibitors are known to increase cholesterol and the risk of diabetes -- both risk factors for heart disease. However, no such affect was seen with these two NRTIs. Moreover, tens of thousands of people have been followed in controlled clinical studies of both of these drugs and no increased risk of heart attacks was noted in those studies. More research is needed to determine whether this is a real effect and if so to understand the mechanism.
It's important to re-emphasize that the overall rate of heart attacks seen in this study were small. This report does not mean that people on Ziagen or Videx should necessarily stop taking their drugs or switch to others. If you take either drug and you have heart disease or are at high risk for it -- due to family history, smoking or other known risk factors -- discuss these findings with your doctor.