 |
|
|
| |
Cardiovascular Update from CROI 2017
|
| |
| |
by Priscilla Hsue MD
Professor of Medicine, UCSF
The Conference on Retroviruses and Opportunistic Infections was held February 13-16, 2017 in Seattle, Washington. Overall, many studies evaluated cardiovascular disease in the setting of HIV infection including large cohort studies, smaller pathogenesis based investigations, and studies performed globally.
Cohort Studies: Large cohort studies both within the US and abroad continue to provide key insights into HIV-associated cardiovascular disease. Each cohort differs in strengths and weaknesses and ascertainment of HIV-related data, traditional risk factors, and outcomes.
The DAD cohort continues to provide key data on ART and CV risk as was presented in in an oral presentation by Ryom L et al. In the > 35,000 individuals in DAD, cumulative use of darunavir and ritonavir but not atazanavir/ritonavir was associated with increased CVD risk (RR 1.59, 1.33-1.91 and RR 1.03, 0.90-1.18/5 years). This finding persisted after adjustment for lipids and other factors including bilirubin (Ryom, L et al, Abstract 128LB).
In the CNICS cohort, ischemic stroke including small vessel and cardioembolic subtypes were common in HIV and were related to illicit drug use or infection (Crane HM et al, Abstract 347).
Using ERCHIVES, Butt A and coauthors performed a large comparison of HCV+ and HCV- individuals and evaluated risk of acute MI. Interestingly, HCV+ individuals with HCV had a higher rates of AMI as compared to HCV- individuals. The risk increased with higher lipid levels and was more profound in younger HCV infected individuals (Abstract 574).
In the Veterans Aging Cohort Study (VACS), a time-updated VACS index was predictive of both acute MI and mortality while nadir CD4 count was predictive of mortality. When nadir CD4 count was combined with the VACS index in a single model, nadir CD4 count was no longer predictive of either AMI or mortality. This finding suggests that current health indices as ascertained by the VACS time updated index is a better gauge of risk than nadir CD4 count which may be further upstream in the disease pathway (Rentsch C et al, Abstract 626).
Using the National Inpatient Sample, HIV infected individuals made up 0.35% of ACS hospitalizations. Individuals with HIV were more likely to be male, African American and to have lower income status. In adjusted analysis, HIV-infected individuals were less likely to undergo cardiac catheterization (OR 0.68, 95%CI 0.62 to 0.73), less likely to undergo revascularization with PCI or CABG (0.80, 95%CI 0.73 to 0.89), and less likely to get drug eluting stents (OR 0.74, 95%CI 0.63 to 0.86). These findings suggest health care disparities in the setting of HIV (Abstract 615, Clement M et al).
Using the North American AIDS Cohort Collaboration which includes 14 cohorts, trends in statin prescription were estimated from 2003-2012 (Althoff KN et al, Abstract 619). Over this time interval, the statin treatment gap was ascertained which is number of individuals who are eligible for statin therapy using guidelines for the general population but who are not prescribed statins. While the gap decreased over this time interval, the gap remained substantial over time (from 70% in 2003 to 58% in 2012) which may be due to differences in the two study populations.
Pathogenesis studies: Unique pathways which connect HIV to end organ disease continue to be explored using novel biomarkers. Many studies were presented that confirmed prior findings that traditional risk factors continue to be key contributors to cardiovascular risk along with indices of untreated HIV.
In the VACS cohort that was an oral presentation, higher bilirubin was associated with lower risk for incident heart failure and acute MI. Among HIV-infected individuals, the finding of elevated bilirubin and lower risk of heart failure persisted but was not significantly associated with AMI (Marconi V et al, Abstract 127).
In a longitudinal analysis of individuals in the Hawaii Aging with HIV Cardiovascular Study, HDL cholesterol efflux was inversely associated with classical monocyte numbers (beta -0.21 p=0.02; Abstract 612, Mannem SR et al). Reverse HDL cholesterol transport from macrophages/monocytes may be inhibited by HIV which may contribute to increased CVD.
In the Multicenter AIDS Cohort Study (MACS), cumulative viral load was significantly associated with coronary artery stenosis ≥ 50% (p<0.05) and increasing VL exposure was associated with > 2 fold increased odds of stenosis (OR 2.4 for log increase in duration of viral load, 95%CI 1.4, 4.0). VL was not associated with non-calcified plaque (Haberen S et al, Abstract 624).
In a case control study of treated and suppressed HIV-infected individuals in the CNICS network with and without Type 1 MI, carnitine was found to be independently predictive of MI (OR 4.95 top quartile vs lower quartiles, 95% CI 1.29, 18.95, p=0.02). The other gut metabolites including TMAO, betaine and choline were not associated with MI. (Sinha A et al, Abstract 625).
In the WIHS/MACS cohorts, metabolomics profiling identified plasma tryptophan as being associated with decreased risk of incident plaque (RR 0.75, 95%CI 0.64 to 0.88) per SD while kynurenic acid and Kyn/Trp ratio were significantly associated with increased risk for incident plaque (RR 1.34, 1.08-1.65 per SD and RR 1.41, 1.22-1.64 per SD). (Qi Q et al Abstract 636LB).
Long term non-progressors had similar carotid artery IMT as compared to uninfected individuals, but overall HIV-infected individuals with detectable viremia were more likely to have carotid plaque as compared to controls (PR 1.37, 95%CI 1.06 to 1.79; McKibben R et al, Abstract 632).
In the WIHS study, hsCRP was not associated with IMT in HIV-infected women while it was associated in uninfected women (Moran CA et al, Abstract 629).
In the ANRS Aquitane cohort, higher ST2 was associated with increased risk of death after adjustment for sCD14 and VACS score, adjusted HR 1.02; P<10-4 (Thiebaut R et al, Abstract 630). ST2 is a member of the interleukin 1 receptor family. The IL-33/ST2 axis is involved in several inflammatory and immune diseases.
Finally, in the Austrian HIV Cohort Study, between 1997-2014, AIDS related, liver related, and non-aids infections declined while CV cause of death remained stable and deaths due to non-AIDS malignancy increased. Low CD4 was not associated with CV events (Leierer G et al, Abstract 631).
In the HIV-Heart cohort in Germany, AIDS, CD4/CD8 ratio, an detectable HIV RNA level was associated with increased CVD risk but not duration of ART. Traditional risk factors remained independently associated with CVD events (Esser S et al, Abstract 633).
Unique imaging studies: The use of novel imaging as a means to explore the pathogenesis of CVD in HIV was highlighted in several studies.
Epicardial fat was associated with both immune activation and arterial inflammation among women with HIV in a study of 55 HIV infected women and 27 controls by Srinivasa S et al, Abstract 627).
Using peripheral arterial tonometry to assess microvascular endothelial function, treated and suppressed HIV infected individuals with untreated HCV had worsened endothelial function (Chew KW et al, Abstract 589).
99mTc-tilmanocept, (a CD206+ binding agent) demonstrated a higher degree of aortic uptake in 6 HIV patients vs. 3 controls. CD206+ macrophage infiltration was also demonstrated using banked aortic tissue in HIV vs. controls (Zanni MV et al, Abstract 635LB). - Systemic monocyte activation and arterial macrophage infiltration are thought to contribute to heightened cardiovascular disease risk in HIV.
In a study by Longenecker C et al, aortic inflammation was found to be higher in HIV, and while there was no difference in aortic inflammation among current smokers (Abstract 628).
Interventional studies: From exercise, statins, and targeted anti-inflammatory approaches, several trials which were reported in HIV which had effects on inflammatory markers and cardiovascular risk.
In the oral sessions of comorbidities, Hsue P et al (Abstract 126) reported on the results of a pilot study of IL-1β inhibition using canakinumab in treated HIV infected individuals with known CVD or at risk for CVD. Overall a single dose of canakinumab did not have any impact on CD4, CD8 count or HIV RNA level or T cell activation. Canakinumab significantly reduced IL-6 by 30 % from baseline (p=0.003), and hsCRP by 41% (p=0.039). Canakinumab also significantly reduced arterial inflammation by 10% (p=0.046) and bone marrow activity by 11% (p<0.001).
The impact of exercise on the endothelium was studied by Dr. Connick and colleagues in Abstract 616. IN a small study of HIV-infected individuals on ART, a home based walking regimen was associated with increased capacity of the endothelium to release t-PA in response to bradykinin which was not blunted by co-administration of Vitamin C. This finding suggests that exercise may have a beneficial impact on endothelial function by reduction in oxidative stress. (Abstract 616).
The SATURN-HIV study evaluated rosuvastatin vs placebo in HIV-infected adults on ART. Plasma 25-hydroxyvitamin D levels did not change after rosuvastatin therapy, however, vitamin D deficiency at baseline appeared to decrease the impact of rosuvastatin on various outcomes, suggesting that Vitamin D supplementation may improve the beneficial impact of statins in a subset of HIV patients (Abstract 617, Hileman CO et al). Using this same trial, rosuvastatin prevented the increase in ADMA over time (p<0.01) and baseline ADMA was associated with increase in IMT. Thus in addition to Vitamin D, arginine supplementation may also enhance the beneficial impact of statins (Abstract 618, Dirajlal-Fargo S et al).
Using data from the START trial, Jason Baker and colleagues (abstract 623) demonstrated that higher IL-6 and D-dimer are strongly associated with serious non-AIDS events and AIDS among HIV+ individuals with higher CD4 counts. This study provides additional evidence that therapies that lower these inflammatory/coagulation markers may have an impact on clinical risk.
Other CV complications of HIV: In addition to atherosclerosis, studies confirmed pulmonary hypertension in HIV and reported on venous thrombotic events as well as heart failure in HIV.
Among individuals who initiated anticoagulation therapy, lower CD4 count was found to be a strong driver of first venous thombotic events (VTE) within the Netherlands (ATHENA) cohort while those with high CD4 T-cell counts had VTE incidence similar to the general population. Withdrawal of anticoagulation therapy was also associated with VTE recurrence in the first year. (Rokx C et al Abstract 620).
Using individuals who were referred for echocardiography at Vanderbilt, 25% of individuals with HIV had RV systolic pressure ≥ 40mm Hg which is consistent with echocardiographic evidence of pulmonary hypertension. Presence of PH by echo was associated with increased mortality in HIV (HR 1.71, 95%CI 1.21-2.40) after adjustment for age, gender, race, heart failure, copd and ART. ART decreased risk of mortality (HR 0.76, 95%CI 0.55 to 1.04, p=0.08, Zola C et al, Abstract 621).
Feinstein M et al (Abstract 622) evaluated the role between BMI and heart failure using electronic health records between 2000-16). In adjusted analysis, HF was more common among underweight and low-normal weight HIV-infected individuals and somewhat less common among obese HIV-infected individuals. This finding suggests that HIV related factors may drive HF among HIV-infected individuals with lower BMI while traditional risk factors may be drive HF among obese HIV-infected individuals.
Global studies of HIV and CVD: A large number of studies done in resource limited setting reported on screening for traditional risk factors.
A study in Swaziland screened over 1800 individuals on ART for CVD risk factors and demonstrated that 37% of individual screened had CVD risk factors and that screening while adding time to the visit was feasible in this patient population (Rabkin M et al, Abstract 637).
In a screening study among 10 rural Ugandan communities, 69% of individuals had controlled HIV and noncommunicable diseases after 2 years. A composite endpoint of both HIV and comorbidities including hypertension and diabetes should be considered as a metric to evaluate progress in the future (Kwariisima D et al, Abstract 638).
The Health and Aging in Africa study (Manne-Goehler J et al, Abstract 639) demonstrated that among HIV-infected individuals who use ART, use of healthcare services for hypertension and diabetes including diagnosis and medication for these conditions were more common.
In a study in Botwana, traditional risk factors and HIV-specific factors including duration of HIV were associated with increased carotid IMT (Mosepele M et al, Abstract 64).
HIV-infected individuals followed after initiation of ART in the REALITY trial in Subsaharan Africa showed that HIV infection is associated with an increase in arterial stiffness as assessed by pulse wave velocity but following ART, this resolved by week 24. (Kelly C et al, Abstract 641).
In summary, a broad and diverse set of studies were presented at the Conference of Retroviruses and Opportunistic Infections. These studies build upon the existing concepts that HIV-infected individuals have high rates of CVD which is now expanding to include heart failure. Excess traditional risk factors are now being reported in HIV-infected individuals in countries with limited health care resources and this global study of CVD in HIV is increasing in importance. Traditional risk factors both in developed and low income countries remains a key contributor to CV risk in HIV; however, controversy remains over CV risk and or benefits associated with different types of ART. Chronic inflammation in the setting of treated HIV remains strongly linked to CVD along with reports of novel markers and novel imaging. As clinicians, we continue to ascertain which risk calculators to use in HIV, which ART regimens are the most cardiac friendly, how aggressively to lower traditional risk factors, and how to lower inflammatory/coagulation markers using therapies with the hopes of reduced clinical events in HIV. We look forward to future studies and progress in this rapidly evolving field and to CROI 2018.
|
| |
|
|
|
|
|
