Novel Blood Biomarkers for a Diagnostic Workup of Acute Aortic Dissection.
Anja ForrerFelix SchoenrathMichael TorzewskiJens SchmidUrlich F W FrankeNora GöbelDrahomir AujeskyChristian M MatterThomas F LüscherFrancois MachDavid NanchenNicolas RodondiVolkmar FalkArnold Von EckardsteinJoanna GawineckaPublished in: Diagnostics (Basel, Switzerland) (2021)
Acute aortic dissection (AAD) is a rare condition, but together with acute myocardial infarction (AMI) and pulmonary embolism (PE) it belongs to the most relevant and life-threatening causes of acute chest pain. Until now, there has been no specific blood test in the diagnostic workup of AAD. To identify clinically relevant biomarkers for AAD, we applied Proseek® Multiplex assays to plasma samples from patients with AAD, AMI, PE, thoracic aortic aneurysm (TAA), and non-cardiovascular chest pain (nonCVD). Subsequently, we validated top hits using conventional immunoassays and examined their expression in the aortic tissue. Interleukin 10 (IL-10) alone showed the best performance with a sensitivity of 55% and a specificity of 98% for AAD diagnosis. The combination of D-dimers, high-sensitive troponin T (hs-TnT), interleukin 6 (IL-6), and plasminogen activator inhibitor 1 (PAI1) correctly classified 75% of AAD cases, delivering a sensitivity of 83% and specificity of 95% for its diagnosis. Moreover, this model provided the correct classification of 77% of all analyzed cases. Our data suggest that IL-10 shows potential to be a rule-in biomarker for AAD. Moreover, the addition of PAI1 and IL-6 to hs-TnT and D-dimers may improve the discrimination of suspected AAD, AMI, and PE in patients presenting with acute chest pain.
Keyphrases
- aortic dissection
- acute myocardial infarction
- pulmonary embolism
- liver failure
- percutaneous coronary intervention
- machine learning
- inferior vena cava
- heart failure
- high throughput
- poor prognosis
- aortic aneurysm
- left ventricular
- spinal cord injury
- spinal cord
- drug induced
- coronary artery disease
- intensive care unit
- big data
- risk assessment
- binding protein
- structural basis