Circulating biomarkers improve prediction of postoperative outcome after aortic valve surgery.

BACKGROUND

Predicting adverse outcomes following aortic valve surgery remains challenging. This study aimed to investigate the association between preoperative circulating biomarkers and postoperative mortality and morbidity, with the goal of improving established risk stratification tools.

METHODS

Between March 2018 and May 2022, 492 patients underwent surgical aortic valve replacement or repair and were included in a registry. Preoperative blood biomarkers, including hemoglobin, creatinine, high-sensitivity troponin I (hsTrop-I), GOT, GPT, INR, CRP, NT-proBNP and WBC, were sampled at baseline. Logistic regression analysis adjusted for EuroSCORE-II tested associations between biomarker levels and VARC-III adjudicated endpoints. Model fit was assessed using Akaike's Information Criterion, and likelihood ratio tests compared different prediction models.

RESULTS

Preoperative hemoglobin (OR 0.70; 95% CI: 0.57, 0.87; p < 0.001), creatinine (OR 4.09; 95% CI: 1.63, 10.26; p = 0.003), high-sensitivity troponin I (OR 1.48; 95% CI: 1.11, 1.95; p = 0.007), GOT (OR 2.49; 95% CI: 1.06, 5.86; p = 0.036), INR (OR 4.74; 95% CI: 1.32, 17.05; p = 0.017), CRP (OR 2.14; 95% CI: 1.53, 3.00; p < 0.001), NT-proBNP (OR 2.06; 95% CI: 1.41, 3.00; p < 0.001), and WBC (OR 5.82; 95% CI: 1.88, 17.97; p = 0.002) were independently associated with 30-day mortality after adjustment for EuroSCORE-II. Models combining biomarkers with EuroSCORE-II outperformed those predicting mortality by EuroSCORE-II or biomarkers alone, as indicated by the lowest Akaike's Information Criterion and likelihood ratio tests.

CONCLUSIONS

Combining established risk stratification models with preoperative biomarkers was associated with improved predictive performance for adverse outcomes after aortic valve surgery and may support heart team decision-making when choosing between surgical and transcatheter aortic valve replacement.