Platelet-rich plasmapheresis in cardiac surgery: a meta-analysis of the effect on transfusion requirements

Our purpose was to determine whether intraoperative platelet-rich plasmapheresis in cardiac surgery is effective in reducing the proportion of patients exposed to allogeneic red cell transfusions.

A systematic search for prospective, randomized trials of platelet-rich plasmapheresis in cardiac surgery, using MEDLINE, HEALTHSTAR, Current Contents, "Biological Abstracts," and EMBASE/Excerpta Medica up to August 1997, was completed. Trials were included if they reported either the proportion of patients exposed to allogeneic red cells or the units of allogeneic red cells transfused. Trials were abstracted by 2 independent investigators and the quality of trial design was assessed with the use of a validated scale.

Seventeen references met the inclusion criteria (1369 patients [675 control: 694 platelet-rich plasmapheresis]). Platelet-rich plasmapheresis reduced the likelihood of exposure to allogeneic red cells in cardiac surgery (odds ratio 0.44; 95% confidence interval 0.27, 0.72, P = .001). Platelet-rich plasmapheresis had a small but statistically significant effect on both the volume of blood lost in the first 24 hours (weighted mean difference -102 mL; 95% confidence interval -148, -55 mL, P < .0001) and the mean units transfused (weighted mean difference -0.33 units; 95% confidence interval -0.43, -0.23, P < .0001). However, platelet-rich plasmapheresis was only marginally effective (odds ratio 0.83, 95% confidence interval 0.34, 2.01, P = .68) for "good" quality trials, whereas it appeared very effective in trials with poor methodologic quality (odds ratio 0.33, 95% confidence interval 0.17, 0.62, P = .0007).

Although platelet-rich plasmapheresis appeared effective in decreasing the proportion of patients receiving transfusions after cardiac operations, the quality of most of the supporting trials was low and the benefit was small in trials of good quality. Further clinical trials should be completed.