Editorial 

Evaluating Cryopreserved and Liquid-Stored Platelets for Bleeding During Cardiac Surgery

Timothy A. Pritt

JAMA Surg Published Online: February 18, 2026

doi: 10.1001/jamasurg.2025.6723

There is no substitute for blood during the treatment of hemorrhage. From the first reported successful transfusion of human blood to treat postpartum hemorrhage in 1825,¹ to the early use of whole blood in combat casualty care during World War I,² and to the development of blood preservation and banking,^3,4 significant advances have been made in the donation, storage, and use of this precious resource. One of the ongoing challenges in the field is related to the duration of safe blood product storage prior to transfusion. Currently, fresh frozen plasma may be stored for up to 1 year prior to transfusion, packed red blood cells for 42 days, and whole blood for 35 days, but platelet storage is limited to 7 days at room temperature.⁵ This relatively brief storage period creates significant challenges related to inventory management, distribution, clinical availability, and waste avoidance.

Platelet transfusion is common in the US, with 2 618 000 units distributed in 2023 alone.⁶ Under standard conditions, platelets are stored at room temperature. Recent work has somewhat extended the storage period, demonstrating that platelets stored at 1 to 4 °C for up to 14 days are effective for bleeding, although there is shorter posttransfusion circulation time.⁷ The US Food and Drug Administration recently approved the use of cold-stored platelets for the treatment of hemorrhage.⁸

Cryopreservation has proven an effective strategy for extending the storage life of several blood products. This approach has become standard for plasma, extending the storage life from 26 days for liquid plasma to 1 year for the fresh frozen equivalent.⁵ Cryopreserved red blood cells may be stored for up to 10 years prior to thawing, deglycerolization, and transfusion.⁹ Whole blood cryopreservation may be possible but needs further development and is not currently in clinical use.¹⁰ Cryopreservation of platelets, first described clinically in 1956, offers the advantage of significantly extending the shelf life of this product. Subsequent research has demonstrated that these units appear to be safe and efficacious, albeit with decreased yield during recovery and decreased circulating survival after transfusion (as reviewed by Kelly and colleagues¹¹). Cryopreservation remains attractive for platelets due to its significant potential to mitigate supply and distribution challenges, as well as to allow specific product storage for patients with prior alloimmunization.¹¹

In the current study by Reade and colleagues reported in JAMA,¹² the investigators seek to answer the question of whether cryopreserved platelets are less effective and safe than conventional liquid-stored platelets in the treatment of cardiac surgical bleeding. They completed a multicenter, double-blind, parallel-group noninferiority randomized clinical trial in 11 Australian tertiary hospitals, randomizing 388 patients who were at elevated risk of needing platelets during or after cardiac surgery. A total of 202 patients received platelets, with 98 receiving conventional liquid-stored units and 104 receiving cryopreserved products. The investigators found that the primary outcome of chest drain blood loss during the first 24 hours after intensive care unit (ICU) admission did not differ between groups, and noninferiority could not be established because of overlapping confidence intervals between the 2 groups. Furthermore, they found that cryopreserved platelet transfusion was associated with increased blood loss during the operative and perioperative period, as well as increased need for red blood cell, plasma, and cryoprecipitate transfusion. Their data also indicate that patients receiving cryopreserved platelets experienced longer times to extubation, as well as delays to ICU and hospital discharge. Taken together, the data suggest that cryopreserved platelets were associated with similar adverse events to standard storage platelets but that they were less effective to treat bleeding. Therefore, in this trial, cryopreserved platelets were not noninferior to standard storage platelets.

Important supplemental data presented in the current study include a planned post hoc analysis of ICU admission bleeding. This analysis suggests that there was worse performance in the cryopreservation group. There was a lower platelet nadir seen in patients treated with cryopreserved platelets—even in those patients who also received open-label transfusion. This seems to confirm previous data that show that cryopreserved platelets, while appearing to be hemostatic in function, demonstrate more rapid clearance from the circulation after transfusion.¹³ There may also be decreased survival in the cryopreserved group, although this trend did not reach statistical significance.

This trial by Reade and colleagues¹² demonstrates several significant strengths. This is a large, multicenter study with administration of platelets performed in a method designed to ensure that the clinical team was blinded to which therapy was being given. This can be difficult to achieve in blood product trials, but the investigators used opaque covers over the platelet units to ensure appropriate blinding. Patients received up to 3 units of study platelets based on clinician discretion and usual Australian practice. The rationale for study power design is well thought out. Based on their own preliminary data, the investigators calculated that transfusion of 202 patients would give 80% power to detect a noninferior limit in the primary outcome of 20%. They chose the 20% noninferior limit based on the mean volume of red blood cell units used for transfusion in Australia, as well as the blood loss associated with the anticipated primary outcome. The chosen primary outcome of postsurgical chest drain bleeding volume is measurable, objective, pragmatic, and practical. There were no preset protocols guiding or dictating transfusion practices or thresholds. Thus, the current study can be viewed as a real-world application of transfusion practices.

However, there are limitations that must be considered. There was less blood loss at 24 hours after ICU admission than was found during the authors’ preliminary observations. This less-than-anticipated blood loss may have led the current trial to be underpowered to detect differences between the 2 groups. In patients receiving cryopreserved platelets, there was a higher intraoperative blood loss, but these data were missing or incomplete for one-third of the patients in the study. Even though the patient groups appear well matched, it is unclear if the higher intraoperative blood loss reflects surgical control of hemorrhage in the operating room, operating technique, undocumented or unplanned difficulty of the operation, or coagulopathic bleeding. The study also found that a higher proportion of patients in the cryopreserved group (25% vs 3.1%) received open-label standard platelets prior to transfusion of 3 units of trial platelets. Additional analysis presented in the supplemental data suggested that there were site-level differences in this practice and that this decreased over time, but these findings may point to subtle loss of the effective blinding of treatment groups. There was no assessment of posttransfusion platelet function or concurrent viscoelastic testing, and thus the potential mechanisms underlying the differences between the groups, such as postthaw functional differences or rapid clearance from the circulation, remain unknown. In addition, the study findings somewhat contradict a previous trial of cryopreserved platelets in cardiac surgery, which found that patients who received these units required less transfusion and demonstrated decreased blood loss.¹⁴ The reasons for these different findings and associated conclusions remain unclear. Also, the current results are most applicable to high-risk patients undergoing cardiac surgery. Additional investigation will be needed to evaluate patients in other common bleeding situations, such as trauma, obstetric hemorrhage, or gastrointestinal bleeding, given the differences in pathophysiology seen in each of these populations.

So, what are the key takeaway messages from this study? Based on the current data, the routine use of cryopreserved platelets for bleeding in high-risk patients undergoing cardiac surgery cannot be recommended. The safety profile of cryopreserved platelets suggests that they would be useful in situations where standard platelets are not available. The logistical advantages and longer shelf life of cryopreserved platelets suggest that these are potentially of benefit in smaller hospitals and austere settings that lack a ready supply of platelets. Additional studies are needed to explore several aspects of these findings. One follow-on study that would shed additional insight would be to explore functional dose equivalence between the 2 strategies, with viscoelastic testing and functional platelet evaluation used to determine if a higher cryopreserved platelet dose is needed to achieve equivalence to standard storage platelets.

Taken together, the current study indicates that cryopreserved platelets are not a frontline therapy for bleeding after cardiac surgery but must remain a second-line option. The results of this trial suggest that there is no need for changes in current transfusion practices, but additional research is needed to determine the potential role of cryopreserved platelets in our transfusion armamentarium. Thus, our approach to platelet transfusion for bleeding after cardiac surgery should remain fresh, not frozen.