BACKGROUND: 

We examined the relationship between blood transfusion and long-term adverse events to evaluate the clinical impact of red blood cell (RBC) transfusion on patients undergoing cardiac valve surgery.

METHODS: 

From the National Health Insurance Service database, individuals undergoing heart valve surgery were verified, including aortic valve (AV), mitral valve (MV), tricuspid valve (TV), and complex valves (more than 2 valve surgeries). The interested outcomes were incidence of death, ischemic stroke, hemorrhagic stroke, and admission for myocardial infarction during follow-up. Associations between perioperative RBC transfusion and long-term cardiovascular events were analyzed with Cox-proportional hazard model.

RESULTS: 

Perioperative RBC transfusion (±2 days from the day of surgery) was categorized into 0, 1, 2, and >3 units based on the number of packs transfused. From 2003 to 2019, the data of 58,299 individuals were retrieved (51.6% were male and 58% were aged above 60 years). The median follow-up duration was 5.53 years. Of the total cohort, 86.5% received at least 1 transfusion. In multivariable analysis, adverse cardiovascular event risk proportionally increased with transfusion in a dose-dependent manner. The adjusted hazard ratios and 95% confidence intervals of outcomes after the transfusion of 1, 2, and ≥3 units compared to those with no transfusion were as follows: death, 1.53 (1.41–1.66), 1.97 (1.81–2.14), and 3.03 (2.79–3.29); ischemic stroke, 1.27 (1.16–1.39), 1.31 (1.19–1.44), and 1.51 (1.38–1.66); hemorrhagic stroke, 1.38 (1.16–1.66), 1.71 (1.43–2.05), and 2.31 (1.94–2.76); and myocardial infarction 1.35 (1.13–1.62), 1.60 (1.33–1.91), and 1.99 (1.66–2.38), respectively (all P < .01).

CONCLUSIONS: 

In the analysis of the national cohort, perioperative RBC transfusion during heart valve surgery was associated with adverse cardiovascular outcomes correlated with the volume of RBC transfusion.