Preoperative haemostasis testing does not predict requirement of blood products in cardiac surgery.

OBJECTIVE

Standard haemostasis screening tests are performed to reveal unknown congenital or acquired disturbances of plasma and/or platelet haemostasis. Since their diagnostic efficacy is often low, routinely performed haemostasis testing has been questioned. We investigated whether preoperatively assessed haemostasis testing can be used to predict the requirement of blood products.

METHODS

We retrospectively assessed haemostasis parameters including platelet function testing by PFA 100 as well as the numbers of red blood cell (RBC) concentrates, fresh frozen plasmas (FFPs), and platelet concentrates (PCs) that were given peri-operatively and during the first two postoperative days in 2,831 cardiac surgery patients. Logistic regression analyses were used to select those parameters, which could predict blood product requirement.

RESULTS

Of our study cohort, 56.5% needed RBCs, 15% FFPs, and 5% PCs. The need for RBCs was associated with significantly altered pre-operative values of most haemostasis parameters. However, by the use of logistic regression analysis fibrinogen was the only haemostasis parameter that was independently associated with the use of RBCs (odds ratio 1.56; 95% CI: 1.27-1.91; P <0.001). The predictive value of other parameters such as age, body weight, haemoglobin, and haematocrit was however much higher in comparison to fibrinogen (odds ratios: 1.92-3.50; P <0.001). It was not possible to develop a score based on haemostasis parameters to accurately identify patients at risk for RBC use. Moreover, we were unable to estimate the need for FFPs and PCs using preoperative haemostasis testing.

CONCLUSIONS

Our data demonstrate that preoperatively performed haemostasis testing is not predictive in estimating the need for blood products in cardiac surgery patients.