Back HIV/AIDS HIV/AIDS Topics HIV Treatment IDWeek 2013: People with HIV Have Higher Risk of Death after Myocardial Infarction

IDWeek 2013: People with HIV Have Higher Risk of Death after Myocardial Infarction


People with HIV had higher mortality than HIV negative individuals of the same age after hospitalization for acute coronary syndrome, but the excess risk disappeared for people on antiretroviral therapy (ART) with high CD4 T-cell counts, researchers reported last week at the Second IDWeek conference in San Francisco.

Research has shown that people with HIV are at higher risk for cardiovascular disease, but whether this is due to the virus itself, resulting inflammatory and metabolic changes, antiretroviral toxicities, or other factors is not yet fully understood.

Few studies have looked at outcomes among people with HIV following acute coronary syndrome (ACS), which includes heart attacks (myocardial infarction), unstable angina (chest pains), and other conditions caused by blockage of blood vessels supplying the heart muscle.

Michael Silverberg and colleagues with Kaiser Permanente, a large managed care organization, asked whether HIV status, immune deficiency, or ART use affects long-term survival following ACS hospitalization. The Kaiser team previously reported that HIV positive people had a 40%-50% higher risk of myocardial infarction compared with HIV negative people seen at their facilities.

This analysis included all patients hospitalized for the first time with ACS at Kaiser Permanente Northern California between 1996 and 2010. Of these, 226 were HIV positive and 86,321 were HIV negative. Silverberg noted that in the San Francisco area, more than half the population of people with HIV are now over 50 years of age.

People were included if they had ST-elevation myocardial infarction (STEMI) -- typically the more severe type -- non-ST myocardial infarction (NSTEMI), or unstable angina. Deaths were determined based on hospital records, California death certificates, and Social Security Administration files.

The HIV positive ACS patients were more likely than their HIV negative counterparts to be men (94% vs 63%, respectively) and were younger on average (54 vs 67 years). Within the HIV positive group 75% were white, 17% were black, and 9% were Hispanic, Asian or other; in the HIV negative group these percentages were 78%, 8%, and 15%.

Among people with HIV, 58% were taking protease inhibitor-based ART, 26% were on protease-sparing ART, and 16% were not on ART. About 20% had CD4 counts less than 200 cells/mm3, nearly half had 200-499 cells/mm3, and one-third had greater than 500 cells/mm3.

People with HIV were more likely to be hospitalized with STEMI (37% vs 25%), both groups were equally likely to have NSTEMI (about 33%), and HIV negative patients were more likely to have unstable angina (28% vs 42%).

About 13% in both groups had elevated (>160 mg/dL) LDL or bad cholesterol, but the HIV positive group was significantly more likely to have low (<40 mg/dL) HDL good cholesterol (59% vs 40%) and elevated (>150 mg/dL) triglycerides (73% vs 50%). People with HIV were also significantly more likely to be current or former smokers (70% vs 55%).


  • Absolute all-cause mortality rates at 1 and 3 year after ACS diagnosis were similar -- actually a bit lower -- in the HIV positive group compared with the HIV negative group: 7.6% vs 9.8% at 1 year and 16.6% vs 18.6% at 3 years.
  • However, after adjusting for confounding factors including age, sex, race, ACS type, blood lipid levels and smoking, the HIV positive group had a significantly greater hazard of death; adjusted hazard ratios were 2.2 at 1 year and 2.5 at 3 years -- more than twice the risk of death.
  • Comparing different CD4 cell levels, absolute 3-year mortality rates were 29.8%, 17.9%, and 4.9% for HIV positive people with less than 200 cells/mm3, 200-499 cells/mm3, and more than 500 cells/mm3, respectively.
  • After adjusting for confounders, people with 200-499 cells/mm3 had a 3-year adjusted hazard ratio of 2.5 relative to HIV negative people, the same as the HIV positive group as a whole.
  • Those with less than 200 cells/mm3 had substantially higher mortality, with an adjusted hazard ratio of 5.6 -- more than a 5-fold increase.
  • On the other hand, people with CD4 counts more than 500 cells/mm3 had a slightly lower risk of death (adjusted hazard ratio 0.7 relative to HIV negative patients), though this difference was not statistically significant.
  • Turning to antiretroviral treatment, HIV positive people who were not on ART had the greatest absolute risk of death at 22.0%, compared with 16.9% and 12.1% for those on protease inhibitor-based and protease-sparing regimens, respectively.
  • Untreated individuals had an adjusted hazard ratio of 3.4 relative to the HIV negative group, or triple the risk of death.
  • Adjusted hazard ratios for patients taking protease and non-protease regimens were 2.5 and 1.8, respectively, compared with HIV negative people, the latter of which was not a significant difference.
  • However, when comparing protease-based and protease-sparing regimens directly, the difference in mortality between them did not reach statistical significance.

"HIV patients, in adjusted models, had higher all-cause mortality at 1 and 3 years after an ACS hospitalization," the researchers summarized. There was "no mortality difference at 3 years for HIV+ ACS patients with CD4 >500 compared with HIV- patients."

As a limitation of their study, they noted that they did not look at cause-specific mortality -- so they could not determine how many died of cardiovascular-related causes -- and they did not account for treatments such as revascularization surgery or cardiac medications.

Silverberg explained that it was hard to separate out the effects of CD4 count and ART use in this analysis, and he said numbers were too small to see if there was a difference in ACS outcomes among people who took abacavir (Ziagen), which has been linked to heart attacks in some but not all studies.

"The observed higher mortality after an ACS hospitalization for HIV patients appears to be driven by HIV-related factors, including low CD4 and lack of ART use," the investigators concluded. "These results strengthen recommendations for earlier initiation of ART."

In response to an audience question, however, Silverberg said there were not enough people in the study to address the current controversy as to whether ART use is beneficial among those with more than 500 cells/mm3.



M Silverberg, LHurley, A Prasad, et al. Mortality following hospitalization for acute coronary syndrome among HIV+ and HIV- patients. 2nd ID Week Conference (IDWeek 2013). San Francisco. October 2-6, 2013.Abstract 75.