A systematic review of both the published literature and conference abstracts on the relationship between protease inhibitor (PI) use and cardiovascular risk has found that with the exception of atazanavir, all currently available PIs do appear to elevate risk factors for heart disease.
However, although a majority of the studies found an interrelationship between the risk factors for heart disease - which include increased blood lipids and thickening of the arteries - and myocardial infarction (heart attack) - only two out of six of the large observational studies reviewed found this translated into an increase in heart attacks amongst patients on PIs.
The paper, which appears in the October 1st issue of Clinical Infectious Diseases, assessed 71 different studies up to September 2002, including 14 randomised controlled trials which are the kind of studies most likely to represent true differences between comparison arms, although the statistical power of these kinds of studies to detect differences in rare events - like myocardial infarction or stroke - is limited. This is why the six large observational studies - which featured 2000 or more patients - included in the systematic review are also important.
Three of the large observational studies evaluated the effect of PIs on the incidence of myocardial infarction and two (67%) found a PI-associated worsening. The other three large observational studies reported the total number of cardiovascular events and coronary artery disease and found no difference between those on PIs and those not on PIs, although exactly which cardiovascular events occurred were not specified.
Thirteen of the 14 randomised controlled trials reported changes in total cholesterol levels of which nine (69%) reported a significant worsening in the PI group or improvement when PI treatment was suspended. In total, of the 48 studies that looked at cholesterol levels on PIs, 36 (75%) found that PIs increased total cholesterol, which is a significant risk factor for heart disease.
Ten studies looked at high density lipoprotein (HDL) levels and 12 at low density lipoprotein (LDL) levels. HDL is known as ‘good’ cholesterol, since it appears to protect against heart disease, whereas LDL is known as ‘bad’ cholesterol. Only four out of the 10 (40%) showed a worsening in the HDL level, whereas all 12 (100%) found that PIs were associated with a worsening of LDL levels.
Seven of the 12 (58%) randomised controlled trials reported a significant worsening of triglyceride levels for PI-naive patients who initiated PI therapy, or a significant improvement in triglyceride levels in patients who switched from PI-based to NNRTI-based HAART. In total, out of the 48 studies that looked at cholesterol levels on PIs, 35 (73%) found that PIs increased triglyceride levels. The cardiovascular implications of hypertriglyceridaemia are less clear than those of high total or increased LDL and/or decreased HDL cholesterol levels, but high triglycerides are probably associated with a slight increase in the risk of cardiovascular disease.
Other factors for cardiovascular risk assessed in this review included hyperglycaemia (high blood sugar levels) or diabetes mellitus (11/19 or 58% reporting a PI-associated worsening); carotid intima thickness or atherosclerotic lesions, which indicates a hardening of the arteries (7/8 or 88% reported PI-associated worsening); and endothelial dysfunction which indicates stroke risk (2/3 or 67% reported a PI-associated worsening).
Although the body of evidence is mounting that PI treatment is associated with elevated risk of heart disease and possibly stroke, recently released guidelines from both the US and the UK on treating PI-associated metabolic abnormalities may help reduce the risk over the long term.
However, given the short period of follow-up since PI-treatment began (cardiovascular disease is usually measured in terms of 10 year risk) the true impact of PI-associated metabolic disorders will only be fully appreciated in years to come.
In conclusion, the authors note that “the PI class remains a very effective class of drug for treating HIV infection,” but that prior to prescribing PIs for the first time physicians “should evaluate their HIV-infected patients for cardiovascular risk factors.”
Further information on this website
HIV-positive Canadians need more heart surgery since HAART - news story September 2003
New US HIV lipid guidelines favour lifestyle changes over drug therapy as first line - news story August 2003
Cardiovascular risks of HAART: More data but little more clarity? - news story February 2003
Lipid increase not needed for HIV protease inhibitors to trigger cardiovascular problems - news story February 2003
Rhew DC et al. Association between protease inhibitor use and increased cardiovascular risk in patients infected with human immunodeficiency virus: a systematic review. Clinic Infect Dis 37: 959-72, 2003.