June 30, 2010
The kidneys are a target of HIV infection. Long-term infection causes inflammation, which slowly degrades these organs.
The EuroSIDA dataset contains health-related information on almost 17,000 HIV-positive people from 35 countries. EuroSIDA can provide a powerful tool to analyze health-related trends in HIV-positive people.
The kidneys process tenofovir and there have been rare cases of tenofovir users developing serious kidney dysfunction. To investigate the issue of chronic kidney disease, researchers from EuroSIDA reviewed their large database of information on HIV-positive people who initiated anti-HIV therapy with or without tenofovir. They found that some tenofovir users had a detectable and significant decline in kidney health.
Researchers reviewed data collected from 2004. They checked eGFR (estimated glomerular filtration rate) to get an idea of how the kidneys were working. Chronic kidney disease was defined in the following way:
Researchers assessed the use of anti-HIV therapy over time to find out if there was any relationship between kidney damage and specific therapies.
In reviewing their dataset, researchers focused on data from 6,843 people who had used tenofovir and for whom multiple eGFR assessments were available.
At the start of the study, the profile of these people was as follows:
Participants were monitored for an average of four years.
Before discussing the latest analysis from the EuroSIDA study, it might be helpful to review how researchers talk about risk.
Normally, when we speak about risk, we mean overall (or absolute) risk, which tells us the probability that a certain event (such as developing chronic kidney disease) will happen over a given period of time. Generally speaking, the overall risk of chronic kidney disease in HIV-positive people who are taking tenofovir is low.
Overall, about 3% of nearly 7,000 people in the EuroSIDA study developed chronic kidney disease, so in the EuroSIDA study, chronic kidney disease is generally uncommon. However, everyone's overall risk for chronic kidney disease is different, so some people living with HIV may have a much higher overall risk depending on risk factors such as whether or not they have high blood pressure, diabetes or a family history of poor kidney health.
When researchers want to study the risk of chronic kidney disease associated with taking a particular drug, they try to measure the percentage change in the overall risk that results from taking that drug. This is called the relative risk. The relative risk provides information about how much more (or less) likely chronic kidney disease will be if the person is taking the particular drug studied compared to if they are not taking the drug. For example, a study may find a 70% increase in the relative risk of chronic kidney disease. This means that the risk of developing chronic kidney disease is 70% higher among people taking the drug than among people who are not taking the drug. The relative risk measures a change in risk, it does not measure the overall risk.
In order to know what the relative risk, reported by researchers, means for you, you need to also consider your overall risk for chronic kidney disease. Your doctor can help you determine your overall risk. If your overall risk of chronic kidney disease is low prior to starting the drug, a 70% increased relative risk might not be significant compared to someone else whose initial overall risk for chronic kidney disease is very high. And bear in mind that a 70% increase in relative risk does NOT mean that there is a 70% chance you will get chronic kidney disease.
Now, onto the EuroSIDA study.
Researchers found that 3% of people (225 people) developed chronic kidney disease over an average of four years. The vast majority of these 225 people entered the study with an eGFR greater than 60.
The researchers took into account many factors that have the potential to play a role in affecting the results, including these:
Taking all of these (and likely other) factors into account, the EuroSIDA analysis found that the following drugs were associated with an increased relative risk for developing chronic kidney disease as follows:
The EuroSIDA team performed several different analyses and each confirmed their results.
Compared to people who never used tenofovir, among participants who stopped taking tenofovir, the risk of developing chronic kidney disease when they stopped using tenofovir was as follows:
The high rate of relative risk after stopping tenofovir therapy probably arose because their doctors thought that these patients were at high risk of kidney damage and wanted them to stop using tenofovir, noted EuroSIDA researcher Dr. Ole Kirk.
In contrast to the findings from tenofovir users, when participants stopped taking atazanavir or Kaletra, their risk of developing chronic kidney disease was low and similar to that seen in people who were not using these drugs.
The EuroSIDA study design is that of a cohort study. Such studies are very good at finding associations -- in this case, between the use of certain drugs and an increased relative risk of kidney damage. However, observational studies by their nature can only find associations; they cannot prove cause and effect. That is, they cannot prove that taking a particular medicine(s) will indeed cause a particular effect (serious kidney damage).
Furthermore, confounding or channeling bias is a problem that bedevils observational studies and makes drawing firm conclusions difficult when interpreting the data. Observational studies are useful for finding associations that can later be explored in studies of a more robust design.
Despite these caveats the EuroSIDA analysis is important and raises a red flag: It is possible that a number of anti-HIV drugs, particularly tenofovir, may have a negative impact on the kidneys over the long-term, even for a year after people stop taking this drug. EuroSIDA also raises concern about the safety of atazanavir for the kidneys.
It is likely that these problems were not seen in the initial clinical trials of these two drugs because they enrolled relatively healthy people (as do most trials testing new compounds). The patients in the EuroSIDA cohort are reflective of the reality of living with HIV for many years and the burden of having multiple health conditions.
At present, EuroSIDA does not have sufficient information to assess the impact of the following drugs on chronic kidney disease:
Long-term studies are needed to continue to provide information about the effect of these and other medicines. Other important datasets, such as the following, need to conduct their own analyses to confirm, extend or refute EuroSIDA's initial findings:
In the years ahead these analyses will be eagerly awaited and much debated and discussed among doctors and statisticians.