Objective: To investigate the association between total, direct, and indirect bilirubin and the presence of carotid lesions in a large sample of HIV-1-infected patients on virological suppression.
Design: Retrospective study on adult HIV-1-infected patients, with a carotid ultrasound (CUS) examination performed between January 2008 and August 2016, with HIV-RNA <50 copies per milliliter at CUS and without previous cardiovascular events.
Methods: Intima media thickness was measured in 4 segments: carotid common artery and bifurcation on the left and right sides. Carotid lesion was defined as an intima media thickness ≥1.5 mm in ≥1 region at CUS. Patients were classified as: normal if all bilirubin values before CUS were below the upper normal limit and with hyperbilirubinemia if ≥1 bilirubin value above upper normal limit before CUS was recorded. Multivariate logistic regression was used to determine whether hyperbilirubinemia showed association with the presence of ≥1 carotid lesion, after adjusting for confounding factors.
Results: Overall, 903 patients were evaluated, 511 with ≥1 and 392 without carotid lesions. At multivariate analysis, total [adjusted odds ratio (95% confidence interval) 0.57 (0.36 to 0.90), P = 0.016] and indirect hyperbilirubinemia before CUS [adjusted odds ratio (95% confidence interval) 0.62 (0.40 to 0.97), P = 0.036] were associated with a lower risk of carotid lesions in addition to younger age, negative hepatitis C virus antibodies, higher nadir CD4, lower low-density lipoprotein cholesterol, higher high-density lipoprotein cholesterol, lower triglycerides, and no use of statin; no effect of atazanavir treatment on carotid lesions was detected.
Conclusions: In HIV-1-treated patients, total or indirect hyperbilirubinemia was likely associated with the absence of carotid lesions.