Association of urine biomarkers of kidney health with subclinical cardiovascular disease among men with and without HIV

Abstract

Objective: The aim of this study was to determine whether urine biomarkers of kidney health are associated with subclinical cardiovascular disease among men with and without HIV.

Design: A cross-sectional study within the Multicenter AIDS Cohort Study (MACS) among 504 men with and without HIV infection who underwent cardiac computed tomography scans and had urine biomarkers measured within the preceding 2 years.

Methods: Our primary predictors were four urine biomarkers of endothelial (albuminuria), proximal tubule dysfunction (alpha-1-microglobulin [A1 M] and injury (kidney injury molecule-1 [KIM-1]) and tubulointerstitial fibrosis (pro-collagen-III N-terminal peptide [PIIINP]). These were evaluated for association with coronary artery calcium (CAC) prevalence, CAC extent, total plaque score, and total segment stenosis using multivariable regression.

Results: Of the 504 participants, 384 were men with HIV (MWH) and 120 were men without HIV. In models adjusted for sociodemographic factors, cardiovascular disease risk factors, eGFR, and HIV-related factors, each two-fold higher concentration of albuminuria was associated with a greater extent of CAC (1.35-fold higher, 95% confidence interval 1.11-1.65), and segment stenosis (1.08-fold greater, 95% confidence interval 1.01-1.16). Associations were similar between MWH and men without HIV in stratified analyses. The third quartile of A1 M showed an association with greater CAC extent, total plaque score, and total segment stenosis, compared with the lowest quartile.

Conclusion: Worse endothelial and proximal tubule dysfunction, as reflected by higher urine albumin and A1 M, were associated with greater CAC extent and coronary artery stenosis.

Figure 1:. Kidney biomarker quartiles and coronary artery calcium prevalence in cross-sectional study of MACS.

Figure 2:. Adjusted associations of kidney biomarker quartiles with coronary artery calcium prevalence and extent in MACS.

A, associations of kidney biomarker quartiles with coronary artery calcium prevalence by Poisson regression in MACS; B, associations of kidney biomarker quartiles with coronary artery calcium extent by Tobit regression in MACS. Note the x-axis scale differs between subfigures. All models are adjusted for study center and enrollment cohort (dichotomized by prior to 2001.), urine creatinine, sociodemographic factors (age, race (categorized into White, Black and Other due to small numbers in the non-white and non-black categories), and Hispanic ethnicity); history of hypertension, history of diabetes, and history of dyslipidemia, SBP, A1c, LDL, HDL, triglycerides, and medications (ACEi/ARB, lipid-lowering, antidiabetic), BMI, and pack-years of tobacco smoking ); HIV serostatus and other comorbidities (HCV), and eGFR (CKD-Epi, race-free).

Figure 3:. Adjusted associations of kidney biomarker quartiles with total plaque score and segment stenosis in MACS.

A, associations of kidney biomarker quartiles with total plaque score, among those with any plaque, by linear regression in MACS; B, associations of kidney biomarker quartiles with total segment stenosis, among those with any stenosis, by linear regression in MACS. All models are adjusted for study center and enrollment cohort (dichotomized by prior to 2001.), urine creatinine, sociodemographic factors (age, race (categorized into White, Black and Other due to small numbers in the non-white and non-black categories), and Hispanic ethnicity); history of hypertension, history of diabetes, and history of dyslipidemia, SBP, A1c, LDL, HDL, triglycerides, and medications (ACEi/ARB, lipid-lowering, antidiabetic), BMI, and pack-years of tobacco smoking ); HIV serostatus and other comorbidities (HCV), and eGFR (CKD-Epi, race-free).