I wish to share the results of a retrospective observational correlation study I completed recently. While it is well established that the platelet count is negatively impacted during cardiopulmonary bypass (CPB),^1-7 what is less clear is the extent to which this occurs per unit time of CPB. While 4 hours on CPB likely causes a greater decline in platelet count than 1 hour on CPB, exactly by how much remains uncertain. If Thromboelastography (TEG), Rotational thromboelastometry (ROTEM), or standard labs are not available, what may be a useful tool to help decide if platelet transfusion may be indicated is a nomogram or a “rule of thumb” for predicting the degree in platelet count drop over time on CPB.

At our institution, platelet counts are routinely obtained preoperatively, after rewarming on CPB, and immediately after protamine administration during procedures requiring CPB. For coronary artery bypass grafting (CABG)–only procedures on adult patients requiring CPB between January 2023 and December 2024, platelet counts were retrospectively obtained to determine the change in platelet counts between CPB start, rewarming, and CPB stop. These data points were then plotted against the readily available timestamps of “CPB start,” “rewarming,” and “CPB stop.” It was assumed that the preoperative platelet count represented that at the initiation of CPB. Additionally, because protamine is administered shortly after separation from CPB, the platelet level associated with protamine administration was assumed to represent that at the time of CPB separation. A simple linear regression model was used to determine the degree of correlation between time on CPB and change in platelet levels. At our institution, Sorin S5 roller pumps are exclusively used for CPB.

Patients were excluded if the procedure was a re-sternotomy; the procedure was emergent; the patient received platelet transfusions at any time prior to protamine administration or after protamine lab acquisition; there were incomplete data; or more than 1 pump run was required.

In total, 540 cardiac surgeries were performed at our institution in the specified time frame. Of these procedures, 228 met the inclusion criteria. Figure 1 depicts the study flow diagram. The overall characteristics of these patients are listed in Table 1. The median CPB time was 86 minutes. On average, the platelet count decreased by a median of 41% of the baseline value by the time protamine was administered.