icon star paper   Articles  
Back grey_arrow_rt.gif
Protease inhibitors and cardiovascular outcomes in patients with HIV-1
  Summary: Protease inhibitors for treatment of HIV-1 have been linked with increased risk of hyperlipidaemia and hyperglycaemia. In a cohort of 5672 outpatients with HIV-1 seen at nine US HIV clinics between January, 1993, and January, 2002, the frequency of myocardial infarctions increased after the introduction of protease inhibitors in 1996 (test for trend, p=0.0125). We noted that 19 of 3247 patients taking, but only two of 2425 who did not take, protease inhibitors had a myocardial infarction (odds ratio 7.1, 95% CI 1.6-44.3; Cox proportional hazards model-adjusted for smoking, sex, age, diabetes, hyperlipidaemia, and hypertension-hazard ratio 6.5, 0.9-47.8). Our findings suggest that, although infrequent, use of protease inhibitors is associated with increased risk of myocardial infarction in patients with HIV-1.
Edit note: In the text of the report below you'll see that risk factors normally associated with risk for premature heart disease play an important role in contributing towards cardiovascular disease. Protease inhibitors were strongly, but not significantly, associated with myocardial infarction in this study (p=0.065). The authors note that the incidence of myocardial infarction is low (1.42/1000 person-years; 19 out of 3247 persons taking protease inhibitors) and occurred infrequently. Taken together with the important role risk factors play, this suggests eliminating risk factors ought to help a great deal in limiting risk or avoiding heart disease.
Research Letter, Lancet 2002; 360: 1747-48
Scott D Holmberg, Anne C Moorman, John M Williamson, Tony C Tong, Douglas J Ward, Kathy C Wood, Alan E Greenberg, Robert S Janssen, and the HIV Outpatient Study (HOPS) investigators*
Shortly after their introduction in 1996, protease inhibitors were shown to be effective in the treatment of HIV-1, but also to be associated with hyperlipidaemia and insulin resistance. Thus, many have wondered whether long-term use of protease inhibitors could result in raised risk of myocardial infarctions, angina, or cerebrovascular accidents (CVA).
The HIV Outpatient Study (HOPS) is an ongoing prospective observational cohort in which patients have been continuously recruited and followed up since 1992. Study sites are nine clinics (seven private, two public) in eight US cities (Atlanta; Chicago; Denver; Oakland; Philadelphia; Stony Brook, NY; Tampa, Fla; and Washington, DC). Information is abstracted and entered electronically on site by trained data abstracters shortly after each outpatient visit. For this analysis, we identified all HOPS patients seen more than once between January, 1993, and January, 2002. We verified all diagnoses of myocardial infarction, angina, and CVA by review of hospital discharge records. We counted each condition only once, so observations were censored after an incident case or diagnosis. Individuals who began to take protease inhibitors during our observation period could contribute data to both the non-protease inhibitor use group and the protease inhibitor use group. Observations were censored after 1 year off protease inhibitors for patients who discontinued treatment and did not restart within 12 months.
We calculated person-time under observation and incidence of myocardial infarction per 1000 person-years overall and by calendar year. We used SAS (version 8.0) for all statistical analyses. We used Mantel-Haenszel, x2, t test, or Fisher's exact test for univariate analyses, which are presented as odds ratios (95% CI), and adjusted with left-censored Cox proportional hazards analyses. We also constructed multivariate (logistic regression) models from covariates associated with myocardial infarction.
We assessed 5672 patients infected with HIV-1 for 17 712.4 person years of observation. Their mean age was 42.6 years. 82% (n=4658) of the cohort were men and 38% (n=2178) were not white. With respect to factors affecting risk of transmission of HIV-1: 63% (n=3586) were homosexual, 12% (n=698) were injection drug users, 19% (n=1066) were heterosexual, and 12% (n=658) had other or unspecified risk factors. The incidence of myocardial infarction in these patients rose after the introduction of protease inhibitors in 1996 (figure).
21 patients-19 of 3247 taking (1.42/1000 person-years of observation) but only two of 2425 who did not take (0.46/1000 person-years) a protease inhibitor drug-had a myocardial infarction. The 19 individuals who were taking a protease inhibitor had a median age of 47 years (range 28-67) at the time of their myocardial infarction, and the two who were not taking a protease inhibitor were aged 39 years and 47 years at the time of their myocardial infarctions. Patients who did or did not have a myocardial infarction did not differ greatly with respect to ethnic origin, HIV-transmission risk group, most recent CD4+ count, nadir CD4+ cell count, or most recent or highest viral load (data not shown).
Use of protease inhibitors was strongly associated with the likelihood of having a myocardial infarction (table; p=0.002), and correlated with diabetes mellitus and hyperlipidaemia (Spearman correlation coefficients, 0.08 and 0.34; p<0.012, for both). Left-censored Cox proportional hazards analysis showed a strong relation between incidence of myocardial infarction and use of protease inhibitors (unadjusted hazard ratio, 8.06; 95% CI 1.14-56.8; p=0.036). A model that controlled for all risk factors listed in the table showed that protease inhibitor use was still strongly, but not significantly, associated with incidence of myocardial infarction (p=0.065). Logistic regression models, controlling for all the factors listed in the table, yielded similar results (adjusted odds ratio 4.92, 95% CI 1.28-32.3; p=0.041). There was no significant difference in the length of time the 19 patients who had a myocardial infarction and the 3228 protease inhibitor-taking patients without myocardial infarction took protease inhibitor drugs (mean 3.5 years and 4.1 years, respectively; p=0.17). (Patients were censored at the time of their myocardial infarctions, so their observation times were shorter than for patients without a myocardial infarction.) No individual protease inhibitor was significantly more likely than the others to be associated with incident myocardial infarction (data not shown).
There were also 15 instances of angina, 11 among 3247 individuals taking protease inhibitors and four among the 2425 not taking protease inhibitors (odds ratio 1.93; 95% CI 0.63-5.96). There were 14 CVAs, seven in each group (0.75, 0.24-2.36). Our results suggest that myocardial infarctions and perhaps angina could arise in patients taking protease inhibitors.
The overall frequency of myocardial infarctions rose greatly after protease inhibitors were introduced, and the incidence in HOPS patients rose after protease inhibitors had been used for a few years. Most patients who had a myocardial infarction or an episode of angina also had other traditional risk factors associated with the diseases (Table: 28% had hypertension, 57% smoked, 28% diabetes, 38% were >50 yrs of age, 90% were men, and 47% had evidence of dislipidemia). Cox proportional hazards and multivariate analyses that controlled for these factors nevertheless showed a strong correlation between protease inhibitor use and myocardial infarction.
To the extent that the HOPS cohort serves as a sentinel population, doctors who treat individuals with HIV-1 should be aware of a possibly increased cardiovascular disease risk in those taking protease inhibitors. Myocardial infarction is still infrequent, usually occurs in people with other risk factors for cardiovascular disease, and should not detract from the appropriate use of these drugs for patients with HIV-1. Early intervention to stop smoking, and appropriate diagnosis and treatment of hyperlipidaemia, insulin resistance, and hypertension are all ways of minimising this complication of protease inhibitor use.
  icon paper stack View Older Articles   Back to Top   www.natap.org