| |
Pericardial Adipose Tissue Volume Is Independently Associated With Human Immunodeficiency Virus Status and Prior Use of Stavudine, Didanosine, or Indinavir
|
| |
| |
Download the PDF here
The Journal of Infectious Diseases, Volume 222, Issue 1, 1 July 2020 - Andreas D Knudsen, Andreas Fuchs, Anne-Mette Lebech, Børge G Nordestgaard, Jens Lundgren, Jørgen T Kühl, Klaus F Kofoed, Lars Køber, Lisanne Krebs-Demmer, Marco Gelpi, Marie B Elming, Natascha I D Bjørge, Per E Sigvardsen, Susanne D Nielsen
Pericardial adipose tissue, delineated by the pericardial sac, is a local fat depot that surrounds the heart and in which the coronary arteries are imbedded. Pericardial adipose tissue, like visceral adipose tissue (VAT), is considered a risk factor for major adverse cardiovascular events in the general population and in PWH ......The association between AZT, d4T, and ddI and pericardial adipose tissue was larger for d4T and ddI, and the estimates for these drugs were less affected by adjusting for IDV, compared with AZT, which suggest that d4T and ddI may be more toxic to adipocytes than AZT.....The association between AZT and pericardial adipose tissue may be entirely explained by prior IDV exposure among those persons who have used AZT..... the association between severe immunodeficiency and larger pericardial adipose tissue volume, and that longer time using cART seems to be beneficial among those not exposed to first-generation thymidine analogues, indicates that early cART initiation should be enforced as suggested by other studies [34, 35] and recommended by current treatment guidelines.
Background: Increased pericardial adipose tissue is associated with higher risk of cardiovascular disease. We aimed to determine whether human immunodeficiency virus (HIV) status was independently associated with larger pericardial adipose tissue volume and to explore possible HIV-specific risk factors.
Methods: Persons with HIV (PWH) were recruited from the Copenhagen Comorbidity in HIV Infection (COCOMO) Study and matched 1:1 on age and sex to uninfected controls. Pericardial adipose tissue volume was measured using cardiac computed tomography.
Results:A total of 587 PWH and 587 controls were included. Median age was 52 years, and 88% were male. Human immunodeficiency virus status was independently associated with 17 mL (95% confidence interval [CI], 10-23; P < .001) larger pericardial adipose tissue volume. Larger pericardial adipose tissue volume was associated with low CD4+ nadir and prior use of stavudine, didanosine, and indinavir. Among PWH without thymidine analogue or didanosine exposure, time since initiating combination antiretroviral treatment (per 5-year use) was associated with l6 mL (95% CI, -6 to -25; P = .002) lower pericardial adipose tissue volume.
Conclusions: Human immunodeficiency virus status was independently associated with larger pericardial adipose tissue volume. Severe immunodeficiency, stavudine, didanosine, and indinavir were associated with larger pericardial adipose tissue volume. Persons with HIV with prior exposure to these drugs may constitute a distinct cardiovascular risk population.
Persons with human immunodeficiency virus (PWH) have increased risk of atherosclerotic cardiovascular disease (CVD) compared with the uninfected population [1-3]. The etiology of atherosclerosis in human immunodeficiency virus (HIV) infection likely involves an intricate interplay between traditional risk factors (eg, dyslipidemia and smoking) and HIV-specific factors such as combination antiretroviral treatment (cART) and chronic inflammation [4]. Pericardial adipose tissue, delineated by the pericardial sac, is a local fat depot that surrounds the heart and in which the coronary arteries are imbedded. Pericardial adipose tissue, like visceral adipose tissue (VAT), is considered a risk factor for major adverse cardiovascular events in the general population and in PWH [5]. In particular, pericardial adipose tissue may directly contribute to progression of coronary atherosclerosis [6-9]. In the context of increased availability of cardiac computed tomography (CT), pericardial adipose tissue volume has been suggested as a marker of both local inflammation and overall cardiovascular risk [10]. Two smaller studies have suggested that PWH have a higher pericardial adipose tissue volume than uninfected controls [7, 11], and within PWH, pericardial adipose tissue has been associated with myocardial ischemia [11].
Human immunodeficiency virus status was independently associated with larger pericardial adipose tissue volume. Among PWH, severe immunodeficiency and previous treatment with d4T, ddI, and IDV were associated with larger pericardial adipose tissue volume. Among PWH without exposure to thymidine analogues, time since initiating cART was associated with lower pericardial adipose tissue volume. When gauging cardiovascular risk, PWH with prior exposure to d4T and ddI may constitute a distinct risk population.
Epicardial adipose tissue is the part of pericardial adipose tissue that lies between the visceral pericardium and the myocardium;
Male sex, BMI, hypertension, diabetes, smoking status, dyslipidemia, and sedentary lifestyle were all associated with larger pericardial adipose tissue volume in unadjusted analyses (all P < .05) (Table 2). In analysis adjusting for BMI, age, and sex (model 1), HIV status was independently associated with 15 mL larger pericardial adipose tissue volume (95% confidence interval [CI], 8-21; P < .001). After further adjusting for hypertension, diabetes, dyslipidemia, and smoking (model 2), HIV remained associated with 16 (95% CI, 10-23) mL larger pericardial adipose tissue volume (P < .001). Additional adjustment for physical activity (model 3) did not change this estimate (17 mL; 95% CI, 10-23). In addition to HIV, age, male sex, BMI, sedentary lifestyle, dyslipidemia, and smoking (current and former) were also independently associated with larger pericardial adipose tissue volume (Table 2). Pericardial adipose tissue volume was not independently associated with hypertension or diabetes.
Our findings suggest that pericardial adipose tissue volume accumulation is a distinct feature in PWH, even in a predominantly virologically suppressed population. Epicardial adipose tissue is the part of pericardial adipose tissue that lies between the visceral pericardium and the myocardium; however, although the vast majority of pericardial adipose tissue, as defined by Framingham, consists of epicardial adipose tissue, the 2 compartments are not fully identical. Epicardial adipose tissue volume has previously been reported to be higher in PWH compared with uninfected controls [21, 22] and has been associated with duration of cART use, systemic inflammatory biomarkers, and cardiovascular events in PWH [21-23]. However, studies have been conducted in distinct populations (eg, homosexual men), and findings are not consistent in all populations of PWH [24].
We found that nadir CD4+ T-cell count below 200 cells/μL, indicating prior or current severe immunodeficiency, was independently associated with larger pericardial adipose tissue volume in PWH. Low nadir CD4+ nadir is known to be associated with increased levels of chronic immune activation and inflammation [28]. ......The effect of time since initiating cART was opposite among those with thymidine analogue/ddI exposure, highlighting that although contemporary cART regimes have improved in terms of cardiovascular safety, the adverse effects of older antiretroviral drugs may still be manifest today among otherwise well treated PWH [31, 32].....The association between AZT, d4T, and ddI and pericardial adipose tissue was larger for d4T and ddI, and the estimates for these drugs were less affected by adjusting for IDV, compared with AZT, which suggest that d4T and ddI may be more toxic to adipocytes than AZT.....The association between AZT and pericardial adipose tissue may be entirely explained by prior IDV exposure among those persons who have used AZT. However, the association between HIV status and pericardial adipose tissue volume was not fully explained by thymidine analogue/ddI exposure, and PWH without exposure had larger pericardial adipose tissue volume than uninfected controls, when adjusting for BMI. Taken together, the association between severe immunodeficiency and larger pericardial adipose tissue volume, and that longer time using cART seems to be beneficial among those not exposed to first-generation thymidine analogues, indicates that early cART initiation should be enforced as suggested by other studies [34, 35] and recommended by current treatment guidelines [36, 37].
Human immunodeficiency virus was independently associated with larger pericardial adipose tissue volume. In PWH, prior or current severe immunodeficiency (low nadir CD4+ T cell <200 cells/μL) was associated with larger pericardial adipose tissue volume. It is interesting to note that thymidine analogue/ddI exposure strongly modified the association between time since initiating cART and pericardial adipose tissue volume with larger volume per 5-year use among exposed and lower volume in the unexposed. However, even in PWH without exposure to thymidine analogue/ddI, the pericardial adipose tissue volume was larger than in controls. Larger pericardial adipose tissue volume is a predictor of incident CVD independent of inflammation, coronary artery calcification, and cardiometabolic risk factors in uninfected individuals [6, 9, 20].
Regardless of HIV status and in accordance with previous literature [25, 26], we found that age and male sex were strongly associated with increased pericardial adipose tissue volume. As expected, modifiable risk factors, such as smoking, physical inactivity, and dyslipidemia, were also independently associated with larger pericardial adipose tissue volume. In those with normal BMI, pericardial adipose tissue volume was 22 mL larger in PWH compared with controls, and HIV status was only associated with pericardial adipose tissue volume after adjusting for BMI.
In adjusted analyses of PWH only, use of AZT, d4T, and/or ddI (considered together or separately) was associated with larger pericardial adipose tissue (Table 4). Prior exposure to IDV was associated with 19.6 (95% CI, 8.6- 30.6) mL larger pericardial adipose tissue volume, and low nadir CD4+ T-cell count (<200 cells/μL) was associated with 10 (95% CI, 1-19) mL larger pericardial adipose tissue volume
In adjusted analysis comparing PWH without thymidine analogue/ddI exposure to controls, HIV status was borderline associated with 9 (95% CI, 0.1-18) mL larger pericardial adipose tissue volume (P = .048), whereas PWH with thymidine analogue/ddI exposure had a 22 (95% CI, 14-30) mL larger pericardial adipose tissue volume (P < .001).
Human Immunodeficiency Virus-Related Factors Associated With Pericardial Adipose Tissue Volume
Pericardial adipose tissue volume was 21 (95% CI, -31 to -11) mL larger in PWH with thymidine analogue/ddI exposure compared with PWH without exposure (P < .001) (Figure 1B), whereas BMI was similar (P = .663). In adjusted analyses of PWH only, use of AZT, d4T, and/or ddI (considered together or separately) was associated with larger pericardial adipose tissue (Table 4). Prior exposure to IDV was associated with 19.6 (95% CI, 8.6- 30.6) mL larger pericardial adipose tissue volume, and low nadir CD4+ T-cell count (<200 cells/μL) was associated with 10 (95% CI, 1-19) mL larger pericardial adipose tissue volume (P = .031), but there was no association between current CD4+ count and pericardial adipose tissue volume (P = .381). Longer duration of exposure to AZT, d4T, ddI, and IDV was not associated with larger pericardial adipose tissue volume (Table 4). Further adjustment for IDV attenuated the association between pericardial adipose tissue volume and thymidine analogues and/or ddI and between pericardial adipose tissue and AZT, ddI, and d4T considered separately (Table 4). In a sensitivity analysis with waist-to-hip ratio included in the model, d4T was associated with 10.8 (95% CI, -.9 to 22.6) mL and ddI with 8.5 (95% CI, -3.1 to 20.1) mL larger pericardial adipose tissue volume, whereas AZT was no longer associated with pericardial adipose tissue volume (P = .454). Time with HIV infection and time since initiating cART treatment were not independently associated with pericardial adipose tissue volume (P = .062 and P = .092, respectively). However, exposure to thymidine analogues and/or ddI modified the association between time since initiating cART treatment and pericardial adipose tissue volume (pinteraction = .002). Thus, among PWH with exposure to thymidine analogues and/or ddI, time since initiating cART treatment (per 5-year use) was borderline associated with 8 (95% CI, -0.2 to 15) mL (P = .058) larger pericardial adipose tissue volume, whereas time since initiating cART treatment was associated with l6 (95% CI, -6 to -25) mL lower pericardial adipose tissue volume among those without exposure (P = .002). In adjusted analysis comparing PWH without thymidine analogue/ddI exposure to controls, HIV status was borderline associated with 9 (95% CI, 0.1-18) mL larger pericardial adipose tissue volume (P = .048), whereas PWH with thymidine analogue/ddI exposure had a 22 (95% CI, 14-30) mL larger pericardial adipose tissue volume (P < .001).
|
|
| |
| |
|
|
|
