Bleeding risk-assessment in elective cardiac surgery - a reply.

Abstract

We appreciate Lancé's interest in our paper on bleeding risk-assessment in patients undergoing elective cardiac surgery 1. Our prospective, observational study was designed to compare both impedance aggregometry methods and their predictive value for bleeding and transfusion at different time-points during elective cardiac surgery, but not to assess the influence of platelet count and haemoglobin concentration on these methods since the dependency of impedance aggregometry results on platelet count is already well known and described in the literature 2-4. Accordingly, we presented only the pre-operative platelet count in Table 1, which showed exactly the same median value of 196 × 109.l−1 in the non-excessive and excessive bleeding group. Unfortunately, we cannot provide whole blood counts at each time-point of platelet function analysis. We used a resting time of 30 min after blood sampling before starting the impedance aggregometry assays in order to perform the analyses in a standardised way. A resting time of 30 minutes is, in fact, unnecessary when platelet aggregation is measured by impedance aggregometry in blood samples anticoagulated with lithium heparin or r-hirudin and results are needed urgently for clinical decision-making 5-7. In our institution, balanced electrolyte solutions are used as first-line for volume replacement. Colloids are used with caution as a second-line option due to their negative impact on fibrin polymerisation and platelet aggregation. Anticoagulation for cardiopulmonary bypass was provided by administering 400 U.kg−1 sodium heparin in addition to priming the bypass circuit with 5000 U. Additional boluses of 50 U.kg−1 heparin were given if needed to achieve a targeted activated clotting time (ACT) > 480 s. If 600 U.kg−1 failed to achieve an ACT > 480 s, 500 U of anti-thrombin concentrate were administered. After weaning from CPB, heparin reversal was achieved with protamine hydrochloride in a 1:1-ratio (10 mg of protamine hydrochloride corresponding to 1000 heparin antidote U for every 1000 U of sodium heparin given initially). We are aware of the effect of protamine on platelet function 8, and so we analysed platelet function not only pre-operatively 9-11 but at the three different time-points as described. This allowed us to assess not only pre-existing platelet dysfunction (for example, due to pre-operative antiplatelet therapy 9-11), but also to assess platelet dysfunction induced during CPB 4, 12, and induced by protamine 8. Accordingly, impedance aggregometry results obtained after protamine administration demonstrated the best correlation with postoperative chest tube drainage in our study 1. We performed no power analysis for our observational study which would be reasonable for sample size calculation in interventional studies. However, the number of blood samples taken for platelet function analysis (303 different samples) is higher than in most other studies published before 8-13. As presented in Tables 1 and 3, there was no significant difference in respect to the pattern of pre-operative antiplatelet therapy between the non-excessive bleeding and excessive bleeding group. Accordingly, the status of antiplatelet therapy was not predictive for bleeding in our study. In contrast, impedance aggregometry results from ROTEM® platelet and Multiplate® were predictive for chest tube drainage and red blood cell transfusion requirements, in particular for blood samples taken after protamine administration. Therefore, our study provides evidence that intra-operative platelet function analysis using impedance aggregometry is superior to pre-operative drug history in predicting postoperative bleeding. This is also supported by the systematic review and meta-analysis recently published by Corredor et al. 14.