Biomarkers and risk scores predicting mortality and acute kidney injury after cardiac surgery

Liisa Petäjä

Research output: ThesisDoctoral ThesisCollection of Articles

Abstract

Annually world-wide, 17.7 million people die of cardiovascular diseases. Cardiac surgery is performed to improve patients’ prognosis and alleviate their symptoms. However, these challenging procedures have inherent risks, including cardiac and renal injury. The objective of this study was to investigate, by biomarkers and risk models, cardiac surgery-related cardiac and renal morbidity, and their association with prognosis. A systematic search, review and meta-analysis were performed to find out the relationship of postoperative injury markers after coronary artery bypass grafting (CABG), and mortality. Additionally, two cohorts, 428 CABG and 220 other cardiac surgical patients, were enrolled to a prospective study. The associations of pre- and postoperative high-sensitivity troponin (hs-TnT) with postoperative major adverse events (MAE) and mortality were explored. Whether these biomarkers offer additive information for an established risk model, EuroSCORE II, was detected. Acute kidney injury (AKI) was defined by complete Kidney Disease: Improving Global Outcomes (KDIGO) criteria including serum creatinine and urine output (UO). The associations of AKI stages and the separate criteria with long-term mortality was assessed in the two cohorts as one. Finally, a recently published risk model predicting cardiac surgery-associated AKI (CSA-AKI), named AKI Risk Score, was validated in this cohort. The potential benefit of adding preoperative biomarkers, hs-TnT and N-terminal pro-B-type natriuretic peptide (NT-proBNP), to AKI Risk Score, was evaluated. Postoperative CK-MB releases after cardiac surgery were related to increased mortality until 3 years. Even though troponins seemed better predictors, the assays lack standardization, and this remains a problem in generalizing the results. Elevated preoperative hs-TnT determined the meaning of first postoperative day hs-TnT in all cardiac surgery. It was related to long-term mortality regardless of postoperative hs-TnT in cardiac surgery other than plain CABG. Adding hs-TnT to EuroSCORE II yielded to better prediction of MAE in all cardiac surgery, and to better prediction of half-a-year mortality in at least cardiac surgery other than plain CABG. AKIs, diagnosed by separate criteria of KDIGO, were all associated with 2.5-year mortality, even AKI diagnosed by only urine output. AKI by KDIGO stages was associated with increased mortality, related to severity. The excess mortality associated with AKI was not predicted by EuroSCORE-based risk of mortality or risk factors for AKI. AKI Risk Score performed acceptably in Finnish cardiac surgical patients. Adding preoperative biomarkers to the score yielded to no clear clinical benefit. A thorough analyses of the added value of the biomarkers, by risk assessment plot and estimation of net benefit, is warranted to reveal true potential of the biomarkers.
Original languageEnglish
Place of PublicationHelsinki
Publisher
Print ISBNs978-951-51-5120-9
Electronic ISBNs978-951-51-5121-6
Publication statusPublished - 2019
MoE publication typeG5 Doctoral dissertation (article)

Bibliographical note

M1 - 166 s. + liitteet

Fields of Science

  • Cardiac Surgical Procedures
  • +adverse effects
  • +mortality
  • Acute Kidney Injury
  • Heart
  • +physiopathology
  • Heart Injuries
  • Creatinine
  • Creatine Kinase, MB Form
  • Biomarkers
  • Oliguria
  • Troponin
  • Risk Assessment
  • Heart Failure
  • Myocardium
  • Postoperative Complications
  • Organ Dysfunction Scores
  • Mortality
  • 3126 Surgery, anesthesiology, intensive care, radiology

Cite this