CROI 2009: Elevated Rate of Heart Attacks and Strokes in HIV Patients on HAART Begins to Decline at California Kaiser Permanente

As HIV positive people began to live longer thanks to the development of effective antiretroviral therapy, cardiovascular disease became a growing concern, especially given evidence that HIV infection itself and the drugs used to treat it can increase cardiovascular risk.

According to a Kaiser Permanente study presented at the 16th Conference on Retroviruses and Opportunistic Infections (CROI 2009) last month in Montreal, elevated rates of heart attacks and other cardiovascular events among HIV positive patients appear to be declining, and may be approaching those of HIV negative individuals.

Daniel Klein and colleagues identified HIV positive adult members of the Kaiser Permanente California health system -- which cares for more than 6 million patients in California -- matching them 10:1 with HIV negative members of the same age, sex, and year of enrollment in the cohort.

Overall, they analyzed data from 20,305 HIV positive and 203,500 HIV negative participants, contributing 89,683 and 1,063,567 person-years (PY) of follow-up data, respectively. Almost all (90%) were men, the average age was 41 years, and the HIV positive group included a slightly larger percentage of white patients (56%) than the HIV negative group (47%).

Using a standardized disease classification system, the investigators collected information on hospitalizations for myocardial infarctions (MIs or heart attacks) and cerebral vascular disease (strokes).

Results

• In unadjusted data from 1996 through 2008, the MI hospitalization rate was 3.0 per 1000 PY for HIV positive patients (264 events) compared with 1.7 per 1000 PY for HIV negative participants (1800 events) (P < 0.001).
• Adjusted for age and sex, the annual MI rate fell by 5% in the HIV positive group and declined by 1% in the HIV negative group.
• The unadjusted stroke hospitalization rate was 1.8 per 1000 PY for HIV positive patients (160 events) compared with 1.1 per 1000 PY for HIV negative participants (1136 events) (P < 0.001).
• Adjusted for age and sex, the annual stroke rate stayed the same (0% change) in the HIV positive group but rose by 3% in the HIV negative group.
• Over the successive time periods, crude MI hospitalization rates among HIV positive patients were as follows:

2.6 per 1000 PY in 1996-1997;
3.3 per 1000 PY in 1998-1999;
2.8 per 1000 PY in 2000-2001;
3.7 per 1000 PY in 2002-2003;
3.1 per 1000 PY in 2004-2005;
2.5 per 1000 PY in 2006-2008.

• Crude stroke hospitalization rates among HIV positive patients were as follows:

1.7 per 1000 PY in 1996-1997;
0.9 per 1000 PY in 1998-1999;
1.8 per 1000 PY in 2000-2001;
2.0 per 1000 PY in 2002-2003;
2.0 per 1000 PY in 2004-2005;
2.0 per 1000 PY in 2006-2008.

• Looking at adjusted rate ratios for MI hospitalizations among HIV positive compared with HIV negative participants, the excess rate fell to borderline significance in the most recent period:

RR 1.9 in 1996-1997 (P = 0.011);
RR 2.3 in 1998-1999 (P < 0.001);
RR 2.0 in 2000-2001 (P < 0.001);
RR 2.2 in 2002-2003 (P < 0.001);
RR 1.7 in 2004-2005 (P < 0.001);
RR 1.3 in 2006-2008 (P = 0.057).

• A similar pattern was observed in the latter 3 periods for adjusted rate ratios for stroke hospitalizations:

RR 2.3 in 1996-1997 (P < 0.001);
RR 1.3 in 1998-1999 (P = 0.513);
RR 1.7 in 2000-2001 (P = 0.010);
RR 2.1 in 2002-2003 (P < 0.001);

RR 1.9 in 2004-2005 (P < 0.001);
RR 1.4 in 2006-2008 (P = 0.052).

• The percentage of HAART-treated patients receiving stavudine (d4T; Zerit) declined over time, while percentages taking tenofovir (Viread, also in the Truvada and Atripla combination pills) and atazanavir (Reyataz) increased.
• Use of lipid-lowering medications increased over time.
• The mean Framingham cardiovascular risk score for HIV positive members fell from 7.2 in 2000-2001 to 6.4 in 2006-2008.

"Rates of MI and stroke in a matched sample of HIV negative patients were significantly lower than among HIV positive patients," the investigators stated.

However, they continued, "during 1996-2008, the rates of MI among HIV positive and HIV negative patients converged such that in 2006-2008 the difference in rates between the two groups became statistically non-significant."

"The convergence was due to a decline in the rate of MI among HIV positive patients while the rate among HIV negative patients was stable," they added.

With regard to strokes, they stated, "We observed the same convergence in stroke rates. However for stroke, the convergence was due to a rise in the rate of stroke among HIV negative patients while the rate among HIV positive patients was stable."

Explaining their findings, the investigators concluded, "Among HIV positive patients, the observed decline in rate of MI and stable rate of stroke is consistent with 1) a shift to more lipid friendly antiretroviral retroviral regimens, 2) increased use of lipid-lowering therapy, and 3) effective management of traditional cardiovascular risk factors as evidenced by stable Framingham risk scores despite an aging population."

3/13/09

Reference

L Hurley, W Leyden, L Xu, and others. Updated Surveillance of Cardiovascular Event Rates among HIV-infected and HIV-uninfected Californians, 1996 to 2008. 16th Conference on Retroviruses and Opportunistic Infections (CROI 2009). Montreal, Canada. February 8-11, 2009. Abstract 710.