Relationship between perioperative serum albumin levels and paravalvular leak after transcatheter aortic valve replacement

BACKGROUND: Paravalvular leak (PVL) remains a significant complication after transcatheter aortic valve replacement (TAVR). This study investigated the relationship between perioperative serum albumin levels and post-TAVR paravalvular leak. METHODS: This retrospective observational study analyzed 1463 consecutive patients who underwent TAVR between 2013 and 2023. Serum albumin levels were measured preoperatively, immediately after procedure, and on postoperative days 1–5 and before discharge. Additional biomarkers including NT-proBNP and high-sensitivity troponin T were collected. Postprocedural echocardiography was used to assess PVL severity according to American Society of Echocardiography and European Association of Cardiovascular Imaging guidelines. Primary endpoints included correlation between various albumin parameters and PVL severity. RESULTS: The cohort included 851 males (58.2%) and 612 females (41.8%) with a mean age of 72.5 ± 7.5 years. Serum albumin showed a characteristic decline pattern, reaching nadir immediately postoperatively (35.1 ± 4.3 g/L) and gradually recovering to 38.1 ± 4.1 g/L by discharge. Among patients with evaluable PVL (n = 717), 539 (75.2%) had no or trace PVL, 91 (12.7%) had mild PVL, and 87 (12.1%) had moderate or severe PVL. Patients with moderate or severe PVL showed greater maximum albumin decline (19.8 ± 11.2%) compared to those without significant PVL (12.4 ± 10.6%, p < 0.01). Maximum albumin decline showed the strongest correlation with PVL severity (r = 0.142, p < 0.001). ROC analysis revealed maximum albumin decline had the highest predictive value for moderate or severe PVL (AUC = 0.604), with optimal cutoff at 18.9% (sensitivity 68.6%, specificity 48.2%). Furthermore, patients with both elevated maximum albumin decline and moderate or severe PVL demonstrated significantly higher one-year mortality compared to those with preserved albumin dynamics and no significant PVL. CONCLUSIONS: Perioperative albumin dynamics, particularly maximum albumin decline, are associated with paravalvular leak after TAVR. These findings suggest albumin level changes may serve as a simple biomarker to identify patients at higher risk for significant PVL, with potential implications for clinical outcomes.