Association Between Serum MicroRNAs and Abdominal Aortic Aneurysm Diagnosis and Growth

Abstract

This study aimed to examine the association between serum microRNAs (miRNAs) and diagnosis and growth of abdominal aortic aneurysm (AAA), and to test their diagnostic and prognostic value. The expression levels of 800 miRNA tags were assessed in 108 patients with AAA, 12 age and sex matched healthy controls (HCs), and 12 patients with peripheral artery disease (PAD) using NanoString technology. Findings were assessed in an independent sample of 66 patients with AAA and 29 age and sex matched HCs by reverse transcriptase polymerase chain reaction. AAA growth was assessed by a median of three (interquartile range [IQR] 2, 3) repeat ultrasound scans over a median follow up of 1.1 (IQR 1.0, 2.0) years. The association between the miRNA and AAA diagnosis and growth was examined by regression and linear mixed effects analyses. The diagnostic and prognostic potential of the miRNAs were examined using area under the receiver operator characteristic curve (AUC), net re-classification index (NRI), and Cox hazard analyses. In comparison with HCs, a model combining clinical risk factors, let-7b-5p and miR-548n had an AUC of 98.0% (95% confidence interval [CI] 95.6 - 100.0; p= .003) for diagnosing AAA, which was a significant improvement over clinical risk factors alone (NRI 1.74; 95% CI 1.61 - 1.87; p < .001). Compared with PAD, a model combining clinical risk factors and miR-548n had an AUC of 99.6% (95% CI 98.9 - 100.0, p= .037) for diagnosing AAA, which was a significant improvement over clinical risk factors alone (NRI 1.79, 95% CI 1.68 - 1.91; p < .001). In the longitudinal cohort, none of the miRNAs were able to predict the likelihood of reaching surgical threshold diameter better than clinical risk factors alone. Serum let-7b-5p and miR548n significantly improved the ability to diagnose AAA. None of the miRNAs had independent prognosis value in predicting AAA growth.