icon-    folder.gif   Conference Reports for NATAP  
  17th CROI
Conference on Retroviruses
and Opportunistic Infections
San Francisco CA
February 16-19, 2010
Back grey_arrow_rt.gif
Higher Levels of Heart Cell Peptide
Tied to Cardiovascular Disease in SMART Study Patients
  17th Conference on Retroviruses and Opportunistic Infections, February 16-19, 2010, San Francisco
Mark Mascolini
- The SMART team proposed that the higher heart disease risk in some study participants "is potentially due to upregulation of the expression of NT-proBNP in the setting of HIV-induced subclinical vascular disease and increased myocardial strain."

Higher concentrations of NT-proBNP, a polypeptide secreted by ventricular myocytes, signaled a heightened risk of cardiovascular disease in HIV-infected SMART study participants [1]. Statistical adjustment for traditional risk factors and for inflammation markers weakened but did not abolish this association.
Cell studies indicate that heart cells release natriuretic peptides in response to myocardial ischemia, according to Daniel Duprez (University of Minnesota) and SMART Study Group colleagues. Ventricular myocytes secrete B-type natriuretic peptide when ventricular stretch and wall tension increase. That peptide plays a part in regulating blood pressure, blood volume, and sodium balance. When secreted, the peptide precursor splits into the biologically active peptide and a more stable N-terminal fragment (NT-proBNP), which is the marker Duprez analyzed in 186 SMART participants who had a "cardiovascular event" during follow-up and 329 matched SMART participants who did not.
The 186 new cardiovascular diagnoses consisted of 95 cases of coronary heart disease (CHD), 47 of atherosclerotic non-CHD, 17 of congestive heart failure, and 27 of cardiovascular disease or unwitnessed death.
Cases (with a new cardiovascular diagnosis) and controls (without new cardiovascular disease) matched in age (median 50 in cases and 49 in controls), gender proportion (18.8% and 20.4% female), race (39.2% and 37.7% black), current CD4 count (medians of 579 and 606), lowest-ever CD4 counts (medians of 236 and 244), and percent with a viral load below 400 copies (68.3% and 70.9%),
A higher proportion of cases than controls had an AIDS diagnosis (36.6% versus 24.9%, P = 0.007), were current smokers (49.4% versus 36.8%, P = 0.008), took antihypertensives (43.5% versus 28.9%, P = 0.0009), and took antilipid drugs (29.6% versus 21.6%, P = 0.05). A higher proportion of cases had diabetes (18.3% versus 7.3%, P = 0.0004), a major ECG abnormality (19.8% versus 9.3%, P = 0.002), or HBV infection (2.7% versus 0.6%, P = 0.07). "Good" high-density lipoprotein (HDL) cholesterol was significantly lower in cases than controls (median 38 versus 41 mg/dL, P = 0.04), and total/HDL cholesterol ratio was significantly higher (worse) in cases (5.6 versus 4.7, P = 0.03).
NT-proBNP quotients were higher among cases than controls (median [IQR] 48.1 pg/mL [18.5 to 112.9] versus 25.7 pg/mL [12.4 to 50.2]), a highly significant difference (P < 0.0001). Levels of the coagulation marker D-dimer (P = 0.002) and the inflammation markers high-sensitivity C-reactive protein (hsCRP) (P < 0.0001) and IL-6 (P < 0.0001) were also significantly higher in cases than controls.
Duprez and coworkers divided NT-proBNP levels into four even clusters (quartiles) for the whole group: under 14.1 pg/mL, 14.1 to 29.6 pg/mL, 29.7 to 62.7 pg/mL, and 62.8 pg/mL or more. Each higher quartile had a higher proportion of cases (17.2%, 22.6%, 23.1%, 37.1%) and a lower proportion of controls (29.2%, 26.4%, 26.1%, 18.2%).
To reckon the impact of NT-proBNP concentration on risk of a new cardiovascular diagnosis, the investigators designed two multivariate models. Model 1 factored in baseline age, body mass index, race, CD4 count, viral load, antiretroviral status, smoking, prior cardiovascular disease, diabetes, total/HDL cholesterol ratio, use of antihypertensives, hepatitis coinfection, and major baseline ECG abnormalities. Model 2 considered all of these factors plus levels of D-dimer, IL-6, and hsCRP.
In both multivariate models, having an NT-proBNP concentration in the highest quartile independently raised the risk of cardiovascular disease compared with having a concentration in the lowest quartile. In both models every 10-fold higher NT-proBNP level boosted the risk of heart disease, though this association stopped short of statistical significance in model 2:
Model 1
· Quartile 4 versus quartile 1: odds ratio [OR] 2.8, P = 0.003
· 10-fold higher NT-proBNP: OR 1.4, P = 0.002
Model 2
· Quartile 4 versus quartile 1: OR 2.3, P = 0.02
· 10-fold higher NT-proBNP: OR 1.2, P = 0.07
Associations between NT-proBNP levels and cardiovascular disease did not differ significantly between SMART treatment arms.
The findings replicate results of a recent meta-analysis of NT-proBNP's impact on cardiovascular risk in the general population [2]. Comparing the highest one third (tertile) of NT-proBNP levels with the lowest one third, the meta-analysis figured a 2.68 times higher risk of heart disease in the highest tertile after statistical adjustment for conventional risk factors. When Duprez split SMART's NT-proBNP levels into tertiles, he found a 2.63 times higher risk of cardiovascular disease with model 1 and a 2.30 times higher risk with model 2. All of these associations were statistically significant.
The SMART team proposed that the higher heart disease risk in some study participants "is potentially due to upregulation of the expression of NT-proBNP in the setting of HIV-induced subclinical vascular disease and increased myocardial strain."
1. Duprez D, INSIGHT SMART Study Group. N-terminal-proB-type natriuretic peptide predicts cardiovascular disease events in HIV-infected patients: results of the SMART study. 17th Conference on Retroviruses and Opportunistic Infections. February 16-19, 2010. San Francisco. Abstract 712. http://www.retroconference.org/2010/PDFs/712.pdf.
2. Di Angelantonio E, Chowdhury R, Sarwar N, et al. B-type natriuretic peptides and cardiovascular risk: systematic review and meta-analysis of 40 prospective studies. Circulation. 2009;120:2177-2187.